Malaria incidence and prevention among European and North American travellers to Kenya.

Malaria incidence and prevention among European and North American travellers to Kenya H.O. Lobel,' P.A. Phillips-Howard,2 A.D. Brandling-Bennett,'3 R. Steffen,4 C.C. Campbell,1 A.Y. Huong,1 J.B.O. Were,3 & R. Moser4 A longitudinal survey was conducted among travellers departing from Nairobi airport to determine the use of malaria prevention measures and assess the risk for malaria while travelling in Kenya. Among 5489 European and North American travellers, 66 different drug regimens were used for prophylaxis, and 48% of travellers used both regular chemoprophylaxis and more than 1 antimosquito measure during travel; 52% of 3469 travellers who used chemoprophylaxis did so without interruption during their travel and for 4 weeks after departure. Compliance was lowest among travellers who visited friends and relatives, who were young, or who stayed more than 3 weeks. Sixty-seven (1%) travellers experienced symptoms of malaria, but the diagnosis could be verified for only 16 of these. Long-stay travellers appeared to be at higher risk for malaria than short-stay travellers, and health information needs to be targeted especially to the former. Similar investigations are needed among international travellers to other malaria-endemic countries. With comparable data available, consistent and effective malaria prevention guidelines can be developed. Introduction The incidence of malaria among European and North American travellers returning from Africa has increased markedly during the past decade (1,2).' Especially significant is the rise of malaria incidence among travellers to Kenya, which receives more than 400 000 travellers from Europe and North America each year.b The number of Plasmodium falciparum infections acquired in Kenya and imported into the United Kingdom and the USA tripled between 1977 and 1986 (1).' The increased number of travellers to Kenya, and the lack of effective and safe chemopro- phylactic regimens, contributed to the increased number of cases. The lack of optimal chemopro- phylaxis has led to an increased risk for acquiring malaria during travel (3). ' Division of Parasitic Diseases, Center for Infectious Diseases, Centers for Disease Control, Public Health Service, U.S. Depart- ment of Health and Human Services, Atlanta, GA, USA. Requests for reprints should be sent to Dr. H.O. Lobel, Malaria Branch, Centers for Disease Control, Atlanta, GA 30333, USA. 2 London School of Hygiene and Tropical Medicine, London, England. 3Clinical Research Centre, Kenya Medical Research Institute, Nairobi, Kenya. 4 Institute of Preventive Medicine, University of Zurich, Zurich, Switzerland. g Malaria surveillance annual summary. Atlanta, GA, Centers for Disease Control, 1986. " Economic survey. Nairobi, Ministry of Planning and National Development, 1988. Reprint No. 5064 The lack of highly safe and effective drugs to prevent infection with chloroquine-resistant P. fal- ciparum has created a dilemma for travellers and for those who develop recommendations for malaria prevention. In North America and Europe, at least sixteen different drug regimens are recommended by various authorities for travellers to East Africa (4-6). Because many chemoprophylactic regimens do not eliminate the risk for infection, many experts recom- mend that travellers carry a curative dose of a drug. This dose is to be used to treat a febrile illness, when medical care is not available, in an effort to prevent death from P. falciparum infection. In addition to chemoprophylaxis, antimosquito measures, such as repellents and mosquito nets, are frequently recom- mended because they are considered safe and effec- tive ways of reducing mosquito contact (7). To formulate malaria prevention recommenda- tions for travellers and to evaluate the use 'and efficacy of these measures, it is necessary to determine which measures are used and to assess the risk for malaria among travellers (8). The lack of consistent recommendations is understandable because such information is not available to travellers from most countries. A cohort of travellers departing from Kenya was therefore surveyed, in order to determine the use of