A programme for controlling acute respiratory infections in children: Memorandum from a WHO Meeting*

Me7oranda are state- Les Mimorandums ments concerning the exposent les conclu-Memt]Gl{({tJI U"oranda/ conclusions or recom- sions et recomman- mendations of certain dations de certaines WHO scientific meet- reunions scientifiques X lMl Pw7rrnnmllorand ms ings; they are signed de l'OMS; ils sontby the participants in signes par les partici- the meeting. pants d ces rcunions. Bulletin ofthe World Health Organization, 62 (1): 47-58 (1984) © World Health Organization 1984 A programme for controlling acute respiratory infections in children: Memorandum from a WHO Meeting* The unacceptably high mortality related to acute respiratory infections (ARI) in chil- dren, recognition of the importance of bacteria in the causation of severe acute lower respiratory infection in developing countries, and the established effectiveness of anti- microbial and supportive treatment in averting death make a strong casefor the initiation of an ARI control programme. This should be spearheaded by prototype ARI service activities, delivered through primary health care and backed up by well-coordinated health systems research. Since 1976, when the Twenty-ninth World Health Assemblya initiated action in the field of acute respiratory infections (ARI), several expert groups have met to analyse the problem and a number of re- search programmes have been initiated. The control of ARI is an essential component of primary health care (PHC) programmes and its implementation poses questions that must be studied if the mortality from ARI is to be reduced. A concerted attack on acute respiratory infections in children in developing countries must be based on the available technology and the infrastructure of the evolving PHC programmes. The past few years have seen considerable progress in our understanding of the ARI problem and its susceptibility to inter- vention. In particular, the importance of bacteria as the principal cause of mortality from severe acute lower respiratory infections in developing countries is now clear; effective antimicrobial and supportive treatment is available and therefore many of these deaths can be averted; existing clinical experience has been consolidated into simple case management * This Memorandum was drafted by the signatories listed on pages 57-58 on the occasion of the first meeting of the Technical Ad- visory Group on Acute Respiratory Infections, which was held in Geneva on 7-11 March 1983. Requests for reprints should be addressed to the Chief, Tuberculosis and Respiratory Infections unit, World Health Organization, 1211 Geneva 27, Switzerland. A French translation of this Memorandum will appear in a later issue of the Bulletin. a WHO Official Records, No. 233, 1976, Annex 7, p. 94. plans; and the PHC infrastructure required to utilize these plans is being strengthened rapidly. Thus the time is now opportune for a phased introduction of an ARI control programme. MAGNITUDE OF THE PROBLEM Acute respiratory infections, diarrhoeal diseases, and malnutrition are the principal causes of illness and death in children in developing countries. Where- as diarrhoeal diseases and malnutrition have received considerable attention, acute respiratory infections have often been ignored until recently. Many develop- ing countries are now recognizing the problem ofARI and are seeking ways to deal with it. Constituting a complex and heterogeneous group of conditions, the acute respiratory infections now involve over 300 antigenic types of viruses and bac- teria. These infections can be divided into two main groups: upper respiratory infections and lower respiratory infections. The upper and lower respira- tory tracts are often affected simultaneously or con- secutively and there are also diffuse forms such as the influenza syndrome. This complexity is a source of much variation in the criteria for reporting and classifying ARI as a cause of illness and death. Suffice to say that the group includes influenza, measles, diphtheria, pertussis, sinusitis, acute otitis mnedia, 4376 -47- MEMORANDUM nasopharyngitis, tonsillitis, epiglottitis, laryngitis, tracheitis, acute bronchitis, bronchiolitis and pneu- monia. Mortality The main respiratory syndromes that threaten the lives of children are pneumonia, bronchiolitis, and acute obstructive laryngitis (croup). Bacterial pneu- monia (including bronchopneumonia) as a primary infection or as a complication of viral infection is by far the most frequent cause of death. Registered death rates from influenza and pneu- monia in children are often 20-50 times higher in developing countries than in the developed countries. This difference in mortality between the developed and developing countries cannot be quantified pre- cisely because the methods of diagnosis and reporting are not always comparable. However, some reliable intercountry comparisons can be made with data from special surveys such as the study on childhood mortality in the Americas, 1968-72. In some areas of Bolivia and Brazil, respiratory diseases (mostly infec- tious diseases) were the underlying or associated cause of death in 40-44 per 1000 children below the age of 1 year (a risk 11-14 times higher than in compar- able studies in the USA (California) and Canada (Quebec)) and in 6-8 per 1000 children aged 1-4 years (a risk 50 times higher than those for the USA and Canada). In the developing areas, respiratory infec- tions were frequently an associated cause of death, especially as a complication of measles and pertussis. Another way of estimating the likelihood of death from acute respiratory infections is through the case- fatality ratio. While only a very small proportion of children suffering from pneumonia or bronchopneu- monia die in the developed countries, the case-fatality is 5-1007o in the hospitals of large towns and over 1007o in some rural areas of the developing countries. Morbidity Because in most countries respiratory infections are not notifiable, there is little information about their incidence in the general population. Data from a few community-based longitudinal studies, however, indicate that they are very common. During the first five years of life, a child living in an urban area has (on average) each year 5-8 episodes of acute respira- tory disease, including infections of the upper respira- tory tract. In rural areas the annual incidence per child could be lower, but the published studies do not cover urban and rural areas in a comparable way. Although the overall incidence of ARI is similar in the urban areas of developing and developed countries, there are large differences in the relative frequency and severity of lower respiratory tract infections, in particular pneumonia. While