' l A Vol. 50, No. 1/2, 1997 World Health STATISTICS Quarterly Rapport trimestriel de STATISTIQUES sanitaires mondiales Food safety and foodborne diseases Salubrite des aliments et maladies d'origine alimentaire World Health Organization Organisation mondiale de la Sante Geneve The World health statistics quarterly replaces (since 1978) the monthly World health statistics report(published since 1967) and its forerunner the Epidemiological and vital stallstics report (published since 1947). It deals with the detailed analysis of selected health topics of current interest. Starting with Vol. 41 (1988), the Quarterly contains articles 1n either French or English with a summary in both languages. Annual subscription Sw. fr 121 - Price per copy Sw. fr. 39.- Material from the Quarterly may be reproduced providing due acknowledgement is made. Symbols used in tables Preliminary, approximate or estimated data Data not available Nil or magnitude negligible Category not applicable T Total M Male F Female N Absolute numbers The World Health Organization welcomes requests for perm1ss1on to reproduce or translate its publications, in part or 1n full. Applications and enquiries should be addressed to the Office of Publications, World Health Organ12ation, Geneva, Switzerland, which will be glad to provide the latest information on any changes made to the text, plans for new ed1t1ons, and reprints and translations already available © World Health Organization 1997 Publications of the World Health Organ12ation en1oy copyright protection in accor- dance with the provisions of Protocol 2 of the Universal Copyright Convention All rights reserved. The des1gnat1ons employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the Secretariat of the World Health Organ12at1on concerning the legal status of any country, territory city or area or of its autorit1es, or concerning the delim1tat1on of its frontiers or boundaries. The mention of specific companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organ12ation 1n preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are d1st1nguished by initial capital letters Le Rapport trimestriel de statisllques sanita,res mond1ales remplace (depuis 1978) le Rapport de statistiques sanitaires mondiales (publie depuis 1967) et son precurseur leRapportepidem1ologiqueetdemogra- phique (publie depuis 1947) II presente des analyses detaillees sur des sujets specifiques d'interet courant A compter du Vol. 41 (1988), le Tnmestriel presente des articles originaux en frarn;:a1s ou en anglais, accompagnes d'un resume dans les deux langues. Prix de l'abonnement annuel Fr. s. 121 - Le numero Fr. s. 39.- La reproduction d'extraits duTrimestriel est autorisee, sous reserve d'indication de la source. Explication des signes Donnee preliminaire, approximative ou estimative Oonnee non disponible Zero ou quantite negl1geable Categorie non applicable T Total M Masculin F Feminin N Nombres absolus L'Organisat1on mondiale de la Sante est touiours heureuse de recevo1r des deman- des d'autorisat1on de reproduire ou de traduire ses publications, en part1e ou integralement. Les demandes a cet ettet et les demandes de rense1gnements do1vent ~tre adressees au Bureau des Publications, Organisation mond1ale de la Sante, Geneve, Suisse. qu1 se fera un plaisir de fournir les rense1gnements les plus recents sur les changements apportes au texte. les nouvelles editions prevues et les re1mpress1ons et traductrons deja d1sponibles. © Organisation mondiale de la Sante 1997 Les publ1cat1ons de l'Organisat1on mond1ale de la Sante benef1c1ent de la protection prevue par les dispositions du Protocole No. 2 de la Convention universelle pour la Protection du Dro1t d'Auteur. Taus dro1ts reserves. Les appellations employees dans cette publ1cat1on et la presentation des donnees qui y f1gurent n'impl1quent de la part du Secretariat de !'Organisation mondiale de la Sante aucune prise de pos1t,on quant au statut 1und1que des pays, ternto1res, v1lles ou zones. ou de leurs autor1tes. n1 quant au trace de leurs front1eres ou llmites. La mention de f1rmes et de produ1ts commerc1aux n'1mplique pas que ces firmes et produ1ts commerc1aux sont agrees ou recommandes par !'Organisation mond1ale de la Sante de preference a d'autres Sauf erreur ou om1ss1on, une maiuscule 1nit1ale indique qu'il s'ag1t d'un nom depose. IX ISSN 0043-8510 Printed in Switzerland 97/11561 -Atar SA. Geneva- 5300 ex. World Health Statistics Quarterly Rapport trimestriel de statistiques sanitaires mondiales Vol. 50, N° 1/2, 1997 Food safety and foodborne diseases Contents Food safety: a commonly underestimated public health issue (Introduction) Fritz K. Kaferstein Global estimation of foodborne diseases Yasmine Motarjemi & Fritz K. Kaferstein Surveillance of foodborne diseases: what are the options? M. W Borgdorff & Yasmine Motarjemi Epidemiology of foodborne diseases: tools and applications Morris E. Potter & Robert V. Tauxe Epidemiology of foodborne diseases: a worldwide review Ewen G.D. Todd Chronic health effects of microbial foodborne disease V. Kelly Bunning et al. Economic costs and trade impacts of microbial foodborne illness Jean C. Buzby & Tanya Roberts Foodborne listeriosis Jocelyne Rocourt & Jacques Bille Enterohaemorrhagic Escherichia coli Yoshifumi Takeda Foodborne salmonellosis Tomas M. Gomez et al. Virus transmission via food Dean 0. Cliver Foodborne diseases in travellers R. Y Cartwright & M. Chahed The role of food in the epidemiology of cholera 3 5 12 24 30 51 57 67 74 81 90 102 M. John Albert et al. 111 Establishment of microbial safety criteria for foods in international trade International Commission on Microbiological Specifications for Foods 119 Wld hlth statist. quart., 50 (1997) Salubrite des aliments et maladies transmises par Jes aliments Somma ire Food safety: a commonly underestimated public health issue (Introduction) [anglais seulement] Fritz K. Kaferstein Estimation mondiale des maladies d'origine alimentaire [resume] Yasmine Motarjemi & Fritz K. Kaferstein Surveillance des maladies d'origine alimentaire: quelles options? [resume] M.W. Borgdorff & Yasmine Motarjemi Epidemiologie des maladies d'origine alimentaire: instruments et application [resume] Morris E. Potter & Robert V. T auxe Epidemiologie des maladies d'origine alimentaire: situation mondiale [resume] Ewen G.D. Todd Effets chroniques des maladies d'origine alimentaire sur la sante [resume] V. Kelly Bunning et al. Coot economique et impact commercial des maladies microbiennes d'origine alimentaire [resume] Jean C. Buzby& Tanya Roberts Listeriose d'origine alimentaire [resume] Jocelyne Rocourt & Jacques Bille Escherichia coli enterohemorragique [resume] Yoshifumi Takeda Les salmonelloses d'origine alimentaire [resume] Tomas M. Gomez et al. Transmission des virus par les aliments [resume] Dean 0. Cliver Maladies d'origine alimentaire chez les voyageurs [resume] R. Y. Cartwright & M. Chahed ROie des aliments dans l'epidemiologie du cholera [resume] M. John Albert et al. Etablissement de criteres de securite microbiologique pour les aliments faisant l'objet d'un commerce international [resume] International Commission on Microbiological Specifications for Foods 5 11 20 29 46 55 65 72 79 88 100 109 116 123 Improving the safety of street-vended food Gerald Moy et al. Monitoring and assessment of dietary exposure to chemical contaminants 124 Ramesh V. Bhat & Gerald G. Moy 132 Modernizing food control systems: the Tha"i experience Pakdee Pothisiri 2 150 Moyens d'ameliorer la salubrite des aliments vendus dans la rue [resume] Gerald Moy et al. Suivi et evaluation de !'exposition d'origine alimentaire aux contaminants chimiques [resume] 131 Ramesh V. Bhat & Gerald G. Moy 147 Modernisation des systemes de contrOle des denrees alimentaires - !'experience de la Tha"ilande [resume] Pakdee Pothisiri 154 Rapp. trimest. statist. sanit. mond., 50(1997) Food safety: a commonly underestimated public health issue Introduction For a number of decades, the World Health Orga- nization (WHO) has recognized the widespread nature of foodborne diseases and their impact on communities in both the developing and de- veloped world. The annual incidence of some 1.5 billion episodes of diarrhoea in children under 5 years of age, and the more than 3 million result- ant deaths are an indication of the magnitude of the problem, as a significant proportion of the diarrhoeal disease cases are foodborne in origin. The 1992 FAO/WHO International Conference on Nutrition recognized that hundreds of millions of people worldwide suffer from communicable and non-communicable diseases caused by con- taminated food. These diseases take a heavy toll in human life and suffering, particularly among infants and children, the elderly and other suscep- tible persons. They also create an enormous social, cultural and economic burden on communities and their health systems. Developing countries bear the brunt of the problem, with a wide range of diseases, including cholera, campylobacteriosis, Escherichia coli infec- tions, salmonellosis, shigellosis, typhoid and para- typhoid fevers, brucellosis, hepatitis A, poliomyeli- tis, and foodborne parasitic diseases, such as amoe- biasis, trematode infections, and many others. As a result of improvements in food safety education and standards of personal hygiene, in the develop- ment of basic sanitation, safe water supplies, effec- tive vaccination programmes ( especially for polio- myelitis), improved food control infrastructures, and the application of technologies such as pas- teurization, many foodborne diseases ( e.g. polio- myelitis, brucellosis, cholera, typhoid and para- typhoid fevers and milk-borne salmonellosis) have been either eliminated or considerably reduced in industrialized countries. Nevertheless, many food- borne diseases remain a major public health prob- lem. In recent years, a number of industrialized countries have even experienced a significant in- crease in the incidence of foodborne diseases. In several countries, surveys have pointed to an an- nual incidence of 5-10% of the population in- volved. In addition, a number of foodborne patho- gens such as multi-antibiotic resistant Salmonella typhimurium, Listeria monocytogenes and E. coli 0157 have emerged as significant threats to public health. Recent E. coli infections in Australia.Japan, Scotland, and the United States of America have involved thousands of individuals, with many Wld hlth statist. quart., 50 (1997) deaths, predominantly among children and eld- erly people. The health consequences of foodborne dis- eases are varied, and depend on the nature of the disease, the stage at which the treatment is initi- ated, the patient's age, and the individual's suscep- tibility. Acute symptoms include diarrhoea, vomit- ing, abdominal pain, cramps, fever and jaundice. For most foodborne diseases, recovery takes from a few days to a few weeks. However, many foodborne infections can lead to serious and chronic sequelae and affect the cardiovascular, renal, articular, res- piratory or immune systems. For example, salmo- nellosis may cause reactive arthritis in some sub- jects; infections due to E. coli 0157:H7 can evolve, particularly in children, into haemolytic uraemic syndrome, characterized by acute renal failure. The clinical manifestation of listeriosis may in- clude septicaemia, meningitis, encephalitis, osteo- myelitis, endocarditis, and in pregnant women the disease may lead to abortion, stillbirth, or malfor- mation of the foetus. The overall case-fatality rate is 20-30%. Transplacental infections due to Toxo- plasma gondii occur in about 45% of infected preg- nant women. It is estimated that in about 3 out of every 1 OOO pregnancies, the foetus is affected by toxoplasmosis. Infections due to helminths, both in terms of magnitude and health consequences, are a cause of great concern. Examples are helminthic infec- tions due to Trichinella spiralis, Taenia saginata, and Taenia solium which are acquired through the con- sumption of raw or uncooked meat. These para- sites present a serious public health problem, par- ticularly in countries where there is habit of con- suming raw meat and clandestine slaughtering of animals occur. Ascariasis, transmitted among others through contaminated vegetables, is one of the most common parasitic infections, and is esti- mated to affect over 1 OOO million people. Trema- todes, such as Clonorchis, Fasciola, Dpisthorchis and Paragonimus spp., infect some 40 million people, particularly in Asia, Africa and Latin America. Overall, more than 10% of the world's population is at risk of becoming infected by these parasites, which are transmitted through the consumption of raw or inadequately processed freshwater fish, shellfish or aquatic plants. Foodborne infections are also one of the most important underlying factors of malnutrition and, indirectly, of respiratory infections. Repeated epi- 3 sodes of foodborne infections can, over a period of time, lead to malnutrition, with a serious impact on the growth and immune system of infants and chil- dren. The infant whose resistance is suppressed becomes more vulnerable to other diseases (in- cluding respiratory infections) and is subsequently caught in a vicious circle of malnutrition and infec- tion. Many infants and children do not survive under these circumstances. In addition to the human suffering caused by foodborne diseases in terms of death and ill- health, substantial economic costs are incurred, affecting individuals and families, industries, health care systems and entire communities. These include, among other costs, loss of income due to absenteeism, medical care costs, costs of investi- gation of foodborne diseases outbreaks, loss of income due to closure of business, product recalls and dispositions, legal pursuits and claims. At the national level, epidemics of foodborne disease af- fect both tourism and trade in food. When cholera broke out in Peru in 1991, over US$ 700 million were lost in fish and fishery products exports. In the three months following the start of the epi- demic, US$ 70 million were lost because of closure of food service establishments and a decrease in tourism. Although food safety is a major public health problem of increasing significance, many public health authorities do not appreciate its importance for community health and development. National programmes for the prevention of foodborne dis- eases are too frequently considered a matter of low priority, and some countries have yet to develop legislation and a national infrastructure specifically addressing food safety. The food industry also of- ten fails to prioritise the issue. While larger compa- nies may be aware of their responsibility and com- mitted to producing safe food, many smaller busi- nesses remain unaware of both their responsibili- ties and the best approaches to ensuring the safety of their products. Consumers are vital partners in achieving safe food, but all too frequently they are unaware of their role and fail to recognize the 4 significant risks associated with foodborne disease until a deadly outbreak affects their community. Consequently, consumers do not often play as ac- tive a role as they ought to in preventing food- borne diseases. In terms of strategies for preventing foodborne disease, reference should be made to recommenda- tions of the FAO/WHO International Conference on Nutrition. Effective national food safety strate- gies require an approach which integrates regula- tion and education. The regulatory approach, in- cluding legislation and its enforcement, is impor- tant for ensuring the safety of food produced and processed for sale, but it should be noted that a substantial quantity of food preparation and handling occurs in the domestic situation. Thus, education of the general public and food handlers is also an essential element of preventing food borne disease. WHO continues to promote the concept of shared responsibility among government, industry and consumers in the fight against foodborne dis- eases. To ensure the safety of food, each group must integrate its efforts through involvement in re- search, regulatory control, infrastructure develop- ment, epidemiology and training, education and learning. This issue of the World health statistics quarterly is dedicated to food safety. It reviews the epidemiology of foodborne diseases, identifies a number of emerging food borne pathogens, aims to enhance awareness of the health and economic costs of foodborne disease and specific problems encountered in the developing countries. It further provides guidance, particularly to developing coun- tries, by identifying how the Hazard Analysis and Critical Control Point (HACCP) approach can be applied to improve the safety of food, particularly street-vended food, by reviewing the establishment of microbiological criteria and examining the mod- ernization of the food control system in Thailand, as a model for developing countries. Fritz K. Kaf erstein Director, Programme of Food Safety and Food Aid World Health Organization, Geneva, Switzerland Rapp. trimest. statist. sanit. mond., 50 (1997) Global estimation of foodborne diseases Yasmine Motarjemia & Fritz K. Kafersteinb For the purpose of decision-making and establish- ing priorities, the importance of having informa- tion on prevailing health problems cannot be over- emphasized. This is even more important in the case of countries with frail economies and only limited resources for resolving the most urgent of their health problems. Foodborne diseases are one of the most wide- spread health problems in the contemporary world and they have implications both on health and development (1). However, because of the absence of foodborne disease surveillance systems in some countries, or weaknesses in existing pro- grammes, reliable information on the magnitude of the problem is not available. It is estimated that the reported incidence of foodborne diseases rep- resents less than 10%, or maybe even less than 1 %, of the real incidence. Surveys in a few coun- tries indicate that foodborne diseases may be 300- 350 times more frequent than the reported cases tend to indicate (2-4 ). Furthermore, because of differences in the collection and reporting sys- tems, data from different parts of the world are not usually comparable, thus making a global es- timation of foodborne diseases a rather difficult undertaking. This article describes the constraints in the sur- veillance of foodborne diseases at national and international levels. Based on a number of assump- tions, and following an analysis of the nature of foodborne diseases, their epidemiology, and the possible exposure of populations to the various foodborne pathogens, a semi-quantitative estima- tion is made of the magnitude of foodborne dis- eases of microbiological and parasitical origin in various regions of the world. Constraints in collection of information Collection of information at international and national levels. With the exception of cholera (which is subject to the International Health Regulations), there is no obligation to report foodborne diseases interna- a Scientist, Food Safety and Food Aid Programme, World Health Organization, Geneva, Switzerland. b Director, Food Safety and Food Aid Programme, World Health Organization, Geneva, Switzerland. Wld hlth statist. quart., SO (1997) tionally. Attempts to provide a global picture of foodborne diseases are usually hampered by differ- ences in national surveillance systems, where such systems exist. In most countries, only a few diseases which are or may be of food borne origin appear on the list of notifiabl.e diseases. In addition, the reported diseases are not pre- sented in a uniform manner, for example, while one country may report the incidence of shigellosis and amoebiasis separately, another may report them jointly under the term dysentery. Again, sever- al foodborne diseases are sometimes reported col- lectively under the term food poisoning. However, the meaning of the term food poisoning varies from country to country, and not infrequently it is used to represent different groups of diseases. Refer- ence to the International Classification of Disease code, which would help to identify the disease in question, is also frequently omitted. Some countries report the total number of cases of a foodborne disease, including sporadic cases, while others mainly collect information on the number of outbreaks and the number of cases involved in the outbreaks. As a result, information on sporadic cases is not collected in many coun- tries and data from different countries are difficult to compare. Another constraint- but of a different nature - in collecting information at the international level relates to the repercussions that information on foodborne disease epidemics may have on food trade and tourism. Concern about the possibility of food exports being rejected and/ or loss of tourism have been a disincentive for many governments to release information on foodborne disease epi- demics (see the Introduction to this issue). Weakness of infrastructure. In most countries the surveillance infrastructure is weak or non-existent. Resources for investigation purposes are meagre: the number of foodborne disease outbreaks may often outstrip the human resources available for investigation. Not infre- quently, patients are discouraged from visiting health centres unless their symptoms are serious or persistent. As a result many cases remain unre- ported. Furthermore, in many countries, particularly in the developing world, laboratory resources and skills to identify pathogens are scarce, and etiology- 5 specific surveillance is often not possible. In the industrialized countries, decreasing resources for the public health infrastructure as well as for medi- cal care have also restricted possibilities for labora- tory-based surveillance. In addition, for some agents, e.g. enteric viruses, the absence of a simple and reliable diagnostic test makes surveillance dif- ficult. Unfortunately, even in countries where such a laboratory-based surveillance is carried out, the data are not regularly communicated to the food control agencies responsible for corrective and preventive actions, such as controlling food opera- tions and/ or providing health education. Problems in the investigation ol loodborne diseases Investigating foodborne disease outbreaks is often difficult, particularly when the source of the out- break is a food which has been prepared in private households, in small food service establishments, or by street food vendors. By the time the investiga- tion begins, there may not be any food left over for analysis. Consequently, evidence is often only cir- cumstantial, and the investigation inconclusive. In such cases, the identification of various risk factors can, over time, provide a statistical basis for assess- ing the most likely factors in the transmission of foodborne diseases. However, in many cases health authorities are not sufficiently trained or experi- enced to conduct rigorous investigations into food- borne disease outbreaks. The identification of some pathogens (e.g. rotavirus) in foods is also difficult, particularly when these are present in small numbers. Health authorities' perception of foodborne diseases Foodborne diseases have often received low prior- ity in public health programmes because they have been perceived as mild, self-limiting diseases. Their severe and chronic health consequences have often been overlooked. Sequelae such as can- cer, congenital blindness, reactive arthritis and meningitis resulting from foodborne illnesses do not usually figure in statistics. In addition, some foodborne diseases such as diarrhoeal diseases - particularly in infants and children - or cholera have traditionally been per- ceived as waterborne or as being transmitted from person-to-person. The relation between infant diarrhoea (and associated malnutrition) and the contamination of weaning food with pathogens is too often ignored. Not infrequently, foodborne diseases are mis- diagnosed and mistreated, particularly when the disease is of a new type or one which has been acquired abroad and is unknown to the health workers. 6 The general perception that foodborne dis- eases are benign means that there has been little incentive for investigating, reporting and monitoring the incidence of foodborne dis- eases beyond their treatment aspects. The eco- nomic consequences have also been over- looked. Lack of information on the real magnitude, health and economic consequences has meant that few resources have been allocated to programmes for the prevention of food borne diseases i.e. food safety (Box ]). /". '"'''~·· .. ~ Neglect in foodbome Lack ofapprcciation disease investigation of health significance of foodbome diseases and surveillance ' No priority and resources / assigned to food safety Public perception of toodborne diseases The public's perception of foodborne diseases varies. In some societies diarrhoea is not perceived as a symptom of disease and may even be consid- ered as a normal/ natural occurrence. People may also ignore the role of food and food handling in the transmission of diarrhoeal diseases, and may attribute the diarrhoea to other factors such as indigestion, teething, eating hot (spicy) foods, and superstition. For economic reasons, some popula- tion groups may also hesitate to seek medical assis- tance. In the better-informed societies, people might not seek a physician's help unless the symptoms are severe or long-lasting. Studies in industrialized countries indicate that a relatively small percent- age of people suffering from suspected foodborne illnesses consult a health worker: 5% in the Netherlands, 13.5% in New Zealand and 6.2% in Sweden (5,6,2). Although similar data are not available from developing countries, there is rea- son to believe that an even smaller proportion of cases come to the notice of health services, partic- ularly in so far as the adult population is con- cerned. The motivation to consult a physician not only depends on the severity of the illness but also on the costs involved. Experiences in Zambia indi- cate that when patients were requested to pay a fee for the medical visit, the number of patients consulting for diarrhoeal diseases, including chol- era, decreased. Rapp. trimest. statist. sanit. mond., 50 (1997) > . • African Region - Region africaine ~ South-East Asia Region - Region de (AFR) l'Asie du Sud-Est (SEAR) la Eastern Mediterranean Region - Region de la Mediterranee orientale (EMR) CJ Region of the Americas - Region D European Region - Region des Ameriques (AMR) europeenne (EUR) - Western Pacific Region - Region du Pacifique occidental (WPR) Global estimation of foodborne diseases The above-mentioned constraints make it difficult to compare data from different countries and to provide global estimates. Reported incidence data are generally used for the purpose of assessing trends in foodborne diseases in a specific country. To provide a global picture of foodborne dis- eases, while offsetting to some extent the uncer- tainty in the existing data, an attempt has been made to present their estimated occurrence in a semi-quantitative manner. Table 1 presents the re- sult of this work by WHO Regions (see also Map 1) . Here, foodborne diseases are classified into 4 groups, i.e. (i) not occurring(-) (ii) occurring occasionally or rarely ( +): the reported annual incidence is in the range of up to 1 case per 100 OOO (iii) occurring frequently (++):the reported an- nual incidence is in the range of 1 to 100 cases per 100 OOO (iv) occurring very frequently (+++): the re- ported annual incidence is over 100 cases per 100 OOO The following principles were applied when making the estimates: • When a disease is considered severe, e.g. botu- lism, and the degree of under-reporting likely to be low, the available data are taken as an indication oflevel of occurrence. The source of Wld hffh statist. quart., 50 (1997) data may either be from the literature or from national epidemiological reports. • When there is reason to believe that the inci- dence of a disease is highly under-reported - for instance the disease is mild or self-limiting - the classification is adjusted to correct for the under-reporting. For example, illnesses such as those caused by Staphylococcus aureus and Bacil- lus cereus are likely to be highly under-reported. Therefore, they are classified one level higher than the available data would suggest. • Whenever reliable data are not available for a countryorregion, the estimation has been based on: (i) reported incidence of all diarrhoeal dis- eases in infants and children, when the disease in question is related to diarrhoeal diseases. Tabl,e 2 shows the distribution of pathogens frequently identified in children with acute diarrhoea seen at treatment cen- tres in developing countries; and/or (ii) extrapolation from other regions, taking into account the likelihood of under- reporting, the nature of the disease, the foods eaten in the region, and the general standard of food safety as compared to the regions where data are available. The following examples illustrate how the semi- quantitative estimations of foodborne disease oc- currence have been carried out. 7 CD Ta ble 1 Es tim ati on o f o cc ur re nc e of dis ea se s w hic h ar e or m ay be fo od bo rne , W HO R eg ion s Ta ble au 1 Es tim ati on d e la fr6 qu en ce d es m ala die s d 'or igi ne a lim en tai re c er tai ne o u su pp os 6e , p ar R6 gio n OM S AM R W PR Dis ea se sa - Ma lad ies a AF R No rth Ce nte r & EM R EU RD SE AR Ne w- Ze ala nd , A us tra lia Au tre s p art ies No rd So uth C en tre & Ja pa n Ot he r P art s & Su d Nl le- Ze lan de , A us tra lie & Ja po n Ba cte ria l I nfe cti on s a nd In tox ica tio ns - To xl- lnf ec tio ns b ac ter ien ne s Ba cil lus ce re us ga str oe nte riti s - Ga str o-e nt6 rite a Ba cil lus ce re us ++ + ++ ++ + ++ + ++ ++ + ++ ++ + Bo tul ism - Bo tul ism e + + + + + + + + Br uc ell os is - Br uc ell os e +/+ +C + ++ +/+ +C · /+ /+ +C +/+ +C + +/+ +C Ca mp ylo ba cte rio sis - Ca mp ylo ba ct6 rio se ++ + ++ ++ + ++ + ++ ++ + ++ ++ + Ch ole ra - Ch ol6 ra +/+ +C - !+ +/+ +C + + - /+ + Cl os trid ium pe rtr ing en s e nt er itis - En t6r ite a Cl os trid ium pe rfr ing en s ++ + ++ ++ + ++ + ++ ++ + ++ ++ + Es ch eri ch ia co li d ise as e - Ma lad ies a Es ch eri ch ia co li ++ + + ++ + ++ + +/+ +C ++ + + ++ + Lis ter ios is - Lis t6r ios e + + + + + + + + Ty ph oid a nd p ara typ ho id fev ers - Ty ph o'id e et pa rat yp ho 'ide ++ + ++ ++ + ++ + ++ Sa lm on ell os is - Sa lm on ell os e ++ + ++ ++ + ++ + ++ /+ ++ C ++ + ++ ++ + Sh ige llo sis - Sh ige llo se ++ + ++ ++ + ++ + +/+ +C ++ + ++ ++ + St ap hy loc oc cu s a ur eu s int ox ica tio n - Int ox ica tio n pa r S tap hy loc oc cu s au re us ++ + ++ ++ + ++ + ++ ++ + ++ ++ + Vi bri o pa rah ae mo lyt icu s e nt er itis - En t6r ite a Vi bri o pa rah ae mo lyt icu s + + ++ +/+ +C ++ Vi bri o vu lni fic us se pti ce mi a - Se pti c6 mi e a Vi bri o vu lni fic us + +/+ +C +/+ +C ++ Ye rsi nio sis - Ye rsi nio se + +/+ +C + ~ ~ Vir al inf ec tio ns - Inf ec tio ns vi ra les §" He pa titi s A - H6 pa tite A ++ ++ ++ ++ ++ ++ ++ ++ m No rw alk vi ru s ga str oe nte riti s - Ga str o-e nt6 rite a vir us de N orw alk + + + + + + + + ,. .. .. . ,, Po lio m ye litis - Po lio my 61 ite ++ + + + + ++ + +!+ + Et ~ Ro tav iru s ga str oe nte riti s - Ga str o-e nt6 rite a ro ta vir us ++ + ++ ++ + ++ + ++ ++ + ++ ++ + ,.. ... ~ ;::;: Pr oto zo a inf ec tio ns - Inf ec tio ns a pr oto zo alr es :3 g Am oe bia sis - Am ibi as e ++ + + ++ + ++ I++ + + ++ + + ++ + ~ - Cr yp tos pri dio sis - Cr yp tos pri dio se ++ + ++ ++ + I ++ + ++ ++ + ++ ++ + . . . . . . . Gi ard ias is - Gi ard ias e ++ + ++ ++ + ++ I++ + +/+ +C ++ + ++ ++ + ij To xo pla sm os is - To xo pla sm os e ++ + ++ +!+ + + +!+ + ++ ++ - . . . . :: I § I i ~ 8: ........ is ~ "' Dis ea se sa - Ma lad ies a As ca ria sis - As ca rid ias e Clo no rch ias is - Clo no rch ias e Fa sc iol ias is - Di sto ma tos e Hy da tid os is - Hy da tid os e Op ist ho rch ias is ( 0. fe lin eu s) - Op isth orc hia se ( 0 . f eli ne us ) Op ist ho rch ias is ( 0. v ive rrim ) - Op isth orc hia se ( 0 . v ive rrim ) Ta en ias is/ cy sto ce rco sis - T6 nia se /cy sti ce rco se Tr ich ine llo sis - Tr ich ino se Tr ich ur ias is - Tr ich oc 6p ha los e AF R ++ + · /t+ C t/t +C t/t ++ C +I+ + ++ + AM R No rth Ce nte r & No rd So uth C en tre & Su d He lm int hla sls - He lm lnt hla se s + + + + ++ + t+ /tt +C ++ +!+ + ++ + a no t o cc ur rin g - ab se nte ; + o cc as ion al or r ar e - oc ca sio nn ell e o u ra re ; + + fre qu en t- fre qu en te; ++ + ve ry fre qu en t - Ire s f req ue nte . EM R ++ + ++ +I+ + + · /tC ++ + EU RD SE AR t/t +C ++ + · /t+ C · /t+ +C + · /tt +C · /tt +C + t/t +C + /t+ C t/t +C ++ + W PR Ne w- Ze ala nd , A us tra lia Au tre s p art ies & Ja pa n Ot he r P art s Nl le- Ze lan de , A us tra lie & Ja po n +!+ + ++ + · /+C t+ /tt +C · I++ + + /t+ tC +!+ + ++ ++ ++ D Ar me nia , A za rba ijan , G eo rgi a, Ka za ks tan , K yrg ys tan , T ad jikis tan , T urk me nis tan an d U zb ek ist an ar e inc lud ed in E MR or S EA R- L'A rm en ie, l'A ze rba idja n, la Ge org ie, le K az ak sta n, le Kir gh izis tan , le T ad jikis tan , le T urk me nis tan e tl' Ou zb ek ista n, so nt co m pri s d an s S EA R et EM R. c G rea t re gio na l v ar iat ion s - lm po rta nte s v ar iat ion s r eg ion ale s. Table 2 Pathogens frequently identified in children with acute diarrhoea seen at treatment centres in developing countries T1lll11u 2 Agents pathogenes frequemment identifies chez les enfants atteints de diarrhee aigue vus dans les centres de soins dans les pays en developpement Pathogen - Pathogene % Rotavirus 15-25 Escherichia coli - enterotoxigenic - enterotoxinogene 10-20 - enteropathogenic - enteropathogene 1-5 Shigella spp. 5-15 Campylobacter jejuni 10-15 Vibrio cholerae O 1 5-10 Salmonella (non typhi) 1-5 Cryptosporidium 5-15 Source :Ref.-Ref. (7) Botulism: Although mild cases do occur, the dis- ease usually has severe health consequences. Se- verely affected persons would seek medical advice. The degree of under-reporting is likely to be lower than for other diseases. The incidence of this dis- ease is reported by a number of countries, and data are available in national epidemiological surveil- lance programmes. Scientific literature is also a source of data for many countries. The microor- ganism is ubiquitous in nature and can therefore occur all over the world. Although a proportion of cases may remain unreported, available data justify classifying this disease as a rare or occasional world- wide. Brucellosis: This is a severe disease which is noti- fiable in numerous countries; its incidence has de- creased in European and North American coun- tries as a result of veterinary measures and the pasteurization of milk. In some southern Euro- pean countries e.g. Italy, France, Spain and Greece, where the traditional production of soft cheese from raw milk and the consumption of raw milk are still common, the disease has a higher level of occurrence. In developing countries where milk pasteurization is not yet systematic, the disease also occurs with higher frequency than in industrialized countries. Based on available infor- mation, the disease is classified as rare in most industrialized countries and frequent in developing countries, in particular in the Eastern Mediterra- nean Region. Cholera: Regarded as severe, this disease is sub- ject to international notification. The estimation of its occurrence is based on the number of notified cases. Based on the reported data, cholera is esti- 10 mated to be a rare to frequent disease, depending on the region. Campylobacteriosis, salmonellosis, shigel/osis: Data on incidence are available from a number of industrialized countries ( e.g. Japan, Australia, United States of America, and several European countries). Their global incidence in most devel- oping countries is unknown. However their etio- logical agents are among the most important for diarrhoeal diseases in infants and children as well as for travellers' diarrhoea. Therefore, for coun- tries or regions where data on these diseases are not available, the calculation is based on the esti- mated number of episodes of diarrhoeal diseases in infants and children and travellers' diarrhoea. Escherichia coli infections: Their incidence in most countries is unknown. Data from a few indus- trialized countries point to a relatively low occur- rence compared to other foodborne infections and to other regions. They are therefore classified as rare or occasional in industrialized regions. How- ever, pathogenic strains of E. coli are the leading cause of diarrhoeal diseases in developing regions. They are thus estimated to occur very frequently in regions where diarrhoeal diseases are highly preva- lent. Staphylococcus aureus intoxication, Bacillus cereus gastroenteritis, Clostridium perfringens enteritis: These diseases are reported in several countries under the general term food poisoningc. Their occurrence is highly underestimated because they are often mild and self-limiting. Some laboratories may also choose not to look for these agents. Therefore, although the reported incidence in industrialized countries would indicate that these diseases should be classified as rare or occasional it is likely that they occur frequently. Where data are not available, pre- dicted occurrence is based on data from other regions, taking into account the food safety stan- dards of the region under study in comparison to the region where data are available. Rotavirus gastroenteritis: Although its incidence is unknown, it is one of the major causes of diar- rhoeal diseases, occurring in both industrialized and developing countries. Amoebiasis: Its incidence is reported by a few countries only. In regions where the incidence is unknown, the estimation is based on comparison with regions having similar standards of food safety. Cryptosporidiosis, Giardiasis: Incidence is report- ed by a number of countries, are also available data and literature. They are frequently the cause of diarrhoeal diseases in both industrialized and de- c The term 'food poisoning' is ill-defined and non-scientific; its use is discouraged by WHO. Rapp. trimest. statist. sanit. mond., 50 (1997) veloping countries. In the former, both infections are frequently associated with waterborne out- breaks. Conclusion Weaknesses and variations in foodborne disease surveillance systems, where such programmes exist, make a global estimation of foodborne dis- eases difficult. However, such data are essential for raising awareness about existing problems, setting priority food safety measures, using resources in a cost-effective way, and evaluating the impact of measures. At the national level, countries should consider developing or strengthening their food- borne disease investigation and surveillance sys- tem. At the international level, there is need to provide further guidance in investigation and sur- veillance and to harmonize reporting systems. Therefore, although an exact global estimation is not possible because of uncertainty about infor- mation or lack of data, it is possible, by using a number assumptions, to provide a picture of the occurrence of foodborne diseases on a semi-quan- titative scale in various regions of the world. Acknowledgements The authors gratefully acknowledge the contribu- tion of the following persons in reviewing the manuscript: Dr Anthony Hazzard, University of Western Sydney, New South Wales, Australia; Dr K.E. Mott, Division of control of Tropical Dis- eases, World Health Organization, Geneva, Switzerland; Dr Morris E. Potter, Centers for Dis- ease Control, Atlanta, USA; Dr L. Savioli, Division of control of Tropical Diseases, World Health Or- ganization, Geneva, Switzerland; Dr M.C. Thuri- aux, Division of Emerging and other Communica- ble Diseases Surveillance and Control, World Health Organization, Geneva, Switzerland; and Dr Ewen C.D. Todd, Health Canada, Ottawa, Ontario, Canada. Summary Foodborne diseases are one of the most widespread health problems, but because of weaknesses in food- borne disease surveillance and variation in reporting Wld hlth statist. quart., 50 (1997) systems between countries, it is difficult to make an estimation of their true incidence. This paper describes the constraints in the collection of information on the incidence and/or prevalence of foodborne diseases, including investigation and reporting at national and international levels. It also makes an attempt to semi- quantifythe occurrence of food borne diseases of micro- bial and parasitical origin in different regions of the world. Resume Estimation mondiale des maladies d'origine alimentaire Les maladies d'origine alimentaire sont l'un des proble- mes de sante les plus repandus, mais ii est difficile d'en evaluer !'incidence vraie en raison de la faiblesse des systemes de surveillance et des differences entre les modalites de notification des cas d'un pays a l'autre. Cet article decrit les problemes qui se posent au niveau de la collecte des informations sur !'incidence et/ou la prevalence des maladies d'origine alimentaire, de !'in- vestigation et de la notification aux niveaux national et international. II tente egalement d'evaluer de fa<;:on semi-quantitative la frequence des maladies d'origine alimentaire microbiennes et parasitaires dans differen- tes parties du monde. References - Reference 1. WHOTechnicalReportSeries,No. 705.1984. (Therol.eoffood safety in health and development: report ofajoint FAO/WHO Expert Committee on Food Safety). OMS Serie de rapports techniques, N° 705, 1985 (Lasecurite tks produits alimentaires et son rol.e dans la sante et l.e deueloppement:rapportd'un Comite d'experts FAO/OMS de la securite des produits alimentaires). 2. Norling, B. Food Poisoning in Sweden: results of a field study. Report No. 41/94. National Food Administration, Uppsala, Sweden, 1994. 3. Todd, E.C.D. Preliminary Estimates of Costs of Foodborne Disease in the United States.Jou.ma/ of food protection, 52(8): 586-594 (1989). 4. Notennans, S. &: Hooenboom-Verdegall, A.S. Existing and emerging foodborne diseases. International journal of food microbiology, 15: 197-205 ( 1992). 5. Notennans, S. &: Van de Giessen. Foodborne diseases in the 1980s and 1990s. Food control, 4(3):122-124 (1993). 6. Hodges, I. Raw to coo/red food, Community awareness of safe food handling practices, Wellington, Ministry of Health, 1993. 7. World Health Organization. Readings on diarrhoea, student manual Geneva, WHO, 1992. 11 Surveillance of foodborne diseases: what are the options? M. W Borgdortta & Yasmine Motarjemib In this article, the word 'surveillance' refers to the systematic collection and use of epidemiological information for the planning, implementation and assessment of disease control; surveillance there- fore implies 'information for action' (1). The ob- jectives of foodborne disease surveillance are to: (i) determine the magnitude of the public health problem posed by foodborne diseases, and monitor trends; (ii) identify outbreaks of foodborne disease at an early stage in order to take timely remedial action; (iii) determine to what extent food acts as a route of transmission for specific pathogens, and identify high-risk foods, food practices and populations; (iv) assess the effectiveness of programmes to im- prove food safety; and (v) provide information to enable the formula- tion of health policies regarding foodborne diseases: this includes the formulation and pri- oritization of preventive strategies. In order to achieve the above objectives, various surveillance methods may be employed. Any choice of method will depend partly on the ob- jective under consideration. For instance, one method may be very useful in the early detection of outbreaks, but have severe limitations in estimating the size of the foodborne illness problem. The following methods for surveillance will be discussed in this paper: • Records for registration of deaths and hospital discharges • Disease notification • Sentinel surveillance • Laboratory surveillance • Outbreak investigation The objectives of foodborne disease surveil- lance can sometimes be more effectively and effi- ciently achieved by using epidemiological studies rather than by carrying out a continuous surveil- lance programme. Therefore, a section covering methods that go beyond the 'normal' methods of surveillance/ epidemiological research is also in- cluded in this article. a National Institute of Public Health and the Environment, Department of Infectious Diseases Epidemiology, Bilthoven, The Netherlands. b Scientist, Programme of Food Safety and Food Aid, World Health Organization, Geneva. 12 For each of these methods a description is given of the surveillance system, followed by an assess- ment of its usefulness in relation to the objectives presented in Box 1. The structure of the descrip- tion of the surveillance system is adapted from that proposed by the United States Centers for Disease Control and Prevention for the evaluation of sur- veillance systems (1-4). This article aims to provide managers of nation- al food safety programmes with a description of the strengths and weaknesses of the major tools of foodborne disease surveillance in order to assist them in strengthening existing surveillance activi- ties and possibly develop additional ones. The objectives and methods of surveillance of foodborne diseases may vary between countries de- pending on the availability of resources and the relative importance of foodborne diseases com- pared with other causes of morbidity and mortality. Suggestions on an approach to selecting surveil- lance methods are therefore presented in conclu- sion. Registration of deaths and hospital discharge diagnoses In most countries, physicians complete a death certificate when a person under their care dies. The certificate indicates the cause of death, name, address, date of birth, and gender of the deceased, and the date of death. In all hospitals in some countries and in some hospitals in others, hospital discharge diagnoses are registered by age and gen- der of the patient and include other information such as the duration of stay. The causes of death and hospital discharge diagnoses are usually classified according to the International Classification of Diseases (ICD), which is updated regularly, most recently in 1992 (ICD-10) (5).C Population under surveillance For death certificates the population under surveil- lance includes all those whose cause of death is medically certified; the coverage of the system is estimated at 35% (3). For hospital discharge diag- noses the population under surveillance includes those who may be admitted to a hospital participat- ing in a discharge diagnosis registration scheme. < In many countries, the ninth revision (ICD-9) is still in use. Rapp. trimest. statist. sanit. mond., 50 (1997) Information on death certificates may be ana- lysed centrally by an institution such as a Central Statistical Office. Annual reports may be produced, tabulating the number of people who died from various diseases by age and gender, possibly together with information on denominators ( the population at risk). For some diseases occasional reports may be produced by the central office, or by independent investigators. Hospital discharge diagnoses may be analysed locally in the hospital concerned, or nationally. Utilization of results Registration of causes of death may in a limited way contribute to estimating the size of the public health problem of foodbome diseases and their trends. Infectious diseases overall have been shown to be an important cause of death in the United States in recent decades (6). Diarrhoeal diseases were a specific problem as a cause of death among the elderly and young children. Although the ma- jority of these cases were classified as presumably non-infectious diarrhoea, a large proportion was considered by some authors to have been infec- tious (7). For conditions which are very likely attrib- utable to foodborne infections, such as non- typhoid salmonellosis (ICD code A02), campy- lobacteriosis (ICD code A04.5), enteritis due to Yersinia enterocolitica (ICD code A04.6), and other bacterial foodborne intoxication (ICD code A05), the number of hospital admissions and deaths is extremely small in comparison to the total number of cases of foodbome disease estimated from sources such as laboratory surveillance (8). Box 1 Other problems also arise. Firstly, diagnosis may be incomplete. For instance, it may be known that the patient had gastroenteritis, but salmonel- losis was not diagnosed. In elderly patients with multiple diseases, gastroenteritis may not be re- corded on the death certificate at all if other dis- eases are present. Secondly, the diagnosis may not be specific for foodbome infections. For instance, of all the intestinal infectious cases (ICD codes AOO-A09), an unknown and probably variable pro- portion is acquired through food. Thirdly, al- though in hospitals the quality of the diagnosis is usually good, patients are selected on the basis of severity of disease and access to hospital, making it difficult to calculate rates (3). Results can sometimes be used to quantify the occurrence of severe cases of particular infections (9, 10). Hospital-based registries are often used to improve patient care (3, 10), but they may also contribute to disease cost estimates, as cases admit- ted to hospital may be costly, compared to those receiving outpatient care only (9). In conclusion, notification of deaths and regis- tration of hospital discharge diagnoses play a lim- ited role in estimating the public health impor- tance of diseases that may be foodborne and in monitoring trends over time, in particular of the most serious outcomes (Box I). The usefulness of these systems for surveillance of foodborne dis- eases is limited since sensitivity is low, except for rare and serious conditions, and attribution to spe- cific foods is uncommon. In countries with limited other surveillance systems for foodborne disease, causes of death tend to be underreported, and Methods of foodborne surveillance which are useful in achieving specific objectives Specific objective• 2 3 4 5 6 7 Magnitude of the public health problem and monitor trends over time + ++ +++ ++ + N/A +++ ii Identify outbreaks and take action + ++ + ++ +++ N/A N/A iii Identify high risk foods, high risk food practices and high risk populations for specific pathogens + + + + +++ +++ + iv Health impact of food safety programmes + + ++ + N/A N/A ++ v Information for health policy preventive strategies: formulation and priority setting (integrating i-iv above) + ++ ++ ++ ++ ++ ++ Methods: 1 = Registration of deaths and hospital discharge diagnoses 2= Disease notification 3= Sentinel surveillance 4= Laboratory surveillance 5= Outbreak investigation 6= Case-control studies of sporadic cases 7= Population-based surveillance + to +++=degree of usefulness for achieving the specific objective ( + = of little use to+++ best use) • See objectives listed at the beginning of this article. Wld hlth statist. quart., 50 (1997) 13 coverage of hospital care may be incomplete as well. Countries with comprehensive death registra- tion and universal access to hospitals tend to have other, more sensitive systems in place for surveil- lance offoodborne diseases, including notification of selected foodborne illnesses, and sentinel and laboratory surveillance. Disease notification Notification of diseases may be legally required from physicians or other health workers only for selected conditions, and may be optional for others. Information often collected in addition to the diagnosis and date or week of diagnosis in- cludes: age and sex of the patient, sometimes name and address and possibly details of clinical symp- toms and exposure/risk factors. This information is usually analysed centrally, e.g. in the ministry of health, for the description of national trends, but may be done at regional or district level, in particu- lar for the detection of outbreaks. Population under surveillance The population under surveillance includes all persons at risk of conditions which are notifiable or are included in a national reporting scheme who would consult a physician or other qualified health worker for their condition. As the reporting system is passive, underreporting is a common problem ( 11). Notification requires that the physician diag- nose the diseases validly (i.e. with reasonable sensi- tivity and specificity) and reliably (i.e. if repeated by others give similar results). The physician should be aware that the condition is notifiable and have some incentive to notify, for instance because notification is legally required, or because meaningful action can and will be taken. Notifica- tion is likely to be more complete if the disease is more serious, is perceived to spread easily from person to person, and if preventive measures are available. Utilization of results Reports by medical practitioners may be used to identify outbreaks as discussed below under out- break investigation (12). Recently published out- break investigations which were started by health workers reports include some on botulism ( 13 ), cholera (14, 15), and haemolytic uraemic syn- drome or bloody diarrhoea in the presence of Escheri.chia coli O 157:H7 infection ( 16-20). Notifica- tions and reported diseases may also be used to describe the epidemiology of infectious diseases and monitor trends in disease incidence, e.g. seasonal variation of foodborne intoxication in the United Kingdom (8), and a change in the epidemi- ology of brucellosis in California from a mainly occupational disease among slaughterhouse 14 workers, to a mainly foodborne disease associated with consumption of unpasteurised milk and cheese (21). The system is extremely useful in detecting seri- ous problems that are relatively rare ( cholera, hae- molytic uraemic syndrome due to E. coli 0157), and may be used as an early warning sign leading to outbreak investigation, thereby contributing to understanding the epidemiology offoodborne dis- eases (Box 1). For the surveillance of most fopd- borne diseases the system has limited usefulness, as the reported clinical syndromes are often not spe- cific for foodborne disease. Sentinel surveillance In sentinel surveillance, selected health events (ac- cording to specific case-definitions) are monitored by selected health providers or facilities. For food- borne diseases relevant health events might in- clude notifiable and non-notifiable syndromes such as diarrhoea, dehydration, haemolytic urae- mic syndrome, or specific infections such as campy- lobacteriosis, salmonellosis or Escheri.chia coli 0157:H7 infection (22-25). Laboratory testing may be limited to those requested by the treating physi- cian (26), it may include examination of all samples submitted for a range of pathogens (25), or it may be performed for all patients attending with a se- lected condition such as gastroenteritis (23, 24, 27). Reporting may be on a weekly or monthly basis, or, if computer systems are used, continuous- ly. Computer systems are used, for instance, in sentinel networks of general practitioners in France for convenience of reporting, early feed- back, and early detection of outbreaks (26, 28). Population under surveillance Although efforts are usually made to select sentinel facilities in such a way that their catchment popula- tion is representative of the total population, they may be self-selected or purposely selected, as notifi- cation of selected conditions requires extra effort, and motivation of staff at sentinel stations is crucial for the completeness and quality of the data col- lected (3). The population under surveillance in- cludes people in the catchment area who report to the sentinel facility if they have complaints compat- ible with foodborne illness. Ideally, the catchment should be clearly defined although this depends on the organization of the health services in a particular country. If the catchment population is defined, the number of new cases of specified con- ditions can be translated into consultation rates per 10 OOO population (3). Incidence rates of dis- ease are not easily calculated, as only a proportion of cases consult a health worker and are diagnosed with the selected condition (24, 26). Consultation rates may be used to calculate trends, although results should be interpreted with care. Rapp. trimest. statist. sanit. mond., 50 (1997) Utilization of results In the Netherlands, sentinel surveillance has been used to determine the consultation rates of salmo- nellosis and campylobacteriosis with general prac- titioners (23, 30-31). Of the cases of gastroenteritis presenting to a general practitioner, approximate- ly 5% were attributable to Salmonel/,a infection and 15% to Campylobacterinfection. The overall consul- tation rates for gastroenteritis varied from 9 to 15 per 1 OOO practice population per year. In the Unit- ed Kingdom, sentinel surveillance is part of a set of studies aiming at determining the incidence, agents, risk factors, and socioeconomic costs of infectious intestinal disease, a substantial part of which is likely to be foodborne (24). In France, sentinel surveillance led to the identification of an epidemic of gastroenteritis, possibly attributable to rotavirus (and probably not foodborne) (28). In the United States, sentinel surveillance was used to determine the consultation rates for Campylobacter and Escherichia coli 0157:H7 infection at health maintenance facilities (22, 25). The use of sentinel surveillance in developing countries has been difficult since it is a resource- intensive method. However, in countries with well- developed primary health care systems, selected PHC centres may carry out sentinel surveillance for foodborne diseases. In conclusion, sentinel surveillance may be use- ful for estimating the public health importance of foodborne diseases and may also play an important role in estimating the impact of food safety pro- grammes, thereby contributing to achieving the first and fourth objectives of foodborne disease surveillance (Box 1). Its usefulness for detection of outbreaks of foodborne diseases is limited, unless these occur within one or more of the areas under surveillance. Laboratory surveillance Public health and clinical laboratories record re- sults of specimens obtained from patients, most commonly faecal samples from patients with diar- rhoea ( other entry criteria may be bloody diar- rhoea, bacteraemia, or haemolytic uraemic syn- drome (14, 19, 32)). Pathogens which may be in- cluded are Salmonella typhi, other Salmonella spp, Campylobacter spp, Listeria monocytogenes, Vibrio cho- /,erae and V parahaemolyticus, Brucel/,a spp, and more recently E. coli0157 (8, 12, 14, 19, 33-43). Laboratory surveillance for pathogens such as Shigella, rotavirus, small round structured viruses (SRSV) hepatitis A, and Cryptosporidium has lim- ited value for monitoring the incidence of food- borne diseases, as these are frequently transmitted from person-to-person through direct contact. However, laboratory surveillance for these patho- gens may lead to outbreak investigation implicat- ing food as the route of transmission (8, 11, 12, 44-49). Wld hlth statist. quart., 50 (1997) Population under surveillance The population covered by laboratory surveillance includes patients attending medical care facilities for whom a laboratory test is requested in those laboratories that participate in the surveillance sys- tem. As !aboratory participation in surveillance is often on a voluntary basis, the system may not cover all laboratories (22, 36, 39-40, 50). As laboratories may not all perform the same tests, coverage may be more comprehensive for some agents than for others (37, 39-41, 43, 50). Recording denominators (numbers of patient samples tested) for each pathogen would be very informative, but it is not usually carried out routinely (51). Reporting results Laboratories record results continuously, and may report at intervals (weekly, monthly, quarterly or annually) to a central surveillance centre (36-37, 42, 52). A high frequency of reporting is particular- ly important if laboratory surveillance is also used for the detection of outbreaks. For monitoring trends, a low reporting frequency may be ade- quate. In addition to written reports, some labora- tory surveillance systems send patient material or isolates to a central reference laboratory for confir- mation, typing, or determining resistance patterns (8, 36, 38-40, 50, 53-55). Ensuring quality and com- parability of the data from different laboratories is an important feature of this type of surveillance. Utilization of results Results of laboratory surveillance may be used in combination with other data to estimate the mag- nitude of the public health problem of specific food borne infections (56 ). These estimates may be crude as many selection biases are involved as sug- gested above. As selection biases may change over time, secular trends need to be interpreted with care. For instance, the increase in the reported incidence of Campylobacter infections observed in the 1980s in many industrialized countries was probably largely due to increased case ascertain- ment because of improved laboratory techniques. On the other hand, because no important changes occurred in the diagnostic or reporting character- istics for salmonellosis in the past two decades, and because selection biases would affect all Salmonel/,a serotypes in a similar manner, when Salmonella en- teritidis became the predominant serotype in West- ern Europe and the United States in the late 1980s and early 1990s, the change in reported incidence of this serotype likely reflected an important public health event (8, 52). In animals, Salmonella enteriti- dis was found mainly in chickens and eggs, and a number of outbreak investigations identified eggs as the source of the outbreak. This combined evi- dence led to the development of various interven- tion programmes in the egg production sector. 15 Laboratory surveillance may be used to detect outbreaks of foodborne disease, when cases are scattered and symptoms not very specific ( 19). Re- cent examples include outbreaks of E. coli 0157:H7 (57), Listeria monocytogenes (58), Salmonella paratyphi B (36), Salmonella enteritidis (35, 59), Salmo- nella javiana and Salmonella oranienburg ( 60 ), Shigel- la sonnei (47), and Vibrio chol.erae (61). Collaboration between European countries in laboratory surveil- lance of Salmonella under the Salm-Net project has indeed been shown to be useful, to cover other pathogens, such as verocytotoxin-producing E. coli. Although laboratory surveillance data derived from stable surveillance programmes can be used to determine the direction of trends, they are rare- ly adequate to estimate the magnitude of problems or to attribute observed changes in reported num- bers to specific causes. When combined with epide- miological data from other systems, laboratory sur- veillance improves the overall picture of the public health situation by providing etiological certainty for clinical-based surveillance data. In conclusion, laboratory surveillance is valu- able in that it makes use of available data to assess the occurrence of specific microbiological agents in sporadic cases and clusters, possibly associated with information on clinical symptoms and on ex- posure; it is also useful in detecting outbreaks. Its major limitation is selection bias, making interpre- tation of results complicated and usually requiring complementary sources of information. Labora- tory surveillance makes a particularly useful contri- bution in achieving the first and third objective of foodborne surveillance, namely to monitor trends and identify outbreaks (Box 1). Outbreak investigation Outbreak detection Outbreaks are detected by various means. Health workers, including medical practitioners, may note a shared exposure among self-reporting cases and report the cluster of cases to public health authori- ties, or they may routinely report selected condi- tions voluntarily or as required by law (e.g. botu- lism, cholera, haemolytic uraemic syndrome, bloody diarrhoea, listeriosis) (13, 16-17, 58, 62-64). Members of the general public or institutions such as schools, universities, or work places may detect an outbreak, for instance after a shared meal at a restaurant, canteen, party, reception, field day, or conference (8, 12, 35, 44-45, 65-72). Outbreaks may be relatively easily detected and investigated if those who have been at risk of exposure are known (for example, airline and cruise ships passengers) (14, 46, 65, 73-75) and populations in institutions such as hospitals, nursing homes and prisons (12, 46, 76-77). Laboratory surveillance may detect out- breaks, particularly those spread out over a region or country (8, 35-36, 47, 57, 60, 62). An outbreak 16 may be anticipated and investigated after detecting an increased risk of exposure, for instance by con- taminated water (78). Population under surveillance The population under surveillance includes those at risk of being involved in outbreaks provided they are accessible to investigating officers. They are likely to include people attending large functions (weddings, funerals), those eating in restaurants and obtaining food from take-aways or street ven- dors, and those obtaining pre-cooked or ready-to- eat foods from food industries. Those consuming food prepared at home are usually under-repre- sented in outbreak surveillance as the size of the outbreak is smaller and people are less likely to report it to public health officials (75, 79-80). For outbreaks detected by laboratory surveillance, se- lection bias may occur because of limited access to health care and laboratory services, and limited scope of available laboratory facilities, as discussed under the section laboratory surveillance. Reporting results Outbreak investigation is carried out by various disciplines, including public health physicians, epi- demiologists, food safety officers, environmental health officers and microbiologists (12, 18, 47, 64, 81-82). Outbreak investigators should analyse and report as soon as possible, to prevent additional cases. National and international compilation of reported outbreaks may be done annually or once every few years for monitoring trends (8, 12, 66, 83- 84 ). Methods of outbreak investigation Investigations may use epidemiological methods, microbiological methods or both. In the epidemio- logical approach a case-control or cohort design may be used. In a case-control study, cases with foodborne disease are compared with controls re- garding their food intake, food preparation prac- tices, and other possible risk factors in a given period of exposure. Controls should be representa- tive for the population from which cases were drawn but should not have a foodborne disease in the relevant period of time. The main outcome of a case-control study is an estimate of the relative risk of illness after various exposures. This is estimated by the odds ratio. In unmatched case-control stud- ies the odds ratio is calculated as the ratio of the odds of exposure in cases and the odds of exposure in controls. In matched case-control studies, the odds ratio is calculated as the number of exposed cases with non-exposed controls divided by the number of exposed controls with unexposed cases. In the recent literature, there are many examples ofa case-control design (14, 17-18, 35-36, 47-48, 60, 62, 64, 67, 70, 73, 76, 85). A more elaborate de- Rapp. trimest. statist. sanit. mond., 50 (1997) scription of case-control studies is given in basic textbooks on epidemiology ( e.g. 86). The applica- tion of case-control studies in diarrhoeal diseases is described by Cousens et al. (87). Schesselman and Breslow & Day present advanced discussions of methodological issues (88-89). A cohort study compares attack rates of the illness concerned in those having eaten certain food items and those who have not. The ratio of these two attack rates is expressed as a relative risk. Cohort studies are used somewhat less often than case-control studies, but may be a convenient and powerful tool if those at risk are easily listed, e.g. as passengers on a cruise, or guests at a party (44-45, 58, 68, 71-72, 74-76, 85, 90). Its advantage over a case-control study is that absolute risks (attack rates) can be obtained. When microbiological (including parasitologi- cal) methods can be employed, the pathogen caus- ing the outbreak may be identified in the patients (14, 16, 18, 35-36, 44-45, 47, 57, 58, 60, 63, 67, 73, 75-76, 82, 90), the suspected food (16, 18, 35-36, 44, 57, 60, 64, 68, 90), or both. Microbiological evidence is strongest if the etiological agent is char- acterized as specifically as possible, for instance through serotyping, phage typing, plasmid analy- sis, determining the pattern of antimicrobial resis- tance, or by using new molecular typing methods (8, 12-13, 35-37, 47, 64, 66-67, 71, 73, 76, 91-94). The strongest causal evidence is obtained if both epidemiological and microbiological evi- dence is obtained (18, 36, 44, 60, 90). However, this is not always feasible for a variety of reasons. Epidemiological evidence may be difficult or im- possible to obtain because epidemiological skills are not available, because the outbreak was de- tected too late for a successful epidemiological investigation or because it was too small to obtain statistically significant results. Microbiological evi- dence may be unavailable because food samples were not available or because the pathogen can- not be demonstrated easily in patients ( e.g. toxin of B. cereus or S. aureus (91, 95), or short period of excretion of the pathogen, or late contacting of patients) or because of laboratory limitations. For instance, small round structured viruses (SRSV, or Norwalk agent) cannot be identified in most laboratories and have therefore been underesti- mated as causes of outbreaks of foodborne illness (46, 96-98). Utilization of results Results of an outbreak investigation should be utilized immediately for taking rational measures to control the outbreak itself, such as withdrawal of the implicated product or adjustment of the pro- duction process. Results may find a wider applica- tion if the identification of high-risk foods or high- risk food practices is used for the prevention of further outbreaks (8, 12, 18, 22, 44-45, 48-49, 59- Wld hlth statist. quart., 50 (1997) 60, 62, 66, 68-70, 72-76, 82, 85-90, 95, 99-101). Publication of results is therefore very useful for health policy makers, and the producers, distribu- tors, preparers and consumers of the food con- cerned. Outbreak investigation may also contrib- ute to knowledge of the symptoms, dose-response relationships, and incubation period of the infec- tion concerned and contribute to risk assessment (49, 67, 68, 90, 102-103). For monitoring the incidence of foodborne ill- nesses over time, or making comparisons between countries, data from outbreak investigations have severe limitations, mainly because sporadic cases are not included. In addition, selection biases in the detection and investigation of outbreaks are likely to be strong and may be subject to differ- ences in time, place and groups of persons (75, 83). In conclusion, outbreak investigation is useful in that it responds to public demand and can pro- vide timely information to prevent further cases of foodborne disease. It may also be a unique source of information on routes of transmission of specific pathogens and the identification of high-risk envi- ronments and high-risk food-handling practices. Its major limitations are the high resource require- ments in terms of skilled manpower and laboratory facilities, and a strong selection bias towards large or serious outbreaks and against small clusters and mild disease. Outbreak investigation is particularly useful in achieving the second and third objectives of surveillance of foodborne diseases: take remedial action to limit the size of outbreaks, determine to what extent food acts as transmission routes of specific patho- gens, and identify high risk foods and food handling practices and high risk populations ( Box 1 ). Beyond surveillance: epidemiological research The objectives of foodborne disease surveillance can sometimes be achieved more completely or more efficiently by epidemiological studies than by a continuous surveillance programme. Categories of such epidemiological studies will be discussed below, with recent examples on how they may con- tribute to achieving the objectives of surveillance of foodborne diseases. A distinction is made be- tween epidemiological studies based on health fa- cilities and those based on population. Health care facility based Case reports: Reports on individual cases or case series may be used to generate hypotheses on asso- ciations between exposure and disease. Recent ex- amples include the suggestion that Cydospora caye- tanensis infection may be foodborne, based on a case report ( 104 ), and a description of bacterial pathogens, in particular E. coli 0157 among pa- tients with haemolytic uraemic syndrome ( 106). In the absence of control groups associations may be suggested by biological plausibility in individual 17 cases or a high prevalence of exposure in case series, although they cannot be determined epide- miologically. Case-control studies of sporadic cases: Health fa- cility based case-control studies may be particularly suitable to determine the extent to which disease can be attributed to specific pathogens, which foods act as a route of transmission for specific pathogens, and to identify high-risk foods and high-risk food practices. For instance, in Thailand, pathogens associated with diarrhoeal disease in children were determined in a hospital-based case- control study ( 106). In Ecuador, Guatemala, Peru, and the Philippines hospital-based case-control studies on cholera showed that foodborne trans- mission occurred through street vendors (61, 107- 109). Other examples include a hospital-based case-control study in Thailand on risk factors for Vihrio chol.erae 0139 infection (112) and a health centre-based case-control study on diarrhoeal dis- eases in Malysia ( 111 ). Similarly, cases detected through laboratory surveillance may be enrolled in a case-control study to identify high risk foods and high risk food prac- tices, especially for sporadic cases. Examples are case-control studies on Camp,ylobacter (79, 112), E.coli 0157:H7 (37), Listeria monocytogenes (113), Salmonella enteritidis ( 114 ), Salmonella javiana and oranienburg ( 60 ), and Y ersinia enterocolitica ( 115 ), re- sulted in the identification of high-risk foods for acquiring these infections. Population-based studies When Snyder & Merson attempted to estimate the magnitude of the global problem of diarrhoeal diseases, they decided to base their estimates on the most valid data obtained from longitudinal, prospective, community-based studies of stable populations with low migration rates ( 116 ). Similar- ly, such studies would also provide the most valid estimates of many foodborne diseases, as they can be used not only to estimate the incidence rates of diarrhoeal diseases, but also to identify pathogens and risk factors. However these studies have a num- ber of methodological problems such as the defini- tion of an episode of diarrhoea and the optimal intensity of follow-up ( 117). Bern et al. suggested in their update of the study by Snyder & Merson, that the incidence rates of diarrhoeal diseases reported increased with an increasing frequency or intensity of surveillance. The case definition used had no apparent pattern of effect on the incidence rates (118). Longitudinal population-based studies can de- termine the incidence rate of diarrhoeal diseases in well-defined populations. Provided the inci- dence is sufficiently large ( depending on the inci- dence rate, the size of the study population, and the duration of follow-up) risk factors may be de- termined as well. These epidemiological studies 18 require extensive resources and need to be careful- ly planned. Some of the issues needing consider- ations are selection of the study population (age groups, geographical areas), duration of the study (seasonality of specific infections), data collection methods (e.g. home visits, self-administered ques- tionnaires, telephone interviews), laboratory re- quirements (laboratory investigations on all cases? also on a control group without complaints?), fre- quency of follow-up, case-definition, and plan of analysis (how to attribute cases to foodborne trans- mission). For foodborne diseases other than diar- rhoeal diseases, the incidence rate may be too low for a population-based incidence study to be feas- ible. For those diseases, other methods of estimat- ing incidence need to be used (e.g. laboratory surveillance or disease notification). As population-based longitudinal studies pro- vide the most valid estimates of the incidence rates of foodborne diseases, they may be used to esti- mate the sensitivity of other surveillance systems in detecting cases of foodborne illness. For instance, in a population-based study in the Netherlands, the incidence rate of gastroenteritis was estimated at 45 per 100 person-years, 4.5% and 1.5% were attributable to Campylobacter and Salmonella infec- tion, respectively ( 119). This would correspond to an estimated 300 OOO cases of campylobacteriosis and 100 000 of salmonellosis in the Netherlands (population 15.2 million). Based on sentinel sur- veillance, it was estimated that annually approxi- mately 17 000 cases of campylobacteriosis and 5 OOO cases of salmonellosis are seen in general practice. In laboratory surveillance some 3 OOO cas- es of salmonellosis and 4 OOO cases of campylobac- teriosis are found annually. Finally, in outbreak investigations carried out by the Food Inspection Services in the period 1991-1994, 18 people were found in outbreaks attributed to Camp,ylobacter, and 290 in outbreaks attributed to Salmonella (80) (see Figs. la & lb). Similarly, in the United Kingdom, longitudinal population-based studies have been carried out, in combination with senti- nel surveillance and general practice-based case- con trol studies (24). Conclusion Managers of food safety programmes need infor- mation for advocacy of food safety issues and for decision-making in food safety programmes. In re- viewing information needs, it may be helpful to review the objectives of surveillance of foodborne diseases and determine which surveillance systems are best suited to provide the required information (Box 1). Once a decision is made on the type of information required, a review of available infor- mation should be made. Obviously, information which is available from routine data collection (and probably collected for other purposes as well) is the least costly to obtain. Rapp. trimest. statist. sanit. mond., 50 (1997) Fig. 1 A. Annual incidence of campylobacteriosis in the Netherlands according to different systems of surveillance A. Incidence annuelle de la campylobacteriose aux Pays-Bas selon les differents systemes de surveillance a A total of 18 cases were detected during 1991-1994. -Au total, 18 cas ont ete detectes entre 1991 et 1994. Source: Adapted from: P. Sockett, Ph.D. thesis. - D'apres: P. Socket!, these de doctoral. If it is decided that the available information is insufficient, additional surveillance activities may be needed. As additional data collection requires additional resources, this decision is likely to in- volve a process of prioritization: how important is surveillance of foodborne diseases compared with other surveillance or other public health activi- ties? An example of criteria which may be used in priority setting for surveillance is presented by Teutsch. In general, surveillance should only be established if the information obtained may be expected to be used for control measures (120). It may be helpful to decide who should act on the information: local authorities or national? The sur- veillance system should be designed in such a way that it reaches those who need to take action in time. Whatever options are chosen, no single system is likely to provide all the required information, or to provide perfect information. Rather, the follow- ing statement on surveillance in general also ap- plies to surveillance of foodborne diseases: "Since no one source is usually adequate, good public health decision-making requires the synthesis of data of varying quality from a wide range of sources as well as the critical interpretation of findings" (3). Acknowledgements The authors would like to acknowledge with thanks the contribution of the following persons in Wld hlth statist. quart., 50 (1997) B. Annual incidence of salmonellosis in the Netherlands according to different systems of surveillance B. Incidence annuelle de la salmonellose aux Pays-Bas selon les differents systemes de surveillance a A total of 290 cases were detected during 1991-1994. - Au total . 290 cas ont ete detectes entre 1991 et 1994. Source: Adapted from: P. Socket!. Ph.D. thesis . - D'apres: P. Socket!, these de doctoral. reviewing the document: Ilse van Asperen, Arie Havelaar, Matty de Wit (National Institute of Pub- lic Health and Environmental Protection, Bilthoven, The Netherlands); Jocelyn Rocourt (Institut Pasteur, Paris, France); Morris Potter (Centers for Disease Control , Atlanta, Georgia, United States of America); Michel Thuriaux, Guenael Rodier (Disease Surveillance and Control Unit of the Division of Emerging and other Com- municable Diseases Surveillance and Control, WHO, Geneva, Switzerland); Anthony Hazzard (WHO Collaborating Centre for Environmental Health, Sydney, Australia). Summary Epidemiological data are needed for a variety of reasons, namely, informing public health authorities about the nature and magnitude of foodborne illnesses and their epidemiology, for the early detection of food- borne disease outbreaks, and for the planning, imple- mentation and evaluation of food safety programmes. Thus, epidemiological surveillance of foodborne dis- eases is fundamental to any food safety programme. Various methods of foodborne surveillance may be utilized: (i)records for registration of deaths and hospital discharges; (ii) disease notification, (iii) sentinel surveil- lance, (iv) laboratory surveillance, (v) outbreak investi- gation and (vi) epidemiological research. 19 This article reviews each method, its advantages and disadvantages, and its relevance for meeting the vari- ous objectives and needs. Resume Surveillance des maladies d'origine alimentaire: quelles options ? Les donnees epidemiologiques sont necessaires a di- vers titres: informer les autorites de sante publique de la nature et de l'ampleur des maladies d'origine alimen- taire et de leur epidemiologie, detecter des leur debut les flambees de maladies d'origine alimentaire, et plani- fier, mettre en oeuvre et evaluer les programmes de salubrite des aliments. La surveillance epidemiologique des maladies d'origine alimentaire est done fondamen- tale pour tout programme portant sur la salubrite des aliments. Diverses methodes de surveillance peuvent ~tre utili- sees : (i) les registres des deces et des sorties d'hOpi- taux; (ii) les declarations de cas de maladies; (iii) la surveillance par reseau sentinelle; (iv) la surveillance par les laboratoires; (v) les investigations sur les flam- bees; (vi) la recherche epidemiologique. Cet article examine chaque methode, ses avantages et ses inconvenients, ainsi que son inter~! selon les divers objectifs et besoins. References/References 20 l. Berkelman, R.L. & Buehler,J.W. Sutveillance. In: Oxford textbook of public health. Vol. 2. Methods of public health. Oxford, Oxford University Press, 1991, pp 161-176. 2. Centers for Disease Control. Guidelines for evaluating sutveillance systems. Morbidity and mortality weekly report, 37 (suppl. S-5): 1-18 (1988). 3. Stroup, N.E. et al. Sources of routinely collected data for sutveillance. In: Principles and practice of public health surveillance. Teutsch, S.M. & Churchill R.E. (eds). New York, Oxford University Press, 1994, pp 31-85. 4. Klaucke, D.N. Evaluating public health sutveillance. In: Principles and practice of public health surveillance. Teutsch, S.M. & Churchill, R.E. (eds). New York, Oxford University Press, 1994, pp 158-174. 5. International statistical classifuation of diseases and related health problems. Tenth Reuision. Geneva, World Health Organization, 1992. Classification statistique internationale des maladies et des problemes de sante connexes. Dixieme Revision. Geneve, Organisation mondiale de la Sante, 1992. 6. Pinner, R. W. et al. Trends in infectious diseases mortality in the United States.JAMA, 275: 189-193 (1996). 7. Lew,J.F. etal. Diarrheal deaths in the United States, 1979 through 1987.JAMA, 265: 3280-3284 (1991). 8. Sackett, P.N. et al. Foodborne disease sutveillance in England and Wales: 1989-1991. Communicable disease report - review 159-Rl 73. 9. Koopman, J.S. et al. Patterns and etiology of diarrhea in three clinical settings. American journal of epidemiol.ogyogy, 119: 114-123 (1984). 10. Lengerich, E.J. et al. Severe giardiasis in the United States. Clinical infectious diseases, 18: 760-763 ( 1994). 11. Harkess,J .R. etal. Is passive sutveillance always insensitive? Americanjournalofepidemiol.ogy, 128: 878-881 (1988). 12. Djuretic, T. etal. General outbreaks ofinfectious intestinal disease in England and Wales 1992 to 1994. CDR review, 7- R63. 13. D'Argenio, P.etal. Type B botulism associated with roasted eggplant in oil - Italy 1993. Morbidity and mortality weekly report, 44: 33-36 (1995). 14. Eberhart-Phillips, J. et al. An outbreak of cholera from food setved on an international aircraft. Epidemiol.ogy and infection, 116: 9-13 (1996). 15. Finelli, L. et al. Outbreak of cholera associated with crab brought from an area with epidemic disease. journal of infectious diseases, 166(6): 1433-5 (1992). 16. Cameron, A.S. et al. Community outbreak of hemolytic uremic syndrome attributable to Escherichia coli O 111 :NM - South Australia, 1995. Morbidity and mortality weekly report, 44: 550-558 (1995) . 17. Tozzi, A.E. et al. A community outbreak of haemolytic- uraemic syndrome in children occurring in a large area of Northern Italy over a period of several months. Epidemiol.ogy and infection, 113: 209-219 (1994). 18. Alexander, E.R. et al. Escherichia coli 0157:H7 outbreak linked to commercially distributed dry-cured salami - Washington and California, 1994. Morbidity and mortality weekly report, 44: 157-160 (1995). 19. Bell, B.P. et al. A multistate outbreak of Escherichia coli 0157:H7 - associated bloody diarrhea and hemolytic uremic syndrome from hamburgers.JAMA, 272: 1349-1353 (1994). 20. Shukla, R. et al. Escherichia coli O 15 7 infection associated with a farm visitor centre. CD R Communicable diseases report, 5(6): R86-90 (1995). 21. Chomel, 8.8. et al. Changing trends in the epidemiology of human brucellosis in California from 1973 to 1992: a shift toward foodborne transmission. journal of infectious disease, 170(5): 1216-23 (1994). 22. Tauxe, R. V. Epidemiology of Campyl.obacter jejuni infections in the United States and other industrialized nations. In: Campyl.obacter jejuni - current status and future trends. Nachamkin, I. et al. (eds). Washington, American Society for Microbiology, 1992, pp 9-19. 23. Notennans, S. & Hoogenboom-Verdegaal, A. Existing and emerging foodborne diseases. International journal food microbiol.ogy, 15: 197-205 ( 1992). 24. Roderick, P. et al. A pilot study of infectious intestinal disease in England. Epidemiol.ogy and infection, 114: 277-288 (1995). 25. MacDonald, K.L. et al. Escherichia coli 0157:H7, an emerging gastrointestinal pathogen: results of a one-year, prospective, population-based study. JAMA, 259: 3567- 3570 (1988). 26. Flahault, A. et al. Sentinelle traces of an epidemic of acute gastroenteritis in France. Lancet, 346: 162-163 (1995). 27. Borgdorff, M.W. et al. Sutveillance of gastroenteritis (letter). Lancet, 346: 842-843 (1995). 28. Maurice, S. et al. Telematics: a new tool for epidemiological sutveillance of diarrhoeal diseases in the Aquitaine sentinel network. British medical journal, 300: 514-516 (1990). 29. Taylor, D.N. & Blaser, MJ. Campylobacter infections. In: Bacterial infections of humans, epidemiol.ogy and control. Evans, A.S. & Brachman, P.S. (eds). New York: Plenum, 1991, pp 151-172. 30. Hoogeboom-Verdegaal, A.M.M. et al. Epidemiological and microbiological research on the occurrence of acute gastrointestinal illnesses in sentinel populations in Amsterdam and Helmond, 1987-1991. Bilthoven, RIVM report 149101011, 1994. Rapp. trimest. statist. sanit. mond., 50 (1997) 31. Goosen, E.S.M. et al. Incidence of acute gastroenteritis in sentinels in the Netherlands, 1992-1993. Report 149101012. Bilthoven, RIVM, 1995. 32. Skirrow, M.B. et al. Campylobacterbacteraemia in England and Wales, 1981-91. Epidemiology and infection, 110: 567-573 (1993). 33. Skirrow, M.B. Foodborne illnesses - Campylobacter. Lancet, 336: 921·923 (1990). 34. Skirrow, M.B. &: Blaser, MJ. Clinical and epidemiological considerations. In: Campylobacter jejuni - current status and future trends. Nachamkin, I. et al. (eds). Washington, American Society for Microbiology, 1992, pp ~8. 35. Evans, M.R. etal. Consecutive Salmonella outbreaks traced to the same bakery. Epidemiology and infection, 116: 161-167 (1996). 36. Desenclos,J.C. et al. Large outbreak of Salmonella enterica serotype paratyphi B infection caused by a goats' milk cheese, France, 1993: a case finding and epidemiological study. British medicaljourna~ 312: 91-94 (1996). 37. Finelli, Let al. Enhanced detection of sporadic Escherichia coli 0157:H7 infections - New Jersey.July 1994. Morbidity and mortality weekly report, 44: 429 (1995). 