and compliance with measures to prevent malaria, to assess the frequency of self-treatment for episodes of presumed malaria during travel, and to determine the occurrence of adverse reactions to antimalarial drugs as well as of episodes of malaria during and after travel in Kenya. Bulletin of the World Health Organization, 66 (2): 209-215 (1990) ©D World Health Organization 1990 209 H.O. Lobel et al. Materials and methods A questionnaire, available in four languages (English, French, German, and Italian), was administered to travellers departing from Nairobi airport for Europe on all scheduled flights between 1 and 21 September 1987. The questionnaire was distributed and collected before boarding. Data collected con- cerned the duration and purpose of travel, countries and areas visited within Kenya, preventive anti- malarial measures used, adverse drug reactions experienced, and episodes of suspected or confirmed malaria. Eight to ten weeks later a follow-up ques- tionnaire was mailed to these travellers. The follow- up questionnaire asked about compliance with pro- phylaxis and about malaria episodes after leaving Kenya. To determine the non-response bias, a ran- dom sample of travellers from North America and the United Kingdom who did not return the second questionnaire were contacted and interviewed by telephone. All cases of malaria diagnosed by a physician and all cases of persons hospitalized because of adverse reactions to malaria drugs were investigated by contacting the physician who had treated the patient. The diagnosis of malaria was considered verified if a record was available of a blood-smear examination that indicated the presence of Plasmodium parasites. Compliance with chemoprophylaxis was defined as the regular, uninterrupted use of prophylactic drugs during travel in Africa and for four or more weeks after leaving. Use of adequate antimosquito measures was defined as the use of two or more such measures (bed nets, insect repellents, protective cloth- ing at night, insecticides, staying indoors at night). Travellers in Africa for less than four weeks were defined as short-stay travellers, and those remaining for four weeks to one year as long-stay. Residents of countries outside North America or Europe, travellers who did not go outside Nairobi (malaria transmission is considered not to occur in Nairobi), and travellers who were in Africa for more than twelve months, were excluded from our analysis because their behaviour and experience differ from that of temporary visitors from North America and Europe to malarious areas of Kenya. Statistical significance was determined with the x2 test. Multiple logistic regression analysis was performed in order to identify the factors that influence compliance with chemoprophylaxis. Results Population Questionnaires were completed at Nairobi airport by 8533 (60%) of 14228 travellers boarding 93 flights. Not all passengers, particularly those who arrived late at the departure gate, were given questionnaires. When the number of these passengers was excluded from the denominator, the response rate for com- pleted questionnaires was 75%. A follow-up ques- tionnaire was sent to 6205 of the travellers from Europe and North America, 4612 (74%) of whom responded. Samples of non-respondents in the United Kingdom (120) and the USA (115) were contacted. Use of prophylaxis after departure from Nairobi and the occurrence of adverse reactions and of malaria were comparable for respondents and non-respondents. Respondents at Nairobi airport were from 52 countries, including 3735 residents of European countries, 2707 from North America, 1743 from African countries, and 348 from other countries. To assess the representativeness of the survey popula- tion, the country of residence and the purpose of travel of the respondents were compared with that of all European and North American travellers to Kenya in 1987 (Table 1). Those countries whose residents frequently travel on charter flights directly to the coast of Kenya, such as Germany and Swit- zerland, were under-represented because we did not survey persons using such flights. For the same reason, tourists made up a smaller proportion of respondents than of all travellers to Kenya