their an- nual incidence is between 30 and 40 per 1000 chil- dren under 5 years old in urban areas of the United States, it appears to be at least double (from 70 to 100 per 1000) in developing countries and can reach levels of 500 per 1000 among malnourished children. Acute respiratory infections are a leading cause for people having to use the health services in both developed and developing countries. They account for 20-400%o of the children attending outpatient clinics and 12-35%o of admissions of children into hospitals. The problem of nosocomial ARI in children has rarely been investigated in developing countries. At the National Children's Hospital, San Jose, Costa Rica, acute respiratory infections were responsible for 35% of all nosocomial infections in 1982. Among 580 nosocomial respiratory infections, 148 were pneu- monia and bronchopneumonia. The case-fatality rate was 16%o; Gram-negative bacilli were isolated in 9007o of the infections diagnosed by lung or pleural puncture. ETIOLOGICAL AGENTS Most of our knowledge on the etiological agents of respiratory infections in children is derived from observations in developed countries where viral infec- tions predominate. Viral agents have been estimated to be responsible for over 9007o of cases of commu- nity-acquired acute disease of the upper respiratory tract, and a considerable, if lesser, proportion of cases of the lower respiratory tract. ARI of viral origin may be complicated (or accompanied) by bacterial infection. Virological studies in the USSR of children with indisputable bacterial pneumonia revealed viral antigens in more than 50%o of cases. It has been postulated that viral infections pre- dispose to bacterial invasion. Respiratory viruses were shown to impair mucociliary clearance and the functions of the polymorphonuclear neutrophils, alveolar macrophages, and T-lymphocytes, thus allowing pathogenic bacteria to invade the lower respiratory tract and start an infection. Measles and influenza viruses appear to be particularly notorious in this respect. The most frequent viral agents of lower respiratory tract illnesses in infants and young children are the respiratory syncytial viruses (RSV), the adenoviruses, and the parainfluenza and influenza A and B viruses. Some agents are more often associated with a specific syndrome, e.g., RSV with bronchiolitis, and the para- influenza viruses I and 2 with croup. However the same syndrome may be caused by different agents and the same agent may cause a wide range of different syndromes. 48 ARI CONTROL PROGRAMME 49 Among the bacteria, Streptococcus pneumoniae and Haemophilus influenzae are the most frequent agents of pneumonia in young children. H. influenzae may also cause croup. Staphylococcal pneumonia predominates in the first six months of life and is probably a complication of measles in older children. Mycoplasma pneumoniae is more frequent in chil- dren over the age of 3 years. Streptococcus pyogenes is important as a cause of pharyngo-tonsillitis with a risk of subsequent development of rheumatic fever, heart disease, nephritis, and permanent damage to the kidneys, especially in children of 5-12 years of age. Recent reports suggest that Chiamydia trachomatis may be an important agent of subacute or protracted pneumonia in the newborn in nontrachomatous areas such as the USA. Infection has been associated with genital carriage of the organism by the mother. So far there is no information about the incidence of chlamydial respiratory infections in the newborn of trachomatous or nontrachomatous areas in develop- ing countries. Evidence is now accumulating that bacterial path- ogens play a greater role as a primary or secondary cause of lower respiratory tract disease in developing than in developed countries. The prevalence of bac- terial infections may be favoured by the impairment of immunity in malnourished children, insufficient coverage of immunization programmes, low level of education, poor hygienic conditions, and the lack of early health care or of any care at all. In investigations conducted among children with pneumonia who had not received previous antimicrobial treatment (in Brazil, Chile, India, Nigeria and Papua New Guinea), around 60'7 of lung aspirates yielded a bacterial growth.b The pattern of colonization and carriage of pathogenic bacteria in the upper respiratory tract also differs considerably. It has been found that S.pneu- moniae can be isolated from up to 1000o of healthy children in developing countries, whereas this propor- tion is usually not higher than 500o in the developed countries (except in the underprivileged groups where higher rates may be observed). AVAILABLE CONTROL TECHNOLOGIES Immunization The available measures for preventive immuniz- ation against respiratory infections are rather limited at present. They include measles vaccine which is safe h The mention of lung aspirates does not imply any endorsement by the group of this method for epidemiological surveys. But it acknowledges that it offers the best way to determine the bacterial etiology of pneumonia in children up to now. Although lung punc- ture may be safe in experienced hands, it is not a routine procedure to be recommended for epidemiological surveys. It may be useful in clinical practice in hospitals for the establishment of the etiological diagnosis in critically ill children who fail to respond to therapy. and effective in preventing the disease and its pul- monary complications, and pertussis and diphtheria vaccines which are recommended for widespread use. These three vaccines are included in the WHO Ex- panded Programme on Immunization, but more effort is needed if they are to achieve their full poten- tial for reducing morbidity and mortality in develop- ing countries. Polyvalent pneumococcal vaccines have been made that include serotypes causing most of the pneu- mococcal infections in many parts of the world. Although effective in older individuals, they cannot at this stage be advocated for use in infants and young children; critical vaccine serotypes are inadequately immunogenic in childhood when morbidity and mor- tality are at their highest. A preliminary field trial in Papua New Guinea nevertheless showed some effect on mortality and the results of a current large-scale trial in children are awaited with interest. Meanwhile, efforts to improve immunogenicity in early childhood should be pursued. The same problem has been observed in the first trials with the H. influenzae vaccine. Staphylococcal vaccines at present are not available. Immunization against viruses producing severe lower respiratory infections is desirable. Inactivated parainfluenza vaccines tested so far do not prevent infections and may exacerbate disease. Live attenu- ated RSV vaccines are