38. Cohen, M.L. &: Tauxe, R.V. Drug-resistant Salmonella in the UnitedStatcs-anepidemiologicalperspective. Science, 234: 964-969 ( 1986). 39. Mshar, P.A. et al. Laboratory screening for Escherichia coli Ol57:H7-Connecticut, 1993. Morbidity and mortality weekly report, 43: 192-194 (1994). 40. Ostroff, S.M. et al. Surveillance of Escherichia coli O 157 - Isolation and confirmation, United States, 1988. Morbidity and mortality weekly report, 40 ( suppl. SS. l): 1-6 ( 199 l). 41. Pierard, D. et al. Results of screening for verocytotoxin· producing Escherichia coli in faeces in Belgium. European journal of clinical microbiology and infectious diseases, 9: 198- 201 (1990). 42. Sharp, J.C.M. et al. Escherichia coli 0157 infections in Scotland. Journal of medical microbiology, 40: ~9 ( 1994). 43. Schuchat, A. et al. Epidemiology of human listeriosis. Clinical microbiology review, 4 (2): 169-83 ( 1991). 44. Millard, P.S. et al. An outbreak of cryptosporidiosis from fresh-pressed apple cider.JAMA, 272: 1592-1596 (1994). 45. Patterson, T. etal. Outbreak ofSRSV gastroenteritis at an international conference traced to food handled by a post symptomatic caterer. Epidemiology and infection, 111(1): 157-62 (1993). 46. Kaplan,J.E. et al. The frequency ofa Norwalk-like pattern of illness in outbreaks of acute gastroenteritis. American journal of public health, 72: 1329-1332 ( 1982). 47. Kapperud, G. et al. Outbreak of Shigella sonnei infection traced to imported iceberg lettuce. journal of clinical microbiology, 33: 609-614 ( 1995) . 48. Skala, M. etal. Food borne hepatitis A Missouri, Wisconsin, and Alaska, 1990-1992. Morbidity and mortality weekly report, 42(27): 526-529 (1993). 49. Kohn, M.A. et al. An outbreak of Norwalk virus gastroenteritis associated with eating raw oysters - implications for maintaining safe oyster beds. JAMA, 273: 466-471 (1995). 50. Food and Drug Administration, US Dept of Agriculture, Centers for Disease Control. Establishment of a national surveillance program for antimicrobial resistance in Salmonella. Morbidity and mortality weekly report, 45: 110-111 (1996). 51. Skirrow, M.B. A demographic survey of campylobacter, salmonella and shigella infections in England. A Public Health Laboratory Service survey. Epidemiology and infection, 99: 647-657 (1987). 52. Rodrigue, D.C. et al. International increase in Salmonella enteritidis: a new pandemic? Epidemiology and infection, 105: 21-27 (1990). 53. Endtz, H.P. et al. Quinolone resistance in Campylobacter isolated from man and poultry following the introduction Wld hlth statist. quart., 50 (1997) of fluoroquinolones in veterinary medicine. Journal of antimicrobial chemotherapy, 27(2): 199-208 (1991). 54. Lee, L.A. et al. Increase in antimicrobial-resistant Salmonella infections in the United States, 1989-1990. Journal of infectious diseases, 170: 128-134 ( 1994). 55. MacDonald, K.L. etal. Changes in antimicrobial resistance of Salmonella isolated from humans in the United States. JAMA, 258: 1496-1499 (1987). 56. Waters, J.R. et al. Infection caused by Escherichia coli 0157:H7 in Alberta, Canada, and in Scotland: a five-year review, 1987-1991. Clinical infectious diseases, 19: 834-843 (1994). 57. Tumey, C.etal. Escherichiacoli0157:H7 outbreak linked to home-cooked hamburger- California,July 1993. Morbidity and mortality weekly report, 43: 213-216 ( 1994). 58. Riedo, F.X. et al. A point source foodborne listeriosis outbreak: documented incubation period and possible mild illness. Journal of infectious diseases, 170(3): 693-6 (1994). 59. Mishu, B. et al. Outbreaks of Salmonella enteritidis infec- tions in the United States, 1985-1991. Journal of infectious diseases, 169: 547-552 (1994). 60. Hedberg, C.W. et al. A multistate outbreak of Salmonella javiana and Salmonella oranienburg infections due to consumpion of contaminated cheese. JAMA, 268: 320~ 3207 (1992). 61. Ries, A.A. et al. Cholera in Piura, Peru: a modern urban epidemic. journal of infectious diseases, 166(6): 1429· 33(1992). 62. Davis, M. et al. Update: multistate outbreak of Escherichia coli 0157:H7 infections from hamburgers - Western United States, 1992-1993. Morbidity and mortality weekly report, (1993) 42: 258-263. 63. Knubley, W. et al. Foodborne botulism-Oklahoma, 1994. Morbidity and mortality weekly report, 44: 200-202 ( 1995). 64. Orr, P. et al. An outbreak of diarrhea due to verotoxin producing Escherichia coli in the Canadian Northwest Territories. Scandinavian journal of infectious diseases, 26(6): 675-84 (1994). 65. Benoit, V. et al. Foodborne outbreaks of enterotoxigenic Escherichia coli - Rhode Island and New Hampshire, 1993. Morbidity and mortality weekly report, 43: 81-91 ( 1994). 66. Cowden, J.M. et al. Outbreaks of foodborne infectious intestinal disease in England and Wales: 1992 and 1993. Communicab/,e disease report review, 1995: R109-Rll 7. 67. Holtby, I. et al. Outbreak of Salmonella enteritidis phage type 4 infection associated with cheese and onion quiche. Communicab/,e disease report review, 1995: Rll8-Rll9. 68. Luby, S. et al. A large outbreak of gastroenteritis caused by diarrheal toxin-producing Bacillus cereus. Journal of infectious diseases, 1452-1455 ( 1993). 69. Wood, R.C. et al. Campylobacter enteritis outbreaks associated with drinking raw milk during youth activities. JAMA, 268: 3228-3230 (1992). 70. Mintz, E.D. et al. Foodborne giardiasis in a corporate office setting. journal of infectious diseases, l 67 ( 1): 250-3 (1993). 71. Rodrigue, D.C. et al. A university outbreak of Escherichia coli 0157:H7 infections associated with roast beef and an unusually benign clinical course. journal of infectious diseases, 172(4): 1122-5 (1995). 72. Slaten, D.D. et al. An outbreak of Bacillus cereus food poisoning are caterers supervised sufficiently. Public health report, 107(4): 477-80 (1992). 73. Lew, J .F. et al. An outbreak of shigellosis aboard a cruise ship caused by a multiple-antibiotic-resistant strain of Shigella Jl.exneri. American journal of epidemiology, 134: 41 ~20 (1991). 74. Boyce, T.G. et al. Vibrio cho/,erae 0139 Bengal infections among tourists to Southeast Asia: an intercontinental foodborne outbreak. Journal of infectious diseases, 172(5): 1401-4 (1995). 21 75. Hedberg, C.W. etal. An international foodborne outbreak of shigellosis associated with a commercial airline. JAMA, 268(22): 3208-12 (1992). 76. Evans, M.R. et al. A hospital outbreak of Salmonella food poisoning due to inadequate deep-fat frying. Epidemiology and infection, 116: 155-160 (1996). 77. Lo, S.V. et al. The role of the pre-symptomatic food handler in a common source outbreak of food-borne SRSV gastroentritis in a group of hospitals. Epidemiology and infection, 113: 513-521 (1994). 78. Gutteridge, W. &: Haworth, E.A. An outbreak of gastrointestinal illness associated with contamination of the mains supply by river water. Communicable disease report review, 4: R50-R51 (1994). 79. Adak, G.K. et al. The Public Health Laboratory Service national case-control study of primary indigenous sporadic cases of campylobacter infection. Epidemiology and infection, 115: 15-22 (1995). 80. Goosen, E.S.M. et al. Investigations of food borne diseases by Food Inspection Services in the Netherlands, 1991- 1994. Journal of food protection (in press). 81. Archer,D.E.&:Kessler,D.A.Foodborneillnessin the 1990s (letter).JAMA, 269: 2737 (1993). 82. Aristeguieta, C. et al. Multistate outbreak of viral gastroenteritis associated with consumption of oysters - Apalachicola Bay, Florida, December 1994-January 1995. Morbidity and mortality we,ekly report, 44: 37-39 ( 1995). 83. Bean, N.H. et al. Foodborne disease outbreaks, 5-year summary, 1983-1987. Morbidity and mortality weekly report, 39 (SS-1): 15-57 (1990). 84. WHO surveillance programme f<rr control of foodb<rrne infections and intoxications in Europe. Sixth report, 1990-1992. Berlin, Federal Institute for Health Protection of Consumers and Veterinary Medicine, 1995. 85. Gessner, B.D. &: Beller, M. Protective effect of conventional cooking versus use of microwave ovens in an outbreak of salmonellosis. American journal of epidemiology, 139: 903-909 (1994). 86. Beaglehole, R. et al. Basic epidemiology. Geneva, World Health Organization, 1993. 87. Cousens, S.N. et al. Case-control studies of common chil.dhood diseases, the example of diarrhoea. London, Macmillan Press, 1995. 88. Schlesselmann, JJ. Case-control studies, design, conduct, analysis. Oxford, Oxford University Press, 1982. 89. Breslow, N.E. &: Day, N.E. Statistical methods in cancer research. The analysis of case-control studies. Lyon, Inter- national Agency for Research on Cancer, 1980. 90. Khodr, M. et al. Bacillus cereus food poisoning associated with fried rice at two child day care centers- Virginia, 1993. Morbidity and mortality we,ekly report, 43: 177-178 (1994). 91. Aucott, J.N. Food poisoning. In: Blaser, MJ. et al. (eds). Infections of thegastrointestinaltract. New York, Raven Press, 1995. 92. Brian, M.J. etal. Evaluation of the molecular epidemiology of an outbreak of multiply resistant Shigella sonnei in a day- care center by using pulsed-field gel electrophoresis and plasmid DNA analysis. Journal of clinical microbiology, 31: 2152-2156 (1993). 93. Yagupsky, P. et al. Use of multiple markers for investigation of an epidemic of Shigella sonnei infections in Monroe County, New York. Journal of clinical microbiology, 29(12): 2850-5 (1991). 94. Taylor, D.N. &: Echevenia, P. Diarrhoeal disease: current concepts and future challenges. Molecular biological approaches to the epidemiology of diarrhoeal diseases in developing countries. Transactions of the Ruyal Society f<rr Tropical Medicine and Hygiene, 87(suppl 3): 3-5 (1993). 95. Terranova, W &: Blake, P.A. Bacillus cereus food poisoning. New England journal of medicine, 298: 143-144 ( 1978). 96. Centers for Disease Control. Viral agents of gastroenteritis. 22 Morbidity and mortality weekly report. 39(suppl. RR-5): 1-23 (1990). 97. Centers for Disease Control. Recommendations for collection of laboratory specimens associated with outbreaks of gastroenteritis. Morbidity and m<rrtality weekly report, 39(suppl. RR-14): 1-13 (1990). 98. Hedberg, C.W. &: Osterholm, M.T. Outbreaks of food- borne and waterborne viral gastroenteritis. Clinical microbiology review, 6: 199-210 ( 1993). 99. Rampling, A. Raw milk cheeses and Salmonella - pasteurisation and strict hygiene prevent serious morbidity and death. British medicaljourna~ 312: 67-68 (1996). 100. Petersen, C. Cryptosparidium and the food supply. Lancet, 345(8958): 1128-9 (1995) . 101. Rabold, J.G. et al. Cyclosp<rra outbreak associated with chlorinated drinking water (letter). Lancet, 344: 1360-1361 (1994). 102. Glynn,J.R. &: Palmer, S.R. Incubation period, severity of disease, and infecting dose: evidence from a Salmonella outbreak. American journal of epidemiology, 136: 1369-1377 (1992). 103. Oishi, I. et al. A large outbreak of acute gastroenteritis associated with astrovirus among students and teachers in Osaka, Japan. Journal of infectious diseases, 170: 439-443 (1994). 104. Connor, B.A. &: Shlim, D.R. Foodborne transmission of Cyclospora (letter). Lancet, 346(8990): 1634 (1995). 105. Bitzan, M. et al. The role of Escherichia coli 0157 infections in the classical (enteropathic) haemolytic uraemic syndrome: results of a Central European, multicentre study. Epidemiology and infection, llO: 183-196 (1993). 106. Echeverria, P. et al. Case-control study of endemic diarrhea disease in Thai children. Journal of infectious diseases, 159: 543-548 (1989). 107. Weber,J.T. etal. Epidemic cholera in Ecuador: multidrug resistance and transmission by water and seafood. Epidemiology and infection, 112(1): 1-11 (1994). 108. Koo, D. et al, Epidemic cholera in Guatemala, 1993: transmission of a newly introduced epidemic strain by street vendors. Epidemiology and infection, 116: 121-126 (1996). 109. Lim-Quizon, M.C. et al. Cholera in metropolitan Manila: foodborne transmission via street vendors. Bulletin of the W<rrl.d Health Organization, 72(5): 745-749 (1994). Lim-Quizon, M.C. et al. Le cholera a Manille : trans- mission alimentaire par les vendeurs des rues [resume). Bulletin de ['Organisation mondiale de la Santi, 73(5): 748- 749 (1994). 110. Hoge, C.W. et al. Epidemiologic study of Vibrio cholerae 01 and 0139 in Thailand: at the advancing edge of the eighth pandemic. American journal of epidemiology, 143(3): 263-8 (1996). 111. Knight, S.M. et al. Risk factors for the transmission of diarrhoea in children: a case control study in rural Malaysia. International journal of epidemiology, 21 ( 4): 812-8 (1992). 112. Kapperud, G. et al. Risk factors for sporadic Campylobacter infections: results of a case-control study in south eastern Norway. Journal of clinical microbiology, 30: 3117-3121 (1992). 113. Schuchat, A. et al. Role of foods in sporadic listeriosis. I. Case control study of dietary risk factors. The Listeria Study Group.JAMA, 267(15): 2041-5 (1992). 114. Cowden, J.M. et al. Case-control study of infections with Salmonella enteritidis phage type 4 in England. British medicine, 299: 771-773 (1989). 115. Ostroff, S. M. etal. Sources of sporadic Yersiniaenterocolitica infections in Norway: a prospective case-control study. Epidemiology and infection, 112: 133-141 (1994). 116. Snyder,J.D. &: Merson, M.H. The magnitude of the global problem of acute diarrhoeal disease: a review of active surveillance data. Bulletin of the W<rrl.d Health Organization, 60(4): 605-613 (1982). Rapp. trimest. statist. sanit. mond., 50 (1997) Snyder,J.D. & Merson, M.H. Ampleur du probleme des diarrhees aigues: analyse des donnees de surveillance active [resume]. Bulletin de ['Organisation mondiale de ui Santi, 60(4): 612-613 (1982). 117. Baqui, A.H. et al. Methodological issues in diarrhoeal diseases epidemiology: definition of diarrhoeal epi- sodes. International journal of epidemiology, 20: 1057-1063 (1991). 118. Bern C. et al. The magnitude of the global problem of diarrhoeal disease: a ten-year update. Bulletin of the World Health Organization, 70(6): 705- 714 (1992). Wld hlth statist. quart., 50 (1997) Bern C. etal. L 'importance des maladies diarrheiques dans le monde : dix ans apres [resume]. Bulletin de [ 'Organisation mondialedeuiSanti, 70(6): 711-714 (1992). 119. Wit, de M.A.S. et al. A community-based suroey in four regi.ons of the Netheruinds into the incidence and burden of disease of gastro-enteritis, and Campywbacter and Salmonella infections. Report 149101014. Bilthoven, National Institute of Public Health and the Environment, 1996. 120. Teutsch, S.M. Considerations in planning a surveillance system. In: Principles and practice of public health suroeillance. Teutsch, S.M. & Churchill, R.E. (eds). New York, Oxford University Press, 1994, pp 18-30. 23 Epidemiology of foodborne diseases: tools and applications Morris E. Pottera & Robert V. Tauxea Goals of public health surveillance and epidemiology: Our food supply brings us into intimate contact with a variety of microorganisms that are present in the production and processing environments from around the world. Omnivorous creatures that we are, we may intercept a broad variety of pathogens present in the intestines, flesh and other tissues of the animals and plants that we eat. Defining what is safe to eat, and what is not, and which processes make food safe and keep it safe, has been a major human preoccupation since the dawn of humanity. Food safety is a complex matter that depends on a number of interrelated environmental, cul- tural, and socioeconomic factors. The purpose of epidemiology and surveillance is to define these factors, how they interact, and their relative impor- tance in foodborne infections. In the last century, the application of epidemiological methods has led to dramatic improvements in the safety of food and water in many parts of the world. However, new challenges continue to arise, meaning that new solutions continue to be needed. Epidemiology is the study of the occurrence and distribution of specific disease in populations, and of the factors that explain the distribution ( 1 ). The principal goal of these activities is to improve prevention. Epidemiologists work backwards from diagnosed illnesses to the sources of those illnesses to devise strategies that will prevent similar illnesses in the future. The targets for concern are those with the greatest actual or potential harm, which may vary from one population to another. Epide- miological research identifies new pathogens or diseases, defines how they spread, identifies poten- tial areas for control and prevention, and monitors the results of prevention efforts. For some airborne infections, such as measles or influenza, transmis- sion can be difficult to interrupt by physical bar- riers, and most prevention efforts are focused on vaccinating high-risk populations. For most food- borne infections, vaccines are not available, but physical barriers can lower the risk of transmission substantially. Therefore, controlling foodborne in- fections depends on understanding the mechan- ism of transmission well enough to prevent it. • \.VHO Collaborating Centre for Foodborne Disease Sur- veillance, Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention (CDC), Atlanta, GA, United States of America. 24 Detailed investigations of how foods become contaminated often reveal specific points at which the safety of the food was compromised. The chain of events involved in the harvest, processing, distri- bution, and final preparation of food is frequently quite long, and preventing contamination could be viewed as a form of industrial quality control. Therefore, the ultimate user of epidemiological efforts on foodborne disease is, to a great extent, the food industry itself. Public health surveillance is the organized col- lection of reports of specific disease in a popula- tion from those who make the diagnosis. Surveil- lance depends on a functioning medical care sys- tem that can identify specific conditions, the wil- lingness of clinicians and clinical microbiologists to report the cases they diagnose, and the re- sources needed to gather, verify and summarize the information. Most surveillance is passive, that is, it depends on reports voluntarily sent in by clinics or laboratories. Active surveillance means that public health authorities actively solicit re- ports of new cases on a regular basis by contacting clinicians or laboratories. Because some cases will always be undiagnosed and unreported, any sur- veillance system is inherently incomplete. The more active and the more complete reporting is, the more expensive is the surveillance system. For many diseases under most conditions of surveil- lance, the fraction of cases that go unreported remains relatively constant, so the cases that are captured by the surveillance system usually remain a useful and consistent sample of the whole. Once the disease of interest has been defined, two principal epidemiological approaches are used to characterize it further. The first is descriptive epidemiology, in which the diagnosed cases of the disease are counted and characterized as to time of onset, symptoms of illness, and the age, gender, and location of persons affected. The goals of this description are to define the magnitude of the problem, suggest hypotheses that might explain its distribution, and monitor the effects of control measures. The second approach is that of analyti- cal epidemiology, the systematic comparison of ex- posures or possible risk factors among those who are ill with other comparable persons who have remained well. In this comparison, statistical meth- ods are used to identify specific food exposures or other risk factors that are associated with the food- borne disease. Diagnostic laboratories play a criti- Rapp. trimest. statist. sanit. mond., 50 (1997) cal role in both epidemiological activities by defin- ing the group of patients who are truly infected with the same organism and excluding patients infected with unrelated microorganisms. Once a suspected food has been identified, laboratory in- vestigations can provide supporting e,.,idence that it was indeed the source. Data on the public's health are used for differ- ent purposes by different levels of government. At the local level, where disease occurs and surveil- lance begins, public health units are responsible for collecting communicable disease data from health care providers, for investigating the source of infections, intervening in locally apparent out- breaks, and for routine prevention measures. Em- phasis is placed on those diseases for which imme- diate local public health response is both possible and necessary. At the provincial level, surveillance data are collated and analysed to identify out- breaks missed at the local level, monitor effective- ness of larger scale control programmes, define high-risk population groups, guide additional in- vestigations, and document the need for control programmes so that resources can be appropriate- ly allocated. Public health laboratories at the pro- vincial level sometimes provide important refer- ence confirmation and subtyping services that help to detect outbreaks and monitor long-term trends in specific pathogens. At the national level, the data are analysed and interpreted to assess public health impact of foodborne diseases and their trends, to recognize clusters that span pro- vincial or national boundaries, demonstrate the need for control programmes and allocate re- sources, monitor the effectiveness of control measures, establish appropriate public health and safety standards for international trade in foods, and develop hypotheses about risk factors for the acquisition and transmission of foodborne disease. At the national and international levels, analysis and interpretation may be conducted by persons who know little about the local systems or the methods by which the data were collected; without adequate attention to the limitations inherent in the data, erroneous conclusions may be drawn about the importance of public health problems and effectiveness of control programmes. At every level, the government authorities responsible for surveillance may be distinct from those responsible for food safety control, and their programmes may have evolved independently. If their procedures, goals, and paradigms are incompatible, the data generated by surveillance may have limited value to those who produce food or who regulate its safety at the national level. Because the quality of surveillance data is in part governed by the re- sources that are committed to gathering and anal- ysing them, the cost of surveillance should be sup- ported by the entities that will use the data to make Wld hlth st.atist. quart., 50 (1997) decisions, e.g. the food industry, the food control agency, or the international trade office. The specific diseases of interest to public health epidemiologists will vary from one country or time to another. In many parts of the developing world, the principal foodborne diseases of concern are cholera, typhoid fever, and dysentery. These are infections that are more or less adapted to the human host, and can spread from one human to another if contaminated faeces from an infected person in turn contaminates the food and water of others. This commonly occurs in the great periph- eral shanty towns of the developing world, where rapidly growing populations have outstripped water and sewage system capacities and where street vendors provide much of the food. The same conditions prevailed in European and North American cities in the I 9th Century, before the "sanitary revolution" systematically separated the streams of human sewage leaving the cities from the streams of human food and water coming into them. In more recent decades, the foodborne dis- eases most prevalent in the industrialized world have been zoonotic diseases such as non-typhoid salmonellosis, Campywbacter infections (2) and E. coli 0157:H7 infections (3) that are caused by contamination of human food with animal faeces (4). At the same time that humans gathered in the cities, populations of food animals also became more concentrated, were crowded together, fed processed food, and served by contaminated cen- tral water systems. The slum-like environments in which food animals live provide niches for many potential foodborne pathogens, which then spread to consumers. The early and persistent preoccupation of food control authorities with examining food animals for visible signs of illness at slaughter means that few truly ill animals are consumed in industrialized countries. However, the foodborne pathogens of the modern era tend not to cause illness in their animal hosts, but are present in the guts of many healthy looking ani- mals at slaughter; preventing the food from be- coming contaminated requires barriers between animal faeces and human foods, or guaranteeing that microbes that contaminate human foods are inactivated before the food is consumed. It is like- ly that in many parts of the world, foodborne concerns such as infections with SalmoneUa sero- type enteritidis, Campywbacter and the agent of transmissible spongiform encephalopathies will be controlled by greater attention to the safety of the food and water that the animals themselves consume. This future arena for food safety that brings the sanitary revolution to the agricultural sector will follow the application of epidemiolog- ical methods to answer the question of why cer- tain animals or production facilities harbour food- borne pathogens while others do not. 25 Tools tor epidemiology and surveillance Public health epidemiologists use five principal tools to monitor and investigate foodborne illness: surveillance of specific infections in humans, mon- itoring of contamination by specific pathogens in foods and animals, intensive outbreak investi- gations, collecting reports of outbreaks at the re- gional and national level, and studies of sporadic infections. Surveillance of food borne infections in humans For some infections, the clinician who makes the diagnosis may be required by public health author- ities to report the case. This permits reporting on the basis of clinicaljudgment, and does not require the availability of diagnostic laboratories. However, clinician-based reporting generally is less accurate, less complete, and less timely than systematic re- porting from clinical microbiology laboratories. Clinical laboratories routinely identify patho- gens that may have been foodborne by testing clin- ical specimens, such as blood or stool samples, from ill persons. Regular reporting of the isolation of specific pathogens provides an important source of surveillance data. Of course, such laboratory- based surveillance requires an infrastructure of competent laboratories that provide routine diag- nostic services. It also often requires a central refer- ence laboratory that can confirm unusual isolates and provide quality assurance; this is often part of the public health laboratory system. If this infra- structure is present, laboratory-based surveillance is neither difficult nor particularly expensive. Ac- tive follow-up of cases identified through labora- tory diagnostics provides additional epidemiologi- cal data, including information on possible sources of infection and whether the cases are sporadic or are associated with other cases in an outbreak. Without this follow-up, the use of these laboratory data is more limited, but still useful for programme planning, designing food safety educational mate- rials, and evaluating control programmes. Microbiological monitoring of food animals and foods Monitoring food-producing animals on the farm for microorganisms that may cause foodborne ill- ness in humans can provide producers and food control authorities scientifically sound and statisti- cally valid information on the occurrence, distribu- tion, and trends of these agents. Programmes that collect on-farm data for agricultural and economic reasons can generate information on the fre- quency of animal infections with selected food- borne pathogens in addition to descriptive sta- tistics on animal health, productivity, and manage- ment. Microbiological surveys of foods can help de- fine the risk for exposure to potential pathogens, and the impact of control measures designed to 26 reduce contamination of food pathogens. Because microorganisms may die or may multiply between the time of sample collection and food consump- tion, these survey data are only an indirect measure of exposure potential. In the absence of epidemio- logical data on human illness, food microbiology data collected for international trade or other pur- chase specifications may be the only available indi- cation of the risk for foodborne disease. Investigation of foodborne disease outbreaks One of the fundamentals of epidemiology is that the study of the departure of the observed patterns of occurrence of disease from the expected may lead to identification of the causes of the devia- tion (5). When the observed rate is higher than expected and cases are unusually close together in time or space or within the same demographic group, that group of cases is called an outbreak. A foodborne outbreak is generally defined as an inci- dent in which 2 or more persons experience a similar illness after ingestion of a common food or meal. A sporadic case is an ill person whose illness is not known to be related by common exposure to other ill persons. Although sporadic infections are far more common, outbreaks are easier to investi- gate and characterize than sporadic illnesses, and large dusters of cases provide an opportunity to rapidly identify a common source of exposure by epidemiological methods. The most informative outbreak investigations start with the collection of relevant exposure data from a defined group of ill persons, and a defined group of well persons who have had a similar op- portunity to become ill, such as well classmates of ill schoolchildren, persons attending the same wedding reception, or patrons of the same restau- rant. Once statistical analysis of exposure data im- plicates a food source, further investigation into its mode of preparation and the sources of the raw ingredients may be warranted to identify how it became contaminated. Intensive laboratory investi- gation of leftover food, raw ingredients, and the source farms may identify the causative agent and clarify the mode of contamination. An epidemio- logical investigation by Kapperud et al. (6) showed that contaminated iceberg lettuce imported from Spain caused a large outbreak of Shigella sonnei infections in Norway; concurrent investigations identified related cases in other European coun- tries as well. These outbreak investigations were important in preventing additional cases and sug- gesting a possible original source of product con- tamination during production. As Kapperud et al. showed, strong epidemiolog- ical information combined with selective microbio- logical investigation of implicated items is most likely to yield relevant information. Broad micro- biological sampling of foods and food workers rarely identifies the correct cause of the outbreak, Rapp. trimest. statist. sanit. mond., 5D (1997) and is not a substitute for an epidemiological case- control study. The actual source of contamination may no longer be available at the time of investiga- tion, and many irrelevant items may have been cross-contaminated in a kitchen. Similarly, food workers are often infected as a result of eating the contaminated food that caused the outbreak, and their stool cultures rarely provide useful informa- tion. Because microbiological results are often de- layed or simply unobtainable (for example, if the foods of interest are no longer available or if micro- biological assays do not exist), emergenc;y correc- tive actions to halt the outbreak must sometimes be based on epidemiological data alone. Surveillance of foodborne disease outbreaks Given the presence of a functioning public health infrastructure that can detect and investigate food- borne outbreaks, these events can be reported and analysed as a surveillance system. Such outbreak surveillance is often conducted at the provincial and national levels by collecting summary reports on local investigations. The value of such data de- pend heavily on the quality of the original investi- gations, including the use of standard methods for implicating etiological agents and food vehicles, and the constancy of reporting. Such surveillance data are occasionally the source of new discoveries, such as the linkage of outbreaks of SalmoneUa sero- type Enteritidis infections with eggs (7). More typi- cally, well collected data provide useful informa- tion on the spectrum of outbreaks associated with a particular etiological agent, a particular type of food, or a particular setting (8-10). For example, a review of foodborne outbreaks in nursing homes identified Salmonella enteritidis from eggs as recur- rent sources, as well as specific points of interven- tion that later were used to develop a training programme for nursing home staff (11). A recent review of outbreaks aboard cruise ships docking in United States ports identified specific policies that could have prevented nearly half of such out- breaks (12). Routinely collecting information on foodborne disease outbreaks from medical care providers and diagnostic laboratories has several advantages. It does not require costly systems of primary diagno- sis and data generation, and can provide nation- wide or regional statistics and identify many of the common causes of acute epidemic foodborne dis- ease. However, foodborne disease outbreak surveil- lance depends on the existence of a reason- ably well-developed public health infrastructure capable of detecting and appropriately investigat- ing outbreaks, and a mechanism for reporting and summarizing the results of those investigations. As currently supported in most countries, food- borne disease outbreak surveillance tends to be insensitive, especially for conditions without good diagnostic tests and for diseases with incubation Wld hlth statist. quart., 50 (1997) periods of longer than one week. The data ob- tained are frequently incomplete and of poor qual- ity, and the surveillance fails to completely charac- terize the epidemiological features of the out- breaks it describes. Nonetheless, surveillance systems for food- borne disease outbreaks that have been in place for an extended period provide a valid impression of outbreak trends in the geographic area under sur- veillance, and may be adequate for local public health purposes. They also identify dominant hu- man behaviours associated with illness that can be tracked over time to judge the effectiveness of pub- lic health efforts to modify high-risk behaviours. Epidemiological studies of laboratory-confirmed sporadic foodborne disease Sporadic cases of foodborne infection are far more frequent than cases that are part of identified out- breaks. The prevention lessons learned from out- break investigations may not always apply to spo- radic cases. Specific epidemiological studies of spo- radic cases, comparing their food exposures pre- ceding illness to those of healthy neighbours, may be needed to clarify the sources of sporadic food- borne illness. For example, in the United States, outbreaks of Campywbacter jejuni infections are most typically caused by consuming raw unpasteur- ized milk or untreated surface water, and occur in the spring and fall. In contrast, most sporadic cases of Campylobacter infection occur in the summer, and appear to be related to touching or consuming undercooked poultry (2). A control strategy based on outbreaks alone would focus attention on the bovine reservoir and the need for pasteurization. While this would prevent outbreaks, it would do little to reduce the much larger number of spor- adic cases related to poultry. During the past decade, investigators have learned that Listeria monocytogenes can be found in a wide variety of processed foods ( 13). Growth of the organism during prolonged refrigerated stor- age of food products presents a considerable risk to susceptible consumers, even though the disease caused by the microorganism is uncommon in the population as a whole. In the United States, case- control studies of sporadic listeriosis identified soft cheeses, processed foods purchased from store delicatessen counters, undercooked chicken and non-reheated hot dogs as risk factors for in- fection. This resulted in an educational campaign on specific dietary recommendations for consum- ers at high risk, and industry and regulatory measures to prevent contamination of ready-to-eat foods. The most recent epidemiological studies of sporadic listeriosis in the United States suggest that the combination of risk management strate- gies has resulted in a 44% reduction in illness and a 48% reduction in deaths associated with listeri- osis (14). 27 Applications for global food safety Epidemiological data linking infections with food sources, information from microbiological surveys of foods, and data from controlled studies together provide the basis for local and provincial food safety control, Hazard Analysis Critical Control Point (HACCP) development, and risk assessment. In developing countries where foodborne disease surveillance is particularly weak, the attention of policy makers and the food industry is frequently not directed toward the role of food in transmis- sion of disease, and important opportunities for control are missed. In these countries, the system- atic investigation of sporadic cases of cholera or typhoid fever with case-control studies may identify important sources of foodborne infection, such as street vended foods and beverages or raw seafood, that are amenable to control and prevention (15, 16). In more developed countries, increasing international trade in foods can lead to large inter- national outbreaks ( 17). When such outbreaks occur, it is imperative that the importing and ex- porting countries work together to conduct an ade- quate epidemiological investigation. Identification of a food hazard associated with an imported food can, in fact, reveal an unrecognized food hazard present in the country of origin. For example, in 1991, investigation of cholera cases associated with a frozen coconut milk product imported from Thailand led to the identification of an unlicensed food processor unknown to Thai authorities (18). Risk assessment begins with showing that an infection can be foodborne, and investigation of foodborne disease outbreaks is a useful means of identifying hazards in foods ( 19). In addition, out- break investigations can provide information on rates of hospitalization and death, economic costs of illness, and other indicators of severity of dis- ease. Prospective studies of sporadic foodborne diseases provide additional data on infection rates in the populations studied, which can help rank foodborne pathogens; the proportion of illness at- tributable to specific foods, which can help rank food vehicles; and other information useful for characterizing risk. Data needs for dose-response assessment include information on the variability in host susceptibility, attack rates, and pathogeni- city, and the effects of the characteristics of the food matrix and competing microflora. Epidemio- logical studies of foodborne disease outbreaks and sporadic illnesses can help characterize who be- comes ill and under which conditions, and can, therefore, sometimes provide valuable clues to dose response. Conclusions Nations at different stages of economic develop- ment have very different public health needs for epidemiological and surveillance data, and great differences in available resources to collect and 28 interpret those data. Many of the countries that are least able to conduct sophisticated epidemiological studies earn their foreign currency through their food exports to more developed trading partners that may demand extensive quantitative risk data on which to establish safety standards for interna- tional trade in food. National economies at the beginnings of their sanitary awakening need first to develop their public health foundations, and the epidemiological and surveillance programmes jus- tifiable for public health purposes need only pro- vide gross estimates of disease burden and efficacy of basic control measures. This can be accom- plished by establishing a system of vital records, laboratory-based surveillance for Salmonella typhi, and clinic-based surveillance for dysentery and de- hydrating watery diarrhoea. Middle-class economies can support more elab- orate systems of epidemiology and surveillance of foodborne diseases. Data are needed to stop out- breaks, and to establish sanitary standards for food production, processing, and preparation. In gen- eral, qualitative data on foodborne diseases are adequate to set priorities, and to determine the major causes of disease and prevent their reoccur- rence. The richest nations can establish sophisticated systems of surveillance and investigation that will provide quantitative risk data for foodborne dis- ease. These data will create the knowledge neces- sary to identify emerging causes of foodborne dis- ease, populations at greatest risk of illness and of severe outcomes, and diffuse outbreaks caused by intermittent low-level contamination of widely dis- tributed foods, and will permit the wisest allocation of food safety resources nationally and internation- ally. Summary Food safety is a complex matter that depends on a number of interrelated environmental, cultural, and so- cioeconomic factors. The purpose of epidemiology and surveillance is to define these factors, how they interact, and their relative importance in foodborne infections. The tools epidemiologists use to study foodborne dis- ease include surveillance of specific infections in hu- mans, monitoring of contamination with specific patho- gens in foods and animals, intensive outbreak investiga- tions, collecting reports of outbreaks at the regional or national level, and studies of sporadic infections. With sufficiently elaborate systems of surveillance and inves- tigation, it is possible to provide quantitative risk data for foodborne diseases that will permit the wisest allocation of food safety resources. Rapp. trimest. statist. sanit. mond., SO (1997) Resume Epidemiologie des maladies d'origine a/imentaire : instruments et application La salubrite des aliments est un probleme complexe qui depend de plusieurs facteurs environnementaux, cultu- rels et socio-economiques lies entre eux. L'epidemiolo- gie et la surveillance ant pour but de definir ces facteurs et d'en etudier !'interaction et !'importance relative dans les infections transmises par les aliments. Les instru- ments qu'utilisent les epidemiologistes pour etudier les maladies d'origine alimentaire sont: la surveillance de certaines infections chez l'homme, la surveillance de la contamination par des agents pathogenes determines des denrees alimentaires et des animaux, les enqu~tes approfondies sur les flambees epidemiques, la collecte de rapports sur les flambees epidemiques au niveau national au regional, et les eludes d'infections sporadi- ques. II est possible, au moyen de systemes de sur- veillance et d'investigation suffisamment elabores, de fournir des donnees quantifiees concernant le risque de maladies d'origine alimentaire, qui permettront d'affec- ter le plus judicieusement possible les ressources en matiere de salubrite des aliments. References - References 1. Kelsey, J.L et al. Methods in observational epidemiology. New York, Oxford University Press, 1986. 2. Tauxe, R. V. Epidemiology of Camp,ylobacter jejuni infections in the United States and other industrialized nations. In Nachamkin, Blaser, and Tompkins, eds., Camp,ylobacterjejuni: Current status and future trends, pp. 9-19, 1992. 3. Griffin, P.M. &: R. V. Tauxe. The epidemiology of infections caused by Escherichia coli 0157:H7, other enterohemor- rhagic E. coli, and the associated hemolytic uremic syndrome. Epidemiological reuiew; 13: 60-98 (1991). 4. Tauxe, R.V. &: M.L. Cohen. Epidemiology of diarrheal diseases in developed countries. Chapter 3. In: Blaser, M.J. et al. (eds.) Infections of the gastrointestinal tract. Raven Press, New York, 1995. Wld hlth statist. quart., SO (1997) 5. Teutsch, S. M. &: R. E. Churchill. Principl,es and practice of public health surueillance. New York, Oxford University Press, 1994. 6. Kapperud, G., L. et al. Outbreak of Shigella sonnei infection traced to imported iceberg lettuce. Journal of clinical microbiology; 33: 609-614 ( 1995). 7. St. Louis, M.E. et al. The emergence of Grade A eggs as a major source of Salmonella enteritidis infections: Implications for the control of salmonellosis. journal of the American Medical Association, 259: 2103-2107 ( 1988). 8. Holmberg, S.D. &: P.A. Blake. Staphylococcal food poi- soning in the United States - new facts and old misconcep- tions. Journal of the American Medical Association, 25 l: 487-489 (1984). 9. Potter, M.E. et al. Unpasteurized milk - the hazards of a health fetish. Journal of the American Medical Association, 252: 2048-2052 (1984). 10. Hedberg, C.W. et al. A multistate outbreak of Salmonella javiana and Salmonella oranienburg infections due to consumption of contaminated cheese.Journal of the American Medical Association, 268: 3203-3207 ( 1992). 11. Levine, W.C. et al. Foodborne disease outbreaks in nursing homes, 1975-1987.JournaloftheAmericanMedicalAssociation, 266: 2105-2109 (1992). 12. Koo, D. et al. Epidemiology of diarrheal disease outbreaks on cruise ships, 1986 through 1993. journal of the American Medical Association, 275: 545-547 (1996). 13. Schuchat,A. etal. Epidemiology of human listeriosis. Clinical microbiology reuieur, 4: 169-183 ( 1991). 14. Tappero,J.W. et al. Reduction in the incidence of human listeriosis in the United States. journal of the A merical Medical Association, 273: 1118-1122 (1995). 15. Tauxe, R.V. et al. Epidemic cholera in the New World. Translating field epidemiology into new prevention strategies. Emerging infectious diseases, l: 141-146 ( 1995). 16. Black, R.E. etal. Case-control study to identify risk factors for paediatric endemic typhoid fever in Santiago, Chile. Bull,etin of the Worl.d Health Organization, 63: 899-904 ( 1985). 17. Killalea, D. et al. International epidemiological and micro- biological study of outbreak of Salmonella agona infection from a ready to eat savoury snack-I: England and Wales and the United States. British journal of medicine, 313: 1105-1107 (1996). 18. Taylor, J.L. et al. An outbreak of cholera in Maryland associated with imported commercial frozen fresh coconut milk.Journal of infectious diseases, 167: 1330-1335 (1993). 19. Potter, M.E. The role of epidemiology in risk assessment: a CDC perspective. Dairy food environmental sanitation, 14: 738-41 (1994). 29 Epidemiology of toodborne diseases: a worldwide review Ewen G.D. Todda Introduction Every society has some interest in food safety and disease control. After the Second World War, gov- ernments had shifted away from addressing public health problems with communicable diseases to long-term effects of chemicals in the food supply, because it was thought that infectious diseases were under control. In the last decade, however, there has been a change in food safety priorities from chemical issues back to microbial hazards, at least in North America. This is due to the recognition of new life-threatening agents, the occurrence of con- tinuing well-publicized outbreaks transmitted through food and water, and the apparent inability of public health authorities to prevent them. The current concerns are raw foods of animal origin and fresh produce contaminated by Escherichia coli 0157:H7, Listeria monocytogenes and Salmonella, along with enteric viruses in shellfish and parasites such as Toxoplasma, Cryptosporidium and Cydospora in meat and fruit. Also, traditional pathogens have been recognized as causing illnesses from new foods, such as Salmonella in melons and Clostridium botulinum in garlic bottled in oil. However, al- though reporting is improving, it is still very lim- ited. The number of cases of enteric illnesses based on laboratory isolations of pathogens from stool specimens has been underestimated by a factor of 20 to 100 (1). In addition, for most of these no source of infection is identified. Despite these limi- tations, some countries have been collecting and reporting data for years and increasing numbers of others are starting the process. However, food- borne disease cases are significantly underesti- mated, and relatively few well-investigated out- breaks are described although these usually give information on the etiological agent and the rea- son for the contamination. Therefore, no national system has yet determined accurately the extent of foodborne diseases, and data from different coun- tries may show distinct differences which may re- flect as much the way that these are collected and interpreted as the geographic location and food habits of the people. Thus, caution must be advised in making too critical a comparison of foodborne disease statistics from nation to nation. This in- a Bureau of Microbial Hazards, Health Protection Branch, Health Canada, Sir Ban ting Research Centre, Ottawa, Ontario, Canada. 30 creasing commitment to the documentation and control of foodborne disease indicates an awaken- ing awareness to its significance in terms of mor- bidity, mortality, economic loss, and effects on trade between countries and common markets. The following discussion on worldwide surveil- lance of foodborne disease updates previous re- views (2, 3). Objectives of a foodborne disease surveillance system A well-constructed food safety programme for the protection of the population comprises investiga- tion of foodborne disease incidents, laboratory analysis of foods and clinical specimens for patho- gens and contaminants, regular inspection of foods and their production chain, appropriate leg- islation for the prevention of foodborne disease, and education of those working at the different sectors of the food industry and the consumer. Surveillance is necessary to identify foodborne dis- eases, their causes, and their socioeconomic im- pact, in order to help determine the best short- and long-term control measures. It should include the systematic collection of relevant incident data, i~ evaluation for completeness and accuracy, compi- lation into standard formats, and interpretation of trends with examples from specific outbreaks. There are at least 4 components to a well-designed surveillance system: (i) early alert of disease; (ii) laboratory isolation data; (iii) reports of food- borne illness incidents; and (iv) use of special epi- demiological studies and surveys, including senti- nel site studies, to determine a more realistic level of morbidity. Surveillance principles are covered more fully by Borgdorffand Motarjemi (4). Foodborne disease surveillance is carried out at the local level and reports are actively or passively transferred to a national centre, or at least to a central coordinating body, for standardizing the material and preparing annual or other types of reports. Reports include laboratory isolation data and foodborne disease outbreaks. Specific epide- miological studies may not be coordinated at the national level, but appear in the scientific litera- ture for review. Some countries have a federated system of states, provinces or cantons that take different responsibilities for health care and sur- veillance. These can enhance a national system but another layer of bureaucracy tends to slow down the reporting. A full discussion on foodborne dis- Rapp. trimest. statist. sanit. mond., 50 (1997) ease surveillance is given in Bryan et al. (5, 6), Guzewich et al. (7) and Todd et al. (8). Selected regions of the world are reviewed for their contri- bution to our knowledge of foodborne disease. Europe Summary of foodborne disease surveillance programme in Europe Since 1980, many European and a few other coun- tries have participated in an early warning and routine reporting system, coordinated through the FAQ/WHO Collaborating Centre for Research and Training in Food Hygiene in Berlin (9); by 1996, 46 countries in the WHO European Region had participated in this programme with varying degrees of commitment ( JO). These reports stan- dardized as much as possible so that data can be compared, but because of the different degrees of commitment to reporting in these countries, direct comparisons of data between countries are ques- tionable. However, some general statements can be made. Where foodborne diseases are notifiable, the incidence rates are increasing for Austria and England/Wales, relatively stable for Spain and Scotland and decreasing for Poland. For salmonel- losis, generally rates grew until 1992 and decreased slighly in 1993, perhaps indicating the start of more effective control measures against S. enteri- tidis. In contrast, campylobacteriosis cases appear to be continuing to rise well beyond rates for sal- monellosis. Travel abroad to resorts on the Medi- terranean Sea and to Asian and African countries was cited as a source for some of these infections. Twenty-one countries specifically reported foodborne outbreaks ranging from 5 (Albania) to 2 818 (Spain) from 1990 to 1991/92. In 19.5% of these outbreaks the agent was unknown or unspec- ified. Where the agents were identified, Salmonella spp. caused 84.5% of all outbreaks (S. enteritidis 50.9%), Staphylococcus aureus 3.5%, C. perfringens 3.0%, C. botulinum 1.1 % and Bacillus cereus 1.0%. All other causative agents were <l % except for Trichinella (1.5%) and mushrooms (1.3%). The etiology of the outbreaks in 16 of these countries is shown on Tab/,e 1. Most campylobacteriosis illnesses are sporadic cases of unknown origin, and relative- ly few outbreaks were documented (mostly from the United Kingdom). Only 12 E. coli 0157:H7 outbreaks were reported and these were all from England and Wales, although cases and outbreaks are known to have occurred in other countries (11). However, unspecified E. coli outbreaks were recorded in Germany, the Netherlands, Romania, Spain and Scotland, and some of these were prob- ably Verotoxin-producing strains (VTEC) in ori- gin. Trichinosis outbreaks were documented in Lithuania (146), Bulgaria (18), Spain (16) and Wld hlth statist. quart., 50 (1997) France (1), and many cases in Romania (4 705) and Poland (738). Outbreaks of mushroom poi- sonings were reported in Hungary ( 158) and Israel ( 1). The most frequent foods associated with out- breaks, where these were identified, were eggs and egg products (25.4%), meat and meat products (23.4%), confectionery, sweets, cakes, pastry, pud- dings, desserts and ice cream ( 17.0%). Fish and shellfish, milk and dairy products, and poultry and poultry products were each implicated in less than 5% of outbreaks. There have been limited attempts by countries in this report to associate agents with foods, but there is evidence that Salmonella enteriti- dis infections have arisen from the consumption of raw or undercooked eggs in a variety of products, and poultry. E. coli 0157 infections have been linked to beef, milk and yoghurt. A few Campy- lobacter outbreaks were associated with under- cooked chicken and meat, unpasteurized milk, and bottled milk pecked by birds. Where the place of contamination or mishandling was clearly iden- tifiable, the farm was where the pathogens were most frequently identified as entering the food chain (50%), with subsequent mishandling mainly at restaurants (22%), homes (13%) and catering establishments (7%). Outbreaks occurred from people eating in homes ( 46%), restaurants/hotels (15%), catered events (8%), medical care facilities and canteens (6% each), and schools (5%). Fac- tors contributing to outbreaks were identified in 7 countries in 55% of outbreaks. Temperature abuse, particularly inadequate cooling and im- proper cooking, was responsible in 44% of these, contaminated or toxic raw products in 16%, con- tamination by personnel or equipment in 15%, lack of hygiene in processing, preparing and hand- ling in 10%, and cross-contamination in 4%. The above data are more detailed than in previous reports and are indicative of an increasing com- mitment to the documentation of foodborne dis- ease in the contributing countries which may even- tually lead to control. A few examples at the national level will be given for annual numbers of laboratory-confirmed and notifiable cases, outbreaks of foodborne dis- ease, and data on specific outbreaks. Specific epidemiological studies In Denmark, there was a very gradual decrease in cases of campylobacteriosis and yersiniosis, but for salmonellosis there were two distinct peaks in 1989 and 1992 (Tab/,e 2). The increase in 1992 was mainly due to S. enteritidis and S. typhimurium (a phage type found in pigs). The notifiable food- borne and waterborne disease cases increased steadily between 1984 (192) and 1992 (1 189) with a slight peak in 1988 and a profile similar to that for salmonellosis, indicating that many of the foodborne diseases may have been caused by Salmonella. 31 Table 1 Foodborne disease in 16 countries by percentage of outbreaks of known etiology, WHO Region for Europe Tableau 1 Maladies d'origine alimentaire dans 16 pays, en pourcentage des flambees d'etiologie connue, Region europeenne de l'OMS Etiology - Etiologie Bacillus cereus Campylobacter Clostridium botulinum Clostridium perfringens Albania- Albanie 1990-91 Salmonella 100 Shigella Staphylococcus aureus Other bacterial b - Autres bacteries b Total bacterial {%) - Total des etiologles bacterlennes (%) 100 Viral - Virus Seafood toxins - Toxines de poissons et fruits de mer Plant - Vegetaux Chemical - Produits chimiques Parasitic - Parasites Other - Divers Total number of outbreaks - Nombre total de flambees Etiology - Etiologie Bacillus cereus Campylobacter Clostridium botulinum Clostridium perfringens Salmonella Shigella Staphylococcus aureus Other bacterialb - Autres bacteriesb Total bacterial {%) - Total des etiologles bacteriennes {%) Viral - Virus Seafood toxins - Toxines de poissons et fruits de mer Plant - Vegetaux Chemical - Produits chimiques Parasitic - Parasites Other - Divers Total number of outbreaks - Nombre total de flambees 3 Bulgaria - Denmark- Finland - Bulgarie Danemark Finlande 1990-92 1990-92 1990-92 66 6 73 27 6 6 12 38 14 76 4 4 7 6 13 2 33 29 16 93 7 France - Germany- Hungary- 1990-92 Allemagne Hongrie 5 85 1 7 98 <1 2 <1 <1 <1 1990-92 1990-92 3 2 88 <1 2 97 3 3 7 3 73 <1 9 2 76 24 lceland- lslande 1985-93 26 4 19 33 19 100 Israel- Israel 1990-92 18 21 7 18 16 80 5 2 12 2 67 69 55 1 666 319 648 27 56 Netherlands - Pays-Bas 1990-91 25 17 17 10 27 96 4 59 Portugal- 1990-91 7 34 59 100 29 Romania- Roumanie 1990-92 3 57 2 24 13 100 125 Spain- Espagne 1990-92 <1 1 2 84 <1 6 2 96 2 1723 Sweden- Suede 1990-92 3 3 17 51 5 12 7 98 2 60 England and Wales- Angleterre et Pays de Galles 1992-93 2 2 17 68 <1 2 2 93 6 403 Malta- Malte 1990-93 4 81 11 96 4 27 Scotland- Ecosse 1990-92 <1 11 1 85 99 486 • Source: 6th Report of the WHO Surveillance Programme (Ref. 10). -& Rapport du Programme OMS de Surveillance (Ref. 10). b "Other bacterial" refers to E. cot;, in the case of England & Wales and Scotland it is specified whether VTEC or 0157:H?. - •Autres bacteries» designe E. coti; pour l'Angleterre et le pays de Galles et pour l'Ecosse. ii est precise qu'il s'agit de VTEC ou de 0157:MH?. 32 Rapp. trimest. statist. sanit. mond., 50 (1997) Table 2 Cases of enteric disease in Denmark, 1988-19923 Tableau 2 Cas de maladies intestinales au Danemark, 1988-19923 Disease - Maladie 1988 1989 1990 1991 1992 Mean no./100 OOO in 1992 - N. moyen/100 OOO en 1992 Campylobacteriosisb - Campylobacterioseb 1 276 1 432 1 367 1 261 1129 22 Salmonellosisb - Salmonelloseb 3 200 3 500 2 081 2 203 3 373 65 Yersiniosisb - Yersinioseb 865 879 817 929 909 18 Foodborne and waterborne diseasesc- Maladies d'origine alimentaire et hydriquec 1123 926 898 954 1189 23 • Source: 6th Report of the WHO Surveillance Programme (Ret 10). -6• Rapport du Programme OMS de Surveillance (Ref. 10). b Laboratory-confirmed cases. - Cas confirmes au laboratoire. c Notifiable cases. - Cas a declaration obligatoire. In Germany, there are fewer Campflobacterjejuni cases reported than for Salmonella, unlike other European countries, but campylobacteriosis is con- sidered an important disease ( 10, 12 ). A survey showed that 10.9% of 485 retail food samples sur- veyed contained the organism - turkey liver 66%; poultry 50%; milk/milk products 2.9%; fish 2.3%; meat products 0.8%; shellfish and sausage 0% (12). Deep frozen, raw products were more fre- quently contaminated than fresh, raw products (23.8% vs. 12.4%). In Switzerland, a case-control study showed that most infections were caused by Salmonella en- teritidis PT 4 (13), and associated with the con- sumption of raw or undercooked eggs, especially in desserts made with raw eggs. Infections from other serovars were more likely to be contracted in other countries and were not associated with eggs. VTEC seem to be a major cause of haemolytic uraemic syndrome (HUS). Children with this dis- ease in the Czech Republic had evidence of VTEC infection (verotoxin in stools, antibodies to E. coli LPS in sera, or isolation of VTEC strains) (14). Serotypes 026:Hll and 0157 were the most fre- quently isolated. However, no source of the infec- tions was identified. In the Netherlands, Belgium and Germany, 78% of HUS patients had evidence ofVTEC infections (15). E. coli0157 was the most frequent causative agent and Vf-2 the most fre- quent toxin produced. Where E. coli 0157 was isolated in family members, the strains were identi- cal by subtyping analysis. In Scotland after 13 years of experience with O 157 infections, the epidemiol- ogy is becoming better understood (16). The out- breaks have been associated with ground meat, unpasteurized milk, unpasteurized milk farm cheese, contaminated pasteurized milk, vegetables fertilized with cow manure, drinking and recre- ational water supplies, and person-to-person spread. Livestock, particularly cattle, are consid- ered to be major reservoirs of this group of organ- isms. Wld hlth statist. quart., 50 (1997) One of the reasons for the establishment of the international Salm-Net programme was to identify trends in salmonellosis in Western Europe. Be- tween 1993 and 1995 there was a fall of6.8% in all isolates in 7 contributing countries, with S. enteriti- dis and S. typhimurium representing about 75% of these. However, the trend for these two serovars was different; S. enteritidis decreased each quarter by a total of 16.4%, and S. typhimurium rose each quarter during the same 3 years by 15.2% ( 17). The decrease in S. enteritidis may represent efforts to limit infection in flocks and use unheated eggs less frequently in food items. It is also noted that many of the S. typhimurium strains are multi drug resistant and this leads to an emerging problem in enteric infections. Salm-Net also aims to type strains beyond their serological designation, and is re- questing phage typing for the most common sero- types. Foodborne disease annual reports In France, the number of outbreaks rose from 594 in 1990 to 732 in 1992. Where the agent was identi- fied, Salmonella was responsible for 83-87% of out- breaks. Eggs and egg products were associated with many outbreaks, particularly after they were con- taminated with S. enteritidis (Tabl,e 3). Meat and meat products, and also mixed foods, were the important vehicles for causing Salmonella, C. perfrin- gens and S. aureus outbreaks. Most fish- and shell- fish-borne outbreaks were caused by histamine and diarrhoeic shellfish poison, respectively. Out- breaks involving dairy products, such as cheese and ice cream, and frequently-handled prepared foods most likely arose from S. aureus intoxication. Out- breaks typically occurred after people had eaten at homes, schools, restaurants, canteens, hospitals and homes for the aged, holiday resorts, prisons and religious gatherings. The contributing factors most commonly documented were contaminated equipment, faults in processing, inadequate cool- ing, contaminated raw ingredients, preparation 33 Table 3 Foodborne disease outbreaks by known etiology and associated foods, France, 1990 -1992a Tableau 3 Flambees de maladies d'origine alimentaire selon l'etiologie et !'aliment associe, France, 1990-1992a Year- Etiology MeaVpoultry- Egg/egg products - Fish/shelttish - Milk/dairy products - Other- Total An nee Etiologie- Viande/volaille CEufs/produtts Poissons/ LaiVproduits laitiers Divers ~ base d'reufs fruits de mer 1990 Salmonella 19 (15%) 91 (71%) 6 (5%) 3 (2%) 9 (7%) 128 (100%) C. perfringens 16 (67%) 8 (33%) 24 (100%) S. aureus 4 (16%) 3 (12%) 1 (4%) 9 (36%) 8 (32%) 25 (100%) Other - Divers 2 (9%) 19 (86%) 1 (5%) 22 (100%) 1991 Salmonella 22 (12%) 113 (60%) 5 (2%) 7 (4%) 41 (22%) 188 (100%) C. perfringens 15 (68%) 7 (32%) 22 (100%) S. aureus 6 (16%) 2 (5%) 1 (3%) 12 (32%) 16 (43%) 37(100%) Other - Divers 1 (4%) 1 (4%) 8 (35%) 13 (57%) 23(100%) 1992 Salmonella 13 (7%) 118 (65%) 10 (6%) 39 (22%) 180 (100%) C. perfringens 6 (30%) 14 (70%) 20 (100%) S. aureus 5 (14%) 5 (14%) 2 (6%) 11 (32%) 12 (34%) 35 (100%) Other - Divers 1 (8%) 1 (8%) 8 (61%) 3 (23%) 13 (100%) • Source: 6th Report of the WHO Surveillance Programme (Ref. 10). -6• Rapport du Programme OMS de Surveillance (Ref. 10). too far in advance of eating and contamination through personnel. Foodborne disease outbreaks Powdered infant formula was responsible for 48 known cases of salmonellosis in infants under 7 months old from 14 regions in Spain fromJanu- ary to June, 1994 (18). The implicated strain was a lactose-fermenting S. virchow. It took some time for the illnesses to be linked with the Salmonella in the milk powder and recalls to be made. Interestingly, in 1993 a lactose-fermenting S. tennessee in pow- dered milk products caused infant illnesses in the United States and Canada. Five persons who were suffering from S. enteritidis PT 4 infection in England had attended the same gym and had con- sumed a protein-based beverage composed of milk powder and a raw egg as a body-building drink ( 19). This is an example of "voluntary" salmonello- sis when lifestyle appears to outweigh risks of en- teric illness. In Germany, about 1 OOO cases of salmonellosis were associated with consumption of paprika and paprika-powdered potato chips (20). The infective dose was 4 - 45 organisms with an attack rate of 1 in 10 OOO exposed persons; several serovars were im- plicated. This is the largest documented outbreak from contaminated spices. A scombroid outbreak with 15 cases also oc- curred in Spain from a fresh tuna sold at a super- market in June 1994 (21). Typical symptoms of facial flushing, headache, diarrhoea, nausea and abdominal pain followed about 45 minutes after a peppery taste was noticed in the tuna. Histamine up to 580 ppm was found in leftover tuna samples, and >35 µg/L in 4 urine specimens from the cases. 34 Although botulism from dairy products is con- sidered rare, some recent outbreaks show that this does occur. In June 1989, the largest outbreak in the United Kingdom took place when commercial- ly prepared hazelnut yogurt was consumed after contamination of the nut paste. In Italy, in August and September, 1996, at least 8 persons and one death resulted after tiramisu made with mascar- pone cheese was eaten (22). Most of those ill were children. The cheese contained spores of C. botuli- num but it is uncertain how the contamination took place. The widely exported product was re- called on a massive scale. In Scotland, there was a large meatborne E. coli 0157 outbreak with 396 cases and 11 deaths be- tween November 22 and December 11, 1996. Cases had eaten cold cooked meats or meat sandwiches from a butcher or had eaten cooked steak in gravy at a church lunch supplied by the same butcher. This is the second largest foodborne outbreak in Scotland (23), and the latest in a series of out- breaks; it prompted an enquiry into the reason for the dramatic increase in foodborne disease cases in that country (24). In 1970 there were 858 cases. By 1980 this figure was 2 689; in 1990 and 1995 the numbers were 6 510 and 9 297, respectively, and for 1996, a predicted number of 10 OOO. The Pen- nington Report recommended research into prev- alence in livestock of E. coli 0157 and more sensi- tive typing procedures for isolates (24). There should be a more systematic approach to surveil- lance including electronic reporting and analysis of data. Also, there should be uniform standards of hygiene throughout the food industry. When out- breaks occur a team empowered to take action should quickly be set up. Rapp. trimest. statist. sanit. mond., 50 (1997) Africa Epidemiological studies and foodborne disease data Little in the way of regular foodbome disease sur- veillance is done in Africa, although there is an awareness of the importance of diarrhoeal dis- eases, and some studies have been carried out. Because food and water are thought to be the main means of transmitting these types of diseases, a survey of hygiene in houses where diarrhoea had occurred was carried out in Liberia (25). Between 40 and 88% of stored water samples contained >103 enterobacteria cfu/100 ml, and 19-32% of adult food samples had >105 enterobacteria cfu/ 100 g. Infant food such as formula, baby cereal and traditional "rice water" were even more contami- nated (56% with >105 CFU/100 g) because they had been stored at room temperature for up to 8 hours. In an urban slum community it was a common practice to prepare food in advance of a meal and store it at ambient temperatures for up to 24 hours. It was only occasionally reheated, partly because of lack of fuel. Also, women had only time to make food once per day because they worked far from their homes. Unfortunately, no foods were examined for pathogens. These observations indi- cate that foodborne and waterborne diseases can be easily spread in such communities and cause a high morbidity. However, a case-control study in Liberia showed that diarrhoea in households was less related to poor food hygiene practices and more to improper disposal of faeces (26). Fortu- nately, fermented foods, such as those traditionally made in Zimbabwe, are not good vehicles for trans- mitting pathogens since most pathogens died with- in a few hours in sour porridge and mahewu be- cause the pH was <3 (27); unfermented porridge (pH 6) allowed the growth of Salmonella, Shigella, enterotoxigenic E. coli (ETEC), and enteropatho- genic E. coli (EPEC). Ogi, a fermented maize por- ridge used for weaning infants in Nigeria, also has a low enough pH to prevent growth of Salmonella and EPEC (28). Another problem with cereal prod- ucts is contamination by mycotoxins. Moldy grain, which contains aflatoxins, is often consumed un- der drought conditions. In Kenya and the Sudan, aflatoxins were present in the sera of children suf- fering from kwashiorkor, but were metabolized in a different way than in children with other forms of malnutrition or normally nourished children (29). The aflatoxins, therefore, may be a contributory factor to kwashiorkor. Another consequence of mouldy food is hepatocellular carcinoma (HCC); levels of aflatoxin B1 have been found as high as 1.5 mg/kg food. There was a strong association with mutations of the p53 gene in HCC and dietary aflatoxin intake in an international study of pa- tients in 14 countries (30). The incidence of HCC is higher in Mozambique than any other country. Exposure to Toxoplasma gondii is relatively common Wld hlth statist. quart., SD (1997) in Africa with a 18.2 - 61 % prevalence of antibodies in the population of 8 countries ( 31 ). In the Sudan, both Toxoplasma and Brucella infections are prob- ably associated with consumption of raw liver and intestines (32). In Egypt, the foods at highest risk in 1985 and 1986 were white cheese, fermented cream, meat/ chicken, flour/butter oil, cabbage/rice and pota- toes, and many illnesses occurred in homes (33). These food vehicles seemed to be similar to those associated with foodbome disease in Iraq. Between 1983 and 1985 in Egypt, 3 outbreaks from white cheese and 2 from cream/fermented cream were caused by EPEC. Shigella, B. cereus, C. perfringens or C. botulinum, however, were not confirmed as food- borne agents. However, in this country where cere- al products are widely eaten, it is probable that at least some mild illnesses arise from consumption of food containing B. cereus enterotoxin, but are not reported. In addition, street-vended food in Egypt has been shown to contain pathogens and many foods awaiting sale were at temperatures favour- able for their growth (15-44 °C) (34). Foodborne disease outbreaks Occasionally, acute illnesses directly associated with a food are documented, such as the following 3 examples. Because of the need to conserve as much meat as possible for human consumption, animals that die unexpectedly may be butchered and eaten. An example of this occurred in 1976 when 155 villagers in Uganda feasted on a zebu cow which was later found to have died from anthrax septicaemia (35). Within 15-72 hours 143 of these villagers developed acute gastroenteritis and 9 children died. Four to five cattle died from anthrax in 1974 and 1975 in the same county where the outbreak occurred. Wild animals such as hippopotamuses and elephants also succumb to the disease, and carcasses may be utilized as human food. Therefore, the potential for more human anthrax poisonings exists. Jimson weed (Datura stramonium) seeds can contaminate grain during harvesting and cause hallucinations from atropine poisoning when ingested. Twenty-two such episodes occurred in the United Republic of Tanzania between 1974 and 1981 involving wheat and millet products (36). In 1991, the first major botulism outbreak with at least 18 deaths arose from ingestion of locally-made faseikh ( uneviscerat- ed fish) (37). In 1994, 3 children died and 6 others suffered from severe diarrhoea caused by E. coli 0157:H7 after eating hamburgers, koshari and dairy products in Egypt. As a follow-up to this, a survey of 175 foods obtained from slaughter- houses, supermarkets and farmers' homes was conducted for E.coli 0157. This pathogen was de- tected in 6% of unpasteurized milk, 6% of fresh retail beef, 4% of boneless chicken, and 4% oflamb meat samples (38). 35 Cholera outbreaks have continued to occur in Africa since the seventh pandemic began in 1970. These have caused severe morbidity and mortality in refugee camps in Malawi, Somalia, Ethiopia and the Sudan (39). A study of one outbreak in Malawi in 1988 indicated that there was a common source, which could have been food or water, but no envi- ronmental Vibrio isolates were found. Heavy rains destroyed some latrines 15 days before the out- break, probably contaminating the local water table (39). During 1992, in the Lisungwi refugee camp holding 60 OOO Mozambicans, 772 cases of abdominal cramps and bloody diarrhoea were doc- umented (40). The major factor contributing to illness was consumption of cooked food from the market. Based on analysis of stool cultures and the presence of the Vfl gene in some of these, the authors concluded that most of the cases were caused by E. coli 0157:H7 and some by Shigella dysenteriae type 1. The case fatality rate was 4.7%. Refugees are an increasing concern because of civil wars and national conflicts. For instance, 500 OOO refugees returned from East Zaire to Rwanda be- tween November 15 and 21, 1996. The epidemio- logical sentinel stations registered about 14 OOO consultations and 4 7 deaths. Diarrhoeal disease accounted for two-thirds of all consultations. The crude mortality rate was estimated at 0.1/10 000/ day (based on the 500 OOO denominator). The mortality rate in the Mugunga Camp where the refugee population had stayed earlier was esti- mated at 0.3/10 000/day (source: WHO data). How much of this diarrhoeal disease is due to food and water is not known, but they are suspected to be major vehicles of enteric pathogens in such situations. Asia Epidemiological studies and foodborne disease statistics Except for a few countries such as Japan, for in- stance, relatively little in the way of surveillance of foodborne disease is carried out in Asia, and most information is gleaned from specific but limited investigations and studies. From the number of enteric pathogens isolated from stools of patients in a major hospital in Hong Kong (41), it would seem that the distribution of these pathogens in the population is similar to that in Japan, except that the proportion caused by Shigella is higher and that of Vibno is lower (Salmonella 52.5%, Campy- wbacter 16.6%, Shigella 11.3%, Vibrio 5.3% and EPEC 4.4%). Although a link has been made be- tween Campywbacter and Guillain-Barre syndrome (GBS) in the western world (42, 44), this has now also been shown in Kerala State, India (45). From serum samples taken from patients, 26% showed high antibody titres to Campylobacterjejuniand 38% 36 of stool specimens of new GBS cases were positive for C. jejuni/coli. Foodborne campylobacteriosis, therefore, may be very significant from a social and economic point of view globally. Mycotoxins have been responsible for illnesses in 1974 from afla- toxin in maize (15.6 ppm) and in 1987 from deoxynivalenol and other trichothecenes in wheat (46, 48). In India and many other Asian countries, cysticerosis from Taenia solium is a major public health problem because the widespread consump- tion of insufficiently cooked pork (49). A comparative study of foodborne outbreaks in the Republic of Korea and Japan between 1971 and 1990, conducted by Lee et al. (50), (Tabl.es 4a, 4b, 4c) showed that there were considerable differ- ences in the morbidity and mortality (higher for Japan) and agents involved ( Vibrio spp. important in both countries, but more Salmonella than S. au- reus associated outbreaks in the Republic of Korea and the converse for Japan). Most incidents oc- curred in the workplace and the home in the Re- public of Korea, whereas it was more in restaurants and hotels in Japan. Seafood was often implicated in both countries, but food of animal origin was much more frequently associated with outbreaks in the Republic of Korea. The authors did not know how much of these differences were due to report- ing mechanisms and how much to culture. Pan et al. (51) examined foodborne disease in China (province of Taiwan). From 1987 to 1993 the number of outbreaks reported ranged from 57 to 93. In 1994 there were 102 outbreaks and 4 276 cases. Where the etiology was known, the outbreaks totaled 42 V parahaemolyticus, 15 S. au- reus, 11 B. cereus, 6 Salmonella, 1 chemical, and 1 seafood toxin. In particular, the Vibno outbreaks occurred in the warmer months (April to Octo- ber). It would appear from this study and that of Lee et al. (50) that the characteristics of outbreaks in China (province of Taiwan) are more similar to those in Japan than in the Republic of Korea. In Vietnam, it was estimated that 30-57% of students in university hostels between 1984 and 1988 suf- fered from diarrhoea, mainly because food was poorly prepared and stored (52). Street-vended meals in Hanoi in a 1990/91 survey often con- tained E. coli and C. perfringens. Most of the 5 714 documented illnesses between 1983 and 1988 were caused by Salmonella, E. coli and S. aureus, and 156 of these cases died (fatality rate of 2. 7%). Because anthrax-infected cattle are used for meat, hun- dreds of persons develop bactaeremia and 3 to 7 die each year (52). Chemical residues in foods are not adequately controlled and many cases are thought to be due to chemical poisonings. Illegal additives deliberately introduced in alcohol, can- dies, and sweet products have caused intoxications; one such adulterated liquor caused 14 deaths. Foodborne disease was considered to be the most widespread public health problem in Vietnam and Rapp. trimest. statist. sanit. mond., 50 (1997) Table 4a Foodborne disease outbreaks by etiology in the Republic of Korea and Japana Tableau 4a Flambees de maladies d'origine alimentaire, par etiologie, en Republique de Coree et au Japona Country-Pays Morbidity- Deaths/cases - Salmonella S. aureus Clostridium Vibriospp E.coli Other- Morbidite/ Deces/cas spp Divers 100 OOO Rep. of Korea - Rep.de Coree 3.0 2.48% 23.1% 14.9% 0.5% 37.6% 6.8% 17.1% Japan - Japan 29.2 0.07% 14.8% 24.8% 0.2% 47.3% 3.5% 9.6% Table 4b Foodborne disease outbreaks by place of eating in the Republic of Korea and Japana Tableau 4b Flambees de maladies d'origine alimentaire, par lieu de consommation, en Republique de Coree et au Japona Country-Pays Home- Restaurant Hotel- Workplace- School- Retail store - Other- Unknown- Domicile Hotel Lieu de travail Ecole Magasin Divers lnconnu Rep. of Korea - Rep. de Coree 48.8% 10.6% 5.3% 19.1% 2.5% 0.5% 11.9% 1.3% Japan - Japan 17.2% 32.7% 11.5% c. 3% c. 4% c. 3% c. 16% c. 15% Table 4c Foodborne disease outbreaks by implicated food in the Republic of Korea and Japana Tableau 4c Flambees de maladies d'origine alimentaire, par aliment incrimine, en Republique de Coree et au Japona Country - Pays Seafood- Animal products - Grain/vegetable Confectionery- Multiple foods - Other- Unknown- Poissons et Produits + mushrooms - Patisserie Aliments Divers lnconnu fruits de mer d'origine Cereales/legumes multiples animale Rep. of Korea - Rep. de Coree 31.8% 25.0% Japan - Japan 21.7% 3.6% • Source: Lee et al. (Ref. 50). - Lee et al. (Ret 50). the second leading cause of illnesses and death, even though these were highly underreported (53). Infected foodhandlers and pesticide residues in foods were considered to be important risk fac- tors for foodborne disease in Vietnam, although no direct links were made between these and inves- tigated foodborne illnesses. In Thailand, the focus appears to be mainly on chemical poisonings, particularly from insecti- cides, although these accounted for only 0.33% of the reported 207 580 cases of foodborne disease between 1981 and 1986 (54). Between 1981 and 1987, insecticides accounted for 27.4% of out- breaks and 58.4% of cases (54, 55 ). Because of the widespread use of insecticides, some of these have accidentally contaminated desserts, beverages, fruits and other foods. Methomyl, which looks like sugar or flour and has little odour, was responsible for 15 of the 18 insecticide-related outbreaks. Since 1987 this chemical was sold blue-colored in an attempt to reduce these poisonings. In addition, an Wld hlth statist. quart., SO (1997) + champignons 17.5% 2.9% 18.3% 1.9% 2.6% 14.6% 1.2% 9.6% 1% 48.3% alcoholic beverage containing methanol affected 10 males during a party; 5 died and one had perma- nent visual impairment. Poisonous plants, such as mushrooms, raw cassava roots and wild plant seeds were responsible for 58.9% of outbreaks and 34% of cases, and poisonous animals, including toxins in seafood, for 11.0% of outbreaks and 6.5% of cases. There was one episode of paralytic shellfish poisoning (PSP) with 63 cases and 1 death follow- ing consumption of green mussels containing 465- 714 mouse units PSP toxins/ g, 4 outbreaks associ- ated with meals made with horseshoe crab meat, and 3 outbreaks resulting from ingestion of puffer fish probably containing tetrodotoxin. The incidence of food borne disease in Malaysia was 9.62/100 OOO in 1981, with the most frequent etiological agents being S. aureus, V. parahaemolyti- cus and Salmonella (56). In Singapore, the main bacterial pathogens isolated from 7 344 patients with diarrhoea were Salmonella (10.1 %), Campy- lobacter ( 1. 2 % ) , Shigella ( 1.1 % ) , Vibrio parahaemolyti- 37 cus (0.8%) and V. chol.erae 01 (0.2%) (57). The sources of these were not determined but the V. parahaemolytuus infections probably arose from consumption of contaminated shellfish. Pigs play an important economic and cultural role in the tribes that live in the Central Highlands of New Guinea. In the Indonesian part of the island (IrianJaya) the prevalence of cysticerosis is the highest in the world according to Muller et al. (58), with a rate of over 30%. Pork is often con- sumed insufficiently cooked to destroy parasites or bacterial spores. Pig-bel, a severe form of Clostri- dium perfringens enteritis, is associated with ritual feasts involving consumption of roast pork in Papua New Guinea (59). There were 721 foodborne outbreaks and 1 199 sporadic cases in the twin cities of Hyderabad/ Secunderabad, India, between 1984 and 1989 ( 60 ). The majority of outbreaks affected 2-10 persons, and occurred between February and June. The main vehicles of transmission were "stale" food (36.5%), rice dishes (23.5%), sweets (12.6%) and curry (9.8%). "Stale" food is probably food that has been left over from a previous meal at room tem- perature for a lengthy period of time, usually over- night (61). Chicken, pork, goat and fish were the most frequent components of curries and rice dishes, and were most often eaten at parties or in homes. S. aureus and Bacillus spp. were the most likely causative agents since they are often present in these foods. Foodborne disease outbreaks A very large outbreak of hepatitis A occurred in Shanghai in January and February, 1988, with 292 301 cases and 32 deaths. The virus was trans- mitted through clams contaminated with sewage water ( 62, 63 ). From information available up to 1989 there were 745 outbreaks and 2 861 cases attributed to botulism in China with a 14. 