in 1987. Of the 6442 European and North American travellers, 953 (15%) stayed within Nairobi or were in Africa for more than twelve months and were there- fore excluded from the analysis. The 5489 visitors who travelled outside Nairobi and stayed for twelve months or less included 2418 from North America and 3071 from Europe (for specific countries, see Table 2); 2865 (52%) were female and 2949 (54%) were under 40 years of age. Tourism was the main reason for 76% (4182), business for 6% (325), and visiting friends and relatives for 10% (533). Of all European and North American travellers, 3930 (72%) were in East Africa for less than 4 weeks. American travellers were significantly older than the Europeans (mean age, 44.4 years versus 33.8 years (P <0.001)), stayed a shorter time (1799 (83%) stayed less than 4 weeks versus 1994 (65%, P <0.001)), and were more likely to be tourists (1814 (84%) versus 2184 (71 %, P < 0.001)). The coast ofKenya was visited by 2010 (65%) European travellers but by only 575 (26%) from the USA. In contrast, 823 (27%) Europeans visited game parks only, compared with 1442 (66%) from the USA. Only 236 (4%) visited Lake Victoria (western Kenya). Pre-travel medical advice Of all the European and North American travellers, 5223 (95%) were aware of malaria risk in Africa. WHO Bulletin OMS. Vol 68 1990.210 Malaria Incidence and prevention among travellers to Kenya Table 1: Place of residence and proportion of tourists among all survey respondents and all travellers to Kenya from Europe and North America" Travellers surveyed All travellers Country/region of residence Number % tourists Number % tourists USA 2433 (37.8)b 78.6 60 400 (14.3) 81.3 United Kingdom 1805 (28.0) 68.3 73 100 (17.3) 76.2 Italy 631 (9.8) 75.8 38 100 (9.0) 90.2 France 381 (5.9) 69.2 27 200 (6.4) 86.4 Federal Republic of Germany 302 (4.7) 39.5 111 700 (26.4) 92.2 Canada 274 (4.3) 70.8 8 500 (2.0) 76.5 Scandinavia 175 (2.7) 31.2 15 100 (3.5) 78.1 Switzerland 102 (1.6) 50.3 49 500 (11.7) 93.5 Other 339 (5.2) 71.7 39 800 (9.4) 90.8 Total 6442 (100) 76.7 423 400 (100) 85.9 ' Based on Economic Survey. Nairobi, Ministry of Planning and National Development, 1988. bFigures in parentheses are percentages of the total. Table 2: Percentage of people surveyed who usd malaria prevention measures during travel, by place of residence % using % using % using prophylaxis prophylaxis and AMM' prophylaxis % using Country/region No. regularly with but one or or AMM' no preventive of residence of persons two or more AMM' both inadequate only measures Scandinavia 106 29.2 26.4 36.8 7.5 France 316 32.6 23.7 38.9 4.7 Federal Republic of Germany 219 48.4 30.6 15.1 5.9 Italy 523 36.7 24.3 33.8 5.2 Netherlands 165 32.1 37.0 27.3 3.6 United Kingdom 1559 42.9 33.2 20.3 3.7 USA 2173 58.6 19.8 19.6 2.0 Canada 245 51.8 23.7 22.9 1.6 Other 183 41.8 23.7 30.5 4.0 Total 5489 48.0 25.6 23.1 3.3 'AMM = antimosquito measures (e.g., bed nets, insect repellents, insecticides, protective clothing or staying indoors at night). Most of them (4140 or 75%) obtained advice from medical sources (e.g., physicians, health departments, traveller clinics), prior to departure, about prevention of malaria. Only 623 (11%) relied on their own knowledge of protective measures. This was most frequent as follows: 31 (19%) Dutch travellers, 60 (18%) business travellers, 133 (25%) persons visiting friends or relatives, and 278 (18%) long-stay travellers. Use of prevention measures Almost all European and North American travellers (5309 or 97%) used one or more measures to prevent malaria. During travel, chemoprophylaxis was taken by 5216 (95%) travellers, 4407 (84%) of whom used it regularly. Antimosquito measures were used by 4134 (75%), 3096 (75%) of whom used at least two such measures. However, only 2630 (48%) used prophy- laxis regularly and used adequate antimosquito measures while in Africa. Those least likely to use regular prophylaxis and adequate antimosquito measures were as follows: 84 (26%) business travellers, 173 (32%) persons visiting friends and relatives, 1279 (43%) travellers under 40 years of age, and 599 (38%) long-stay travellers. Country of residence also influenced the use of regular pro- phylaxis and adequate antimosquito measures during travel (Table 2). Only 180 (3%) did not use any preventive measures, including 74 (2%) tourists, 38 (12%) business travellers, and 37 (7%) persons visit- ing friends and relatives. The differences between tourists and business travellers and between tourists WHO Bulletin OMS. Vol 68 1990. 