being developed but no strain has yet been brought to clinical trial. In the case of influenza vaccine, its short efficacy and the antigenic drift/shift of the virus put it generally beyond the resources of developing countries; it has no place at present in preventing ARI in children or in the control of outbreaks. Chemoprophylaxis There is evidence that antimicrobials used prophy- lactically are of no benefit against bacterial infec- tions, except in a few indications such as the use of penicillin (or erythromycin) for children with re- current streptococcal infections. It is now possible to prevent some virus infections by antiviral prophylaxis: amantadine and riman- tadine for influenza A, enviroxime for rhinovirus infections, and intranasal interferon for coronavirus and rhinovirus infections. At present, none of these can be recommended for mass application. Non-specific measures Improvements in socioeconomic conditions almost certainly contributed to the decrease in mortality from ARI in the developed countries before the intro- duction of antimicrobial chemotherapy. While the mechanism responsible for the decrease is not known, reduced overcrowding in sleeping areas, increased 50 MEMORANDUM access to health services, better child care facilities, and improved nutrition must all have contributed. Recent surveys confirm that exposure to adverse social conditions increases the risk of acute lower respiratory infections in young children; the risk is further increased in the case of exposure to high levels of air pollution. Cigarette smoking by parents has been associated with an increased incidence of acute respiratory dis- eases in childhood. The effect of smoke (from burn- ing of wood or other fuel) on the diseases of the respiratory tract has not been adequately studied in developing countries. Resistance to ARI in infants is also related to birth weight and feeding habits. Low birth weight increases the risk of ARI and death during infancy; the incidence of low birth weight in many developing countries varies between 2007o and 40%. Breast-feeding during the first 4-6 months of life is associated with a reduced incidence of pneu- monia and bronchiolitis as compared to formula feeding. Changes in non-specific factors (such as improve- ment in living standards) and the control of domestic and environmental pollution are slow; their effect on the incidence of respiratory diseases can therefore only be measured in the long term. Infant feeding and parental smoking habits can be modified through a community education programme which may con- tribute to reducing the incidence and severity of respiratory infections in very young children. Case management ofARI Discrimination of casesfor action in management. A number of decision trees and flow charts based on selected, easily recognizable, signs and symptoms of acute respiratory infections have been developed to facilitate the process of discrimination and timely decision-making.' The critical decisions are whether or not to give antimicrobials and whether or not to refer to higher levels of care. A difficult part of the programme is the develop- ment of a system that can be used by semi-literate parents and primary health care workers. Detailed and clear instructions must be provided to PHC workers, both through manuals and training programmes. Their capacity to discriminate ARI into different management categories is crucial to the success of the programme. Antimicrobial therapy. By far the most common causes of bacterial pneumonia in children are S. pneu- moniae and H. influenzae. Therefore the use of anti- microbial drugs is a life-saving measure in both devel- oped and developing countries but it has often been ' Clinical management of acute respiratory infections in children: a WHO Memorandum. Bulletin of the World Health Organization, 59: 707-716 (1981). applied inadequately or irrationally. The main issues concerning antimicrobial drugs that require consider- ation are (a) decision to use antimicrobial drugs at the PHC level; (b) choice of drugs; (c) dosage; and (d) duration of therapy. The indications for the use of antimicrobial therapy in the clinical management of ARI may vary from country to country depending on the following factors: (a) information on the prevailing bacterial path- ogens; (b) pattern of bacterial sensitivity to drugs; (c) policy regarding the use of antimicrobials and of intramuscular injection by the PHC worker; (d) acceptability of the form in which medication is given to the local people, e.g., injections, tablets, syrups; (e) cost of treatment; and (Jf) the available referral support. Parenteral penicillin is generally the drug of choice for the initial treatment because of its high effective- ness and low cost. The selection of the best initial drug treatment is more difficult when the PHC worker is unable or is not allowed to give injections. Oral penicillin, long-acting sulfonamides, ampicillin (or amoxicillin) or even co-trimoxazole can be considered as possible choices. Provisions should be made for a second standard antimicrobial to be used at the first referral level. Chloramphenicol and oxacillin (or a similar drug effective against beta-lactamase-producing bacteria) may be used. Supportive treatment at the PHC level. Supportive treatment has an important role in the management of acute respiratory infections. Children with ARI may suffer from anorexia or have difficulties in sucking due to blocked airways and troublesome breathing. In many regions, the hydration and feeding of infants during an acute respiratory infection and following recovery from the illness are inadequate because of parental ignorance and taboos. When this happens repeatedly as a result of consecutive attacks of ARI, there may be severe malnutrition and an increased risk of a fatal outcome from acute lower respiratory tract infection. During ARI, breast-feeding should be continued; moreover, the quantity of food and liquids taken during the ill- ness and during the recovery phase from ARI should be increased if the child is to recuperate rapidly. Further supportive measures include clearing the child's nose, providing warmth in cold weather, and increasing the moisture in the air (by simple measures, e.g., hanging wet clothes in the room) to soothe the upper respiratory passages. Tight clothing should be loosened because of possible interference with the breathing. ARI CONTROL PROGRAMME Most sick children need not be brought to the health centres and hospitals. They can be managed at home provided that the primary health care workers and the families know how to give supportive treatment. Referral support. Cases of severe acute lower respiratory infection must be referred immediately if they cannot be managed at the PHC level. Other