7% fatal- ity rate. Most of the outbreaks (71.8%) were from home-fermented bean or cereal products (64). As more commercial products become available to the population, it is expected that fewer home- made products will be consumed and there will be correspondingly fewer botulism outbreaks. Be- tween 1985 and 1990 in Taiwan there were 28 cases of Vibrio vulnificus infections (65). Most of these cases had ingested seafood or had exposed abraded skin to seawater. In Japan, for the most recent years, Salmonella has become much more frequent because S. enteri- tidis infections arising from consumption of eggs or egg products have been increasing (2). Further- more, in 1996, there was widespread public con- cern after a number of large outbreaks due to E. coli 0157:H7 occurred and several children died. In July and August, in Sakai City, a total of 6 309 schoolchildren and 92 staff from 62 elemen- tary schools were affected and a further 160 sec- 38 ondary infections developed, mainly in family members of the schoolchildren (66). In Habikino City another outbreak affecting 98 persons in a home for the elderly and 3 other small outbreaks in the same region occurred. All of the strains from the 5 outbreaks had identical DNA patterns, and radish sprouts from one farm were consumed by those ill. However, no isolates could be obtained from samples of soil, water and sprouts on the farm. A total of9 578 cases of E.coli 0157:H7 and H- and 11 deaths were documented from out- breaks and sporadic cases as of26 August, 1996. No responsible foods have yet been identified, except in a few isolated cases. Tetrodotoxin puffer fish poisoning is one of the most serious in Japan; in 1982 there were 26 episodes and 8 deaths, but only 1 death in 1990 (67). In fact, the reduction in deaths has been consistent over the years from 411 in 1949 to 1 in 1990. In 1991, 21 persons in an inn near Tokyo suffered from cholera and one died after they consumed contaminated imported Kore- an clams (68). There had been a cholera outbreak in the Republic of Korea at that time. In Malaysia, in 1983 at a school canteen, 48 students eating meehoon (fried rice noodles) de- veloped S. aureus intoxication. The organism was isolated from the vomitus and nasal swabs of 3 food handlers (56). The source of the S. aureus was not identified, since only scanty growth of the organism was obtained from the positive food handlers. S. aureus was also implicated in one outbreak in Hyderabad, India, when over 100 persons fell ill after eating a sweet porridge (69). About 1 OOO persons suffer from foodborne dis- ease in this city each year at an estimated cost of $10 480. As in India, sweet dishes are vehicles for S. aureus intoxication in Pakistan (70). One such outbreak from khoa, a confectionery with concen- trated buffalo milk, caused 8 persons to be hospi- talized in the early 1980s. More recently, some samples of khoa obtained from manufacturers in a large city contained up to 10s S. aureus/g. Salmo- nel/.a was also found in khoa and in cheese-based confectioneries (70). Pulses, ground meat dishes and chick peas vended at bus and train stations in the same city contained 104-107 C. perfringens/g, when the holding temperature was not hot enough (28-46 °C) (71). Home-prepared foods in small communities also contained pathogens, such as S. aureus, C. perfringens and B. cereus, but not Salmonella. The main hazard identified was holding foods for long periods of time, e.g., over- night, at ambient temperatures (72). An outbreak involving soy milk affected 35 of 263 school- children at a midday meal in Dehli (73). The 30-minute incubation period and mild symptoms were indicative of S. aureus or B. cereus intoxica- tion, although only E. coliat >105/ml were found in the milk. Rapp. trimest. statist. sanit. mond., SO (1997) In a village in Sri Lanka the carcass of a freshly- dead monkey, including its entrails, was made into a curry and 9 persons who ate this were subse- quently infected with S. enteritidis phage type 8 (74 ); one died. Other dead monkeys found later in the same area were suspected to have suffered from Salmonell.a septicemia, but no samples were taken for microbiological analysis. It was assumed that the curry contained Salmonell.a-contaminated meat and that the heat was insufficient to destroy the pathogen. In Bombay, 132 persons were ill and 4 died after eating fish that were harvested from algae-rich water and an algal toxin was believed to be responsible for their symptoms (75). Around the coastline of India, Pakistan, Myanmar and Thailand, seafood toxins such as paralytic shellfish poison (PSP), diarrhetic shellfish poison (DSP) and histamine, are a risk. However, Vibrio and Sal- monell.a infections, through consumption of fish and crustaceans, are more frequent (75). Oceania Epidemiological studies In Australia, some trends in notifications of enteric diseases are apparent for 1991-95 (Tabl,e 5). Labora- tory isolates for Campylobacter and Salmonell.a isola- tions are increasing, those for Shigell.a and Yersinia are decreasing and those for Listeria monocytogenes vary slightly from year to year. New Zealand up- dated its notifiable diseases in 1996 to include bot- ulism, campylobacteriosis, cryptospridiosis, giar- diasis, listeriosis, toxic shellfish poisoning, VIEC, and yersiniosis (76). A surveillance of listeriosis was conducted in 1995: there were 15 cases compared with 11 in 1994 (77). Three of the cases were peri- natal and 2 of the infants died. Of the remaining 12 cases, 11 had underlying disease or were elderly. None of the cases were in clusters and there were no links to food. Risk factors for campylobacterio- sis were determined through a case-control study fromjune 1994 to February 1995 (78). The main Table 5 factors were consumption of raw or undercooked foods ( especially poultry and unpasteurized dairy products) and untreated drinking water, overseas travel and contact with animals. Thorough cooking of chicken could significantly reduce the incidence of campylobacteriosis. Foodborne disease annual reports A recent summary for the years 1980 to 1995 by Crerar et al. (79) indicates that the agents respon- sible for illness are similar to those in other industri- alized countries with Salmonella being the predom- inant cause of morbidity and mortality (Tabl,e 6). However, S. enteritidis does not seem to be a major problem in Australia. Vibrio parahaemolyticus infec- tions were greater than in North America or Europe, but this probably reflects the fact that most of Australia's population live close to the sea and consume seafood regularly. Scombrotoxin and viral outbreaks were also associated with seafood. Foodborne disease outbreaks S. mbandaka caused at least 54 cases of illness in Victoria and South Australia in 1996 and was shown to result from consumption of one brand of peanut butter (80). Individuals infected with this rare serotype occurred in other states and territo- ries and about half of the cases were under 5 years of age; links to peanut butter were only established in 3 Western Austalian cases. One case also oc- curred from the same brand of peanut butter in New Zealand. The product was recalled in both countries. Although sporadic cases of haemolytic uraemic syndrome (HUS) have been associated with E. coli 0111 and other VIEC in Australia, the first out- break was in 1995 when mettwurst produced by a small manufacturer infected many people and caused HUS in 23 children with 1 death (81). This stimulated considerable interest in VIEC infec- tions and HUS. In a study of 55 cases of Queens- land children with HUS, most of those preceded by Notifications of selected enteric diseases in Australia, 1991-1995a Tableau 5 Cas notifies de maladies intestinales, Australie, 1991-1995a Year- Campylobacteriosis - Salmonellosis - Shigellosis - Listeriosis - Yersiniosis - An nee Campylobacteriose Salmonellose Shigellose Listeriose Yersiniose 1991 8 672 5 440 902 44 515 1992 9135 4 614 894 38 567 1993 8 311 4 731 708 53 450 1994 10 117 5 283 724 34 414 1995 10 933 5 895 734 58 306 • Source: National Salmonellosis Surveillance Scheme, Communicable Diseases Intelligence Reports, Canberra, Australia. Wld hlth statist. quart., 50 (1997) 39 Table 6 Foodborne disease outbreaks in Australia, 1980-1995, by etiological agenta Tableau 6 Flambees de maladies d'o~igine alimentaire, Australie, 1980-1995, par agent etiologiquea Agent No. of outbreaks - N. de flambees Salmonella 27 C. perfringens 14 S. aureus 9 Campy/obacter 5 B. cereus 5 V. parahaemotyticus 4 L. monocytogenes 2 E. coli 0111 1 C. botulinum Small round structured viruses - Petits virus ronds structures 11 Rotavirus 1 Hepatitis A virus - Virus de l'hepatite A Toxoplasma 1 Scombrotoxin - Scombrotoxine 2 Ciguatera Mushroom poison - Champignons veneneux 1 Total known - Total pour les agents connus 86 Unknown - lnconnu 42 Total 128 a Source: Crerar et al. (Ref. • Ref. 79). diarrhoea were under 5 years of age (85%), had reduced or no urine output (71 % ) , experienced hypertension (31 %), and developed seizures (29%). Eighty-five percent required transfusion, 56% antihypertensive therapy, 56% peritoneal dialysis and 2% haemodialysis which lasted 3 to 29 days, and 10% ventilation (82). One child died. In February and March 1996, the first 0157 out- break occurred in Australia when 6 persons were infected from food served in a delicatessen on the Gold Coast in south east Queensland (83). A food handler was the index case and may have contract- ed the infection from her pet dog which had bloody diarrhoea in the week before she devel- oped symptoms. Recent outbreaks in New Zealand include 3 separate episodes of S. typhimurium phage type 35 associated with the consumption of bakery prod- ucts in Christchurch in 1993 (84), two incidents in 1994, with people ill after eating curry probably contaminated with Clostridium perfringens, one with 31 persons at an Aukland wedding reception in March (85), and the other with 59 attendees at a fashion show in April (86), and hepatitis A in Well- ington from delicatessen food contaminated by the 40 % of outbreaks - No. of cases - % of cases- No. of deaths - % des flambt!es N. de cas % des cas N. de d6ces 21 1 323 22 1 11 280 5 0 7 99 2 1 4 106 2 0 4 27 0 3 181 3 2 2 13 0.2 0 0.8 23 0.4 1 0.8 0.02 0 9 2 267 38 0 0.8 55 0.9 0 0.8 7 0.1 0 0.8 13 0.2 1 2 8 0.1 0 0.8 30 0.5 0 0.8 5 0.08 0 67 4 438 75 6 33 1 514 25 0 100 5 952 100 6 owner/ operator who was the index case in 1996 (87). The two C. perfringens outbreaks probably in- volved the same supplier, although this is not stat- ed in the reports, and in one of them the practice for years had been for large pots of meat to be left at room temperature because they were too big for the cold room. Canada and the United States Epidemiological studies In Canada, reports of enteric diseases show that campylobacteriosis is increasing, salmonellosis is decreasing, and that E. coli 0157:H7 enteritis rose to a peak after its recognition as a pathogen in 1982 and has since declined (Tah/,e 7). In the United States, salmonellosis and shigellosis appeared to be increasing from 1992 to 1995 (Tah/,e 8). E. coli 0157:H7 data are available only since 1994 with 1 420 for that year and 2 139 for 1995; despite the recent publicity on this pathogen, there are ap- proximately 0.5 cases per 100 OOO population caused by it, much less than reported in Canada (4.0 cases per 100 OOO). A survey of used cellulose Rapp. trimest. statist. sanit. mond., 50 (1997) Table 7 Cases of enteric diseases in canada, 1985-1994a Tableau 7 Cas de maladies i ntestinales, Canada, 1985-1994a Disease - Maladie 1989 1990 1991 1992 1993 1994 1995 Mean noJ 100 OOO in 1995- N. moyen /100 OOO en 1995 Campylobacteriosis - Campylobacteriose 9 653 9 081 9 786 7 666 13 669 11 767 10 499 35 Salmonellosis - Salmonellose 8 762 8 742 8 984 7 265 8 057 7 441 7138 24 Escherichia coli 0157:H7 enteritis - Enterite a Escherichia coli 0157:H7 2 432 1 585 1 565 1 521 1 212 1 014 1 277 4 • Source: National Enteric Reference Centre, Laboratory Centre for Disease Control, Ottawa. - Centre de reference national de bacteriologie enterique, Laboratoire de lutte contre la maladie, Ottawa. Table 8 Cases of notifiable enteric diseases in the United States of America, 1992-1995a Tableau 8 Cas de maladies intestinales a declaration obligatoire, Etats-Unis d'Amerique, 1992-1995a Enteric disease - Maladie intestinale 1992 1993 1994 1995 Salmonellosis - Salmonellose 40 912 41 641 43 323 45 970 Shigellosis - Shigellose 23 931 32196 29 769 32 080 Escherichia coli 0157:H? enteritis - not reported - not reported - 1 420 2139 Enterite a Escherichia coli 0157:H7 non rapporte non rapporte • Source: Summary of notifiable diseases (Ref. 133). - Donnees recapitulatives sur les maladies a declaration obligatoire (Ref. 133). sponges and cotton dishcloths collected from households in 4 United States cities showed that these had a considerable faecal coliform load in liquid wrung from the sponges and dishcloths (4.5 x 102 CFU/ml and 2.0 x 1Q3 CFU/ml, respec- tively). Salmonel/,a was present in 15.4% of the sponges and 13.8% of the dishcloths, and S. aureus in 20.0% of the sponges and 18.6% of the dish- cloths (88). These soiled cleaning materials may be an important souce of bacterial contamination in domestic kitchens. A study of 80 farm families in Ontario, Cana- da, found that many individuals experienced mild or subclinical immunizing VTEC infection at an early age (89). VTl antibodies were found in 41 % of persons tested. E. coli 0157:H7 and 8 other serotypes were isolated from 21 persons on 16 dif- ferent farms. Four of these serotypes were isolated from cattle living on the same farms. Non-0157 verotoxin producing organisms have been shown to be pathogens, but the methodology to detect them in stools or food samples is currently limited (90). Outbreaks have not been reported from Canada, but they have from the United States, Japan, the Czech Republic, Germany, Italy, and Australia; the source of these is largely unknown but in 2 of them parsley and mettwurst were implicated. Wld hlth statist. quart., 50 (1997) Foodborne disease annual summaries In both the United States and Canada, salmonello- sis cases seemed to have reached a plateau of about 40 OOO and 9 OOO each year, respectively (2), de- spite the fact that S. enteritidis has become a major egg-borne pathogen in the United States over the last 15 years. Moreover, illnesses from E. coli 0157:H7 and other VTEC are being documented in outbreaks from both countries, not only from ground beef but also from vegetables, milk and apple juice. Unfortunately, foodborne disease re- ports at the national level lag behind the laboratory isolation data by several years in both countries; the last published material was for 1991/92 (91, 92). Therefore, relatively current trends may not always be apparent from a passive surveillance system which is not coordinated with laboratory analysis. This delay has been a function of low priority for foodbome disease surveillance at the responsible state/provincial and federal departments over many years. However, because of increased public health concern recently over enteric disease in both countries, sentinel sites are being established in the United States (44), and an enhanced surveil- lance system is being developed in Canada. In the United States, data for outbreaks between 1988 and 1992 were recently published (92). The num- ber of outbreaks and cases had not changed sub- 41 stantially during this period, although there was a peak in 1990 with 528 outbreaks and 19 883 cases (Tabk 9). Many of these were caused by Salmonella. Other pathogens responsible for foodborne ill- nesses were the traditional ones: S. aureus, C. per- fringens, C. botulinum, Shigella, B. cereus and hepati- tis A. Few E. coli O 157 outbreaks were documented during this time period. Both ciguatera and scom- broid poisonings represented the bulk of the chemical illnesses. Foodborne outbreaks in prisons were assessed in the United States from 1974 to 1991, with 88 outbreaks and 14 307 cases (93). The three main pathogens implicated were Salmonella includ- ing S. enteritidis (37%), C. perfringens (34%), and S. aureus (22%). About 50% of the outbreaks had no agent identified. Beef, poultry, meat and fish salads and Mexican foods were the main foods implicated. Improper food storage and inadequate cooking were the main factors contributing to out- breaks. Because the HIV seroprevalence in prisons is higher than average (18.9% in 1989), the risk of serious infections is high in prisoners from any source including food. Table 9 In Canada, there were in 1989, 1990 and 1991, 1 248, 1 044 and 986 outbreaks, and 6 879, 6 027 and 6 408 cases, respectively (91). The number of bacterial pathogens causing illness included Salmo- nella, responsible for most incidents (51 in 1989, 28 in 1990, 28 in 1991), followed by Staphylococcus aureus (21, 13, 21, respectively for the same years), Bacillus cereus (19, 15, 11), Campylobacter (13, 13, 15), E coli 0157:H7 (9, 7, 16) and Clostridium per- fringens (14, 3, 1). The most dramatic decrease was for C. perfringens, down to a single case incident being reported in 1991, whereas E. coli 0157:H7 incidents rose from 7 in 1990 to 16 in 1991, al- though most of these were small in size. There was also one outbreak of 50 cases with 5 foods contain- ing Salmonella heidelherg and high levels of B. cereus. Incidents caused by chemical agents were next most frequent, mainly extraneous matter, rancid compounds, and heavy metals. A few incidents of paralytic shellfish poisoning, scombroid poison- ing, ciguatera poisoning and mushroom poisoning occurred during the 3-year period. One large out- break involved uncooked wild morel mushrooms served at an hotel banquet. Meat, poultry, bakery Foodborne disease outbreaks in the United States of America, by etiology, 1988 to 19923 Tableau 9 Flambees de maladies d'origine alimentaire, Etats-Unis d'Amerique, par etiologie, 1988 a 19923 Etiology - Etiologie 1988 1988 1989 1989 1990 1990 1991 1991 1992 1992 outbreaks - cases - outbreaks- cases- outbreaks- cases- outbreaks- cases- outbreaks- cases- flambees cas flambees cas flambees cas flambees cas flambees cas Salmonella 94 2 987 117 4 920 136 6 290 122 4146 80 2 834 C. perfringens 0 0 7 436 11 1 240 10 1 213 12 912 S. aureus 8 245 14 524 13 372 9 331 6 206 C. botulinum 20 49 13 24 12 22 11 25 4 13 Shigella 6 3 581 6 257 8 834 4 112 1 4 B. cereus 5 51 3 61 5 43 5 253 3 25 Campy/obacter 4 134 8 295 3 72 6 93 6 138 E.coli 2 109 1 3 2 80 3 33 3 19 II. cholerae 0 0 0 0 5 26 2 6 1 2 Other - Divers 0 0 2 37 5 23 100 1 3 Hepatitis A - Hepatite A 12 795 7 329 9 452 7 114 8 419 Norwalk-like virus - Virus type Norwalk 0 0 1 42 0 0 0 0 250 Parasitic - Parasites 3 34 5 36 5 234 3 73 2 Seafood toxins - Toxines de poissons et fruits de mer 21 79 36 146 24 262 26 125 16 143 Other chemical - Produits chimiques divers 8 60 7 3 8 5 34 3 63 Total confirmed - Total pour les etiologies confirmees 183 8124 221 7 117 237 9 958 214 6 658 146 5 033 Unknown - lnconnu 268 7 608 284 8 750 295 9 925 314 8 218 261 5 982 Total 451 15 732 505 15 867 532 19 883 528 14 876 407 11 015 a Source: Bean et al. (Ref. 92- Ref. 92). 42 Rapp. trimest. statist. sanit. mond., 50 (1997) products and fish/ shellfish were the main foods associated with illness. Foodbome disease outbreaks Many outbreaks are published each year for the United States, and only a few are selected here. In 1994, 593 cases of S. enteritidis were identified in Minnesota after a nationally-distributed brand of ice cream was eaten (94). Pasteurized ice cream mix had been transported in a tanker previously used to carry non-pasteurized liquid egg. If it is assumed that the ice cream was contaminated in the sample taken at the same level of 0.093 S. enteritidis PT 8/ g throughout the whole load, a possible 224 OOO persons in the United States could have been ill. This represents a reporting rate of0.3% even when there was widespread pub- licity surrounding the event. This is about the same degree of underreporting of outbreaks estimated by Todd (95, 96), about 1 case in 350. If these data are correct, they indicate that in many outbreaks the existing local/state reporting mechanism is very limited. A commercially canned cheese used as a cheese sauce on barbecue stuffed potatoes in Georgia in 1993 caused 8 cases of type A botulism (97). The mild cases and the fact that an unusual vehicle, cheese, was implicated made the diagnosis of the disease difficult. The cheese was probably contaminated with spores after the can was opened, possibly from the potatoes. In recent years Cryptosporidium had been impli- cated in large waterborne outbreaks, but there have been 2 foodborne episodes that indicate the parasite may also be transmitted through faecal contamination of food. In 19'93, consumers of ap- ple juice in Maine contracted cryptosporidiosis where there was evidence that the cysts came from cattle manure contaminatin~; the apples (98). In 1995, 15 of26 persons attending a social function in Minnesota suffered from a Cryptosporidium infec- tion after eating a chicken salad (99). It is probable that the host had contaminated her hands through changing a diaper from an asymptomatic child in her licensed day-care home, although she washed her hands before preparing the salad. The strongest evidence that Cyclospora may be foodborne was in 1996 when raspberries prob- ably from Guatemala caused severe diarrhoea in more than 1 400 persons in 15 states and Ontario, Canada. Because of the lack of suitable methodolo- gy to isolate the cysts from food and its probable low infectious dose, no laboratory confirmation of the parasite in fruit has yet been made. One possi- ble source for the Cyclospora was a pesticide spray using untreated water (Marta Ackers, Florida Cen- ters for Disease Control, personal communica- tion). In Ontario in 1996, roast turkey served at a church supper was implicated in an outbreak of E.coli 0157:H7 with 36 cases (14 confirmed and Wld hlth statist. quart., 50 (1997) 4 seriously ill) (100). Three of the 6 ill food prepar- ers had 0157 in their stools. The turkeys appeared to be properly cooked and may have been contam- inated during the cooling on the kitchen counter. This is one of the first VfEC outbreaks associated with a church event. Other foods implicated in VfEC infections in Ontario between 1990 and 1995 were ground beef, chicken, beef, unpasteur- ized milk, pork, goat, seafood, eggs and apple cider (101). Turkey, chicken and eggs are not foods tra- ditionally associated with VfEC infections. Also in Canada, 5 separate incidents of ciguatera poison- ing occurred from September 1996 to February 1997. Barracuda, Spanish mackerel and doctorfish purchased in Ontario or Quebec were implicated and leftover fish were confirmed by mouse bioassay to contain ciguatoxin or related toxins. This in- crease in illness may reflect a growing market for tropical fish in temperate regions. An example of an international outbreak affect- ing both the United States and Canada involved a kosher snack item imported from Israel into North America as well as Europe. During December 1994 and February 1995, 27 cases of S. agonawere identi- fied in England and Wales, more than double the number for the same period the year before ( 102). A case-control study showed a strong association between infection with S. agona phage type 15 and consumption of a peanut-flavoured savoury snack imported from Israel. Information relayed to Israel helped identify the cause of more than 2 200 phage type 15 S. agona infections in that country during the same time period (103). The combined testing of food from the United Kingdom, the United States and Canada, where the product was imported, showed that the contaminated snacks were manufactured on at least 7 separate dates during a 5-month period, between October 1994 and February 1995. Ten cases were documented in the United States but none in Canada and in other European countries contacted through Salm-Net. Voluntary recalls were carried out to avoid further illnesses. Central and South America Notifiable and foodbome disease reporting All Central, South American and Caribbean coun- tries have some form of notifiable disease system. Diarrhoeal diseases are one of the main causes of death in young children, e.g., 967 deaths/100 OOO infants < 1 year old in Nicaragua compared with 0.5 deaths in Canada, and between 4 to 7 bouts of diarrhoea are experienced each year by a child < 5 years old ( 104 ). The causes of these are not generally known, but amoebic dysentery, trichino- sis, giardiasis, shigellosis, brucellosis, typhoid fever, E. coli and hepatitis infections are all documented from Latin America and the Caribbean, and there 43 is increasing evidence to link some of these to foods. Such is the case with cholera. This disease was first documented in Peru in 1991 with a total of 600 OOO cases (105). It rapidly spread to other countries, and in 1994 caused 112 611 cases and 1 229 deaths, mainly in Peru, Brazil, El Salvador, Nicaragua, Honduras, Guatemala, Mexico, Bolivia, Ecuador, and Colombia (106). The total number of cases and deaths from 1991 to 1994 was 1 061 188 and 9 989, respectively. The source of the cholera was probably contaminated food. In 1991, V. chol.era.e O 1 was found throughout Peru in water, sewage, finfish (skin and intestines), mol- luscs and plankton with counts up to 105/100 mL (107). The disease was spread partly through con- sumption of street-vended foods and beverages containing ice (108). Undercooked or raw seafood may also have been implicated as these have been associated with cholera in the past ( 109). Shellfish may be contaminated not only with local sewage, but waste water pumped from ships in har- bour (110). Seafood (crab, shrimp, ceviche) was implicated in several incidents among United States travelers to Ecuador and Mexico in 1992 ( 109, 111). The economic loss in Peru in 1991 from lost markets and tourism, absenteeism from work, medical care and the value of lives lost due to the cholera epidemic was estimated at US$ 495.3 billion ( 112). In Mexico, colonization of infants by heat-labile ETEC increased 400-500% during the rainy season. There was a similar in- crease for children who received oat gruel, but the risk of symptomatic infection was reduced by the Table 10 Foodborne disease in 9 Central and South American countriesa Tableau 10 ingestion of herbal tea and LT-ETEC-specific anti- bodies in breast milk. Also, the likelihood of non- infection increased 200% for each increment in the mother's level of education and for each pre- vious LT-ETEC-posirive episode experienced by the child (113). Salmonella may also be an impor- tant agent in Mexico, since one study showed that even with a method of limited sensitivity, 4.5% of chocolate samples in Guadalajara tested positive ( 114 ). Homes were the main places where the con- taminated food was acquired in Venezuela (70% outbreaks) and in Parana, Brazil (42%) (115). In Parana, the other locations of acquisition were in- stitutions (25%), restaurants (15%), and unknown (17%). In Cuba, S. aureus is the most frequently-occur- ring etiological agent, responsible for 22.6% of all, and 49.3% of known, outbreaks, followed by C. perfringens, E. coli ( types not specified) Salmonella and B. cereus. The number of outbreaks caused by microorganisms and seafood toxins increased from 1985 to 1988 (104). Chemically-caused out- breaks were more sporadic in nature, but episodes of pesticide poisoning were reported each year. Beef, pork and chicken products, fish and shellfish were the main foods associated with illness. Cream- filled desserts were also implicated in S. aureus illnesses. Those ill primarily ate in school cafeterias or canteens at work. In 8 other countries, food- borne outbreaks and cases were documented, but with little information available on etiology or foods implicated. Tabl.e 10 shows the current state of knowledge for Latin America. Maladies d'origine alimentaire dans 9 pays d'Amerique centrale et d'Amerique du Suda Country - Pays Years- No. of outbreaks - No. of cases/year - Other information - Annees N. de flambees N. de cas/an Autres renseignements Brazil - Brasil 1985-89 42-90/year/an 5 627-9 758 3,0-5,7% of cases hospitalized/year - 3,0-5,7%de cas hospitalises/an Columbia - Colombie 1983-88 5 281-8 668 Dominican Republic - 1989-90 45 196 Republique dominicaine (6 months) - (6 mois) El Salvador 1989 509 Guatemala 1987-89? 9 32 20 hospitalized - hospitalises Mexico - Mexique 1981-90 363 14 412 No. of cases/outbreak= 41 (median) N. de cas/flambee = 41 (mediane) Venezuela 1989 23 293 1990 14 400 Argentina - Argentine 1986-90 35b 3 500 Buenos Aires Province - Province de Buenos Aires 1988-90 522 a Source: VETA (Refs. -R6t 104, 115). b Salmonella enteritidis only-Salmonella enteritidis uniquement. 44 Rapp. trimest. statist. sanit. mond., 50 (1997) Foodborne disease outbreaks In Argentina, S. enteritidis(l16)andE. coli0157:H7 ( 117) have been responsible for foodborne ill- nesses. Between 1986 and April 1990, 35 outbreaks of S. enteritidis affected 3 500 persons, largely through consumption of insufficiently cooked poultry and eggs used in mayonnaise. In Provincia de Buenos Aires, S. enteritidis caused 23% of the outbreaks, other bacteria 44%, chemicals 27% and unknown origin 6%. However, in Venezuela, Colombia and Brazil, S. aureus was the agent re- sponsible for the majority of outbreaks, with inci- dents caused by C. perfringens, ETEC and B. cereus also documented. Bergdoll et al. (118) reported on 30 S. aureus outbreaks from two regions of Brazil in 1988-90; cream-filled cakes, white cheese, sausage and milk were the foods implicated. Recently, an outbreak in Brazil was caused by a new form of staphylococcal enterotoxin, enterotoxin H, present in cheese ( 119). In a remote mountain community in Peru a health hazard analysis of foods prepared in homes was conducted and it was discovered that opportunities for foodborne dis- ease were limited ( 120). Most foods were thorough- ly cooked and eaten within a few hours. There were usually few leftovers. In Venezuela records have been kept since at least 1981 and, from that year to 1988, S. aureus caused 64. 7% of outbreaks, Salmonella and B. cereus 5.9% each, and "faecal coliforms" 23.5%. Cheese was the most common vehicle of illness (35.3%), followed by cakes (23.5%), tuna (23.5%) and bolo- gna, turkey and a maize beverage (each 5.9%). Cheese is often contaminated with S. aureus in this and other countries. Street-vended foods in the Dominican Republic had counts as high as 109 aerobic colony counts/ g with holding temper- atures usually <30 °C (121). Pathogens, however, were <103/g. Although street-vended food is fre- quently suspected as a source of foodborne out- breaks, there is rarely microbiological proof found during investigations. However, some babies in the Dominican Republic that were being treated for enteric infections came from homes where formu- la and other foods were prepared in a hazardous manner (122). Pathogens were found in these foods and the kitchen environment. Botulism has been reported from Argentina, Brazil, Chile, Guatemala, Peru, Mexico and Venezuela (123, 124), but incidents are not listed along with other food borne diseases ( 104 ). Seafood toxins are important in coastal regions. Ciguatera is a major concern in the Dominican Republic where it can affect the tourist trade as well as the residents (125, 126). There have been major outbreaks in the United States Virgin Islands, Cuba and the Dominican Republic, and fish containing ciguatoxin is an ongoing problem in many of the northern Caribbean Islands. One exceptionally large outbreak of ciguatera poisoning affected Wld hffh st.atist. quart., 50 (1997) 57 Canadian tourists staying in a Cuban resort ( 127). Paralytic shellfish poisoning has occurred in Argentina, Chile, Costa Rica, El Salvador, Guate- mala, Honduras, Mexico, Nicaragua and Vene- zuela (128). One hundred and eighty-seven cases and 26 deaths arose from the first documented outbreak in Guatemala in 1987 when a red tide of Pyrodinium bahamense var compressa contaminated shellfish on many beaches on the coast (129). Scombroid poisoning through histamine in fish is recorded from Cuba, but must also occur in other countries. For instance, a marlin exported from Ecuador to Canada gave rise to 12 cases of hista- mine poisoning ( 130); a similar episode was docu- mented in the United States ( 131). Presumably, these fish are sold locally and to other South American countries. Most Latin American countries now recognize that foodborne disease is important enough to jus- tify some kind of surveillance scheme and are try- ing to develop better ways of determining numbers of cases and their causes ( 104, 115 ). Conclusion This worldwide survey shows that microbiological issues are the major ones in all the countries con- ducting foodborne disease surveillance pro- grammes. Salmonella, S. aureus and the better rec- ognized agents continue to plague most countries, and E. coli O 15 7 and other VTEC strains, parasites, and other emerging pathogens are presenting new challenges to the food industry and national authorities. Despite our knowledge of the source of these organisms and factors contributing to dis- ease, there seems to be little decrease in numbers affected, and the socio-economic impact of food- borne disease remains very high. One reason that control is difficult to achieve is because surveil- lance is inadequate and the burden of foodborne disease is not fully understood by policy-makers. Another reason is that a consistent and coordi- nated effort by industry and government is re- quired. With increasing world trade and travel, improvement of surveillance on a worldwide basis is all the more important. In the United States and the United Kingdom, sentinel general practitioner studies are under way to find out the nature and causes of gastroenteritis ( 44 ). Both epidemiologi- cal and laboratory components are being incorpo- rated into these programmes. The United States commitment to this was reinforced by the presi- dential proposal for the 1998 budget to allocate $43 million for a programme to detect foodbome outbreaks before they become widespread and also to reduce the estimated 6.5 million-33 million peo- ple who become ill and 9 OOO who die every year from foodbome diseases (132). These initiatives should stimulate other countries to conduct appro- priate surveillance programmes, so that the real burden of food borne disease can be determined at 45 various national levels. If it is now being recognized as a major concern in industrialized countries such as the United States and the United Kingdom, how much more of a problem is it in countries where urban growth is faster than the public health infras- tucture can support, and in rural areas where drinking water is frequently contaminated? This issue is becoming increasingly important now that immigrants from developing countries are becom- ing more frequent, international travel is common- place, and trade barriers between blocks of coun- tries are coming down. International organiza- tions, such as WHO and FAO need to take a lead role in accomplishing better surveillance for both industrialized and developing countries, by build- ing on the existing expertise, and insisting on the required funding to accomplish this. Since risk assessments, used to prioritize programmes and be consistent with the World Trade Organization (WTO), are dependent on good outbreak data, food surveys, and population demographics, all countries have to be involved. Summary Acute foodborne disease infections and intoxications are much more of a concern to governments and the food industry today than a few decades ago. Some of the factors that have led to this include the identification of new agents that have caused life-threatening condi- tions; the finding that traditional agents are being asso- ciated with foods that were of no concern previously; an increasing number of large outbreaks being reported; the impact of foodborne disease on children, the aging population and the immunocompromised; migrant pop- ulations demanding their traditional foods in the coun- tries of settlement; the ease of worldwide shipment of fresh and frozen food; and the development of new food industries, including aquaculture. However, to mean- ingfully monitor increases or decreases in foodborne disease requires an effective surveillance system at the local, national and international levels. To date, resour- ces have been limited for most countries and regions to do this, and our current knowledge is based, for the most part, on passive reporting mechanisms. Laboratory isolation data and reports of notifiable diseases have some value in observing timely changes in case num- bers of some enteric diseases, but they usually do not indicate the reasons for these trends. Special epidemi- ological studies are useful for the area covered, but it is often questionable whether they can be extrapolated to other areas or countries. Outbreak investigations tell us that a certain set of circumstances led to illness and that another outbreak may occur under similar but not nec- essarily identical conditions. Control programmes have often been triggered by the conclusions from investiga- tions of specific outbreaks. Unfortunately, the agent/ food combination leading to illness in many of the reported incidents were not predicted from existing databases, and no doubt foodborne agents will contin- ue to surprise food control agencies in the foreseeable future. 46 Nevertheless, data from around the world do show some common elements. Salmonella is still the most important agent causing acute food borne disease, with Salmonel- la enteritidis and S. typhimurium being of most concern. Foods of animal origin, particularly, meat and eggs, were most often implicated. Desserts, ice cream and confectionery items were products also mentioned, but some of these would have egg as a raw or incompletely cooked ingredient. Incidents most frequently occurred in homes or restaurants, and the main factors contribut- ing to outbreaks were poor temperature control in pre- paring, cooking and storing food. Clostridium botuli- num, Salmonella and VTEC are more frequently docu- mented in industrialized than in developing countries. ETEC, EPEC, Shigella, Vibrio cholerae and parasites are the main scourges in developing countries, but it is uncertain how many cases are attributed to food, to water or to person-to-person transmission. The appar- ent decrease of S. aureus and C. perfringens outbreaks in industrialized countries may be related to improved temperature control in the kitchen. An increasing number of illnesses are international in scope, with contamination in a commercial product occurring in one country and affecting persons in sever- al other countries, or tourists being infected abroad and possibly transmitting the pathogen to others at home. For Salmonella, a rapid alert and response coordination is being encouraged through Salm-Net and other inter- national programs. However, unless such a network is worldwide, tracking clusters of illnesses is going to fall on the countries where the first cases occur, and some of these have very limited resources for investigation and control. It was heartening to see funds recently being allocated to foodborne disease surveillance and control in several industrialized countries, but the same commitment is required by the World Health Organiza- tion for the international community. Resume Epidemiologie des maladies d'origine alimentaire: situation mondiale Les taxi-infections alimentaires aigu~s sont actuelle- ment beaucoup plus preoccupantes pour les pouvoirs publics et l'industrie alimentaire qu'il y a quelques decennies. Parmi les facteurs a l'origine de cette situa- tion figurent la decouverte de nouveaux agents patho- genes provoquant des maladies parfois mortelles, !'as- sociation nouvelle d'agents traditionnels avec des ali- ments qui ne posaient jusqu'alors pas de problemes, le nombre croissant de flambees rapportees, l'impact des maladies d'origine alimentaire sur les enfants, les per- sonnes Agees et les personnes immunodeprimees, la demande, par les populations migrantes, de leurs ali- ments traditionnels sur leurs lieux d'etablissement. la facilite d'expedition des aliments frais et congeles, et le developpement de nouvelles industries alimentaires, comme !'aquaculture. Cependant, pour suivre valable- ment !'augmentation ou la diminution de ces maladies, ii faut un systeme efficace de surveillance aux niveaux local, national et international. Actuellement, la plupart Rapp. trimest. statist. sanit. mond., 50 (1997) des regions et des pays manquent de ressources a cette fin, et la situation est le plus souvent connue par des mecanismes de notification passifs. Les donnees des isolements realises au laboratoire et les rapports concernant les maladies a declaration obligatoire sont utiles pour suivre en temps voulu les modifications du nombre de cas de certaines maladies intestinales, mais ne donnent en general aucune indication sur les causes de ces tendances. Les etudes epidemiologiques spe- ciales sont utiles dans la region consideree, mais leurs resultats peuvent ne pas etre extrapolables a d'autres regions ou pays. Les investigations sur les flambees nous apprennent qu'une serie de circonstances a con- duit a la maladie et qu'une nouvelle flambee pourrait survenir dans des circonstances similaires mais non necessairement identiques. Souvent, des programmes de lutte ant ete decides au vu des conclusions d'inves- tigations menees sur des flambees. Malheureusement, dans nombre de cas, !'association agent pathogene/ aliment ayant conduit a la maladie n'avait pas ete prevue d'apres les bases de donnees existantes et ii ne fait pas de doute que les agents de ces maladies continueront a prendre de court les organismes de contrOle des denrees alimentaires. Neanmoins, les donnees recueillies a l'echelle mondiale font ressortir certains points communs. Salmonella est toujours le plus important facteur de maladies aigues d'origine alimentaire, Salmonella enteritidis et S. typhi- murium etant les especes les plus dangereuses. Les aliments d'origine animale, en particulier la viande et les reufs, sont le plus souvent impliques. Les desserts, cremes glacees et pAtisseries sont egalement mention- nes, mais certains d'entre eux contiennent des reufs crus ou incompletement cuits. Les incidents surviennent le plus frequemment en milieu familial ou dans les restau- rants; le principal facteur contribuant a !'apparition de flambees est l'insuffisance du contrOle de la temperatu- re au cours de la preparation, de la cuisson et de la conservation des aliments. Clostridium botulinum, Sal- monella et Escherichia coli verotoxinogene sont plus souvent signales dans les pays industrialises que dans les pays en developpement. Dans ces derniers, les principaux agents en cause sont E. coli enterotoxino- gene et E. coli enteropathogene, Shigella, Vibrio chole- rae et les parasites, mais on ne sait pas exactement quelles sont les proportions de cas imputables aux aliments, a l'eau ou a une transmission interhumaine. La diminution apparente des flambees dues a Staphylo- coccus aureus et a C. perfringens dans les pays indus- trialises pourrait etre liee a une amelioration du contrOle de la temperature dans les cuisines. De plus en plus de maladies ant une portee internatio- nale : un produit commercial peut etre contamine dans un pays et provoquer des maladies dans plusieurs autres, ou des touristes peuvent etre infectes a l'etran- ger et transmettre l'agent pathogene a d'autres person- nes une fois rentres chez eux. Pour Salmonella, Salm- Net et d'autres programmes internationaux encoura- gent l'alerte rapide et la coordination de la reponse en cas de flambee. Cependant, tant qu'un tel reseau ne sera pas mondial, le depistage des groupes de cas incombera aux pays ou surviennent les premiers cas, pays dont certains ne disposent que de moyens tres limites d'investigation et de lutte. II est encourageant de Wld h«h statist. quart., 50 (1997) constater que des fonds ant recemment ete alloues a la surveillance des maladies d'origine alimentaire et a la lutte contre ces maladies dans plusieurs pays industria- lises, mais !'Organisation mondiale de la Sante aurait besoin d'un meme engagement a l'echelle de !'en- semble de la communaute internationale. References/References 1. Tauxe, R.V. Salmonella. a postmodern pathogen. Journal of food protection, 54: 563-568 ( 1991). 2. Todd, E.C.D. Surveillance of foodborne disease. In: Hui, Y.H. et al. Foodbarne disease handbook, Vol. 1, Diseases caused fyy bacteria, 461-536. New York, Marcel Dekker, 1994. 3. Todd, E.C.D. Worldwide surveillance of foodborne disease: the need to improve.Journalof food protection, 59: 82- 92 (1996). 4. Borgdorff, M.B. &: Motarjemi, Y. Surveillance of foodborne diseases: what are the opinions? World health statistics quarterly, 50 (1/2): 12-23 (1997). 5. Bryan, F.L et al. Surveillance offoodborne disease. Part II. Summary and presentation of descriptive data and epidemiologic patterns; their value and limitations.Journal of food protection, 60: 567-578 (1997). 6. Bryan, F.L. etal. Surveillance offoodborne disease. Part III. Summary and presentation of data on vehicles and contributory factors. Journal of food protection, 60: 701-714 (1997). 7. Guzewich, J.J. et al. Surveillance of foodborne disease. Part I. Purposes and types of surveillance systems and networks. Journal of food protection, 60: 555-566 ( 1997). 8. Todd, E.C.D. et al. Surveillance of foodborne disease. Part IV. Dissemination and uses of surveillance data. Journal of food protection, 60: 715-723 ( 1997). 9. WHO surveillance programme for control of foodbarne infections and intoxications in Europe. Fifth report, 1985-1989, 1-235. Institute of Veterinary Medicine, Berlin ( 1992). 10. WHO surveillance programme for control of foodborne infections and intoxications in Europe. Sixth report, 1990-1992, 1-340. Berlin, Institute of Veterinary Medicine ( 1996). 11. Advisory Committee on the Microbiological Safety ofF ood. &port on verocytotoxin-producing Escherichia col~ 1-148. London, HMSO (1995). 12. Loewenherz-Luning, K. et al. Untersuchungen zum vorkommen von Campywbacter jejuni in verschiedenen lebensmitteln tierischen ursprungs. Fl,eischwirtschaft, 76: 958-961 (1996). 13. Schmid, H. et al. Risk factors for sporadic salmonellosis in Switzerland. European journal of clinical microbiowgy and infectious diseases, 15: 725-732 (1996). 14. Bielaszewska, M. et al. Verocytotoxin-producing Escherichia coli in children with hemolytic uremic syndrome in the Czech republic. Clinical nephrology, 46: 42-44 (1996). 15. van de Kar, N. C. A. J. et al. Verocytotoxin-producing Escherichia coli infection in hemolytic uremic syndrome in part of western Europe. European journal of pediatrics, 155: 592-595 (1996). 16. Reilly, W.J. VerotoxigenicEscherichiacoli0157 in Scotland. Proceedings of the Soci.ety for Veterinary Epidemiowgy and Preventative Medicine, Glasgow, March 27 -29, 1996, 60-71 (1996). 17. Fisher, I. S. T. Salmonella enteritidis and S. typhimurium in Western Europe for 1993-1995: a surveillance report from Salm-Net. Eurosurveillance, 2(1): 4-6 (1997). 18. Usera, M. A. et al. Interregional foodborne salmonellosis outbreak due to powdered infant formula contaminated with lactose-fermenting Salmonella virchow. European journal of epidemiowgy, 12: 377-381 (1996). 47 19. Barnes, G. H. &: Bousfield, I. Body builders, raw eggdrink, and Salmonella enteritidis. Communicable disease report, 6:R67 (1996). 20. Lehrnacher, A. et al. Nationwide outbreak of human salmonellosis in Germany due to contaminated paprika and paprika-powdered potato chips. Epidemiology and infection, 115: 501-511 (1995). 21. Pena, A. A. et al. Scombroid fish poisoning outbreak by tuna ingestion in Castellon, Spain. Medicina clinica (Barcelona), 107: 645-648 ( 1996) . 22. Simini, B. Outbreak of foodborne botulism continues in Italy. Lancet, 348: 813 (1996). 23. Cowden, J. M. Scottish outbreak of Escherichia coli 0157 November-December 1996. Eurosurveillance, 2: 1-2 (1997). 24. The Pennington Group. Interim report and priority recommendations. Investigation into the outbreak of E.coli 0157:H7 infection in Central Scotland. Scottish Office, Edinburgh (1997). 25. Melbak, K. et al. Bacterial contamination of stored water and stored food: a potential source of diarrhoeal disease in West Africa. Epidemiology and infection, 102: 309-316 (1989). 26. Ekanem, E. E. etal. Food hygiene behaviour and childhood diarrhoea in Lagos, Nigeria: a case-control study. Journal of diarrhoeal diseases research, 9: 219-226 (1991). 27. Simango, C. &: Rukure, G. Survival of bacterial enteric pathogens in traditional fermented foods.Journal of applied bacteriology, 73: 37-40 (1992). 28. Odugbemi, T. et al. Study on the pH of ogi, Nigerian fermented weaning food, and its effect on entero- pathogenic Escherichia coli, Salmonella typhi and Salmonella paratyphi. Journal of tropical medicine and hygiene, 94: 219-223 (1991). 29. De Vries, H. B. et al. Aflatoxins and kwashiorkor in Kenya: a hospital based study in a rural area of Kenya. Annals of tropical paediatrics, 7: 249-257 (1987). 30. Ozturk, M.etal. p53 mutation in hepatocellularcarcinoma after aflatoxin exposure. Lancet, 338: 1356-1359 ( 1991). 31. Abdel-Hameed, A. A. Sero-epidemiology of toxoplasmosis in Gezira, Sudan. Journal of Tropical medicine and hygiene, 94: 329-332 (1991). 32. Mohd, M. G. Brucellosis in the Gezira area, Central Sudan. Journal of tropical medicine and hygiene, 92: 86-88 ( 1989). 33. Rakha, M. et al. Foodborne disease outbreaks due to specific bacteriologic etiologies in Egypt. Egyptian journal of microbiology, 25: 1-8 ( 1990) . 34. EI-Sherbeeny, M.R. et al. Microbiological profiles of foods served by street vendors in Egypt. International journal of food microbiology, 2: 355-364 ( 1985). 35. Ndyabahinduka, D. G. K. et al. An outbreak of human gastrointestinal anthrax. Annali dell'Istituto Superiore de Sanita, 20: 205-208 (1984). 36. Rwiza, H. T.Jimson weed poisoning: an epidemic at Usangi rural government hospital. Tropical and geographical medicine, 43: 85-90 (1991). 37. Weber, J.T. et al. A massive outbreak of type E botulism associated with traditional salted fish in Cairo. Journal of infectious diseases, 167: 451-454 (1993). 38. Abdul-Raouf, U.M. et al. Isolation of Escherichia coli 015 7: H7 from some Egyptian foods. International journal of food microbiology, 29: 423-426 (1996). 39. Moren, A. et al. Practical field epidemiology to investigate a cholera outbreak in a Mozambican refugee camp in Malawi, 1988. Journal of tropical medicine and hygiene, 94: 1-7 (1991). 40. Paquet, C. etal. Aetiology of haemorrhagic colitis epidemic in Africa. Lancet, 342: 175 (1993). 41. Tsang, R.S.W &: Yung, R.W.H. Relative frequency and antimicrobial susceptibility of gastrointestinal bacterial pathogens. Laboratory medicine, 22: 793-797 (1991). 42. Jacobs, B.C. et al. Campylobacter jejuni infections and Guillain-Barre syndrome. Annals of neurology, 40: 181-187 (1996). 48 43. Bunning, V. Kelly et al. Chronic health effects of microbial foodborne disease. World health statistics quarterly, 50 ( 1 /2): 51-56 (1997). 44. Buzby,Jean C.&: and Roberts, T. Economic costs and trade impacts of microbial foodborne illness. World health statistics quarterly, 50 (l/2): 57-66 (1997). 45. Hariharan, H. et al. Detection of Campylobacterjejuni/C. coli infection in patients with Guillain-Barre Syndrome by serology and culture. Microbiologica, 19: 267-271 (1996). 46. Outbreaks due to microbial toxins in wheat and fish. Weekly epidemiological record, 64: 145-146 ( 1989) . Flambees dues a des toxines microbiennes dans le hie et le poisson. Rel£ue epidemiologique hebdomadaire, 64:145-146 (1989). 47. Sinha, K.K. Mycotoxin in maize in North Bihar, India. Newsletter, WHO Surveillance Programme for Control of Foodborne Infections and Intoxications in Europe. Berlin, Institute of Veterinary Medicine, 23: 6 (1990). 48. Microbial toxins in wheat and fish in India. Newsletter, WHO Surveillance Programme for Control of Foodborne Infections and Intoxications in Europe. Berlin, Institute of Veterinary Medicine, 23: 5-6 (1992). 49. Pathak, K.M.L Taenia solium cysticercosis in India: a meatborne disease. In: Proceedings of the 3rd World Congress of Foodborne Infections and Intoxications,June 16-19, 1992, Berlin. Berlin, Institute of Veterinary Medicine, 171 (1992). 50. Lee, W.-C. et al. An epidemiological study of food poisoning in Korea and Japan. International journal of food microbiology, 29: 141-148 (1996) . 51. Pan, T.-M. et al. Food-borne disease outbreaks in Taiwan. Journal of the Formosan Medical Association, 95: 417-420 (1996). 52. Nguyen van Xoang. Epidemiology of foodborne infections and intoxications. In Proceedings of the 3rd World Congress of Foodborne Infections and Intoxications,June 16-19, 1992, Berlin. Berlin, Institute of Veterinary Medicine, 184 (1992). 53. Bui Minh Due et al. A review of the study of food hygiene quality and foodborne infections and intoxications in Vietnam. In: Proceedings of the 3rd World Congress ofFoodborne Infections and lntoxications,June 16 - 19, 1992, Berlin. Berlin, Institute of Veterinary Medicine, 55 (1992). 54. Swaddiwuthipong, W. et al. Surveillance of food poisoning outbreaks in Thailand, 1981-1986. Southeast Asian journal of tropical medicine and public health, 19: 327-331 (1988). 55. Swaddiwudhipong, W. et al. Foodborne disease outbreaks of chemical etiology in Thailand, 1981-1987. SoutheastAsian journal of tropical medicine and public health, 20: 125-132 (1989). 56. Rampal, LA food poisoning outbreak due to Staphylococcus aureus, Kapar, Malaysia, 1983. Medical journal of Malaysia, 38: 294-298 (1983). 5 7. Lim, Y S. &: Tay, L. A one-year study of enteric Campylobacter infection in Singapore. Journal of tropical medicine and hygiene, 95: 119-123 (1992). 58. Muller, R. et al. Human cysticercosis and intestinal parasitism amongst the Ekari people oflrian]aya.Journalof tropical medicine and hygiene, 90: 291-296 (1987). 59. Labbe, R. Clostridium perfringens. In: Doyle, M.P. (ed), FoodbomeBacterialPathogens. New York, Marcel Dekker Inc., 191-234, 1989. 60. Shekhar, S.R. et al. Foodborne disease outbreaks in the twin cities of Hyderabad and Secunderabad (India) during 1984 to (1989). In: Proceedings of the 3rd World Congress on Foodborne Infections and Intoxications, June 16-19, 1992, Berlin. Berlin, Institute of Veterinary Medicine, 54 (1992). 61. Rao, R.N. et al. A study of recorded cases of foodborne diseases at Hyderabad during 1984 and 1985. Journal of tropical medicine and hygiene, 92: 320-324 ( 1989). 62. Halliday, M.L Hepatitis A epidemic in Shanghai triggered by raw clams. Journal of infectious diseases, 164: 852-859 (1991). Rapp. trimest. statist. sanit. mond., SO (1997) 63. Cheng, X.K. et al. An epidemic of foodborne hepatitis A in Shanghai. In: Proceedings of the 3rd World Congress on Foodborne Infections and Intoxications, June 16-19, Berlin. Berlin, Institute ofVeterinary Medicine, 119 (l 992). 64. Gao, Q.Y. et al. A review of botulism in China. Bio=dical environ~tal science, 3: 326 - 336 ( 1990). 65. Chuang, Y.-C. et al. Vibrio vulnificus infection in Taiwan: report of28 cases and review of clinical manifestations and treatment. Clinical infectious diseases, 15: 271-276 (1992). 66. Enterohaemorrhagic Escherichia coli infection. Weekly Epidemiological Record, 71: 267-268 (1996). Infection due a Escherichia coli enterohemorragique. Rel£ve epidemiologiquehebdomadaire, 71: 267-268(1996). 67. Health and Welfare Statistics in Japan 1992, Health and Welfare Statistics Association, Ministry of Health and Welfare, Tokyo.Japan (1992). 68. Cholera associated with seafood. Newsletter, WHO surveillance program= for control of foodborne infections and intoxications in Europe. Berlin, Institute of Veterinary Medicine, 31: 4 (1992). 69. Sudhakar, P. et al. The economic impact of a foodborne disease outbreak due to Staphylococcus aureus. journal off ood protection, 51: 898-900 (1988). 70. Teufel, P. et al. Risks of salmonellosis and staphylococcal food poisoning from Pakistani milk-based confectioneries. Journal of food protection, 55: 588-594 (l 992). 71. Bryan, F.L et al. Hazards and critical control points of vending operations at a railway station and a bus station in Pakistan. journal off ood protection, 55: 534-541 ( 1992). 72. Bryan, F.L etal. Hazards and critical control points of food preparation and storage in homes in a village and a town in Pakistan. Journal off ood protection, 55: 714-72 l ( 1992). 73. Singh,J. et al. Outbreak of soya milk poisoning in Dehli. Journal communicable diseases, 23: 167 ( 1991). 74. umabadusuriya, S.P. et al. An outbreak of salmonellosis following consumption of monkey meat. Journal of tropical =dicine and hygi.ene, 95: 292-295 ( 1992). 75. Ramamurthy, V.D. Seafood infections in Indian Ocean region. In: Proceedings of the 3rd World Congress on Foodborne Infections and Intoxications, June 16-19, 1992, Berlin. Berlin, Institute of Veterinary Medicine, 95 (1992). 76. Baker, M. &: Roberts, A. A new schedule of notifiable diseases for New Zealand. New Zealand public health report, 3(5): 33-37 (1996). 77. Listeriosis in 1995. New Zealand public health report, 3: 53 (1996). 78. Risk factors for campylobacteriosis identified in study. New Zealand public health report, 3: 20, ( 1996). 79. Crerar, S.K. et al. Foodborne disease: current trends and future surveillance needs in Australia. Medical journal of Australia, 165: 672-675 (1996). 80. Ng, S. et al. Human salmonellosis and peanut butter. Communicable disease intelligence, 20: 326 (1996). SI. Cameron, S. et al. Enterohaemorrhagic Escherichia coli outbreak in South Australia associated with the consumption of mettwurst. Communicable disease intelligence, 19: 70-71 (1995). 82. Mizusawa, Y. et al. Survey of haemolytic-uraemic syndrome in Queensland 1979-1995. Medical journal of Australia, 165: 188-191 (1996). 83. McCall, B. et al. An outbreak of Escherichia coli 0157 infection on the Gold Coast. Communicable diseases intelligence, 20: 236-239 ( 1996). 84. Mitchell, P. et al. Outbreak of Salmonella in Christchurch bakery. New Zealand public health report, l: 27 (1994). 85. Jones, N. et al. Outbreak of food poisoning at Aukland wedding. New Zealand public health report, l: 3-4 (1994). 86. Cavaney, N. &: Mamood, F. Outbreak at food fashion show. New Zealand public health report, l: 27 ( 1994). 87. O'Hallahan, J. An outbreak of foodborne hepatitis A associated with a Wellington delicatessen. New Zealand public health report, 3: 41-4 (1996). Wld hlth statist. quart., 5D (1997) 88. Enriquez, C.E. et al. Bacteriological survey of used cellulose sponges and cotton dishcloths from domestic kitchens. Dairy, food and enuiron~tal sanitation, 17: 20-24 (1997). 89. Wilson, J.B. et al. Vero cytotoxigenic Escherichia coli infection in dairy farm families. Journal of infectious diseases, 174: 1021-1027 (1996). 90. Johnson, R.P. et al. Growing concerns and recent outbreaks involving non-0157:H7 serotypes of verotoxigenic Escherichia coli. Journal of food protection, 59: 1112-1122 (1996). 91. Todd, E.C.D. &: Chatman, P. Annual summaries of foodborne and waterborne disease in Canada, 1990 and 1991, 1-287. Morin Heights, Quebec, Polyscience Publications, Inc., 1997. 92. Bean, N.H. et al. Surveillance for foodborne disease outbreaks - United States, 1988-1992. Morbidity and mortality weekly report, 45, SS-5: 1-68 ( 1996). 93. Cieslak, P.R. et al. Preventable disease in correctional facilities. Desmoteric foodborne outbreaks in the United States, 1974-1991. Archives of internal =dicine, 156: 1883- 1888 (1996). 94. Hennessey, T.W. et al. A national outbreak of Salmonella Enteriditis infections from ice cream. New England journalof =dicine, 334: 1281-86 (1996). 95. Todd, E.C.D. Preliminary estimates of costs of food borne disease in Canada and costs to reduce salmonellosis.Journal of food protection, 52: 586-594 ( 1989). 96. Todd, E.C.D. Preliminary estimates of costs of food borne disease in the United States.Journal of food protection, 52: 595- 601 (1989). 97. Townes, J.M. et al. An outbreak of type A botulism associated with a commercial cheese sauce. Annals of internal =dicine, 125: 558-563 ( 1996). 98. Millard, P.S. et al. An outbreak of cryptosporidiosis from fresh-pressed apple cider. Journal of the American Medical Association, 272: 1592-1596 ( 1994). 99. Besser-Wiek, J.W. et al. Foodborne outbreak of diarrhcal illness associated with Cryptosporidium parvum- Minnesota, 1995. Journal of the American Medical Association, 276: 1214 (1996). 100. Imbiscuso, L. E. coli 0157:H7 outbreak associated with a church supper. Public health and epidemiology reports Ontario, 445-447, Dec 20 (1996). IOI. LeBer, C.A. Reported cases/outbreaks of verotoxin- producing E. coli in Ontario, 1987 to 1995. Public health and epidemiology reports Ontario, 448-449, Dec 20 (1996). 102. Killalea, D. et al. International epidemiological and microbiological study of outbreak of Salmonella agona infection from a ready to eat savoury snack- I: England and Wales and the United States. British medical journa~ 313: l 105-ll07 (1996). 103. Shohat, T. et al. International epidemiological and microbiological study of outbreak of Salmonella agona infection from a ready to eat savoury snack- II: Israel. British =dicaljourna~ 313: 1107-1109 (1996). l 04. Prim era reunion de la Red Latinoamericana de Vigilancia Epidemiologica de las Enfermedades. VETA II. Informe final. PAHO, 1-35 (1990). 105. Madico, G. et al. Active surveillance for Vibrio cholerae Ol and vibriophages in sewage water as a potential tool to predict cholera outbreaks. Journal of clinical microbiology, 34: 2968-2972 (1996). 106. Cholera situation in the America. Update number 13. Communicable Disease Program, Division of Disease Prevention and Control, PAHO, Washington, DC, 15 July, 1995. l 07. Tamplin, M. &: Parodi, C.C. Environmental spread of Vibrio cholerae in Peru. Lancet, 338: 1216-1217 ( 1991). 108. Ries, A.A. et al. Cholera in Piura, Peru: a modern urban epidemic. Journal of infectious diseases, 166:1429-1433 (1992). 49 109. Finelli, L. et al. Outbreak of cholera associated with crab brought from an area with epidemic disease. journal of infectious diseases, 166:1433-1435 (1992). 110. Ship ballast, holding tank waters checked for cholera. Food chemical news, 25:40 (1991). 111. Cooper, G. et al. Cholera associated with international travel, 1992. Morbidity and mortality weekly reports, 41: 664-667 (1992). 112. Petrera, M. & Montoya, M. The economic impact of the cholera epidemic, Peru, 1991. Epidemiowgical bulletin, PAHO, 13(3): 9-12 (1992). 113. Long, K.Z. Multi-state competing risk hazards analysis of feeding patterns and enteric infections due to entero- toxigenic Escherichia coli in Mexican infants. FASEB Journal, 9(4): A732 (1995). 114. Torres-Vitella,M.A.R.etal. Riskofsalmonellosisassociated with consumption of chocolate in Mexico. Journal of food protection, 58: 4 78-481 (1995). 115. Primera reunion <lei Gruppo de Trabajo sabre Vigilancia Epidemiol6gica de las Enfermedades Transmitidas por Alimentos. VETA I. Informe final. PAHO, 1-52 (1989). 116. Eiguer, T. et al. Importancia de la Salmonella enteritidis en brotes de enfermedades transmitidas por alimentos en Argentina, anos 1986-1988. Re11ista argentina de microbiologia, 22: 41-46 (1990). 117. Lopez, E.L etal. Hemolytic uremicsyndrome anddiarrhea on Argentine children: the role of Shiga-like toxins.journal of infectious diseases, 160: 469-475 (1989). 118. Bergdoll, M.S. et al. Staphylococcal food poisoning in Brazil. In: Proceedings of the 3rd World Congress on Foodborne Infections and Intoxications, June 16-19, Berlin. Berlin, Institute of Veterinary Medicine, 320-323 (1992). 119. Pereira, M.L et al. Enterotoxin Hin staphylococcal food poisoning. Journal of food protection, 59: 559-561 ( 1996). 120. Bryan, F.L. et al. Hazard analyses of foods prepared by inhabitants near Lake Titicaca in the Peruvian Sierra. Journal of food protection, 51: 412-418 ( 1988). 121. Bryan, F.L. et al. Critical control poinL~ of street-vended foods in the Dominican Republic.Journal of food protection, 51: 373-383 (1988). 50 122. Miehanie,S.etal. Critical control points for foods prepared in housholds in which babies had salmonellosis. International journal of microbiology, 5: 337-354 ( 1987). 123. Hauschild, A.H.W. Epidemiology of botulism. In: Hauschild, A.H.W. & Dodds, K.L. Clostridium botulinum: ecology and control in foods, 60-104. New York, Marcel Dekker, 1993. 124. Quevedo, F. et al. Several clusters of foodborne human botulism: first reported outbreak in Peru. In: Proceedings of the 3rd World Congress onFoodborne Infections and Intoxications, June 16-19, Berlin. Berlin, Institute ofVeterinary Medicine, 1098 (1992). 125. Todd, E.C.D. How ciguatera affects Canadians. In: Tosteson, T.R. (ed) Proceedings of the 3rd International Conference on Ciguatera Fish Poisoning, Puerto Rico, 1990, 181-195, Quebec, Polyscience Publications, 1992. 126. Alvarez, P. et al. In: Tosteson, T. R. ( ed) Proceedings of the 3rd International Conference on Ciguatera Fish Poisoning, Puerto Rico, 1990. Polyscience Publications, Quebec, 169-179, 1992. 127. Frenette, C. et al. A large common-source outbreak of ciguatera fish poisoning. Journal of infectious diseases, 158: 1128-1130 (1988). 128. de Campos, M. & Canahui, E. Paralytic shellfish poisoning in Central America and Mexico. Proceedings of the 3rd World Congress on Foodborne Infections and Intoxications,.fune 16-19, Berlin. Berlin, Institute of Veterinary Medicine, 391-394, 1992. 129. Rodrigue, D.C. et al. Lethal paralytic shellfish poisoning in Guatemala. American Journal of tropical medicine and hygiene, 42: 267-271 (1990). 130. Todd, E.C.D. et al. Scombroid poisoning - an outbreak in two Ontario communities. Canada communicable disease report, 18-3: 17-19 (1992). 131. Morrow,J.D. etal. Evidence that histamine is the causative agent of scombroid-fish poisoning. New England journal of medicine, 324: 716-720 (1991). 132. Burros M. Clinton to fight foodbomc illness. New York Times, p Al.Jan. 25 (1997). 133. Summary of notifiable diseases, United States 1995. Morbidity and mortality weekly report, 44 (53): 1-88 (1996). Rapp. trimest. statist. sanit. mond., 50 (1997) Chronic health effects of microbial foodborne disease V. Kelly Bunninga, James A. lindsayb & Douglas L. Archerb Introduction As the incidence of foodborne microbial disease increases, chronic disease sequelae may also be expected to rise (I, 2). Archer and Kvenberg (3) estimated that chronic, secondary after-effect ill- nesses may occur in 2-3% of cases of foodborne infections, and suggested that the long-term conse- quences to human health and the economy may be more detrimental than the acute disease. Except in rare circumstances, chronic complications to food- borne infections are unlikely to be identified, or epidemiologically linked to a foodborne cause, simply because these data are not systematically collected(], 4). Moreover, host symptoms induced by a specific pathogen are often wide-ranging and overlapping, and therefore difficult to link tempo- rally to a foodborne illness incident. These imped- iments manifest themselves because the pathology of chronic disease can result from an infection without overt illness (5). Alternatively, the chronic sequelae may be unrelated to the acute illness, and occur despite the immune system successfully elim- inating the primary infection. This suggests that activation of the immune system can initiate the chronic condition due to an autoimmune re- sponse (1). Several microbial pathogens are highly adapted to parasitization, exhibiting environmentally re- sponsive and adaptive traits which allow attach- ment, invasion, and replication in the host (6, 7). Microbial pathogenicity should however, not be viewed unidimensionally, that is, from only the microbe's perspective (2). A major selective force that regulates the phenotype of an infecting micro- bial pathogen population is the host immune sys- tem, which is also highly adaptive, especially in discriminating "self' and "non-self' antigens ( 7, 8). When holistically examining the host-parasite rela- tionship, apparent mechanisms that "successful" pathogens have evolved to elude the immune sys- tem include: (z) antigenic heterogeneity or varia- tion, (ii) sequestration, either intracellularly or in certain specific host sites, (iii) molecular mimicry, through either imitation (cross-reaction) or ad- a Division of Virulence Assessment, Center for Food Safety and Applied Nutrition, Food and Drug Administration, Laurel, MD, United States of America. b Department of Food Science and Human Nutrition, University of Florida, Gainesville, FL, United States of America. Wld hlth statist. quart., 5D (1997) sorption of host protein, and (iv) direct immune stimulation and/ or suppression (2, 9). How food- borne pathogens serve as "triggers" in chronic dis- ease pathology, within the context of mechanistic theories is discussed, and the risk-assessment para- digm will be evident(]). Rheumatoid disease Septic arthritis Several bacterial pathogens induce septic arthritis by haematogenous spread to the synovial space resulting in inflammation. By definition, viable or- ganisms are recoverable from synovial fluid, and treatment usually involves antibiotic therapy. The subsequent prognosis is dependent on host factors and virulence of the organism. Either complete resolution or permanent joint damage can occur (2, 5). Aseptic arthritis Several Gram-negative species of food borne patho- gens appear to initiate aseptic or reactive arthritis (ReA), an acute, non-purulentjoint inflammation following infection elsewhere in the body, for ex- ample the bowel. Although a distinct clinical dis- ease, ReA also occurs in the Reiter's Syndrome (RS) triad with conjunctivitis and uveitis. A subset of patients with ReA and RS symptoms develop ankylosing spondylitis (AS), a rheumatoid inflam- mation of synovial joints and en theses within and distal to the spine. The relative risk of developing these sero-negative spondyloarthropathies after a Gram-negative enterobacterial infection is high for individuals positive for the MHC antigen B27 and the cross-reacting MHC B7 group. These chronic complications, then, are based on a genetically determined host risk factor in combination with an environmental trigger (2, 5, 9, 10). Implicated pathogens for ReA include Y. entero- colitica, Y. pseudotuberculosis, Shigella flexneri, Sh. dys- enteriae, Salmonella spp., Campywbacter jejuni, Es- cherichia coli, and Klebsiella pneumoniae (2, 5) al- though the latter bacterium is more often con- nected only with faecal carriage by AS probands. No cause and effect relationship of enteric patho- gens in AS has been established (11); however, a low but consistent incidence (0.2-2.4%) of ReA occurs after outbreaks of S. typhimurium, Sh. flexneri and C. jejuni. Biotypes and phage types of Y. entero- 51 colitica 0:3 and 0:9, endemic to Scandinavia, ap- pear to be either highly "arthritogenic" or affect a more genetically predisposed population with per- sistent and debilitating symptoms which may last for years (2, 5). The sharing of antigenic determinants by a microbe and its host is a frequent natural occur- rence, and bacterial antigens from the aforemen- tioned pathogens that directly cross-react with MHC B27 have been demonstrated (12, 13). Addi- tionally, the plasmid-mediated synthesis of bacteri- al B27 "modifying factor" a protein that binds to, and subsequently alters the conformation of B27 has been reported (14). In both of these models, immune recognition of the foreign antigen leads to an autoimmune anti-B27 response. Alterna- tively, B27 may act non-immunologically, because dissemination of bacterial antigens to infected joints stimulates a local T-cell inflammatory re- sponse. Here, B27 may act as a receptor for bacte- ria or antigens thereof, facilitating invasiveness from mucosal surfaces in the gut (13). Indeed, transfected B27 on the surface of mouse L-cells reportedly altered bacterial invasion capability (15). There are several caveats to the above hypothe- sis for explaining the role of microbial pathogens in stimulating rheumatoid disease. Despite the strong familial association related to the MHC B27 gene, B27-negative individuals are known to devel- op disease, albeit less often, but with apparently equal severity as evidenced by an epidemiological investigation of RA following the 1985 S. typhimu- rium gastroenteritis outbreak due to contaminated milk (16). Unfortunately, the microbial antigen/ organism persistence model suffers from the fail- ure to either culture the microorganism, or detect its DNA by polymerase chain reaction (PCR) in affected tissues and/ or synovial fluid. The persis- tence of bacterial antigen, detected in a percent- age of probands is plausible, but a model to ex- plain the development of chronic arthritis remains unproven (12). Inflammatory bowel disease (IBD) IBD is the collective term for Crohn's disease (CD) and ulcerative colitis (UC). While both infections are chronic inflammatory diseases exhibiting histo- logical infiltrates of macrophages and lympho- cytes, coupled with a prolonged clinical course, the primary clinical and pathological effects are gas- trointestinal. The two infections can be difficult to discriminate since the disease conditions are often similar. Familial and nationality associations are strong, suggesting genetic predisposition ( 17). The acute clinical characteristics are diarrhoea, abdom- inal pain, fever, and weight loss, and the acute pathology reveals a constant flux of neutrophils into inflamed mucosa eventually penetrating the epithelium into the intestinal lumen. CD involves 52 the ileum and/or colon, while UC appears re- stricted to the colon. Although the etiology of IBD, and the mechanism(s) for spontaneous exacerbations and remissions are unknown, much research has fo- cused on transmissible agents, including food- borne pathogens. An association between bacterial L-forms and IBD has been sporadically reported, with isolation of Pseudomonas, Mycobacterium, Strep- tococcus fecalis and E. coli from affected tissue, but not from appropriate controls. Currently, an open debate exists as to whether L-forms are pathogenic in humans or persist in affected tissue (18). Mycobacterium paratuberculosis, the etiological agent of Johne's disease in ruminants, is possibly associated with CD, through the production of L- forms of the bacterium. The model suggests that a susceptible human neonate first contacts the or- ganism after ingestion of commercial dairy prod- ucts. Subclinically infected cows are known to shed M. paratuberculosis, and the organism has been identified in pasteurized milk using PCR specific for the M. paratuberculosis insertion sequence IS900. Infection of the human neonatal host in- vokes an antigen-poor (by virtue of lacking a cell wall) L-form which grows slowly and persists in the lamina propria, stimulating a chronic low-grade inflammation. The immune response increases in severity over a period of years without bacterial replication, ultimately producing the pathology of CD (19, 20). Since the association of CD with M. paratuberculosis is not absolute, the model is, and will likely remain unproven. Other current models favour nutritional factors or autoimmune phenomenon mediated by alterations in inflamma- tory cytokine profiles ( 17). Recent immunocytochemical techniques were used to demonstrate antigens to Listeria monocyto- genes, E. coli and Streptococcus species in CD tissues, in 75%, 57% and 44% of probands examined. Macrophages and giant cells immunolabelled for antigen specific to these organisms were found beneath ulcers, around abscesses, along fissures, within the lamina propria, in granulomas, and in germinal centers of mesenteric lymph nodes (21). Superantigens and autoimmunity Most superantigens are protein virulence factors produced by a number of microbial pathogens, and are so named because they elicit extreme im- mune responses. The mechanism whereby super- antigens stimulate T-cells is different from conven- tional antigens. Conventional antigens are pro- cessed by the antigen-presenting cells (APC) into peptides which combine with MHC class II mole- cules for presentation to T-cells. In contrast super- antigens need not be processed by APC in order to combine with MHC class II because they bind at sites distinct from the normal antigen binding site. They interact with the variable side of the Vf3 chain Rapp. trimest. statist. sanit. mond., 50 (1997) of the T-cell receptor (TCR) by recognizing ele- ments shared by a subset of T-cells. This ability to interact with specific VB elements regardless of antigen specificity is unique. Depending on the type of interaction, recognition of a superantigen by a T-cell can have different consequences, in- cluding profileration and expansion, suppression (clonal deletion), or alternately the induction of prolonged unresponsiveness (anergy) or cell death (apoptosis) (22-24). The mechanisms whereby the various alternative responses can result, is un- known. Activation can potentially cause the expansion of naturally occurring autoreactive T-cell clones, which in turn could mediate autoimmune destruc- tion in the host (22-24). Alternatively, autoimmuni- ty could be induced by eliminating regulatory T- cells that modulate self-reactive T-cells. Superanti- gens could stimulate autoantibody by activating normal T-cells and providing a molecular bridge that allows their interaction with autoantibody-pro- ducing B-cells in a manner similar to graft-vs-host disease (12). The expansion of autoreactive B-cells could lead to a secondary stimulation of autoreac- tive T-cells. Superantigens may also trigger autoim- munity by directly facilitating the activation of an- ergic autoreactive T-cells. This could be accom- plished by promotion of cell division forcing unre- sponsive T-cells from this state, or by lowering the threshold level of cryptic self-antigen that is re- quired for autoreactive T-cell stimulation (22-24). Alternatively, interaction with MHC class II mole- cules can activate B-cells and macrophages to pro- duce excessive quantities of inflammatory cyto- kines. Any imbalance in cytokine regulation could provide an environment for the development of autoimmunity (23, 24 ). Superantigens may also act as adjuvants, in that they alone do not trigger the disease, but enhance the development or exacer- bate existing conditions (24). Superantigens from several foodborne bacteria Staphylncoccus, Streptococcus, Yersinia and Clostridium have been isolated and characterized. Many are thought to be associated with several autoimmune disorders, for example, rheumatic heart disease, rheumatoid arthritis, multiple sclerosis, Graves dis- ease, Sjogren's syndrome, autoimmune thyroiditis, psoriasis, Kawasaki disease (23), Crohn's disease (25) and insulin-dependent diabetes mellitus (IDDM) (26). It should be stressed that, while the idea that superantigens have a role in autoimmune disorders is accepted per se, there is no definitive evidence showing that they are responsible for causing human autoimmune dis~rders. The pro- posal is accepted based on extensive animal model studies (12, 22-24), but limited human clinical studies. The following clinical studies serve as exam- ples. While synovial infiltrates from some patients with RA appear to have selective expansion of Wld hlth statist. quart., 50 (1997) T-cells expressing VB 14, other studies showed con- siderable variability in VB expansion (12, 22). This suggested that either a variety of superantigens was involved, or that other factors might influence these T-cell expansions. The link between super- antigens and Kawasaki disease was suggested from a clinical study where 13 of 16 patients grew super- antigen-producing bacteria (Staphylococcus and Streptococcus) from culture, compared to I of 15 from controls (27). The bacteria produced TSST-1 and 2 toxin known to stimulate the expansion of VB2+, which was also expanded in the blood of children with Kawasaki disease. Similarly, in pa- tients with Crohn's disease, VB8+ were significantly expanded compared to controls (25). Finally, in patients with IDDM there was selective expansion of VB7+ when compared to controls (26). That exposure to superantigens can be a triggering event in the pathology of autoimmune diseases is a compelling hypothesis, but it has to be balanced with some caution. In human diseases where super- antigens have been clearly demonstrated as the cause, for example, toxic shock syndrome (TSS), initial T-cell proliferation and TCR-mR.i"\JA up-reg- ulation have been observed, but the long-term sequelae in terms of T cell function are unknown (27). Perhaps the most important consideration is: why do bacteria produce superantigens? It has been suggested that superantigens act as bait, by immuno-compromising the host, in preparation for the pathogen to take over the host (23). This could be achieved by three mechanisms, firstly, the ability to induce superantigen-dependent cellular cytotoxicity; secondly, the overstimulation of in- flammatory cytokines which allow apoptosis in T-cells and thirdly, the generation of a non-specific immune response that diverts attention from what should have been the host's primary response to the pathogen. While these are interesting argu- ments, it is also important to consider that not every superantigen has the same effect, nor does the same superantigen have the same effect on differ- ent individuals. Biological activitiy is dependent on the host, and factors such as age, stress, hormonal changes, and concurrent infections (23). E.coli and the haemolytic uraemic syndrome. Follow- ing colitis caused by E. coli 0157:H7 and other enterohaemorrhagic strains of E. coli, a proportion of individuals, particularly children, will develop haemolytic uraemic syndrome (HUS). HUS is the leading cause of acute renal failure in children. HUS is a worldwide problem which mirrors the distribution of E. coli 0157:H7 and other Shiga and Shiga-like toxin producing microorganisms. Out- breaks of haemorrhagic colitis and subsequent cases of HUS have developed as a result of various food vehicles. Ground beef is still the most com- mon food vehicle, but other food vehicles include lunch meats, potatoes, unpasteurized cow's milk, drinking water, unpasteurized apple cider, mayon- 53 naise (28), parsley contaminated with bovine ma- nure (29), salad contaminated by under-cooked beef (30), and very recently, unpasteurized apple juice (31). The cider, mayonnaise and apple juice outbreaks underscore the acid resistance of the 0157:H7 strain. Besides 0157:H7, other Shiga-like toxin-producing E. coli are also causative of illness and HUS (28). The fact that Citrobacter Jreundii was capable of causing severe gastroenteritis and HUS (29) underscores the ability of the genes respon- sible for the virulent attributes of E. coli O 157:H7 to be transferred to other bacteria. The toxin-mediated damage to the kidneys may not be limited to the glomerular endothelial cells as once thought, but may include the tubular epi- thelial cells (32). The toxic activity of the exotoxins of the 0157:H7 strain was originally defined by cytotoxic activity against the Vero cell line, derived from the kidney of African Green monkeys. This led to the toxin(s)' original designation as verocy- totoxins. Binding of the toxins to Vero cells, and the human cell line Daudi, was shown to be specific for the glycosphingolipid globotriaosylceramide (Gb3), and mutants of the Vero or Daudi lines deficient in Gb3 were not damaged by the toxin(s) (33). Studies conducted with human renal and umbilical endothelial cells showed that the pres- ence of Shiga toxin receptor Gb3 on renal, but not umbilical endothelial cells may account for the differential sensitivity of renal cells to toxin- induced damage (34). Studying human renal sec- tions, and the binding of fluorescein-conjugated VTl, Lingwood (35) showed that Gb3 was present in the glomeruli of infants less than two years old, but not in adult glomeruli, suggesting that the presence of Gb3 in the paediatric renal glomerulus may be a risk factor for the development of HUS. Characterization of the Shiga toxin receptor has led to a potential preventive treatment (36). Guillain-Barre syndrome and Camp,ylobacter. Guil- lain-Barre Syndrome (GBS), an acute imflamma- tory demyelinating polyradiculoneuropathy, fre- quently occurs following infection. It occurs world- wide, and is the most common cause of neuromus- cular paralysis. Rees et al. (37) recently discussed the relationship between one pathogen, Camp,y- lobacter jejuni, and GBS. In a prospective case-con- trol study involving 96 patients diagnosed with GBS, there was either culture or serologic evidence of prior C. jejuni infection in 27 patients, and 19 of the 27 recalled having had diarrhoeal illness within 12 weeks before the onset of neurological symp- toms (37). Compared with household- or age- matched controls, the association of GBS with pre- ceeding C. jejuni infection was highly significant (37). This association has been suspected for over a decade (38). Unusually severe GBS cases were de- scribed in which electrophysiological and patho- logical studies suggested primary axonal degenera- tion (39). Similar patients from Mexico, Spain, 54 South America, Japan, The Republic of Korea, India, and China have been described (40). Cases of this severe form of GBS tend to occur in sum- mer, and have been linked to infection with C.jejuni (41). There is ongoing controversy as to whether GBS is an autoimmune disease (42). Sufficient data exist to classify GBS as an autoimmune disease (42) as 4 major Rose-Witebsky criteria are almost com- pletely met. Studies on the relationship between GBS and C. jejuni have lent validity to the hypothe- sis of molecular mimicry. Peripheral nerves may share epitopes with surface antigens of C. jejuni. This has been supported by studies in which anti- GMl IgG antibodies recognized surface epitopes on intact C. jejuni, and the reaction was strain- specific for certain Penner serotypes (43). In a 32- patient study conducted in northern China, 66% had serological evidence of recent C. jejuni infec- tion, which differed significantly from village con- trols, and anti-GMl IgG antibodies were present in 42% of GBS cases, again, significantly different from village controls ( 44). It is likely that the im- munologic mechanisms at work in GBS triggered by C. jejuni are complex. In the United States, C. jejuni is the leading bacterial cause of diarrhoeal illness, and it should be noted that other "chronic" diseases such as Reiter's syndrome and reactive arthritis are also triggered by C.jejuni (45). Future studies may show these patterns elsewhere around the world. Conclusions In the future, more foodborne pathogens may be linked to chronic disorders. This will include cur- rently known pathogens, and those that have yet to evolve. Summary The acute effects of foodborne disease are sometimes not the end of the illness. Several significant foodborne pathogens are capable of triggering chronic disease, and even permanent tissue or organ destruction, prob- ably via immune mechanisms. Arthritis, septic and reac- tive, inflammatory bowel disease, haemolytic uraemic syndrome, Guillain-Barre syndrome, and possible sev- eral autoimmune disorders can be triggered by food- borne pathogens or their toxins. Research is needed to more fully understand the mechanisms by which the immune system is inappropriately activated by these common foodborne disease-causing agents. Rapp. trimest. statist. sanit. mond., 50 (1997) Resume Effets chroniques des maladies d'origine alimentaire sur la sante Les effets aigus des maladies d'origine alimentaire, une fois surmontes, ne signifient pas toujours la fin de la maladie. Plusieurs agents pathogenes majeurs a l'ori- gine de ces maladies sont capables d'entraTner une maladie chronique, et m~me des destructions irreversi- bles de tissus ou d'organes, probablement par des mecanismes immunologiques. L'arthrite septique et reactionnelle, la maladie de Crohn et la recto-colite hemorragique, le syndrome hemolytique-uremique, le syndrome de Guillain-Barre, et eventuellement certai- nes maladies auto-immunes peuvent ~tre declenches par des agents des maladies d'origine alimentaire ou leurs toxines. References - References I. Foegeding, P.M. et al. Foodhorne pathogens: risks and consequences. Vol. 122, Ames, IA, USA, Council for Agricultural Science and Technology, 1994. 2. Archer, D.L. & Young, F.E. Contemporary issues: diseases with a food vector. Clinical microbiof.ogical reviews, I ( 4): 377- 398 (1988). 3. Archer, D.L. & Kvenberg, J.E. Incidence and cost of foodborne diarrhoeal disease in the United States. Journal of food protection, 48(10): 887-894 (1985). 4. Inman, E.D. et al. Post dysenteric reactive arthritis: a clinical and immunogenic study following an outbreak of salmonellosis. Arthritis and rheumatism, 31(11): 1377-1383 (1988). 5. Bunning, V.K. et al. Foodborne enterobacterial pathogens and rheumatoid disease. Journal of applied bacteriology. Symposium supplement: 87S.107S (1988). 6. Mekalanos, J.J. Environmental signals controlling expres- sion of vimlence determinants in bacteria. Journal of bact,e- riof.ogy, 174(1): 1-7 (1992). 7. Branham, R.C. et al. Bacterial antigenic variation, host immune response and pathogens-host co-evolution. Infec- tion and immunity, 61(6): 2273-2276 (1993). 8. Sinha, A.A. et al. Autoimmune diseases: the failure of self tolerance. Science, 248: 1380-1388 ( 1990). 9. Bunning, V.K. Immunopathogenic aspects of foodborne microbial disease. Food microbiology, ll: 89-95 (1994). 10. Yu, D.T.Y. & Thompson, G.T.D. Clinical, epidemiological and pathogenic aspects ofreactive arthritis. Food microbiology, ll: 97-108 (1994). 11. Keat, A. Is spondylitis caused by Kkbsiella? lmmunof.ogy today, 7(5): 144-148 (1986). 12. Behar, S.M. & Porcelli, S.A. Mechanisms of autoimmune disease induction: the role of the immune response to microbial pathogens. Arthritis and rheumatism, 38: ( 4) 458-4 76 (1995). 13. Gaston,J.H.S. How does HLA-B27 confer susceptibility to inflammatory arthritis? Clinical and experimental immunology, 82: 1-2 (1990). 14. McGuigan, L.E. et al. Significance of non pathogenic cross- reactive bowel flora in patients with ankylosing spondylitis. Annals of rheumatic diseases, 45: 566-571 (1986). 15. Kapasi, K, & Inman, R.D. HLA-B27 expression modulates Gram-negative bacterial invasion into transfected L-cells. Journal of immunof.ogy, 148: 3554-3559 (1992). 16. Ike, R et al. Reactive arthritis syndrome (RAS) following an epidemic of Salmonella gastroenteritis. Clinical research, 34: 618A (1986). Wld hlth statist. quart., 50 (1997) 17. McDermott, R.P. &Stenson W.F. Alterations of the immune system in ulcerative colitis and Crohn 's disease. Advances in immunof.ogy, 42: 285-328 (1988). 18. Belsheim, M.R. et al. Bacterial L-form isolation from inflammatory bowel disease patients. Gastroenterof.ogy, 85: 364-369 (1983). 19. Chiodini, R.J. Crohn's disease and the mycobacteriosis: a review and comparison of two disease entities. Clinical microbiof.ogical reviews, 2 ( 1): 90-117 (1989). 20. Millar, D. et al. IS900 PCR to detect Mycobacterium paratuhercuf.osis in retail supplies of whole pasteurized cow's milk in England and Wales. Applied and environmental microbiology, 62(9): 3446-3452 ( 1996). 21. Lui, Y. et al. Immunocytochemical evidence of Listeria, Escherichia coli, and Streptococcus antigens in Crohn 's disease. Gastroenterol.ogy, 108(5): 1396-1401 (1995). 22. Goodacre, J.A., Brownlee, C.E.D. et al. Bacterial super- antigens in autoimmune arthritis. British journal of rheuma- tology, 33: 413-419 (1994). 23. Koth, M. Bacterial pyrogenic exotoxins as superantigens. Clinical microbiof.ogical reviews, 8 ( 3): 411-426 (1995). 24. Koth, M. Infection and autoimmunity: a story of the host, the pathogen and the copathogen. Clinical immunof.ogy and immunopathof.ogy, 74(1): 10-22 (1995). 25. Conrad, B. et al. Evidence for superantigen involvement in insulin dependent diabetes mellitis aetiology. Nature. 371: 351-355 (1994). 26. Kay, R.A. The potential role of superantigens in inflammatory bowel disease. Clinical Experiments in immunology, 100: 4-6 (1995). 27. Leung, D.Y. et al. Toxic shock syndrome toxin secreting Staphyf.ococcus aureus in Kawasaki syndrome. Lancet, 342: 1385-1388 (1993). 28. Tarr, P.I. facherichia coli 0157:H7: clinical, diagnostic, and epidemiological aspects of human infection. Clinical infectious diseases, 20 (l): 1-10 ( 1995). 29. Tschape, H. et al. Verotoxigenic Citrobacter freundii associated with severe gastroenteritis and cases of haemolytic uraemic syndrome in a nursery school: green butter as the infection source. Epidemiof.ogy and infection, 114(3): 441-450 (1995). 30. Rodrigue, D.C. et al. A university outbreak of Escherichia coli 0157:H7 infections associated with roast beef and an unusually benign clinical course. Journal of infectious disease, 172(4): 1122-1125 (1995). 31. Outbreak of E. coli 0157:H7 infection associated with drinking unpasteurized commercial apple juice. M&rbidity and m&rtality weekly report, 45 ( 44) : ( 1996) . 32. Tesh, V.L. and O'Brien, A.D. The pathogenic mechanisms of Shiga toxin and the Shiga-like toxins. Mokcular microbiology, 5(8): 1817-1822 (1991). 33. Pudymaitis, A., Armstrong, G., and Lingwood, C.A. Vero tox- in-resistant cell clones are deficient in the glyco- lipid globotriosylceramide: differential basis of pheno- type. Archives of biochemistry and biophysics, 286(2): 448-452 (1991). 34. Obrig, T.A. et al. Endothelial heterogeneity in Shiga toxin receptors and responses. journal of biof.ogical chemistry, 268(21): 15484-15488 (1993). 35. Lingwood, C.A. Vero toxin-binding in human renal sections. Nephron, 66(1): 21-28 (1994). 36. Armstrong, G.D. et al. A phase I study of chemically synthesized verotoxin (Shiga-like toxin) Pk-trisaccharide receptors attached to chromosorb for preventing hemolytic uremic syndrome. Journal of infectious diseases, 171 ( 4): 1042- 1045 (1995). 37. ReesJ.H. et al. Campyl.obacter jejuni infection and Guillain- Barre Syndrome. New England journal of medicine, 333(21): 1374-1379 (1995). 