211 H.O. Lobel et al. and persons visiting friends and relatives were highly significant (P <0.001). Sixty-eight drug regimens were reportedly used for prophylaxis, nine of which were used by 79 or more persons (Table 3). The most commonly used drug regimen, chloroquine alone, was used by 2550 (49%) persons who took chemoprophylaxis, ranging from 1672 (77%) American to 3 (2%) Dutch travellers. Chloroquine and proguanil were used by 1184 (23%) persons, ranging from 133 (81%) Dutch to 766 (49%) British and 7 (2%) Italian travellers. Mefloquine was mainly used by French (148 or 47%) and pyrimetha- mine-sulfalene (Metakelfin)e by Italian travellers (132 or 25%). Compliance with a chemoprophylactic regimen during and after travel was examined in travellers who used prophylaxis during travel and who com- pleted both the original and follow-up question- naires. Of 3469 travellers who used chemoprophy- laxis in Kenya, 1793 (52%) used it regularly during travel and for 4 weeks after leaving Kenya. Of the 1676 (48%) travellers who did not comply fully with their regimen, 27% did not use prophylaxis regularly while in Africa and 73% did not do so after depar- ture. Of the latter group, 8% did not take any prophylaxis after departure, 67% did not take pro- phylaxis for a full four weeks after departure, and 24% did not take prophylaxis regularly. Logistic regression analysis showed that compliance was especially poor among people who visited friends and relatives (37%), travelled for more than three weeks (39%), experienced adverse reactions (40%), used proguanil (31%), and among young travellers from the United Kingdom (43%) (P <0.001). C Use of trade names is for identification only and does not imply endorsement by the Public Health Service or the U.S. Department of Health and Human Services. Table 3: Drugs used for prophylaxis by travellers Drug No. of users % of users Chloroquine 2550 48.9 Chloroquine and proguanil 1184 22.7 Proguanil 183 3.5 Mefloquine 177 3.4 Metakelfin 134 2.6 Fansidar 110 4.4 Chloroquine and Maloprim 108 2.1 Chloroquine and Fansidar 94 1.8 Maloprim 79 1.5 Other drugs 578 11.1 Not stated 19 0.4 Total 5216 100.0 Adverse reactions One or more adverse reactions attributed to the use of chemoprophylaxis was reported by 669 (13%) of the 5216 travellers who used chemoprophylaxis. Gastrointestinal side-effects were most frequent. Although no side-effect caused anyone to be hos- pitalized, side-effects reduced compliance signifi- cantly. Side-effects were most frequent for users of chloroquine and proguanil combined, and least so for pyrimethamine-dapsone (Maloprim) users (Table 4). Self-treatnent for presumed malaria Of 5489 travellers, 1715 (31%) stated that they carried drugs to treat a malaria attack. The number that did so ranged as follows: 137 (63%) German travellers to 24 (15%) Dutch travellers; 1308 (76%) of these travellers carried pyrimethamine-sulfadoxine (Fansidar), 100 (6%) chloroquine only, 78 (5%) mefloquine, and 71 (4%) Metakelfin. Only 44 (3%) treated themselves for a presumed malaria attack: 8 (0.6%) of 1242 short-stay travellers and 36 (8%) of 473 long-stay travellers (P <0.001). Malaria Sixty-seven (1%) of 5489 travellers reported having had symptoms of malaria. Of these, 65 (97%) experienced symptoms while abroad and 2 after returning to their country of residence. Eleven (0.3%) of the short-stay travellers and 54 (4%) of the long- stay travellers had a presumed malaria attack while in Kenya. Few of the 65 cases that occurred in Kenya were documented by microscopic examination; malaria was diagnosed by clinical examination by a physician for 25 (38%) persons, in 22 (34%) cases by travellers themselves, and by a physician reportedly after blood smear examination for only 18 (28%) persons. Subsequent contact with physicians revealed that for 14 (78%) of these 18 cases a blood slide was positive for malaria parasites (all P. falciparum); for 4 