cases of ARI may require referral because they fail to respond to initial treatment. Such children should be referred within hours rather than days. The referral facilities should have available some alternative anti- biotics and intravenous therapy. Whenever possible, they should also have available oxygen therapy, aspiration equipment, and aerosol therapy, and the staff should be skilled in their use. Workers at these centres should be conversant with the management of severe acute laryngitis and severe bronchospasm, and the use of digitalis for heart failure in children with ARI. The provision of appropriate support will increase the effectiveness of PHC workers and their accept- ability in the community, particularly if the links between them and the referral facility are good. Health education Effective case management depends on getting the community informed and involved, otherwise people will not utilize the available health care facilities or will do so only late in the course of the disease. Health education must: (a) increase the capability of families in differen- tiating moderate and serious respiratory illness from mild disease; (b) educate the community regarding simple sup- portive therapy; (c) promote timely immunization against measles, pertussis and diphtheria; (d) promote breast-feeding of infants and proper nutrition of all children; (e) reduce parental smoking and other domestic air pollution. The education effort must be based on an under- standing of local health behaviour, on what medical care can offer, and on how to make the best use of the available services. AVAILABLE ARI CONTROL TECHNOLOGIES FOR THE PHC LEVEL IN DEVELOPING COUNTRIES Three of the ARI control measures mentioned above offer imnmediate potential benefits for children in the developing countries. (1) Improved and standardized case management of acute respiratory infections at both the primary health care and the referral levels. (2) Health education in the prevention of mor- bidity and mortality from ARI. (3) Strengthening of immunization to prevent diph- theria, pertussis, measles, and childhood tuber- culosis. Case management There is ample evidence that the introduction of antimicrobials in the industrialized countries was as- sociated with an accelerated reduction of mortality from many bacterial diseases and particularly from pneumonia in children and young adults. Examples are also available to show that the application of stan- dardized case management, including antimicrobial treatment and supportive measures through an inten- sive programme of training and supervision of pri- mary health care workers, produces a measurable impact on the mortality from pneumonia in children in the rural areas of developing countries. In addition, serious and possibly life-threatening and disabling complications can be prevented by the use of anti- biotics in purulent pharyngitis, purulent tonsillitis, sinusitis, and acute otitis media. The rational use of antimicrobials should be guided by feedback from surveillance of common pathogens and information on their susceptibility to drugs. Where such surveil- lance is impossible, the programme's effectiveness at the very least should be monitored by an assessment of the clinical outcomes. In case management, the main emphasis of the supportive measures lies in the continuous oral hydration and feeding of sick chil- dren to keep up the physiological fluid and energy balance. Doubts have been expressed about the effectiveness of standardized case management in the prevention of childhood deaths in situations where little is known about the changing pattern of etiological agents, where the prevalence of malnutrition is high, and where poverty and environmental deprivation are important contributors to the problem. Indeed, the epidemiological, clinical and etiological knowledge concerning ARI may not always be adequate to permit the design of an optimal case management strategy that could confidently be expected to be effective. Nevertheless, the available information is sufficient to support the assumption that it is possible to reduce mortality in an important proportion of severe acute lower respiratory infections in children in developing countries. Given the extent of our present knowledge, there is no justification in delaying any further the commencement of an ARI control programme. In most developing countries insufficient guidance 51 52 MEMORANDUM is given to health workers at the primary health care and referral levels on how to deal with the large numbers of children suffering from ARI who daily attend the health units. A service-oriented pro- gramme is therefore needed to rationalize the use of antimicrobials, standardize the application of sup- portive measures, and provide clear criteria for the referral of severe cases on the basis of current clinical practice. Such a programme would be within the spirit and letter of the Alma Ata Declaration on Primary Health Care which calls for, inter alia, appropriate treatment of common diseases and injuries, and pro- vision of essential drugs.d Purely preventive programmes alone, e.g., the expanded programme on immunization (EPI) and pollution control, are inadequate and must be complemented by an active control programme for promoting better management of ARI. Community support will increase if it is demonstrated that most common diseases in children can be tackled in the villages and mortality thereby decreased. Education Education of families in child care practices related to ARI is essential for the implementation of any ser- vice programme. The primary health care approach in most developing countries holds out great promise for the introduction of simple case management through active involvement of primary health care workers and families. By the use of proper teaching and management techniques, the community's accept- ance of such technology can be ensured. Immunization Immunization against diphtheria, pertussis, measles and childhood tuberculosis is already part of the expanded programme. These four disease con- ditions contribute heavily to childhood mortality in many developing countries, where strengthening of EPI activities should be given very high priority. PHASED IMPLEMENTATION OF A SERVICE-ORIENTED PROGRAMME The service-oriented programme on ARI should be started in phases so that the effectiveness of the vari- ous components could be tested gradually and then built up at various levels before establishing a national control programme with wide application. Initially the programme should be in localities where primary health care services, the EPI, and diar- rhoeal disease control programmes have already been d Alma Aa, 1978. Primary Health Care. Geneva, World Health Organization, 1978 (Health for