38. Kaldor, J., and Speed, R.B. Guillain-Barre Syndrome and Campyl.obacter jejuni: A serological study. British medical journal, 228: 1867-1870 (1984). 55 39. Feasby, T.E. et al. An acute axonal form of Guillain-Barre polyneuropathy. Brain, 109: 1115-1126 (1986). 40. McKhann G.M. et al. Acute motor axonal neuropathy: a frequent cause of acute flaccid paralysis in China. Annals of neurowgy, 33: 333-342 ( 1993). 41. Bolton, C.F. The changing concepts of Guillain-Barre Syndrome. New Engl.and journal of medicine, 333(21): 1415- 1417 (1995). 42. Shoenfeld Y., George,J., and Peter,J.B. Guillain-Barre as an autoimmune disease. International archives of aUergy and immunowgy, 109: 318-326 (1996). 56 43. Oomes, P.G. et al. Anti-GM! IgG antibodies and Campylobacter bacteria in Guillain-Barre Syndrome: evidence of molecular mimicry. Annals of neurowgy, 38: 170- 175 (1995). 44. Ho, T.W. et al. Guillain-Barre Syndrome in northern China: Relationship to Campywbacter jejuni infection and anti-glycolipid antibodies. Brain, 118(Pt 3): 597-605 (1995). 45. Peterson, M.C. Clinical aspects of Campywbacter jejuni infections in adults. Western journal of medicine, 161 (2): 148- 152 (1994). Rapp. trimest. statist. sanit. mond., 50 (1997) Economic costs and trade impacts of microbial toodborne illness Jean C. Buzbya & Tanya Robertsa, b Introduction Foodborne pathogens include a diverse range of bacteria, viruses, parasites, and fungi. While these pathogens commonly cause mild and self-limiting gastrointestinal illnesses, severe acute illness and life threatening complications may occur. Perhaps, 2-3% ofall foodborne-illness cases develop compli- cations ( 1) such as arthritis, haemolytic uraemic syndrome (HUS), mental retardation; heart dis- • United States Department of Agriculture b The views expressed in this paper are solely those of the authors and should not be interpreted as official policy of the U.S. Department of Agriculture. Review comments are appreciated from Scott Crerar, Kate DeRemer, Sally Hasell, Georg Kapperud,Jean Kinsey, Fred Kuchler, Clark Nardinelli, Steve Neff, Terry Nicholls.Jenny Roberts and others. Table 1 Pathogens included in the CAST reporta Tableau 1 Agents pathogenes figurant dans le rapport du CASTa Viruses - Virus Hepatitis A - Hepatite A Norwalk Norwalk-like - Type Norwalk Bacteria - Bacteries Aeromonas hydrophifab Bacillus cereus Bruce/la abortus Campy/obacter jejuni Clostridium botulinum C/ostridium perfringens Coxiella bumetti Escherichia coli Listeria monocytogenes Mycobacterium bovis Salmonella spp. Shigel/a spp. Sta.phylococcus aureus Vibrio cho/erae Vibrio parahaemolyticus Vibrio vulnificus Yersinia enterocolitica ease or Guillain-Barre syndrome (nontrauma neuromuscular paralysis) (2). The Council for Agricultural Science and Tech- nology (CAST) (3) provided an assessment of risks from foodborne pathogens in the United States of America and identified 40 potential pathogens (Tab/,e 1 ). CAST's review of the literature concluded that there may be 6.5-33 million cases of acute foodborne illness and up to 9 OOO associated deaths in the United States each year. While these same pathogens are found in many other countries, there are geographical differences in pathogen reservoirs as well as cultural differences in food consumption habits that make each country's risks unique. This article addresses the approaches to calcu- lating the economic costs offoodborne disease, the Parasites Anisakid nematodes - Anisakides (nematodes) Cryptosporidium parvum Diphyllobothrium spp. Enta.moeba histo/ytica Giardia lamblia Taenia saginata Taenia solium Toxop/asma gondii Trichinella spiralis Toxins - Toxines Ciguatoxin - Ciguatoxine Diarrheic shellfish poisons - Toxines responsables de !'intoxication diarrheique par fruits de mer Domoic acid - Acide domoi'que Histamine Histamine-like compounds (scombroid) - Composes de type histamine Neurotoxic shellfish poison (brevetoxins) - Toxines responsables de !'intoxication neurotoxique par fruits de mer Paralytic shellfish poisons - Toxines responsables de !'intoxication paralytique par fruits de mer Tetrodotoxin - Tetrodotoxine a Source: Foegeding, P.M. & Roberts, T. Assessment of risks associated with food borne pathogens: an overview of a Council for Agricultural Science and Technology report. Journal offood protection. 19-23 (1996 supplement). Reprinted with permission from Journal of food protection. Copyright: International Association of Milk, Food, and Environmental Sanitarians, Inc. Foegeding works for North Carolina State University and Roberts works for the Economic Research Service. -Reproduit avec autorisation du Journal of food protection. Foegeding travaille pour la North Carolina State University et Roberts pour !'Economic Research Service. b Although Aeromonas hydrophi/a has not been conclusively established as foodborne, it is covered briefly in this report. - Bien que Aeromonas hydrophi/a ne soil pas formellement implique dans les maladies d'origine alimentaire, ii est brievement mentionne dans ce rapport. Wld hlth statist. quart., 50 (1997) 57 economic costs to a country from foodborne dis- ease, and the interaction between microbial food safety issues and international trade in food. The human illness costs due to foodborne pathogens are estimated most completely in the United States. Seven foodborne pathogens found in ani- mal products were selected for analysis: their costs in human illness were estimated at US$ 6.5-34.9 billion annually. However, these estimates under- value true societal costs, because they focus on medical costs and productivity losses and omit or do not fully include valuations of other costs to individuals (such as psychological costs), costs to industry, or costs to the public sector (Tah/,e 2). Calculating costs of foodborne illness in the United States Each year, 7 foodborne pathogens (Camp,ylobacter jejuni, Clostridium perfringens, E. coli O 157:H7, Liste- Table 2 Societal costs of foodborne illness Tableau 2 Coot des maladies d'origine alimentaire pour la societe Costs to indivlduals/householdsa Human illness costs: 58 Medical costs: Physician visits Laboratory costs Hospitalization or nursing home Drugs and other medications Ambulance or other travel costs Income or productivity loss for: Ill person or person dying Caregiver for ill person Other illness costs: Travel costs to visit ill person Home modifications Vocational/physical rehabilitation Child care costs Special educational programs Institutional care Lost leisure time Psychological costs: Pain and other psychological suffering Risk aversion Averting behaviour costs: Extra cleaning/cooking time costs Extra cost of refrigerator, freezer, etc. Flavor changes from traditional recipes (especially meat, milk, egg dishes) Increased food cost when more expensive but safer foods are purchased Altruism (willingness to pay for others to avoid illness) ria monocytogenes, Salmonella, Staphylococcus aureus, and Toxoplasmagondii) cause an estimated 3.3-12.3 million cases of food borne illness in the United States and up to 3 900 deaths (Tah/,e 3). Cost-of- illness estimates for these pathogens were devel- oped, based on the estimated number of acute illnesses, deaths, and secondary complications. In general, for each pathogen, the number of cases were divided into 5 severity groups: those who did not consult a physician, those who did, those who were hospitalized, those who developed complica- tions, and those who died prematurely because of their illness. Some long-term complications are in- cluded, such as HUS for E. coli 0157:H7 disease, mental retardation for toxoplasmosis and listerio- sis, and hearing and visual impairments for toxo- plasmosis. For each severity group, medical costs were esti- mated for physician and hospital services, supplies, Couts pour les personnes et les menagesa Couts des maladies humaines : Couts medicaux : Consultations medicales Frais de laboratoire Hospitalisation ou soins a domicile Medicaments et autres traitements Ambulance et autres frais de transport Perte de revenus ou de productivite pour : Le malade ou la personne decedee La personne s'occupant du malade Autres coOts imputables a la maladie : Frais de voyage ou de transport pour rendre visite au malade Modifications du domicile Readaptation professionnelle ou physique Frais de garde des enfants Programmes speciaux d'education Soins en institution Pertes au niveau du temps de loisir Couts psychologiques : Douleurs et autres souffrances psychologiques Aversion pour le risque Couts lies au comportement: Frais supplementaires pour le nettoyage et la cuisine Frais supplementaires en refrigeration Modification du gout des recettes traditionnelles (notamment pour les plats a base de viande, de lait ou d'ceufs) Accroissement du coot de la nourriture lors de l'achat d'aliments plus sOrs mais plus chers Altruisme (consentement a payer pour que d'autres evitent la maladie) Rapp. trimest. statist. sanit. mond., 50 (1997) Industry costsb Costs of animal production: Morbidity and mortality of animals on farms Reduced growth rate/feed efficiency and increased time to market Costs of disposal of contaminated animals on farm and at slaughterhouse Increased trimming or reworking at slaughterhouse and processing plant Illness among workers because of handling contaminated animals or products . Increased meat product spoilage due to pathogen contamination Control costs for pathogens at all links in the food chain: New farm practices (age-segregated housing, sterilized feed, etc.) Altered animal transport and marketing patterns (animal identification, feeding/watering) New slaughterhouse procedures (hide wash, knife sterilization, carcass sterilizing) New processing procedures (pathogen tests, contract purchasing requirements) Altered product transport (increased use of time/temperature indicators) New wholesale/retail practices (pathogen tests, employee training, procedures) Risk assessment modeling by industry for all links in the food chain Price incentives for pathogen-reduced product at each link in the food chain Outbreak costs: Herd slaughter/product recall Plant closings and cleanup Regulatory fines Product liability suits from consumers and other firms Reduced product demand because of outbreak: Generic animal product - all firms affected Reduction for specific firm at wholesale or retail level Increased advertising or consumer assurances following outbreak Wld hlth statist. quart., SO (1997) Couts pour l'lndustrieb Coots au niveau de la production animate : Morbidite et mortalite des animaux dans les exploitations agricoles Reduction du rythme de croissance, de l'indice de conversion alimentaire et temps d'elevage plus long avant la mise sur le marche Frais d'elimination des animaux contamines chez l'exploitant et a !'abattoir Augmentation du travail de preparation et de retraitement a !'abattoir et dans les usines de fabrication Maladies chez les employees a cause de la manipulation d'animaux ou de produits contamines Accroissement des pertes pour les produits carnes a cause de la contamination par des agents pathogenes Frais inherents a la lutte contre les agents pathogenes a tous les mail/ons de la chaine alimentaire : Nouvelles pratiques d'exploitation agricole (separation du cheptel en fonction de l'age, aliments sterilises, etc.) Modifications du transport des animaux et de la commercialisation (identification des animaux, alimentation, abreuvage) Nouvelles techniques dans les abattoirs (lavage des animaux avant abattage, sterilisation des couteaux, des carcasses) Nouvelles techniques de traitement (recherche des agents pathogenes, clauses d'achat etablies par contrat) Modifications du transport des produits (augmentation de !'utilisation d'indicateurs temps/temperature) Nouvelles pratiques au niveau des grossistes et de la vente au detail (recherche des agents pathogenes, formation des employes, techniques) Modelisation de !'evaluation du risque a tous les maillons de la chaine alimentaire par l'industrie Incitations au niveau des prix pour reduire la quantite d'agents pathogenes dans les produits a tous les maillons de la chaine alimentaire Couts des flambees epidemiques : Abattage des troupeaux/rappel des produits Fermeture et nettoyage des usines Amendes reglementaires Proces intentes par les consommateurs et les entreprises sur la garantie des produits Diminution de la demande pour les produits a cause de la flambee epidemique : Production animale en general - toutes les entreprises sont touchees Reductions pour certaines entreprises au niveaux des grossistes ou de la vente au detail Augmentation de la publicite et des garanties pour les consommateurs a la suite des flambees epidemiques 59 Regulatory and public health sector costs for foodborne pathogens Disease surveillance costs to: Monitor incidence/severity of human disease by foodborne pathogens Monitor pathogen incidence in the food chain Develop integrated database from farm to table for foodborne pathogens Research to: Identify new foodborne pathogens for acute and chronic human illnesses Establish high-risk products and production and consumption practices Identify which consumers are at high-risk for which pathogens Develop cheaper and faster pathogen tests Risk-assessment modeling for all links in the food chain Outbreak costs: Costs of investigating outbreak Testing to contain an outbreak (for example, serum testing and administration of immunoglobulin in persons exposed to Hepatitis A) Costs of cleanup Legal suits to enforce regulations that may have been violatedc Other considerations: Distributional effects in different regions, industries, etc. Equity considerations, such as special concern for children Couts dans les secteurs de la sante publique et de la reglementatlon associes aux agents pathogenes d'orlglne alimentalre Couts de la surveillance epidemio/ogique pour: Surveiller !'incidence et la gravite des maladies humaines causees par les agents pathogenes d'origine alimentaire Surveiller !'incidence des agents pathogenes dans la chaine alimentaire Elaborer une base de donnees integrant toutes les etapes, de !'exploitation agricole a la table du consommateur, pour ces agents pathogenes Recherche pour: Identifier de nouveaux agents pathogenes transmettant par l'alimentation des maladies aigues au chroniques chez l'homme Etablir la liste des produits et des pratiques de production et de consommation a haut risque Identifier les consommateurs les plus exposes au risque pour chaque agent pathogene Mettre au point des epreuves plus rapides et mains couteuses de recherche des agents pathogenes Modelisation de !'evaluation du risque pour tous les maillons de la chaine alimentaire coots des flambees epidemiques : Couts des enquetes sur les flambees Epreuves diagnostiques pour limiter la flambee epidemique (par exemple les recherches seriques et I' administration d'immunoglobulines aux personnes exposees a l'hepatite A) Frais de nettoyage Proces en justice concernant !'application des reglements qui ant pu etre violesc Autres considerations : Effets sur la distribution dans differentes regions et industries, etc. lnterets speciaux comme les preoccupations pour les enfants en particulier • Willingness-to-pay estimates for reducing risks of food borne disease is a comprehensive estimate of all these categories (assuming that the individuals have included employer funded sick leave and medical programs in their estimates). The estimate is comprehensive and covers reduced risks for everyone-those who will become ill as well as those who will not. - Les estimations du desir de payer pour la reduction des risques des maladies d'origine alimentaire comprennent les estimations pour toutes ces categories (en partant de l'hypothese que les individus ant inclus les conges-maladie et les programmes d'assurance medicale payes par les employeurs). II s'agit d'une estimation exhaustive qui couvre la reduction du risque pour tous, ceux qui tomberont malades et ceux qui ne le deviendront pas. b Some industry costs may fall with better pathogen control, such as reduced product spoilage, possible increases in product shelf-life, and extended shelf-life permitting shipment to more distant markets or lowering shipment costs to nearby markets. -Certains coats pour l'industrie pourraient disparaitre avec une meilleure lutte centre les agents pathogenes, comme la diminution des produits a mettre au rebut, l'accroissement possible de la duree de conservation, permettant !'expedition vers des marches plus lointains au abaissant les frais de transports vers les marches proches. c In adding up costs, care must be taken to assure that product liability costs to firms are not already counted in the estimated pain and suffering cost to individuals. However, the legal and court expenses incurred by all parties are societal costs. - En faisant la somme des couts, ii taut bien s'assurer que les frais de garantie des produits pour les entreprises ne sont pas deja comptes dans l'estimation du pretium doloris pour les personnes. Ouoi qu'il en soit, les frais juridiques assumes par toutes les parties representent des couts pour la societe dans son ensemble. Source: USDA, Economic Research Service, based on table in (12). - USDA, Economic Research Service, d'apres le tableau dans (12). medications, and special procedures unique to treating the particular foodborne illness. Such costs reflect the number of days/treatments of a medical service, the average cost per service/treat- ment, and the number of patients receiving such service/treatment. 60 Most people with foodborne illnesses only miss 1 or 2 days of work. This lost productivity is approxi- mated by wage rates, published by the United States Bureau of Labor Statistics. However, some patients die and some develop complications so that they never return to work. For these patients, Rapp. trimest. statist. sanit. mond., 50 (1997) Calculating the value of a statistical life estimates of the value of a statistical life for an unidentified individual are used as a proxy for lost productivity. The total cost of lost productivity is the sum for all individuals affected, including the patients and, in the case of ill children, their par- ents or paid caretakers. The United States Department of Agriculture's Economic Research Service (ERS) now obtains value of statistical life estimates from two different approaches. The first approach estimates the value of a statistical life by using records indicating con- Table 3 Estimated annual costs of selected foodborne illnesses in the United States a Tableau 3 Coots annuels estimes de certaines maladies d'origine alimentaire aux Etats-Unis d'Amerique a Pathogen and disease/complication - Agent pathogene et maladie/complication Bacteria: - Bacteries: Campylobacter jejuni or coli Campylobacteriosis - Campylobacteriose Clostridium perfringens C. pertringens intoxications - Intoxications par C. perfigens Escherichia coli 0157:H7 £ coli 0157:H7 disease - Maladie a£ coli Estimated food borne illness - Estimated food borne illness costs, assuming: - Maladies d'origine alimentaire (estimations) Couts estimes des maladies d'origine alimentaire, selon hypotheses : Cases-Cas Deaths - Deces US$ 5 mil. per lifet1- US$ 5 millions par vieb Ref. - Ref. (Sf Number - Nombre Billion US$ - Milliards de US$ 1 100 000-7 OOO OOO 110-511 1.2-6.6 0.7-4.3 10 OOO 100 0.5 0.1 8 000-16 OOO 80-200 0.4-1.0 0.1-0.3 Hemolytic uremic sydromed - Syndrome hemolytique-uremiqued 320-656 96-233 0.5-1.2 0.2-0.4 Subtotal - Sous-total 176-433 0.9-2.2 0.3-0.7 Listeria monocytogenese Listeriosis - Listeriose 928-1 767 230-485 1.2-2.2 0.12-0.25 Complications 22-41 0 0.1-0.2 .03-.05 Subtotal - Sous-total 230-485 1.3-2.4 0.1-0.3 Salmonella (non-typhoid) - Salmonella (non typhoidique) Salmonellosis - Salmonellose 696 000-3 840 OOO 870-1 920 4.8-12.2 0.9-3.5 Staphylococcus aureus S. aureus intoxications - Intoxications par S. aureus 1 513 OOO 454 3.3 1.2 Parasite: Toxoplasma gondiif Toxoplasmosis - Toxoplasmose 217 40 0.1 0.04 Complications 1 541 0 7.6 3.15 Subtotal - Sous-total 1 581 40 7.7 3.2 Total 3 300 000-12 300 OOO 1900-3900 19.7-34.9 6.5-13.3 • Cost estimates are in 1995 US$. - Estimation des couts en dollars US de 1995. b The $5 million value of a statistical life was estimated from wage-risk studies. -La valeur de US$ 5 millions etablie pour une vie statistique a ete estimee apres des eludes de risque-salaire. c This human capital approach, increased by a willingness-to-pay multiplier, estimates the value of a statistical life, depending on age, to range from roughly $15 OOO to $1 979 OOO in 1995 US$. - Cette approche fondee sur le capital humain, augment!! d'un coefficient tenant compte du consentement a payer, est une estimation de la valeur d'une vie statistique, en fonction de l'age, et allant d'environ US$15 OOO a US$1 979 OOO (valeur 1995). d Kidney failure. - lnsuffisance renale. • Includes only hospitalized patients because of data limitations. - Concerne uniquement les malades hospitalises, en raison de l'insuffisance de donnees. t Includes only toxoplasmosis cases related to fetuses and newborn children who may become blind or mentally retarded. 217 develop severe acute illness at birth, of which 40 die and 177 survive acute illness but develop complications by age 17. In addition to these 177 acute cases with complications, 1 364 cases do not have noticeable acute illness at birth but develop complications by age 17 for a total of 1 541 food borne illness cases with complications. Does not include all other cases of toxoplasmosis. Another high-risk group for this parasite is the immunocompromised, such as patients with AIDS. -N'inclut que les cas detoxoplasmose concernant des fcetus et des nouveau-nes risquant de devenir aveugles ou de presenter un retard mental. Parmi ces cas, 217 ant une maladie aigue a la naissance, dont 40 decedent et 177 survivent mais font des complications avant l'age de 17 ans. Outre ces 177 cas aigus avec complications, 1 364 cas ne presentent pas de maladie aigue notable a la naissance mais font des complications avant l'age de 17 ans, sur un total de 1 541 cas de maladies d'origine alimentaire avec complications. Ce chittre n'inclut pas les autres cas de toxoplasmose. Les sujets immunodeficients, parexemple les malades atteints de SIDA, constituent un autre groupe a haut risque pourcette maladie. Wld hlth statist. quart., 50 (1997) 61 sumers' willingness-to-pay to reduce the risks of death and poor health. The hedonic-wage ap- proach uses labour market data to estimate con- sumers' willingness-to-pay. Workers, often subcon- sciously, place a value on life and health when they earn higher wages in jobs that incur risks. Typical- ly, employers must offer workers higher wages to induce them to take such a job, as opposed to similar jobs with no such risks. Viscusi ( 4) compared wage differences in 24 wage-risk studies and found that the extra wages associated with the increased overall hazard of 1 death from risky jobs are between $3 million and $7 million (in 1990 US$). Several United States regulatory agencies use either Viscusi's range of estimates or the $5 million midpoint when ana- lysing the benefits of proposed public-safety rules. [Note: this value is usually not updated to a more recent year for these analyses]. The $5 million value is used when estimating costs under this first approach. One caveat when using this estimate for foodborne disease is that it was developed for United States workers employed in risky jobs whereas more se- vere cases of foodborne illnesses often occur among the very young, the elderly, and the immu- nocompromised. Without further research it is un- clear whether the $5 million over- or under-esti- mates the per-person value of a statistical life for a particular sub-population with foodborne illness. The second approach was developed by Lande- feld and Seskin (5) and estimates the value of a statistical life by focusing primarily on lost produc- tivity. For those who die or are unable to return to work, this lost productivity is calculated using a combination of human capital and willingness-to- pay estimates. Human-capital estimates are the value in today's dollars of the difference between an individual's lifetime stream of income if the illness had not occurred, and the income stream given the illness. Landefeld and Seskin increased human capital estimates by a risk-aversion multi- plier that captures people's willingness-to-pay to avoid death, as reflected in life insurance premi- ums. These estimates of the value of a statistical life range, depending on age, from roughly $15 OOO to $1 979 OOO (in 1995 US$). The major limitation of this approach is that it does not fully consider the value that individuals may place on (and pay for) feeling healthy, avoiding pain and suffering, or using their free time. Because the approach does not cover all of these valuable aspects of health, the approach understates true societal costs. ERS uses both approaches here because econo- mists have not reached a consensus on which esti- mates to use, though they may now be leaning toward the hedonic-wage approach. Previously, ERS used the Landefeld and Seskin estimates in cost-of-illness anal- yses because these estimates are more conservative and provide values for patients of different ages, instead of one value for patients of all ages (6). 62 High costs of foodborne illness in the United States When Landefeld and Seskin's value of a statistical life estimates were used in the cost-of-illness ana- lyses, estimated annual United States costs of the seven foodborne illnesses totaled $6.5-$13.3 bil- lion. Using the $5 million value from the hedonic- wage studies, annual costs totalled $19.7-$34.9 bil- lion. Both sets of estimates are in 1995 US$ and undervalue true societal costs of foodborne ill- nesses, however, because the analyses covered only 7 pathogens and did not incorporate all of the costs identified in Tab/,e 2. Estimated costs would also increase if the costs for all complications linked to foodborne illnesses, such as arthritis and Guillain-Barre syndrome were included. Research on estimating the costs of these complications is needed. Cost estimates In other countries The primary limitation in estimating foodborne disease costs is that for most pathogens there are no accurate data on incidence and the distribution of outcome severity. A landmark study by the United States Centers for Disease Control and Pre- vention (CDC) did provide "best estimates" of foodborne illness for selected pathogens and was the foundation for our cost estimates (7). For other countries, surveillance systems are generally re- stricted to reporting foodborne disease outbreaks (which misses the majority of cases that are unre- ported) and/or laboratory isolates from patients who have consulted a doctor (which captures more cases, but misses the milder cases, persons who prefer self-treatment as well as cases where the lab test does not isolate a pathogen or where the doc- tor does not ask for a lab test). Todd (8) extrapolated estimates of foodborne illness cases and deaths from the Bennett study (7) to Canada with some adjustments. The most not- able adjustment is a reduction of the estimated death rate. Foodborne illnesses were estimated at 2.2 million annually at a cost of CAN$ 1.3 billion (1985 CAN$, using Landefeld and Seskin esti- mates). Razem & Katusin-Razem (9) estimated the costs of notified foodborne disease poisoning cases in Croatia. They assumed that the cost per salmonel- losis case would be representative of the average foodborne illness. To extrapolate cost estimates from studies in other countries to Croatia, they used the ratio of Gross National Product per capita between Croatia and the reference country. Esti- mated foodborne illness costs in Croatia exceeded US$ 2 million annually. The cost of inpatient care in hospitals for acute infectious intestinal disease in England from 1991 to 1994 has been estimated at £24 million annually ( 10). Roberts ( 11) estimated the medical costs and Rapp. trimest. statist. sanit. mond., SO (1997) value of lives lost from five infections in England and Wales at £300-£700 million annually. The foodborne percentage was not discussed in these studies, but implied to be the bulk of the cases. Other cost estimates in the literature are limited to costs for specific pathogens and are not reviewed here (see (12) for a brief discussion of these esti- mates). Inter-country extrapolations of foodborne ill- ness cases, deaths, and costs are hindered by differ- ences in surveillance systems.c Although the basis of surveillance systems will never be quite the same for all countries, perhaps in the future, surveil- lance systems may be sufficiently similar for reason- able comparison. Foodborne illness risks differ among countries Countries face real differences in foodborne illness risks and as a result the incidence of foodborne illness and associated deaths varies. International differences in food production practices that can influence the probability of pathogen contamina- tion include: farming and husbandry practices ( e.g., degree of industrialization), slaughtering and processing technologies, storage and preserva- tion techniques (including refrigeration, gas-pack- ing etc.), and final preparation (see Tabl,e 2 in (17)). Pathogens vary in their geographical distribu- tion in the environment and in food production systems. Livestock vary in their genetic suscepti- bility to a specific pathogen. Climate, such as a harsh winter, limits the lifetime and spread of some pathogens. Human host factors include differences in the number of people who are more susceptible to infection, such as the elderly, infants, and persons c Improvements in surveillance systems that are undeiway in many countries will improve the ability of economists to estimate societal costs of foodbome illness. In the United States, a jointeffort by CDC, Food Safety and Inspection Service (FSIS), and Food and Drug Administration will survey all patients with diarrhoeal disease at selected sentinel sites to develop national incidence estimates for specific pathogens causing diarrhoea ( 13). In the United Kingdom, a 3-year survey of 70 general practitioners for patients with gastroenteritis is nearing completion. This is a collaborative study between the Public Health Laboratory Service, the Medical Research Council, the London School of Hygiene and Tropical Medicine, and the Department of Health. It will include identification of pathogens, acute and chronic impact, and estimates of economic costs ( 14 ). Salm-Net is improving European surveillance of human salmonellosis, based on laboratory reports. The goal is on-line availability to use in spotting emerging outbreaks across national borders (15 ). New Zealand has estimated there may be as many as 300 OOO cases of foodbome illness annually (16), but a breakdown by specific pathogen is not available for most pathogens. Australia is currently planning a population based study of diarrhoeal disease which will employ similar methodology to that of the FDA/FSIS/CDC study undeiway in the United States. A national foodbome disease outbreak surveillance system is currently being developed and will be piloted in a number of states in 1997 (S. Crerar, unpublished observations, 1996). Wld hffh statist. quart., 50 (1997) who are immunocompromised due to medical conditions, malignancy, immunosuppressive treat- ment, and malnutrition. The unavailability of an adequate food supply leaves people with inade- quate strength to build up resistance to pathogens (3). Even if food is available in sufficient quantities for nourishment, poor-quality food ingredients may contain relatively higher pathogen loads. These groups are more susceptible than others to opportunistic and low-dose pathogens, and more likely to develop serious manifestations of the infections, including secondary complications and death. The increasing prevalence of AIDS patients worldwide means a growing population at risk from foodborne diseases. The death rate from foodborne illnesses is considered to be higher in developing countries where pre-existing medical conditions are common, nutritional status is poor, and where there is a reduced likelihood of immedi- ate medical attention ( 18). Overall, costs of food- borne illnesses are likely to be higher in countries with larger at-risk populations. Cultural and individual food-handling prac- tices, cooking preferences, and kitchen hygiene routines may put some individuals at higher risk of foodborne illness. Such factors may vary from one country or region to another. For example, per- sons who consume greater quantities of a food likely to contain pathogens and persons who prefer raw or rare animal protein products (such as cheeses made of unpasteurized milk ( 19), raw shell- fish/ seafood, raw or rare meat products, and many egg dishes) face higher risks offoodborne illness.cl International trade and food safety International trade in agricultural products and commodities is extensive and growing. The global value of this trade was estimated at US$ 381 billion in 1993 (23). Any trade poses some risk of introduc- ing new foodborne pathogens into countries ( e.g., new Salmonella strains (24 )) or spreading patho- gens across boundaries (i.e., from endemic areas to low-endemic areas). Perceptions of these and oth- er food safety risks, as well as the value of a statisti- cal life, vary among countries. These differences contribute to the diversity of international stan- dards for food production and inspection, as well as the diversity of regulatory procedures. d The 1993 outbreak of E. coli0157:H7 in the United States was associated with medium-rare hamburgers. The consumption of raw limpets by the Portuguese population in the United States has been responsible for cases of typhoid fever and Norwalk-lite gastroenteritis (20). Latin American dishes, such as raw meat or fish marinated in lime juice, pose risks. And, human infection with the liver fluke Opistharchis uivcnini associated with the consumption ofrawfish is the leading cause offoodbome parasitic illness in Thailand (21 ). Consumption of uneviscerated fish led to the first major botulism outbreak in Egypt (22). 63 Each year, a proportion of food exports are rejected by importing countries on the grounds of unacceptable food contamination or food product attributes. Individual food shipments or groups of shipments can be rejected as in the case of food that is decomposed or contaminated with filth or hazardous loads of pathogens. But on a larger scale, countries can impose food safety-based tech- nical barriers to trade that have more far reaching trade impacts. These barriers to trade include re- strictions due to the presence or use of growth hormones or antibiotics, and zero or infeasibly low tolerance levels for Salmonella, Campylobacter, or other pathogens in food. Other food-related bar- riers to trade that do not involve product contami- nation include restrictive shelf-life, labelling, and temperature requirements as well as changing and ambiguous standards that make exporting to the country difficult. These barriers can effectively block or limit international trade of food and can result in substantial economic losses. The worldwide economic consequences from these food rejections are unquantified but likely to be substantial. To date, data could not be found on the total annual volume and economic impact of international shipments of food that are rejected for real or alleged food safety reasons, either for individ- ual countries or for all countries combined. These data are highly sensitive because dissemination of the information could jeopardize ongoing techni- cal negotiations with foreign trading partners, espe- cially in cases where official estimates released by a government differ from estimates found in the data. Estimates, when available to a government agency, are commonly used to underpin strategic planning efforts and are not for external release. Some food safety barriers to trade are reason- able, for example when a sovereign State decides to implement a higher science-based standard of food safety to protect its residents than similar stan- dards in other countries. It is believed (K. DeRemer, unpublished observations, 1996) that questionable barriers to trade may include policies which are not supported by sound science, are not least trade restrictive, or violate one of the many other principles in the General Agreement on Tar- iffs and Trade (GATT) and the Agreement on Sanitary and Phytosanitary Measures (SPS). For example, a barrier to trade may be imposed under the guise of food safety but which was developed primarily to shield domestic producers from inter- national competition. A hypothetical example of a barrier to trade that is not least trade restrictive is a country that requires frozen meat to be both frozen and test negative for trichinae even though trichinae is killed by freezing and therefore poses no health hazard. Food safety regulations can also be used to manipulate trade arrangements. A zero tolerance for Campylobacter in poultry can effectively close off all poultry imports because 64 Campywbacteris a pathogen common to the poultry industry worldwide. Perhaps one of the largest food safety issue in blocking international trade is Salmonella. During the 1970s and 1980s, Salmonella on raw meat, mung beans, dried milk, and other foods disrupted world trade (25 ). However, many exporters found requirements for Salmonella un- reasonable because most microbiologists concur that this bacteria cannot be completely eradicated, using current technology, from most raw meats (25). Regulations that require zero tolerance for Salmonella and Campywbacter in imports but do not test or hold similar standards for domestic prod- ucts where the pathogens are also endemic violate theGATT. Steps towards harmonization As a result of these varying standards for food safety and their subsequent impact on trade, there has been substantial and continuing debate on the harmonization of international standards. Harmo- nization could potentially provide a level playing field by opening markets, bringing cost savings to industry, providing an arena to resolve conflicts fairly, and perhaps reducing the cost and shorten- ing the time needed to introduce new products. For example, cost savings to an industry can occur if an exporter needs only one uniform set of micro- bial testing for a product instead of a series of redundant and costly procedures and tests. Several steps towards harmonization have al- ready occurred. The Codex Alimentarius Commis- sion ( commonly referred to as Codex) was created in the 1960s by two agencies of the United Nations (the World Health Organization (WHO) and the Food and Agriculture Organization (FAO)) to prepare voluntary standards for the safe use and levels of food additives, pesticides, chemi- cals, and contaminants (26). In 1994, the Uruguay Round of the GATT was signed by the United States and over 100 other nations. The GATT agreement provides a framework for distinguish- ing protectionist regulations from legitimate SPS regulations by requiring all SPS measures to be science-based. As part of the GATT agreement, the World Trade Organization (WTO) was created to settle trade disputes such as for product standards. WTO has no mechanism to devise standards, so it refers problematic issues that may be solved by having an international standard to Codex or to other international standard-setting bodies. Cur- rently, countries either adopt Codex standards and are in compliance with WTO's trade rules or they have more rigorous standards that they must de- fend using public health and safety reasons (26). System approaches such as Hazard Analysis Critical Control Point (HACCP) and ISO 9000 standards convey verifiable and valuable information to trad- ing partners and thus will likely have important implications on international trade (27). Rapp. trimest. statist. sanit. mond., 50 (1997) Discussion The presence of foodborne pathogens in a coun- try's food supply not only affects the health of the local population, but also represents a potential for spread of pathogens to visiting tourists and cus- tomers in other countries to which food products are exported. Likewise, the presence of animal dis- eases in a country's livestock may result in the dissemination of pathogens through exportation of meat products. Awareness of these problems complicates international trade in food and has stimulated international efforts to harmonize food safety standards. Knowledge of the current extent of food borne illness cases worldwide, associated costs of illness, and costs of rejections in international food ship- ments due to food safety concerns and regulations could be used as an indicator by which changes in food safety can be monitored. The estimated $6.5- $35 billion annual costs for 7 foodborne illnesses in the United States provides an indication of the high costs of foodborne illness. Including cost esti- mates for other pathogens and for other countries would greatly magnify these costs. New tests for pathogens; improved epidemio- logical techniques; new animal production, slaugh- ter, and processing techniques; and application of HACCP methods to systematically identify and control risks in food production will fundamentally increase our ability to reduce risks from pathogens in foods. However, improved surveillance pro- grammes are needed to better quantify existing risks and target the most costly risks for control. Quantifying the economic and health consequenc- es caused by foodborne pathogens can also help policymakers weigh the costs and benefits of imple- menting or expanding food safety programmes for health promotion and protection. Summa,y This article presents the economic costs of foodborne diseases for selected countries, the approaches used to calculate these costs, and a discussion on the interac- tion between microbial food safety issues and interna- tional trade in food. The human illness costs due to foodborne pathogens are estimated most completely in the United States of America, where, each year, 7 food borne pathogens ( Campylobacter jejuni, Clostri- dium perfringens. Escherichia coli 0157:H?, Listeria monocytegenes, Salmonella, Staphylococcus aureus, and Toxoplasma gondi1) cause an estimated 3.3- 12.3 million cases of foodborne illness and up to 3 900 deaths. These 7 pathogens are found in animal products and cost the United States an estimated $6.5- $34.9 billion (1995 US$) annually. The presence of food borne pathogens in a country's food supply not only affects the health of the local population, but also repre- sents a potential for spread to pathogens to visitors to the country and to consumers in countries which import food products. With more complete data on foodborne Wld hlth statist. quart., 50 (1997) illnesses, deaths, costs and international trade rejec- tions in each country, indicators could be developed by which changes in food safety can be monitored. Resume coot economique et impact commercial des maladies microbiennes d'origine alimentaire Cet article presente le coat economique des maladies d'origine alimentaire dans certains pays, avec les ap- proches utilisees pour les calculs et une discussion sur les rapports entre les questions de securite microbiolo- gique des aliments et le commerce international des denrees alimentaires. Le coot des maladies humaines dues a des agents pathogenes alimentaires est estime de fac;:on tres complete aux Etats-Unis d'Amerique ou, chaque annee, 7 agents pathogenes transmis par les aliments ( Campylobacter jejuni, C/ostridium perfrin- gens, Escherichia coli 0157:H?, Listeria monocytoge- nes, Salmonella, Staphylococcus aureus et Toxop/as- ma gondi1) provoquent entre 3,3 et 12,3 millions de cas de maladie et jusqu'a 3 900 deces. Ces agents se rencontrent dans les produits animaux et coOtent cha- que annee aux Etats-Unis de US$ 6,5 a 34,9 milliards (valeur 1995). La presence d'agents pathogenes dans la chaine alimentaire d'un pays affecte non seulement la sante de sa population mais constitue egalement un risque de propagation aux visiteurs etrangers et aux consommateurs des pays importateurs de denrees. En disposant de donnees plus completes sur les maladies d'origine alimentaire, les deces, les coots et les opposi- tions au commerce international pour chaque pays, on pourrait elaborer des indicateurs qui permettraient de surveiller les modifications de la salubrite des aliments. References/References 1. Archer, D.L. &: Kvenberg, J.E. Incidence and cost of food borne diarrheal disease in the United States. Journal of food protection, 48: 887-894 (1985). 2. Mossel, D.A.A. Impact of foodborne pathogens on today's world, and prospects for management. Animal and human health, l: 13-23 (1988). 3. CAST Report. Foodborne pathogens: risks and consequences. Task force report no. 122, Washington DC, Council for Agricultural Science and Technology, Sept. 1994. 4. V1SCusi, W.K. The value of risks to life and health. Journal of economicliterature, 31: 1912-1946 (1993). 5. Landefeld, J. &: Seskin, E.P. The economic value of life: linking theory to practice. American journal of public health, 6: 555-566 (1982). 6. Buzby, J.C. et al. Bacterial foodbarne disease: medical cost1 and productivity wsses, AER-741. USDA, Economic Research Service. Aug. 1996. 7. Bennett, J.V. et al. Infectious and parasitic diseases. In: Amler and Dull (eds), Cwsing the gap: the burtkn of unnecessary illness. New York, Oxford University Press, 1987, pp 102-114. 8. Todd, E.C.D. Preliminary estimates of costs of foodborne disease in Canada and costs to reduce salmonellosis. Journal of food protection, 52(8): 586-594 (1989). 9. Razem, D. &: Katusin-Razem, B. The incidence and costs of food borne diseases in Croatia.Journal of food protection, 57 (8): 746-752 (1994). 65 10. Djuretic, T. et al. The cost of inpatient care for acute infectious intestinal disease in England from 1991 to 1994. CDR R.euiew, 6: R 78-R80 (1996). 11. Roberts,J.A. Economic evaluation of surveillance. London, Department of Public Health, and Policy, Aug. 1996. 12. Roberts, T. &: Todd, E. APfrroaches to estimating the benefits and costs of foodborne disease control choices. \\'HO/USDA Consultation on economic implications of foodborne diseases on animal production food safety, Washington, DC, June 8-10, 1995. 13. U.S. Food and Drug Administration, Food Safety and Inspection Service, and Centers for Disease Control and Prevention. FSIS, FDA, CTJC, State health departments collaborate on foodborne illness project. Backgrounder. July 1995. 14. Tompkins, D. The infectious intestinal disease (IID) in England study: interim report, PHLS Microbiology digest, 13: 84-85 (1996). 15. Public Health Laboratory Service (PHIS). Global role for Salm-Net? At-Your Service: PHLS Newsletter, 2: 4-5 (1996). 16. Gilbert, S. &: Allman, R. Improving food safety in New Zealand. The New Zealand public health report, 3: 65-77 ( 1996). 17. Roberts, T. et al. Risk assessment for foodborne microbial hazards. In: Trackingfoodborne pathogens from farm to table: data needs to evaluate control options. U.S. Dept. of Agr., Econ. Res. Serv., MP No. 1532, 1995. 18. Sockett, P. Social and economic aspects of food-borne disease. Food policy, 18 (2): 110-119 ( 1993). 19. Nichols, G.M. et al. The microbiological quality of soft cheese. PHLS microbiowgy digest, 13: 68-75 (1996). 66 20. Townes,J.M. et al. Outbreak of norwalk virus gastroenteritis associated with eating raw imported limpets. Program and abstracts of the 34th interscience conference on antimicrobial agents and chemotherapy. Washington, DC, American Society for Microbiology, AbstractJ188, 1994. 21. Loaharanu, P. &: Sommani, S. Preliminary estimates of economic impact ofliver fluke infection in Thailand and the feasibility of irradiation as a control measure. Southeast asian journal of tropical medicine and public health, 22: 384-390 (1991). 22. Todd, E.C.D. Worldwide surveillance of foodborne disease: the need to improve. Journal of food protection, 59 ( 1): 82-92 (1996). 23. Henderson, D.R.&: Handy C.R. International dimensions of the food marketing system. In: Schertz & Daft (eds). Food and agricultural markets: the quiet revolution. Washington DC, National Planning Association, 1994. 24. D'Aoust, J.Y. Salmonella and the international food trade. International journal of food microbiowgy, 24: 11-31 ( 1994). 25. Eyles, M.J. Microbial concerns of the Pacific rim countries and implications for harmonizing free trade. Dairy.food, and environmental sanitation, 14(8): 467-470 (1994). 26. Leonard, R.E. Safe food, environment compliments global trade. Nutr-week. Washington DC, Community Nutrition Institute. Mar. 8, 1996. pp 4-5. 27. Adams, C.E. Microbial concerns of the North and South American countries and scientific implications for harmonizing free trade. Dairy, food, and environmental sanitation, 14(8): 471-472 (1994). Rapp. trimest. statist. sanit. mond., 50 (1997) Foodborne listeriosis Jocelyne Rocourta & Jacques BilJeb Introduction While rarely diagnosed prior to 1960, more than 10 OOO cases of listeriosis were recorded in the medical literature between 1960 and 1982 and thousands more have been reported annually worldwide. This trends reflects a change in social patterns which favours the expression of this microorganism. Listeria monocytogenes is a bacte- rium responsible for opportunistic infections, preferentially affecting individuals whose immune system is perturbed, including pregnant women, newborns, immunocompromized persons ( cancer patients, transplant recipients, people on haemo- dialysis, AIDS patients, etc.) and the elderly. Thus, longer life expectancy and medical progress al- lowing immunodeficient individuals to survive partially explain the increasing incidence of liste- riosis. Moreover, L. monocytogenes is ubiquitous and can grow at temperatures as low as +3 °C. Thus facilitating contamination of various food- stuffs. The expansion of the agro-food industry, the widespread use of systems of cold storage and changes in consumer demand ( convenience foods that have freshly-cooked taste, can be pur- chased ready-to-eat, refrigerated or frozen, pre- pared rapidly and require essentially little heat application before consumption) have led to a large increase in the pool of Listeria from which man can be contaminated. Listeriosis Human listeriosis is a foodborne disease occurring either at a low prevalence endemic mode or as outbreaks of variable size and duration. In both settings, the disease is severe with a high mortality rate. In nonpregnant human adults, L. monocytogenes primarily causes meningitis, encephalitis, or septi- caemia ( 1, 2). Elderly patients or persons with pre- disposing conditions that lower cell-mediated im- munity, such as those with transplanted organs, lymphomas, and AIDS, are especially susceptible. However, on rare occasions, patients have no rec- • WHO Collaborative Centre for Foodborne Listeriosis, Pasteur Institute, Paris, France. E-mail:jrocourt@pasteur.fr b WHO Collaborative Centre for Foodborne Listeriosis, Institute of Microbiology, University Hospital, 1011 CHUV Lausanne, Switzerland. E-mail:jacques.bille@chuv.hospvd.ch Wld hlth statist. quarf.., 50 (1997) ognizable predisposing conditions. The yet unex- plained tropism of L. monocytogenes for the central nervous system leads to severe disease, often with high mortality (20-50%) or with neurological se- quelae among survivors (3). In pregnant women, L. monocytogenes often causes an influenzalike bactaeremic illness which, if untreated, may secondarily lead to amnionitis and infection of the fetus, resulting in abortion, stillbirth, or premature birth. Early diagnosis can be made by detecting L. monocytogenes in maternal blood cultures; at birth, the diagnosis is made by detecting the organism in cerebrospinal fluid (CSF), blood, amniotic fluid, respiratory secre- tions, placental or cutaneous swabs, gastric aspi- rate, or meconium of the neonate. Direct micro- scopic visualization of gram-positive rods in these specimens could be invaluable in early diagnosis of the disease ( 4 ). Focal infections rarely occur after an episode of bactaeremia. However, primary cutaneous listerio- sis with or without bactaeremia has been reported among veterinarians and abattoir workers, who ac- quire the illness through direct contact with in- fected animal tissues. Endocarditis, arthritis, osteo- myelitis, intra-abdominal abscesses, endoph- thalmitis, and pleuropulmonary infections have been described, although infrequently. The incubation period and infective dose have not been firmly established, although they may be inversely related. Reported incubation times vary from a few days to 2-3 months. Gastrointestinal symptoms such as diarrhoea have been observed in some individuals with listeriosis but are not com- monly associated with ingestion of contaminated food. A transient carrier state exists in 2-20% of animals and humans. Epidemiology of listeriosis Human listeriosis can occur either as a sporadic disease (probably most often of foodborne origin) or as part of an outbreak. The annual incidence of listeriosis usually does not distinguish between spo- radic and epidemic cases. In addition, the system of reporting cases to health authorities vary greatly from one country to another, being even totally absent in many countries, particularly outside Europe and United States of America. The first tentative to put these data together was done at a WHO informal working group in Geneva in 1988 67 (see Tab/,es 1 & 2, refs. 5-7 and c). Since then, a number of countries with surveillance systems have regularly reported the yearly incidence of human listeriosis. The annual rate varies from 2 to 15 cases per million population, and tends to increase in some European countries. It is difficult however to know whether this increase is real, or whether it represents greater awareness of the disease and/ or a better reporting system. In Switzerland for example, where a National Reference Center for listeriosis was established in 1987, following the Swiss cheese-related outbreak of listeriosis (9), the annual rate has been slightly decreasing (see Tab/,e 1). What is clear, however, is that the high- risk population is increasing, particularly in the Western hemisphere (elderly, immunocompro- mised patients, transplant patients, HIV-positive patients in particular). Likewise, the potential sources of acquisition of listeriosis by contami- nated food have also increased, because of major changes in food production, conservation and con- sumption. Strains causing diseases. It has long been ob- served that among the many different strains of L. monocytogenes (as determined by phenotypic or genotypic typing methods), a rather restricted number of them caused the m3:jority of human cases, and an even more restricted number was responsible for several of the most recent and large outbreaks (see Tab/,e 1). Of the 13 commonly described serovars of L. monocytogenes, only 3 (sv 4b, sv l/2a, and sv 1/ 2b) are responsible for the majority of all sporadic cases, as well as for all outbreaks reported so far (see Tab/,e 2). Serovar 4b accounted for almost half of the cases in Europe, whereas serovars 4b, l/2a, and l/2b were equally represented in North Amer- ica. No clear cut association seems to exist between a given serotype and a particular clinical form of listeriosis (bactaeremia versus meningitis or me- ningoencephalitis) or a particular host. Of particular interest, L. monocytogenes serovar 4b (or 4b related) strains have caused almost all recent outbreaks of foodbome listeriosis, posing the question of an increased virulence of this sero- var, a better adaptation to foods or to humans, or a broader distribution in the environment. No sig- nificant difference in pathogenicity has been con- vincingly established so far. L. monocytogenes strains of serovar 4b have a restricted genetic diversity when compared to strains ofserovar l/2a or l/2b, for example. Despite this limitation, it has been shown that L. monocytogenes serovar 4b strains isol- ated from animals (8) or from fish and fish prod- ucts (9) belong to different lineage of strains, when analysed by discriminatory subtyping methods. c J. Rocourt. Human listeriosis-1989. Document WHO/HPP/ FOS/91.3. World Health Organization, Geneva, 1991. 68 Thus the hypothesis of a host-specific adaptation of strains is a possible explanation for these recent observations. Carriage of L. monocytogenes. L. monocytogenes can be found as part of the normal flora of many ani- mal species and man. Highly variable carriage rates have been reported in the past, reflecting both the high diversity of the population studied, as well as the absence of separation between the nonpatho- genic (mainly L. innocua) and the pathogenic spe- cies. Recent studies have contributed to clarify this issue. Jensen in Denmark has looked at the car- riage rate of L. monocytogenes both in normal per- sons ( 10 ), and in patients with listeriosis ( 11) using discriminant selective media. In normal persons, the faecal carriage rate of L. monocytogenes was infe- rior to 1 % (3/348), whereas a vaginal carriage was absent from more than 800 genital samples. We can conclude from this and similar studies that an ascending route is unlikely in neonatal listeriosis, and also that faecal excretion of L. monocytogenes is not necessarily an indication of disease. However, the role of healthy carriers in the epidemiology of listeriosis is still unclear. Among patients with listeriosis, 21.6% (16/74) excreted L. monocytogenes in faeces, as opposed to none of 103 patients with diarrhoea. When deter- mined, the level of excretion was 104 colony-form- ing units (CFU)/g early on, and ceased after 2 weeks, arguing for a rather short duration of excretion. This study also pointed out that house- hold contacts of a listeriosis patient could excrete L. monocytogenes, and that an early screening could indicate common food items to investigate. In household contacts of United States patients with listeriosis, Schuchat et al. ( 12) have found a percentage of faecal carriers of 18%. Investigations carried out in relation with the Californian out- break in 1985 suggested that secondary transmis- sion by faecal carriage could occur ( 13). Listeria and food L. monocytogenes is a ubiquitous microorganism which has been isolated from various environ- ments (decaying vegetation, soil, polder, sewage water, etc.). Despite being asporogenous, it is resis- tant to diverse environmental conditions including low pH and high NaCl concentrations, and is microaerophilic and psychrophilic. The various ways in which the bacterium can enter a plant, its tenacity in the industrial environment, its ability to grow at refrigeration temperatures and to survive in food for prolonged periods under adverse con- ditions have made it one of the hottest topics for the food industry during the last decade. Entry of L. monocytogenes into food-processing plants is of- ten primarily due to animals which excrete the bacterium, raw food of animal origin and possibly healthy human carriers. The growth of Listeria is favoured by high humidity and nutrient waste. Rapp. trimest. statist. sanit. mond., 50 (1997) Table 1 Incidence of human listeriosis in European countries (cases/106 total population/ year) Tableau 1 Incidence de la listeriose humaine en Europe (cas/106 population totale/an) Country- Pays Data from ref. 7- Data from ref. 5- Data from ref. ~ Donnees de Ja ref. 7 Donnees de Ja ref. 5 Donnees de la ref. 8 (1989) Belgium - Belgique 4.8 Denmark - Danemark 2.3 8.0 6.0 Finland - Finlande 5.9 France 3.3 11.3 8.0 Germany - Allemagne 1.0 5.8 Netherlands - Pays-Bas 1.7 Norway - Norvege 4.0 1.6 Scotland - Ecosse 5.7 Switzerland - Suisse 2.1 6.0 5.0 Sweden - Suede 1.5 8.0 3.8 United Kingdom - Royaume-Uni 0.7 5.0 4.3 Yugoslavia - Yougoslavie 3.0 • J. Rocourt & J. Bille, Human listeriosis: 1991-1992 (Document WHO/FNU/FOS/97.1) 1997 Table 2 Major outbreaks of human listeriosis (>30 cases) Tableau 2 Principales flambees de listeriose humaine (>30 cas) Year(s)- No. of cases No. of perinataV Serovar- Annee(s) (deaths) non perinatal Serovar Nb de cas cases- (Nb de deces) Nb de cas perinatals/ non perinatals Europe Anjou, France 1976 162 134/28 4b Vaud, Switzerland - Vaud, Suisse 1983-87 122 (31) 63/59 4b United Kingdom - Royaume-Uni 1989-90 300 NK 4b,4bx France 1992 279 (85) 92/187 4b France 1993 39 31/8 4b France 1995 36 18/18 4b Outside Europe - Hors Europe Halifax, Canada 1981 41 (17) 34/7 4b Massachussetts, USA - Massachussetts, Etats-Unis d'Amerique 1983 49 (14) 7/42 4b California, USA - Californie, Etats-Unis d'Amerique 1985 142 (48) 93/49 4b Philadelphia, USA - Philadelphie, Etats-Unis d'Amerique 1986-87 36 (16) 4/32 multiple NK: not known - NK : inconnu (?): epidemiological link only-(?) : lien uniquement epidemiologique Wld hlth statist. quart., 50 (1997) Recent data·- Donnees recentes • (1992) 4.4 4.8 6.0 13 4.9 3.5 4.1 (1991) 2.2 Vehicle - Vehicule NK soft cheese - fromage a pate molle pate pork tongue in aspic - langue de pore en gelee rillettes (potted mince) - rillettes soft cheese - fromage a pate molle coleslaw- salade de chou cru milk ? - lait ? soft cheese - fromage a pate molle NK (multiple) - NK (multiples) 69 L. monocytogenes has been detected in drains, mois- ture and stagnant water, floors, and food-process- ing equipment. L. monocytogenes can attach itself to various kinds of surfaces (stainless steel, glass, rub- ber, etc.) and even biofilms. It survives on fingers after handwashing and in aerosol suspensions. Fur- thermore, contaminated effluents from food- processing plants, where the organism can grow, increase the spread of Listeria in the environment (14). L. monocytogenes is present in a wide variety of foods, both raw and processed. It can survive and readily multiply in several foods during storage. A survey by PHLS (Public Health Laboratory, London, UK) indicated 6% of 18 OOO foods to be contaminated with L. monocytogenes and 5% of the positive samples to contain more than 1 OOO CFU / g. In another study, 11 % of foods sampled from the refrigerator of patients suffering from listeriosis in United States of America were positive, with 10% of positive samples containing more than 100 CFU/g (15-17). Milk and cheeses (18-19, d) Milk. Milk can be contaminated by environmental sources including cow dung, soil, straw and, more rarely, by mastitis. Cows suffering from mastitis caused by L. monocytogenes are very rare but in such cases up to 103 CFU/ml may be excreted. The incidence of L. monocytogenes in raw milk in bulk tanks is low, ranging from 1 to 8%, possibly de- pending upon geographical location and season; concentrations in bulk tanks are usually very low. Nevertheless, raw milk must be considered by the dairy processor as a source of contamination com- ing into the plant. Conflicting results have ap- peared historically concerning the heat resistance of L. monocytogenes which can be partly explained by differential procedures and by the strains in- volved. However, a World Health Organization (WHO) informal working group on foodborne listeriosis concluded in 1988 that "pasteurization is a safe process which reduces the number of L. monocytogenes occurring in raw milk to levels that do not pose an appreciable risk to human health''.d Efforts to ensure that milk is safe from L. monocytogenes contamination should focus on promoting appropriate methods of pasteurization and on identifying and eliminating sources of post- pasteurization contamination. Cheeses. Large surveys including hundreds of samples suggest that 1-10% of cheeses (especially soft cheeses) are contaminated with L. mono- cytogenes. Studies on the behaviour of L. mono- cytogenes in various artificially contaminated d World Health Organization. Foodborne listeriosis - R.eport of a WHO informal Working Group, Geneva, 15-19 February, 1988. (Document WHO/EHE/FOS/88.5) 1988. 70 cheeses show that substantial numbers of L. mono- cytogenes cells survive the manufacture and ripen- ing of Trappist, brick, cheddar, blue, feta, colby, cottage and more particularly camembert cheeses. There is a highly significant correlation between Listeria growth and cheese pH values >5.5 and the absence of starter cultures during manufacturing. The concentration of L. monocytogenes in cheeses is usually low but levels higher than 1 OOO CFU/g may occur in 0,5-5 % of samples. Meat and meat products (20-22) The presence of Listeria on carcass is usually attrib- uted to contamination by faecal matter during the slaughter. The proportion of healthy carriers har- bouring Listeria in their intestine ranges from 11 to 52%. Studies in a slaughter line indicated that 45% of pigs harbour L. monocytogenes in the tonsils and 24% of cattle have contaminated internal retro- pharyngeal nodes. L. monocytogenes has been recov- ered from unclean and clean sections ( especially on workers' hands) in slaughterhouses and the most heavily contaminated working posts are cow dehiding and pig stunning and hoisting. A wide variety of meats can be contaminated with L. monocytogenes, usually on the surface. High per- centages of contaminated samples have been re- ported for ground meat and processed meat prod- ucts (from 10 to 80%). L. monocytogenes can survive in various meat (ground beef, liver) during storage at 4 and 25 °C. Different internal (composition, pH, Aw) or external (temperature, gaseous atmo- sphere, competitive flora) factors influence the de- velopment of L. monocytogenes in processed meat. The levels of contamination usually observed are low in raw meat and meat products, with 80-90% of samples below 10-100 CFU/g. However, higher levels have been reported for some ready-to-eat products, as observed for vehicles of outbreaks. Poultry (23-24) Studies in poultry slaughter houses failed to detect L. monocytogenes in feather samples, scalding tank water overflow, neck skin, liver, heart, caecum and large intestine; in contrast, L. monocytogenes was re- covered from feather picker drip water, chiller wa- ter overflow, recycling water for deaning gutters and in mechanically deboned meat. These find- ings demonstrate the significance of defeathering machines, chillers and recycled water in product cross-contamination. The prevalence of Listeria on the hands and gloves of the persons hanging birds after chilling, cutting carcasss and packaging parts were 16-20%, 33-45% and40-59% respectively. The prevalence of L. monocytogenes in turkey wings, legs and tails increased through processing and distri- bution. At this last stage, the prevalence of L. monocytogenes in turkey wings, legs, and tails was 20, 13 and 11 % respectively and in chicken legs, Rapp. trimest. statist. sanit. mond., 50 (1997) wings, and whole liver, it was 40, 13, and 26% respectively. Other surveys have indicated that 23- 66% of broiler chicken, 28% of ready-to-eat, pre- cooked chilled chicken and 15-54% of frozen chicken were contaminated with L. monocytogenes. Seafood and fishproducts (25-27) Estuary environments are continously exposed to potential contamination with L. monocytogenes from sewage effluents, processing effluents and agricul- tural runoff, and the salinity of estuary environ- ement does not affect the presence of L. mono- cytogenes. L. monocytogenes has been recovered from unprocessed shrimps in the Gulf of Mexico (11 % of samples positive) and from 1.5 % of retailed fresh-water fish. In a study done by the Food and Drug Agency (FDA) on frozen seafood products from various countries 26% of positive samples were shrimp (raw and cooked), cooked crabmeat and surimi. L. monocytogenes was found in 13% of fish products, ready-to-eat shrimps, crab and smoked salmon sampled at the wholesale level. Nine to 28% of smoked fish products were found positive for L. monocytogenes. Experiments examin- ing the behaviour of L. monocytogenes during the production and storage of smoked salmon indicat- ed that its concentration remained about the same during marination and smoking but a significant multiplication occurred during storage at 4-10 oc ; freezing to - 25 °C had no influence on the Listeria concentration. Low levels have been reported in ready-to-eat lobster and shrimps (0.2-2 CFU/g) and for frozen fish fingers (<lOOCFU/g) but higher levels (up to 104CFU/g) have been observed in smoked fish. Eggs Contamination of eggs with L. monocytogenes is rare and a survey of900 hen's eggs from retail markets failed to detect this pathogen. The presence of L. monocytogenes in raw, commercially broken egg due to shell contamination from faeces and the processing environment seems inevitable but it has rarely been isolated from commercial raw liquid whole egg. Prevention Prevention of listeriosis involves different types of actions: (i) prevention and control of food contamination at the farm level (silage of pH > 5 should not be fed to animals, fields should not be irrig- ated with effluents from sewage treatment plants, or fertilized with unheated sludge from sewage treatment plants; sound milk collec- tion techniques should be adopted, including teat-dipping, and sanitation of all milk contact surfaces before use, maintaining temperature Wld hlth statist. quart., 5D (1997) of the bulk-tank as low as possible; and measures based on the Hazard Analysis Criti- cal Control Point system (HACCP) approach, etc. should be applied); at the slaughterhouse (improvement of hygiene, clean animals on entry into the slaughterhouse, no piercing of intestinal wall ofanimals, HACCP approach); at the processing plant (good manufacturing practices, HACCP systems, cleanability of equipment, etc.); and at the retail level (HACCP approach, etc.); (ii) disease recognition (informing physicians to encourage early diagnosis of the disease, espe- cially in pregnant women); (iii) surveillance systems and epidemiological in- vestigations to develop active surveillance sys- tems to determine morbidity and mortality and to evaluate incidence in the various popu- lations at risk; in addition, case-control studies should be developped during outbreaks to identify the food vehicle, and for sporadic cases to identify foods at risk; (iv) dietary recommendations adapted to popula- tions at risk. Persons at increased risk for list- eriosis (i.e., pregnant women, the elderly and those with immmunosuppressive conditions) can lower the risk of listeriosis by avoiding the consumption of foods known to be associated with the disease and following basic food- handling hygiene practices that may also help prevent other foodborne diseases. Booklets with these recommendations have been issued in a number of countries. Summary Various epidemiological investigations of outbreaks and sporadic cases have clearly demonstrated that the consumption of contaminated food is responsible for a high proportion of listeriosis cases and Listeria monocy- togenes has been increasingly recognized as an impor- tant foodborne pathogen over the last 15 years. The emergence of listeriosis is the result of complex interac- tions of different factors: medical progress which in- creases the lifespan and allows immunodeficient people to survive, expansion of the food industry and cold storage systems as well as changes in food habits. None of these factors on its own is entirely responsible. Considerable research has attempted to characterize the organism, define the magnitude of the public health 71 problem and its impact on the food industry, identify the risk factors associated with the disease, and devise appropriate control strategies. Nevertheless, a number of crucial questions remains incompletely elucidated ( extent of the food borne transmission of listeriosis, health status of apparently "healthy patients" with the possible role of an intercurrent infection or genetic susceptibility, how to distinguish highly virulent from less virulent strains of L. monocytogenes, factors con- tributing to the emergence of outbreaks, the possible role of healthy carriers in the epidemiology of listeriosis, etc.). To investigate the complexity of listeriosis requires the close collaboration of clinicians, epidemiologists, clinical and food microbiologists, food scientists and the food industry. A large amount of data has been accumu- lated during the past 10 years but more research is required to elucidate the epidemiology of the disease and the virulence of the causative agent. Resume Listeriose d'origine alimentaire Diverses investigations epidemiologiques realisees sur des flambees et des cas sporadiques ant nettement demontre que la consommation d'aliments contamines est responsable d'une grande proportion des cas de listeriose, et ces 15 dernieres annees, !'importance de Listeria monocytogenes en tan! qu'agent de maladies d'origine alimentaire a ete de plus en plus largement reconnue. L'emergence de la listeriose est due a la conjonction de divers facteurs, dont aucun n'est entie- rement responsable: les progres de la medecine, qui allongent l'esperance de vie et permettent a des sujets immunodeficients de survivre, !'expansion des indus- tries alimentaires et des systemes de conservation par le froid, ainsi que les modifications des habitudes ali- mentaires. D'importants travaux ant ete consacres a la caracterisa- tion de Listeria, a la definition de l'ampleur du probleme de sante publique et de son impact sur l'industrie alimentaire, a !'identification des facteurs de risque associes a la maladie et a !'elaboration de strategies de lutte appropriees. Neanmoins, un certain nombre de questions cles restent a elucider (etendue de la trans- mission de la listeriose par voie alimentaire, etat de sante des malades «apparemment sains» et rOle even- tuel d'une infection intercurrente au d'une predisposi- tion genetique, distinction des souches hautement viru- lentes et peu virulentes de L. monocytogenes, facteurs contribuant a !'apparition de flambees, rOle possible des porteurs sains dans l'epidemiologie de la maladie, etc.). En raison de cette complexite, I' exploration de la listerio- se exige l'etroite cooperation de cliniciens, d'epidemio- logistes, de specialistes de microbiologie clinique et alimentaire, de chercheurs en sciences de l'alimenta- tion et de representants de l'industrie alimentaire. Une masse considerable de donnees ant ete rassemblees ces dix dernieres annees mais ii reste beaucoup a faire pour elucider l'epidemiologie de la maladie et la viru- lence de l'agent responsable. 72 References/Referencese 1. Nieman, R.E. & Lorber, B. Listeriosis in adults, a changing pattern: report of eight cases and a review of the literature, 1968-1978. Review of infectious diseases, 2: 207-227 (1980). 2. Schuchat,A. etal. Epidemiologyofhuman listeriosis. Clinical microbiology review, 4: 169-183 (1991). 3. Biila, CJ. et al. An epidemic of foodborne listeriosis in Western Switzerland: description of 57 cases involving adults. Clinical infectious diseases, 20: 66-72 (1995). 4. Swaminathan, B. et al. Listeria. In: Manual of Clinical Microbiology (Sixth ed.): 341-348. Washington DC, ASM Press, 1995. 5. Bille,J. et al. Anatomy of a listeriosis outbreak. In: Foodborne listeriosis. Proceedings of a symposium: 29-36. Hamburg, B. Behr's Gmbh & Co., 1989. 6. Ralovich, B. Listeriosis research-present situation and perspective. Akademiai Kiado, Budapest, 1984. 7. Bille, J. Epidemiology of human listeriosis in Europe, with special reference to the Swiss outbreak. In: AJ. Miller, J.L. Smith, and G.A. Somkuti (ed.), Foodborne listeriosis. Society for Industrial Microbiology: 71-74. New York, Elsevier Science Publishing, Inc., 1990. 8. Boerlin, P. & Piffaretti,J.C. Typing of human, animal, food, and environmental isolates of Listeria monocytogenes by multilocus enzyme electrophoresis. Applied environmental microbiology, 57: 1624-1629 (1991). 9. Boerlin, P. et al. Typing Listeria monocytogenes isolates from fish products. Applied and environmental microbiology ( 1997 in press). 10. Jensen, A. Listeria in faecal and genital specimens. Medical microbiology utters, 2: 125-130 (1993). 11. Jensen, A. Excretion of Listeria monocytogenes in faeces after listeriosis: rate, quantity and duration. Medical microbiology utters, 2: 176-182 (1993). 12. Schuchat, A. et al. Gastrointestinal carriage of Listeria monocytogenes in household contacts of patients with listeriosis. Journal of infectious diseases, 167: 1261-1262 (1993). 13. Mascola, L. et al. Fecal carriage of Listeria monocytogenes - Observations during a community-wide, common-source outbreak. Clinical infectious diseases, 15: 557-558 (1992). 14. Rorvik, L.M. et al. Contamination pattern of Listeria monocytogenesand other Listeria in a salmon processing plant. International journal of food microbiology, 25: 19-27 (1995). 15. Farber,J.M. & Peterkin, P.I. Listeria monocytogenes, a food- borne pathogen. Microbiology review, 55: 4 76-511 (1991). 16. Mcl..auchlin, J. & Gilbert, R.J. Listeria in food. PHLS Microbiology digest, 7: 54-55 (1990). 17. Pinner, R.W. et al., and the Listeria Study Group. Role of foods in sporadic listeriosis. II. Microbiologic and epidemiologic investigation. Journal of the American Medical Association, 267: 2046-2050 (1992). 18. EI-Gazzar, F.E. & Marth, E.H. Listeria monocytogenes and listeriosis related to milk, milk products and dairy ingredients: a review - II. Listeria monocytogenes and dairy technology. Milchwissenschaf~ 46: 82-86 (1991). 19. Husu, J.R. et al. Contamination of raw milk by Listeria monocytogenes on dairy farms. Journal of veterinary medicine B, 37: 268-275 (1990). 20. Buncie, S. The incidence of Listeria monocytogenes in slaughtered animals, in meat, and in meat products in Yugoslavia. International Journal of food microbiology, 12: 173- 180 (1991). 21. Gilbert, R.J. et al. The contamination of pate by Listeria monocotygogenes in England and Wales in 1989 and 1990. Epidemiology and infections, llO: 543-551 (1993). e Additional references are available on request from the authors. Rapp. trimest. statist. sanit. mond., 50 (1997) 22. Skovgaard, N. 8c Morgen, C.A. Detection of Listeria spp. in faeces from animals, in feeds, and in raw foods of animal origin. International journal of food microbiowgy, 6: 229-242 (1988). 23. Genigeorgis, C.A. et al. Prevalence of Listeria. spp. in poultry meat at the supermarket and slaughterhouse level.journalfar food protection, 52: 618-624 (1989). 24. Kerr, K.G. et al. Incidence of Listeria spp. in pre-cooked, chilled chicken products as determined by culture and Wld hlth statist. quatt., 50 (1997) enzyme-linked immunoassay (ELISA). journal Jar food protection, 53: 606-607 (1990). 25. Dillon, R. et al. Occurrence of Listeria in hot and cold smoked seafood products. International journal of food microbiowgy, 22: 73-77 (1994). 26. Farber,J.M. Listeria monocytogenesin fish products.Journal far food protection, 54: 922-934 ( 1991). 27. Fuchs,R.S.&Nicolaides,L.IncidenceofListeriainhot-andcold- smoked fish. Letters of apPlied microbiowgy, 19: 394-398 ( 1994). 73 Enterohaemorrhagic Escherichia coli Yoshifumi Takedaa History of the discovery Enterohaemorrhagic Escherichia coli (EHEC) is characterized by its ability to produce a toxin that is cytotoxic to Vero cells. This toxin was first de- scribed by Konowalchuk et al. ( 1 ). While they were screening a heat-labile enterotoxin (LT) produced by enterotoxigenic E. coli, they found that strains of certain serotypes such as 018, 026, 068, 0111, and 0138 produced a cytotoxin to Vero cells that was distinct from LT. This finding was confirmed soon after by Wade et al. (2) and Scotland et al. ( 3 ), but these results did not receive significant atten- tion for some years. O'Brien et al. (4), on the other hand, had been working a cytotoxin to a certain cell line of HeLa Cells produced by a strain of E. coli 026, and in 1982 they reported that the toxin was neutralized by antitoxin to Shiga toxin produced by Shigella dysenteriae type 1. They named this toxin Shiga-like toxin. Almost at the same time, there was a report on a unique outbreak due to E. coli which occurred in the States of Oregon and Michigan in the United States of America (5). This outbreak was due to consumption of hamburgers and the unique fea- tures of this foodborne infection were, firstly, that one of the major symptoms was excretion of bloody diarrhoea which in some cases was described as "all blood and no stool", thus the symptom being named haemorrhagic colitis; secondly, a causative E. coli isolated belonged to a rare serotype 0157:H7; and lastly, the organism neither pro- duced any reported E. coli toxins, such as a heat- labile enterotoxin (LT) and a heat-stable entero- toxin (ST), nor was it an enteroinvasive E. coli (5). Having recognized that the E. coli 0157:H7 iso- lated was a new enteropathogen, Levine ( 6) named it enterohaemorrhagic E. coli. Nomenclature of the toxin Konowalchuk et al. ( 1) designated the toxin they found as Verocytotoxin because it is cytotoxic to Vero cells.Johnson et al. (7) recognized that E. coli 0157:H7 isolated from patients with haemorrhagic colitis in Canada produced a Verocytotoxin similar to that which was reported by Konowalchuk et al. ( 1). On the other hand O'Brien et al. (4), who had a Director-general, Research Institute, International Medical Center of Japan, Tokyo, Japan. 74 named the toxin a Shiga-like toxin, recognized that a Verocytotoxin reported by Johnson et al. was the same as the Shiga-like toxin (8). Moreover they found that the E. coli 0157:H7 isolated from the hamburger case (5) produced a Shiga-like toxin. Since then the same toxin has been called by two names, Verocytotoxin (or Verotoxin) and Shiga-like toxin. In early 1996, 16 scientists from the United States, Germany and Japan published a note in ASN news (9) proposing that the toxin be grouped into a Shiga toxin family, thus avoiding different names for the same toxin, and that the E. coli producing the toxin be called Shiga toxin- producing E. coli. Shiga toxin family (Verotoxins, Shiga-like toxins) Enterohaemorrhagic E. coli (Shiga toxin-produc- ing E. coli) produce two distinct types of the toxin, that is, Shiga toxin 1 (Stxl=VTl and SLTl) and Shiga toxin 2 (Stx2=VT2 and SL T2), and several variants of both toxins have been reported ( JO). The nucleotide sequences of both Stxl and Stx2 genes (stx1 and st"2, respectively) as well as those of the variants of both toxins have been determined. The amino acid sequences of the Stxl were essen- tially identical to those of Shiga toxin produced by Shigella dysenteriae type 1, while those of the Stx2 shared about 55-60% homology to those of Stxl. Purification ofStxl and Stx2 were first achieved by O'Brien et al. (11) and Yutsudo et al. (12), re- spectively. Purified toxins showed lethal toxicity to mice and other small experimental animals, and enterotoxicity resulting in fluid accumulation in the rabbit ileal loop, in addition to the cytotoxicity to Vero cells and a certain line of HeLa cells. The mode of action of the Shiga toxin and Shiga toxin family has been well characterized. As early as 1976, it was demonstrated that Shiga toxin inhibits protein synthesis of eukaryotic cells ( 13). Soon after that it was found that the inhibition was due to the inactivation of 60S ribosomal subunits ( 14 ), which was demonstrated to be due to an inhi- bition ofEF-1 dependent aminoacyl-tRNA binding to the subunit in both Shiga toxin ( 15) and Shiga toxin family (VTl and VT2) ( 16, 17). Further mo- lecular analysis of the mode of action by Endo et al. (18) elegantly showed that Shiga toxin and VT2 exhibited an RNA N-glycosidase activity that cleaved the N-glycosidic bond of the adenosine residue at position 4324 from the 5' terminus of Rapp. trimest. statist. sanit. mond., 50 (1997) Table 1 Major outbreaks of enterohaemorrhagic E. coli infection Tableau 1 Principales flambees d'infection a E. coli enterohemorragique Year- Location - Lieu Setting - Cadre Causative agent - Agent causal Serotype- Ref.- An nee Serotype Ref. 1982 Oregon, USA Community - Communaute Hamburger 0157:H7 (5) 1982 Michigan, USA Community - Communaute Hamburger 0157:H7 (5) 1982 Ontario, Canada Nursing home - Hamburger 0157:H7 (28) Maison de retraite 1983 Labrador, Canada Community - Communaute NIA- S/0 0157:H7 (28) 1984 Nebraska.USA Nursing home - Hamburger 0157:H7 (29) Maison de retraite 1984 North Carolina - Caroline du Nord, USA Day care center - Garderie NIA- S/0 0157:H7 (30) 1985 Ontario, Canada Nursing home - Cold sandwich - 0157:H7 (31) Maison de retraite Sandwich froid 1985 United Kingdom - Royaume-Uni Community - Communaute Raw potato- 0157:H7 (32) Pommes de terre crues 1985 Washington, USA Community - Communaute Ground beef - Bmuf hache 0157:H7 (33) 1986 Ontario, Canada School - Ecole Raw milk - Lait cru 0157:H7 (34) 1987 Utah, USA Institution for the mentally Ground beef - Bmuf hache 0157:H7 (35) handicapped - Etablissement pour handicapes mentaux 1988 Minnesota, USA School - Ecole Precooked meat patties - 0157:H7 (36) Petits pates a la viande precuits 1989 Missouri, USA Community - Communaute Municipal water - 0157:H7 (37) Eau de distribution 1990 North Dakota - Dakota du Nord, Community - Communaute Roast beef - Roti de bmuf 0157:H7 (38) USA 1990 Hamilton, USA Psychogeriatric ward - NIA- S/0 0157:H7 (39) Service de psychogeriatrie 1990 Edinburgh, UK - Restaurant 0157:H7 (40) Edimbourg, Royaume-Uni 1991 Oregon, USA Community - Communaute NIA- S/0 0157:H7 (41) 1991 Massachusetts, USA Community - Communaute Freshly-pressed apple cider - 0157:H7 (42) Cidre frais (moOt) 1991 Canada Community - Communaute NIA- S/0 0157:H7 (43) 1991 London, UK- Community - Communaute Yogurt 0157:H7 (44) Landres, Royaume-Uni 1992 Germany - Allemagne Day care center - Garderie NIA- S/0 0157:H7 (45) 1992 Italy - ltalie Community - Communaute NIA- S/0 0111:H- (46) 1993 Washington, Idaho, Community - Communaute Hamburger 0157:H7 (47) California & Nevada - Washington, Idaho, Californie & Nevada, USA 1993 Italy- ltalie Community - Communaute NIA- S/0 0158, 0111 (48) and- et 086 1995 Australia - Australie Community - Communaute Metwurst - Saucisse de viande 0111 :H- (49) 1996 Scotland, UK - Community - Communaute Cold cooked meat - 0157 (57) Ecosse, Royaume-Uni Viande cuite froide N/A: not associated with food. - S/0: flambee non associee a des aliments. Wld hlth statist. quart., 50 (1997) 75 Table 2 Outbreak of enterohaemorrhagic E. coli (EHEC) food poisoning in Japan, 1996 Tableau 2 Flambees d'intoxications alimentaires a E. coli enterohemorragique (EHEC), Japan, 1996 Month-Mais Place -Lieu Setting - Cadre Number of patients - Serotype- Nombre de malades Serotype May-Mai Okayama Primary school/Kindergarten - 468 0157:H7 Ecole primaire/Jardin d'enfants June-Juin Gifu Primary school - Ecole primaire 371 0157:H7 June-Juin Hiroshima Primary school - Ecole primaire 185 0157:H7 June-Juin Aichi Secondary school - Ecole secondaire 21 0157:H7 June-Juin Fukuoka Nursery - Creche 48 0157:H7 June-Juin Okayama Primary school - Ecole primaire 364 0157:H7 June-Juin Osaka Nursery- Creche 50 0157:H7 June-Juin Tokyo Company - Entreprise 191 0157:H7 July-Juillet Gunma Primary school - Ecole primaire 144 0157:H7 July-Juillet Osaka (Sakai) Primary school - Ecole primaire 5 727 0157:H7 July-Juillet Osaka Nursing home - Maison de retraite 98 0157:H7 July - Juillet Wakayama Nursing home - Maison de retraite 12 0157:H7 July - Juillet Kyoto Company - Entreprise 74 0157:H7 July-Juillet Wakayama Nursing home - Maison de retraite 22 0157:H7 July- Juillet Ishikawa Secondary school - Ecole secondaire 247 0118:H2 August - AoOt Hokkaido Hospital - Hiipital 10 0157:H7 September - Septembre Iwate Primary school - Ecole primaire 218 0157:H7 October - Octobre Hokkaido Nursery - Creche 182 0157:H7 Source: Based on/D'apres: A report of the Ministry of Health and Welfare, Japan. - Rapport du Ministere de la Sante et des Affaires sociales, Japan. Table 3 Outbreak of enterohaemorrhagic E. coli (EHEC) food poisoning in Japan, 1984-1994 Tableau 3 Flam bees d'intoxications alimentaires a E. coli enterohemorragique (EHEC), Japan, 1984-1994 Year-Annee Place- Lieu Setting - Cadre Number of patients - Serotype- Nombre de malades Serotype 1984 Tokyo Primary school - Ecole primaire 100 0145:H- 1986 Ehime Nursery- Creche 22 0111:H- 1990 Saitama Kindergarten - Jardin d'enfants 319 0157:H7 1991 Osaka Nursery- Creche 161 0157:H7 1991 Niigata Primary school - Ecole primaire 234 0111:H- 1991 Tokyo Primary school - Ecole primaire 89 OUT:H19 1992 Saga Nursery - Creche 11 0157:H7 1993 Tokyo Primary school - Ecole primaire 165 0157:H7 1993 Tokyo Nursery - Creche 40 0157:H7 1994 Nara Primary school - Ecole primaire 250 0157:H- Source: Based on/D'apres: Reports of the Ministry of Health and Welfare, Japan. - Rapports du Ministere de la Sante et des Affaires sociales. Japan. 