people the result of microscopic examination was negative. Both cases that occurred after departure from Nairobi were verifiable. Thus, the diagnosis was verified for only 16 (24%) of the travellers who reportedly had malaria. The incidence of these verified episodes of malaria was 3 cases per 1000 travellers per month (Table 5). The relative risk was 14-fold higher for travellers to Lake Victoria than for travellers to game parks (P <0.001). The incidence of all reported episodes of malaria among travellers to Kenya was 12 cases per 1000 travellers per month. Only 1 of the 16 travellers with a verified episode of P. falciparum malaria did not take any chemo- prophylaxis. Prophylaxis was used regularly by 13 of them: 6 used chloroquine, 2 used chloroquine and WHO Bulletin OMS. Vol 68 1990.212 Malaria Incidence and prevention among travellers to Kenya Table 4: Percentage of travellers surveyed who reported adverse reactions to chemopro- phylaxls, by drug Adverse reactions No. of Drug users Eye Cutaneous N/V/D' Other Total Chloroquine and proguanil 1184 1.2 1.0 14.7 1.7 18.6 Chloroquine and Fansidar 94 1.1 4.4 12.0 0 16.4 Chloroquine and Maloprim 108 1.9 0.9 9.3 1.8 13.9 Mefloquine 177 0 1.2 12.1 4.1 16.2 Proguanil 183 0.5 0 10.6 1.2 12.3 Metakelfin 134 0.8 1.5 9.9 0.7 12.2 Chloroquine 2550 1.0 0.8 7.4 3.5 10.9 Fansidar 110 0 1.9 3.8 1.9 7.6 Maloprim 79 0 0 1.3 1.3 2.6 Nausea, vomiting, diarrhoea. Table 5: Incidence of malaria' among travellers surveyed, by area of Kenya visited Person-months of No. Relative Destination exposure of cases Incidence" risk Coast 3341 11 3.3 4.7 Lake Victoria 301 3 10.0 14.3 Game parks 1523 1 0.7 1 Other 237 1 4.2 6.0 Total 5403 16 3.0 " Number of verified cases of P. falciparum infection per month per 1000 travellers. proguanil, 2 used chloroquine and chlorproguanil, 1 used chlorproguanil only, 1 used proguanil only, and 1 used Metakelfin. Prophylaxis was used irregularly by 2 persons, 1 of whom used chloroquine and 1 chloroquine and proguanil. The incidence of P. fal- ciparum malaria was not significantly different among groups by type of prophylactic regimen, but the number of cases was too small to permit inferences about prophylactic efficacy. Discussion Information about travellers' knowledge of malaria, the preventive measures they employ, their ex- perience with malaria, and the occurrence of adverse drug reactions can be obtained from questionnaire surveys of travellers. The accuracy of the information about use of prevention measures cannot be verified, but reported episodes of malaria or of adverse reac- tions that require medical care can be validated by contacting physicians. Traveller surveys have been conducted at airports, among members of tour groups, among clients of a medical advisory service, and aboard charter and scheduled flights (9-1 1).° Campbell, H. Imported malaria in Britain: a study of British residents travelling to malaria endemic areas. M.Sc. thesis, University of London, London, 1984. Such surveys are often cross-sectional and focus on select groups. Also, surveys have usually been limited to travellers from the USA and a few European countries. Our investigation, which was conducted among travellers from many countries returning from an African destination frequently visited by European and North American travellers, demon- strates the importance of follow-up procedures to determine compliance with chemoprophylaxis after travel. Many different drugs and drug combinations were used by travellers for prophylaxis, which ref- lected the different recommendations in the countries of residence and the availability of the drugs. Incon- sistent recommendations may confuse travellers and decrease their compliance. Developing consistent recommendations requires a better understanding of the use of prevention measures among travellers from different countries and of their risk of malaria (8). Less than half the persons surveyed used both regular chemoprophylaxis and more than one anti- mosquito measure during their travels. Only 52% of those interviewed complied fully with chemopro- phylactic regimens during and after travel, and most noncompliance occurred after they had left Kenya. A similar trend was found among travellers interviewed in the United Kingdom (11). Health providers, who WHO Bulletin OMS. Vol 68 1990. 