All Series, No. 1). adequately implemented. The first phase should provide answers to questions concerning the variations in ARI that are character- istic of the country, and indeed different parts of the country. The main questions are: - What are the clinical types ofARI cases found in rural and urban areas? - What are the apparent causative organisms in severe cases? - Which cases require therapeutic intervention? - How are cases diagnosed, referred, and treated? - What forms of case management should be ap- plied by workers at the primary health care and referral levels? - What are the appropriate contents and method- ology of the educational process for changing the behaviour of families and for increasing their ability to recognize mild and severe ARI, provide supportive care, and bring their children earlier for specific treatment? Supervision and monitoring of the prototype control programme in the selected areas will provide information for evaluating the activities. However, the main objective of the ARI control programme is to reduce the deaths from acute respiratory infections in children under 5 years old; eventually the critical element of evaluation, as in the diarrhoeal diseases control programme and EPI, is the accurate surveil- lance of specific mortality. An evaluation design, based on observations and assessment, has been outlined by a WHO Working Group on Operational Research in ARI for intro- duction as part of overall evaluation of primary health care interventions in selected areas of a country. If applied in different areas and started at different times, additional comparisons could be made. The evaluation should make it possible to determine such variables as the degree of utilization of the referral services for serious cases of ARI by primary health workers, the kind of resources used by them in case management, the extent of delay in seek- ing treatment, and the type of management and treat- ment used in children who died and in those who recovered from severe acute lower respiratory infections. When answers to these problems are available, health administrators should be able to revise the initial programmes and provide an appropriate ARI control programme for the whole country. National implementation of the service-oriented programme will depend on strengthening the ability of health workers at all levels to manage acute respira- tory infections and to educate the entire community about these infections. Thus, the national programme will require: ARI CONTROL PROGRAMME 53 (a) the designation of a national manager, who should also be closely connected with the EPI and diarrhoeal disease control programmes; (b) training of health workers at all levels in the discrimination of severe respiratory infections, the selection and use of different forms of treatment (including supportive measures and antimicrobial treatment) and referrals, and the education of the community. (c) training of managers at the intermediate level and their supervision; (d) an uninterrupted supply of equipment and drugs; and (e) a management information system. The programme must be coordinated and in- tegrated with an effective referral system within the existing health care structure. The cost of the programme in terms of essential drugs at the primary level has been estimated to be in the range of $0.20-0.50 per case of moderate or severe ARI. SURVEILLANCE AND MONITORING OF A NATIONAL ARI CONTROL PROGRAMME The development of a simple mortality reporting system that can reflect the effects of the ARI control programme should be a priority. Close collaboration with the EPI and diarrhoeal diseases control pro- gramme is recommended for this purpose. Surveillance is an important component of any service-oriented programme. A number of ap- proaches of varying complexity are available and a number of models have been tested, including the ARI sentinel unit approach which has already been describede and examples of which have been estab- lished in a number of countries. Resource limitations will dictate what can be done. At the very least there should be monitoring of mortality. Every effort should also be made to determine the prevailing respiratory pathogens and bacterial drug sensitivity. Routine surveillance can be initiated by the country implementing an ARI control programme and can be complemented by systematic independent evaluation. Constraints in the training and service components and in the implementation of the case management programme should be identified early and the pro- gramme should be reviewed regularly so that appro- priate adjustments can be made. The formation ofARI committees at national level, and support to a limited number of collaborating centres are considered as the starting point for an See, for example, WHO Technical Report Series, No. 642, 1980(Viral respiratory diseases: report of a WHO Scientific Group). effective monitoring programme. This should be followed by a network of centres built up at the national and international level. Evaluation of the programme as a whole, in order to demonstrate how far the reduction of ARI mor- tality can be achieved, is equally necessary but demands special resources and will not be possible everywhere. The epidemiological evaluation of the programme should be conducted in such a way that the effects of the intervention can be distinguished from other factors influencing the results. A classification of acute respiratory infections should be designed which will provide rational guid- ance for case management decisions and can also be used for epidemiological purposes. The classification must be simple to ensure the uniformity required for national and international comparisons of data. The development of such classifications is in progress, but wide experience should be gathered in the field before any definite system can be recommended. RESEARCH Research is an essential component of the pro- gramme to control acute respiratory infections in chil- dren/ Clinical, laboratory, epidemiological and oper- ational studies may be carried out either on patients who seek health care or may be population-based. Health systems research (operational studies) After a standard plan of case management has been elaborated on the basis of clinical observations and special studies, its applicability within the compre- hensive health system and its effectiveness in terms of mortality reduction must be determined. A prototype programme should be carried out in a population of sufficient size (e.g., 4000-5000 children aged 0-4 years). In countries with limited resources it may not be practical to do anything more than study the following: - the application of the standard plan for case management by primary health care workers; - the ability of families