76 Rapp. trimest. statist. sanit. mond., 50 (1997) the 28S ribosomal RNA of 60S ribosomal subunit of rabbit reticulocytes. It was subsequently demon- strated that other toxins of the Shiga family also showed the same enzymatic activity ( 19, 20). Inter- estingly, this molecular mode of action of Shiga toxin family is exactly the same as that of a plant lectin, ricin, reported by Endo et al. (21, 22). Later, it was found that the sequence of amino acids around a putative active centre of ricin and Shiga toxin family is quite homologue (23-27), although overall homology is only about 20-21 %. Foodborne outbreaks due to EHEC Since the first report of a foodborne outbreak due to E. coli 0157:H7 in the United States in the early 1980s (5), many more have been reported. Tabl.e 1 summarizes major outbreaks reported in the United States, Canada, Australia and European countries, which suggests that E. coli 0157:H7 out- breaks, mainly foodborne, first frequently oc- curred in the United States and Canada and then in European countries. Swerdlow et al., at the thirty-second Joint Con- ference of US-Japan Cooperative Medical Science Program, Cholera and Related Diarrhoeal Diseases Panel, during November 14-16, 1996 in Nagasaki, Japan, reported that in the United States during the last 14 years, more than 100 outbreaks were reported, and that in 1996 the number of infected patients and deaths had been estimated at 20 OOO and over 100, respectively.b In Scotland, a foodborne outbreak due to E. coli 0157:H7 was reported in November 1996 in which over 400 persons were affected, with 17 fatal- ities, mainly the elderly, by the end of January 1997 (57). The investigation has concluded that gravy, which was prepared in a butcher's shop and served with cooked steaks for a church lunch, was associ- ated with this outbreak. In addition to these E. coli 0157:H7 outbreaks, several other serotypes ofEHEC also caused similar outbreaks associated with contaminated foods as listed in Tabl.e 1. Interestingly, they were outside the American continent. In 1996, an unexpectedly large number of out- breaks due to E. coli 0157:H7 occurred in Japan (Tabl.e 2). A total of 18 were reported from May through October. Among them, one at Sakai City in July and August was extremely large and more than 5 700 patients were reported (56). A unique characteristic of these Japanese outbreaks was that 8 of 18 occurred in primary and secondary schools due to luncheons provided by schools. In Japan, most schools in the country prepare luncheons for b Swerdlow et al. The thirty-second Joint Conference of US. Japan Cooperative Medical Science Program, Cholera and Related Diarrheal Diseases Panel, November 14-16, 1996, Nagasaki, Japan. Wld hlth statist. quart., 50 (1997) children in their own school-kitchen or in a central kitchen in a city or a province. In Sakai, the central kitchen of the city prepared the luncheon for al- most 50 OOO children and one of the dishes con- tained radish sprouts contaminated by E. coli 0157:H7.c An outbreak of E.coli Ol 18:H2 was also reported in July in which 247 schoolchildren were affected (Tabl.e 2). Fortunately, however, the num- ber of deaths in 1996 was only 5 of 8 432 patients. The lower mortality, compared to that in other countries, remains to be examined. Before these events in 1996, several EHEC out- breaks were reported in Japan, they are summa- rized in Tabl.e 3. These data suggest that since around 1990, EHEC is spreading all over the coun- try of Japan. Nevertheless, it is difficult to explain why so many outbreaks have occurred, spreading suddenly throughout the country from north to south, in 1996. Sporadic cases of EHEC infection in Japan Sporadic cases of EHEC infection have been re- ported in Japan since the 1980s. Data in Tabl.e 4 reported by Kai et al. (50) indicate that several sporadic infections have occurred every year since 1979. However, the number of reported cases have increased significantly since 1990, when a big out- Table 4 Sporadic cases of EHEC infection in Japan, 1979-1995 Tableau 4 Cas sporadiques d'infections a EHEC au Japan, 1979-1995 Year-Annee Number of cases - Serotype 0157:H? - Others- Nombre de cas Serotype 0157:H? Autres 1979-1983 3 3 1984 6 2 4 1985 8 6 2 1986 3 2 1 1987 15 9 6 1988 5 4 1989 5 3 2 1990 17 13 4 1991 87 76 11 1992 68 63 5 1993 80 78 2 1994 70 60 10 1995 46 40 6 Total 413 353 60 Source: Ref. - Ref. (50). c Michino, H. et al. The thirty-second Joint Conference of US. Japan Cooperative Medical Science Program, Cholera and Related Diarrheal Diseases Panel, November 14-16, 1996, Nagasaki,Japan. 77 Table 5 Sporadic cases of haemolytic uraemic syndrome (HUS) in Japan, 1983-1995 Tableau 5 Cas sporadiques de syndrome hemolytique-uremique au Japan, 1983-1995 Year-Annee Number of cases - Nombre de cas 1983-1988 11 1989 10 1990 17 1991 29 1992 37 1993 41 1994 42 1995 45 Total 232 Source: Takeda, T. et al. Unpublished data. - Donnees non publiees. break, with 2 victims due to E. coli 0157 occurred in Saitama, near Tokyo metropolitan area (Tabl,e 3). Interesting data from the National Children's Medical Research Center (Tabl,e 5) may suggest that the number of EHEC, especially E. coli 0157: H7 infections in Japan, is much more than that listed in Tabl,e 4. Tabl,e 5 shows the number of annual incidence of haemolytic uraemic syndrome (HUS) reported to the National Children's Medical Cen- ter in Tokyo. If we assume that more than 90% of HUS is due to Verotoxin-producing E. col~ as re- ported by Milford et al. (51), and about 10% of EHEC infected children developed HUS, figures in Tabl,e 5 would suggest that there were approxi- mately 2 OOO sporadic EHEC infections during 1983-1995 in Japan. It should be noted that the data in Tabl,e 5 did not cover the whole country. Prevention and control of foodborne EHEC infection Foods associated in outbreaks are reported to be predominantly meat products such as hamburgers. The reservoir of this pathogen appears to be main- ly cattle so that transmission may occur principally through consumption of contaminated foods, such as raw or undercooked meat products and raw milk. In addition, contamination of water and other foods with manure from the carriers in cattle and human populations, as well as cross contamination in food preparation, will also lead to infection. Apple cider and yogurt which are normally consid- ered not to be associated with foodborne diseases because of their acidity, and fresh vegetables such as radish sprouts are also reported as causative foods. Person to person infection, which may occur as secondary infections in large scale endemics, 78 and infections from swimming in contaminated surface water have been reported. To prevent foodborne EHEC infection, control measures at all stages of the food chain are re- quired, particularly the introduction of bacteri- cidal treatments such as heating ( e.g., pasteuriza- tion) or irradiation. The application of preventive measures such as the Hazard Analysis Critical Con- trol Point (HACCP) system should be emphasized. For mass catering establishments WHO has issued important rules,d and for families the strict adher- ence to the WHO golden rules for safe food prepa- ratione are recommended. As a follow-up to the endemics in Japan, the Government has put in place a number of preven- tive measures, with emphasis on guidance and veri- fication to ensure hygienic practice in mass cater- ing facilities in schools as well as other food hand- ling establishments. A possible development of candidate vaccine Development of a vaccine against EHEC is another promising strategy to control the infection. Thus, several investigators have been working to develop candidate vaccines (52-55). The strategy is to iden- tify and replace the amino acids responsible for RNA Nglycosidase activity of the Shiga toxin family by site-directed mutagenesis of the nucleotides to encode these amino acids (25-27). Wild type gene of the host E. coli can be replaced by these geneti- cally modified genes using an appropriate suicide vector. The resulting E. coli 0157:H7 with a non- toxic Verotoxin (either Stxl or/and Stx2) gene was non-toxic in animal experiments, thus it would be possible, when orally administered to animals (such as cattle, sheep or others) or humans, to stimulate local mucosal immunity to protect against E. coli 0157:H7 infections. Summary Enterohaemorrhagic Escherichia coli produces Vero- toxins (Shiga-like toxins) and causes diarrhoea, hae- morrhagic colitis and haemorrhagic uraemic syndrome. The mode of action of the Shiga toxin family has been well characterized in numerous studies which have demonstrated that the toxin has ANA N-glycosidase activity, which inhibits the binding of EF-1-dependent aminoacyl-tRNA to the 60S ribosomal subunit of eukary- otic cells, thus resulting in an inhibition of protein synthe- sis of the cells. In 1996, an unexpectedly large number of outbreaks of food poisoning due to E. coli 0157:H7 occurred in d Hygiene in food-seivice and mass catering establishments, important rules, WHO, Geneva, 1994 (Document WHO/ FNU/FOS/94.5). e The WHO golden rules for safe food preparation (poster), Food Safety and Food Aid Programme, WHO, Geneva, 1992. Rapp. trimest. statist. sanit. mond., 50 (1997) Japan. In addition, many sporadic cases of E. coli 0157:H? were also reported and a total of about 8 400 patients were recorded during that year. In most of these cases, the food vectors could not be identified. The possibility of developing a candidate vaccine is also discussed in this article. Resume Escherichia coli enterohemorragique Les Escherichia coli enterohemorragiques produisent des verotoxines (toxines «Shiga-like») et provoquent des diarrhees, une colite hemorragique et un syndrome hemolytique-uremique. Le mode d'action des toxines de type Shiga est connu, de nombreuses etudes ayant demontre une activite ARN N-glycosidase, qui inhibe la liaison de l'aminoacyl-t-ARN dependant de l'EF-1 a la sous-unite ribosomale 60S des cellules eucaryotes, ce qui entraine une inhibition de la synthese proteique. Une serie de flambees d'intoxications alimentaires d'ampleur inattendue, due a E. co/i0157:H7, a frappe le Japan en 1996. De nombreux cas sporadiques ont en outre ete rapportes et au total. pres de 8 400 cas ant ete enregistres pendant l'annee. Dans la plupart des cas, !'aliment responsable n'a pu etre identifie. Cet article expose egalement les possibilites demise au point d'un vaccin candidat. References/References 1. Konowalchuk, J. et al. Vero response to a cytotoxin of Escherichia coli. Infection and immunity, 18: 775-779 (1977). 2. Wade, W.G. et al. Cytotoxic enteropathogenic Escherichia coli. Lancet, ii: 1235-1236 (1979). 3. Scotland, S.M. et al. Production ofa cytotoxin affecting Vero cells by strains of Escherichia coli belonging to traditional enteropathogenic serogroups. FEMS microbiology letters, 7: 15- 17 (1980). 4. O'Brien, A.D. et al. Production of Shigella dysenteritll type 1- like cytotoxin by Escherichia coli. Journal of infectious disease, 146: 763-769, (1982). 5. Riley, L.W. et al. Haemorrhagic colitis associated with a rare Escherichia coli serotype. New England journal of medicine, 308: 681-685 (1983). 6. Levine, M.M. Escherichia coli that cause diarrhea: Entero- toxigenic, enteropathogenic, enteroinvasive, entero- haemorrhagic, and entero-adherent. Journal of infectious disease, 155: 377-389, (1987). 7. Johnson, W.M. et al. Cytotoxic Escherichia coli 0157:H7 associated with haemorrhagic colitis in Canada. Lancet, i: 76 (1983). 8. O'Brien, A.D. et al. Escherichia coli O 15 7:H7 strains associated with haemorrhagic colitis in the United States produce a Shigella dysenteritll 1 (Shiga)-like cytotoxin. Lancet, i: 702 (1983). 9. Calderwood, S.B. etal. Proposed new nomenclature for SL T (VT) family. ASN News, 62: 118-119 (1996). 10. Takeda, Y. etal. Vero toxins (Shiga-like toxins) produced by enterohaemorrhagic Escherichia coli (Verocytotoxin- Wld hlth statist. quart., SO (1997) producing E. colt). Microbiology and immunology, 37: 591-599 (1993). 11. O'Brien, A.D. et al. Purification and characterization of a Shigella dysenterilll I-like toxin produced by Escherichia coli. Infection and immunity, 40: 675-683 (1983). 12. Yutsudo, T. etal. Purification and some properties ofa Vero toxin from Escherichia coli 0157:H7 that is immunologically unrelated to Shiga toxin. Microbial pathogenesis, 3: 21-30 (1987). 13. Thompson, M.R. etal. Inhibition ofin vitro protein synthesis by Shigella dysenteriae 1 toxin. Biochemical biophysical research communication, 71: 783-788 (1976). 14. Reishig, R.S. et al. The cytotoxic activity of Shigella toxin. Evidence for catalytic inactivation of the 60S ribosomal subunit. Journal of biological chemistry, 256: 8739-8744 ( 1981). 15. Brown, J.E. et al. Shiga toxin from Shigella dysenteritll 1 inhibits protein synthesis in reticulocyte lysate by inactivation of aminoacyl-tRNA binding. Microbial pathogenesis, l: 325-334 (1986). 16. Igarashi, K. et al. Inhibition of elongation factor 1- dependent aminoacyl-tRNA binding to ribosomes by Shiga- like toxin I (VTl) fromEscherichiacoli0157:H7 and by Shiga toxin. FEMS Microbiowgy letters, 44: 91-94 (1987). 17. Ogasawara, T. etal. Inhibition of protein synthesis bya Vero toxin (VT2 or Shiga-like toxin II) produced by Escherichiacoli 0157:H7 at the level of elongation factor I-dependent aminoacyl-tRNA binding to ribosomes. Microbial pathogenesis, 4: 127-135 (1988). 18. Endo, Y. et al. Site of action of a Vero toxin (VT2) from Escherichia coli 0157:H7 and of Shiga toxin on eukaryotic ribosomes. European journal of biochemistry, 171: 45-50 ( 1988). 19. Saxena, S.K. etal. Shiga toxin, Shiga-like toxin II variant, and ricin are all single-site RNA N-glycosidases of28S RNA when microinjected into &nopus oocytes. Journal of biowgical chemistry, 264: 596-601 (1989). 20. Furutani, M. et al. Demonstration of RNA N-glycosidase activity of a Vero toxin (VT2 variant) produced by Escherichia coli 091: H21 from a patient with the hemolytic uremic syndrome. Microbiology and immuno/ngy, 34: 387-392 (1989). 21. Endo, Y. et al. The mechanism ofaction ofricin and related toxic lectins on eukaryotic ribosomes. Journal of biowgical chemistry, 262: 5908-5912 (1987). 22. Endo, Y. et al. RNA N-glycosidase activity of ricin A-chain. journal of biological chemistry, 262: 8128-8130 (1987). 23. Hovde, CJ. etal. Evidence thatglutamicacid 167 isan active- site residue of Shiga-like toxin I. Proceedings of the National Academy a/Sciences of the United States of America, 85: 2568-2572 (1988). 24. Jackson, M.P. et al. Mutational analysis of the Shiga toxin and Shiga-like toxin II enzymatic subunits. journal of bacteriology, 172: 3346-3350 ( 1990). 25. Yamasaki, S. et al. Importance of arginine at position 170 of the A subunit of Vero toxin I produced by enterohae- morrhagic Escherichia coli for toxin activity. Microbial pathogenesis, 11: 1-9 (1991 ) . 26. Ohmura, M. et al. Characterization of non-toxic mutant toxins of Vero toxin I that were constructed by replacing amino acids in the A subunit. Microbial pathogenesis, 15: 169- 176 (1993). 27. Cao, C. et al. Construction of mutant genes for a non-toxic Verotoxin 2 variant (VT2vpl) of Escherichia coli and characterization of purified mutant toxins. Microbiology and immunology, 38: 441-447 (1994). 28. Hockin,J.C. &: Lior, H. Haemorrhagic colitis and hemolytic uremic syndrome caused by Escherichia coli 0157:H7 in Canada. Canada disease weekly repart, 13: 203-204 (1987). 29. Ryan, C.A. etal. Escherichiacoli0157:H7 diarrhea is a nursing home: clinical, epidemiological, and pathological findings. journal of infectious disease, 154: 631-638 (1986). 30. Spika,J. S. et al. Hemolytic uremic syndrome and diarrhea associated with Escherichia coli0157:H7 in a day care center. Journalofpediatrics, 109: 287-291 (1986). 79 31. Carter, A.O. et al. A severe outbreak of Escherichia coli 0157:H7-associated haemorrhagic colitis in a nursing home. New England journal of medicine, 316: 1496-1500 (1987). 32. Morgan, G.M. etal. First recognized community outbreak of haemorrhagic colitis due to verotoxin-producing Escherichia coli0157:H7 in the UK. Epidemiol.og;y and infection, 101: 83-91 (1988). 33. Ostroff, S.M. et al. A statewide outbreak of Escherichia coli 0157:H7 infections in Washington state. American journal of epidemiol.og;y, 132: 239-247 (1990). 34. Duncan, L. et al. Outbreak of gastrointestinal disease in Samia, Ontario. Ontario disease surveillance report, 7: 604-611 (1986). 35. Pavia, A. T. et al. Hemolytic-uremic syndrome during an outbreak of EscherichiacoliO l57:H7 infections in institutions for mentally retarded persons: clinical and epidemiologic observations. Journal of pediatrics, 116: 544-551 ( 1990). 36. Belongia, E.A. et al. An outbreak of Escherichia coli O 15 7:H7 colitis associated with consumption of precooked meat patties.Journal of infectious disease, 164: 338-343 (1991). 37. Swerdlow, D.L. et al. Waterborne outbreak in Missouri of Escherichia coli O 157: H7 associated with bloody diarrhea and death. Annals of internal medicine, 116: 812-819 (1992). 38. Foodborne outbreak of gastroenteritis caused by Escherichia coli 0157:H7 - North Dakota, 1990. Morbidity and mortality weekly report, 40: 265-267 (1991). 39. Kohli, H.S. et al. A severe outbreak of Escherichia coli O 15 7 in two psychogeriatric wards. Journal of public health medicine, 16: 11-15 (1994). 40. Marsh,J. et al. A restaurant-associated outbreak of Escherichia coli 0157 infection.Journal of public health medicine, 14: 78-83 (1992). 41. Keene, W.E. et al. A swimming-associated outbreak of haemorrhagic colitis caused by Escherichia coli O 15 7:H7 and Shigella sonnei. New England journal of medicine, 331: 5 79-584 (1994). 42. Besser, R.E. et al. An outbreak of diarrhea and hemolytic- uremic syndrome from Escherichia coli Ol57:H7 in fresh- pressed apple cider. journal of the American Medical Association, 269: 2217-2220 (1993). 43. Orr, P. et al. An outbreak of diarrhea due to verotoxin- producing Escherichia coli in Canadian Northwest Territories. Scandinavian journal of infectious diseases, 26: 675-684 ( 1994). 44. Morgan, D. et al. Verotoxin producing Escherichia coli0157 infections associated with the consumption of yoghurt. Epidemiol.og;y and infection, 111: 181-187 (1993). 45. Reida, P. et al. An outbreak due to enterohaemorrhagic Escherichia coli 0157:H7 in a children day care center characterized by person-to-person transmission and 80 environmentalcontamination. lnternationaljournalofmedical microbiol.og;y, virol.og;y, parasitol.og;y and infectious diseases, 281: 534-543 ( 1994). 46. Caprioli, A. et al. Community-wide outbreak of hemolytic- uremic syndrome associated with non-0157 verotoxin- producing Escherichia coli.Journalof infectious disease, 169: 208- 211 (1994). 47. Bell, B.P. et al. A multistate outbreak of Escherichia coli 015 7:H7-associated bloody diarrhea and hemolytic-uremic syndrome from hamburgers. Journal of the American Medical Association, 272: 1349-1353 (1994). 48. Tozzi, A.E. et al. A community outbreak of haemolytic- uremic syndrome in children occurring in a large area of northern Italy over a period of several months. Epidemiol.og;y and infection, 113: 209-219 (1994). 49. Community outbreak of hemolytic uremic syndrome attributable to Escherichia coli O 111 :NM - South Australia 1995. Morbidity and mortality weekly report, 44: 550-551, 55 7-558 (1995). 50. Kai, A. et al. Isolation ofVerotoxin-producing Escherichia coli in Japan (in Japanese). Rinsho to Biseibutsu, 23: 827-834 (1996). 51. Milford, D.V. et al. Haemolytic uraemic syndrome in the British Isles 1985-1988: association with verocytotoxin producing Escherichia coli. Part 1: Clinical and epide- miological aspects. Archives of diseases of child, 65: 716-721 (1990). 52. Gordon, V.M. et al. An enzymatic mutant of Shiga-like toxin II variant is a vaccine candidate for edema disease of swine. Infection and immunity, 60: 485-490 (1992). 53. Ryd, M. et al. Induction of a humoral immune response to a Shiga toxin B subunit epitope expressed as a chimeric LamB protein in a Shigellaflexnerilive vaccine strain. Microbial pathogenesis, 12: 399-407 ( 1992). 54. Butterton, J.R. et al. Heterologous antigen expression in Vibrio chol.eraevector strains. Infection and immunity, 63: 2689- 2696 (1995). 55. Acheson, D.W.K. et al. Protective immunity to Shiga-like toxin I following oral immunization with Shiga-like toxin I B- subunit-producing Vibrio chol.eraeC\!D 103-HgR. Infection and immunity, 64: 355-357 (1996). 56. Enterohaemorrhagic Escherichia coli infection. Weekly epidemiol.ogical record, 71: 267-268 (1996). Infection a Escherichia coli enterohemorragique. Rei.eve epidemiologique hebdomadaire, 71: 267-268 (1996). 57. Outbreak of Escherichia coli 0157 infection. Weekly epidemiol.ogical record, 72: 14-15 (1997). Flambee d'infection a Escherichia coli 0157. Rei.eve epidemiol.ogique hebdomadaire, 72: 14-15 (1997). Rapp. trimest. statist. sanit. mond., SO (1997) Foodbornesa/monellosis Thomas M. Gomeza, Yasmine Motarjemib, Shoji Miyagawac, Fritz K. Kafersteind & Klaus Stohre Introduction Foodborne diseases caused by non-typhoid Salmo- neUa represent a very important public health problem in many parts of the world. Data from foodborne disease surveillance programmes in de- veloped countries indicate that Salmonella is the leading cause of food borne infections ( 1 ). This situation has led to substantial costs in public health terms and in serious economic losses for the food industry (2). Although salmonellae are ubiquitous, the pri- mary reservoir of SalmoneUa is the intestinal tract of infected or colonized domestic and wild animals and humans. The primary source of Salmonella in- fection in humans worldwide is the ingestion of contaminated food and water (3, 4). Foods of ani- mal origin such as poultry, eggs, milk, beef, and pork are the main sources. In addition, fresh fruits and vegetables have been implicated as vehicles in Salmonella transmission (5, 6, 7, 8). Contamination of these foods can occur during production, pro- cessing, distribution, retail marketing, and hand- ling/preparation. Other sources of human infec- tion with Salmonella include person-to-person transmission and contact with animals and pets. This article reports global incidence rates of salmonellosis, its economic consequences, and dis- cusses foodborne prevention measures. Situation analysis The World Health Organization (WHO) maintains a Global Databank on Foodborne Diseases, which contains national clinical and laboratory-con- firmed data on salmonellosis. These data are col- lected through national epidemiological bulletins and from reports received directly from Member countries. For the purpose of this article, countries are classified in the text as "developed market- economy countries", "developing countries", and • United States Department of Agriculture, Animal and Plant Health Inspection Service, Veterinary Services, Atlanta, GA. b Scientist, Food Safety and Food Aid Programme, World Health Organization, Geneva. c Formerly, Scientist, Food Safety and Food Aid Programme, World Health Organization, Geneva. d Director, Food Safety and Food Aid Programme, World Health Organization, Geneva. e Scientist, Division of Emerging and other Communicable Diseases Surveillance and Control, World Health Organi- zation, Geneva. Wld hlth statist. quart., 50 (1997) "economics in transition" (United Nations World Economic and Social Survey 1994, New York, 1994) and by WHO Regions in the Figures. We reviewed and analysed these reports, data from published literature, and the WHO Surveil- lance Programme in Europe (9) for the years 1985- 1995. For some countries data were lacking; coun- tries were excluded from analysis if they had fewer than 5 annual reports during the 11-year period. Annual incidence rates for salmonellosis were either reported directly by countries or were calcu- lated by dividing the country's total yearly reported cases of illness due to Salmonella by the mid-year population. Data on salmonellosis were available from 51 countries for the period 1985-1995. Twelve (23%) of these had fewer than 5 annual reports and were excluded from analysis. Of 39 countries, 20 (51 %) were classified as "developed market- economy countries", 9 (23%) as "developing coun- tries", and 10 (26%) as "economics in transition". During the last decade reported incidence rates of salmonellosis in humans increased in many parts of the world (Figs. 1-4). In 1994 (the last year for which at least 25 % of countries reported), particu- larly high rates were reported in the Czech Repub- lic, Hungary, and Germany (Fig. 5). A comparative analysis of salmonellosis surveil- lance data between countries is difficult for many reasons. Data on salmonellosis are available only from a limited number of countries, mainly from the developed world. The contributions from countries vary considerably, ranging from statisti- cal data on numbers of cases only ( clinical and laboratory-confirmed isolates) to full epidemiolog- ical data on foodborne disease outbreaks. Differ- ences also exist in the systems of statutory notifica- tions and surveillance, and in levels of underre- porting within a country. Trend analyses are not possible for individual countries, since these data are surveillance and non-linear (see 10). The above shortcoming notwithstanding, these data provide evidence of a steady and significant increase in the incidence of salmonellosis during the past two to three decades, particularly since the mid 1980s in most countries. The major recent change in the epidemiology of Salmonella concerns the emergence and increase of Salmonella enteritidis (SE) in industrialized countries ( 11) and Salmonella typhimurium definitive type 104 (DT104) in the United Kingdom and the United States. Increases 81 Fig. 1 Salmonellosis incidence rates, WHO Region of the Americas, 1985-1995 Taux d'incidence de la salmonellose, Region OMS des Ameriques, 1985-1995 (/) I~ c::: ·"::::: .9~ "' .c: sg g-o a.o oO o~ 0 ~ 100 80 g g_ -:= -~ 60 "' ·-c.= (/)Q "'c:: (/) "'(/) u"' "CJ u "' Q) t::""CJ O Q) 2-.0 a: E 0 2 Fig. 2 40 20 1985 1986 1987 1988 1989 1990 Year-Annee 1991 Salmonellosis incidence rates, WHO Eastern Mediterranean Region, 1985-1995 l;J Canada D Mexico - Mexique • Panama [] United States - Etats-Unis 1992 1993 1994 1995 Taux d'incidence de la salmonellose, Region OMS de la Mediterranee orientale, 1985-1995 1985 1986 1987 82 1988 1989 1990 Year-Annee 1991 Bahrain - Bahre"in D Cyprus - Chypre • Kuwait - Kowe"it [] Sudan - Soudan 1992 1993 1994 1995 Rapp. trimest. statist. sanit. mond., 50 (1997) Fig.3a Salmonellosis incidence rates, WHO European Region, 1985-1995 Taux d'incidence de la salmonellose, Region europeenne de l'OMS, 1985-1995 !!l I c: "' c: .'t:'. o-"' ·-"' ;;; .c :ig §-o O.c:, c:,O o~ o~ o::i O g_ ::.~ "'·-C.!!:: "'o ~ c: rl ~ "C (.) "'"' c-c O Q) 0. ~ "'.0 a: E 0 z Fig.3b "' I~ c: .-t;:: o-"' -~~ :ig §-o o.o c:,O o~ 0 ~ 0 ::I O g_ ::.~ "'·-0..':=. "'o ~ c: "'"' (.)"' "C (.) "'"' c-c 0"' 0. ~ "'.0 a: E 0 z 500 450 400 350 300 250 200 150 100 50 0 250 200 150 100 50 0 1985 1986 1987 1988 1989 1990 D Austria - Autriche 121 Czech Republic - Republique tcheque D Belgium - Belgique fZl Denmark - Danemark } fa f i [ % { 1985 1986 1987 1988 1989 1990 D Federal Republic of Germany (before reunification on 3 October 1990) - Republique federale d'Allemagne (avant la reunification le 3 octobre 1990) 1991 1991 0 (Former) German Democratic Republic - (Ex) Republique democratique allemande 121 Federal Republic of Germany (after reunification on 3 October 1990) - Republique federale d'Allemagne (apres la reunification le 3 octobre 1990) fZ] Greece - Grece Wld hlth statist. quart., 50 (1997) 1992 1993 1994 1995 Ifill Estonia - Estonie • France • Finland - Finlande 1992 1993 1994 1995 D Hungary - Hongrie • Iceland - lslande • Israel - Israel 83 Fig.3c .l!l I~ c:~ 0 .0 ·-"' «i .c: -o :::,0 g.o o.o 00 o~ o~ 0:::, ~ g_ 4)~ c.~ .,,c5 ~c "'"' u "Cu Cl) Cl) t: "C O Cl) 0. ~ Cl) .0 er. E 0 z 84 100 50 . j 0-1LL...ll..lad:.mLLJ....t'...!:;LJILJ...LJ..ol:.l!llll....l...1....1':1.:L.111....l...LL.l:..t:lm..L..LL.l::.L':l-...t..l...1.J.:..U-...t..L1.L:.U-...J..L1..J.:EJ..._~~·:L'.._~~___J 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 O Italy- ltalie [2l Lithuania - Lituanie El Norway - Norvege • Poland - Pologne • Portugal D Latvia - Lettonie IZI Luxembourg 1985 1986 1987 1988 1989 [] Romania - Roumanie [2l Slovakia - Slovaquie O Russian Fed. - Fed. de Aussie IZI Sweden - Suede 1990 1991 1992 El Switzerland - Suisse • United Kingdom: England - Royaume-Uni: Angleterre 1993 1994 1995 • United Kingdom: Scotland - Royaume-Uni: Ecosse Rapp. trimest. statist. sanit. mond., 50 (1997) Fig. 4 Salmonellosis incidence rates, WHO South-East Asia and Western Pacific Regions, 1985-1995 Taux d'incidence de la salmonellose, Regions OMS de l'Asie du Sud-Est et du Pacifique occidental, 1985-1995 "' 35 "EI "' c: :5 .S! <U 30 -:;; .c: :lg g-o 25 o.o 0 0 0 ~ 0 ~ 0 :::, 20 0 0 0. ~ -.~ Ql.- 15 C.!: "' 0 Q) c: "' "' "' 10 u "' -c u Q) Q) t: -c 0 Q) 5 o.~ Q) .c, cr:E 0 2 0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 O Bangladesh [] Australia - Australie • Japan - Japan O Republic of Korea - Republique de Goree Fig. 5 Salmonellosis incidence rates, selected countries, 1994 Taux d'incidence de la salmonellose pour certains pays, 1994 Czech Republic - Republique !cheque Hungary- Hongrie Germany - Allemagne Poland - Pologne Denmark - Danemark Switzerland - Suisse Norway - Norvege Australia - Australie Kuwait - Kowe"it United States of America - Etats-Unis d'Amerique Japan - Japan Luxembourg Romania - Roumanie Bangladesh Portugal 0 50 100 150 200 250 300 350 400 450 'O ~ ..... "' 0 i 500 Reported cases per 100 OOO population - Nombre de cas notifies pour 100 OOO habitants Wld hlth statist. quart., 50 (1997) 85 in salmonellosis during the last two decades have largely been due to SE, which has caused signifi- cant increases in foodborne illness in a number of countries, such as Argentina, Austria, Bulgaria, Fin- land, France, Hungary, Italy, Spain, Sweden, Swit- zerland, the United Kingdom, and the United States (9, 12). Numerous factors may have contrib- uted to the emergence of these pathogens ( 13 ). These include microbial adaptation and dissemi- nation of the pathogen through increasing global distribution of food and international travel, in- creasing susceptible populations, as well as lifestyle changes leading to an increase in food service es- tablishments and food outlets combined with inad- equate knowledge of foodhandling. In England and Wales, Salmonella isolations re- ported from human infections doubled from 10 251 in 1981 to 22 627 in 1991 (9). This increase was primarily because of an increase in SE, particu- larly phage type 4, and most recently, Salmonella typhimurium DT104. Salmonella typhimurium DT104 is now the second most prevalent Salmonella sero- type, after SE phage type 4, in humans in England and Wales with isolations increasing from 259 in 1990 to 2 873 in 1994 and 3 837 in 1995 (14). Of particular importance in this increase has been the epidemic spread of a strain of Salmonella typhimuri- um DT104 with an antimicrobial multiresistance pattern characterized by resistance to ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline (R-type ACSSuT) ( 14). The emer- gence of these multidrug resistant strains reduce the therapeutic options in cases of invasive infec- tions (15,16). A national case-control study showed infection with Salmonella typhimurium DT104 R-type ACSSuT was associated with higher hospitalization and fatality rates than other Salmonella serotypes ( 17). In addition, this study identified contact with ill farm animals and consumption of chicken, pork sausages, and meat paste as risk factors for infec- tion. This strain has also been isolated from farm workers, a wide range of mammals and poultry and from domestic pets ( 14 ). In the United States, a study of Salmonella typhi- murium isolates from the Pacific Northwest showed that 4% of human isolates obtained in 1989 had the R-type ACSSuT compared to 43% in 1994 ( 18). Among cattle isolates obtained before 1986 none had this R-type, compared to 13% of isolates obtained between 1986 and 1991, and 64% of iso- lates obtained between 1992 and 1995. Of 25 iso- lates from either human or cattle sources with the R-type ACSSuT that were phage typed, all were phage type 104 (18). The Centers for Disease Con- trol and Prevention (CDC) have also confirmed the presence of Salmonella typhimurium DT104 R- type ACSSuT in human isolates submitted as part of national studies (CDC, unpublished data). In the United States, approximately 25 OOO Sal- monella infections were reported annually in the 86 1970s to the CDC's National laboratory-based Sal- monella Surveillance System. There has been a con- tinual increase, to more than 50 OOO reported in- fections in 1985. In 1995, 41 222 infections were reported to the CDC ( 19). The proportion of total Salmonella isolates that were Salmonella enteritidis (SE) increased from 5% in 1976 to 26% in 1994 and decreased to 25% in 1995. Salmonella enteritidis outstripped Salmonella typhimurium as the most commonly reported Salmonella serotype in the United States in 1990, 1994, and 1995. Salmonella enteritidis phage type 4, the predominant phage type in other parts of the world, has recently emerged in the United States (20,21). A similar trend in Salmonella serotypes is reported from European countries where SE, especially phage type 4, is currently the most frequently isolated serotype having a share greater than 50% of all isolates. Salmonella typhimurium, which caused most of the human cases in the early 1980s, now ac- counts for less than 20% of total Salmonella isolates (9). However, as discussed, the emergence of multi- drug resistant S. typhimurium in other European countries may change the relative incidences of these serotypes. Although not clearly defined, the potential threat of SE phage type 4 to both human health and the poultry industry may be greater than that of other phage types. In most countries, both outbreak and sporadic cases of human SE infection frequently have been associated with con- sumption of raw or undercooked shell eggs and/ or poultry meat ( 11, 12, 22-27). The continuing increase in the incidence of cases of human salmonellosis imposes consider- able burdens on a country public health and the economy. The health consequences and spectrum of illness caused by Salmonella is dependent upon a number of factors, including the susceptibility of the person infected, and ranges from a mild gastro- enteritis, which may not warrant seeking medical advice, to more severe debilitating illnesses which may require hospitalization. Certain populations are more susceptible to Salmonella infections. In the industrialized world, the distribution of re- ported cases by age is bimodal with a large propor- tion of cases either less than 4 years or more than 60 years of age (28). Immune deficiencies un- related to age also influence susceptibility to infec- tion and the clinical course of disease. For exam- ple, in states that had high rates of HIV infection, the proportion of Salmonella isolates reported from blood of male salmonellosis cases increased from 2.8% in 1978-1982 to 14.2% in 1983-1987, with substantial increases from serotypes Enteritidis and Typhimurium (29). In addition to the acute health effects, which last for about one week, salmonello- sis may also lead to serious chronic health effects. Infections due to S. enteritidis and S. typhimurium are known to lead to complications such as reactive arthritis or Reiter's syndrome (30). Rapp. trimest. statist. sanit. mond., 50 (1997) Economic consequences The economic impact of salmonellosis falls upon the public sector, the food industry (in particular upon the retail and wholesale food industry), and most importantly, upon infected persons and their families. The national cost of salmonellosis in England and Wales in 1992 has been estimated at between £350 and £502 million with an average cost per case between £789 and £861 (31). Over 73% of costs were direct costs associated with treat- ment and investigation of cases, and costs to the economy from sickness related to absence from work. Social costs of production loss were the larg- est component of the total. The total annual cost of salmonellosis in the United States was calculated at US$3.99 billion with an average cost per case of $1,350 (32). Adjusting for foodborne causes (87- 96%), an estimated 696 OOO to 3 840 OOO salmonel- losis cases stem from food sources each year in the United States with an associated total cost esti- mated to range from $0.6 billion to $3.5 billion annually (33). Prevention A large proportion of Salmonel/,a infections is po- tentially preventable, hence the burden of disease and its repercussions are largely avoidable. There are many approaches to controlling Salmonel/,a in foods, the basis being that of a preventive nature. In order to reduce the incidence of human salmo- nellosis, measures should be taken simultaneously on several levels to prevent the introduction and multiplication of Salmonella in food. An integrated approach should involve preventive measures in the areas of animal production, slaughter /process- ing, and food handling in the home and food service establishments. Risk and cost-benefit analy- ses will provide the rationale for the choice of prevention strategies at the most effective point(s) of intervention. At the level of animal production, measures taken to reduce the prevalence of Salmonella and other microbial pathogens should be based on good production practices (GPP) which include biosecurity, cleaning and disinfection, and animal hygiene. These GPP should be incorporated in industry efforts such as commodity food safety I quality assurance programmes. Integral to the suc- cess of further interventions at the animal produc- tion level are an in-depth understanding of the epidemiology, ecology, and microbial characteris- tics of specific pathogens, animal identification, and diagnostics. Further, intervention strategies modelled in demonstration (pilot) programmes can be used not only to test basic research under practical yet controlled situations, but also to en- able producers to understand the benefits of such interventions in a production setting. Food industries, including animal slaughter and processing, should be encouraged to develop Wld hlth statist. quart., 50 (1997) Hazard Analysis and Critical Control Point (HACCP) programmes. HACCP, which has been endorsed and adopted by international bodies and governments as an effective, scientific, risk-based system for protecting the public from foodborne illness, emphasizes preventing contamination by focused intervention targeted at specific points in food processing. These interventions are moni- tored and recorded to document whether a critical control point is within the established critical limit. Generic HACCP plans can be developed by indus- trial trade groups with approval and verification/ oversight by a central authority; individual proces- sors can then modify these generic plans to meet their specific requirements. At the consumption end of the continuum, the public, especially food handlers, needs education in safe food handling. Vulnerable groups such as the elderly, the immunocompromized (including the sick, infants, and pregnant women) should be made aware of their increased susceptibility to foodborne disease. Furthermore, important rec- ommendations about several measures that should reduce the general risk of infection from foods, in particular foods of animal origin, should be publi- cized. These include guidelines regarding the stor- age, use, handling, and preparation of foods of animal origin and use of pasteurized milk and, where available, pasteurized eggs, in certain high- risk settings. For example, it is recommended that hospitals, nursing homes and commercial kitchens should use pasteurized egg products (where avail- able) for all recipes requiring pooled eggs or raw or undercooked eggs. Finally, considering that it is not possible to produce raw foods of animal origin free from pathogens, there is an additional need to apply control measures, i.e., technologies such as irradiation (radiation pasteurization), to reduce/ eliminate Salmonella from raw materials. An educa- tional process needs to be instituted so that the health professionals and consumers are made aware that the irradiation (radiation pasteuriza- tion) levels that are needed to reduce/ eliminate Salmonel/,a are safe and do not produce a food that is unwholesome and, when used in conjunction with proper food processing and preparation tech- niques, greatly decreases the probability that food- borne pathogens associated with these foods will reach consumers. Globally, the challenge to control of salmonel- losis needs to be supported by an effective surveil- lance programme. Salmonella surveillance data, in reality, underestimate the actual number of per- sons affected. It has been estimated that from 10 to as many as 100 or more cases go unreported for each case reported (34, 35). With few exceptions, in developing countries, where a lack of resources prevent effective foodborne disease surveillance, pertinent data on the prevalence and incidence of foodborne diseases are not available. The impor- 87 tance of laboratory-based surveillance (and further characterization of these isolates by serotype and other subtyping methods) of these infections in the development of a sound control programme cannot be overemphasized. Effective surveillance allows the recognition and investigation of out- breaks and emerging pathogens/strains and the ability to assess the need for and evaluate interven- tions by monitoring longer term trends. Acknowledgement The authors wish to acknowledge the contribution of Dr K. Schmidt, coordinator of the WHO Surveil- lance Programme for Control ofFoodborne Infec- tions and Intoxication, WHO Regional Office for Europe, in the preparation of this article. Dr Schmidt works at the Federal Institute for Health Protection of Consumers and Veterinary Medicine, Berlin, Germany. Summary Foodborne diseases caused by non-typhoid Salmonel- la are a very important public health problem and an economic burden in many parts of the world. Salmonel- losis data from the WHO Global Data bank on Food borne Disease, from the literature and from the WHO Surveil- lance Programme in Europe were reviewed for the years 1985-1995, showing an apparent increase in the inci- dence of salmonellosis in many parts of the world. In industrialized countries, this increase may be due to the emergence and increase of S. enteritidis and S. typhi- murium DT104. In order to reduce the incidence of human foodborne salmonellosis, measures should be taken simultaneous- ly during the production, processing, distribution. retail marketing and handling/preparation of food to prevent the introduction of Salmonella and its multiplication. These control measures need to be supported by effec- tive food borne disease surveillance programmes which make it possible to recognize and investigate outbreaks and emerging pathogens, and to assess the need for and evaluate interventions by monitoring longer term trends. Resume Les salmanel/ases d'arigine alimentaire Les maladies d'origine alimentaire dues aux Salmonella non typho'idiques sont un probleme de sante publique tres important et representent un fardeau economique dans de nombreuses regions du monde. Les donnees sur les salmonelloses provenant de la base de donnees mondiale OMS sur les maladies d'origine alimentaire, de la documentation existante et du programme de surveillance OMS en Europe ont ete passees en revue pour les annees 1985-1995, ce qui a permis de mettre en evidence une augmentation apparente de !'inci- dence des salmonelloses dans de nombreuses parties 88 du monde. Dans les pays industrialises, cette augmen- tation pourrait etre due a !'emergence ou a la recrudes- cence de S. enteritidis et de S. typhimurium DT104. Afin de reduire !'incidence des salmonelloses humaines transmises par les aliments, des mesures doivent etre prises simultanement au cours de la production, de la transformation, de la distribution, de la commercialisa- tion au detail et de la manipulation/preparation des aliments pour eviter !'introduction de Salmonella dans les aliments et leur multiplication. Ces mesures de lutte doivent etre etayees par des programmes efficaces de surveillance des maladies d'origine alimentaire qui per- mettent de reconnaitre et d'etudier les flambees epide- miques et les agents pathogenes emergents, et d'ap- precier la necessite d'interventions et de les evaluer en surveillant les tendances a long terme. References - References 1. CAST R£port. Foodhorne pathogens: risks and consequences. Task force report No. 122, Washington, DC, Council for Agricultural Science and Technology, Sept. 1994. 2. Sockett, P.N. The economic implications of human Salmonella infection. Journal of applied bacteriowgy, 71: 289-295 (1991). 3. Tauxe, R. T. An update on Salmonella. Health and environment digest, 10(1): 1-4 (1996). 4. Gangarosa, E.J. Commentary: The challenge of global control. Healthandenuironmentdigest, 10(1): 4-6 (1996). 5. Hedberg, C.W. et al. Changing epidemiology of food-borne disease: a Minnesota perspective. Clinical infectious diseases, 18: 671-82 (1994). 6. D'Aoust, J. Salmonella and the international food trade. International journal of food microbiowgy, 24: 11-31 ( 1994). 7. Beuchat, L.W. Pathogenic microorganisms associated with fresh produce. Journal of food protection, 59(2): 204-216 (1995). 8. CDC Surveillance Summaries, October 25, 1996. Morbidity and mortality weekly report, 45 ( SS-5) ( 1996) . 9. \\'HO surveillance programme for control of foodborne infections and intoxications in Europe: sixth report 1990- 1992, Berlin, Federal Institute for Health Protection of Consumers and Veterinary Medicine, 1995. JO. Motarjemi, Y. &: Kaferstein, F.K. Global estimation of foodborne diseases. World health statistics quarterly, 50(1/2): 5-11 (1997). Motarjemi, Y. &:Kaferstein, F.K. Lessalmonelloses d' origine alimentaire [resume]. Rapport trimestriel de statistiques sanitaires mondia/,es, 50(1/2): 11 (1997). 11. Rodrigue, D.C. et al. International increase in Salmonella enteritidis: A new pandemic? Epidemiology and infection, 105: 21-27 ( 1990) 12. Rampling, A. Salmonella enteritidis five years on. Lancet, 342: 317-318 (1993). 13. Altekruse, S.F. &: Swerdlow, D.L. The changing epide- miology of foodborne diseases. American journal of medical science, 311(1): 23-29 (1996). 14. Threlfall, EJ. et al. Increasing spectrum of resistance in multiresistant Salmonella typhimurium. Lancet, 347: !052-53 (1996). 15. Holmberg, S.D. et al. Health and economic impacts of antimicrobial resistance. Review of infectious diseases, 9(6): !065-!078 (1987). 16. Cohen, M.L. Antimicrobial resistance: prognosis for public health. Trends in microbiowgy, 2 (I): 422-425 ( 1994). 17. Wall, P.G. et al. A case control study of infection with an epidemic strain of multiresistant Salmonella typhimurium DTI04 in England and Wales. CDR review, 4: Rl26- Rl31(1994). Rapp. trimest. statist. sanit. mond., 50 (1997) 18. Besser, T.E. et al. Emergence of Salmonella Typhimurium DT 104 in humans and animals in the Pacific Northwest. International Conference on Diseases in Nature Communicable to Man. August, 1996. Seattle, Washington. 19. Centers for Disease Control and Prevention. CDC Laboratory Confirmed Salmonella Surveillance, Annual summary, 1993-1995. 20. Boyce, T.G. et al. Recurrent outbreaks of Salmonella enteritidis infections in a Texas restaurant: phage type 4 arrives in the United States. Epidemiol.ogy and infection, 117: 29-34 (1996). 21. Passaro, DJ. et al. Epidemic Salmonella enteritidis infection in Los Angeles County, California - The predominance of phage type 4. Western journal of medicine, 165: 126-130 (1996). 22. Mishu, B. et al. Outbreaks of Salmonella enteritidis infections in the United States, 1985-1991.Journal of infectious diseases, 169: 547-52 (1994). 23. Scuderi, G. et al. Foodborne outbreaks caused by salmonella in Italy, 1991-4. Epidemiology and infection, 116: 257-265 (1996). 24. Hedberg, C.W. et al. Role of egg consumption in sporadic Salmonella enteritidisand Salmonella typhimuriwn infections in Minnesota. Journal of infectious diseases, 167: 107-11 ( 1993). 25. Salmonella in eggs. PHL5 evidence to Agriculture Committee. PHLS microbiology digest, 6: 1-9 (1989). 26. Rampling, A. et al. Salmonella enteritidisphage type 4 infection of broiler chickens: a hazard to public health. Lancet, 436-438 (1989). Wld hlth statist. quart., 50 (1997) 27. Humphrey, T.J. et al. Poultry meat as a source of human salmonellosis in England and Wales. Epidemiol.ogy and infection, 100: 175-84 (1988). 28. Bean, N. et al. Salmonella isolates from humans in the United States, 1984-86. MarlJidity and mortality weekly repart, 37, (SS-2): 25-31 (1988). 29. Levine, W.C. et al. Epidemiology of nontyphoidal Salmonella bacteremia during the human immunodeficiency virus epidemic.Journal of infectious diseases, 164: 81-7 (1991). 30. Smith,J.L Arthritis and foodbornc bacteria.Journal of food protection, 57(10): 935-941 (1994). 31. Roberts,J.A. & Socket, P.N. The social-economic impact of human Salmonella enteritidis infection. International journal of food microbiol.ogy, 21(1/2): 117-129 (1994). 32. Todd, E.C.D. Preliminary estimates of costs of foodborne disease in the United States.Journal of food protection, 52(8): 595-601 (1989). 33. Buzby, J.C. et al. Bacterial Foodborne Disease: Medical costs and productivity losses. Food and Consumer Eco- nomics Division, Economic Research Service, U.S. Depart- ment of Agriculture. Agricultural economic repart, No. 741 (1996). 34. Chalker, R.B. & Blaser, M.J. A review of salmonellosis: III. Magnitude of Salmonella infections in the United States. Review of infectious diseases, 10 (1): 111-124 ( 1988). 35. Cohen, M.L & Tauxe, R.V. Drug-resistant Salmonella in the United States: An cpidemiologic perspective. Science, 234: 964-969 ( 1986). 89 Virus transmission via food Dean 0. Clivera Introduction Hepatitis A virus was the fourth leading cause (fol- lowing Salmonella spp., ShigeUa spp., and Clostridium perfringens) of foodborne disease among outbreak- associated illnesses compiled by the United States Centers for Disease Control and Prevention (CDC) for the period 1988-1992, the most recent period for which data have been issued (]). Norwalk-like viruses ranked ninth. By contrast, Norwalk virus was fifth, hepatitis A virus sixth, and other viruses (principally rotaviruses) were tenth among the top 10 determined causes of foodborne disease in the United States during the period 1983-1987 (2). The numbers of reported illnesses are small because the CDC has a passive data collection system and only records illnesses occurring as outbreaks, rather than sporadically. Hepatitis A, which is noto- riously underreported in this country (3), is the only reportable foodborne virus disease (i.e., offi- cial reporting is mandatory for all diagnosed cases), so it seems certain that incidence of the others is even less accurately recorded. Some other new agents will also appear among the top 10 caus- es in future compilations, but it is clear that viruses have now "emerged" as causes of foodborne dis- ease. A few European countries also report food- borne viral disease, but differences in emphasis make international comparisons difficult (4). It would appear that some countries do not.consider viruses among potential causes of observed food- a Professor, Department of Population Health and Reproduction, School of Veterinary Medicine, University of California, Davis, CA, United States of America. E-mail: docliver@ucdavis.edu Table 1 Major groups of nonenveloped human enteric viruses Tableau 1 borne disease, perhaps because of diagnostic diffi- culties and the unlikelihood of detecting virus in an implicated food. Viruses transmitted via foods The hepatitis A and small gastroenteritis viruses are more often transmitted via foods than are other viruses. All known foodborne viruses except the agent of tick-borne encephalitis are human-spe- cific and transmitted by a faecal-oral cycle. Perhaps incidentally, all of these "enteric" viruses contain RNA rather than DNA (parvovirus, if it is really foodborne, is an exception) and have no lipid en- velope around their protein coats (Tabl.e 1). Ve- hicles implicated have most often been mussels from contaminated waters, but direct contamina- tion of other foods through handling by infected persons occurs frequently. Hepatitis A virus When hepatitis transmission via food was first re- corded, it was not known that multiple types of viral hepatitis existed (5). Now, at least five sero- logical types of hepatitis viruses are recognized, belonging to various taxonomic groups; but only hepatitis A is documented as being foodborne (6) (Tabl.e 2). Hepatitis E virus (a calicivirus) is appar- ently not present in North America; it has been implicated in water-associated, but as yet not food- associated outbreaks (6). Hepatitis A is one of the more severe of foodborne diseases, especially among those caused by viruses (Tabl.e 3). Both hepatitis A and E viruses are listed as Severe Haz- ards in Appendix V of the 1995 United States Food and Drug Administration's Food Code (7). Virus Principaux groupes de virus intestinaux humains non enveloppes Diameter (nucleic acid strands) - Diametre (mono/bicatenaire) 25-35 nm (single) - (monocatenaire) 70·85 nm (double) - (bicatenaire) 90 Nucleic acid type - Type d'acide nucleique RNA-ARN astroviruses - astrovirus caliciviruses - calicivirus picornaviruses - picornavirus reoviruses - reovirus rotaviruses - rotavirus DNA-AON parvoviruses - parvovirus adenoviruses - adenovirus Rapp. trimest. statist. sanit. mond., 50 (1997) Table 2 Some viral causes of hepatitis in humansa Tableau 2 Quelques causes virales de l'hepatite chez l'hommea Type A B c D E Former name - Ancien nom infectious hepatitis - hepatite infectieuse serum hepatitis - hepatite serique non-A, non-B delta agent - agent delta non-A, non-B Mode of transmission - Voie de transmission faecal-oral - feco-orale parenteral - parenterale parenteral - parenterale parenteral - parenterale faecal-oral - feco-orale a Adapted from - Adapte de: Cliver, D.O. Foodborne viruses. In: M.P. Doyle et al. (eds), Fundamentals offood microbiology. Washington, DC, American Society for Microbiology (in press) - (sous presse). Table 3 Hepatitis A, virus and disease Tableau 3 Hepatite A: le virus et la maladie Plcomavlrus Particles featureless spheres - 28 nm in diameter, single-stranded RNA coated with protein Infection Infection via intestine to liver, incubation period 15-50 days (mean 28 days) Illness Illness from immune destruction of infected liver cells: fever, malaise, anorexia, nausea, abdominal discomfort, often followed by jaundice; severity tends to increase with age - ranges from inapparent infection to weeks of debility, occasionally with permanent sequelae Shedding Shedding of virus peaks during the second half of the incubation period (10-14 days), usually ends by 7 days after onset of jaundice Diagnosis Diagnosis is based on detection of lgM-class antibody against hepatitis A virus in the patient's blood serum (kits available) Immunity Immunity is durable (possibly lifelong) after infection Source: Ref. - Ref. (6) Wld hlth statist. quart., SO (1997) Particules spheriques sans caracteristiques particulieres -28 nm de diametre, a ARN monocatenaire et revetement proteique L'infection L'infection progresse de l'intestin vers le foie, periode d'incubation 15-50 jours (moyenne 28 jours) La maladie La maladie est due a la destruction, par des mecanismes immunologiques, des cellules hepatiques infectees: fievre, malaise general, anorexie, nausees, gene abdominale, souvent suivis par un ictere; la gravite tend a augmenter avec l'age, la maladie allant d'une infection inapparente a une maladie laissant le sujet affaibli pendant plusieurs semaines avec parfois des sequelles irreversibles L'excr6tion L'excretion du virus culmine pendant la deuxieme moitie de la periode d'incubation (10-14 jours), et cesse habituellement dans les 7 jours suivant !'apparition de l'ictere Le diagnostic Le diagnostic est fonde sur la detection des anticorps de classe lgM diriges contre le virus de l'hepatite A dans le serum du malade (ii existe des trousses de depistage) L'immunite L'immunite est durable (peut-etre a vie) apres !'infection. 91 particles produced in the liver are shed via the common bile duct and occur at levels above I Q6 / g of faeces for days or weeks before the onset of illness. Food becomes contaminated via faecally soiled hands of infected persons or via faecally contaminated water, as is usual with shellfish. The virus is more heat resistant than most enteric virus- es and is also quite resistant to drying. Now that a formalin-killed vaccine has been licensed for use in the United States (it had been licensed earlier in Europe), food workers could be immunized to pre- clude the possibility that they might contaminate food with the virus (8). The vaccine is produced from a mutant strain of the virus that replicates in cell culture: wild type virus either does not repli- cate in cell culture or replicates very slowly, often without cytopathic effects (6). Small round structured viruses causing gastroenteritis The Norwalk virus was the first gastroenteritis virus reported to be foodborne (9). Later, other serolog- ically and genetically related viruses belonging to the calicivirus group were recognized (Tabk 4 ). Table 4 Norwalk-like, small round-structured viruses of gastroenteritis Tableau 4 The term, "small round structured viruses", was applied to the agents when they were detected by electron microscopy or immune electron micro- scopy (10). These viruses have only occasionally been associated with sporadic gastroenteritis in the United Kingdom but are apparently among the leading causes of foodborne illness there, whether from shellfish or from other foods contaminated in handling ( 11 ). Astroviruses, which also have visible surface structure, are variably included in this group; they are discussed separately below. Gastro- enteritis caused by the Norwalk-like viruses often includes vomiting, and virus shed in vomitus may contaminate food. Because immunity is transient, persons who have been infected and ill with the Norwalk virus are subject to reinfection with it after perhaps a year, as well as with other serotypes ( 12). Susceptibility is common, and attack rates in out- breaks are often quite high. Outbreaks have been traced both to ill persons ( 13) and to food workers who had recovered from illness days earlier ( 14 ). The virus evidently is shed in large quantities, and small (not yet measured) quantities are infectious perorally, but the individual viral particles do not Petits virus ronds structures, de type Norwalk, de la gastro-enterite Particles spheres 25-35 nm in diameter, single-stranded RNA coated with protein that has characteristic cupped surface depressions Infection Infection of intestinal lining, incubation 24-48 hours Illness Nausea, vomiting, diarrhoea, etc., lasting 24-48 hours Shedding During illness (in vomitus and faeces), possibly 7 days after onset Diagnosis Detection of virus in stool by ELISA or PER or of antibody against the virus in patient's blood serum; no standard methods, reagents not readily available for most agents Immunity Apparently transient Source: Ref. - Ref. (10). 92 Calicivirus Particules spheriques de 25-35 nm de diametre, a ARN monocatenaire et revetement proteique presentant des depressions caracteristiques en forme de cupules L 'infection L'infection touche la muqueuse intestinale, avec une incubation de 24-48 heures La maladie Nausees, vomissements, diarrhees, etc., pendant 24-48 heures L 'excr6tion Pendant la maladie (dans les vomissements et les selles), et peut-etre pendant les 7 jours suivant le debut de la maladie Le diagnostic Detection du virus dans les selles par ELISA au PER au des anticorps diriges centre le virus dans le serum du malade; ii n'existe pas de methodes standard, ni de reactifs facilement disponibles pour la plupart de ces agents L'immunite Apparemment de courte duree Rapp. trimest. statist. sanit. mond., 50 (1997) Table 5 Diseases caused by viruses less commonly transmitted by foods Tableau 5 Maladies provoquees par des virus moins frequemment transmis par des aliments Virus group- Groupe de virus Illness {major signs) - Maladie {signes principaux) Incubation Duration- Remarks-Duree Remarques astroviruses - astrovirus diarrhoea - diarrhees 3-4 days- 2-3 daysa- some replicate in cell culture - 3 4 jours 2-3 joursa certains se repliquent en cultures cellulaires picornavirusesb- picornavirusb meningitis, myalgia, etc. - meningites, myalgies, etc. 3-5 daysc- varies- food vehicle rare - vehicule 3-5 joursc variable alimentaire rare rotaviruses - rotavirus fever, vomiting, diarrhoea, dehydration - fievre, vomissements, diarrhee, deshydratation 1-3 days- 4-6 days- often severe in children - 1-3 jours 4-6 jours souvent grave chez l'enfant tick-borne encephalitis (Flavivirus) - encephalite a tiques (flavivirus) fever, headache, nausea, vomiting, encephalitis or aseptic meningitis - fievre, cephalees, nausees, vomissements, encephalite ou meningite aseptique 7-14 days- varies- virus shed in milk of tick-bitten 7-14 jours variable animals - les virus sont excretes a Sometimes longer. - Quelquefois plus. b Other than hepatitis A virus. - Autres que le virus de l'hepatite A. c Variable. - Variable. appear to be exceptionally resistant to inactivation by heat or chlorine. Members of this group have not been shown to replicate in cell cultures. Other viruses occasionally transmitted via foods Many other groups of human enteric viruses are known, yet these are reported to be transmitted via foods only infrequently or not at all (Tabl.e 5). Fac- tors such as duration and level of faecal shedding of the virus, efficiency of peroral infection, or sta- bility of the virus in the food vehicle may play a role. Given that some of these viruses are able to replicate in laboratory cell cultures and cause cyto- pathic effects, they may be better characterized than the more important foodborne viruses dis- cussed above. Astroviruses comprise a distinct group of small round gastroenteritis viruses that have surface pro- jections in patterns often resembling five- or six- pointed stars ( 10). The non-enveloped protein coat surrounds single-stranded RNA (Tabl.e 1). Usual features of the disease vary slightly from those of the Norwalk-like viruses: the incubation period is somewhat longer, vomiting is less common, and very young (<l year) children are more often af- fected. Epidemiological evidence of transmission via foods is limited. Rotavirusescontain segmented, double-stranded RNA (Tabl.e 1) surrounded by a double protein coat (15). They, too, infect young children most often but are occasionally associated with food and water. Wld hlth statist. quart., 50 (1997) dans le lait d'animaux piques par des tiques Picornaviruses other than hepatitis A virus also have single ( +) stranded RNA (Tabl.e 1) in a simple protein coat that is featureless as seen by electron microscopy. The polioviruses were the first viruses shown to be foodborne, but virulent strains are now rare, and the vaccine strains are among poten- tial indicators of the possible presence of other, virulent viruses in food and water ( 16). Although the coxsackieviruses have sometimes been re- ported in association with food and water, only one outbreak of foodborne coxsackievirus illness has been reported ( 17). Two food borne outbreaks of echovirus illness have been recorded (18, 19). Tick-borne encephalitis virus is the only known foodborne virus that is not transmitted by a faecal- oral route (20). The agent infects dairy animals in central Europe (principally Slovakia) via the bites of vector ticks, Ixodes persul.catus and I. ri.cinus. In- fected animals shed the virus in their milk, which, if ingested without pasteurization, infects humans. Products made from unpasteurized milk may also be vehicles. The virus belongs to the genus Flavi- virus, meaning that the particle contains single ( +) stranded RNA like most foodborne viruses, but has a lipid envelope around the protein coat. This is the only enveloped virus known to be foodborne. It is highly specific to its vectors and has a limited range, so that outbreaks are now rare. Seven people in Slovakia were affected during a recent incident (21). Hepatitis E virus belongs to the calicivirus group, so it has single ( +) stranded RNA coated with pro- 93 tein that shows characteristic cup-like depres- sions (6). It occurs widely in Asia, Africa, and Latin America, but rarely elsewhere and not at all in the United States and Canada (except for rare import- ed cases). It is the non-A, non-B hepatitis virus that was known to be transmitted by the faecal-oral route. Waterborne outbreaks are common; but for some reason, foodborne outbreaks have not yet been documented. Parvoviruses are proposed as causes of human gastroenteritis. Although documented food-associ- ated outbreaks are evidently quite rare, Appleton ( JO) describes an outbreak associated with cockle consumption in England that involved at least 800 people. Special features of viruses among foodborne disease agents The viruses transmitted via foods (e.g., the hepati- tis A virus, Norwalk-like viruses, and rotaviruses mentioned above) share properties that distin- guish them from other foodborne disease agents. Particle as the transmissible form Viruses pass from host to host in the form of inert particles. The particles are roughly spherical, with diameters of 25 to 35 nm (hepatitis A and Norwalk- like, or small round structured, viruses) or as large as 75 nm ( rota viruses; Tab/,e 1). The smaller viruses contain single-stranded RNA, whereas the rota- viruses contain double-stranded RNA. None of the DNA-containing enteric viruses has been shown with certainty to be foodborne. The outer surface of the particle is a highly specific protein coat that protects the RNA, interacts with a susceptible host cell to initiate infection, and acts as the antigen against which the host's immune responses are mounted. Because these particles are totally inert, they cannot multiply in foods or anywhere outside the host. Neither can they carry out any metabolic activity, nor respond to stresses encountered in the environment. Viral infection The virus particle will enter only a suitable host cell. Specificity depends on the interaction of the coat protein with receptors on the host cell. Only certain cells in the bodies of certain species can be infected: essentially all viruses transmitted to hu- mans via foods are specific for humans, and per- haps a few other primates. In practical terms, zoo- notic viruses are not transmitted via foods. When the viral coat protein reacts with homolo- gous receptors on the cell membrane, the host cell engulfs and uncoats the viral RNA (22). The RNA is translated to various virus-specific proteins and replicated (with the help of virus-specific, RNA- dependent RNA polymerase) into additional 94 copies of viral RNA. The replicative cycle takes place in the host cell's cytoplasm, without partici- pation of DNA (reverse transcription of the viral genome does not occur). As coat protein and viral RNA accumulate in the host cell, progeny particles assemble themselves and eventually leave the cell via leakage or in blebs that pinch off from the cell's surface membrane. Viral disease may occur when progeny virus spreads and infects enough host cells to interfere with some normal bodily function. The viral repli- cation may kill or subvert the host cells; but in the case of hepatitis A, the host's infected liver cells apparently are little affected until the body mounts an immune response and destroys the infected cells by means of cytotoxic T-cells. Whether the liver or the lining of the small intestine is the site of viral infection, the hepatitis or gastroenteritis that results is seldom fatal. Epidemiology of viruses Enteric (faecal-oral) transmission Essentially all foodborne viruses are transmitted enterically: they are shed with faeces and infect by being ingested (22). Like many other infectious agents that are enterically transmitted, the majority of infections are probably contracted by person-to- person contact, perhaps via faecally soiled hand to mouth (3). If vomiting is part of the illness, viral particles may be shed with vomitus. Indirect trans- mission of enteric agents may occur via vectors such as flies, fomites such as soiled diapers, but most importantly via the vehicles, food and water. In the United States, more foodborne than water- borne viral illnesses are recorded (23). Foods as vehicles for viruses Although the majority of transmissions of enteric viruses probably occur by person-to-person con- tact, many foodborne outbreaks of viral disease have been recorded. A few enteric viruses of hu- mans have not been reported to be transmitted via foods; whereas others are foodborne with some frequency. The only virus for which transmission via food and water can be compared with total recorded incidence is hepatitis A virus, because the disease it causes is reportable (3). The proportion of cases attributed to food- and waterborne trans- mission, collectively, in the United States is only 3% to 9%, with most cases foodbome (23). Howev- er, it must recognized that the numerator (e.g., foodborne illnesses) and the denominator (total reported illnesses) are compiled in very different ways. That is, the foodborne illnesses are only those occurring in outbreaks that happen to have been investigated (many are not), so illnesses in un- recorded outbreaks and those occurring sporadi- cally, though also foodborne, would not be includ- Rapp. trimest. statist. sanit. mond., 50 (1997) ed. It is also certain that not all diagnosed cases of hepatitis A are reported through official channels, but the proportion missed in this way is probably less than for the foodborne category. In any event, it seems likely that if all of the recorded food borne illnesses had been prevented, no statistically signif- icant change in the total rate of recorded hepa- titis A in the United States would be detected. Until diagnosis and reporting of hepatitis A are pursued more vigorously, transmission via food must be prevented for its own sake, rather than to reduce overall incidence. Data from other countries are compiled differently and are therefore difficult to compare with those from the United States (4) (Tah/,e 6). Germany evidently records all hepatitis A as possibly foodbome, as some countries do for salmonellosis. Categories of vehicles for viral dis- ease have been reported from the United King- dom (4) (Tab/,e 7). Table 6 Viruses among foodborne disease agents Various groups of viruses occupied 3 of the top 10 positions among causes of reported foodborne disease in the United States during 1983-1987 (2). This was a period when the Norwalk virus was re- garded as one agent, rather than a group of vari- ably related viruses. Therefore, one would have expected that the num her of reported food- associated cases would increase as the diagnostic "net" broadened. However, Norwalk-like viruses fell to ninth position {two outbreaks comprising 292 cases) during 1988-1992 ( 1 ), which suggests that the many improved diagnostic methods that have appeared in the literature are being applied very sparingly. Among reported outbreaks of food- borne disease in the United States in 1988-1992, etiologies were determined in 1 001 ( 41 % ) , com- prising 36 890 (48%) of the cases: 4% of these outbreaks and 6% of the cases were attributed to Reported viral and other foodborne disease in five European countries, 1990-1992a (4) Tableau 6 Maladies d'origine alimentaire virales et autres rapportees dans cinq pays d'Europe, 1990-1992a (4) Country - Pays France Germany - Allemagne Spain - Espagne Sweden - Suede UK-England & Wales - Royaume-Uni - Angleterre & Pays de Galles Hepatitis A outbreaks (cases)- Flam bees d'hepatite A (cas) 1 8 (18 348) 3 1(8) 3 a For the UK, 1989·1991. - Pour le Royaume-Uni, 1989-1991. Small round Total foodborne disease structured outbreaks where agent virusesb- identified - Total des Petits virus flambees de maladies ronds d'origine alimentaire dont structuresb l'agent a ete iden:ifie nrc 1 669 nr 319 nr 1 723 nr 2 609 29 496d b Small round structured viruses, also called Norwalk-like. - Petits virus non structures, egalement appeles virus de type Norwalk. c nr = not reported - non rapporte. Total foodborne disease outbreaks - Total des flambees d'origine alimentaire 1 974 386 2 818 3 750 575d d These data are for general outbreaks and do not include family outbreaks-Ces donnees concernent les flambees collectives et ne comprennent pas les flambees familiales. Source: Ref. - Ref. (4). Table 7 Outbreaks of hepatitis A and small round structured virus foodborne disease in the United Kingdom (England and Wales), 1989-1991 Tableau 7 Flambees d'hepatite A et de maladies d'origine alimentaire a petits virus ronds structures au Royaume-Uni (Angleterre et Pays de Galles), 1989-1991 Virus Molluscan Seafood- Salad/buffeVbread/ Meat1barbecue/ Vehicle not shelttish- Poisson et sandwiches - pizza - Viande/ specified- Coquillages autres fruits Salade/buttevpain/ barbecue/pizza Vehicule non de mer sandwiches precise Hepatitis A - Hepatite A nra nr Small round structured viruses - Petits virus non structures 2 2 10 4 11 a nr = not reported. - non rapporte. Source: Ref. - Ref. (4). Wld hlth statist. quart., 50 (1997) 95 viruses. Until methods for detecting viruses in foods are improved and widely applied, knowledge of foodborne virus transmission results from suc- cessful diagnoses of human illnesses. Here, too, it should be noted that much foodborne illness is endured without consulting a physician and that physicians are unlikely to order diagnostic tests for virus disease because, if the viral diagnosis is con- firmed, the physician has no means of treating the illness. The hepatitis A virus now appears to be causing more food borne illnesses than many of the better-known bacterial pathogens. Among the 1 422 recorded foodborne outbreaks (presumably gastroenteritis in most instances) during 1988- 1992 that were of undetermined etiology, it was surmised on the basis of incubation periods longer ~ 15 h that 35% might have been caused by viruses ( 1 ). The etiology of recorded food borne disease is evidently much less likely to remain undetermined in Europe than in the United States (4). Shellfish as special vehicles for viruses Bivalve molluscs, such as clams, cockles, mussels, and oysters, are especially prone to transmit viruses. In England and Wales during 1965 to 1988, 17 outbreaks of hepatitis A, 37 outbreaks of viral gastroenteritis, and 101 outbreaks of unknown eti- ology (with features often characteristic of viral gastroenteritis) were among 169 recorded out- breaks of mollusc-associated illness (24). The wa- ters in which molluscs grow are increasingly subject to human faecal contamination, sometimes from sewage discharges and sometimes from the shell- fish harvesters. The shellfish collect viruses in the course of their filter feeding activity. Human viruses do not infect these species, but they are harboured for long periods in the shellfish diges- tive tract and are apparently more difficult to re- move than bacteria during processes intended to cleanse the shellfish (25). Unlike many other sea- foods, shellfish are usually eaten with their diges- tive tracts in place. Shellfish are often eaten raw or lightly cooked. The shell may protect against heat penetration during cooking, thus making thermal inactivation more difficult than in the case of other foods (26). The first recorded outbreak of shellfish-associ- ated viral disease resulted from storing clean oys- ters in a faecally contaminated harbour while await- ing sale (27). Over 600 cases of hepatitis A resulted. More recently, outbreaks of viral gastroenteritis (28) and of hepatitis A (29) have been associated with eating various shellfish, usually without cook- ing. A clam-associated outbreak of hepatitis A in Shanghai comprising nearly 300 OOO cases may have been the largest recorded outbreak of food- borne disease in history (30). Among workers at a Shanghai factory, 18% of those who had eaten raw clams and 7% of those who had eaten cooked clams got hepatitis A (31). Sporadic hepatitis A 96 associated with shellfish has also been demon- strated (32); it is difficult to avoid bias entirely in such studies because, at least in coastal states, a diagnosis of hepatitis A regularly leads to asking the patient about shellfish consumption, to the exclusion of other foods. Shellfish-growing waters are typically moni- tored for faecal contamination by testing for bacte- ria of the faecal coliform group or for Escherichia coli. The presence of these bacteria has been shown to be a poor predictor of the presence of human enteric viruses (33, 34). Unfortunately, no more accurate index of the presence of viruses in shell- fish or their growing waters has yet been identified. For want of a way to guarantee the safety of raw cockles, the United Kingdom government allows sale only of cockles cooked by an approved method. Outbreaks: special features, costs The span of onsets in an "explosive" outbreak tends to equal the median incubation period of the disease. Therefore, an outbreak of hepatitis A in which many people were infected on the same occasion may show onsets of illness over a 28- or 30- day period (6). Under similar circumstances, out- breaks of Norwalk virus gastroenteritis would prob- ably show onsets of primary illness over no more than 2 days (10). Secondary cases may occur in both instances, with transmission of the infectious agent from those who ate contaminated food, via contact, to others who did not. Norwalk gastro- enteritis is often characterized by shedding of the virus in vomitus as well as in faeces, so opportuni- ties for secondary transmission are relatively great. Because immunity after Norwalk virus infection is not protective, attack rates (number of persons ill+ number of persons exposed) are often quite high ( e.g., >60%). In outbreaks from food contaminated by a sin- gle infected person, the long duration of shedding (10-14 days) may further spread the period of on- sets of hepatitis A in situations where the infected person contaminates food on several days. Shed- ding of Norwalk-like viruses for a week is known, so "diffuse" outbreaks of viral gastroenteritis are also possible. Either virus can persist in contaminated food and thus infect people who eat the food on different days, further "blurring" the outbreak. Given all these complications, it is remarkable that epidemiologists have been able to solve as many of the mysteries of hepatitis A transmission as they have. As yet, sporadic transmission resulting in sin- gle illnesses has defied characterization, except in the study concerning shellfish that was mentioned earlier (32). Hepatitis A is one of the more severe of food- borne diseases: a few weeks of debility are com- mon, and permanent impairment of some liver functions occurs occasionally. In contrast, viral gas- Rapp. trimest. statist. sanit. mond., 50 (1997) troenteritis typically lasts 1 or 2 days. In the absence of directly applicable data, costs per case of food- borne hepatitis A have been estimated at US$ 5000, versus US$ 887 for Norwalk gastroenter- itis (35). Because costs of specific outbreaks have not been determined, these must be estimated by multiplying the recorded case number by the ap- plicable, estimated cost per case. Risk assessment The process of risk assessment has generally com- prised hazard identification, dose-response assess- ment, exposure assessment, and risk characteriza- tion (35). Because foodborne infectious agents present some special features not seen with chemi- cal hazards, the risk assessment process may need some modification in this application (36). Beyond this, viruses transmissible via foods present some problems all their own. Hazard identification, based on the epidemiological record, focuses at- tention on hepatitis A virus and the Norwalk-like agents of gastroenteritis. Dose-response assessment for these viruses is complicated by the lack of labo- ratory hosts that they will infect: when human vol- unteer trials have been conducted, precise quanti- fication of the doses administered was impossible (37, 38). Even when cell-cultured enteric viruses have been administered perorally, the number of viral particles comprising a cell culture infectious dose (e.g., a plaque-forming unit) was not deter- mined. Assumptions based on various models for infectivity yield quite variable predictions. Expo- sure assessment is also difficult for foodborne viruses because there are no standardized methods for qualitative detection of viruses in foods, and even the best methods typically are not quanti- tative. Distribution of the few human viruses present in the food supply can be expected to be heterogeneous and non random. The virus detec- tion methods applicable to foods are unlikely to be used on a routine basis to determine which foods contain viruses because the cost of performing pre- dominantly negative tests would be huge. For these reasons, precise risk characterization is not really an option. All the same, attempts have been made; results do not conform closely to present knowl- edge of the incidence of viral infections in the United States (39). Detecting loodborne viruses Detection of viruses in food is most likely to be undertaken when an outbreak has occurred; how- ever, it would indeed be helpful if some routine testing method were available to apply to foods such as shellfish that often serve as vehicles for viruses. These problems will be addressed briefly here, as they have recently been reviewed fairly extensively elsewhere ( 40, 41 ). A fundamental problem is that the viruses of greatest concern, Wld hlth statist. quart., 50 (1997) hepatitis A viruses and the Norwalk-like gastroen- teritis viruses, replicate slowly and inapparently or not at all in laboratory cell cultures. If foodborne viruses cannot be detected on the basis of their infectivity, alternate bases include their mor- phology (as seen by electron microscopy), their antigenic specificity (as demonstrated by reactions with homologous antibody), their genetic specifici- ty (as demonstrated with complementary probes or polymerase chain reaction (PCR) primers), or combinations of these. These methods may be less sensitive than tests based on infectivity, and by na- ture they all carry some risk of yielding a posi- tive result with virus that has been inactivated (no longer infectious). The final detection method to be used must be considered when the food sample is being processed for testing. An additional prob- lem encountered with the hepatitis A virus is that the incubation period of the resultant illness is so long (average 4 weeks) that pertinent food samples are unlikely to be available once the disease has been recognized. Processing food samples for testing Most foods other than milk, water, and a few others require liquefaction as a first processing step (40, 41). Addition of liquid for this purpose is usu- ally kept to a minimum because the virus present in food will be diluted, to the detriment of the sen- sitivity of the method. Solid food samples are shaken, comminuted, or otherwise dispersed in a diluent that has been selected to encourage disso- ciation of virus from the food solids, which will be removed by centrifugation, filtration, or other means. Additives may be used to encourage separa- tion of the food solids from the liquid suspension containing the virus. Because foods suspected of viral contamination are likely to contain bacteria from both the food and faeces, a step to remove or kill bacteria in the suspension is usually included. And, because the detection methods available can usually accommodate only very small volumes of sample, as much water as possible is removed from the food sample extract before testing begins. Ap- plicable concentration methods include adsorp- tion-elution, differential precipitation, ultracen- trifugation, ultrafiltration, etc. Methods for detecting viruses extracted from foods As was mentioned above, infectivity tests are not applicable to the viruses that are most often food- borne. Nevertheless, infectivity testing may be appropriate when astroviruses, rotaviruses, and perhaps a few others are sought. The remaining detection methods are based on the morphology, antigenic specificity, or genetic specificity of the viral particle. Morphology of viral partides affords an important basis for their classification (Tabl.e 1). However, many viruses not of human origin are indistin- 97 guishable by simple electron microscopy from vi- ruses suspected as food contaminants. One way to demonstrate that the viruses seen in the electron micrograph are indeed the suspected type is to attach them to the grid or to each other with ho- mologous antibody ("immune electron microsco- py"), which is a combination of serological and morphological criteria. When the serological type of the virus is unknown, an alternative is to com- pare immune electron micrographs done with acute and convalescent serum from a patient, as- suming that paired samples are available. If the convalescent phase serum reacts with the virus and the acute phase serum does not, this is likely to have been the agent that caused the illness. Unfor- tunately, these methods are not nearly sensitive enough to detect the quantity of foodborne virus that would suffice to cause an infection. Antigenic specificity of the viral coat protein can serve directly as a means of detection by enzyme immunoassay or by radioimmunoassay. These methods have been of considerable use in diagnos- tic virology, where they are applied to faecal sam- ples containing levels of virus often exceeding a million particles per gram. Because contaminated foods are likely to contain only imperceptible levels of faeces, the quantities of virus present in food samples are likely to be much smaller, and immunoassay tests are seldom adequately sensitive. Therefore, serological reactions between virus and antibody are most often applied in combination with electron microscopy or with nucleic acid- based tests. Nudeic acid tests are generally based on the spe- cific interactions of portions of the viral genome with complementary probes, PCR primers, or both. PCR and other genetic amplification methods af- ford highly sensitive means of detecting small quantities of virus. Because almost all known food- borne viruses are RNA agents, the viral nucleic acid must be extracted and reverse-transcribed to com- plementary DNA before amplification by PCR can begin. The specificity of the copies is demonstrated on the basis of their appropriate length ( in terms of number of nucleotide bases), their reaction with a complementary probe that is directed at a por- tion of the amplified segment, or both. Probes used for this purpose are labeled with a radionu- clide, an enzyme, or some other means of express- ing their presence in association with the PCR- amplified product. In addition to the difficulties of performing the challenging PCR and probe-speci- ficity tests themselves, testing food samples has been found to present some special problems, in that certain food components interfere with re- verse transcription or with PCR amplification. Var- ious specialized means of surmounting these prob- lems are being reported (42). Although such nucle- ic acid-based test methods are exquisitely specific, it is possible to amplify or probe for a segment of 98 the viral genome that is common to more than one type, to afford a broader test when needed (43). Combined methods consist typically of a serologi- cal method followed by another detection proce- dure, as with the immune electron microscopy ap- proach described above. Virus has also been cap- tured from sample extracts with homologous anti- body for later detection by PCR (44). This method seems to obviate some of the problems with food inhibitors of reverse transcription and PCR, and it allows RNA to be released from the viral particle simply by heating. Detection methods combining nucleic acid tests and morphological tests seem not to have been devised. Outbreak investigation and routine monitoring present very different sets of priorities. When sever- al people have already been made ill, the relatively high costs of attempting to detect viruses in food samples may be acceptable, especially iflitigation is in prospect. However, as was mentioned earlier, an outbreak of hepatitis A affecting several people is likely to be recognized only 4 weeks or more after the contaminated food was consumed, so that per- tinent food samples may well be unavailable for testing. With other viral diseases that have shorter incubation periods, clinical histories and diagnos- tic samples of faeces and blood serum may afford some indication of which virus is to be sought in the food, which could prove very helpful. In con- trast, testing foods for virus contamination in the hope of preventing human illness presents the spe- cial problems that costs are likely to outweigh de- monstrable benefits and that almost all foodborne viruses are transmitted by faecal-oral cycles, so that one is as likely as another to occur in food subject to faecal contamination. As ingenious as the meth- ods are that have been devised for detecting food- borne viruses, none could yet be properly de- scribed as routine. Indicators as alternatives for monitoring foods Faecal coliform bacteria and Escherichia coli have long been used for monitoring shellfish and their growing waters, but bacteria are essentially irrele- vant to the presence of viruses in shellfish and in other foods (33, 34). Human enteric viruses capa- ble of replicating in laboratory cell cultures and producing cytopathic effects have been proposed as alternate indicators of virus contamination. At least in instances of contamination via community sewage, there may be enough of such viruses, par- ticularly vaccine polioviruses, to afford an indica- tion of viral contamination. Infectivity tests are in- herently quite sensitive, but are expensive and may take several days to a few weeks to produce firm results. Other candidate indicators of viral contam- ination of food have been bacteriophages that in- fect intestinal bacteria, such as E. coli or Bacteroides Jragilis. These bacteriophages are relatively easy to detect, with readout times of as little as 6 h, but they Rapp. trimest. st.atist. sanit. mond., 50 (1997) do not seem to be more reliably present than cyto- pathic human viruses. In general, any of these indi- cators might be of value in identifying foods at risk of harbouring viruses via sewage contamination, but not when direct contamination by faeces-soiled hands was at issue. Preventing the transmission of virus via foods Clearly, the reason for compiling information re- garding virus transmission via foods is in the hope of preventing this foodborne disease. Viruses present some unique features among foodborne disease agents. Since they cannot multiply in foods, one might suppose that control would be relatively easy, but this has not proven to be the case. Preventing contamination With the sole exception noted, foodborne viruses come from human faeces and can be kept out of foods by preventing human faecal contamination. Obviously, this has proven a difficult task, even in the matter of pre-harvest monitoring of shellfish and their growing waters. Conditions that had been judged acceptable can change very rapidly, with highly unforeseen results (30). Non-potable water used in food preparation can introduce viral contamination, as with Norwalk virus and celery in a recorded outbreak (45). Contamination via the hands of infected food handlers is quite another matter. A study in Germany has shown that kitchen staff, especially over the age of 30, often show evi- dence of having been infected with hepatitis A (46). Staff infections that are completely inappar- ent (47) have led to recorded outbreaks, but incu- bating infections in the case of hepatitis A (48) and briefly persistent infections in persons conva- lescing from viral gastroenteritis (14) have been greater problems. Proper, diligent hand washing is the general preventive in these instances ( except for very few instances in which contamination by vomitus has been suspected). Faeces on the hands of infected persons may contaminate food, wheth- er the person is a field worker, a kitchen worker, or a server. Even baked bread has been found suscep- tible to hands-on contamination, as in an outbreak in the United Kingdom (49). Gloves may be of some value in preventing hand contact, but avail- ability and enforced use of proper hand washing facilities are the most important precautions to prevent direct human contamination of foods with viruses (50, 51). Vaccination against hepatitis A is also of value in locations where natural immunity from infection is not regularly attained early in life (8). Inactivation of viruses in foods Virus cannot multiply in food and it can be inacti- vated before being consumed. Storage at room temperature may encourage inactivation of viruses Wld hlth statist. quart., 50 (1997) in food, but may lead to bacterial hazards. Thermal processing is generally effective, although some further attention to the adequacy of milk pasteur- ization to inactivate hepatitis A virus is needed (52). The British Government recommends heating mollusks to 85-90 °C for at least 90 seconds to de- stroy viruses (53). Alternately, shellfish may be de- purated (held in facilities supplied with clean sa- line water) or relayed from where they grew into clean water to purge themselves of contaminants. Although these practices have served well in re- moving pathogenic bacteria from shellfish, success with viruses is not guaranteed. Longer periods of treatment are required for removal of viruses than for bacteria (25, 38), and individual shellfish may not cleanse themselves. Other food processes that are not based on heat are less effective: viruses present a small target for ionizing radiation, en- teric viruses are generally acid-resistant, and the hepatitis A virus, at least, is quite resistant to drying. Viruses in water or on surfaces can be inactivated by strong oxidizing agents such as chlorine or ozone and by ultraviolet light. Hazard Analysis Critical Control Point (HACCP) Food safety, at least in the developed world, is increasingly predicated on hazard analysis critical control point (HACCP) systems, along with appro- priate general precautions. Foods subject to faecal contamination via wastewater or contact by the faeces-soiled hands of infected persons are at risk of contamination with viruses; this is the essence of hazard analysis in this case. Critical control points are those at which contamination can be prevented (as in disinfecting potentially contaminated water so that virus is not canied to food) or undone (such as by cooking an "at-risk" food thoroughly after it is last handled). Thorough hand washing, or prevention of hand contact with foods that will not be cooked thereafter before being served, are also very important measures. Although depura- tion or relaying of potentially contaminated shell- fish may have some preventive value, these prob- ably should not be considered critical control points in preventing virus transmission via shellfish eaten raw. Acknowledgement This paper is adapted from a Scientific Status Sum- mary written for the Institute of Food Technolo- gists, Chicago, United States of America. The author is grateful to the Institute of Food Technol- ogists for permission to do this. Summary Viruses are transmitted to humans via foods as a result of direct or indirect contamination of the foods with human faeces. Viruses transmitted by a faecal-oral 99 route are not strongly dependent on foods as vehicles of transmission, but viruses are important among agents of foodborne disease. Vehicles are most often molluscs from contaminated waters, but many other foods are contaminated directly by infected persons. The viruses most often foodborne are the hepatitis A virus and the Norwalk-like gastroenteritis viruses. Detection methods for these viruses in foods are very difficult and costly; the methods are not routine. Indicators that would rapidly and reliably suggest the presence of viral contamination of foods are still being sought. Contamination can be prevented by keeping faeces out of food or by treating vehicles such as water in order to inactivate virus that might be carried to food in this way. Virus cannot multiply in food, but can usually be inactivated by adequate heating. Other methods of inactivating viruses within a food are relatively unreliable, but viruses in water and on exposed surfaces can be inactivated with ultraviolet light or with strong oxidizing agents. Resume Transmission des virus par les aliments Les virus se transmettent a l'homme par l'intermediaire des aliments a la suite d'une contamination directe ou indirecte de ceux-ci par des matieres fecales humaines. Bien qu'utilisant peu les aliments comme vehicule, les virus transmis par voie feco-orale figurent parmi les principaux agents de maladies d'origine alimentaire. Le vehicule est souvent un mollusque provenant d'eaux contaminees, mais de nombreux autres aliments sont directement contamines par des personnes infectees. Les virus les plus souvent transmis par les aliments sont le virus de l'hepatite A et les virus des gastro-enterites de type Norwalk. Les methodes de detection de ces virus dans les aliments, coOteuses et d'application dif- ficile, ne sont pas utilisees en routine. On recherche des indicateurs qui permettraient de signaler rapidement et avec une bonne fiabilite la presence d'une contamina- tion virale dans les aliments. On peut empecher la contamination en evitant tout contact entre les matieres fecales et les aliments ou en traitant les vehicules tels que l'eau afin d'inactiver les virus qui pourraient etre transmis aux aliments par cette voie. Les virus ne peuvent se multiplier dans les aliments, et on peut habituellement les inactiver par un chauffage suffisant. Les autres methodes d'inactivation des virus dans un aliment donne sont relativement peu fiables mais les virus presents dans l'eau ou a la surface de !'aliment peuvent etre inactives par exposition au rayonnement ultraviolet ou a des oxydants puissants. References/Refer +ences 1. Bean, N.H. et al. Surveillance for foodborne-disease outbreaks-United States, 1988-1992. CDC suroeillance summarie.