213 H.O. Lobel et al. were consulted by 75% of the travellers, need to emphasize the importance of the correct use of anti- mosquito measures and the need for uninterrupted use of chemoprophylaxis together with several anti- mosquito measures. Health information needs to be targeted especially to long-stay travellers who are less likely to use adequate prevention measures, and who appear to be at higher risk for malaria. Of the long-stay travellers in our study, more than 60% were mission- aries and persons visiting friends or relatives and less than 30% were tourists and business travellers. The recommendation that travellers treat them- selves when they suspect a malaria attack places great responsibility on them. They must interpret the etiology of often vague symptoms and understand the appropriate regimens for self-treatment. This may result in abuse of potentially dangerous drugs or failure to take life-saving action. Recommendations for self-treatment may be more important for long- stay than for short-stay travellers. Few of the malaria episodes in Kenya reported by the travellers surveyed were diagnosed by blood- slide examination, even when the traveller consulted a physician. Comparable data have been reported previously for expatriates in Africa (12,13). This lack of blood-slide examination may be dangerous for the patient because the symptomatology of malaria is not pathognomonic, especially in the early stages of the disease. Because measures to prevent malaria are not always effective, travellers should be aware of the available medical facilities in their area of travel. When they receive medical care, they should request that a laboratory test be performed to diagnose their illness, and they should obtain written documenta- tion of the diagnosis and treatment. Estimates of malaria risk for travellers and of the efficacy of prophylaxis are often based on the incidence of malaria after return to the country of residence (3,14). This may result in an underestima- tion of the real risk, because many malaria episodes occur during travel. In our investigation the incidence of suspected malaria among travellers to Kenya was similar to the incidence of P. falciparum infection in Peace Corps volunteers in East Africa (15 cases per 1000 persons per month), which suggests that most suspected episodes of malaria were indeed malaria (15). International cooperation is needed to develop comparable data bases on the use of malaria preven- tion measures by travellers and their risk for malaria (8). Our investigation has initiated this cooperation. We recommend that similar investigations, using comparable methodologies, be carried out among travellers from different countries to other areas with a risk for malaria. Acknowledgements We appreciate the generous assistance of the Nairobi station managers and personnel of Air France, Alitalia, British Airways, Kenya Airways, KLM Royal Dutch Air- lines, Lufthansa, Pan American Airways, Olympic Air- ways, Sabena, and Swissair. We also acknowledge the assistance of Dr C.G. Nevill and the staff of the African Medical and Research Foundation (AMREF) in Nairobi in contacting physicians in East Africa. R6sum6 Incidence et pr6vention du paludisme chez les voyageurs d'Europe et d'Amerique du Nord se rendant au Kenya En septembre 1987, on a distribue un question- naire en quatre langues a des voyageurs au depart de Nairobi, afin de determiner quelles etaient les mesures de prevention employees contre le paludisme et le risque de contracter la maladie chez les voyageurs se rendant au Kenya. Le taux de reponse a 6te de 75%. Un second questionnaire leur a ete adresse huit a dix semaines plus tard, pour lequel le taux de reponse a ete de 74%. Les 5489 voyageurs provenant d'Europe et d'Am6rique du Nord ayant repondu au questionnaire ont utilise pour la prophylaxie 68 schemas therapeutiques diff6rents, qui 6taient le reflet des diverses recom- mandations formulees dans les pays de residence et des medicaments disponibles. Seuls 48% des voyageurs ont associe une chimioprophylaxie reguliere a au moins deux mesures de protection antimoustiques au cours de