to recognize mild and severe forms of ARI; - the provision by mothers of supportive care for the sick child; - the behaviour of families in relation to health care practices, and their utilization of the health services; - the logistics of drugs and supplies; - the recording and reporting system; - the referral system; f Guidelines for research on acute respiratory infections: Memorandum from a WHO meeting. Bulletin of the World Health Organization, 60: 521-533 (1982). MEMORANDUM - the links with the other related components of primary health care, such as EPI, control of diar- rhoeal diseases, essential drugs, and nutrition. Example ofeffective health systems research. In an ongoing health systems research project in Egypt, the feasibility of a standard case management plan for ARI in children in a rural community setting is being investigated. During the first phase, nurses, after two weeks of orientation, demonstrated competence in differentiating upper from lower respiratory tract infections and in screening sick children for manage- ment by the physicians. The study showed that peni- cillin G is effective in the treatment and this can be successfully applied at the PHC level. Cough and runny nose can be used as an entry point for education of the mothers and families regarding ARI manage- ment. The effect of the programme on ARI mortality will be assessed during the second phase of the study, which is still continuing. Studies based on hospital and health unit data These studies are the usual starting point for ARI research. As they are concerned with ill children, they are also called disease-based studies. Their aims are: - to provide a detailed clinical description of ARI; - to compare the characteristics of children with mild illnesses and those with more serious disease; - to correlate the etiological agents with the symp- toms and syndromes; - to determine current management practices before a service programme is introduced; - to carry out controlled trials of antimicrobial therapy and of supportive measures. The information will be used to arrive at a clinical classification and standardized case management methods for both the primary health care and the referral levels. Example of effective disease-based research. An example of effective institution-based research is the study now being carried out in 5 hospitals and 13 rural health centres in the United Republic of Tanzania. The aim is to describe the relative frequency of clinical ARI presentations in children, as well as how they are now managed by doctors and with what results, before standardized diagnostic and therapeutic pro- cedures are introduced. From a preliminary analysis of the data, it appears that deaths among 907o of ARI cases presenting for care was due to pneumonia alone or in combination with another diagnosis. The extensive use of antibiotics for both upper and lower respiratory tract infections in children who presented at the health centres and the use of multiple drugs were also noted. The study illustrates the need for rationalizing the use of both antibiotics and other drugs. Another important difference noted is that between the symptoms reported and the diagnosis offered. The high mortality from pneumonia reported in Tanzanian hospitals and communities appears to be due to patients being taken to hospital too late or not at all. This is a strong reason for extending the health services to include those who are not covered by the existing health facilities. A pilot programme to extend primary health care services to unserved villages and to improve the management of cases in hospitals and health centres is now under way in one district. Population-based epidemiological studies Population-based studies provide a comprehensive view of the natural history of respiratory infections in the community through the measurement of mor- bidity and mortality rates, the description of indi- viduals and groups at special risk, and the identifi- cation of determinants and severity of illness. They also make it possible to carry out controlled trials of vaccines and other interventions that can be applied to prevent morbidity or mortality. Population-based research is costly because it requires large populations and specially trained field workers. Such research should therefore be under- taken after careful planning. Example of an epidemiological study. The results of a longitudinal study conducted within a monitored urban population (comprising 400 households) and on hospitalized children from a surveillance popu- lation of 25 260 in Metropolitan Manila, Philippines, identified poor socioeconomic status, protein-energy malnutrition, overcrowding, and lack of DPT im- munization as important risk factors for ARI. Investi- gation of the prevailing knowledge, attitudes, and practices in this population revealed poor utilization of the existing health care facilities by the com- munity for children with ARI. The prognosis was poor in children with severe lower respiratory infec- tion if they were severely malnourished, if the illness had complications, or if the patients were brought for medical care more than 2 weeks after the onset of the illness. Considering the above-mentioned risk fac- tors, an intervention programme has been planned for strengthening primary health care through the intensification of the existing EPI and nutrition pro- grammes, for training health providers at the PHC level towards improving case management, and for providing community health education. WHO ACTIVITIES IN ARI CONTROL In WHO's Sixth General Programme of Work (1978-83), it was recommended that the Organization 54 ARI CONTROL PROGRAMME 55 should extend its programme for control of tubercu- losis to "control of communicable diseases of the respiratory tract which as a group form one of the principal causes of morbidity and mortality in many countries". In 1979 the Thirty-second World Health Assembly adopted a resolution (WHA32.33) request- ing the Director-General "to stimulate and to inten- sify the involvement of Member States in the control of respiratory diseases..." and "to accord high priority to research activities for the development of simple and effective methods..." of control. Between 1979 and 1982 several meetings and consultations were organized by WHO to review the possible scientific and operational background for the ARI programme and to advise the Organization in specific problems. These meetings were held to review in the ARI field the following: epidemiology and etiology, diagnosis, clinical management, control measures, and priorities for research. Although each of these aspects is complex, the expert groups made provisional recommendations on the clinical manage- ment of ARI, use of rapid laboratory techniques for viral and bacterial diagnosis, general guidelines for research in ARI, and guidelines for operational research on ARI. To ensure uniformity of approach and better coordination of ARI activities carried out by WHO in the Regions and also by other organiz- ations and institutions, several basic protocols cover- ing epidemiological, institution-based, and oper- ational research have been developed, arrangements for the provision of reagents for diagnosis and surveil- lance of ARI have been made, and several publi- cations have been issued. In 1982, the World Health Assembly approved the Seventh General Programme of Work (1984-89) in which ARI control appears as one of the seventeen programmes in the section on disease prevention and control in the classified list of programmes. As stated in this programme, the aim of ARI control activities will be to foster national and international action so that: (1) by 1985 a set of alternative strategies for inter- vention at the community level will have been devel- oped on the basis of operational and basic research to meet different national situations for the reduction of mortality from acute respiratory infections, par- ticularly in children; (2) by 1989 most developing countries will have formulated and, to the extent of available diagnostic and treatment facilities, implemented a national pro- gramme for the control of acute respiratory disease as an integral part of maternal and child care within primary health care. Research on acute respiratory infections was also discussed at the twenty-fourth session of the Advisory Committee on Medical Research (ACMR) in Geneva in October 1982. The ACMR agreed that both basic and applied research on acute respiratory infections should be further promoted, and recommended that the subject should be discussed at the regional level by the respective regional ACMRs, and the WHO Regional Offices should prepare a status report on current ARI activities for review at the next session of the ACMR.9 The subject of acute respiratory infections was also discussed, for the first time, at the UNICEF/WHO Intersecretariat Meeting in October 1982 and by the UNICEF/WHO Joint Committee on Health Policy in February 1983. A medium-term programme on ARI for the years 1984-89 has been drafted with the participation of the WHO Regional Offices. CONCLUSIONS AND RECOMMENDATIONS Development of the ARI control programme Enough knowledge and technology is already avail- able for countries to initiate an ARI control pro- gramme in a phased manner. The programme, com- prising a service and a research component, should start with the introduction of simple measures at the primary health care level and progressively provide technical support at higher levels. Service component. Every child with a severe acute respiratory infection, or at special risk of developing such an infection, should receive appropriate medical care which will ensure, as far as possible, his survival and restoration to health. An additional benefit from such a programme is the likely reduction of disability stemming from ARI. The proposed ARI service programme comprises the following: (a) Case management, based on early discrim- ination of mild and severe ARI by families and primary health care workers, supportive treatment, antimicrobial treatment, and timely referral to a higher level of care. (b) Health education on the causes of ARI, recog- nition of the various forms of ARI, and action to be taken by the community. (c) Immunization against measles, diphtheria, per- tussis, and tuberculosis. It is essential to recognize that the success of the case management strategy will require: - extensive training of primary health care workers; 9 The ACMR in October 1983 agreed with the research pro- gramme on ARI and its priority ranking. A realistic approach to start with would be health systems research to find the most appropriate control technology now available and to identify the managerial requirements for its application in developing countries. MEMORANDUM - the strengthening of maternal and child health care within the context of primary health care activities; - the provision of effective first-line referral ser- vices for severe cases identified by primary health care workers; and - a built-in system of evaluation which will guide the operation of strategy and monitor mortality. Health education and community involvement should be actively pursued to ensure early recognition of severe cases, prompt institution of appropriate therapeutic measures and maternal and child health care practices which may contribute to the prevention of ARI in children. The Expanded Programme on Immunization is essential for preventing morbidity and mortality from measles, pertussis, diphtheria, and childhood tuber- culosis and should therefore complement the ARI control programme. Research component. The research component of the ARI control programme is essential for further development, implementation, and evaluation of the ARI service component. It should conform to the standards prescribed by WHO on ethical aspects of research on human subjects. Health systems research will have emphasis on the following: - improvement of primary health care through better case management of ARI; - development and evaluation of a simple management-oriented classification of ARI; - evaluation of procedures for case management at different levels of health care; - evaluation of community involvement; - development and evaluation of improved child care practices. Institution-based studies should be concerned with detailed clinical classification, evaluation of the avail- able techniques for diagnosis, and development of new rapid techniques, and with controlled treatment trials and studies on the immunology and pathophysi- ology of ARI. Epidemiological studies should be concerned with the clinical and microbiological aspects, other factors determining morbidity and mortality, identification of high-risk groups, and social and behavioural deter- minants of ARI. Collaborating centres should be developed, strengthened and reorientated in support of research activities. Centres should be encouraged to give priority to studies directed to strengthening the scien- tific basis of case management, in accord with the operational priorities. Implementation of the programme Effective implementation of the service and re- search components of the programme is highly rel- evant to Member States and therefore calls for ad- equate management staff at the international level. The pace of the proposed medium-term pro- gramme can and should be increased and the targets brought forward; to do this will require an increase in regular budgetary and extrabudgetary funds. Although the development of an ARI control pro- gramme is a national responsibility, bilateral and multilateral international cooperation will be most beneficial at the outset and will be needed to over- come any obstacles. Member States should be encouraged and assisted in the development of the activities described above to assess the magnitude of the problem and to take appropriate action. The ARI control programme should respond to the needs of the individual countries. In countries where mortality from ARI has been reduced substantially, emphasis should be on the development and im- plementation of measures to reduce morbidity. The ARI control programme should link up with other primary health care programmes, particularly for control of diarrhoeal diseases and the expanded programme on immunization, because they have a common final path and can enrich each other and pro- duce a multiplier effect. Common needs in service delivery, research and programme evaluation, super- vision of primary health care, and training should be identified. National workshops will be required to sensitize health administrators to the immediate need for ARI control within the primary health care programmes. Provisions should be made for incorporating the ARI component in the training programmes for primary health care workers and in education of community leaders, school teachers, and families. Development of suitable manuals, especially those for primary health care workers, and health education messages for use in training programmes and case management is a priority. Laboratory support for surveillance, monitoring and research should be secured. This requires: - organizing training workshops for laboratory workers; - making available quality control reagents; - organizing reference facilities for quality con- trol in the performance of tests and for further identification of agents. The existing information systems should be ex- panded for the benefit of those concerned with the operation of ARI control programmes and research 56 ARI CONTROL PROGRAMME activities; an ARI news bulletin should be issued periodically. Global relevance of the programme The needs of industrialized countries in the control of acute respiratory infections should be considered, particularly with reference to age groups other than those under 5 years. Special attention should be directed to basic research, including the development of new diagnostic methods, drugs, and vaccines. Comprehensive surveillance in the service of a global ARI control programme Surveillance should constitute an integral part of the programme. In selected areas a programme oriented surveillance system containing epidemio- logical and laboratory components should be es- tablished. Surveillance and monitoring of the pro- gramme should be used as a management tool in strengthening the national ARI control programme. Classification A classification of acute respiratory infections is required to meet more adequately the requirements of epidemiological surveillance and in accord with case management criteria. Attempts are being made to develop ARI classifications which will serve the needs of epidemiologists in the conduct of field surveys and of clinicians in reaching management decisions. The results of these studies should be analysed and taken into account in formulating the classification. * * R. M. Douglas (Chairman), The University of Adelaide, Department of Community Medicine, Royal Adelaide Hospital, Adelaide, Australia V. Kumar (Rapporteur), Department of Community Health, Postgraduate Institute of Medical Sciences, Chandigarh, India D. L. Miller, Academic Department of Community Medicine, St Mary's Hospital, London, England E. Mohs, Children's Hospital, San Jose, Costa Rica F. D. Mtango, Ministry of Health, Dar es Salaam, United Republic of Tanzania V. Tatochenko, (Vice-Chairman), Acute Respiratory Disease Department, Institute of Pediatrics, Academy of Medical Sciences of the USSR, Moscow, USSR. T. E. Tupasi, Laboratory Research Division, Min- istry of Health, Research Institute for Tropical Medicine, Manila, Philippines Other participants A. Aslund, Permanent Mission of Sweden to the United Nations Office and other International Organizations at Geneva, Switzerland J. Cowcher, Permanent Mission of Australia to the United Nations Office at Geneva, Switzerland L. Gothefors, Department of Pediatrics, University of UmeA, UmeA, Sweden B. Hamza, National Institute for Child Health, Tunis, Tunisia A. M. Makki, UNICEF, San'a, Yemen Arab Re- public A. Meltzer, Tropical and Infectious Diseases, Health Sciences Division, International Development Research Centre, Ottawa, Canada D. Neuvians, German Agency for Technical Co- operation Ltd, Eschborn bei Frankfurt am Main, Federal Republic of Germany J. Pillet, Population, Health and Nutrition Depart- ment, World Bank, Washington, USA WHO Secretariat B. Abdelmoumene (Consultant), Research Pro- motion and Development, World Health Organiz- ation, Geneva, Switzerland G. Alleyne, WHO Regional Office for the Americas, Washington, DC, USA F. A. Assaad, Division of Communicable Diseases, World Health Organization, Geneva, Switzerland D. Barakamfitiye, WHO Regional Office for Africa, Brazzaville, Congo T. A. Bektimirov, Virus Diseases, World Health Organization, Geneva, Switzerland J. Borgofio (Temporary Adviser), International Health Department, Ministry of Health, Santiago, Chile H. G. ten Dam, Tuberculosis and Respiratory In- fections, World Health Organization, Geneva, Switzerland T. Kereselidze, Bacterial and Venereal Infections, World Health Organization, Geneva, Switzerland J. Kostrzewski (Representative of the Advisory Committee on Medical Research), Department of Epidemiology, State Institute of Hygiene, Warsaw, Poland R. Krzysko, WHO Regional Office for South-East Asia, New Delhi, India S. K. Lam, Virus Diseases, World Health Organiz- ation, Geneva, Switzerland J. Leowski, Tuberculosis and Respiratory Infections, World Health Organization, Geneva, Switzerland S. K. Litvinov, World Health Organization, Geneva, Switzerland F. Luelmo, WHO Regional Office for the Americas, Washington, DC, USA 57 MEMORANDUM M. P. Mitrofanov (Consultant), Division of Non- communicable Diseases, World Health Organ- ization, Geneva, Switzerland A. B. Mobarak (Temporary Adviser), Ministry of Health, Cairo, Egypt P. Pasquini, WHO Regional Office for Europe, Copenhagen, Denmark A. Pio (Secretary), Tuberculosis and Respiratory Infections, World Health Organization, Geneva, Switzerland P. M. Shah, Maternal and Child Health, World Health Organization, Geneva, Switzerland D. Tarantola, Expanded Programme on Immuniz- ation, WHO Regional Office for the Western Pacific, Manila, Philippines D. A. J. Tyrrell (Temporary Adviser), Medical Re- search Council Common Cold Unit, Harvard Hos- pital, Salisbury, England 58