-, Morbidity and mortality weekly report, 45(SS-5): 1-66 (1996). 2. Bean, N.H. et al. Foodborne disease outbreaks, 5-year summary, 1983-1987. Journal of food prot,ection, 53: 711-728 (1990). 3. Cliver, D.O. Vehicular transmission of hepatitis A. Public health reviews, 13: 235-292 ( 1985). 100 4. World Health Organization. WHO surveillance programme for control of foodborne infections and intoxications in Europe. Sixth report ( 1990-1992), Copenhagen, EURO, 1993. 5. Cliver, D.O. Implications of food-borne infectious hepatitis. Public health reports, 81: 159-165 (1966). 6. Cromeans, T. et al. Hepatitis A and E viruses. In: Hui, Y. H. et al. (eds), Foodborne disease handbook: Vol. 2. Diseases caused by viruses, parasites, and fungi. New York, Marcel Dekker, 1994, pp. 1-56. 7. United States Public Health Service. Food Code. Washington, DC, Food and Drug Administration, 1995. 8. Public health control of hepatitis A: Memorandum from a WHO meeting. Bulletin of the World Health Organization, 73: 15-20 (1995). 9. Greenberg, H.B. et al. Role of Norwalk virus in outbreaks of non bacterial gastroenteritis. Journal of infectious disease, 139: 564-568 (1979). 10. Appleton, H. Norwalk virus and the small round viruses causing foodborne gastroenteritis. In: Hui, Y.H. et al. (eds), Foodborne disease handbook: Vol. 2. Diseases caused by viruses, parasites, and fungi. New York, Marcel Dekker, 1994, pp. 57-79. 11. Viral gastroenteritis sub-committee of the PHLS virology committee. Outbreaks of gastroenteritis associated with SRSV's. PHLS microbiology digest, 10(1): 2-8 (1993). 12. Parrino, T.A. etal. Clinical immunity in acute gastroenteritis caused by Norwalk virus. NewEnglandjournalofmedicine, 297: 86-89 (1977). 13. K.uritsky, J.N. et al. Norwalk gastroenteritis: a community outbreak associated with bakery product consumption. Annalr of internal medicine, 100: 519-521 (1984). 14. White, K.E. et al. A foodborne outbreak of Nonvalk virus: evidence for post-recovery transmission. American journal of epidemiolog)', 124: 120-126 (1986). 15. Sattar, S.A. et al. Rotavirus. In: Hui, Y.H. et al. (eds), Foodborne disease handbook: Vol. 2. Diseases caused by viruses, parasites, and fungi. New York, Marcel Dekker, 1994, pp. 81-111. 16. Alhaijar, BJ. et al. Transport modelling of biological tracers from septic systems. \.tater research, 22: 907-915 (1988). 17. Osherovich, A.M. & Chasovnikova, G.S. Study on the isolation of enteroviruses from environmental objects (in Russian). In: Chumakov, M.P. (ed), Materialy problemnoi komissiiAkademia Meditsinkikh Nauk SSSR "Poliomyelit i virusnye entsefality," Vypusk 1, Ent,erovirusy. Moscow, Akademia Meditsinskikh Nauk SSSR, 1967, pp. 89-90. 18. New York Department of Health. A review of foodborne disease outbreaks in New Ym* State 1988. Albany, NY, Bureau of Community Sanitation and Food Protection, 1989. 19. United States Department of Health Education and Welfare. Aseptic meningitis outbreak at a military installation in Pennsylvania. Aseptic Meningitis Suroeillance, annual summary 1976, HEWCDC publication No. 79-8231. Atlanta, GA, U.S. Department of Health, Education, and Welfare, 1979, p. 11. 20. Gresikova, M. Tickborne encephalitis. In: Hui, Y.H. et al. (eds), Foodborne disease handbook: Vol. 2. Diseases caused lry viruses, parasites, and fungi. New York, Marcel Dekker, 1994, pp. 113-135. 21. Outbreak of tick-borne encephalitis, presumably milk- borne. Weekly epidemiological recard, 19: 140-141 (1994). Flambee d'encephalite a tiques, probablement transmise par le lait. Rei.eve epidemiologique hebdomadaire, 19: 140-141 (1994). 22. Cliver, D.O. Viruses. In: Cliver, D.O. (ed), Foodbornediseases. San Diego, Academic Press, 1990, pp. 275-292. 23. Centers for Disease Control and Prevention. Viral hepatitis surveillance program, 1990-1992. Hepatitis suroeillance report No. 55. Atlanta, Ga., Centers for Disease Control and Prevention (1994) pp. 19-34. 24. Appleton, H. Foodborne illness: foodborne viruses. Lancet, 336: 1362-1364 (1990). Rapp. trimest. statist. sanit. mond., 50 (1997) 25. Power, U.F. &: Collins, J.K. The production of microbiologically safe shellfish - lessons from the classification of shellfish at source. Environmental health, 97: 124-130 (1989). 26. DiGirolarno, R. et al. Survival of virus in chilled, frozen, and processed oysters. Applied microbiology, 20: 58-63 ( 1970). 27. Gard,S.1957. Discussion. In: Hartman, F. W. (ed), Hepatitis frontiers. Boston, Little, Brown & Co., 1957, pp. 241-243. 28. Gunn, R. A. et al. Norwalk virus gastroenteritis following raw oyster consumption. American journal of epidemiology, 115(3): 348-351 (1982). 29. Desenclos,J.-C.A. et al. A multistate outbreak of hepatitis A caused by the consumption of raw oysters. American journal of public health, 81(10): 1268-1272 (1991). 30. Halliday, M.L et al. An epidemic of hepatitis A attributable to the ingestion of raw clams in Shanghai, China. Journal of infectious diseases, 164: 852-859 (1991). 31. Wang, J.-Y. et al. Risk factor analysis of an epidemic of hepatitis A in a factory in Shanghai. International journal of epidemiology, 19(2): 435-438 (1990). 32. Koff, R.S. etal. Viral hepatitis in a group of Boston hospitals. III. Importance of exposure to shellfish in a nonepidemic period. New England journal of medicine, 276: 703-710 (1967). 33. Berg, G., ed. Indicators of viruses in water and food. Ann Arbor, Mich., Ann Arbor Science, 1978. 34. Wait, D.A. et al. Enteric bacterial and viral pathogens and indicator bacteria in hard shell clams. Journal of food protection, 46: 493-496 (1983). 35. Council for Agricultural Science and Technology. Foodborne pathogens: risks and consequences. Task Force Report No. 122. Ames, Ia., Council for Agricultural Science and Technology, 1994. 36. Potter, M.E. Risk assessment terms and definitions.journal of food protection, 1996 Supplement: 6-9 ( 1996). 37. Dolin, R. etal. Biological properties of Norwalk agent of acute infectious non bacterial gastroenteritis. Proceedings of the Soci- ety for Experimental Biology and Medicine, 140: 578-583 (1972). 38. Grohmann, G.S. et al. Norwalk virus gastroenteritis in volunteers consuming depurated oysters. Australian journal of experimental biology and medicine, 59(Pt. 2): 219-228 (1981). 39. Rose,J.B. &: Sobsey, M.D. Quantitative risk assessment for viral contamination of shellfish and coastal waters. Journal of food protection, 56: 1043-1050 ( 1993). Wld hlth statist. quart., 50 (1997) 40. Cliver, D.O. et al. Foodborne viruses. In: Vanderzant, C. & Splittstoesser, D.F. (eds), Compendium of methods for the microbiological examination of foods, Jrd ed. Washington, DC, American Public Health Association, 1992, pp. 763-787. 41. Cliver, D.O. Detection and control of foodborne viruses. Trends in food science and technology, 6: 353-358 ( 1995). 42. Gouvea, V. et al. Indentification of Norwalk virus in artificially seeded shellfish and selected foods. journal of virolological methods, 48: 177-187 (1994). 43. Jaykus, LA. et al. Development of a molecular method for the detection of entcric viruses in oysters. journal of food protection, 58: 1357-1362 (1995). 44. Deng, M.Y. et al. Detection of hepatitis A virus in environmental samples by antigen-capture polymerase chain reaction. Applied and environmental microbiology, 60: 1927-1933 (1994). 45. Warner, R.D. et al. A large non typical outbreak of Norwalk virus. Archilles of internal medicine, 151: 2419-2424 (1991). 46. Koster, D. et al. Hepatitis A immunity in food-handling occupations. European journal of clinical microbiology and infectious diseases, 9: 304-305 ( 1990). 47. Eisenstein, A.B. et al. An epidemic of infectious hepatitis in a general hospital. Journal of the American Medical Association, 185: 171-174 (1963). 48. Common source outbreak of hepatitis A - Ohio. Morbidity and mortality weekly report, 27: 86 (1973). 49. Warburton, A.R.E. et al. Hepatitis A outbreak involving bread. Epidemiology and infection, 106: 199-202 (1991). 50. Ansari,S.A. etal. In vivo protocol fortestingefficacyofhand- washing agents against viruses and bacteria: experiments with rotavirus and F..scherichia coli. Applied and environmental microbiology, 55: 3113-3118 (1989). 51. Cliver, D.O. &: Kostenbader, K.D.,Jr. Disinfection of virus on hands for prevention of food-borne disease. International journal of food microbiology, l: 75-87 (1984). 52. Parry,J.V. &: Mortimer, P.P. The heat sensitivity of hepatitis A vimsdetermincd by a simple tissue culture method.Journal of medical virology, 14: 277-283 (1984). 53. Institute of Food Science and Technology. IFST position statement on foodborne viral infections, dated 14 September 1996. London, Institute of Food Science and Technology, 1996. 101 Foodborne diseases in travellers R. Y. Cartwrighta & M. Chahedb Introduction Travel is a major economic force throughout the world. The number of people who go abroad on business or for pleasure increases every year (I). The growth of long-haul destinations takes travel- lers to increasingly exotic destinations exposing them to foods and micro-organisms which they have not previously met. The success of business negotiations or the enjoyment of a holiday can be marred by unexpected illness. The health of travel- lers is important however not only for the traveller but also to those whose livelihood depends on the travel industry. A resort with a high record of ill- ness does not encourage return visits or recom- mendations to friends. This may result in a signi- ficant drop in tourist income even if only for a limited time. Gastrointestinal disturbances have always been linked with travel and in spite of major food hy- giene programmes remain the principal travel as- sociated disease. The spectrum of disease and the causes are varied and although the routes of trans- mission involve contaminated food or water it is possible to divide foodborne diseases in travellers into 3 categories: those found in every country throughout the world, those with a marked geo- graphical distribution, and the characteristic con- dition known as 'travellers' diarrhoea'. A major difficulty in providing meaningful figures is the difference in microbiological and epidemiological services between countries. The approach to diar- rhoeal disease will also vary and is influenced by the level of such disease in the local community and the importance attached to its prevention. Al- though of great importance to travellers, a few days' diarrhoea is of little significance to the local population compared to their overall health prob- lems. However, the health of tourists, especially in those countries relying heavily on tourist income, may have a significant impact on the economy. Consequently, it is essential to understand dis- eases in travellers so that appropriate preventive measures can be taken. It is also important to recognize that measures to reduce travel-associated diseases may result in improvements to the public a Group Director, Consultant Microbiologist, PHLS South Thames, Guildford, England. b Head & Scientific Director of Laboratory, Sanitary Advisor to the ONTT, Ministry of Tourism, Tunis, Tunisia. 102 health infrastructure which will be beneficial to the local population. The provision of safe water is to the advantage of the entire community, perma- nent or transitory. A traveller may be at a greater risk of acquiring a foodborne disease for different reasons, includ- ing the increased use of hotels and restaurants, exposure to entero-pathogenic organisms not pre- viously met, variable levels of public health hygiene and environmental contamination, and differing concepts of food hygiene. Any illness may become apparent while the traveller is in the country where the infection was acquired, during another part of a journey or not until return back home. Medical and microbiological diagnostic facilities will vary from clinical assessment only, to clinical assess- ment with a full parasitological, virological and bacteriological work up. Epidemiological services will vary to an even greater extent as will the re- sponses to the report of infections in individuals or of a suspected outbreak. A further complicating factor is that there may be no dear division of responsibility between gov- ernment departments. Is illness in tourists a prob- lem for the ministry of health or the ministry of tourism or, if water is involved, the ministry of the environment? Foodborne diseases in travellers do not differ from any foodborne disease in that they are all preventable, providing the appropriate pre- cautions are taken, but they do differ in the poten- tial economic consequences. There is a considerable volume of literature on travellers' diarrhoea although the majority of pub- lications are concerned with the causative organ- isms, treatment or drug prophylaxis. There is a paucity of information on the epidemiology, routes of transmission and public health preven- tive measures. An improvement in the present situ- ation requires cooperative efforts by governments, public health personnel, both medical and non- medical, and all sections of the tourist industry. There is also a responsibility on the tourists them- selves, all too often simple personal hygienic pre- cautions such as hand-washing appear to be left at home. Travel and tourism have a very wide spectrum, from the business person travelling first class and staying in five-star hotels, the package-holiday tour- ist who may be self-catering, the independent trav- eller and the back packer. They all have two ele- ments in common: the journey and a destination Rapp. trimest. statist. sanit. mond., 50 (1997) away from their home, and exposure to food and drinks (including water) produced and/or pre- pared differently. Causes of travel-associated loodborne disease Travellers can be exposed to any cause of food- borne disease and the fact that they are travelling classifies it as travel associated. The following list illustrates some of the more common causes. It should be remembered that travel-associated infec- tions and intoxications will also occur in the local population and form part of the normal pattern of illness in that locality. Enterotoxigenic Escherichia coli Regarded as the major cause of travellers' diar- rhoea (2) the natural history and routes of spread of this organism are not well described. It usually causes a relatively mild illness and predominantly affects tourist from industrialized countries visiting tropical and sub-tropical areas (3-5). It is consid- ered to be spread through food and water (6) but is not normally recognized as causing point-source outbreaks, although an outbreak of diarrhoea on a cruise ship has been attributed to this organism (7). The countries and regions associated with E. coli diarrhoea generally are those with developing pub- lic health hygiene. Among Finnish tourists in Morocco a seasonal difference was observed (7). Other E. coli associated with travellers' diar- rhoea are enteropathogenic, enteroinvasive, enteroad- herent and enterohaemorrhagicstrains. Little is known about their routes of spread, although E. coli 0157 strains have been linked with raw or undercooked beef. Salmonella Salmonella are worldwide causes of foodborne dis- ease. Historically, food- and waterborne infections due the enteric fever organisms, Salmonella typhi and the Salmonella paratyphi, have stimulated the development of public health methods and legisla- tion to control their spread. These have been large- ly successful but the outbreaks in Zermatt and Kos illustrate how travellers can be infected (8, 9). Pre- travel immunization with typhoid vaccine is recom- mended for people travelling in conditions of doubtful hygiene ( JO). The gastroenteritis-causing salmonellas are re- sponsible for an unknown number of cases of food- borne illness. In England and Wales, from 1992 to 1994 there were over 92 OOO cases and nearly 6 OOO outbreaks of salmonellosis ( 11) in spite of generally good levels of public health and food hygiene. Outbreaks are associated with many types of food but poultry and eggs feature predomi- nantly. The production of mayonnaise in hotels and restaurants is a well recognized factor in food- borne outbreaks due to Salmonella enteritidis. There Wld hlth statist. quart., 50 (1997) have been outbreaks of salmonellosis associated with the consumption of contaminated airline meals (12, 13). Campy/obacters These can be found worldwide as a cause of diar- rhoea. They account for up to 15% of cases of travellers' diarrhoea from Thailand and Bangladesh (14, 15). In Morocco a seasonal varia- tion has been observed (7). Shigella Although shigella can be acquired through envi- ronmental contamination, it is likely that many of the infections in travellers are through the con- sumption of contaminated food or water. Shigella sonnei is common throughout the world, whereas Sh. baydii, dysenteriae and flexneri are more common- ly associated with developing countries. Shigella infection has also been associated with air travel (16). Vibrios Although Vibrio cholRrae can cause water- or food- borne illness in travellers, with the exception of pilgrimages, this is unusual. Vibrio parahaemolyticus may cause sporadic cases and outbreaks following consumption of inadequately cooked seafood. It is not uncommon in Japan and South-East Asia al- though cases have been reported from other re- gions. Bruce/la Brucellosis has been eradicated in many countries and infections that are seen may be associated with the consumption of inadequately pasteurized dairy produce while visiting countries where the infec- tion is still endemic in cows, sheep and goats. Viruses Many viruses may be spread through contaminated food and water although the predominant ones which are also associated with travel are hepatitis A, hepatitis E, rotavirus ( 17) and astrovirus. Although the small round structured virus may be spread through contaminated food it is usually associated with environmental contamination ( 18, 19). The two hepatitis viruses have been associated with both sporadic cases and outbreaks of disease (20- 23 ). The other viruses cause acute gastroenteritis and are associated in particular with shellfish con- sumption. Parasites Gut parasites are acquired through the consump- tion of contaminated food or water (24 ). Giardia lambliaand Cryptosporidiurn (25), both of which may be spread through drinking water, are distributed 103 worldwide (26-28), although travel-acquired giar- diasis has long been associated with the consump- tion of tap water in St. Petersburg (Leningrad). Entamoeba histolytica infection is a hazard for visitors to tropical and some subtropical countries. lsosospora belli (29, 30) and Blastocystis hominis (31) infections have been reported in travellers. Foodborne trematode infections, acquired by consumption of raw or inadequately processed freshwater fish, shellfish and aquatic plants, affect Table 1 more than 40 million people throughout the world; over 10% of the global population, particu- larly tourists travelling to endemic areas, is at risk of infection (32). Toxins Disease due to the consumption of food contain- ing toxins has a worldwide distribution affecting travellers and non-travellers alike. Staphylococcal The average incidence of subjective travelers' diarrhoea (STD) in British package holiday tourists during the 1996 Summer season Tableau 1 Incidence moyenne de la diarrhee des voyageurs «subjective» (DVS) chez des touristes britanniques ayant effectue un sejour organise pendant la saison d'ete 1996 Holiday region• %STD- No. of questionnaires analysed - Destination de vacances• % DVS Nombre de questionnaires analyses Egypt - Egypte 63 13 134 Dominican Republic - Republique dominicaine 57 31 218 Kenya 56 3 469 Mexico - Mexique 49 3 502 Ionian Coast, Turkey - Cote ionienne, Turquie 43 2 750 Tunisia - Tunisie 37 6 409 Aegian Riviera.Turkey - Cote de lamer Egee, Turquie 37 30 395 Jamaica - Jamaique 30 2 272 Saint Lucia - Sainte-Lucie 26 1 909 Antigua -Antigua 22 1 612 Costa Dorada, Spain - Costa Dorada, Espagne 16 26 306 Tenerife South - Tenerife Sud 13 21 660 Algarve, Portugal 13 23 358 Zakynthos, Greece - Zakynthos, Grece 13 19 752 Gran Canaria - Grande Canarie 13 27 381 Barbados - Barbade 12 3149 Rhodes 12 25 619 Greek Mainland - Grece continentale 12 978 Costa Brava, Spain - Costa Brava, Espagne 12 18 045 Crete - Crete 12 20 530 Corfu - Corfou 11 34 949 Majorca - Majorque 11 151 043 Kos-Kos 11 13 727 Malta - Malte 10 22 698 Minorca - Minorque 9 46 928 Cyprus - Chypre 9 40 950 Tenerife North - Tenerife Nord 8 7 638 Florida - Floride 7 21121 Adriatic Riviera, Italy - Cote adriatique, ltalie 6 2 015 Austria - Autriche 5 3198 Switzerland - Suisse 3 116 Source: R. Y. Cartwright, unpublished data, 1997. - Donnees non publiees, 1997. • The holiday regions are those designated by the tour operator and have no other significance. - Les destinations de vacances sont celles qui sont indiquees par le voyagiste et n'ont pas de signification particuliere. 104 Rapp. trimest. statist. sanft. mond., 50 (1997) Table 2 The incidence of travelers· diarrhoea in tourists visiting Tunisia during late August and the beginning of September, the peak season for TD Tableau 2 Incidence de la diarrhee des voyageurs chez des touristes ayant sejourne en Tunisie fin aoOt et debut septembre, periode maximale pour cette affection French- German- English- Italian- Total% (all nationalities) - Total number Franca is Allemands Anglais ltaliens "lo du total (toutes of all international nationalites) visitors questioned - Nombre total de visiteurs de toutes nationalites interroges 1985 27 55 47 25 40 8 352 1987 18 40 46 25 35 7 398 1989 15 33 38 15 25 10 816 1992 15 33 50 18 27 11 213 Note: With each survey, a comparative study was carried out to compare results with a tourist-sending country that surveyed its own travelers upon their return from Tunisia.- Pour chaque enquete, une etude comparative a eM realisee avec les resultats d'un pays ayant interroge ses ressortissants a leur retour de Tunisie. Source: M. Chahed, ONTT, unpublished data, 1996. - Oonnees non publiees, 1996. food poisoning is one of the commonest causes of foodborne intoxication. Ciguatera fish poisoning is caused by the presence in the consumed fish of toxins elaborated by the dinoflagellate Gambierdis- cus toxicus and other algae that grow on reefs under the sea. It is seen in the Caribbean, southern Florida, Hawaii, the South Pacific and Australia - all popular holiday destinations (33, 34). Epidemiology of travel-associated foodborne disease The size of the problem is unknown as surveillance systems round the world are very variable in both the collection of data and the reporting of results. It is however a problem well recognized by health professionals, the tourist industry and travellers. Travellers' diarrhoea has been the subject of many studies and as the causative organisms are most likely to have been ingested in contaminated food or water, the incidence is a measure of the back- ground level of food and public health hygiene in the places visited. The anecdotal observation that the incidence is higher in developing countries and areas with poor public health hygiene is generally supported. A study involving 16 568 travellers returning home on charter flights to Switzerland and Germany used questionnaires completed during the flight: 5.1 % of travellers to North America had travellers' diar- rhoea compared to levels exceeding 30% in those returning from tropical and subtropical countries (35 ). An Austrian study involving 3 696 tourists also supports high rates in subtropical countries and indicates seasonal variations (36). The incidence among 485 persons attending a European Congress in the United Kingdom was 2.8% for the 143 British residents and 0.6% for 342 visitors (37). Wld hlth statist. quart., 50 (1997) An ongoing study over 15 years in conjunction with a major British package tour operator uses a health question in the client satisfaction survey undertaken during the return flight (38). In excess of one million data sets are analysed each year covering in 1996, 88 holiday regions, 454 resorts and 3 288 hotels. The data although relatively crude provide a useful measure of the incidence of subjective travellers' diarrhoea in different resorts at the same time and the same resort at different times. In summer 1996, the incidence of subjective travellers' diarrhoea varied from 3% in Switzerland to 63% in Egypt (Tab/,e 1). In Tunisia 4 statistical studies have been made among departing tourists during the following years: 1985, 1987, 1989 and 1992. These studies were carried out at the end of the summer when incidents of travellers' diarrhoea occurred at their highest frequency. The twin ob- jectives of these studies were to (i) EVALUATE the incidents of tourist diarrhoea by nationality, region and hotel category; and (ii) to SDISITTZE those responsible in the tourist industry to the impor- tance of PREVE:-JTIO:-.i of health hazards for travellers (Tab/,e 2). The value of these studies is to provide a mea- sure for public health authorities on the efficacy of control measures that are instituted. Reliance on routine reporting of foodborne illness in trav- ellers is not adequate for monitoring the situa- tion. The use of questionnaires at airports is time consuming and expensive, although it does en- able discriminating data to be obtained. The in- clusion of a basic health question in tour opera- tors' satisfaction questionnaires has proved a rel- atively crude but effective method and has the added advantage that it maintains awareness of the problem in the tourist industry encouraging remedial actions to be taken. 105 The prevention al load- and waterbarne disease in tourists The problem is recognized if not precisely defined. This does not however prevent effective actions from being taken to reduce the incidence. Two international meetings sponsored by the World Health Organization (WHO) and the World Tour- ist Organization (WTO) have been held on the topic of Food Safety and Tourism. The first was a Regional Conference for Africa and the Mediterra- nean held in Tunis, Tunisia in 1991 and the second an Inter-American Conference on Food Protection and Tourism held in Cancun, Mexico in 1992. At both meetings it was recognized that throughout the world it was essential to maintain high standards of food hygiene in all catering establishments and that any lapses could result in outbreaks of food- borne disease. In addition, in areas associated with a high incidence of travellers' diarrhoea, there was a need for significant improvements in general public health hygiene and, in particular, there were often defective water supply and sewage disposal systems. It was also recognized that the occurrence of food borne diseases in tourists can be minimized, with the involvement of the traveller, tour operator, management of tourist establishments, and health and tourism authorities. This combined approach was regarded as essential for any real progress to be made. Governments also need to consider the im- portance of safeguarding the health of tourists within the larger context of the value of tourism to the whole community. It has been said that healthy tourists equal healthy profits. The health of tourists cannot be separated from the economic conse- quences of failure (39). At the Tunisian meeting it was recognized that, in addition to poor sanitary environments, certain common infrastructural/ administrative,Jactors in countries in Africa and the Eastern Mediterranean contribute to the high incidence of foodborne dis- ease, including: • lack of awareness of relevant issues by health and tourism sectors when considering con- sumer protection; • insufficient information on the magnitude of the food contamination problem; • inadequate legislation and poor implementa- tion of existing legislation; • inadequate evaluation procedures for both food safety programmes and activities; • insufficient involvement of the food industry in food safety issues; problems associated with street food-vending; • inadequately equipped laboratories which are poorly staffed; • poorly-trained food inspectorate; • lack of adequate collaboration between differ- ent government agencies, the private sector and consumer associations concerned with food safety; and 106 • lack of inter-country cooperation on food safety. The evidence available suggests that neither regulatory control nor law enforcement measures alone can contribute significantly to the control of foodborne diseases. One effective strategy for their prevention and control lies with the catering estab- lishments and restaurants themselves as well as an informed travelling public. To this end, WHO has issued important Rules for hygiene in food-service and mass catering establishments,C a Guide on safe food for travellers,d as well as a Training guide on safe food handling for managers of food service establishments (40). It was recognized that new technologies modify the way today's catering professionals must work. This in turn creates certain demands on the con- ception and organization of restaurant catering services. Safe, wholesome and attractive food is essential for the quality of tourism and health of travellers. As such, it should be identified with tourist destina- tions, considered a tourist asset and promoted by tourism organizers. The 30 recommendations from the Tunisian meeting recognize the important roles of govern- ments, tour operators, hoteliers and restaurateurs, and emphasize adequate training of food handlers. • a system for the investigation and epidemiolog- ical surveillance of foodbome diseases should be established in each country of Africa and the Eastern Mediterranean as an essential food safe- ty measure. Public health authorities should develop a protocol for carrying out investiga- tions and surveillance of foodbome diseases. Foodborne disease surveillance data should be used to identify critical food-handling practices as a means of setting priorities; • the Hazard Analysis Critical Control Point (HACCP) approach should, wherever possible, be widely adopted by the food service industry as an accepted tool for the prevention and con- trol of food contamination at all points in the food chain. Often food service enterprises find difficulty in the application of this approach; • to brief the mass media on food safety issues, seminars and workshops should be organized by tourist-receiving countries; • WHO, in collaboration with WTO, should pro- duce documentation to inform the tourism sec- tor on measures and methodologies to follow when an outbreak of foodbome diseases occurs at tourist resorts; c World Health Organization. Hygi,ene in food-seruice and mass catering establishments, important ru/,es. Geneva, WHO, 1994 ("WHO/FNU/FOS/94.5). d World Health Organization. A guide on safe food fur travellers. Geneva, WHO, 1994 (WHO/FNU/DOS/94.2). Rapp. trimest. statist. sanit. mond., 50 (1997) • to ensure the safety of food during its prepara- tion, countries should develop strategies, pro- grammes and activities to ensure that food service workers are sufficiently well trained to identify food safety risks and to take preventive actions. Such training for food-handlers should be made mandatory in all food service establishments; • governments and tourist associations should consider supporting jointly, special pro- grammes to improve the safety and quality of street foods; • countries choosing tourism as a development option should foresee that the supply of food (whether locally produced or imported) for tourists and its safety are given due importance. Measures taken to this effect should benefit the whole population; • tour operators (TOs) should develop and use a standard hygiene checklist for hotels. There should be a pre-season inspection covering facili- ties and services including buildings, equip- ment, and an in-season inspection covering opera- tional aspects including personnel. Such check- lists should be based on the HACCP approach; • TOs as well as local tourist organizations should periodically review the advice given to tourists and develop more effective methods for in- forming and educating them; • TOs should develop or support the develop- ment of a common data base on illnesses ac- quired by their clients. This will require closer co-operation than hitherto between TOs and health authorities in both originating and re- ceiving countries; • a liaison mechanism including TOs, tourism and health administrations should be estab- lished in all receiving countries; • TOs should ensure that tour guides employed by them are knowledgeable on food and safety issues; • TOs should be more actively involved in pre- ventive programmes aimed at improving stan- dards of food safety; • academic institutions are encouraged to con- duct basic research in foodbome diseases in travelers; • in collaboration with TOs, academic institutes and other generating bodies should undertake intervention studies. Cooperation between in- formation generating institutions should be en- couraged; • travelers diarrhoea (TD) can primarily be pre- vented by improving the hygiene standards of food and beverages served to visitors. Interven- tion studies, using the most refined tools can ascertain how this can best be achieved; • until such times as the incidence of TD can be drastically reduced, travelers must be instructed about the possibilities of self-therapy of TD when visiting high-incidence destinations; Wld hlth statist. quart., 50 (1997) • managers of tourist establishments are encour- aged to notify all occurrences of food borne dis- eases in their establishment, to facilitate better diagnosis and assessment of the situation, as well as self-control and better collaboration with health authorities, TOs and eventually con- cerned international organizations; • the role of the mass media should be informa- tive, undiscriminating, unbiased, accurate, e.g. location of food borne disease outbreaks should clearly be identified; • evaluation is an essential component for all na- tional food safety programmes, in order to ver- ify whether or not activities undertaken are/ or have been effective. Likewise, evaluation is an effective tool that should be used by all sectors associated with improving food safety standards for travelers; • an effective co-ordinating mechanism to facili- tate collaboration between responsible minis- tries and other authorities as appropriate (health, tourism, agriculture, industry, com- merce, education, etc.) should be established to examine at the national level all aspects relat- ing to food safety and tourism; • as a practical measure to minimize food safety hazards to tourists, local health authorities in collaboration with tourism authorities should ensure hygiene standards in those eating estab- lishments close to or within easy access of tour- ist hotels; • the curricula of hotel schools should place due emphasis on hygienic aspects, with special at- tention being given to the role of food safety. Such schools should include training in the basics of food microbiology and biochemistry. Such training should be practical and easily adapted to the working situation; • food safety issues associated with tourist estab- lishments cannot be undertaken effectively in isolation from their environment. Any action to improve the safety and quality of food in these establishments must therefore be fully inte- grated in the overall effort to improve environ- mental hygiene; • when food-safety training programmes are formu- lated, priority should be given to training pro- grammes for management, who in tum should be able to train their staff in safe food-handling practices and ensure they are adhered to; • management of hotels, restaurants and similar establishments should facilitate the training of their employees in the basic principles of food safety and safe food-handling practices. The training received should be appropriate to the level of responsibility; • in consultation with ministries of health, the establishment of hygiene/tourist health sec- tions within ministries of tourism (as has been done in Tunisia) should be encouraged; 107 • as a practical measure to minimize food-safety hazards for tourists and the local population alike, national health authorities in collabora- tion with tourism authorities, catering and food service establishments and their respective asso- ciations, should establish closer working rela- tions with producers and distributors of pri- mary raw ingredients and processed food prod- ucts to encourage the use of accepted quality control, production, processing and distribu- tion procedures; • health education of the public (not forgetting schoolchildren) on food safety issues is an es- sential component in any programme to reduce the incidence of foodborne diseases. Health and tourism administrations should work close- ly together to develop appropriate "health mes- sages". Periodic evaluation of such activities should be undertaken to assess their effective- ness; • all countries need to formulate coherent poli- cies to ensure the safety of food as it relates to tourism. The implementation of such policies can only be effective if it is supported by sound food legislation which is supported by national standards and codes of practice relevant to countries' needs. To this end, all countries should intensify their cooperation with United Nations specialized agencies and collaborating organizations within their field of competence (e.g. WHO, WfO, ILL, FE). The deliberations in Mexico were similar, with particular emphasis on ensuring that all ed- ucational programmes take into account the edu- cational level of the target audience. Northern European and American food handling and hy- giene training packages may be ideal for workers in the country of production but totally inade- quate in many major tourist destinations. The importance of ensuring that the public health hygiene for the local population is commensu- rate with the standards required for tourists was stressed. The tourist industry depends on the lo- cal population whose health will also be affected by unsafe water supplies and sewage disposal systems. In Tunisia, apart from regular surveillance and the control by relevant government departments, a number of initiatives and programmes have been under way (from 1987-1992): (i) to evaluate statistically the situation and the incidence of TD during the high frequency occurrences at the end of the summer (Tabl.e 2); (ii) to alert the responsible parties and educate the hotel personnel to initiate a self-check sys- tem in hotels, to appoint a "hotel hygienist" responsible for the overseeing of critical as- pects of food handling, and educating the hotel personnel on this topic; 108 (iii) to develop the cooperation with international organizations and research institutes with the twin aims of identifying problem areas and exchanging experiences wi.th a view to formu- lating appropriate strategies. In the second phase (from 1993 onwards) ex- panding actions are being undertaken as a result of the Regional Conference on Food Safety and Tour- ism in Tunis in November 1991, namely: • the training and implementation of the first steps of the HACCP system in tourist establish- ments; • the development of a system of investigation and epidemiological follow-up of outbreaks of foodborne diseases; • the creation of a technical committee on a cen- tral as well as regional level, paving the way for national strategies and evaluations of action plans for the prevention of outbreaks of food- borne diseases; and • the strengthening of collaboration between government departments concerned with food hygiene. The International Tourist Health Association has suggested that ministries of health should have a "tourism unit" and ministries of tourism a "health Unit". In Turkey, for example, a multi-sectoral working group for tourist health was estalished in 1993. This group is led by the assistant under- secretary of the Ministry of Tourism. The members are from the Ministries of Health, Tourism, Agri- culture, Environment and Education. There are also representatives from the Turkish tourist indus- try whose presence is regarded as essential for the success of the Group. They are actively involved in the planning of preventive measures which will promote the health of tourists. Summary It is encouraging that the problem of food- and water- borne disease in travellers is being increasingly recog- nized, but it is disappointing that to a large extent there is little effective action being taken. It is essential that the problem be recognized across government depart- ments and joint programmes developed, particularly between ministries of health and tourism. Foodborne disease will always be a risk to travellers but this risk can be reduced by relatively simple programmes. The initial stage is to recognize the problem not just as an incon- venience to travellers but as an important economic factor which, unless controlled, can have severe effects on local, national and international trade. Rapp. trimest. statist. sanit. mond., 50 (1997) Resume Maladies d'origine alimentaire chez les voyageurs II est, d'une part, encourageant de constater que le probleme des maladies transmises par les aliments ou par l'eau chez les voyageurs est de plus en plus large- ment reconnu et, d'autre part, decevant de voir que dans la plupart des cas, rares sont les mesures effecti- vement prises. II est indispensable que le probleme soit reconnu a un niveau eleve et que des programmes conjoints soient elabores, notamment entre les ministe- res de la sante et du tourisme. Les maladies d'origine alimentaire seront toujours un risque pour les voya- geurs, mais ce risque peut ~tre reduit avec des pro- grammes relativement simples. II taut en premier lieu reconna1'tre ce probleme non seulement comme desa- grement pour les voyageurs mais aussi comme impor- tant facteur economique qui, faute d'attention, peut avoir de graves effets sur le commerce local, national et international. References-References 1. Gl.obal tourism forecasts to the year 2000 and beyond - Executive summary. Madrid, World Tourism Organization, 1993. 2. Black, R.E. Epidemiology of travelers' diarrhea and relative importance of various pathogens. [Review]. Reviews of infectious diseases, 12: 5 73-5 79 ( 1990). 3. Chapman, P.A. &: Mitchelmore, D.L A two-year survey of the incidence of heat-labile enterotoxin-reducing Escherichia coli and other enteric pathogens in travellers returning to the Sheffield area. Epidemiol.ogy and infection, IOI: 239-247 (1988). 4. Jertbom, M. &: Svennerholm, A.M. Enterotoxin-producing bacteria isolated from Swedish travellers with diarrhoea. Scandinavian journal of infectious diseases, 23: 4 73-9 ( 1991). 5. Gascon, J. et al. Etiology of traveller's diarrhea in Spanish travellers to developing countries. European journal of epidemiol.ogy, 9: 217-23 (1993). 6. DuPont, H.L. et al. Emporiatric enteritis : Lessons learned from U.S. students in Mexico. Transactions of the American Clininical and Climatol.ogical Association, 97: 32-42 ( 1985). 7. Mattila, L. et al. Seasonal variation in etiology of travelers' diarrhea. Finnish-Moroccan Study Group. Journal of infectious diseases, 165: 385-8 ( 1992). 8. Bernard, R.P. The Zermatt typhoid outbreak in 1963.Journal of hygiene, 63: 537-63 (1965). 9. Christeson, B. &: Andersson, Y. Typhoid fever among Swedish visitors to Kos in 1983. Journal of infection, 10: 223-7 (1985). 10. lnl£rnational trauel and health - uaccination requirements and health advice. Geneva, WHO, 1997. 11. Djuretic, T. et al. General outbreaks of infectious intestinal disease in England and Wales 1992 to 1994. Communicab/,e disease review, 6: R57-63 (1996). 12. Burslem,C.D.etal.Food poisoning-a majorthreatto airline operations [published erratum appears in: Journal of the Society for Occupational Medicine, 41 (2): 95 ( 1991)]. Journal of the Society of Occupational Medicine, 40: 97-100 ( 1990). Wld hlth statist. quart., SO (1997) 13. Hatakka, M. Salmonella outbreak among railway and airline passengers. Acta ueterinaria scandinauica, 33: 253-60 (1992). 14. Taylor, D.N. &: Escheverria, P. Etiology and epidemiology of travellers' diarrhea in Asia. Reviews of infectious diseases, 8(suppl): S 136-141 (1986). 15. Taylor, D.N. et al. Treatment of travellers' diarrhea: ciprofloxacin plus loperamide compared with ciprofloxacin alone. A placebo-controlled, randomized trial. Annals of inl£rnal medicine, II 4: 731-4 (1991). 16. Hedberg, C.W. et al. An international foodborne outbreak of shigellosis associated with a commercial airline. Journal of the American Medical Association, 268: 3208-12 (1992). 17. Fukui, Y, et al. Rotavirus infections in traveler's diarrhea. [in Japanese]. Kansenshogaku [lasshi Journal of the Japanese Association/or Infectious Diseases], 63: 1296-300(1989). 18. Ho, M.S. et al. Viral gastroenteritis aboard a cruise ship. Lancet, 2: 961-5 (1989). 19. McEvoy, M. et al. An outbreak of viral gastroenteritis on a cruise ship. Communicab/,edisease review, 6: Rl88-192 (1996). 20. Le Bras, M. et al. Epidemioiogie et clinique des maladies tropicales d 'importation. Revue de midecine inl£rne, 13: 205-10 (1992). 21. Maguire, H. etal. The epidemiology and control of hepatitis A. Com111unicab1Rdiseasereport, 2: Rll4-7 (1992). 22. van Zeijl,J.H. et al. Unrecognized hepatitis following a visit to Pakistan: a case of hepatitis E in The Netherlands. [in Dutch]. Nederl,ands Tijdschrift voor Geneeskunde, 136: 2153-5 (1992). 23. Xu, Z.Y. et al. Ecology and prevention of a shellfish- associated hepatitis A epidemic in Shanghai, China. Vaccine, 10: S67-8 (1992). 24. Wright, S.G. Parasites and travellers' diarrhoea. Scandinavian journal of gastroenterology, 84(suppl): 25-29. (1983). 25. Shiota, T. et al. [Imported cryptosporidiosis-reportofa case in Japan and review of the literature - in Japanese]. Kansenshogaku lasshi [journal of the Japanese Association for Infectious Diseases], 68: 941-5 (1994). 26. de Lalla, F. et al. Outbreak of Entamoeba histolytica and Giardia lamblia infections in travellers returning from the tropics. Infection, 20: 78-82 (1992). 27. Gray, S.F. &: Rouse, A.R. Giardiasis-a cause of travellers' diarrhoea. Communicab/,ediseasereport, 2: R45-47 (1992). 28. West, P.A. Human pathogenic viruses and parasites: emerging pathogens in the water cycle. Society for Applied Bacteriology, Symposium series, 20: 107S-114S (1991). 29. Shaffer, N. &: Moore, L. Chronic travellers' diarrhea in a normal host due to Isospora belli [letter] .Journal of infectious diseases, 159: 596-7 (1989). 30. Rijpstra, A.C. &: Laarman, J.J. Repeated findings of unidentifyed small Isospora-like coccidia in faecal specimens from travellers returning to The Netherlands. Tropical and geographical medicine, 45: 280-2 (1993). 31. Kain, K.C. et al. Epidemiology and clinical features associated with Blastocystis hominis infection. Diagnosis and microbiol.ogy of infectious diseases, 8: 235-244 ( 1987). 32. WHO Technical report series, N° 849, 1995 (Control of foodbrtrne trematode infections. P..eport of a WHO Study Group). OMS Serie de Rapports techniques N° 849, 1995 (Luttecontre /,es trematodoses d 'origine alimentaire. Rapport d 'un Grouped 'etude de l'OMS). 33. Levine, D.Z. Ciguatera: current concepts. [Review]. Journal of the American Osteopathic Association, 95: 193-8 ( 1995). 34. Lange, W.R. et al. Travel and ciguatera fish poisoning. Archiues of internal medicine, 152: 2049-53 ( 1992). 35. Steffen, R. Epidemiology of travellers' diarrhoea. Scandinauian journal of gastroenterol.ogy - SupplRment, 84: 5-17 (1983). 36. Kollaritseh, H. Traveller's diarrhea among Austrian tourists to warm climate countries: II. Clinical features. European journal of epidemiol.ogy, 5: 355-62 ( 1989). 109 37. Freedman, B.J. Travellers' diarrhoea: does it occur in the United Kingdom? journal of hygi.ene (Camlrridge), 79: 73-75 (1977). 38. Cartwright, R.Y. Epidemiology of travellers' diarrhoea in British package holiday tourists. PHLS microbiology digest, 9: 365-370 (1992). 110 39. Cartwright. R.Y. Eco,wmic aspects of travel health. In: Cook, G.C., (ed) Travel-associated disease. London, Royal College of Physicians, 1995. 40. Jacob, M. Safe food handling: a training guide for managers of food service establishments. Geneva, World Health Organization, 1989. Rapp. trimest. statist. sanit. mond., 50 (1997) The role of food in the epidemiology of cholera M. John Albert,a Maria Neirab & Yasmine Motarjemic Introduction Food serves as an important vehicle for most of the enteric infections. Although cholera has customar- ily been associated with water, recent epidemiolog- ical studies have shown that foodstuffs also play an important role in its transmission. In this report, we will review the role of foodstuffs in the epidemi- ology of cholera, although in most instances the roles of water and food cannot be dissociated. Cholera Cholera is an acute, watery diarrhoeal disease. In its severe form, cholera gravis, the disease is charac- terized by the passage of voluminous stools of rice water in appearance and vomiting that rapidly lead to dehydration. Hypovolemic shock, acidosis, and death can follow, if prompt and appropriate ther- apy is not initiated. Pandemics of cholera Since 1817, the world has been affected by a series of 7 pandemics of cholera. The causative agents of • Interim Director, Laboratory Sciences Division, International Centre for Diarrhoeal Disease Research, Dhaka, Bangladesh. b Medical officer, Division of Emerging and other Communicable Diseases Surveillance and Control, World Health Organization, Switzerland. c Scientist, Food Safety and Food Aid Programme, World Health Organization, Geneva, Switzerland. Fig. 1 the first 4 pandemics are not known since they occurred when culture facilities were not yet devel- oped, but those of the fifth and sixth pandemics were due to the classical biotype of Vibrio cholerae 01. The first 6 pandemics are thought to have originated from the Gangetic delta in Bengal. However, the causative agent of the present, sev- enth pandemic cholera, the El Tor biotype of V choleraeOl began in 1961 in Sulavesi, Indonesia. The seventh pandemic extended across Asia into the Middle East during the 1960s and through Africa, southern Europe, and the Pacific Islands during the 1970s ( 1). The seventh pandemic reached the South American continent when chol- era cases were reported in Peru in January 1991. This marked the return of cholera to this continent after a century of absence. Within a year, the major- ity of countries of the South and Central America were affected (2). In 1995, V choleraeOl biotype El Tor was reported from all regions of the world (Fig. I). In all, 208 755 cases and 5 034 deaths were officially reported by 78 countries to WHO. The American continent accounted for the largest pro- portion ofall reported cases worldwide in 1995 (3). By late 1992, epidemic cholera due to a new organism (V cholerae 0139 synonym Bengal) ap- peared in the Indian subcontinent (4). Since the first reports of V cholerae O 139 in India, the organ- ism has been isolated in 10 countries or areas in Asia: Bangladesh, China, Hong Kong, India, Number of cases of cholera reported to WHO, by continent and by year, 1984-1995 Nombre de cas de cholera signales a l'OMS, par continent et par annee, 1984-1995 "' ~ §_ SOO ~ u -2 400 "' .0 E ~ 300 I ;;, -0 ~ 200 6 :§ ~ 100 ~ 0 ai .0 E ~ z Wld hlth statist. quart., 50 (1997) 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 Year - Annee [2] ".frica - Afr,que O Americas - Ameriques 111 Malaysia, Myanmar, Nepal, Pakistan, Singapore and Thailand (5). When explosive outbreaks of cholera due to V. cho/,erae 0139 occurred in some Asian countries during 1992-93, it was suspected that V. cho/,erae 0139 might be the causative agent of the "eighth" pandemic. However, the strain has so far been confined to Asia (apart from a few imported cases in industrialized countries), and has not invaded new countries since 1993; more- over, it accounts for only a small proportion of cholera cases in countries where it has become endemic (6). Therefore, the fear that the organism may be the cause of a new pandemic cannot be substantiated since cholera cases due to V. cho/,erae 0139 appear to have declined. In 1995, only 1 country reported cases. Since August 1996 a resur- gence of Vibrio cho/,erae 0139 (Bengal) was reported in Calcutta (7). Etiological agents of cholera Vibrio which are gram-negative curved rods with single polar flagella are divided into more than 100 serogroups. Until recently, toxigenic strains belonging to serogroup O 1 were associated with epidemics of cholera, and the non-01 serogroups were associated with sporadic diarrhoeas and extraintestinal infections. However, after the emer- gence of V. cho/,erae 0139 in late 1992, both toxi- genic V. cho/,erae O 1 and O 139 serogroups are now considered as etiological agents of cholera (8). Strains of V. cho/,erae 01 are divided into two biotypes, classical and El Tor, and each biotype is further divided into two serotypes, Inaba and Ogawa. The two biotypes are differentiated by a number of tests, and the serotypes are differentiat- ed by slide agglutination with specific antisera. Se- rotype conversion in either direction is known to occur and its genetic basis has been established (9). Pathophysiology and clinical features After ingestion through water or food, V. cho/,erae attaches to the small intestinal mucosa. Although a variety of colonization factors have been impli- cated in the attachment process, toxin coregulated pilus (TCP) has been identified as the most critical adherence factor ( 10). During multiplication in the small bowel, a variety of secretogenic factors are released, the most important of which is chol- era toxin ( CT). CT attaches to the ganglioside GMl receptor on the enterocyte, gets translocated intracellularly and activates adenyl cyclase-C'AMP. This results in net secretion of water and electro- lytes into the gut lumen, which leads to severe diarrhoea, characteristic of cholera. The severe form of cholera, cholera gravis, is characterized by severe diarrhoea and vomiting which lead to rapid dehydration with the appear- ance of sunken eyes and poor skin turgor. Painful muscle cramps may occur, probably resulting from hypokalaemia. However, cholera has a wide clini- 112 cal spectrum: only a small minority of infected persons develop cholera gravis, about one-fourth develop mild to moderate diarrhoea and the vast majority are asymptomatic ( 11 ). Rehydration is the mainstay of therapy for cholera. However, anti- biotics may constitute an adjunct therapy; they shorten the volume and duration of diarrhoea and the duration of the excretion of the organism. Source of organism causing cholera Cholera is exclusively a human disease and faeces from cholera patients are the primary source of infection. Generally, the infective dose is high, ap- proximately 106 organisms. It is estimated that a cholera patient passing 10 litres of stool containing 106-109 organisms per ml of stool can produce a staggering 1013 organisms, sufficient to infect up to 10 million people. In addition, V. cho/,erae 01 also exists in environmental reservoirs, and two such reservoirs have been identified: one in rivers in north-eastern Australia (12) and the other in the United States Gulf Coast waters (13). Vehicles of transmission The primary vehicles of transmission of cholera are contaminated water and food. Waterborne trans- mission of cholera was first recognized in Europe in the mid nineteenth century when studies by John Snow and others showed cholera to be trans- mitted by sewage-contaminated municipal water. There is a potential for contamination of water if stored in wide-mouthed open containers in en- demic areas. Such containers will allow the entry of contaminated hands and other objects. A study in Calcutta, India found that introduction of narrow- mouthed water storage containers (into which hands and other objects cannot be introduced) resulted in a significant reduction in disease trans- mission (14 ). Contamination of food Food can be contaminated by cholera organisms in the following ways: (i) by the faeces of an ill person or an asymptomatic carrier who is handling and preparing food, (ii) when sewage-contaminated water is used for watering crops or for preparing food, and when seafood is harvested from areas contaminated by sewage, and (iii) by exposure of seafood to environmental reservoirs when the or- ganisms occur naturally independent of human faecal contamination (for example, contamination of seafood in the United States Gulf Coast waters). Factors influencing the survival and growth of V. cholerae 01 in food A number of intrinsic and extrinsic factors affect survival and growth of V. cho/,erae in food. Intrinsic factors include nutrients, pH (acidity), water activ- ity (~), red-ox potential, presence of antimicrobial Rapp. trimest. statist. sanit. mond., 50 (1997) agents or competing microbial flora. V. cholerae survives and grows over a pH range of 6-11, but optimally at pH 7 .6, at water activity above 0.97 and moderate amount of salt (0.25-3%). Drying, high osmotic pressure and sunlight reduce its survival. Also, V. cholerae grows better in foods with high carbohydrate content and in cooked foods where the competing flora has been killed (15, 16). Extrinsic factors which affect survival and growth are temperature of storage, humidity and the gaseous composition of the environment. Sur- vival is extended at refrigeration, and even more so at freezing temperatures, although growth is slowed or stopped. V. cholerae grows best at temper- ature 18-42 °C, but optimally at about 37 °C. On the other hand, the organism does not resist heat, and is sensitive to temperatures over 48 °C. At tem- peratures above 70 °C, it is killed in a matter of seconds. High humidity is favourable to survival, whereas gases used as preservatives e.g. C02 have adverse effects. Furthermore, in studies in which food has been deliberately contaminated, it has been observed that the El Tor biotype usually reaches higher numbers than the classical biotype, and in certain foods, the former survives better than the latter ( 17). Survival ofV. cholerae 01 in specific food items Seafood. The food that has been most frequently implicated in outbreaks is seafood, both mollus- can shellfish and crustaceans. Seafood may be contaminated if it is harvested from water pol- luted by sewage or from environments where V. cholerae O 1 occurs naturally independent of human faecal contamination. Since V. cholerae O 1 possesses chitinase, it attaches to chitinous surfac- es of zooplankton and crustaceans. Zooplankton is ingested by molluscs and crustaceans from sea- water, and the ingested V. cholerae O 1 organisms survive inside crabs and molluscs for several weeks (18). In artificially contaminated oysters and clams, V. cholerae 01 survives for more than 3 weeks when refrigerated ( 19). Crabs boiled for less than 10 minutes or steamed for less than 30 min- utes may still harbour viable V. cholerae O 1 organ- isms (20). Fruit and vegetables. Fresh fruit and vegetables may become contaminated if they come into con- tact with faecally contaminated soil, or contaminat- ed water is used for cleaning or injected into fruits to increase weight and improve turgor (21). Surviv- al of vibrios is shorter in acidic fruits (pH <4.5). Milk and milk products. Standard pasteurization will kill vibrios. However, contaminated milk and milk-products will permit the survival of V. cholerae 01 for as long as 2 weeks. In contaminated milk, vibrios may survive for as long as 4 weeks if refriger- ated (22). Wld hlth statist. quart., 50 (1997) Meat. Warm-blooded animals are free from V. cholerae O 1 infection, but their meat may be- come contaminated while processing. V. cholerae 01 survives well in raw and cooked meat. Other foods. V. cholerae 01 grows well in some foods such as hard boiled egg, cooked rice and pasta, and cooked red lentils. Extremely high levels of organisms are achieved rapidly if some foods are stored at 37 °C. V. cholerae 01 can survive for sever- al days even in some spices such as cinnamon and pepper (23). Detection and characterization ofV. cholerae from foods As for isolation of V. cholerae from stool, enrich- ment in alkaline peptone water (APW) is recom- mended for isolation from contaminated foods. Usually about 25 g of contaminated food is blend- ed and mixed with 225 ml of APW. One set of broth is incubated at 37 °C and another at 42 °C for 6-8 h. Since vegetable surfaces have high con- centrations of enteric bacteria, blended samples are diluted, and enriched in APW at the two tem- peratures as above. The broth is then subcultured onto a selective medium, such as thiosulfate- citrate-bile salt-sucrose (TCBS) agar and incubated at 37 °C for 18-24 h. Typical yellow colonies resem- bling V. cholerae are subcultured onto a non-selec- tive medium, and the growth from the medium used for oxidase test. If the oxidase test is positive, slide agglutination of the colonies can be per- formed first with V. cholerae 01 polyvalent antiser- um, and if positive, then with Inaba - and Ogawa - serotype specific antisera. If agglutination is nega- tive with V. choleraeOl antiserum, V. cholerae0139 can be suspected, which can be confirmed by ag- glutination with antiserum specific for V. cholerae 0139. - Several screening media are useful for pre- sumptive identification of V. cholerae. Arginine- containing media are used for differentiation of V. cholerae from arginine dihydrolase-positive or- ganisms. V. cholerae requires sodium ions for growth, and most commercially available media will contain this critical component in the required concentration. In contrast, other vibrio species will require higher concentration of NaCl than what is present in the commercial media for optimal growth. Thus, a medium, such as tryptophan broth without added NaCl ( 1 % tryptophan, 0% NaCl [T1N0]) can be used to screen for V. cholerae (24). Since only cholera toxin (CT)-producing V. cholerae 01 and 0139 are capable of causing epidemics, the isolates from food samples should be screened for CT or the gene ctx encoding the toxin. CT can be detected by tissue culture assay using Yl adrenal tumour cells or Chinese hamster ovary cells, which show distinct morphological changes on exposure to CT (25-26). A GMl gan- 113 glioside enzyme-linked immunosorbent assay or coagglutination test can also be used for detection of CT in culture supernatants (27). Colonies can be hybridized with both cloned DNA fragments or synthetic oligonucleotide probes to detect ctx gene (28-29). A polymerase chain reaction (PCR) assay is also available which uses primers complementary to certain regions of ctxoperon (30). Molecular typing o!V. cholerae Phenotypic characteristics are not sufficient for dis- crimination of strains. However, several molecular techniques that are sufficiently discriminatory have been developed, and some of them have been suc- cessfully employed to trace foodborne outbreaks. The molecular techniques that have been devel- oped for differentiation of strains include restric- tion fragment length polymorphism (RFLP) of CT gene (ctx) (ctx genotyping) and genes encoding rRNA (ribotyping), sequence divergence within the structural gene for the CT B subunit, multilo- cus enzyme electrophoresis (MEE) based on varia- tions in the mobility of several enzymes, and pulsed field gel electrophoresis (PFGE) which separates large DNA fragments created by digestion of total genomic DNA with restriction endonucleases that cut DNA infrequently. By these techniques, it has been shown that the strains of cholera associated with the seventh pandemic, the Latin American epidemic, the United States Gulf coast waters and the rivers of north-eastern Australia are distinctly different (31-32). Epidemiology of foodborne outbreaks Various foods have been implicated in cholera out- breaks over the past 30 years and have been identi- fied by case-con~ol studies. There has been a plethora of cholera outbreaks associated with con- taminated food; however, all these outbreaks will not be described here; their descriptions can be found elsewhere (33-35). Only some outbreaks as- sociated with selected food items will be described here. Seafood. Seafood is the most commonly impli- cated food item associated with outbreaks. It in- cludes molluscan shellfish, crustaceans and finfish. In 1969, in Malaysia, mussels were implicated in an outbreak involving 135 people (36). Raw cockles imported from a North African country were impli- cated in an outbreak involving 2 400 people in Portugal (37). Home boiled crabs contaminated with V chol.erae 01 caused two separate outbreaks involving 11 people in Louisiana in 1978 ( 16). Crabs and shrimps harvested from widely scattered sites in Louisiana caused an outbreak involving 18 people (38). An outbreak affecting 37 construc- tion workers occurred in Singapore in 1982. The outbreak was traced to contaminated squid pre- pared by foodhandlers who were found to be in- 114 fected with V chol.eraeOl (39). There are reports of spread of cholera in Latin America due to con- sumption of ceuiche, a cocktail of raw fish or shell- fish marinated in lemon juice (40). Although lemon juice is acidic, which kills V chol.eraeOl, the organisms might have survived due to improper marination. The fish might have been contami- nated naturally in the estuary from where they were harvested or during preparation. On two occasions crabs transported from Ecuador have been associ- ated with outbreaks in the United States (41-42). On both occasions, the crabs were purchased and boiled in Ecuador, brought to the United States in personal baggage and served cold at small parties more than 24 h later. V. chol.erae 01 might have survived initial boiling or might have been subse- quently introduced during handling. In 1992, cold seafood salad prepared in Lima, Peru and served to passengers on board a flight from Buenos Aires, Argentina through Lima to Los Angeles, California was incriminated in an outbreak affecting over 75 airline passengers (43). Shellfish beds can be contaminated by ballast and bilge water from arriving ships. This was thought to be the case when the Latin American strain of V chol.erae O 1 was isolated from a com- mercial shellfish bed in Alabama, United States. Samples of ballast, bilge and sewage from several ships arriving from Latin American ports con- tained the same toxigenic strain of V chol.erae 01 (44). Flushing of ballast and bilge water in the open ocean before entering ports should avoid contami- nation of seafood beds. Fruits and vegetables. An outbreak in Israel in 1970 was attributed to raw vegetables irrigated with inadequately treated wastewater. V chol.erae O 1 was isolated from irrigation water, but an epidemiolog- ical study implicating raw vegetables was not con- ducted (45). In Chile, vegetables watered with sew- age-contaminated river water was thought to have caused an outbreak of cholera (46). In Peru, con- sumption of unwashed fruits and vegetables, and cabbage harvested from field where raw sewage was used for irrigation, and which was cooked only briefly were associated with illness (2). Other foods. In November 1972, cholera infec- tion occurred among 47 of 331 passengers on board an aircraft travelling from London to Syd- ney. The aircraft made a stop in Bahrain and load- ed food. Bahrain was experiencing an outbreak of cholera at that time. The implicated food was an hors d'oeuvre served in the aircraft between Bahr- ain and Sydney (47). Two instances of cholera transmission by meat have been implicated in Thailand. These involved meat that was held for hours at ambient tempera- ture before being served raw (48-49). Cooked grains and legumes support the growth of V chol.erae O 1 and they have been implicated in Rapp. trimest. statist. sanit. mond., 50 (1997) the spread of epidemic cholera in Africa. Typically, the foods are prepared, held at ambient tempera- ture for a long time and often eaten without re- heating. In Mali, consumption of leftover steamed millet gruel was associated with cholera; however, adding curdled goat's milk (an acidic food) to gruel gave protection against cholera (50). An in- vestigation in Guinea defined the role of sauces served with cooked grains. In a case-control study in Conakry, Guinea, eating leftover unheated rice with acidic tomato sauce was protective, but with less acidic peanut sauce was associated with illness (51). Investigation of two cholera outbreaks associ- ated with funerals for cholera victims in Guinea showed that they were caused by rice served at funerals. The rice had been prepared by women after they had cleaned the deceased body and evac- uated the bowel contents and they would have contaminated the rice with their hands. The rice remained at ambient temperature for several hours before being eaten, allowing V. cholerae O 1 to multiply (51). Cooked rice was also responsible for an outbreak involving 16 people in a Gulf Coast oil rig in the United States. Cooked rice was moist- ened with water contaminated by human faeces and then held warm for several hours before served (52). The first commercially imported product known to have caused a cholera outbreak occurred in Maryland, United States. The product was frozen coconut milk imported from Thailand. Four persons became infected after eating a rice pudding dessert topped with frozen coconut milk. Toxigenic V. cholerae 01 was isolated from an im- ported package of the same brand of frozen coco- nut milk. Investigations revealed that contamina- tion occurred during preparation of the food in the country of origin (53). Foodborne outbreak of cholera due to V. cholerae 0139 Bengal There is only one outbreak of diarrhoea reported so far due to the newly emerged second etiological agent of cholera, V. cholerae0139. This occurred in February 1994 among 6 of 630 passengers of a cruise ship visiting countries of the South-East Asia. The infection was acquired in Thailand and was associated with eating yellow rice at a buffet restau- rant in Bangkok. Investigations did not find any deficiencies in sanitation or food preparation tech- niques. However, foodhandlers were not tested for evidence of recent infection with V. cholerae (54). Risk of cholera transmission by commercially imported foods Food items are exported from many countries where cholera is endemic. Surprisingly, there are relatively few documented outbreaks associated with imported foods (53). There are, however, sev- Wld hlth statist. quart., 50 (1997) eral outbreaks related to foods imported by travel- lers themselves (55, 56). Studies have shown that imposition of restrictions on importation of food from countries where cholera is endemic is not warranted. The World Health Organization (WHO) has issued guidelines relating to the risk of cholera transmission by food imports (57, d ). Ac- cordingly, no specific restrictions are called for where the physical or chemical characteristics or the processing of the food are such that V. cholerae 01 is unlikely to be present, even if earlier contam- ination of the food cannot be excluded. Examples of such food are acid foods (pH< 4.5), irradiated foods (minimum dose 1 kGy), foods that have been subjected to heating processes, e.g. cooking, pas- teurization, sterilization/ canning, and foods with low water activity, i.e., dry food such as dried vege- tables, coffee beans, cereals, pulses, dried milk. However, in cholera-affected areas, certain foods may present a risk of contamination. These are: seafoods which may have been harvested from or kept in contaminated water; raw shellfish and un- cooked crustaceans; and fruits and vegetables that grow close to the soil and have been irrigated with faecally-contaminated water or top-dressed with untreated night-soil. Prevention Measures for the prevention of foodborne and waterborne-cholera should be based on minimiz- ing three factors: i.e. contamination, survival, and growth. Contamination. Good sanitation, including ap- propriate facilities for human waste disposal and hand washing, are important for the prevention of contamination of the environment and thus the food and water supplies. In addition to provision of facilities, people should also be taught how to use latrines, about the dangers of defecating on open ground, or in or near water, and the importance of washing their hands with soap, particularly after any contact with excreta and before handling and preparing foods. The food and water supplies also need to be protected as far as possible from possi- ble sources of contamination, e.g. irrigation with untreated sewage water. Survival. Recognizing that in endemic countries V. cholerae is inevitably present in the environment and may contaminate food and drinking-water sources, measures to reduce survival of the organ- ism should also be considered when planning strat- egies for prevention of cholera. Disinfecting water with chlorine or boiling it are effective methods to render drinking-water safe. Proper heat treatment of food will kill the organism. Therefore, the pub- d WHO guidance on farmulation of national policy on the control of chol,era. (Document \.\'HO/CDO/SER/92.16 REV), 1996. 115 lie should be advised to cook their food thorough- ly. Other methods to destroy or remove the organ- ism from food are acidification, irradiation, and sterilization. Growth. As V. chol.erae generally has a high infec- tive dose, limiting growth in food to below disease- causing levels is an important measure to prevent cholera. Opportunities for growth can be avoided by eating prepared food as soon as it is ready. Foods that are prepared in advance should be kept at temperatures below 10 °C or kept hot at a tem- perature above 60 °C, and should be reheated before use. When hot or cold storage is not feas- ible, acidification and fermentation can be consid- ered as a means of slowing down the growth (58, e, f). Epidemiological studies have shown that most cases of foodborne cholera are associated with un- hygienic handling of food in the home or by street food vendors. Therefore, educating food handlers and street food vendors in the hygienic handling of food is a key measure to prevent cholera. Since sociocultural and environmental conditions vary all over the world, more efficient health education in food safety can be achieved through the applica- tion of the Hazard Analysis Critical Control Point (HACCP) system to food preparation. The applica- tion of this system will allow identification of be- haviours which are specific to the local food and food preparation practices, and which are critical for the safety of the food ( 5 9, g ). Conclusions Food, an important vehicle of cholera transmis- sion, can be contaminated in its natural habitat or during preparation. Therefore, attention must be given to the hygienic preparation of food and its consumption. The vehicles of transmission and thus the possible preventive measures can vary markedly from place to place, being influenced by local customs and practices as well as local condi- tions. For example, in Mali, adding fermented goat's milk to millet gruel can be an important preventive measure, while in the Amazon region of Brazil, adding toronja juice to drinking water de- creases the risk of cholera. Although general rec- ommendations can be made about the prevention and control of cholera, it is important to continue to examine the mode and vehicle of transmission e FAO/WHO. Fermentation: assessment and research. Repart of a FAO/WHO Workshop on fermentation as a househo/,d technowgy to improve food safety (Pretoria, South Africa, 11-15 December 1995.) (Document WHO/FNU/FOS 96.1), 1996. t Motarjemi, Y. et al. Food technowgies and public health. (Document \,VHO/FNU/FOS 95.12), 1995. g Application of Hazard Analysis Critical Control Point system far the improvement of food safety. Vv'HO-supparted case studies on food prepared in homes, at street vending operations, and in cottage industries. (Document "\NHO/FNU/FOS/ 93.1 ), 1993. 116 of the disease and to identify protective factors, so that most appropriate control measures that are important locally can be implemented. Summary Cholera is an acute dehydrating diarrhoeal disease, traditionally caused by Vibrio cholerae 01, and also more recently by V. cholerae 0139 (Bengal). Tradition- ally, water was recognized as the primary vehicle for transmission of cholera, but in the past 30 years, out- breaks of cholera associated with eating contaminated food have demonstrated that food also plays an impor- tant role, although in many instances water is the source of contamination of foods. Most commonly associated with cholera is seafood, both molluscan shellfish and crustaceans. Seafood may be contaminated in its natu- ral environment or during preparation. Other food items associated with outbreaks are fruit and vegetables, meat, cooked grains, etc. Vegetables are usually con- taminated by contact with sewage in soil and fruits when injected with contaminated water to increase weight and turgor. Food items initially free from V. cholerae organ- ism may become contaminated when mixed with water, or other contaminated food, or through handling by infected persons who have not observed proper hy- giene. Refrigeration, freezing, alkaline pH, high concen- tration of carbohydrate, humidity and absence of competing flora enhance the survival of V. cholerae in food. Survival of V. cholerae is shorter in food with acidic pH. Foodborne cholera can be averted by the hygienic preparation of food and its consumption. However, since the vehicles of transmission vary markedly from place to place, being affected by local customs and practices, selected control and preventive measures that are most important locally must be implemented. To this end, application of the Hazard Analysis and Critical Control Point system to food preparation is essential in order to identify the practices which may present a risk. Restrictions on importation of foods which do not present a risk of being contaminated from areas where cholera is endemic is not warranted. Resume Role des aliments dans l'epidemiologie du cholera Le cholera est une maladie diarrheique entraTnant une deshydratation aigue, classiquement due a Vibrio cho- lerae 01 et, plus recemment, a V. cholerae 0139 (Ben- gal). L'eau est traditionnellement consideree comme le principal vehicule de transmission du cholera mais, au cours des 30 dernieres annees, des flambees asso- ciees a la consommation d'aliments contamines ont demontre que ceux-ci jouent aussi un rOle important, meme si dans de nombreux cas ils ont ete contamines par de l'eau. L'association la plus frequente avec le cholera concerne les fruits de mer - mollusques et crustaces - qui peuvent etre contamines dans leur milieu naturel ou pendant la preparation. D'autres ali- Rapp. trimest. statist. sanit. mond., 50 (1997) ments sont associes a des flambees, comme les fruits et legumes, la viande, les cereales cuites, etc. Les legu- mes sont en general contamines par contact avec les eaux d'egout dans le sol, et les fruits lorsqu'on leur injecte de l'eau pour augmenter leur poids et ameliorer leur aspect. Des denrees initialement saines peuvent ~tre contaminees lorsqu'elles sont melangees a de l'eau au a des denrees contaminees, au par des manipula- teurs infectes n'ayant pas respecte les regles d'hygiene concernant le lavage des mains. Certains facteurs com- me la refrigeration, la congelation, un pH alcalin, une forte teneur en glucides, l'humidite et !'absence de flare concurrente favorisent la survie de V. cholerae dans les aliments. Le vibrion survit mains longtemps dans les aliments de pH acide. II est possible d'eviter le cholera transmis par les aliments en respectant les regles d'hy- giene lors de la preparation et de la consommation des aliments. Toutefois, comme les vehicules de transmis- sion varient sensiblement d'un endroit a l'autre en fonc- tion des coutumes et pratiques locales, ii taut mettre en place des mesures de prevention et de lutte selectives adaptees a la situation locale. II est pour cela indispen- sable d'appliquer le systeme HACCP d'analyse des risques - points critiques pour leur maTtrise afin d'iden- tifier les pratiques qui peuvent comporter un risque. Les restrictions a !'importation, depuis des regions au le cholera est endemique, de denrees alimentaires pour lesquelles ii n'existe pas de risque de contamination ne se justifient pas. References - References 1. Barua, D. History of cholera. In: Barua, D. & Greenough, W.B. (eds), Chol.era, New York, Plenum, 1992, pp 1-36. 2. Swerdlow, D.L. et al. Waterborne transmission of epidemic cholera in Trujillo, Peru: lessons for a continent at risk. Lancet, 340(8810): 28-32 (1992). 3. Cholera in 1995. Weeklyepidemiologicalrecord, 71(21): 157-164 (1996). Le cholera en 1995. &l.eue epidimiologique hebdomadaire, 71 (21): 157-164 (1996). 4. Albert, M.J. et al. Large epidemic of cholera-like disease in Bangladesh caused by Vibrio chol.erae O 139 synonym Bengal. Lancet, 342(8868): 387-390 (1993). 5. Cholera in 1994. Weekly epidemiological record, 70(28): 201-208 (1995). Le cholera en 1994. &l.eue epidimiologique hebdomadaire, 70(28): 201-208 (1995). 6. Albert, MJ. Epidemiology and molecular biology of Vibrio chol.erae 0139 Bengal. Indian journal of medical research, I04(July): 14-27 (1996). 7. Rupak, Mitra, et al. Resurgence of Vibrio chol.erae (Bengal) with altered antibiogram in Calcutta, India. [Article accepted for publication by Lancet]. 8. Albert, MJ. Vibrio chol.erae 0139 Bengal. Journal of clinical microbiology, 32(10): 2345-2349 (1994). 9. Stroeher, U.H. et al. Serotype conversion in Vibrio chol.erae 01. Proceedings of the U.S. NationalAcademy of Sciences, 89 ( 10): 2566-2570 (1992). 10. Taylor, R.K. etal. Use of PhoA gene fusions to identify a pilus colonization factor coordinately regulated with cholera toxin. Proceedings of the U.S. National Academy of Sciences, 84(9): 2833-2837 (1987). 11. Gangarosa, EJ. &: Mosley, W.H. Epidemiology and surveillance of cholera. In: Barua D. & Burrows W.(eds), Chol.era, Philadelphia, W.B. Saunders Co., 1974, pp. 381-403. Wld hlth statist. quart., 50 (1997) 12. Bourke, A. T.C. et al. Investigation of cholera acquired from the riverine environment in Queensland. Medical journal of Australia, 144(5): 229-234 (1986). 