leur voyage; sur les 3469 personnes soumises a une chimioprophy- laxie, 52% l'ont observee sans interruption pen- dant leur voyage et au cours des 4 semaines qui ont suivi. La plupart des cas de non-observance se sont produits apres le depart du Kenya. L'obser- vance la plus faible a ete le fait de voyageurs qui avaient rendu visite a des amis ou a de la famille, voyage plus de trois semaines, eu des effets indesirables, pris du proguanil; ce meme pheno- mene a ete retrouve chez les jeunes du Royaume- Uni. Le corps m6dical, qui a 6t6 consulte par 75% des voyageurs, doit insister sur l'impor- tance d'utiliser correctement les protections anti- moustiques et sur la necessite d'une chimiopro- phylaxie ininterrompue associee a plusieurs mesures antimoustiques. Un tiers des voyageurs avaient emporte des medicaments destin6s i traiter une crise de paludisme et 3% d'entre eux ont eu a s'en servir. Soixante-sept personnes (1 %) ont pr6sente des symptomes de paludisme, mais le diagnostic n'a pu dtre confirme que pour 16 WHO Bulletin OMS. Vol 68 1990.214 Malaria Incidence and prevention among travellers to Kenya d'entre elles. Les voyageurs se depla9ant pour de longues durees semblent avoir presente un risque plus eleve de contracter la maladie que les voya- geurs venus pour peu de temps, et l'information medicale doit donc dtre cible sur les premiers. Des etudes du meme type seraient necessaires chez les voyageurs internationaux se rendant dans d'autres pays d'end6mie du paludisme. On ne pourra mettre au point de directives coh6rentes et efficaces que si l'on dispose de donnees com- parables. Cela permettra de diminuer la confusion qui rdgne chez les voyageurs dans ce domaine et ameliorera l'observance des mesures de prevention. References 1. Phillips-Howard, P.A. et al. Malaria in Britain: 1977- 86. Br. med. j., 296: 246-248 (1988). 2. Raeber, P.A. et al. Le paludisme en Suisse: 1982 a 186. Schweiz. Med. Wochenschr., 118: 1261-1266 (1988). 3. Lobel, H.O. et al. Efficacy of malaria prophylaxis in American and Swiss travellers to Kenya. J. infect. dis., 155: 1205-1209 (1987). 4. Health information for international travel. Atlanta, Centers for Disease Control, 1987 (Publication No. (CDC) 87-8280). 5. Bradley, D.J. & Phillips-Howard, P.A. Recommenda- tions for malaria prevention in European countries. In: Steffen, R. et al., ed. Proceedings of the First Conference on International Travel Medicine, Zurich, 1988. New York, Springer-Verlag, 1989, pp. 164-166. 6. Kendler, D. & Keystone, J.S. Chemoprophylaxis for chloroquine-resistant Plasmodium falciparum malaria. Canada diseases weekly rep., 13: 55-57 (1987). 7. Pappaloanou, M. et al. A quantitative approach to recommendations on malaria prophylaxis. Bull. Wld Hlth Org., 66: 477-483 (1988). 8. Development of recommendations for the protection of short-stay travellers to malaria endemic areas: Memorandum from two WHO Meetings. Bull. Wld Hlth Org., 66: 177-196 (1988). 9. Steffen, R. et al. Malaria chemoprophylaxis in 28 712 European travellers to Africa: a follow-up study. In: Steffen, R. et al. Proceedings of the First Conference on International Travel Medicine, Zurich, 1988. New York, Springer-Verlag, 1989, pp. 141-144. 10. Lobel, H.O. et al. Use of prophylaxis for malaria by American travellers to Africa and Haiti. J. Amer. Med. Assoc., 257: 2626-2627 (1987). 11. Phillips-Howard, P.A. et al. Malaria prophylaxis: survey of the response of British travellers to pro- phylactic advice. Br. med. j., 293: 932-934 (1986). 12. Jaatinen, M. et al. Falciparum malaria resistant to chloroquine and Fansidar. Br. med. j., 288: 65 (1984). 13. Harrles, A.D. et al. Malaria prophylaxis amongst British residents of Lilongwe and Kasungu districts, Malawi. Trans. R. Soc. Trop. Med. Hyg., 82: 690-692 (1988). 14. Phillips-Howard, P.A. et al. Short-term travel to malarious areas: malaria risk in British residents. Trav. med. int., 6: 51-60 (1988). 15. Bernard, K. Health risks for temporary residents of developing countries: the U.S. Peace Corps as an epidemiologic model. In: Steffen, R. et al., ed. Proceedings of the First Conference on International Travel Medicine, Zurich 1988. New York, Springer- Verlag, 1989, pp. 37-44. WHO Bulletin OMS. Vol 68 1990. 215