13. Kaper,J.B. et al. Molecular epidemiology of Vibrio chol.eraein the U.S. GulfCoast.Journalofclinicalmicrobiology, 16(1): 129- 134 (1982). 14. Deb, B.C. et al. Studies on intervention to prevent El Tor cholera transmission in urban slums. Bulletin of the World Health Organization, 64(1): 127-131 (1986). Deb, B.C. et al. Etudes sur les mesures destinees a prevenir la transmission du cholera El Tor dans les taudis urbains [resume]. Bulletin de l'Organisation mondial.e de la Sante, 64 ( 1): 130-131 (1986). 15. Roberts, D. Growth and survival of Vibrio chol.erae in foods. Microbiology digest, 9(1): 24-31 (1992). 16. Quevedo, F. Foods and cholera. In: Pestana de Castro & WF Almeida (eds), Chol.era on the American Continents. Washington, DC, ILSI Press, 1993, pp 71-86. 17. Kolvin,J.L. &: Roberts, D. Studies on the growth of Vibrio chol.erae biotype El Tor and biotype classical in foods. journal of hygiene (Cambridge), 89(2): 243-252 (1982). 18. Colwell, R.R. &: Spiral, W.M. The ecology of Vibrio chol.erae. In: Barua, D. & Greenough, W.B. (eds), Chol.era, New York, Plenum, 1992, pp.107-127. 19. DePaola, A. Vibrio chol.erae in marine foods and environmental waters: literature review.Journal of foodscienc.e, 46(1): 66-70 (1981). 20. Blake, P.A. et al. Cholera - a possible endemic focus in the United States. Nf!W England1ournal of medicine, 302(6): 305- 309 (1980). 21. Feachem, R.G. Environmental aspects of cholera epidemiology I. A review of selected reports of endemic and epidemic situations during 1961-1980. Tropical diseases bulletin, 78(8): 675-698 (1981 ). 22. Felsenfield, 0. Notes on food, beverages and fomites contaminated with Vibrio cholerae. Bulletin of the World Health Organization, 33(5): 725-734 (1965). 23. Cholera in the Americas. Bulletin of the Pan American Health Organization, 25(3): 267-273 (1991). 24. U.S. Food and Drug Administration bacteriological analytical manual. Arlington, Virginia, Association of Offi- cial Analytical Chemists, 1991. 25. Guerrant, R.L. et al. Cyclic adenosine monophosphate and alteration of Chinese hamster ovary cell morphology: a rapid, sensitive in vitro assay for the enterotoxins of Vibrio choleraeand Escherichia coli. Infection and immunity, 10 (2): 320- 327 (1974). 26. Sack, D.A. &: Sack, R.B. Test for enterotoxigenic Escherichia coli using YI adrenal cells in miniculture. Infection and immunity, 11(2): 334-336 (1975). 27. Almeida, R.J. et al. Comparison of latex agglutination assay and an enzyme-linked immunosorbentassay for detecting chol- era toxin.Journal of clinical mi<:robiology, 28(1): 128-130 (1990). 28. Kaper,J.B. et al. DNA probes for pathogenic Vibrio species. In: Tenover, F.C. ( ed), DNA ,m-obes for infectious diseases, Boca Raton, Florida, CRC Press, 1989, pp 65-77. 29. Wright, A.C. et al. Development and testing of a nonradioactive DNA oligonucleotide probe that is specific for Vibrio chol.erae cholera taxi n. Journal of clinical microbiology, 30(9): 2302-2306 (1992). 30. Shirai, H. et al. Polymerase chain reaction for detection of the cholera enterotoxin operon of Vibrio chol.erae. journal of clinical microbiology, 31 (11 ): 3068-3070 (1993). 31. Wachsmuth, K. etal. Molecular epidemiology of cholera. In: Wachsmuth, I.K. et al (eds), Vibrio chol.erae and chol.era: molecular to global perspectives, Washington, DC, American Society for Microbiology, 1994, pp 357-370. 32. Cameron, D.N. et al. Molecular characterization of Vibrio chol.eraeO 1 strains by pulsed-field gel electrophoresis. journal of clinical microbiology, 32(7): 1685-1690 (1994). 33. Popovic, T. et al. Cholera in the Americas: foodborne aspects.journal of food protection, 56(9): 811-821 (1993). 117 34. Mintz, E.D. et al. Transmission of Vibrio cho1£rtU 01. In: Wachsmuth, I.K. et al (eds). Vibrio chol£rtU and cho1£ra: mo/,ecular to gwbal perspectives, Washington, DC, American Society for Microbiology, 1994, pp 345-355. 35. Kaysner, C.A. & Hill, W.E. Toxigenic Vibrio chol£rtU 01 in food and waterln: Wachsmuth, I.K.etal. (eds), Vibriocho1£rae and chol£ra: mo/,ecular to gwbal perspectives. Washington, DC, American Society for Microbiology, 1994, pp 27-39. 36. Dutt, A.K. et al. A shellfish-borne cholera outbreak in Malaysia. Transactions of the Ruyal Society for Tropical Medicine and Hygiene, 65(6): 815-816 (1971). 37. Blake, P.A. et al. Cholera in Portugal. I. Modes of transmission. American journal of epidemiowgy, 105( 4): 337- 343 (1977). 38. Lowry, P.W. etal. Cholera in Louisiana-widening spectrum of seafood vehicles. Archives of internal medicine, 149 (9): 2079- 2084 (1989). 39. Goh, K. T. et al. A common source foodborne outbreak of cholera in Singapore. International journal of epidemiowgy, 13(2): 210-215 (1984). 40. Weber,J.T. et al. Epidemic cholera in Ecuador: multidrug resistance and transmission by water and seafood. Epidemiowgy and infection, 112(1 ):1-11 (1994). 41. Finelli, L. et al. Outbreak of cholera associated with crab brought from an area with epidemic disease. Journal of infectious diseases, 166(6): 1433-1435 (1992). 42. Centers for Disease Control. Cholera - New York. Morbidity and mortality weekly report, 40(30): 516-518 (1991). 43. Eberhart-Phillips,J. et al. An outbreak of cholera from food served on an international aircraft. Epidemiology and infection, 116(1): 9-13 (1996). 44. McCarthy, S.A. et al. Toxigenic Vibrio cho1£rae01 and cargo ships entering Gulf of Mexico. Lancet, 339(8793): 624-625 (1992). 45. Cohen, J. et al. Epidemiological aspects of cholera El Tor outbreak in a non-endemic area. Lancet, ii(7715): 86-89 (1971). 46. Cholera in the Americas. Weekly epidemiological record, 67(6): 33-38 (1992). 118 Le cholera dans les Arneriques. Rekve epiderniowgique hebdomadaire, 67(6): 33-38 (1992). 4 7. Sutton, R.G. An outbreak of cholera in Australia due to food served in flight on an international aircraft. Journal of hygiene (Cambridge), 72(3): 441-451 (1974). 48. Swaddiwudhipong, W. et al. A cholera outbreak associated with eating uncooked pork in Thailand. journal of diarrhoeal disease research, 8(3): 94-96 (1990). 49. Swaddiwudhipong, W. et al. A common source foodbome outbreak of El Tor cholera following the consumption of uncooked beef. Journal of Medical Association of Thailand, 75(7): 413-417 (1992). 50. Tauxe, R. V. et al. Epidemic cholera in Mali: high mortality and multiple routes of transmission in a famine area. Epidemiowgy and infection, 100(2): 279-289 (1988). 51. St. Louis, M.E. et al. Epidemic cholera in West Africa: the role of food handling and high-risk foods. American journal of epidemiology, 131(4): 719-727 (1990). 52. Johnston, J.M. et al. Cholera in a Gulf Coast oil rig. New England journal of ,nedicine, 309(9): 523-526 ( 1983). 53. Taylor, J.L. et al. An outbreak of cholera in Maryland associated with imported commercial frozen fresh coconut milk. journal of infectious diseases, 167 ( 6): 1330-1335 ( 1993). 54. Boyce, T.G. et al. Vibrio chol£rtU 0139 Bengal infections among tourists to Southeast Asia: An Intercontinental foodbome outbreak. journal of infectious diseases, 172(5): 1401-404 (1995). 55. Roman, D. et al. Cholera: New York. Morbidity and mortality weekly report, 1991, 40(30): 516-518. 56. Cholera associated with food transported from El Salvador· Indiana, 1994. Morbidity and mortality weekly report, 1995, 44(20): 385-386 57. Cholera: small risk of cholera transmission by food imports. Weekly epidemiowgical record, 66(8): 55-56 (1991). Cholera : faible risque de transmission du cholera par des aliments importes. Rekve epidemiowgique hebdomadaire, 66(8): 55-56 (1991). 58 World Health Organization. Guidelines for cho1£ra control Geneva, WHO, 1993. 59. Bryan, F. Hazard Analysis Critical Control Point evaluations. A guide to identifying hazards and assessing risks associated with food preparation and storage. Geneva, World Health Organization 1992. Rapp. trimest. statist. sanit. mond., 50 (1997) Establishment of microbiological safety criteria for foods in international trade International Commission on Microbiological Specifications for Foodsa Introduction Most foodborne diseases can be prevented by the application of the basic principles of food hygiene throughout the food chain, from the primary pro- ducer to the consumer. The responsibility of pro- ducing, manufacturing and preparing safe food is, of course, in the hands of people engaged in these activities. However, governmental agencies also have an important role to play in ensuring the safety of food (i.e. assuring that food will not cause harm to the consumer when it is prepared and/ or eaten according to its intended use). b Traditionally, three principal means have been used by governments to control microbiological hazards of foods: (i) educa- tion and training, (ii) inspection of facilities and operations, and (iii)microbiological testing. Educa- tion and training of food handlers and consumers help them to understand and apply safe food-han- dling practices. Inspection of facilities is intended to provide evidence that operations were designed and practices are applied to ensure that safe food is consistently delivered, i.e. that the Hazard Analysis Critical Control Point (HACCP) concept was ap- plied, and the results of microbiological testing should confirm this. Foods may be tested for micro- organisms that are related to the safety and/ or the quality of food. Safety is related to the presence or absence of pathogenic microorganisms or their tox- ins, the number of pathogens present, and the ease with which these agents can be destroyed or prevent- ed from multiplying. Microbiological quality is relat- ed to the presence or absence of microorganisms capable of causing spoilage or that are indicative of lack of Good Manufacturing Practices (GMP), and to the effectiveness of measures to control or destroy such organisms. Microbiological criteria, used in testing, should distinguish between an acceptable and unacceptable product, or between acceptable and unacceptable food processing and handling practices. With regard to control of imported food, em- phasis has been placed on microbiological testing of incoming consignments, and the International • Submitted on behalf of the International Commission on Microbiological Specifications for Foods by Dr. M. van Schothorst, Secretary, Vevey, Switzerland. b Codex Alimentarius Commission, Joint FAO/WHO Food Standards Programme, &port of the Twenty-Eighth Session of the CodexCommitteeonFoodHygiene, (DocumentALINORM97/13, Appendix II), 1995. W/d hlth statist. quart., 50 (1997) Commission on Microbiological Specifications for Foods (ICMSF) provided guidance on this subject from as early as 197 4 (] ). This situation is now changing, partly as a consequence of the General Agreement on Tariffs and Trade (CATT) Uruguay Round Agreement on the Application of Sanitary and Phytosanitary (SPS) Measures. The implemen- tation of this Agreement is intended to facilitate the free movement of foods across borders, by en- suring that means established by countries to pro- tect human health are scientifically justified, and are not used as non-tariff barriers to trade in food- stuffs (2). In principle, this should result in a reduc- tion in microbiological testing which is frequently unjustified scientifically, since it may not provide conclusive evidence that food satety measures were applied by the producer in the country of origin. The Agreement states furthermore that SPS mea- sures based on appropriate standards, codes and guidelines developed by the Codex Alimentarius Commission are deemed to be necessary to protect human health and consistent with the relevant CATT provisions. Thus, it is very important to understand how Codex approaches the subject of microbiological criteria. The World Trade Organization (WTO) Agree- ment also states that risk assessment should be used to provide the scientific basis for national food regulations on food safety and SPS measures, by taking into account risk-assessment techniques de- veloped by international organizations. At present there is no internationally agreed procedure for carrying out microbiological risk assessment, although a working group of the Codex Commit- tee of Food Hygiene has drafted some guidance on the principles of such an assessmenU One procedure that may be applied as a result of the SPS Agreement is the use of a scientifically- based microbiological criterion as a measure of the acceptability of a food in international trade. The development and application of such a criterion, as it might be applied within the SPS Agreement, is described in an Annex<l to the Codex Alimentarius c Codex Alimentarius Commission, Joint FAO/WHO Food Standards Programme, &port of the Twenty-Ninth Session of the Codex Committee on Food Hygime, (Document ALINOR.i\,f 97 I 13A, Appendix IV), 1996. d Codex Alimentarius Commission, Joint FAO/WHO Food Standards Programme, &port of the Twenty-Ninth Session of the Codex Committee on Food Hygi,me, (Document ALINORM 97 / 13A, Appendix Ill), 1996. 119 Code on "The General Principles of Food Hygiene".b The latter document describes the pro- cedures generally applicable to the hygienic manu- facture of foods and recommends the application ofHACCP as the preferred procedure for ensuring the microbiological safety of foodstuffs. A descrip- tion of principles and use of HACCP is included in another Annexe to this code. A microbiological criterion is intended to pro- vide assurance that a food is safe, when no more effective means of providing such assurance are available, e.g. where there is no knowledge that good hygienic practice and HACCP have been properly applied.<l Management al micrabia/agical hazards tar loads in international trade Within the Codex Alimentarius Commission a sys- tem is under development to ensure the proper management of the microbiological hazards of food in international trade under the SPS Agree- ment. In line with this system, the ICMSF has rec- ommended, in a document submitted to the Co- dex Alimentarius Food Hygiene Committee, that the following 5 steps should be applied. Conduct a risk assessment Apply the "Principles and Guidelines for the Appli- cation of Microbiological Risk Assessment" .c • The risk assessment and consequential risk management decisions provide a basis for de- termining the need to establish microbiological safety objectives. Establish food safety objectives A microbiological food safety objective is a statement of the maximum level of a microbiological hazard considered acceptable for consumer protection. • Each food safety objective should be technically and administratively feasible. • Statements should include quantitative values. • The food safety objectives should be developed by governmental bodies with a view to reaching consensus in relation to a food in international trade. The food safety objectives should be achievable throughout the food chain by application of: The "General principles of food hygiene"b or product-specific codes and the use of the HACCP system and guidelines for its application".e c Codex Alimentarius Commission, Joint FAO/\\'HO Food Standards Programme, Report of the Twenty-Ninth Session of the Codex Committee on Food Hygiene, (Document ALINORM 97 I 13A, Appendix II), 1996. 120 • It should be confirmed that the food safety ob- jectives are achievable. • The HACCP requirements must be developed by the food industry. Establish microbiological criteria, when appropriate Apply the "Principles for the Establishment and Application of Microbiological Criteria for Foods"<l. • This step should be performed by an expert group of food microbiologists, including ex- perts with experience in the food industry. Establish acceptance procedures for the food at port of entry Acceptance could be based upon for example: • A list of approved suppliers as determined by inspection of facilities and operations, certifica- tionf etc.; microbiological testing (if criteria have been established) and/ or other testing (pH (acidity), aw (water activity), container in- tegrity). Establishment al micrabialagical criteria According to Codex, a microbiological criterion for food defines the acceptability of a product or a food lot, based on the absence or presence, or number of microorganisms, and/or quantity of their toxins/metabolites, per unit(s) of mass, vol- ume, area or lot.<l It consists of: • a statement of the microorganisms of con- cern and/or their toxins/metabolites and the reason for that concern in the product; • the analytical methods for their detection and/ or quantification; • a plan defining the number of field samples to be taken and the size of the analytical sample unit; • microbiological limits considered appropriate to the food at the specified point(s) of the food chain; and • the number of analytical sample units that should conform to these limits. It states further that, to fulfil the purposes of a microbiological criterion, consideration should be given to: • the evidence of actual or potential hazards to health; • the microbiological status of the raw materials; • the effect of processing on the microbiological status of the food; r Codex Alimentarius Commission, Joint FAO/WHO Food Standards Programme, Report of the Fourth Session of the Codex Committee on Food lmpmt and Expmt Inspection and Certification Systems, (Document ALINORM 97 /30), 1996. Rapp. trimest. statist. sanit. mond., 50 (1997) • the likelihood and consequences of microbio- logical contamination and/ or growth during subsequent handling, storage and use; • the category(s) of consumers concerned; • the cost/benefit ratio associated with the appli- cation of the criterion; and • the intended use of the food. The targeted consumer group should also be considered. Some groups of consumers such as the very young, the very old, the diseased and immuno- compromised individuals may be more susceptible to a certain microorganism than the "normal" population. The Codex textd describes the principles to be applied, but lacks details concerning sampling plans and their interpretation. Furthermore, the considerations are of a general nature and apply to all foods. When dealing with specific foods, deci- sions must be made where criteria are to be applied in the food chain and what would be achieved by applying them. Sampling plans A sampling plan includes the sampling procedure and the decision criteria to be applied to a food lot, based on examination of a prescribed number of analytical sample units by defined methods. Sam- pling plans should be administratively and eco- nomically feasible. A sampling plan may define the probability of detecting a microorganism ( or group of microorganisms) in a foodstuff or that a specified concentration of microorganisms is not exceeded. It must be recognized that no practical sampling plan can ensure the absence of the target microorganism and that the concentration of microorganisms measured may be exceeded in a part of the lot that was not sampled. The probabili- ty of acceptance of a food lot with a concentration of microorganisms above that specified in the sam- pling plan is termed the "consumer risk". The re- jection of a food with a level of microorganisms below the acceptable level is called the "producer risk". Both must always be considered when setting a sampling plan. Sampling plans should take into account: • risks to public health associated with the hazard ( severity and likelihood of occurrence); • the susceptibility of the target group of consum- ers (very young or old, immunocompromised, etc.); • the heterogeneity of distribution of microor- ganisms or the randomness of sampling, and • the acceptable quality level (percentage of non-conforming or defective sample units tolerated) and the desired statistical probabil- ity of accepting or rejecting a non-conforming lot. Sampling plans, taking these considerations into account, were developed and published by the ICMSF (1). The ICMSF approach distinguished Wld hlth statist. quart., SD (1997) 3 categories of hazards based upon the relative de- gree of severity : • severe hazards, e.g. Cl.ostridium botulinum; Vero- toxinogenic E. coli; S. typhi; • moderate hazards, potentially extensive spread, e.g. Shigella; Salmonella; enterotoxigenic E. coli; and • moderate hazards, limited spread, e.g. Staphylo- coccus aureus; Cl.ostridium perfringens, B. cereus. This categorization and the examples are based on the best epidemiological data available at the time of publication. Those categories may need to be revised as a result of new risk assessment proce- dures. The other factor to be considered is the likeli- hood of occurrence of the hazard (i.e. risk), taking account of the anticipated conditions of use. Here the ICMSF again recognized 3 categories: • situations where the risk would decrease; • situations where the risk would increase; and • situations where the risk would remain the same. Combining the 3 levels of severity with the cate- gories of likelihood of occurrence, leads to differ- ent levels of concern called "cases" by the ICMSF, case 7 being of lowest concern to food safety and case 15 of the highest (Tabk 1). This is a basic form of risk assessment. In situations where the likelihood of occur- rence of pathogens is reduced before consump- tion, for instance by cooking during preparation, cases 7, 10 and 13 apply, depending on the severity of the hazard (Tabk 1 ). Cases 8, 11 and 14 refer to situations where the likelihood of occurrence would remain the same between the time of sam- pling and the time of consumption, i.e. where pathogens are unable to multiply in the food under expected conditions of handling, storage, preparation and use. Cases 9, 12 and 15 refer to situations where multiplication may occur. Based on these 9 cases, the ICMSF developed two-class sampling plans in which "n" indicates the number of sample units to be tested and "c" the number of defective sample units which can be accepted. These sampling plans are summarized in Tabk 1. The plans make it possible to direct more of the available resources for analysis towards those situations with a high level of concern. In a lot with a given percentage of defectives, the number of sample units examined determines the probability of detecting contaminated food lots. The limitation of sampling is that it is neither practical nor cost-effective to attempt to detect, with a high degree of confidence, low levels of contamination in many foods. In particular, it does not work well with many raw foods ( e.g. produce, fruit, etc.) which have been the source of signifi- cant problems lately. It must be realized that only positive results are meaningful, while negative re- sults provide the level of confidence set by the 121 Table 1 Stringency (case) plan in relation to degree of health hazard and conditions of use Tableau 1 Plan d'echantillonnage (cas ) pour determiner le niveau de risque en fonction du risque sanitaire et des conditions d'utilisation Degree of health hazard - Degr6 de risque pour la sant6 Conditions in which food is expected to be handled and consumed after sampling in the usual course of events - Conditions dans lesquelles !'aliment est normalement manipul6 et consomme apres l'echantillonnage Reduced degree of concern - No change in concern - May increase concern - Moderate, direct (limited spread) - Madere, direct (propagation limitee) Moderate, direct (potentially extensive spread) - Madere, direct (propagation eventuellement etendue) Severe, direct - Severe, direct n = number of sample units tested - nombre d'unites 6chantillonn6es examin6es. Peu de risque Case-cas 7 n = 5, c = 2 Case-cas 10 n = 5, c = O Case- cas 13 n = 15, c = 0 Risque stable Risque d'augmentation du risque Case-cas 8 Case-cas 9 n = 5, c = 1 n = 10, c = 1 Case-cas 11 Case-cas 12 n=10,c=0 n = 20, c = 0 Cpse-cas 14 Case-cas 15 ~ = 30, c = 0 n = 60, c = 0 c = number of defective sample units which can be accepted - nombre acceptable d'unites defectueuses. number of sample units tested, assuming that there is a homogeneous distribution of the pathogen in the lot. For example, finding no defectives after testing 5 sample units implies that a lot is less than 50% contaminated; finding no defectives in 30 samples indicates that the lot is less than 10% contaminated; and no defectives in 300 samples means that the lot is < 1 % contaminated (all at a 95% confidence level). Fig. 1 gives the probabilities of acceptance for 3 levels of contamination when using cases 10 to 15 and clearly indicates the limita- Fig.1 Probability of acceptance of lots with different levels of defective units Probabilite d'acceptation des lots pour differentes proportions d'unites defectueuses 100~~~~~~~~~~~~~~~~~~ 80 60 % 40 20 122 15 14 13 12 Case No. - Cas N° D 2% defective - d·unites defectueuses D 5% defective - d'unites defectueuses • 10% defective - d'unites defectueuses 11 10 tion of using microbiological testing of samples to assure food safety or to verify the effective imple- mentation of HACCP. Sampling plans form an essential part of a Codex microbiological criterion. Those criteria should be regarded as the demonstration that a food lot has been produced according to GMP (and HACCP where applicable). Once a criterion has been established, ICMSF wishes to emphasize that routine testing of all lots of food against such a criterion is impractical and often unnecessary. Realistic sampling and testing are hindered by dif- ficulty in obtaining a representative and random sample of a food lot from a consignment of food arriving at a point of entry into a country, particu- larly as it is often made up of a number of different food lots, and unacceptable delays while testing is performed. Indeed, the modern approach to the control of foodstuffs in international trade is in- creasingly based on agreements between coun- tries, e.g. memoranda of understanding and pur- chasing agreements with trading partners and less reliance on testing of product at point of entry. The decision to test must be made by the regu- latory authorities in an importing country on the basis of the availability of information on which to make a judgement as to likely acceptability. Exam- ples of factors that may influence whether or not to test an imported food for which microbiological criteria have been established are: • Supplier's history of compliance with: - GMP - HACCP - Criteria, including microbiological criteria • New information linking the food commodity with foodborne illness • Whether the food is: - commonly involved in disease primarily destined for sensitive population Rapp. trimest. statist. sanit. mond., SO (1997) • The country of origin is: - known to exercise control over the food - not in an area with endemic disease of im- portance to food safety • Practical considerations such as: - cost/benefit - the statistical limitations of the sampling plan for differentiating acceptable from un- acceptable lots, particularly when a low level of defective units is expected. Management of microbiological hazards for foods in international trade In the context of the WTO/SPS measurements, transparency and consistency are important. Coun- tries should manage food safety in a similar or equivalent manner. The use of Codex documents should be instrumental in achieving this goal. However, the development of Codex documents is a relatively slow process, and when several docu- ments are developed at the same time, it should be clear from the start how these texts are intended to be used. For this reason the ICMSF made apropos- al outlining the use of the Codex documents for the establishment of microbiological criteria for specific food commodities, and for reassessing ex- isting microbiological criteria in Codex docu- ments. As a prerequisite, information about the food must be available. Conclusion Microbiological safety criteria are used as one of the tools to protect the health of the consumer. Testing foods for pathogens is however not very effective, because negative results do not guaran- tee that the food is indeed safe to eat. Safety is obtained by the application of good hygienic prac- tices and HACCP as a food safety management tool throughout the food chain, "from farm to fork". However, an importing country may not know whether incoming consignments of foods were in- deed produced under good hygienic and safety conditions, and consequently may want to examine the lot. The microbiological criteria to be used in such a situation should be established according to the documents described, in order to be in line with the WTO/SPS agreements. These Codex doc- uments are not yet finalized, and basic require- ments such as risk assessment are still under devel- opment, but the general matrix is clear and can already be applied. Wld hlth statist. quart., 5D (1997) Summal}' Microbiological safety is achieved by applying good hygienic practices throughout the food chain, "from farm to fork". Governmental food control is traditionally based on inspection of the facilities where foods are handled, and on testing food samples. Testing is usually applied to imported foods, when no information con- cerning the safety of a consignment is available. The microbiological safety is judged by means of microbio- logical criteria. Such criteria should, in the context of the WfO/SPS measures, be scientifically justified, and es- tablished according to the principles described by the CodexAlimentarius. However, microbiological testing is not a very reliable tool for consumer protection; the emphasis is currently shifting to the application of food safety management tools SJCh as the Hazard Analysis Critical Control Point system (HACCP). Resume Etablissement de criteres de securite microbiologique pour les aliments faisant l'objet d'un commerce international La securite microbiologique s'obtient par !'application de bonnes pratiques d'hygiene tout au long de la chaine alimentaire, du producteur au consommateur. Le con- trOle officiel des denrees alimentaires repose tradition- nellement sur !'inspection des etablissements ou les aliments sont manipules et sur !'analyse d'echantillons. Des analyses sont habituellement effectuees sur les aliments importes lorsqu'aucune information concer- nant la salubrite d'un arrivage n'est disponible. La securite microbiologique est estimee d'apres des crite- res microbiologiques. Ceux-ci doivent, dans le cadre des mesures de l'OMC/SPS, ~Ire scientifiquement justi- fies, et etablis selon les principes decrits par le Codex Alimentarius. L'analyse microbiologique n'est toutefois pas un outil tres fiable du point de vue de la protection des consommateurs; on tend a s'interesser davantage a !'application d'outils de gestion de la salubrite des aliments comme le systeme d'analyse des risques - points critiques pour leur maitrise (HACCP). References/References 1. ICMSF, Microarganisms in foods 2- sampling far microbiologi.cal analysis: Principles and specific applications. Toronto, University of Toronto Press, 1986. 2. WfO, Trading into the future. Geneva, World Trade Organisation, 1995. 123 Improving the safety of street-vended food Gerald Moya, Anthony Hazzardb & Fritz Kiitersteinc Introduction Street-vended food provides relatively low-cost ready-to-eat foods for many populations living in urban areas, but also in peri-urban and rural areas. In urban areas, street foods are often an essential part of the food supply, particularly for low-income persons. Street-vending of food also provides a livelihood for a large number of workers (in some countries, women especially) (Fig. 1) and offers business opportunities for budding entrepreneurs. Street-vended foods support the tourist industry by offering culturally unique foods appreciated by many tourists. In contrast to these potential bene- fits, it is also recognized that street-vended foods have been implicated in outbreaks of foodbome diseases and the transmission of communicable diseases, such as cholera. Street food vendors are often poor, uneducated and lack appreciation of safe food handling. If a community is to get the full a Scientist, Food Safety and Food Aid Programme, World Health Organization, Geneva. b Food microbiologist, \\'HO Collaborating Centre for Environmental Health, University of Western Sydney, Australia. c Director, Food Safety and Food Aid Programme, World Health Organization, Geneva. Fig. 1 Socioeconomics of street food, by WHO Region benefits of street-vended foods with minimal risks of foodbome disease, government intervention is required to protect consumers and to ensure that the standard of safety of such foods is the best attainable in the context of the prevailing local situation. However, WHO emphasizes that care should be taken not to stifle this important outlet for ready-to-eat meals by excessive regulation ( 1 ). The risk of increasing malnutrition and hunger may be considered a serious consequence if health authorities overly restrict the availability or accessi- bility of street-vended foods. In its report (2), a Joint FAO/WHO Expert Committee on Food Safety, which was convened at WHO Headquarters in Geneva in 1983, recognized the importance and potential hazards of street- vended foods. Because of the possibility of micro- biological and chemical contamination, including adulteration, which conceivably could occur under street conditions, the Committee concluded that efforts must be made (i) to educate the personnel involved; (ii) to improve the environmental condi- tions in which the trade is practised; and (iii) to provide the essential services to assist street food vendors in assuring the safety of their wares. In 1986, ajoint FAO/WHO Expert Consultation on Food Protection for Urban Consumers (3), in con- sidering the problem of street-vended foods, rec- Aspects socio-economiques de la vente d'aliments dans la rue, par Region OMS 100--r~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--, ~ 80 "' "' "C "#. 60 I "' "' ~ 40 ::, 0 (.) ; 20 124 Urban importance - Importance en milieu urbain D AFR D AMR Important means of employment- Source importante d'emplois Significant number of women - Nombre important de femmes Rural importance - Importance en milieu rural Socioeconomic issues - Aspects soicio-economiques [21 EMR Q EUR El SEAR • WPR • Global - Ensemble des Regions Rapp. trimest. statist. sanit. mond., 50 (1997) Improving the safety of street-vended food Gerald Moya, Anthony Hazzardb & Fritz Kafersteinc Introduction Street-vended food provides relatively low-cost ready-to-eat foods for many populations living in urban areas, but also in peri-urban and rural areas. In urban areas, street foods are often an essential part of the food supply, particularly for low-income persons. Street-vending of food also provides a livelihood for a large number of workers (in some countries, women especially) (Fig. 1) and offers business opportunities for budding entrepreneurs. Street-vended foods support the tourist industry by offering culturally unique foods appreciated by many tourists. In contrast to these potential bene- fits , it is also recognized that street-vended foods have been implicated in outbreaks of foodborne diseases and the transmission of communicable diseases, such as cholera. Street food vendors are often poor, uneducated and lack appreciation of safe food handling. If a community is to get the full a Scientist, Food Safety and Food Aid Programme, World Health Organization , Geneva. b Food microbiologist, WHO Collaborating Centre for Environmental Health , University of Western Sydney, Australia. c Director, Food Safety and Food Aid Programme, World Health Organization, Geneva. Fig. 1 Socioeconomics of street food, by WHO Region benefits of street-vended foods with minimal risks of foodborne disease, government intervention is required to protect consumers and to ensure that the standard of safety of such foods is the best attainable in the context of the prevailing local situation. However, WHO emphasizes that care should be taken not to stifle this important outlet for ready-to-eat meals by excessive regulation (]). The risk of increasing malnutrition and hunger may be considered a serious consequence if health authorities overly restrict the availability or accessi- bility of street-vended foods. In its report (2), a Joint FAO/ WHO Expert Committee on Food Safety, which was convened at WHO Headquarters in Geneva in 1983, recognized the importance and potential hazards of street- vended foods. Because of the possibility of micro- biological and chemical contamination, including adulteration, which conceivably could occur under stree t conditions, the Committee concluded that efforts must be made (i) to educate the personnel involved; (ii) to improve the environmental condi- tions in which the trade is practised; and (iii) to provide the essential services to assist street food vendors in assuring the safety of their wares. In 1986, a Joint FAO/WHO Expert Consultation on Food Protection for Urban Consumers (3), in con- sidering the problem of street-vended foods, rec- Aspects socio-economiques de la vente d'aliments dans la rue, par Region OMS 100 </) 80 >, "' a. </) Q.) "C ";!?- 60 I </) Q.) E 40 <= :::, 0 u 0 20 ";!?- 0 Urban importance - Importance en milieu urbain Et] AFR Et] AMR 124 Important means of employment- Source importante d'emplois Significant number of women - Nombre important de femmes Rural importance - Importance en milieu rural Socioeconomic issues - Aspects soicio-economiques 12} EMR Q EUR fil!l SEAR • WPR • Global - Ensemble des Regions Rapp. trimest. statist. sanit. mond., 50 (1997) ommended that- in addition to those efforts men- tioned above - simple techniques for regulation of street foods be implemented and that the work of international organizations be extended to study street-food-vending practices and develop ade- quate strategies to improve their safety. In respond- ing to this recommendation, WHO has sponsored numerous studies in this field as have other inter- national organizations. However, the WHO ap- proach has focused on the use of the Hazard Anal- ysis and Critical Control Point (HACCP) system to identify behaviours and practices which may be hazardous (4, d). Extensive use of the HACCP ap- proach was also made in studies of street-vended foods in Bogor, Indonesia.e In 1993 the World Health Organization, through its six regional offices (the Americas (AMRO), Africa (AFRO), Eastern Mediterranean (EMRO), Europe (EURO), South-East Asia (SEARO) and Western Pacific (WPRO)) under- took a survey of its Member States to assess the current situation in regard to street-vended food and to obtain the views of responsible author- ities concerning the hazards posed by street- vended foods and contributing factors, as well as approaches for managing these hazards. Over I 00 countries participated in this survey which repre- sents the most extensive report on street-vended food available to date. The survey noted that al- most all countries reported a wide variety of foods, types of preparation, facilities and infrastructure (see Chart 1). d Application of the Hazard Analysis Critical Control Point (HACCP) system for the improvement of food safety • reports of WHO-supported case studies on food prepared in homes, at street food vending operations, and in cottage industries. WHO, Geneva, 1994 (Document 'WHO /FNU /FOS/94.1). e Winarno, F.G. Unpublished reports available from Food Technology Development Centre, Bogor Agricultural University, Kampus IPB Dannaga, P.O. Box 61, Bogor, Indonesia. Chart 1 Summary of WHO street-vended food survey Socioeconomics of street food sector Are street-vended foods a significant part of the urban food supply? Is street-vending of foods an important mean of employment? Are a significant number of women employed in this sector? Are street-vended foods important in rural areas? Description of street vended foods Type of food Grain & cereal Fruit & vegetables Meat & fish Frozen produce Beverages Wld hlth statist. quart., so (1997) 74 69 62 22 64 86 85 76 65 Type of preparation Ready to eat 97 Cooked on site 82 None (raw) 65 Conditions normally encountered in the trade Type of facility Mobile carts 75 Fixed stalls 88 Improved food centres 44 Type of infrastructure Potable water 47 Toilets 15 Hand washing facilities 32 Dish/utensil washing 48 Refrigeration available 43 Waste disposal 47 Factors contributing to foodborne disease outbreaks which are believed or shown to be Important in street-vended food Growth of pathogens Preparation too far in advance Left at room temperature Foods cooked in large pots Improper warm holding Extra large quantities prepared Survival of pathogens Inadequate cooking Inadequate reheating Contamination by pathogens Contamination of raw food Infected handler Cross-contamination Inadequate cleaning of equipment Unsafe source Contaminated water Management of street foods Laws and regulations covering street-vended food exist Registration of street vendors required Code of practice applied Periodical medical exam requiredb Periodic training required Inspection procedure Existing agency with responsibility for street food Ratio of vendors: inspectorc HACCP concept applied to street food a % countries replying "YES" to the particular question. b Not generally regarded as cost-ettective. c Exact figures not generally available. 73 82 33 72 38 58 72 76 70 73 72 56 57 72 32 32 41 34 79 Very large 23 Countries in all WHO Regions reported street- vended foods to be a significant part of both the urban and rural food supplies. However, it is clear that there is a greater reliance on such foods in urban settings (Fig. 1), and less so in the Eastern Mediterranean and Europe compared to Africa, the Americas, Asia and the Western Pacific. Street- vended foods included such as diverse fare as meat, fish, fruits, vegetables, grains, cereals, frozen pro- duce and beverages. Types of preparation m- 125 Fig. 2 Number of respondents indicating improved food centres as a strategy Nombre de pays indiquant !'amelioration des centres de restauration rapide en tant que strategie a: LI... < a: ~ < a: ~ w a: :::i w a: < w (/l WHO Regions - Regions OMS a: c.. 3:: eluded foods without any preparation (65%), ready-to-eat food (97%) and food cooked on site (82%). Vending facilities varied from mobile carts to fixed stalls and food centres. The latter were seen as a strategy in the improvement of street food safety in 44% of responding countries, with a great- er percentage of European and South-East Asia countries identifying improved food centres as fa- cilities normally encountered in the sector (Fig. 2). Although discussed later in this article, it is worth noting here that particular care should be em- ployed to ensure such centres are appropriate for the communities for which they are intended. Infrastructure developments were relatively limit- ed in some regions with restricted access to safe drinking-water, toilets, refrigeration/ice, and hand-washing and waste disposal facilities (Fig. 3). In less than 50% of countries, safe drinking-water was considered a condition normally available to street food vendors. In Africa and South-East Asia only between 30% and 40% of countries reported potable water being available for street food prepa- ration and vending. Refrigeration or ice was not normally encountered in more than half of the countries participating in the survey. The availabil- ity of refrigeration or ice was particularly limited in Africa, where less than I 0% of respondents consid- ered it to be normally accessible to street food- vendors. While a broad picture of limited access to toilets (15%) and hand-washing facilities (about 30%) also emerged, South-East Asian countries re- ported noticeably less capacity for hand-washing (none of this region's countries considered it nor- mal to encounter such facilities) and the washing 126 of dishes and utensils (approximately 25%) than countries of other regions. The majority of coun- tries reported time and temperature abuse, and contaminated raw foodstuff, infected handlers and inadequately cleaned equipment (Fig. 4) as the ma- jor factors contributing to foodborne disease. How- ever, it is interesting to note the significant differ- ence between regions regarding contaminated wa- ter as an important source of pathogens. While more than 90% of African countries saw contami- nated water as a significant factor contributing to foodborne disease, less than 20% of European countries considered it such. This observation re- flects the general availability of safe drinking-water in the European countries reporting, however the situation in the less developed countries of Europe may be different. At the same time, however, little information on the hazards and risk factors of street-vended foods was available to responsible authorities. As a result, current policies and approaches in many countries are not scientifically based and, in some cases, have led to measures that were not cost-effective and even some that were counter-productive with re- gard to the prevention of foodborne diseases. The Hazard Analysis and Critical Control Point (HACCP) System WHO has long advocated the application of the HACCP system to improve the safety of food at any stage of the food chain. It was natural that HACCP should increasingly become an important tool in WHO's food safety activities, particularly with re- spect to food prepared by street vendors and homemakers in developing countries. The ap- proach uses hazard analysis (HA) to assess all fore- seeable hazards and identification of critical con- trol points (CCP) to analyse the potential of each step of the operation to reduce, prevent or elimi- nate these hazards. Establishment of critical limits that must be monitored at identified CCPs in the process provides continuous assurance of the safety of the final product. This information may be used by food safety programme administrators and su- pervisors as the basis for setting priorities for inter- vention activities, including regulatory action. De- tailed HACCP analysis of priority foods, especially CCP monitoring and verification procedures, may be employed for the training of both health and food regulatory officials and street vendors. Fur- thermore, when suitably presented, this informa- tion can be used in health education campaigns for the public. The HACCP system offers a rational approach to the control of health hazards in foods, avoids the many weaknesses inherent in the traditional in- spectional approach and circumvents the short- comings of reliance on end-product testing. In 1993, the Joint FAO/WHO Codex Alimentarius Commission endorsed HACCP as the most cost- Rapp. trimest. statist. sanit. mond., 50 (1997) Fig. 3 Conditions normally encountered in the street-food trade, by WHO Region Conditions normalement rencontrees dans la vente d'aliments dans la rue, par Region OMS 100~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Cl) 80 >, ~ Cl) "' -o 60 -;F. I Cl) "' E 5 40 0 u 0 -;F. 20 till AFR O AMR f.21 EMR 12] EUR ml SEAR • WPR • Global - Ensemble des Regions Safe drinking water - Eau potable Toilets - Toilettes Hand-washing facilities - Lavabo Dish/utensil washing - Evier Refrigerator/Ice available - Refrigerateu r/g lace Waste disposal - Evacuation des dechets Type of infrastructure - Type d'infrastructure Fig. 4 Factors believed to be significant in the contamination of street-vended food by pathogens Facteurs juges importants dans la contamination par des agents pathogenes des aliments vendus dans la rue 100-.-::::-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ u] AFR O AMR C/l 80 ?;;' Cl. Cl) "' "O -;F. 60 I C/l "' E c: 40 :::, 0 u -;F. 20 0 l2J EMR [) EUR llli] SEAR • WPR • Global - Ensemble des Regions Contaminated water - Eau contaminee Infected handler - Manipulateur infecte Contaminated raw material - Matieres premieres contaminees Contamination factors - Facteurs de contam ination effective approach devised to date for assuring the safety of food (5). Unlike standard "check-list" inspections, the HACCP approach is based on an understanding of the factors that contribute to outbreaks of foodborne disease and on applied research on the ecology, multiplication and inacti- vation of food borne pathogens and toxins, the haz- ards of food additives and chemical contaminants and the presence of foreign objects. Even when data on foodborne diseases are not available, haz- ard analysis can detect potential problems and identify the critical control points of a street-vend- ing operation. Thus food safety agencies can target their limited resources on public health risks of greatest concern. HACCP also provides guidance in the selection of enforcement and education pri- orities, rather than general sanitation and aesthetic Wld hlth statist. quart., 50 (1997) improvements. Additionally, valuable information about food and processes will be obtained, and follow-up actions to verify that the vendors are monitoring the critical control points will take less time. The benefits derived from greater assurance of food safety should offset the time spent on the initial hazard analysis and verification. Additional benefits will ensue from inspections of potentially hazardous foods and operations to determine whether they are being monitored effectively, rather than randomly inspected, when only a few high-risk foods or operations may be seen. The HACCP system should be the foundation for promoting safe handling of street-vended foods through studies that identify potential hazards and effective measures for their control. Where epide- miological data on food borne diseases are available, 127 such as common causative agents, places where mishandling occurs, and factors contributing to outbreaks, these should also be taken into consider- ation. Where there is no national or local surveil- lance of food borne disease, epidemiological or re- search data from other countries where the same foods are prepared in a similar way, may indicate probable vehicles of foodborne pathogens or tox- ins. Preliminary studies such as these are essential in developing strategies, both voluntary and mandato- ry, for assuring the safety of street-vended foods. A 1992 WHO publication on the application of HACCP to small food businesses can provide useful guidance in conducting HACCP studies of street- vended foods (6). In specific response to the need for better guidance on cost-effective approaches to street-vended foods , WHO has recently prepared the document "Essential safety requirements for street-vended foods"f which further clarifies how authorities may adopt a HACCP-based strategy to prioritize food control measures specific to the safety of street-vended foods. Application of HACCP to street-food-vending operations As food safety problems include process-specific and food-enterprise-specific issues, the most effec- tive application of the HACCP system, in the im- provement of food safety, requires the food enter- prise to develop and implement its own HACCP plans. However, it is unlikely that street food- vendors would have the food safety knowledge nec- essary to develop, implement and evaluate an HACCP plan. Consequently, the application of f Essential safety requirements for street-vended foods. WHO, Geneva, 1996 (Document WHO / HPP / FOS/ 96. 7) . Fig. 5 Management of street-vended food Gestion des aliments vendus dans la rue Ifill AFR D AMR l2J EMR [] EUR er, 80 >, "' Cl. er, <1) "O 60 ~ I er, <1) ·c c 40 ::::, 0 u 0 ~ 20 0 HACCP to street-food operations must differ from its application to industry. A greater emphasis must be placed on the role of regulatory authorities in the development of the HACCP plan and in the maintenance of verification procedures. Still, the collaboration of vendors in the establishment of HACCP principles will enhance the vendors' un- derstanding of the factors influencing food safety. The implementation of monitoring and corrective action procedures by vendors will create a signifi- cant shift in the balance of responsibility shared by the operator and the regulatory authorities. Greater emphasis will be placed on the operator's responsibility to assure the safety of the vended food. The role of regulatory authorities in conduct- ing the HACCP studies necessary for development ofHACCP plans gives rise to the question of how to match regulatory agency resources to the size and diversity of the street food sector. An HACCP study and plan cannot be completed for each operator or for every type of street food. Even with the identification of priority foods, regulatory agency resource limitations would not permit the develop- ment of enterprise-specific HACCP plans. Pragma- tism would suggest the need for regulatory agen- cies to undertake HACCP studies involving a select number of vendors for each priority food and to apply the findings to risk management on a broader scale. While individual formulations and handling practices for basic types of street food may vary, the information obtained will improve food safety by better focusing interventions on is- sues directly relevant to food safety. Management options with the HACCP approach Interventions that promote the safety of street- vended foods include: (i) the establishment and fj) SEAR • WPR • Global - Ensemble des Regions Laws and regulations exist - Lois et reglements en vigueur Registration required - Inscription exigee Code of practice applied - Code de pratique applique Medical examination required - Examen medical exige Training of vendors required - Formation des vendeurs exigee Mandatory management practices - Pratiques gestionnaires obligatoires 128 Rapp. trimest. statist. sanit. mond., 50 (1997) enforcement of regulations and codes of practice; (ii) training of street food vendors; (iii) awareness education of consumers; (iv) improved design and construction of vending premises and equipment; and (v) provision of infrastructure and services. The establishment and use of general food reg- ulations and codes of practice accompanied by inspection programmes has been the traditional approach taken by regulatory agencies. More than 70% of responding countries indicated they had existing food laws and regulations addressing street foods (Fig. 5). However, it is now widely recognized that regulatory agencies do not have the resources required to enforce regulations with a sector of such size and mobility. In recent years, interest in pro- moting the safety of street-vended foods through the use of codes specific to this sector has been increasing in many countries, although the survey indicates few have yet applied such codes (Fig. 5 ). In response to this increased interest, the Joint FAO/WHO Codex Alimentarius Commission was asked to develop an international code of practice for street-vended foods although this was not strictly within its mandate. The Codex Coordinating Com- mittees for Africa, Asia and Latin America and the Caribbean each developed separate codes which were considered by the Codex Committee on Food Hygiene (CCFH) in 1991 for consolidation into an international code. However, the CCFH concluded that given the variation in foods, practices and conditions around the world, an international code was inappropriate. For the same reasons, the CCFH noted that regional codes were possibly inappropri- ate as well. The Committee recommended that the codes be returned to the Regional Committees for further consideration. At its 20th session in Geneva (1993), the Codex Alimentarius Commission con- curred with the recommendation of the CCFH and invited the relevant Codex Coordinating Commit- tees to reconsider their approach to the question of street-vended foods to determine whether regional codes in their present form were actually useful. Despite such misgivings, the 2lst session of the Codex Alimentarius Commission in 1995 approved the adoption of a Codex Code of Hygiene Practice for the Preparation and Sale of Street-Vended Foods as a Regional Code for Latin America and the Caribbean and the Commission will be presented with a similar Regional Code for Africa. As with promulgated general regulations, regu- latory agencies are unlikely to have the resources necessary to enforce codes of practice addressing the street food sector. The general nature of such codes result in their failing to address certain criti- cal control points while including elements which may be easily enforced, but which may impact only indirectly or not at all on food safety. In order to provide direction on the elements more likely to have an impact on food safety, WHO has issued the previously mentioned document on essential safety Wld hlth statist. quart., 50 (1997) requirements for street-vended foodsf. The docu- ment which took into account principles enunciat- ed in the Codex document "The Recommended International Code of Practice - General Principles of Food Hygiene"e provides general guidance for regulatory authorities, but is primarily intended as a basis for the training of street-food vendors, which is recognized as the most cost-effective inter- vention. However, many codes of practice, thus far developed, may increase the likelihood of failing to focus on critical control points either for inspec- tion purposes or as the basis for training pro- grammes. Viewed from a general perspective, many food- borne hazards may be prevented, reduced to an acceptable level or eliminated by selection of safe raw materials, thorough cooking, hot holding, rapid cooling, cold storage, avoidance of cross con- tamination or combinations of these. In addition, food handlers may themselves cause food contami- nation although the problem commonly has less to do with personal hygiene and more with faulty food-handling practices than is generally recog- nized. Pathogenic organisms present in, or on, the food handler's body or clothing, can come into contact with food, or with surfaces used to prepare food, and, under the right conditions, multiply to levels which constitute an infective dose. However, there has been some debate as to whether the food handler is contaminating the food or the food is infecting the handler. For example, in the case of non-typhi salmonellosis, epidemiological evidence indicates that food handlers have acquired their infections during the course of an outbreak and have been victims themselves rather than the source (7). Nevertheless, the risk of food handlers infected or colonized with pathogens contaminat- ing food needs to be minimized through improved personal hygiene and food handling practices. In many countries, infected handlers are considered a significant source of contamination (Fig. 4) and medical examination of vendors is required (Fig. 5 ). Approximately two-thirds of Eastern Medi- terranean countries report using mandatory medi- cal examinations as a management practice with street food vendors. WHO considers this practice cost inefficient and of limited value due to the short duration for which such examinations are relevant (7). \VHO has provided a number of guidance doc- uments attempting to focus attention of training programmes on the factors contributing to food- borne disease. These documents include its "Ten Golden Rules for Safe Food Preparation"g and "A Guide on Safe Food for Travellers".h However, the g The "'110 goui.en rulRs JM safe food preparation (poster), Food Safety Unit, WHO, Geneva, 1992. h AguideonsafefoodjMtraveUers. WHO, Geneva, 1994 (Document WHO/FNU/FOS/94.2). 129 more general a document, the less likely it is to address all of the critical control points confront- ing a particular food enterprise. Consequently, WHO encourages regulatory authorities to dev- elop, for priority foods, food- or food-group-specif- ic training materials and programmes based upon the findings of the afore-mentioned HACCP studies. Clear direction should be provided to ven- dors on points that must be controlled, critical limits, monitoring procedures and establishment of appropriate corrective actions where necessary. WHO also promotes the development of regula- tions or codes of practice that focus on the ven- dor's responsibility in establishing and maintain- ing a safe food programme and in applying good hygiene practices essential to food safety. Interventions aimed at increasing the impor- tance of food safety as a consumer's product and vendor selection criterion may also prove a cost- effective approach to improving the safety of street- vended food. Unfortunately, consumers are often unaware of the relationship between contaminated food and foodborne disease. Furthermore, con- sumers often will place more emphasis on aesthetic aspects instead of the important factors which can protect against foodborne disease. In advising trav- ellers and consumers, in general, WHO has pro- vided guidance on common foodborne hazards. However, HACCP studies will give rise to more food-specific information on safe and unsafe food- handling practices. As a result, the messages de- rived from such studies may have a more direct impact on consumer selection practices based on food safety. Care should be taken in designing education programmes for the consumer to ensure that messages are concise, focused on critical con- trol points, accurate and practical. The provision of infrastructure and services, such as safe drinking-water supplies, refrigeration, sanitary facilities and waste disposal, are usually prohibitively expensive, if not impossible, but may be considered as part of a long-term strategy to improve food safety. In the interim, WHO has con- sidered other approaches for promoting the avail- ability of safe water, such as chemical disinfection. However, based on WHO's long experience in community water and sanitation programmes, such infrastructure and services are not sufficient by themselves to improve the level of hygiene and sanitation in a community. Hygiene education, particularly as it relates to food handling, is essen- tial. In fact, education in safe food practices can have a major impact on improving the safety of food even in the absence of such infrastructure and services. It is therefore of particular concern to note that many respondants to the WHO survey indicated training of vendors was not required. Where street-food vendors have been accepted as a legitimate part of the food supply system, they may be included in urban development planning. 130 Among other benefits, this recognition may enable some vendors to obtain loans for improving their business. For example, municipal building codes might require that certain large buildings incorpo- rate food-vending centres into their design. Desig- nated vendor areas with facilities may also be created in specific locations which are satisfactory to vendors, consumers and town planners. It should be noted that street food vendors are, in many countries, a vital part of their communities and, therefore, an effort should be made to keep them as close to their current neighbourhood as possible, even though some facilities may not be available. The design and construction of vending pre- mises and equipment has also been the object of increased interest and activity in recent years. In 1995, an FAO technical meeting on street foods (8) was of the opinion that experience in the develop- ment of appropriate technologies gained in recent years should be evaluated regarding vendor I con- sumer acceptability and with respect to food safety improvements. However, it is clear that many such developments are not related directly to food safety and consequently contribute to a significant drain on the regulatory agency resources available to improve street food safety. Interventions relating to design and construction of vending premises and equipment must be focused on relevant criti- cal control points identified by HACCP studies if such interventions are to be effective in reducing health hazards associated with street food. For further information and assistance on the application of HACCP to various stages of the food supply system, readers are encouraged to contact the WHO Representative in their countries or to write directly to the Director, Programme of Food Safety and Food Aid, WHO, CH-1211 Geneva 27, Switzerland. Summary An integrated plan of action for improving street food involving health and other regulatory authorities, ven- dors and consumers should address not only food safety, but also environmental health management, in- cluding consideration of inadequate sanitation and waste management, possible environmental pollution, congestion and disturbances to traffic. However, WHO cautions that, in view of their importance in the diets of urban populations, particularly the socially disadvan- taged, every effort should be made to preserve the benefits provided by varied, inexpensive and often nutritious street food. Therefore, authorities concerned with street food management must balance efforts aimed at reducing the negative aspects on the environ- ment with the benefits of street food and its important role in the community. Health authorities charged with responsibility for food safety control should match risk management action to the level of assessed risk. The Rapp. trimest. statist. sanit. mond., 50 (1997) rigorous application of codes and enforcement of regu- lations more suited to larger and permanent food service establishments is unlikely to be justifiable. Such rigorous application of codes and regulations may result in disappearance of the trade with consequent aggrava- tion of hunger and malnutrition. Moreover, most codes and regulations have not been based on any systematic identification and assessment of health hazards associ- ated with different types of foods and operations as embodied in the HACCP approach which has been recognized by Codex as the most cost-effective means for promoting food safety. WHO encourages the devel- opment of regulations that empower vendors to take greater responsibility for the preparation of safe food, and of codes of practice based on the HACCP system. Resume Moyens d'ame/iorer la salubrite des aliments vendus dans la rue Un plan d'action integre pour !'amelioration des aliments vendus dans la rue, impliquant les autorites sanitaires et autres autorites reglementaires, les vendeurs et les consommateurs, doit porter non seulement sur la salu- brite des aliments mais aussi sur la gestion de !'hygiene de l'environnement, notamment sur des questions telles que l'insuffisance de l'assainissement et de la gestion des dechets, la pollution eventuelle de l'environnement et la perturbation de la circulation routiere. Cependant, l'OMS souligne le fail qu'etant donne leur importance dans l'alimentation des populations urbaines, en parti- culier des groupes socialement defavorises, ii taut s'ef- forcer de preserver les a vantages qu'offrent les aliments varies, peu coOteux et souvent nutritifs proposes par les marchands ambulants. Les autorites chargees de cette question doivent done trouver un equilibre entre la reduction des aspects negatifs sur l'environnement et la preservation des aspects positifs des aliments vendus dans la rue, notamment leur r61e important dans la vie de la communaute. Les autorites chargees du contr61e de la salubrite des denrees alimentaires doivent adapter les mesures de gestion des risques au niveau de risque estime. L'application rigoureuse des codes et regle- ments convenant pour des etablissements permanents du secteur alimentaire est probablement peu justifiee dans le cas des vendeurs des rues; elle pourrait se traduire par une disparition de ce commerce et par une Wld hlth statist. quart., 50 (1997) aggravation des problemes poses par la faim et la malnutrition. De pi us, la pi u part des codes et reg lements ne reposent pas sur une identification et une evaluation systematiques des risques pour la sante associes a differents types d'aliments et de processus alimentaires comme le fait l'approche HACCP (analyse des risques - points critiques pour leur maitrise}, reconnue par le Codex Alimentarius comme le moyen ayant le meilleur rapport coOt-efficacite pour promouvoir la salubrite des aliments. L'OMS encourage l'etablissement d'une re- glementation conferant aux vendeurs une plus grande responsabilite dans la preparation d'aliments sains, et !'elaboration de codes de pratique fondes sur l'appro- che HACCP. References/References 1. Abdussalam, M. &: Kaferstein, F.K. Safety of street foods. World health forum, 14: 191-194 (1993). 2. WHO Technical Report Series, No. 705, 1984 (Therof.eoffood safety in health and devewpment: report of the joint FAO/WHO Expert Committee on Food Safety). OMS Serie de Rapports techniques N° 705, 1984 (La securite des produits alimentaires et son rof.e dans I.a santi et f.e deueloppement: rapport d'un Comite mixte d'Experts FAO/OMS de I.a securite des produits alimentaires ). 3. Food protection for urban consumers. Report of a Joint FAO/ WHO Expert Consultation, 1-5 December 1986, Rome, Italy. FAO, Rome (1987). 4. Desmarchelier, P. et al. Evaluation of the safety of domestic food preparation in Malaysia. Bulf.etin of the World Health Organization, 72(6): 877-884 (1994). Desmarchelier, P. et al. Evaluation de la salubrite de la preparation domestique des aliments en Malaisie [resume]. Bulf.etin de ['Organisation mondiaf.e de /,a Santi, 72(6): 877-884 (1994). 5. Guidelines for the application of the Hazard Analysis Critical Control Point (HACCJ') system, Codex Alimentarius Commission. JointFAO/\o\'HO Food Standards Programme, FAO, Rome, 1993. 6. Bryan, F. Hazard Analysis Critical Control Point Evaluations, a guide to identifying hazards and assessing risks associated with food preparation and storage. Geneva, World Health Organization, 1992. 7. WHO Technical Report Series, No. 785, 1989 (Health surveill.ance and management procedures for food-handling personnel). OMS Serie de Rapports techniques N° 785, 1989 (Le controf.e sanitaire et /,a gestion des manipul,ateurs de produits alimentaires ). 8. Report of anFAO TechnicalMeetingonStreetFood, 6-JONovember 1995, Calcutta, India, Rome, Food and Agriculture Organization of the United Nations, (in press). 131 Monitoring and assessment of dietary exposure to chemical contaminants Ramesh V. Baht a & Gerald G. Moyb Introduction More than 10 million unique chemical entities are now known to scientists and several hundred thousand are in common use throughout the world. While the technologies associated with the use of these chemicals have resulted in a dramatic improvement in living conditions for most of the world's population, these technologies and related human activities have also resulted in the introduc- tion of numerous chemicals, directly or indirectly, into the food supply. Only a few of these chemicals have been fully characterized in terms of their po- tential toxicities to animals and humans, particu- larly in relation to their long-term effects. Chemical contaminants present in food may result from their natural occurrence in soil ( e.g. cadmium, lead, and mercury) or from mycotoxin contamination (e.g. aflatoxin), from pollution aris- ing from industrial and other human activities ( e.g. lead, mercury, cadmium, and polychlorinated biphenyls (PCBs)) from agricultural practices ( e.g. pesticides, fertilizers, and drugs used in food ani- mals) and from food processing and packaging ( e.g. nitrosamines, certain polycyclic aromatic hydrocarbons and lead). These contaminants may present a potential hazard for human health if exposure exceeds tolerable levels. Food contami- nation monitoring is an essential component of ensuring the safety of food supplies and managing health and environment resources, since it pro- vides information on the levels and sources of con- taminants in foods, on the amounts of contami- nants ingested by humans, and on trends in con- tamination levels. Prevention and control of food contamination are highly dependent on adequate and reliable food contamination monitoring programmes. Such programmes offer a number of important benefits to governments and are an integral part of the food safety programmes in most industrialized countries. At the international level, the need for a global assessment oflevels of chemicals in food and in the total diet led to the establishment of the Global Environment Monitoring System - Food Contamination Monitoring and Assessment Pro- a Deputy Director, National Institute of Nutrition, Hyderabad, India. b Food Safety Scientist, Programme ofF ood Safety and Food Aid, Division of Food and Nutrition, World Health Organization, Geneva, Switzerland 132 gramme (GEMS/Food) in 1976 which now in- cludes participating institutions in more than 60 countries throughout the world. The main objective of the Programme is to inform governments, the Codex Alimentarius Commission, other relevant institutions and the public on levels and trends of contaminants in food, their contribution to total human exposure and significance with regard to public health and trade. The Programme is conducted globally, re- gionally, and nationally, as well as in local areas of special concern that serve as models - where ap- propriate - for remedial actions, food control and resource management. Supporting components of the Programme involve technical cooperation, training, analytical quality assurance studies and information exchange. GEMS/Food collects residue and intake data on the following 18 priority contaminations: • Industrial chemicals ( 4) - lead, cadmium, mer- cury, and polychlorinated biphenyls (PCBs); • Organochlorine pesticides (8) - aldrin/ diel- drin, DDT-complex, heptachlor and hepta- chlor epoxide, hexachlorobenzene, total hexa- chlorocyclohexane (HCH) isomers, gamma- HCH (lindane), endosulfan, and endrin; • Organophosphorus pesticides (5) - diazinon, fenitrothion, malathion, parathion, and para- thion-methyl; • Mycotoxins (1) -Aflatoxin. The major source of information for this paper was data submitted to GEMS/Foodc as well as results of the most recent assessment of regional data on estimated total dietary intakes of contami- nants in food (]). Additional data were obtained from national reports and from the literature. Risk assessment of chemical contaminants in food If toxicological information and/ or exposure esti- mates suggest that adverse health effects might be expected as a result of a chemical contaminant in food, more detailed information should be ob- tained on its toxicity (hazard characterization) and on the actual human exposure ( exposure assess- ment) in order to conduct a risk assessment of the c Jelinek, C. Assessment of dwtary intake of chemical contaminants, Joint UNEP/FAO/WHO Food Contamination Monitoring and Assessment Programme, WHO, Geneva, 1992 (Document WHO/ HPP /FOS/92.6). Rapp. trimest. statist. sanit. mond., 50 (1997) potential hazard.cl As the risk becomes a matter of public health concern, various management op- tions can be considered which can be addressed to the source of the contamination or to the affected food. Any risk management decision must also in- clude consideration of relevant economic, social and political factors (2). Once an option is selected and, where appropriate, intervention programmes are designed and implemented, further monitor- ing will provide a basis for assessing the effective- ness of the programmes and for future manage- ment decisions. In the following sections, characterization of the risks posed by potential foodborne hazards has been considered in the light of established safe levels, such as the Acceptable Daily Intake (ADI) d Application of risk analysis to food standards issues. Report of a Joint FAO/WHO Consultation, 13-17 March 1995, WHO, Geneva, 1995 (Document WHO/FNU/FOS/95.3). Table 1 Dietary intake of lead by adultsa Tableau 1 Doses de plomb ingerees avec les aliments par les adultesa and the Provisional Tolerable Weekly Intake (PTWI) when these have been established by the Joint FAO/WHO Expert Committee on Food Ad- ditives (JECFA) or the Joint FAO/WHO Meeting on Pesticide Residues. In cases where these are not established, recommended national levels have been considered. Dietary intake of lead Information from 23 countries, mostly industrial- ized, on the dietary intakes of lead by their adult populations was available for review. Average week- ly lead intakes for adults in each of the various countries are given in Tah/,e 1. For a variety of factors, a strict comparison of the levels among the various countries cannot be made. The differences in years in which the studies were carried out and in the study approaches, and inadequacies in ana- lytical quality control may account for much of the differences in intakes. This applies for all contami- nants and countries discussed in this paper. Intake (µg/kg bw/wk) - Dose ingeree (µg/kg de poids corporeVsemaine) Country/area - Year- Median- Mean- 90th percentile - Reference b/remarks- Pays/territoire An nee Mediane Moyenne 90• centile Reference b/remarques Australia - Australie 1983 2.4 1984 4.2 10.6 1985 4.0 1987 8.0 17.1 95th percentile consumer - Consommateurs du 95e centile Belgium - Belgique 1980 31.5 1982 11.2 20.9 41.7 (3) Highest in Brussels - Valeurs les plus elevees a Bruxelles Canada 1981 4.6 5.7 10.9 GEMS/Food; (4) China - Chine 1988 5.7 6.4 (5) Beijingc - Pekinc Cuba 1984 63.0 63.7 96.6 Denmark - Danemark 1980 7.7 34.3 1980 23.7 With 1/4 liter wine/day-Avec 1/41 de vin par jour 1980 30.0 Near heavy traffic -A proximite d'une forte circulation automobile 1980 32.6 Near smelter - Pres de fonderies 1985 6.6 7.0 10.5 Finland - Finlande 1980 7.0 (6) Mykkanen et al. (1984) 1981 4.2 1986 1.6 1984-88 2.3 (7) Market basket - Panier de la menagere 1984-88 2.7 (7) Duplicate diet - Rations en double France 1983 19.8 Germany - Allemagne 1982 14.7 21.9 GEMS/Food; (8) 1987 14.1 28.6 71.4 Guatemala 1982 21.6 20.8 27.6 Urban - Zone urbaine 1984 15.6 18.5 26.4 Urban - Zone urbaine 1985 15.3 18.3 22.1 Urban - Zone urbaine 1988 30.5 32.3 55.0 Urban - Zone urbaine Wld hlth statist. quart., 50 (1997) 133 Table 1 (continued) Tableau 1 (suite) Intake {µg/kg bw/wk) - Dose ingeree {µg/kg de poids corporel/semaine) Country/area - Yw- Median- Mean- 90th percentile - Reference b/remarks- Pays/territoire An nee Mediane Moyenne 90• centile Reference b/remarques Hungary - Hongrie 1980 10.6 11.3 17.0 1982 9.2 10.3 18.7 1983 11.9 12.2 17.4 1984 12.3 14.0 20.2 India- lnde 1981 64.0 (9) Duplicate diet - Rations en double 1981 56.0 (9) Market basket - Panier de la menagere Ireland - lrlande 1982 17.6 1985 15.2 Italy - ltalie 1981 28.5 Rural families - Families rurales 1982 39.1 Rural families - Families rurales Japan - Japan 1980 6.2 10.1 10.9 1981 5.9 7.6 13.4 1981 4.7d (10) Rural females - Sexe feminin, milieu rural 1983 11.3 9.9 16.8 1985 4.6 5.6 9.8 1987 4.4 5.6 8.1 1988 4.8 9.8 15.2 Netherlands - Pays-Bas 1980 15.4 47.6 Selected foods - Chaix d'aliments 1985 6.4 Duplicate diet - Rations en double 1984-85 3.4 Duplicate diet - Rations en double 1985 3.3 4.0e (12) Duplicate diet - Rations en double 1984-86 3.4 Market basket - Panier de la menagere 1984-86 3.3 3.4 (13) 1988 15.4 Selected foods - Chaix d'aliments 1984-88 5.5 (7) Market basket - Panier de la menagere New Zealand - Nouvelle-Zelande 1982 24.9 Poland - Pologne 1982 7.0 20.2 55.0 Industrial area - Zone industrielie 1982 5.6 16.5 44.8 Non-industrial area - Zone non industrielie 1983 7.9 22.4 60.5 Industrial area - Zone industrielie 1983 1.9 14.0 39.9 Non-industrial area - Zone non industrielie 1981-83 10.5 (14) Wroclaw, Lublin 1987 11.5 (15) Lublin, Olzstyn, Poznan Rep. of Korea - Republique de Coree 1985 4.6d (11) Rural families - Families rurales 1985 12.3 d (11) Seoul families - Families de Seoul Sweden - Suede 1980 2.7 1980 3.0 (16) 1983 3.8 (11) 1984-88 1.75 (7) Average person; market basket - lndividu moyen; panier de la menagere 1988 3.0 3.0 4.4 (5, 18) Stockholmb Switzerland - Suisse 1984-88 3.5 (7) Turkey- Turquie 1984-88 6.4 (7) Rural area - Zone rurale UK - Royaume-Uni 1980 9.2 14.0 19.8 1981 7.7 8.5 1981 30.8 47.2 114.1 (19) High-lead water, Ayr - Eau a forte teneur en plomb, Ayr 1983 6.9 7.4 8.3 1985 6.6 6.8 7.5 1987 6.7 6.9 7.4 1988 6.9 7.1 7.3 1980 7.9 8.3 12.2 1982 5.5 5.7 8.4 1985 4.4 New study design - Nouveau plan experimental 1986 1.8 134 Rapp. trimest. statist. sanit. mond., 50 (1997) Table 1 (continued) Tableau 1 (suite) Intake (µg/kg bw/wk) - Dose ingeree (µg/kg de poids corporel/semaine) Country/area - Pays/territoire Year- Median- Mean- 90th percentile - Reference b/remarks- Reference b/remarques An nee Mediane Moyenne 90• centile 1987 1.3 1986-88 1.1 1988 1.8 (5f a FAO/WHO provisional tolerable weekly intake: 50 µg/kg body weight. - Dose ingeree hebdomadaire tolerable provisoire FAQ/OMS : 50 µg/kg de poids corporel. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. c Non-smoking adult females, duplicate diet. - Femmes adultes non fumeuses, rations en double. d Geometric mean. - Moyenne geometrique. • Three-fold decrease from 1978. - Diminution d'un facteur 3 depuis 1978. Because the P1WI for lead refers to the maxi- mum intake from all sources, not only the 90th percentile intakes, but even the median intakes should be well below 25 g/kg body weight. This is particularly true for intakes of infants and children because some of the young can ingest substantial amounts of lead from paint, dust and soil by hand- to-mouth activities in addition to exposure through air and water. Dietary intakes of lead by infants and children are given in Tab/,e 2. In view of the relative frequency that median and mean intakes of lead approached or exceeded the PTWI, more countries, both industrialized and developing, should carry out dietary intake studies for lead. Surveys should be conducted on a regular basis to determine the impact, if any, of prevention and control measures on levels of lead in food. More duplicate diet studies should be carried out in areas with heavy industrial activities, especially mining and metal processing, in areas with high lead levels in drinking water and in areas of high traffic density if leaded petrol is still in use. Since dietary intakes of lead relative to body weights are highest for infants and young children, who are also especially sensitive to the toxic effects of lead, and since lead passes the placental barrier readily, every effort should be made to reduce the levels of lead in foods for infants and children, in foods consumed by women of childbearing age and in drinking water. In addition, more' duplicate diet studies for lead should be conducted with pregnant and nursing women and with infants and young children. Dietary intake at cadmium The average intakes from recent surveys in a num- ber of countries are presented in Tab/,e 3. Because of their higher consumption, Canada, Denmark, Finland, the Netherlands and the United States of America identified cereals and their products, fol- lowed by potatoes and other vegetables, as the largest contributors to this intake. \Vhile animal Wld hlth statist. quart., 5D (1997) kidneys, molluscs and crustacean had higher lev- els, these foods constitute much less of the average diet. The average weekly intakes of cadmium by in- fants and young children reported in surveys by the 11 countries are given in Tab/,e 4. The average in- takes reported from Cuba and Poland exceed the P1WI. The intakes reported from Finland, Ger- many, Canada, the United Kingdom and Niger were around 50% of the P1WI. In many of the reporting countries, the median or mean cadmium intakes constitute an appreci- able percentage of the P1WI. Where the 90th per- centiles were reported, they approached or ex- ceeded the P1WI in some cases. Therefore, expo- sure to cadmium in the diet is of public health concern. Nationally representative dietary intake studies should be carried out by more countries and appropriate measures should be taken to min- imize its occurrence in the diet. Well-designed duplicate diet studies should be conducted in locations of probable cadmium con- tamination, such as those near mining and metal extraction operations, phosphate fertilizer plants, high-cadmium bearing strata in the soil, municipal sludge-deposition areas, and shellfish areas af- fected by improperly treated industrial or munici- pal discharges. Caution should be observed in con- suming meat organs, shellfish, vegetables, fruit and grains from areas of known cadmium contamina- tion, and appropriate measures should be taken to minimize its occurrence in these areas. Dietary intake at mercury Data reported from 15 countries on weekly adult dietary intake were available for assessment. The average total mercury intakes from each of the countries are presented in Tab/,e 5. The highest average intakes, from Poland and Denmark, are about 60% of the P1WI of 3.3 µg/kg body weight for methylmercury, or about 40% of the P1WI for total mercury. The average intakes from Germany, 135 Table 2 Dietary intake of lead by infants and childrena Tableau 2 Doses de plomb ingerees avec les aliments par les nourrissons et les enfantsa Country/area - Year- Age Median- Mean- 90th Reference b/remarks- Pays/territoire An nee Mediane Moyenne percentile- Reference b/remarques 90• centile Australia - Australie 1983 9mo. 8.5 1984 9 mo. 24.4 1985 9 mo. 12.1 32.9 1987 9 mo. 8.9 12.1 95th percentile consumers - Consommateurs du 95e centile Austria - Autriche 1981 0-6 mo. 118.0 (20) Canada 1987 0-12 mo. 8.3 (21 J Ready-to-use formulae - Formule prate a l'emploic 1987 0-12 mo. 38.4 (21) Ready-to-use formulad - Formule prl!te a l'emploid 1987 0-12 mo. 11.5 (21) Concentrated formulae - Formule concentreec 1987 0-12 mo. 13.7 (21) Powdered formulae - Preparation en poudrec 1987 0-12 mo. 16.5 Selected foods - Chaix d'aliments 1987 0-12 mo. 16.5 Market basket - Panier de la menagere 1988 0-1 mo. 3.4 (22) Breast-fed - Alimentation au sein Cuba 1984 1-4 yr. 5.6 6.6 11.0 Urban - Milieu urbain 1985 3-6 mo. 7.5 7.6 13.7 Urban - Milieu urbain 1985 9-14 mo. 8.2 7.2 14.2 Urban - Milieu urbain Finland - Finlande 1980 3 yr. 22.4 (BJ Germany - Allemagne 1980 1 mo. 18.2 Regular water - Eau normale 1980 1 mo. 106.2 High-lead - Eau a forte teneur en plomb 1980 1 mo. 18.2 Breast-fed - Alimentation au sein Guatemala 1980 3 mo. 2.1 Hungary - Hongrie 1983 3 mo. 11.6 Niger 1983 3 mo. 3.6 Philippines 1983 3 mo. 12.8 Poland - Pologne 1984 1-3 yr. 5.4 33.4 96.6 Industrial area - Zone industrielle 1984 1-3 yr. 2.7 17.8 61.3 Non-industrial area - Zone non industrielle 1985 1-3 yr. 14.6 32.3 85.4 Industrial area - Zone industrielle 1984-85 1-3 yr. 20.0 (14) Wroclaw Sweden - Suede 1979 3 mo. 1.7 (23) Breast-fed - Alimentation au sein 1983 3 mo. 14.1 (24) Switzerland - Suisse 1982 10 mo. 23.5 (25) UK - Royaume-Uni 1981 <4 mo. 226.1 361.2 807.8 (19) High-lead water, Ayr - Eau a forte teneur en plomb, Ayr 1982 3.7 yr. 6.0 7.3 15.9 1982 4 yr. 5.6 6.8 16.1 Asian non-vegetarian - Sujets asiatiques non vegetariens 1982 3.6 yr. 8.3 8.9 15.4 Caucasian non-vegetarian - Blancs non vegetariens 1982 3.4 yr. 6.36 6.83 12.6 1985 2 yr. 12.4 15.3 24.9 (26, 27) Birmingham USA - Etats-Unis d'Amerique 1980 6mo. 29.0 1982 6mo. 16.8 1985 6-11 mo. 12.2 New study design - Nouveau plan d'etude 1986 6-11 mo. 5.0 1987 6-11 mo. 3.6 1986-88 6-11 mo. 3.2 Zaire - Zaire 1983 3 mo. 2.7 • FAO/WHO provisional tolerable weekly intake: 25 µg/kg body weight. - Dose ingeree hebdomadaire tolerable provisoire FAO/OMS: 25 µg/kg de poids corporel. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. c Non-lead-soldered cans. - Boiles de conserves non soudees au plomb. d Lead-soldered cans. - Baites de conserves soudees au plomb. 136 Rapp. trimest. statist. sanit. mond., 50 (1997) Table 3 Dietary intake of cadmium by adults a Tableau 3 Doses de cadmium ingerees avec les aliments par les adultesa Country/area - Year- Median- Mean- 90th percentile - Reference b/remarks- Pays/territoire An nee Mediane Moyenne 9Qe centile ReMrence b/remarques Australia - Australie 1983 0.8 1984 1.4 1985 1.0 3.6 1987 2.1 2.7 3.6 95th percentile consumer - Consommateurs du 95e centile Belgium - Belgique 1980 5.4 1982 1.8 2.1 (3) Brussels, Charleroi, Liege, 2 % > PTWI - Bruxelles, Charleroi, Liege, 2 % > dose ingeree tolerable provisoire Canada 1981 1.3 1.5 3.5 GEMS/Food; (4) China - Chine 1988 1.0 1.0 1.0 (5) Beijingc - Pekin c Cuba 1984 1.9 1.4 3.7 GEMS/Food; (28). Havana - La Havane Denmark - Danemark 1980 3.2 7.5 1980 4.1 Near lead smelter - A proximite de fonderies de plomb 1985 2.2 2.5 3.6 1987 2.0 Finland - Finlande 1980 1.5 (6) 1981 1.4 1986 0.8 (29) High intakes with liver or wild mushrooms in meals - Fortes doses ingerees avec des repas a base de foie ou de champignons 1986 1.1 1984-88 1.2 (7) Market basket - Panier de la menagere 1984-88 1.2 (7) Duplicate diet - Rations en double France 1983 3.5 Germany - Allemagne 1982 3.5 GEMS/Food; (8) 1987 2.7 5.6 13.4 Guatemala 1982 4.6 4.8 7.4 Urban - Zone urbaine 1984 1.8 1.8 2.4 Urban - Zone urbaine 1985 2.2 3.2 4.3 Urban - Zone urbaine 1988 2.8 3.7 7.4 Urban - Zone urbaine Hungary - Hongrie 1980 0.8 0.9 1.5 1982 0.4 0.4 0.7 1983 0.6 0.6 1.0 1984 0.5 0.6 1.0 1985 0.9 Italy - ltalie 1981 7.8 Rural families - Families rurales 1982 6.4 Rural families - Families rurales 1986 4.7 4.7 1987 3.4 Japan - Japon 1980 4.8 4.9 8.0 1981 4.2 4.3 5.7 1981 5.6e (30) Rural families - Families rurales 1981 5.2e (31) Non-polluted areas - Zones non polluees 1982 3.8 4.5 6.0 1983 3.9 4.1 5.5 1984 4.3 4.9 6.1 1985 3.9 3.7 4.7 1986 2.9 4.1 6.1 1987 2.8 3.4 4.2 Korea, Rep. of - Coree, Republique de 1985 2.4e (11) Rural families - Families rurales 3.4e (11) Seoul females - Sexe feminin, Seoul 2.9 9.9 Selected foods - Choix d'aliments 1985 1.1 Duplicate diet - Rations en double 1985 1.1 1.2 (12) Duplicate diet - Rations en double 1984-86 2.0 2.1 (15) Market basket - Panier de la menagere Wld hffh statist. quart., so (1997) 137 Table 3 (continued) Tableau 3 (suite) Country/area - Year- Median- Mean- 90th percentile - Reference b/remarks - Pays/territoire An nee Mediane Moyenne 90• centile Reference b/remarques 1988 2.3 Selected foods - Choix d'aliments 1984-88 2.3 (7) Market basket - Panier de la menagere New Zealand - Nouvelle-Zelande 1982 5.9 Poland - Pologne 1982 8.8 5.3 94.5 Industrial area - Zone industrielle 1982 0.6 4.4 87.5 Non-industrial area - Zone non industrielle 1983 2.9 5.1 23.1 Industrial area - Zone industrielle 1983 0.6 2.4 8.1 Non-industrial area - Zone non industrielle 1981-83 2.2 (14) Lublin, Wroclaw 1987 3.8 (15) Lublin, Poznan, Olzstyn Sweden - Suede 1980 1.0 GEMS/FOOD; {'16) 1988 1.2 (9) Average person; market basket - lndividu moyen; panier de la menagere 1988 0.9 1.0 1.6 (5, 18) Stockholmc Switzerland - Suisse 1984-88 1.8 (1) Turkey- Turquie 1984-88 0.8 (7) Rural - Zone rurale UK - Royaume-Uni 1980 2.1 2.8 3.2 1981 1.9 1.9 2.2 1983 1.8 1.9 2.1 1985 1.8 1.9 2.1 1987 2.2 2.2 2.5 1988 1.9 2.2 2.3 USA - Etats-Unis d'Amerique 1980 2.6 2.8 4.4 1982 2.9 2.8 4.0 1985 1.9 New study design - Nouveau plan d'etude 1986 1.6 1987 1.5 1986-88 1.1 Yugoslavia - Yougoslavie 1988 1.0 1.0 (5) Zagrebc • FAQ/WHO provisional tolerable weekly intake: 7 µg/kg body weight. - Doses ingerees hebdomadaires tolerables provisoires FAQ/OMS: 7 µg/kg de poids corporel. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. c Non-smoking females. - Non-fumeurs de sexe feminin. d Geometric means. - Moyennes geometriques. Table 4 Dietary intake of cadmium by infants and childrena Tableau 4 Doses de cadmium ingerees avec les aliments par les nourrissons et les enfantsa Country/area - Pays/territoire Australia - Australie Cuba Canada 138 Year- An nee 1983 1984 1985 1987 1984-85 1986 1987 1987 Intake (µg/kg bw/wk) - Dose ingeree (µgtkg de poids corporel/semaine) Age Median- Mean- Mediane Moyenne 9mo. 1.2 9 mo. 2.1 9 mo. 1.5 9 mo. 2.2 3.0 3-6 mo. 6.3 9.8 0-1 mo. 1.1 0-12 mo. 2.4 0-12 mo. 4.4 90th Reference b/remarks- percentile- Reference b/remarques 90• centile 5.25 95th percentile consumers - Consommateurs du 95e centile 19.9 Urban - Milieu urbain (22) Breast-fed - Alimentation au sein (21 J Ready-to-use formula, milk-based - Formule prate a l'usage, a base de lait (21 J Ready-to-use formula, soy-based - Formule prate a l'usage, a base de soja Rapp. trimest. statist. sanit. mond., 50 (1997) Table 4 (continued) Tableau 4 (suite) Intake (µg/kg bw/wk) - Dose ingeree (µg/kg de poids corporel/semaine) Country/area - Year- Age Median- Mean- 90th Reference b/remarks - Paystterritoire An nee Mediane Moyenne percentile- Reference b/remarques 90" centile 1987 0-12 ma. 2.8 Market basket - Panier de la menagere 1987 0-12 ma. 2.9 Selected foods - Chaix d'aliments Finland - Finlande 1980 3 yr. 3.9 (6) Germany - Allemagne 1980 1 ma. 3.2 Regular water - Eau normale 1980 1 ma. 7.6 High-cadmium water - Eau a forte teneur en cadmium 1980 1 ma. 3.2 Breast-fed - Alimentation au sein Niger 1983 3 ma. 2.8 Philippines 1983 3 ma. 2.0 Poland - Pologne 1984-85 1-3 yr. 4.4 (14) Wroclaw 1984 1-3 yr. 2.7 6.5 14.5 Industrial area - Zone industrielle 1984 1-3 yr. 1.3 2.7 9.7 Non-industrial area - Zone non industrielle 1985 1-3 yr. 2.7 7.5 15.6 Industrial and non-industrial areas - Zones industrielles et non industrielles Sweden - Suede 1983 3 ma. 0.1 (23) Breast-fed - Alimentation au sein UK - Royaume-Uni 1982 3.7 yr. 2.0 2.0 2.7 1982 4 yr. 1.9 1.9 3.1 Asian non-vegetarian - Asiatiques non vegetariens 1982 3-6 yr. 1.8 2.4 3.6 Caucasian non-vegetarian - Blancs non vegetariens 1982 3-4 yr. 2.5 2.5 3.8 1985 2 yr. 2.7 2.9 3.7 (26) Birmingham USA - Etats-Unis d'Amerique 1980 6mo. 9.8 1982 6mo. 9.4 1985 6-11 ma. 3.5 New study design - Nouveau plan d'etude 1986 6-11 ma. 2.9 1987 6-11 ma. 2.2 1986-88 6-11 ma. 2.3 • FAQ/WHO provisional tolerable weekly intake: 7 µg/kg body weight. - Doses ingerees hebdomadaires tolerables provisoires FAQ/OMS: 7 µg/kg de poids corporel. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. Table 5 Dietary intake of mercury by adultsa Tableau 5 Doses de mercure ingerees avec les aliments par les adultesa Country/area - Year- Paystterritoire An nee Australia - Australie 1984 Belgium - Belgique 1982 Cuba 1983 Denmark - Danemark 1985 Finland - Finlande 1980 1984-88 1984-88 Wld hlth statist. quart., 5D (1997) Intake (µgtkg bw/wk) - Dose ingeree (µg/kg de poids corporel/semaine) Median- Mean- 90th percentile - Mediane Moyenne 90• centile 0.3 0.8 1.6 1.6 5.6C 1.6 1.9 0.4 0.23 0.27 Reference b/remarks - Reference b/remarques (3) Brussels, Charleroi, Liege, 2%>PTW1 - Bruxelles, Charleroi, Liege, 2%> dose ingeree tolerable provisoire (6) (7) Market basket - Panier de la menagere (7) Duplicate diet - Rations en double 139 Table 5 (continued) Tableau 5 (suite) Intake (µg/kg bw/wk) - Dose ingeree (µg/kg de poids corporeVsemaine) Country/area - Year- Median- Mean- 90th percentile - Reference b/remarks- PayS/territoire An nee Mediane Moyenne 9()e centile Reference b/remarques France 1980 1.2 Germany - Allemagne 1981 2.3 2.6 3.9 1982 1.6 (8) Guatemala 1988 .68 1.5 Italy - ltalie 1982 1.3 Netherlands - Pays-Bas 1985 0.2 (12) 1985 0.2 1986 0.07 1986 0.06 0.07 (13) 1988 1.1 New Zealand - Nouvelle Zelande 1982 0.6 (32) Poland - Pologne 1981-83 1.0 (14) Lublin, Wroclaw 1987 2.0 (15) Lublin, Olsztyn, Poznan Sweden - Suede 1988 0.23 Thailand - Thailande 1987 0.7 0.8 1.3 UK - Royaume-Uni 1985 0.3 (33) USA - Etats-Unis d'Amerique 1985 0.3 1986 0.2 1987 0.2 1986-88 0.3 a FAO/WHO provisional tolerable weekly intake: total mercury- 5 µgtkg body weight methylmercury- 3.3 µg/kg body weight - Doses ingerees hebdomadaires tolerables provisoires FAQ/OMS: mercure total 5 µg/kg de poids corporel methylmercure - 3,3 µg/kg de poids corporel. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. c Maximum Guatemala, Belgium, and Cuba, were around 50% of the PTWI for methylmercury. The lowest aver- age intakes were reported from Finland, the United Kingdom, the United States and Sweden. The data available concerning weekly total mer- cury intakes by infants and children are given in Tab/,e 6 and are within the PTWI. In a study of breast milk in Sweden, mainly among fishermen's wives who consumed relatively large amounts of fish, intake levels for breast-feeding infants were about 50% of the PTWI. Because of concern for the toxic effects of methylmercury on the develop- ing foetus, Sweden has advised pregnant women to avoid consuming too much fish containing high levels of mercury. In the data reviewed from 16 countries, the average adult intake did not exceed the PTWI in any instance. However, in several countries, it amounted to an appreciable portion of the PTWI. In view of these results, dietary intake studies should be continued in order to follow up on re- sults obtained to date, and should be carried out in additional countries. Most of the studies reviewed have been carried out in industrialized countries. In the results to 140 date, there did not appear to be any correlation with the state of development of the country, but not enough data were available from developing countries to make a valid comparison. Not many results were available concerning dietary intakes of mercury by infants and young children, the age group most at risk. However, in all cases the intakes were equal to or higher than for adults. Represen- tative studies on the dietary intakes of infants and children should be carried out in more countries. In addition, well-designed duplicate portion stud- ies should be conducted in more countries with targeted groups, such as infants, children, preg- nant and nursing women, those who consume greater amounts of fish and those living near areas of metal pollution. Dietary intake of polychlorinated biphenyls The available data from 9 countries on mean daily dietary intakes of PCBs by adults are summarized in Tab/,e 7. Since fish generally contain higher levels of PCBs than any other food category, diets contain- ing higher amounts of fish may be expected to lead to higher PCB intakes. Only Japan and the United Rapp. trimest. statist. sanit. mond., 50 (1997) Table 6 Dietary intake of total mercury by infants and childrena Tableau 6 Doses de mercure totales ingerees par les nourrissons et les enfants a Country/area - Pays/territoire Australia -Australie Finland - Rnlande Poland - Pologne Sweden - Suede USA - Etats-Unis d'Amerique USA - Etats-Unis d'Amerique Zaire - Za"ire Weekly intake (µgtkg bw/wk) - Dose ingeree (µg/kg de poids corporel/semaine) Year- Age group- Mean- An nee Groupe d'age Moyenne 1984 9mo. o.02c 1984 2 yr. 0.5 c 1980 3 yr. 1.5 1985 3-7 yr. 2.5 1985 1-3 yr. 3.1 1988 3mo. 2.6 1985 6-11 mo. 0.3 1986 6-11 mo. 0.2 1987 6-11 mo. 0.3 1986-88 6-11 mo. 0.4 1985 2 yr 0.6 1986 2 yr. 0.4 1987 2 yr. 0.5 1986-88 2 yr. 0.5 1983 3mo. 1.7 c (6) (14) Reference b/remarks- Reterence b/remarques (34) Among high fish consumers - Parmi les gros consommateurs de poisson • FAD/WHO provisional tolerable weekly intake: 5.0 µg/kg body weight. - Doses ingerees hebdomadaires tolerables provisoires FAD/OMS: 5.0 µg/kg de poids corporel. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. c Median - Mediane. Table 7 Dietary intake of polychlorinated biphenyls by adultsa Tableau 7 Doses de biphenyls polychlores ingerees par les adultes a Country/area - PayS/territoire Australia -Australie Finland - Finlande Guatemala Japan - Japon Netherlands - Pays-Bas New Zealand - Nouvelle-Zelande Switzerland - Suisse United Kingdom - Royaume-Uni USA- Etats-Unis d'Amerique Year- Annee 1987 1984 1986 1988 1980 1982 1984 1986 1988 1984 1982 1983 1981 1985 1980 1982 1985 1986 1988 Mean daily intake (µg/kg bw/wk) - Dose ingeree journaliere moyenne (µg/kg de poids corporel) 0.002 0.21 0.026 0.012 0.046 0.046 0.042 0.030 0.045 0.2 0.9 0.12 0.0005 Not detected - Non decele 0.0075 0.0029 0.0006 0.0025 0.001 Referenceb/Remarks- Referenceb/remarques 95th percentile consumers - Consommateurs du 95e centile (35) (36) • U.S. FDA suggested consumption maximum: 1 µg/kg body weighl/day. - Doses maximales proposees par la FDA des Etats-Unis: 1 µg/kg de poids corporel par iour. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. Wld hlth statist. quart., 50 (1997) 141 States reported intakes over several years. Both countries reported mean intakes of less than 0.05 µg/kg body weight/day over the 1980-1988 period. The United States intakes were an order of magnitude lower than those of Japan, probably because of the lower amount of fish in the United States diet. Substantially higher daily intakes in New Zealand were due primarily to the high PCB intake from dairy products. In this case, the mean intake of 0.9 µg/kg body weight/day approaches the FDA-suggested maximum of 1 g/kg body weight, while the daily mean intake of 1.5 µg/kg body weight for male teens exceeds it. Intakes reported from Australia and the United States for infants and young children were in the same range as those reported for adults. The aver- age daily intakes of PCBs for infants, as estimated from levels in human milk reported from 8 coun- tries, are presented in Tabl,e 8. The data from Den- mark, Germany, Japan and the United Kingdom were submitted to GEMS/Food. The estimated in- takes reported from all countries are above the U.S. Food and Drug Administration-suggested con- sumption maximum of 1 µg/kg body weight, ex- cept for India where the use of PCBs is greatly restricted. A mean intake of about 13 µg/kg body weight was calculated from data reported from PCB con- tent in the breast milk of Inuit women from the Hudson Bay region of Northern Quebec. This high level in breast milk is ascribed to the markedly higher consumption of fish and marine mammals. Table 8 Dietary intake of PCBs by infants from human milka Tableau 8 However, calculated mean intakes above 10 µg/kg body weight were also noted in Germany. In this connection, it should be noted that high PCB levels have been reported in fish and meat in cer- tain areas of Germany, indicating localized high levels of environmental contamination. While the available information on dietary in- takes of PCBs by adults is almost exclusively from industrialized countries, the intakes are usually well below the U.S. FDA guidance figure of 1 µg/kg body weight per day. Monitoring of PCBs in animal-derived foods, especially fish, should be carried out in all countries to determine possible sources of this contaminant in the diet, and dietary intake studies should be carried out, including areas of probable high intakes of PCB-contami- nated food. The estimated intake of PCBs by the breast-fed infant was usually far in excess of this guidance value in virtually all reporting nations. In view of the high levels of PCBs found in human milk, surveys of this food should be conducted on a more regular basis in both industrialized and de- veloping countries. Dietary intake of pesticides Data on levels of residues of organochlorine and organophosphorous pesticides have been submit- ted to GEMS/Food for a wide variety of foods since 1980. Of the total of twenty countries submitting data, few submitted data on all the pesticides re- quested and fewer still on all the requested food Doses de PCB d'origine alimentaire ingerees par des nourrissons avec le lait materne1a Country/area - Pays/territoire Canada Denmark - Danemark Finland - Finlande Germany - Allemagne Hong Kong India- lnde Japan - Japon UK - Royaume-Uni Year- Annee 1988 1988 1982 1982 1982 1983 1985 1982 1980 1981 1982 1983 1984 1985 1980 Mean daily intake (µg/kg bw/wk) - Dose ingeree journaliere moyenne (µg/kg de poids corporel) 13.32 3.12 3.32 1.92 10.5 11.9 2.19 Not detected - Non decele 2.28 2.40 2.52 2.52 2.16 1.80 1.86 (37) Inuit Referenceb/Remarks - Referenceb/remarques (37) caucasians - Blancs (38) Helsinki (39) Ethnic Chinese (40) Ahmedabab • U.S FDA suggested consumption maximum: 1 µg/kg body weighVday. - Doses maximales proposees par la FDA des Etats-Unis: 1 µg/kg de poids corporel par jour. b GEMS/Food unless otherwise referenced. - Reference GEMS/Food sauf indication contraire. 142 Rapp. trimest. statist. sanit. mond., 50 (1997) groups. Certainly, one major reason for this is the non-usage or, in some cases, trivial usage of pesti- cides monitored by GEMS/Food in many coun- tries. On the other hand, participation of develop- ing countries is very low, even though in recent years they have been the main users of organochlo- rine pesticides, which have been banned in many of the industrialized countries. Data have been sub- mitted to GEMS/Food by even fewer countries on dietary intakes of pesticides. For example, only9 of the 20 countries providing food data to GEMS/ Food have also submitted dietary intake data. Nonetheless, the available information does offer some indication of the prevailing situation in cer- tain parts of the world. Organoch/orine pesticides In many countries, restrictions have been placed on the use of organochlorine pesticides for non- food as well as food applications, because of persis- tence and potential for bioaccummulation in the food chain. In foods, they occur primarily in milk and dairy products, eggs, meat and animal fat, and fish. Even when they are used only in non-food applications, they can enter the food supply not only by direct drift onto crops, but they can ulti- mately contaminate animal-derived foods because of their chemical stability in the soil and waterways. GEMS/Food data on dietary intakes of aldrin and dieldrin, DDT-complex, total HCH isomers, gamma-HCH (lindane), hexachlorbenzene, hepta- chlor and heptachlor epoxide, endosulfan and en- drin are largely limited to developed countries where exposure for the adult populations is gener- ally very much below the respective AD Is. The lim- ited data available from developing countries indi- cate a higher average exposure for adults, but, with a few exceptions, still within the ADis. However, based on the levels in breast milk, a significant portion of infants in both developed and develop- ing countries are exposed to levels of organochlo- rine pesticides above the respective ADis. While the ADI was intended as a level which could be consumed safely over a lifetime, excessive expo- sure of infants and children to these substances is a concern. Consequently, additional research is re- quired to better characterize the risks of these pes- ticides from both toxicological and exposure per- spectives. Organophosphorus pesticides Diazinon, fenitrothion, malathion, parathion and parathion-methyl are included in the GEMS/Food Programme, but have been reported by only a few countries. Almost all reports indicate that, with few exceptions, exposures are only a few percent of the corresponding ADis. Among the exceptions are reports oflevels of fenitrothion reaching about 45 % of the diet of adults and children in Australia ( 41 ). Wld hlth statist. quart., 50 (1997) However, acute poisonings by organophospho- rus pesticides are by far the leading cause of pesti- cide-associated deaths, especially in developing countries. For example, China reported 42 812 cases of pesticide poisoning, including 3 900 fatal- ities in 1994. About 80% of the cases were caused by organophosphorus pesticides, and in particular, parathion (52%) and methamidophos (14%) (42). Dietary intake of aflatoxin Aflatoxin has been reported to occur in a wide variety of food commodities ( 4 3 ). Aflatoxin B1, B2, Gi' G2, and M1 are the most common forms of aflatoxin, but aflatoxin B1 is considered to be the most toxic. In plant products, aflatoxin contamina- tion occurs directly as a result of fungal invasion mainly by Aspergi,llus flavus and A. parasiticus. Afla- toxin, mainly in the form of its metabolites, can occur in animal products, such as milk, meat and eggs, as a result of feeding contaminated feed to animals (43). A global survey carried out in differ- ent parts of the world revealed that among the commodities, groundnut and maize are the most frequently contaminated foods (44). Aflatoxin ex- posure in human populations has been assessed using either analysis of foods and the total diet for aflatoxin or measurement of aflatoxin biomarkers in human biological fluids like urine, serum and breast milk. High levels of aflatoxin have been reported in these commodities particularly in regions of Africa, South East Asia and Southern China (45). Health effects of dietary exposure to atlatoxin Dietary aflatoxin exposure has been examined in areas of sub-Saharan Africa and South-East Asia where commodities like maize and groundnuts are consumed as dietary staples. Epidemiological in- vestigations in these areas revealed that aflatoxin intake is associated with a number of acute and chronic liver diseases in the populations. Acute aflatoxin toxicity incidents in human populations have been rare and occurred only under stress conditions like unseasonal rains which damaged food crops and made them susceptible to heavy fungal contamination. The acute diseases mostly in the form of acute hepatitis and cirrhosis were asso- ciated with consumption of foods with very high levels of aflatoxin in the parts per million range (Tab/,e 9). Of the outbreaks with acute toxicities suggestive of aflatoxin, the most convincing evi- dence of aflatoxin involvement came from an acute hepatitis outbreak in western parts of India (50). More than 400 people were affected and more than 100 people died in the outbreak. The implicated food was a recently harvested, badly stored maize which had been affected by un- seasonal rains. Analysis of the maize samples showed heavy Aspergi,llus jlavus infection and high 143 Table 9 Reported exposure to aflatoxins in acute liver diseases in human populations in different countries Tableau 9 Exposition aux aflatoxines dans des cas d'affections hepatiques aigues chez des populations de differents pays Country/area - No. of Nature of toxicity Pays/territoire cases- (as reported) - Nombre Nature de la toxicite (telle de cas qu'indiquee dans le rapport) Senegal - Senegal 2 Hepatitis - Hepatite Uganda - Ouganda Acute hepatitis - Hepatite aigue China (Prov. of 26 Acute liver disease - Taiwan) - Chine Affection hepatique aigue (province de Ta'iwan) India- lnde 20 Indian childhood cirrhosis - Cirrhose indienne de l'enfant India- lnde 400 Acute hepatitis - Hepatite aigue Kenya - Kenya Acute hepatitis - Hepatite aigue Malaysia - Malaisie 13 Acute hepatic and renal failure - Affection hepatique aigue et insuffisance renale aflatoxin levels ranging from 0.25-15.6 mg/kg. In- dividual aflatoxin intakes were calculated to be 2-6 mg/day (up to 120 µg/kg body weights) over a period of one month. Exposure to aflatoxin has been implicated in diseases like liver cirrhosis, kwashiorkor and Reye's syndrome. In individuals suffering from these diseases, aflatoxin, mostly afla- toxin B1, B2, M1 and aflatoxicol, have been de- tected in liver biopsy, autopsied tissue samples and urine samples (53-55). Concern for effects of aflatoxin in humans is increasing in recent times owing to its hepatocarci- nogenic properties. IARC concluded that there was sufficient evidence of carcinogenicity in hu- mans for aflatoxin B1 and classified it as a Group 1 carcinogen ( 4 5 ). Dietary aflatoxin exposure has been examined in areas of the world where pri- mary liver cancer (PLC) is endemic and where aflatoxin contamination of foods is high (Table 10). It has been also assessed in normal populations exposed to high risk commodities in order to esti- mate the extent of risk. Epidemiological surveys in countries of sub-Saharan Africa and South-East Asia showed a positive correlation between dietary aflatoxin intake and liver cancer incidence ( 58, 68). The major source of aflatoxin exposure in the diet was due to maize, groundnuts and some- times rice. The frequency and level of aflatoxin contamination in these foods varied according to region, altitude and season. Regions with high rainfall showed higher levels of contamination. This was particularly observed in rice which showed high aflatoxin levels in rain-affected re- 144 Food implicated - Aflatoxin levels Dietary aflatoxin Reference- Aliments incrimines in food- intake- Reference Concentration Dose d 'aflatoxine d'aflatoxine ingeree par la dans !'aliment voie alimentaire (mg/kg) (µg/day- jour) Groundnut meal - 0.5 - 1.0 35-140 (46) Farine d'arachide Cassava - Manioc 1.7 100 (47) Rice-Riz 0.2 Not reported - (48) Non indiquee Groundnut meal - 0.3 9 - 18 (49) Farine d'arachide Maize-Mais 0.25 - 15.6 2000-6000 (50) Maize-Mais 1.6 - 12.0 Not reported - (51) Non indiquee Chinese noodles - Not reported - Not reported - (52) Nouilles chinoises Non indiquee Non indiquee gions of Thailand and India (64). Higher levels of contamination were observed during the rainy season when compared to the dry season. In The Gambia, the intakes were higher during the dry season when high aflatoxin levels occurred in groundnuts stored for consumption (63). The high prevalence of hepatitis B virus infection in these areas is also considered to be a risk factor for Pri- mary Liver Cell Cancer (PLC) (69, 70). Correlation studies in PLC endemic areas addressing com- bined exposure to hepatitis B virus and aflatoxin concluded that aflatoxin exposure was a more im- portant determinant of PLC incidence than hepa- titis B virus infection (59, 66). Presence of aflatoxin metabolites in human biological fluids Exposure to aflatoxin leads to their absorption, metabolic activation and subsequent covalent binding to cellular DNA and protein. The afla- toxins are primarily metabolized by the microsom- al mixed function oxygenase system to a number of hydroxylated derivatives as well as an unstable highly reactive epoxide. This epoxide binds co- valently to nucleic acids and proteins to exert the toxic and carcinogenic action of aflatoxin (71). A number of aflatoxin metabolites have been de- tected in urine, milk and serum of human popula- tions exposed to dietary aflatoxin. Among these metabolites, aflatoxin-DNA and -protein adducts and aflatoxin M1 are considered as biomarkers of aflatoxin exposure. Their measurement in urine Rapp. trimest. statist. sanit. mond., 50 (1997) Table 10 Dietary aflatoxin exposure in human populations in different countries Tableau 10 Exposition des populations aux aflatoxines alimentaires dans differents pays Country/area - Period of Type of foods sampled - Aflatoxin levels Estimated aflatoxin Reference- PayS/territoire sampling- Type d'aliments echantillonnes in food•- exposure Reference Periode Concentrations (ng/kg bw/day) - d'echantillonnage d'aflatoxine dans Exposition estimative les aliments• aux aflatoxines (µg/kg) (ng/kg de poids corporel par jour) Africa - Afrique Kenya 1969 Prepared foods and beer - 0.1 • 0.4 4 · 12 (56) Preparations alimentaires et biere (81 + 82) (81 + 82) Swaziland 1972-1973 Cooked diet - Aliments cuisines 6 · 53 (57) Not reported - Non indiquee Swaziland 1982-1983 Cooked foods and beer - Aliments 11 • 700 3000 · 18 OOO (59) cuisines et biere (Total AF)- µg/day - jour (aflatoxines (Total AF) totales) Mozambique 1969-1974 Cooked diet - Aliments cuisines 8 (Total AF) - 220 (60) ( aflatoxi nes (Total AF) totales) Transkei 1976-1977 Cooked diet - Aliments cuisines 17 (81) (61) Nigeria - Nigeria 15 • 40 (81) (62) Gambia - Gambie 1988 Cooked foods - Aliments cuisines 1·>100(8,) 4 • 120 (8 1) (63) Asia-Asie Thailand - Thailande 1969-1970 Cooked diet - Aliments cuisines Trace - > 1000 (81) 5-55(8,) (64) Philippines Market foods - Produits du marche 0.6 • 470 (81) 22 OOO (65) µg/day- jour (81) Southern China - Chine du sud 1978-1984 Maize - Mais 12 · 2 OOO (81) (66) lndia-lnde 1990 Cooked maize - Mais cuit 1-9(8,) 4 · 100 (81) (68) USA - Etats-Unis d'Amerique 1960-1979 Market foods - Produits du marche 2.7 (81) (45) • Type of aflatoxin is given in parentheses. - Type d'aflatoxine entre parentheses. and serum using ELISA or HPLC techniques has been shown to reflect DNA damage by aflatoxin and to be directly related to the dietary intake (74, 77). These observations led to a better under- standing and assessment of the relative roles of hepatitis B viral infection and aflatoxin in the etiol- ogy of PLC. The level of urinary adducts reflects recent di- etary exposure to aflatoxin (the previous 24-48 hours) whereas the level of aflatoxin-albumin ad- ducts reflects aflatoxin exposure over the previous 2-3 months ( 45). Studies on levels of aflatoxin- albumin adducts in subjects from several countries in Africa, South East Asia and Europe showed quantitative differences (Tab/,e 11). Exposure was significantly higher in regions of Africa and South- ern China than in the European region (76). The levels of aflatoxin-albumin adducts in sera of popu- lations in different regions surveyed ranged from <5 pg to 720 pg aflatoxin B1-lysine equivalents per mg of albumin. Wld hlth statist. quart., 50 (1997) Aflatoxin has been shown to cross the placental barrier in humans. Their detection in cord sera has been reported from Thailand, Ghana, Nigeria and Gambia (Tab/,e 12). Several studies have demon- strated the presence of aflatoxin M1 in human milk (Tab/,e 13). It was observed that only a small percent- age (0.09-0.43%) of dietary aflatoxin intake was excreted in milk as aflatoxin M1• Because of its importance for health and trade, aflatoxin contamination of maize, groundnuts and rice as well as other affected foods, such as tree nuts and figs, should continue to be a priority monitoring activity. When levels in staple foods are high, biomonitoring should be undertaken to provide an assessment of the integrated exposure to aflatoxin in the diet. As currently recommend- ed by JECF A, the levels of aflatoxins in food should be as low as feasible and interventions during production, harvest and storage of food should be considered to prevent or reduce con- tamination. 145 Table 11 Aflatoxin-albumin adducts in human seraa Tableau 11 Adduits aflatoxine-albumine presents dans les serums humainsa <5b 5-25 Country/area (n)- Pays/territoires (n) Gambia - Gambie May - Mai (323) 7 53 November - Novembre (67) 0 39 Senegal - Senegal (29) 0 20 Kenya (91) 48 26 China - Chine Guangxi (93) 28 35 Shandong (69) 69 0 Thailand - Tha'ilande (84) 73 10 France (44) 44 0 Poland - Pologne (30) 30 0 a Source: Ref. - Ref. (75) b limit of detection. - Concentration minimale detectable. Table 12 In utero exposure to aflatoxins Tableau 12 Exposition aux aflatoxines in utero Country/area - Pays/territoires Thailand - Tha'ilande Ghana Nigeria - Nigeria Gambia - Gambie Table 13 Type of sample - Type d'echantillons Cord blood - Sang du cordon Maternal blood - Sang maternel Cord blood - Sang du cordon Cord blood - Sang du cordon Maternal blood - Sang maternel Cord blood - Sang du cordon Maternal blood - Sang maternel Aflatoxin M1 detected in human milk in different countries Tableau 13 Adduct levels - Concentrations d'adduits (pg d'AFB1 lysine eq/mg albumin/e) 26-50 51-75 76-100 Number of subjects - Nombre de sujets 76 49 40 13 7 3 6 2 1 5 1 5 13 6 2 0 0 0 1 0 0 0 0 0 0 0 0 No. of samples - No. positive (%) - Nombre Nombre d'echantillons d'echantillons posit ifs (%) 35 49 35 6 188 34 78 12 77 21 30 97 30 70 Aflatoxines M1 mises en evidence dans du lait humain de differents pays Country/area - PayS/territoires Zimbabwe Gambia - Gambie Ghana Sudan - Soudan Thailand - Tha'flande Victoria, Australia - Australie No. of samples - Nombre d'echantillons 54 5 264 11 73 No. positive(%)- Nombre d'echantillons positifs (%) 11 100 22 45 15 >100 98 5 0 6 9 0 0 0 0 Reference- Reference (77) (78) (62) (79) Reference -Reference (45) (80) (78) (81) (82) (82) 146 Rapp. trimest. statist. sanit. mond., 50 (1997) Conclusions An assessment of the results of GEMS/Food and other monitoring activities generally confirms the effectiveness of government efforts to prevent or reduce food contamination in industrialized coun- tries. However, it is also recognized that better exposure estimates for infants and children and other vulnerable groups should be conducted. In developing countries, monitoring informa- tion is generally lacking and the health of hun- dreds of millions of people may be threatened. For these countries, and especially those that employ older agricultural and industrial technologies, ba- sic food contamination monitoring and assessment programmes should be established for at least those contaminants of priority concern. In addi- tion, these programmes can supply the essential information necessary for developing effective in- tervention strategies and for efficient management of health and environment resources. In all countries, accidental and sporadic con- tamination is an ever present danger and continual vigilance is necessary to safeguard public health. All countries should identify institutions with the analytical capability to support epidemiological in- vestigations of outbreaks of disease that may be associated with consumption of chemical contami- nants in food. All countries should participate in GEMS/Food to promote health-oriented, popula- tion-based monitoring of food contaminants in their countries. Summary The results of the Global Environment Monitoring Sys- tem I Food Contamination Monitoring and Assessment Programme (GEMS/Food) and other monitoring pro- grammes for priority contaminants in the diet, including lead, cadmium, mercury, polychlorinated biphenyls, organochlorine and organophosphorus pesticide resi- dues and aflatoxin, are presented. These results are assessed with respect to established acc·eptable or tolerable intakes for these contaminants. While the as- sessments generally confirm the effectiveness of gov- ernment efforts to prevent or reduce food contamination in industrialized countries, better exposure estimates for infants and children and other vulnerable groups should be calculated. In developing countries, little reliable information is available on the occurrence of food contamination. Without such information, the health of hundreds of millions of people may be threatened. For these coun- tries, and especially those that employ older agricultural and industrial technologies, basic food contamination monitoring and assessment programmes should be Wld hlth statist. quart., 50 (1997) established for at least those contaminants of priority concern. These programmes form an essential basis for developing effective intervention strategies and for efficient management of health and environment resources. In all countries, accidental and sporadic contamination is an ever present danger and continual vigilance is necessary to protect public health. All countries should identify institutions with the analytical capability to sup- port epidemiological investigations of outbreaks of dis- ease that may be associated with consumption of chem- ically contaminated food. All countries should partici- pate in GEMS/Food to promote health-oriented, popula- tion-based monitoring at the national level. Resume Suivi et evaluation de /'exposition d'origine alimentaire aux contaminants chimiques L'article expose les resultats obtenus par le systeme mondial de surveillance continue de l'environnement, par le programme de surveillance continue et d'evalua- tion de la contamination alimentaire (GEMS/Food) et autres programmes de surveillance, relatifs aux conta- minants alimentaires prioritaires et, en particulier, au plomb, au cadmium, au mercure, aux biphenyles poly- chlores, aux residus de pesticides organochlores et organophosphores, ainsi qu'a l'aflatoxine. Ces resultats sont evalues compte tenu des doses acceptables ou tolerables fixees pour ces contaminants. Ces evalua- tions confirment generalement l'efficacite des efforts deployes par les pouvoirs publics des pays industriali- ses pour prevenir ou reduire la contamination alimen- taire, mais ii conviendrait de proceder a de meilleures estimations de !'exposition des nourrissons et des en- fants ainsi que des autres groupes vulnerables. En ce qui concerne les pays en developpement. on ne possede guere de renseignements fiables, et des cen- taines de millions de personnes peuvent etre menacees sans qu'on le sache. Dans ces pays, et notamment ceux qui ant recours a des techniques agricoles et industriel- les anciennes, ii faudrait mettre en place des program- mes de surveillance et d'evaluation de la contamination des denrees alimentaires de base, au mains en ce qui concerne les contaminants prioritaires. Ces program- mes sont essentiels pour la mise au point de strategies d'interventions efficaces et la bonne gestion des res- sources sanitaires et environnementales. Dans tous les pays, la contamination accidentelle ou sporadique des aliments constitue un danger de cha- que instant. et une vigilance sans faille est necessaire pour proteger la sante puolique Taus les pays de- vraient recenser les etablissements capables d'effec- tuer des analyses a l'appui des enquetes epidemiologi- ques portant sur des flambees de maladies suscepti- bles d'etre imputables a la consommation d'aliments contamines par des produits chimiques. lls devraient egalement tous participer au GEMS/Food afin d'encou- rager une surveillance a vocation sanitaire au sein de la population vivant sur leurs territoires. 147 References/References I. World Health Organization. Concern for Europe's tomorrow. WHO European Centre for Environment and Health, WHO Regional Office for Europe, Wissenschaftliche Verlags- gesellshaft mbH, Stuttgart, Germany, 1995. 2. Food and Agriculture Organization of the United Nations. Risk management and food safety. Report of a Joint FAO /\VHO Consultation, 27-31 January 1997, Rome, Food and nutrition paper No. 65, FAO, Rome, 1997. 3. Buchet, J.P. et al. Oral daily intake of cadmium, lead, manganese, copper, chromium, mercury, calcium, zinc and arsenic in Belgium: a duplicate meal study. Food, chemistry and toxicowgy, 21: 19-24 (1983). 4. Dabeka, R. W. etal. Dietary intakes oflead, cadmium, arsenic and fluoride by Canadian adults: a 24-hour duplicate diet study. Food additives and contaminants, 89-102 (1987). 7 Vahter, M. &: Slorach, S. Exposure monitoring of lead and cadmium: An international pilot study within the WHO/UNEP Human Exposure Assessment Location (HEAL) Programme. "WHO and UNEP, Nairobi, 1990. 8. Weigert, P. et al. Arsenic, lead, cadmium and mercury in and on foods. Zentral Erfassungsund Bewertungstelle fiir Umwelt- chemikalien (ZEBS) - Hefte (Bundesgesundheitamt, Berlin) l: 33-36 (1984). 9. Jathar, V.S. etal. Intake oflead through food in India.journal of food sciences and technowgy, 18: 240-242 (1981). 10. Ikeda, M. et al. Dietary intake of lead among Japanese farmers. Archives of environmental health, 44: 23-29 ( 1989). 11. Watanabe, T. et al. Pb and Cd levels among Korean populations. Bulletin of environmental contamination and toxicowgy, 38: 189-195 (1987). 12. Ellen, G. et al. Dietary intakes of some essential and non- essential trace elements, nitrate, nitrite and N-nitrosoamines by Dutch adults; estimated via a 24-hour duplicate portion study. Food additives and contaminants, 7: 207-221 (1990). 13. Van Dokkum, W. et al. Minerals and trace elements in total diets in the Netherlands. British journal of nutrition, 61: 7-15 (1989). 14. Szymczak,J. etal. Cadmium, lead and mercury in daily food rations of children aged 1-7 years. Roczniki Panstwowego Zakladu Higieny, 38: 230-236 (1987). 15. Marzec, Z. &: Bulinski, R. Evaluation of cadmium, mercury and lead intake with reproduced diets. Roczniki Panstwowego Zakladu Higieny, 39: 344-348 (1988). 16. Sloracb, S. et al. Intakes of lead, cadmium and certain other metals via a typical Swedish weekly diet. Var Foda, Suppl. l: 3- 16 (1983). 17. Abdullah, M. et al. Essential and toxic inorganic elements in prepared meals- 24 hours sampling techniques. In: Bratter, P. & Shramel, P. (eds). Trace element analytical chemistry in medicine and technology, Vol. 2. Berlin, Walter de Gruyter and Co., 1983. 18. Vahter, M. et al. Dietary intake of lead and cadmium in Sweden.VarFoda. 44: Suppl. 2: 1-16 (1990). 19. Sherlock, J. et al. Assessment of lead intakes and dose- response for a population in Ayr exposed to a plumbosolvent water supply. Human toxicowgy, l: 115-122 (1982). 20. Haschke F. &: Stefan, I. Lead intake with food and young infants in the years 1980 / 1981. Wiener Klinische Wochenschrift, 93: 613-616 (1981). 21. Dabeka, R.W. et al. Survey of lead, cadmium, cobalt and nickel in infant formulas and evaporated milks and estimation of dietary intakes of the elements by infants 0-12 months old. Scienceofthetotalenvironment, 89: 279-289 (1989). 22. Dabeka, R.W. &: McKenzie, A.O. Lead and cadmium levels in commercial infant foods and dietary intake by infants 0-1 year old. Food additives and contaminants, 5: 333-342 (1988). 23. Larsson, B. et al. WHO collaborative breast feeding study II. 148 Levels of lead and cadmium in Swedish human milk. Acta pediatrica scandinavia, 70: 281-284 ( 1981). 24. Yengar, G.V. &: Parr, R.M. Trace element considerations in human milk fromseveralgwbalregions. International Workshop on composition and physiological properties of human milk, Kiel, Germany, 1985. 25. Erard, M. et al. Infants' exposure to lead cadmium and selenium by food. Transactions chimiques alimentaires hygiiniques, 73: 394-411 (1982). 26. Smart, G.A. et al. Dietary intakes oflead and other metals: a study of young children from an urban population in the UK. Food additives and contaminants, 5: 85-93 ( 1987). 27. Davies, DJ.A. et al. Lead intake and blood lead in two-year old U.K. urban children. Scienceofthetotalenvironment, 90: 13- 29 (1990). 28. Lierandi, G.B. et al. Estimation of the daily intake of cadmium which may be consumed by students 12-17 years old in secondary schools in the city of Havana. Die Nahrung, 33: 315-318 (1989). 29. Louekari, K. et al. Exposure to cadmium from foods, estimated by analysis and calculation, comparison of methods. Food additives and contaminants, 5: 111-11 7 (1987). 30. Watanabe, T. et al. Relationship of cadmium levels among blood, urine and diet in a general population. Bulletin of environmental contamination and toxicowgy, 38: 196-202 (1987). 31. Watanabe, T. et al. Dietary cadmium intakes of farmers in non polluted areas in Japan, and the relation with blood lead levels. Environmental research, 37: 33-43 (1985). 32. Pickston, N.L. et al. The New Zealand diet: a survey of elements, pesticides, colours and preservatives. New l.ealand journal of technowgy, l: 81-89 (1985). 33. Ministry of Agriculture, Fisheries and Food. Survey of mercury in food: second supplementary repart. London, 1987. 34. Skerfving, S. Mercury in women exposed to methylmercury through fish consumption, and in their newborn babies and breast milk. Bulletin of environmental contamination and toxicowgy, 41: 475-482 (1988). 35. Moilanen, R. et al. Average total dietary intakes of organochlorine compound from the Finnish diet. l.eitschrift filr Lebensmittel-Untersuchung und forschung, 182: 484-488 (1986). 36. Wuthrich, C. et al. Pesticides and other chemical residues in Swiss diet samples. Mitteilungen aus dem Gebiete der Lebensmittel- Untersuchung und hygiene, 76: 260-276 ( 1985). 37. Dewailly, E. etal. High levels of PCBs in breast milk of Inuit women from Arctic Quebec. Bulletin of environmental contamination and toxicowgy, 43: 641-646 (1989). 38. Wickstrom, K. et al. Levels of chlordane, hexachloroben- zene and DDT compounds in Finnish human milk in 1982. Bulletin of environmental contamination and toxicology, 31: 251- 256 (1983). 39. Ip, H.M.H. &: Phillips, D.J.H. Organochlorine residues in human breast milk in Hong Kong. Archives of environmental toxicowgy, 18: 490-494 (1989). 40. Jani, J.P. etal. Levels of organochlorine pesticides in human milk in Ahmedabad, India. International archives of environ- mental health, 60: 111-113 (1988). 41. National Health and Medical Research Council. The 1987 market basket survey. Canberra, Australian Government Publishing Service, 1-84 (1990). 42. World Health Organization. The Newsletter of the International Programme on Chemical Safety, Issue 8, \\-BO, Geneva, December 1995. 43. Bhat, R. V. Aflatoxins: Successes and failures of three decades of research. In: Fungi and mycotoxins in stared products: Proceedings of an international conference, Bangkok, Thailand, 23-25 April 1991. Champ B.R. et al. (eds), Canberra, ACIAR Proceeding No. 36, pp 80-85, 1991. 44. Jelinek, C.F. et al. World-wide occurrence of mycotoxins in foods and feeds-an update.journal of the Association of Official Analytical Chemists, 72: 223-230 (1989). Rapp. trimest. statist. sanit. mond., 50 (1997) 45. IARC. Monographs on the evaluation of carcinogenic risks to humans. Vol. 56. Some naturally occurring substances: rood items and constituents, heterocyclic aromatic amines and myc<>- toxins. Lyon, France, International Agency for Research on Cancer, 1993. 46. Payet, M. et al. Two observations of children who had consumed for a prolonged period flours contaminated with Aspergillus flavus. Presse midica/,e, 74: 649 (1966). 47. Serck-Hanssen, A. Aflatoxin-induced fatal hepatitis? A case report from Uganda. Archives of environmental health, 20: 729- 731 (1970). 48. Tung, T.C. &: Ling, K.H. Study on aflatoxin of foodstuffs in Taiwan. Journal of vitaminowgy, 14: 48-52 ( 1968). 49. Amla, I. et al. Cirrhosis in children from peanut meal contaminated by aflatoxin. American journal of clinical nutrition, 24: 609-614 (1971). 50. Krishnamachari, K.A. et al. Hepatitis due to aflatoxicosis - An outbreak in Western India. Lancet, i: 1061-1063 (1975). 51. Ngindu, A. et al. Outbreak of acute hepatitis caused by aflatoxin poisoning in Kenya. Lancet, i: 1346-1348 (1982). 52. Chao, T.C. et al. An outbreak of aflatoxicosis and boric acid poisoning in Malaysia: a clinicopathological study. Journal of pathology, 164: 225-33 (1991). 53. Dhatt, P.S. etal. Aflatoxins and Indian Childhood Cirrhosis. Indian journal of pediatrics, 19: 407-408 (1982). 54. Hendrickse, R.G. &: Maxwell, S.M. Aflatoxin and child health in the tropics. Journal of toxicowgy. Toxin reviews, 8: 31- 41 (1989). 55. Shank, R.C. et al. Aflatoxin in autopsy specimens from Thai children with an acute disease of unknown aetiology. Food and cosmetics toxicowgy, 9: 501-507 (1971). 56. Peers, F.G. &:Unsell, C.A. Dietaryaflatoxins and liver cancer - a population based study in Kenya. British journal of cancer, 27: 473-484 (1973). 57. Peers, F.G. et al. Dietary aflatoxin and human liver cancer. A study in Swaziland. International journal of cancer, 17: 167- 176 (1976). 58. Peers, F.G. &: Unsell, C.A. Dietaryaflatoxin and human liver cancer. Anna/,es de la nutrition et de l'alimentation, 31: 1005- 1018 (1977). 59. Peers, F. etal. Aflatoxin exposure, hepatitis B virus infection and liver cancer in Swaziland. International journal of cancer, 39: 545-553 (1987). 60. van Rensburg, S.J. et al. Primary liver cancer rate and aflatoxin intake in a high cancer area. South African medical journa~ 48: 2508a-2508d (1974). 61. van Rensburg, SJ. et al. Primary liver cancer and aflatoxin intake in Transkei.Journalof environmentalpathowgy, toxicowgy andoncowgy, 10: ll-16 (1990). 62. Bean, T.A. &: Yourtee, D.M. Aflatoxin metabolites in the urine of Nigerians. Comparison of chromatographic methods. Journal of toxicowgy. Toxin reviews, 8: 43-52 ( 1989). 63. Wild, C.P. et al. Dietary intake of aflatoxins and the level of albumin-bound aflatoxin in peripheral blood in The Gambia, West Africa. Cancer, epidemiowgy, biomarkers and prevention, l: 229-234 (1992). 64. Shank, R.C. et al. Dietary aflatoxins and human liver cancer, III Field survey of rural Thai families for ingested aflatoxins. Food and cosmetics toxicowgy, 10: 71-84 (1972). Wld hlth statist quart., SO (1997) 65. Bulatao-Jayme, J. et al. A case-control dietary study of primary liver cancer risk from aflatoxin exposure. Inter- national journal of epidemiowgy, 11: 112-118 (1982). 66. Yeh, F.S. et al. Hepatitis B virus, aflatoxins and hepato- cellular carcinoma in Southern Guanyxi, China. Cancer re- search, 49: 2506-2509 (1989). 67. Vasanthi, S. et al. Aflatoxin intake from Maize-based diets in a rural population in southern India. Journal of the science of food and agriculture, (in press) (1996). 68. Groopman, J.D. et al. Aflatoxin exposure in human populations: Measurements and relationships to cancer. CRC critical reviews of toxicol.ogy, 19: 113-144 (1988). 69. Campbell, T.C. et al. Non-association of aflatoxin with primary liver cancer in a cross-sectional ecological survey in the People's Republic of China. Cancer research, 50: 5882- 6893 (1990). 70. Stoloff, L. Aflatoxin is not a probable human carcinogen. The published evidence is sufficient. Regulatory toxicowgy and pharmacol.ogy, 10: 272-283 (1989). 71. Eaton, D.L. &: Callagher, E.P. Mechanisms of aflatoxin carcinogenesis. Annual review of pharmacol.ogy and toxicowgy, 34: 135-172 (1994). 72. Groopman, J.D. et al. Molecular dosimetry of aflatoxin exposure. In: Groopman,J.D. & Skipper, P. (eds). Mo/,ecular dosimetry of human cancer: epidemiol.ogical, analytical and social considerations. Boca Raton, FL. CRC Press, pp 302-324, 1991. 73. Groopman, J.D. et al. A molecular dosimetry of urinary aflatoxin-DNA adducts in people living in Guangxi Autonomous Region, People's Republic of China. Cancer research, 52: 45-52 (1992). 74. Groopman, J.D. et al. Molecular dosimetry of aflatoxin-N- guanine in human urine obtained in The Gambia, West Africa. Cancer, epidemiowgy, biomarkers and prevention, l: 221- 227 (1992). 75. Wild, C.P. et al. Molecular dosimetry ofaflatoxin exposure: contribution to understanding the multifactional etiopathogenesis of primary hepatocellular carcinoma with particular reference to hepatitis B virus. Environmental health perspectives, 99: 115-122 (1993). 76. Zhu,J.Q. etal. Correlation of dietary aflatoxin BI levels with excretion of aflatoxin Ml in human urine. Cancer research, 47: 1848-1852 (1987). 77. Wilkinson, A.P. et al. Immunoassay of aflatoxin in food and human tissue. Journal of toxicology. Toxin reviews, 8: 69-79 (1989). 78. Lamplugh, S.M. et al. Aflatoxins in breast milk, neonatal cord blood and serum of pregnant women (short report). British medical journal, 296:968 ( 1988). 79. Wtld, C.P. et al. In utero exposure to aflatoxin in West Africa [letter]. Lancet, 337:1602 (1991). 80. Zarba, A. et al. Aflatoxin M1 in human breast milk from The Gambia, West Africa, quantified by combined monoclonal antibody immunoaffinity chromatography and HPLC. Carcinogenesis, 13: 891-894 (1992). 81. Coulter,J.B.S. et al. Aflatoxins in human breast milk. Annals of tropical paediatrics, 4: 61-66 ( 1984). 82. El Nezami, H.S. et al. Aflatoxin Ml in human breast milk samples from Victoria, Australia and Thailand. Food and chemical toxicology, 33: 173-179 ( 1995). 149 Modernizing food control systems: the experience of Thailand Pakdee Pothisiri a Many parts of the world, including Thailand, have experienced rapid social and economic develop- ment. As a result, the application of modern tech- nology and the utilization of chemicals are on the increase in agriculture and the food industry, in order to boost productivity in the face of high competitivity. Newly emerging and re-emerging biological hazards, particularly those that cause severe diarrhoeal diseases are prevalent in many areas of the country. Food control has thus become even more important for the protection of con- sumers against health hazards and fraud. However, in view of the !imitated financial and human re- sources available to cope with the increasing demo- graphic and industrial growth rate, the Thai Food and Drug Administration (FDA), the main food control authority in Thailand, had to be strength- ened for effective food control by modernizing its structure, its methods of work and its information services. This article aims to provide information concerning modernization of food control sys- tems, with particular emphasis on food safety infor- mation services, based on the experience of Thai- land. Situation and problems Data obtained from monitoring food safety in 1995 ( 1) showed that approximately 30% of food sam- ples collected in regional areas and 18% of those collected in metropolitan Bangkok did not comply with established standards because of contamina- tion by biological or chemical agents. Contamination by biological agents Microbial contamination. Analysis of results (2) indi- cated that approximately 20% of food samples collected throughout the country over the past 5 years contained microorganisms (such as bacteria, yeasts, moulds, coliforms and E. coli) which ex- ceeded mandatory standards, particularly in semi- processed foods, sauces, ice cream, bottled water, ice and beverages, causing "food poisoning"b and diarrhoeal diseases in schoolchildren. It is estim- ated that in recent years 80-90% of outbreaks of • Secretary-General, Food and Drug Administration, Bangkok, Thailand. b The term "food poisoning" is ill-defined and non-scientific; its use is discouraged by WHO. 150 foodborne illness could be attributed to contam- ination of food with pathogenic bacteria. This problem may be associated with ignorance of good hygienic practices on the part of food man- ufacturers. Mycotoxin contamination. Based on research con- ducted during 1989-1991 ( 3) increasing aflatoxin intake among the Thai population was found to be correlated with increasing incidences of detected liver cancer. Staple food frequently contaminated with aflatoxin include grains, legumes, meats, milk and dairy products, fats and oils, and condiments. High humidity in tropical countries like Thailand is suitable for the growth of certain groups of moulds capable of producing toxins such as afla- toxin in some agricultural products. Good agricul- tural practice, appropriate handling and storage of food products as well as consumer education are essential to avoid health problems caused by myco- toxin. Contamination by chemical agents Chemical additives contamination. Food additives are used mainly by the food industry when technologi- cally justified. However, in the face of commercial competition, there has been massive utilization of food additives such as flavouring agents, non- edible colours and preservatives. Preservatives such as nitrates and nitrites have sometimes been un- necessarily added in high amounts to a variety of foods. The use of food colouring has also been a serious problem in the past decade. According to a 1974 report (4) on food products, approximately 34% of confectionery products widely consumed by children, collected from various local markets, were contaminated with prohibited colouring agents. Most prohibited colouring agents have been implicated as carcinogens, but the present situation shows some improvement. Pesticide residues and veterinary drug residues. The use of pesticides and veterinary drugs has been increasing rapidly in order to maximize agricul- tural production. Pesticide residues in foods create health problems due to misuse and mishandling by farmers and other users. National epidemiological data ( 5) have shown that several cases of food borne disease occurring among schoolchildren and fac- tory workers were related to accidental contamina- tion of foods by pesticides. Rapp. trimest. statist. sanit. mond., 50 (1997) Heavy metals contamination. Unintentional con- tamination of food by heavy metals has resulted from manufacturing processes as well as from envi- ronmental pollution. Seafoods, particularly mol- luscs, were found to be contaminated by mercury and cadmium at appreciable levels. Certain fer- mented foods, vinegars and beverages were found to be contaminated by lead from the environment as well as from ceramic containers. Some reports indicated that villagers in the southern part of Thailand, particularly Ronpiboon area of Surath- Thani province, suffered from skin diseases due to the chronic toxic effects of arsenic, after consum- ing local underground water for several years. Food control strategies In order to cope with the ever increasing magni- tude and complexities of food safety problems and foodborne diseases, a food control programme was incorporated into the Seventh National Economic and Social Development Plan (1992-1996). The main strategies adopted and implemented by the Thai FDA may be summarized as follows: • to improve the infrastructure of the food con- trol system, including revision of food registra- tion, in order to effectively protect consumers against unsafe foods; • to increase efficiency of cooperation as well as coordination among agencies involved in food control; • to develop an integrated information system with other agencies leading to the establish- ment of an information network for the collec- tion, monitoring and dissemination of informa- tion to the public; • to initiate food surveillance in order to trace problems and search for appropriate preven- tive or corrective measures; • to promote and implement self-protective mea- sures involving public participation; • to promote and upgrade manufacturing stan- dards of food-processing factories in conformity with good manufacturing practice (GMP), and Hazard Analysis and Critical Control Points (HACCP) system; and • to improve consumer protection in the provin- cial areas, aiming for equity in coverage. Regulatory enforcement The FDA is the prime agency responsible for the implementation of national food control pro- gramme in coordination with other agencies, both within and outside the Public Health Ministry. Its main duty under the Food Act B.E.2522 (1979) is to guarantee the quality and safety of foods by establishing food safety standards or hygienic prac- tices, labelling requirements, production control, and controlling the importation and advertising of food products, as well as the registration of specifi- Wld hlth statist. quart., 50 (1997) cally controlled foods. The principle of the regula- tory procedures for food safety comprises 3 areas: pre-marketing control, post-marketing control and surveillance. Pre-marketing control Establishment of food standards and manufacturing re- quirements. A set of standards, supervised by the subcommittee on food standards and manufactur- ing requirements, constitute the minimum accept- able requirements. Control of food manufacturing. Food manufactur- ers must apply for a licence prior to opening their business. Plant lay-outs must be submitted for the approval of the Food Control Division. The FDA inspectors will then visit and inspect the plant be- fore a manufacturing licence can be issued. It is the responsibility of the licencee to renew his/her licence every 3 years. Control of food importation. A licence is required for importing food into the kingdom, and it must be renewed every three years. A licencee may im- port various kinds of food provided that they are approved by the Office of Food and Drug Adminis- tration. FDA inspectors will visit and examine the appropriateness of the designated storage place or warehouse before a licence is issued. Control of food products. A food product, either manufactured or imported, if categorized as "spe- cifically-controlled food" must be registered. Re- sults of analysis of the product and details of the manufacturing processes and its ingredients must be submitted along with an application for registra- tion. Food additives themselves are subject to ap- proval before they can be used. Control of food advertisement. All forms of food advertisement through any mass media are subject to approval from the FDA. Advertisement of false or deceptive quality or benefit is prohibited. The approval covers the statements and visual materials to be used in food advertisement. Post-marketing control Monitoring of compliance with the regulations. The monitoring process is intended to ensure that food as distributed to consumers is wholesome and that the quality complies with the national food standards. Inspection of food factories and premises throughout the country is regularly con- ducted together with sampling of food products for laboratory testing. In cases of violations, ac- tions such as seizures, product recalls, and prose- cution will be carried out. Technical advice on the development of food production, delivery, hand- ling and storage may be given during the moni- toring processes. 151 Food surveillance The aim of the programme is to assure the safety and quality of food distributed in the market throughout the country. Food surveillance is con- ducted by several ministerial agencies, e.g., Minis- try of Agriculture and Cooperatives, Ministry of Science, Technology and Environment, Ministry of Industry, Bangkok Metropolitan, and the Office of the Prime Minister. However, the FDA plays the major role and has overall responsibilities. FDA inspectors will take samples of food at market places from time to time, and whenever problems arise. The samples are delivered to the Food Analy- sis Division, Department of Medical Science for analysis to identify toxins, pesticide residues, heavy metals, nutritional values and standard conformity. Warning and legal actions, such as seizures, prod- uct recalls etc., are taken depending on the degree of violation. Consumer awareness of the impor- tance of selecting quality food is always encouraged through the dissemination of related information and educational material to the general public. Modernizing the food control system In 1995 the Thai FDA reaffirmed its mandate as "an agency which protects consumers on health products especially their safety, qualities and bene- fit by unique and systematic performances and co- operation among concerned agencies while build- ing up credibility and satisfaction among business operators, as well as acceptability at global level" (6). In order to fulfill its new vision the Thai FDA has proposed and implemented a series of activi- ties which can be summarized as follows: Changing the working system and infrastructure. Due to understaffing and increasing workload, the FDA has developed systems and procedures to in- crease working efficiency by combining pre-mar- keting control processes (such as licensing of food premises, food product registration, food labelling approval, etc.) and post-marketing control process- es (such as inspection of food premises and sam- pling of food products, taking legal action etc.) into an integrated service. As a result, both busi- ness operators and consumers benefit from more efficient and faster services. Information system. To increase its efficiency in accessing relevant information the Thai FDA has recently established an information centre for con- sumer protection which collects, compiles, moni- tors and disseminates information. It is planned to link this data system to all units within the FDA, and to all departments in the Ministry of Public Health (MPH) as well as to 76 provincial health offices throughout the country. Moreover, tele- communications using the Internet via the com- puter centre of the Ministry of Public Health will enhance information exchange with international 152 organizations and with concerned agencies of other countries. Therefore, the level of efficiency in inspection, monitoring and information service is expected to be significantly improved. Establishing regional technical centres. In 1992 the Thai FDA delegated its duties, provided by the Food Act, to the provincial health offices in all regions. In order to keep the same standard opera- tion procedures, it has recently initiated a project to establish 12 regional technical centres. These regional centres are intended to serve in support- ing and coordinating work between provincial health offices and the head office in order to in- crease both the consistency and efficiency of con- sumer protection activities at intra- and inter- provincial levels. The first regional technical centre has already been established and is functioning in Chonburi province. The remaining centres should be com- pleted within the Eighth National Economic and Social Development Plan (1997-2002). Revision of food legislation. Since the existing Food Act has been in force for almost 18 years, some provisions are obsolete and cannot keep up with the present situation. The need for revision of the act itself is supported by all parties concerned. The FDA recently drafted a new text to replace the Food Act 1979 (B.E.2522). In principle, the revised legislation will emphasize deregulation and shift- ing pre-marketing control measures into monitor- ing measures. Moreover, provisions concerning in- novative practices such as direct sale control which is now a major problem, will be added. It will also facilitate harmonization with international stan- dards to promote food export. In addition, HACCP approaches will become compulsory for certain food products. As for violation of the law, the penalties foreseen will be more severe and strictly enforced. The Thai FDA has encouraged consumer par- ticipation in monitoring violations of food laws by issuing a new ordinance which offers a reward for any report of violation. This ordinance was ap- proved by the Finance Ministry and became effec- tive in June 1995. Up to now, several cases of viola- tions such as illegal production or importation, as well as the sale of adulterated, mislabelled, substan- dard and unsafe foods have been reported by con- cerned consumers, resulting in strengthened con- sumer protection. Cooperation and coordination among concerned agencies. In Thailand, various agencies are taking part in food control, resulting in the duplication of some activities. There is a need to strengthen coop- eration and coordination among concerned agen- cies with respect to food safety in all stages of the food chain, starting with production and harvest- ing at the farm, through processing and marketing until the product reaches the consumer. The FDA Rapp. trimest. statist. sanit. mond., 50 (1997) therefore organized a national seminar in April 1995 in order to brainstorm and reach an agree- ment among concerned agencies. The main con- siderations were to avoid duplication and overlap- ping of responsibilities among concerned agencies to maximize utilization of resources and comple- mentarity in order to improve food safety. Partici- pants consisted of representatives from various or- ganizations such as the Ministry of Agriculture and Cooperatives, the Ministry of Industry, the Ministry of Sciences & Technology and Environment, the Ministry of Finance, the Ministry of Public Health and the food industry. They unanimously agreed in principle to establish a network in food safety control as shown in the proposed model (Fig. 1). The outcome of this seminar should lead to the formulation of a national food policy including a food safety scheme. A Food Import & Export In- spection and Certification system will also be an integral part of this national food safety scheme. Upgrading manufacturing practices by implementa- tion of HACCP. The Hazard Analysis and Critical Control Point (HACCP) concepts were introduced into the food inspection programme during 1990 in order to ensure the safety and quality of food products. This system is used to assess hazards and establish specific control measures for prevention Fig. 1 Organizational chart on national food safety and control, rather than depend on finished prod- uct testing and conventional inspection methods. The food industries are encouraged to adopt HACCP approaches and good manufacturing prac- tice (GMP) codes on a voluntary basis through guidelines and manpower development. Priority target groups for the implementation ofHACCP are manufacturers of exported low-acid canned foods, frozen seafoods and meats as well as bottled water. Public education and consumer participation. The objectives of food control could not be achieved solely by law enforcement, the setting up of a con- trol organization or the promulgation of regula- tions. It could be said that benefits derived from a food control service will depend on the degree of consumers' awareness and support. Consumer ed- ucation and participation, therefore, should be considered as an important function of food-con- trol organizations and inspection services as well. The scientific and technical information of current concern regarding food safety such as risk assess- ment and food safety evaluation, aspects of food hygiene and proper handling practices, and pro- tection measures have to be disseminated, in order to raise the awareness and knowledge of con- sumers. In 1996, the Thai FDA started a pro- gramme on "campaign of label reading before Diagramme d'un syst~me national de contrale de la salubrite des aliments Executive Committee • Consumer • Plaintes des Comite execulif complaint consommateurs l! • National and • Information au niveau Research & Service recherche et • International national et t:; National data base t; development Unit developpement information • au niveau international Base de donnees • Cooperation & • Cooperation et coordina- !:j nationale coordination on tion de la recherche-Information & Service information research & developpement dissemination et diffusion de Tl development • Analyse et appui unit !'information • Scientific analysis scientifiques National focal point & support • Evaluation des risques • Private sector • Participation du • Risk assessment • Mise en valeur des participation secteur prive Point focal national • Manpower ressources humaines • Consumer • Education du development education consommateur • Communication • Communication et T! & behaviour elude des study comportements Operation unit Service exploitation • Policy and • Politique et • Surveillance • Surveillance planning planification • Import & export inspection and • Inspection et certification des certification importations et des exportations • Regulatory inforcement • Application de la reglementation • Policy planning & budgetting • Planificalion et budgetisalion ~ ~ • Private laboratory accreditation • Homologation des laboratoires prives "' C) • Epidemiological information • Information epidemiologique ~ Wld hlth statist. quart., 50 (1997) 153 buying" using various media such as radio, tele- vision spots and printed material. In addition to this programme, electronic messages such as "audio text" were initiated in September 1996 to promote consumer education. Under the responsi- bility of FDA, "audio text" is a useful tool for con- sumers who are interested in gathering informa- tion relevant to health products, in terms of safety aspects, status of regulation, etc., through an auto- matic answering telephone system. International cooperation. As a member of the World Trade Organization (WTO), Thailand is obliged to follow the Sanitary and Phytosanitary Measure (SPS) as well as the Technical Barrier to Trade Measure (TBT), and must, therefore, adjust its strategies to regulatory measures and various control procedures for the benefit of international trade. Activities such as (i) harmonizing of national regulations along the line of Codex recommenda- tions, (ii) participating in drafting Codex standards and related codex activities, (iii) disseminating reg- ulatory information to other member countries as required and requested, and (iv) inspection and certification of export foods to build up credibility and acceptability of importing countries are con- sidered high priorities. Summary Food control is essential for the benefit of local con- sumers as well as international trade, which impacts on the overall social and economic concerns of the nation. Nevertheless, success in food control implementation will not be achieved by a single agency. Concerned agencies as well as food industries have to be fully aware of and be responsible for the production of good quality and safe foods. Efficient cooperation and coordi- nation among the agencies concerned, therefore, is indispensable in the development of food safety. The Thai FDA has recently attempted to modernize its food control system by various approaches, including the establishment of a national food safety scheme and the introduction of modern technology as well as restructur- ing and changing the working system. It is anticipated that control along new directions may enhance better consumer protection in this country. 154 Resume Modernisation des systemes de controle des denrees alimentaires: /'experience de la Thai1ande Le contr61e des denrees alimentaires, indispensable a la protection des consommateurs locaux, a egalement des repercussions favorables sur le commerce interna- tional, et done sur le bien-etre economique et social de !'ensemble du pays. Toutefois, le contr61e des denrees alimentaires ne peut etre mene a bien par un organisme unique. Les organismes interesses et l'industrie alimen- taire doivent etre pleinement conscients de tous les aspects de la salubrite des aliments et assumer la responsabilite de la production d'aliments sains de bonne qualite. Une cooperation et une coordination efficaces entre les differents organismes interesses sont essentielles pour assurer la salubrite des aliments. La Thai Food and Drug Administration (FDA) s'est re- cemment efforcee de moderniser son systeme de con- trole des denrees alimentaires par divers moyens, no- tamment la creation d'un systeme national de contr61e de la salubrite des aliments, !'introduction des technolo- gies modernes, la restructuration et la modification des methodes de travail. Ces changements devraient ame- liorer la protection du consommateur dans ce pays. References/References 1. FDA report on summary of analytical result in regional area 1995 [in Thai). Nonthaburi, Food and Drug Administration, Ministry of Public Health, 1996. 2. fDA year book 1995. Nonthaburi, Food and Drug Administration, Ministry of Public Health, 1996. 3. Kan.myavanij, S. et al. Dietary intake of Aflatoxin and risk estimate. Bulletin of the Department of Medical Science, 36 ( 4): (October-December, 1994). 4. fDA report on summary of analytical result in regional area 1994 [in Thai]. Nonthaburi, Food and Drug Administration, Ministry of Public Health, 1995. 5. Annual epidemiological suroeillance report 1993. Nonthaburi, Division of Epidemiology, Office of Permanent Secretary, Ministry of Public Health, 1994. 6. Modification of system and working procedures on consumer protection in health of Thai food and drug administration [in Thai]. Nonthaburi, Food and Drug Administration, Ministry of Public Health, 1997. Rapp. trimest. statist. sanit. mond., 50 (1997) WO R L D HEALTH ORGANIZATION PUBLICATIONS Issued in May 1997, this WHO report on the "crisis of suffering" caused by chronic diseases made headlines the world over. The World Health Report 1997: Conquer- ing Suffering, Enriching Humanity provides an expert assessment of the world health situation in all its complexity, giving par- ticular attention to the problems posed - in developing and industrialized countries alike - by chronic diseases. Compiled by the World Health Organization, the report com- bines the latest global health statistics with profiles for each of the major chronic dis- eases, offering pertinent details about causes, risk factors, global and regional trends, and current and future prospects for prevention, treatment, cure, and rehabilita- tion. The report opens with an overview of recent trends that have made chronic diseases the focus of particular concern in every part of the world. These include population ageing, which puts more people at risk of developing chronic conditions late in life, global popu- lation growth, and the rising prevalence of unhealthy lifestyles -characterized by inap- propriate diet, inadequate physical exercise, and smoking. The report also explains why these trends call for an urgent review of several traditional attitudes towards health. Key issues raised include the need to combat chronic and infectious diseases simultane- ously, rather than sequentially, as in the past; knowledge that many chronic diseases have infectious causes, which reinforces the need for simultaneous action; and the tremendous burden of suffering and disability caused by chronic diseases, which makes health ex- pectancy more important than life expect- ancy. This general discussion is followed by a profile of the state of world health at the end of 1996, including a concise analysis of The World Health Report 1997 Conquering Suffering Enriching Humanity global economic, political, social, and de- mographic trends relevant to health in gen- eral and chronic diseases in particular. Also included are the most up-to-date statistics on life expectancy, mortality by age group and sex, and causes of death and disease. Against this background, the report turns to an in-depth analysis of all the chronic dis- eases that are major causes of death or avoid- able ill-health and disability. These include each of the eight most common cancers, circulatory diseases, asthma, diabetes and other metabolic disorders, hereditary dis- eases, musculoskeletal diseases, and men- tal and neurological disorders. For each disease, the report gives a brief description of clinical features, estimates the global numbers affected, and identifies the main risk factors. Drawing on the latest scientific knowledge, the report also explains current and future methods of prevention, detec- tion, diagnosis, and treatment - whether concerning genetic influences that are be- ing probed in research laboratories or rec- ommendations for a healthy diet. Noting the relatively small number of risk factors shared by many of these diseases, the report points to major opportunities for pre- vention through the use of coordinated strat- egies and integrated packages of disease- specific interventions. Though the focus is on global problems and global solutions, these profiles also offer individuals the best available advice on protecting their own health, particularly in the absence of a cure for most chronic diseases. To show the way forward, the report de- scribes the wide range of ongoing WHO activities aimed at combatting chronic dis- eases, infectious diseases, and a host of other factors that threaten world health. The final chapter, on charting the future, estimates future trends for selected diseases and iden- tifies six priority areas where international action can help combat chronic diseases and thus reduce the enormous suffering and dis- ability that they cause. The World Health Report 1997 Conquering Suffering, Enriching Humanity 1997, vi + 162 pages Available in English and French editions ISBN 92 4 156185 8 SFr 15.-/US$ 13.50 In developing countries: Sw.fr. 9.- 0rder no. 1241997 WHO• DISTRIBUTION AND SALES• 1211 GENEVA 27 • SWITZERLAND ORGANISATION MONDIALE D f L A S A N T E Publie en mai 1997, ce rapport de l'OMS sur "l'aggravation de la souffrance"causee par les maladies chroniques a fait la une des journaux. Rapport sur la Sante dans le Monde 1997 Vaincre la souffrance "Nous devons reconnaitre que sans la qualite de la vie, une longevite accrue ne presente guere d'intereL L 'esperance de sante est aussi importante que l'esperance de vie." Le Rapport sur la sante dans le monde /997; vaincre la souffrance, enrichir l'humanite est un bilan d'experts sur la situation sanitaire dans le monde dans toute sa complexite et en particulier sur les maladies chroniques. Etabli par !'Organisation mondiale de la Sante, ii associe les statistiques sanitaires les plus recentes a une analyse des facteurs multiples qui ont conduit a la situation actuelle - et a un accroissement des souffrances et des incapacites qui accablent l'humanite. Du fait de l'allongement spectaculaire de l'esperance de vie associe a une transforma- tion complete des modes d'existence, des epidemics mondiales de cancer et autres maladies chroniques vont, au cours de prochaines decennies, provoquer un enorme accroissement des souffrances et des incapacites qui accablent l'humanite. Les maladies chroniqucs tuent deja plus de 24 millions de pcrsonnes par an et sont done responsables de pres de la moitie du total des deces dans le monde. Le vieillissement constant de la population mondiale a pour effet de donner aces mala- dies !'occasion d'atteindre, chez un plus OMS• DISTRIBUTION f nrichir / 'humanite grand nombre de personnes, le stade ou elles deviennent mortelles ou invalidantes. Les gens sont maintenant plus preoccupes de leur esperance de sante que de leur espcrance de vie. Le Rapport sur la sante dans le monde I 997 met !'accent sur les principales maladies chroniques telles que les cancers et les cardiopathies, le diabete et les affections rhumatismales, et les troubles mentaux et neurologiques. II montre comment Jes a vantages d'un prolongement de la duree de vie risquentd'etre compromis par !'incidence E T VENTES • 1211 des maladies chroniques s'ajoutant a celle des maladies infeetieuses, ce qui oblige a mener la bataille pour la sante simultane- ment sur ces deux fronts. Pour chaque secteur d'activite, le rapport definit les mesures internationales de prevention, de traitement et de readaptation qui s'imposent en priorite. II donne aussi a chaeun Jes meilleurs conseils possibles pour la protection de sa propre sante, et explique ce que fait \'OMS dans ce domaine crucial, tout en passant en revue !'ensemble de l'activite de !'Organisation. "Toute vie humaine a inevitablement une fin. Veiller a ce qu 'elle survienne de lafaron la plus digne, lapluschaleureuse et la moins douloureuse possible est un objectif dont le rang de priorite vaut tous Les autres. " Rapport sur la sante dans le monde 1997 Vai.ncre la souffrance Enrichir l'humanite 1997, vi + 162 pages (disponible en franr,:ais et anglais) ISBN 92 4 256185 I Francs suisses: 15.- Pays en developpement: Fr.s. 9.- N0 de commande 2241997 G E N E V E 2 7 • S U I S S E Publications of the World Health Organization 1997 WORLD HEALTH FORUM An international journal of health development (Separate editions in English, French, Spanish, Arahi<:, Chinese and Russian) World health forum is aquarterlyjournal for policy-maker,, health planners, administrators, health educators, and public health workers ofall kinds. It pro,·ides a mediun1 !Or the presentation and disrussi<n1 of new concepts in public health and new approaches to health problems, and it is devoted to the irnprovement or health through the.: promotion of health services covering the emire population and the undertaking or a wide variety of public health measures, whether or not they are supported by WI 10. The Forum is the main organ of VVI IO's t\.{ember States for the international exchange or health lnfr)rmation and an active instrument in technical cooperation among developing countries. Subscription (4 numbers) ...................................... . Price per copy ........................................................... . Sw. fr. 80.- Sw. fr. 2fi.- BULLETIN OF THE WORLD HEALTH ORGANIZATION The Bulletin is the principal scientific organ of WI 10: iLs role is to re,inv progress in medical and rdatcd sciences ("update" articles) and bring to light new knowledge by publishing original paper, on scientific research in the laborat01y and field. The Hul/elin comains original articles in either English or French, with a summary in the other language. Bimonthly. Subscription (6 numhers) .... . ....................... . Price per C<)(lY ............................................................... . WEEKLY EPIDEMIOLOGICAL RECORD (Bilingual: English and French) Sw. fr. 194.- Sw. fr. 42.- Prepared for the guidance of health administrations and health authorities, the tt-'eekly eJ1idetniologi.r:al retard contains notifications made under the International Health, Regulations and inf<,rmaLion concerning their application. The Rl!cord also contains epidemiological information on communicable diseases of international irnportance. Annual subscription ......................................................... . Sw. fr. 230.- WORLD HEALTH STATISTICS QUARTERLY The mirld health statistics quarterly replaces (since l 978) the World health .1t11t,,lir.mpctrt (published since 1967) and its forerunner thel!pidrmiologiml r,nd vital statistin repctrt (published since 1947). It deals with the detailed analysis of selected health topics of current interest. The()Jwrter/ycontains articles in either French or English with a summary in both languages. Annual subscriptio11 ........................... . Price per copy ............................................................ . WORLD HEALTH STATISTICS ANNUAL (Bilingual: English and French) Sw. fr. 121.- Sw. fr. 39.- The forerunner of this series was the ,1nnual tjiidemiolllgical report of the League of Nations. It was followed by the Annual epidemiological and vilal statistin issued hy the World Health Organization. Latest publications: 1992. Vital statistics and life tables, implementation of the: Glohal Strategy for Ilealh for A.II by the Year 2000 (single volume), 480 pages l 99'\. Vitnl statistics and life tables, health and demographic daLa, and alternalive sources for the collection of cause-of-de,uh and vital events data in the absence of universal vital registration ( single volume), 662 pages ......... . 1994. Vital statistics and life tables, and worldwide health and demographic data (single volume), 4fi0 pages .......................................................... . 1995. Vital statistics and life table, (including causes of death in the newly independent states of the fonncr t:SSR), and worldwide health and demographic data (single volume), 880 pages Sw. fr. IOO.- Sw. fr. 100.- Sw. fr. IOO.- Sw. fr. 150.- Publications de /'Organisation mondiale de la Sante 1997 FORUM MONDIAL DE LA SANTE Revue internationale de developpement sanitaire (EdiLions <if'parees en fran<,:ais, anglais, c!-ipaguol, arabe, chinois et russe) Forum. mandialdtla santP. est tmc revue Lrimeslrielle deslinee aux responsables des politiques sanii.aires, aux planificatettr'i, administratcurs cl ('ducat(·ur~ ~anitaires, enfin aux travaillt:urs de la sanl(: puhlique de toutes categories. Trihui1c pour la pre~ent..1.tion et la di'icttssion de nou\eaux nntccptscn santt'. publique et de nouvdlcs appnKhes des problernes de sante. forum se consacrc a l'ami~lioration de la sante par la promotion de services de sant{ couvranl la population tout entiCre et d'unc vasle gamme de me~ures de santC puhlique, qu 'ellessoient ou non soutenues par I '0~1[S. II est le pri11dpal organe a la disposition des Etats Membres de l'O~IS pour l'echange i1uen1ational d 'informations sanitaires en uu'':rnc t<"mps q u 'un instrument de cooperalion technique entre pays en d{:vdoppement. Abonnement (4 nuin<:ros) .............................................. . Le llllIIlero ........................................... . Fr. s. 80.- Fr. s. 26.- BULLETIN DE L'ORGANISATION MONDIALE DE LA SANTE Le Hulu,lin est le principal organe scientifique de: !'OMS; il a pour r<ile de passer en revue les progrt's des sc:iences m{dicak~!-i et apparentees ( articles de la n1brique (<Le poinL>~) el de mettre en lumit're le.s ronnaissanccs nouvelles en prCst'nl.anl dans des articles originaux lcs rfsultats de recherches scientifiques au laboratoire et sur le terrain. le Bulktin contient des articles originaux en frarn;ais ou en anglais, accompagnes d'un resume dan.s l'autre languc. Bintcnsuel. Abonnement (6 numi:ros) .... . I.e nun1<::'rc., .................................................................. . RELEVE EPIDEMIOLOGIQUE HEBDOMADAIRE (Bilingue: fran(ais et anglais) Fr. s. 194.- Sw. fr. 42.- Destinf. aux administrations sanitaires et aux services de sanlt'., le RelLvf epidemiologi,que hehdmnadaire contient les notifications exigies par le Reglement sanit.aire internatio11al, ainsi que d'auu·es renseignen1ents concen1ant l'application de ce reglt'ment. I.c &l.n.,eamLienL egalement des informations f::pidCmiologiques concernant les maladies u·ansmissihles d'une import.anc:e inten1ationak. Prix de l'abonnement annud ... Fr. s. 230.- RAPPORT TRIMESTRIEL DE STATISTIQUES SANITAIRES MONDIALES Le Raj,part lrim.eslrielde stalisliques r,wndial.es, remplarc ( dcpuis l 978) lellttptmrt de statistiques mnitaiw mrmdial,s (publie depuis 1967) et son pri'curseur le llapporl epidimiolll{flque et dirrwgraphiq,u (puhlic depuis 1947). ll preseme des analrses df't.ailli:es sur des s1~jets spt:'dfiqucs d'interel couranl. Le Trimestriel prt'sente des articles origi.naux en fran('.ais ou en anglais, accompagnCs d'un resume clans les deux langucs. Prix de l'abonncmenl annuel ......................................... . Le 11umero ................................................................. . Fr. s. 121.- Fr. s. 39.- ANNUAIRE DE STATISTIQUES SANITAIRES MONDIALES (Bilingue: franrai, et anglais) Cet annuaire rempla,·c les Stalisliques epidimiolllgiques et demograJ,hiques annuellP.s publiees par l'Organisation mondiale de la Sante et qui avaient1 cllcs-m<'rnc:,, rem place le Rapport r!pi.d.bniologique annuel puhlie par la Societe des ~ations. Publicarions rt'.cc11tcs: 1992. Mouvcment de la population et tables de survic, mise en cruvre de la Stratt'.gic rnondialc de la sant.e pour tous d'ici l'an 2000 ( I scul volume), 480 pages ........................................................... . 1993. Mouvement de la population et tables de survie, donnees sanitaires et df'tnographiques, methorles permettant de recueillir des donnt'es sur les causes de di'.ct'.s et lcs faiL<; de l'elat civil en l'ahsence de systeme universd d'cnregisLrernent (1 seul volume), fi62 pages ................................. . 1994. \fouvemcnl de la population et tables de sur,ic, ainsi que donni:cs sanitaire~ et demographiques i l'echelle mondialc (l scul volume), 460 pages 1995. Mouvementdt' la population et t.1.blesde survie (y compri~ cau1,es de deces pour lcs nouveaux Elals independant, de l'ex-URSS), ainsi que donni:t:s sanitaire, cl demographiques a l'echdle mon- diale (1 seul volume), 880 pages ..... Fr. s. 100.- Fr. s. 100.- Fr. s. 100.- Fr. s. 150.- Food safety and foodborne diseases Illness due to contaminated food (i.e. foodborne disease) is perhaps the most widespread public health problem in the contemporary world and an important cause of reduced economic productivity. Be it in the form of infant diarrhoea, cholera, salmonellosis, listeriosis, infections due to enterohaemorrhagic Escherichia coli or acute or chronic intoxications caused by chemical contami- nants, to mention but a few, foodborne diseases cause mortality, morbidity, suffering and economic losses which no country can afford. While by no means a comprehensive account of all the food safety problems encountered world- wide, this issue of World health statistics quarterly tries to foster an awareness of a growing and serious public health problem and advocates the need for change. Food safety does not receive the degree of attention and funding which it deserves; the health and economic impact of foodborne diseases associated with the contamination of the food supply has to be recognized nationally and internationally so that resources can be set aside for their prevention. Salubrite des aliments et maladies d'origine alimentaire Les maladies transmises par des aliments contamines (c'est-a-dire les maladies d'origine alimentaire) sont peut-etre le probleme de sante publique le plus courant dans le monde aujourd'hui et se traduisent par une baisse importante de la productivite economique. Qu'il s'agisse de la diarrhee enfantine, du cholera, des salmonelloses, de la listeriose, des infections enterohemorragi- ques dues a Escherichia coli, ou encore des empoisonnements aigus ou chroniques par des produits chimiques toxiques, pour n'en citer que quelques- unes, les maladies d'origine alimentaire causent une mortalite, une morbidite, une souffrance et des pertes economiques dont aucun pays ne peut porter le fardeau. Ce numero du Rapport trimestriel de Statistiques sanitaires mondiales, qui ne donne en aucune fa9on une image complete de tous les problemes de salubrite des aliments dans le monde, cherche a faire prendre conscience de ce probleme de sante publique qui va en s'aggravant, et insiste sur la necessite d'agir. L'attention portee a la salubrite des aliments et les fonds qui y sont consacres demeurent insuffisants; si l'on veut que des ressources soient consa- crees a la prevention des maladies transmises par des aliments contamines, ii faut que les responsables a !'echelon national et international prennent conscience de leurs consequences sur la sante et l'economie.
Всемирная организация здравоохранения (ВОЗ / WHO) · Journal articles
Food safety and foodborne diseases = Salubrité des aliments et maladies transmises par les aliments [full issue]
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст