|Food and Nutrition Bulletin Volume 12, Number 3, 1990 (UNU, 1990, 82 pages)|
|Public health nutrition|
This paper summarizes information on the characteristic-flaws in primary health care services, and their geographic patterns of occurrence, with reference to their local, medical, and bureaucratic contexts. The data are derived from published medical ethnographic reports, evaluations of programmes, surveys of physicians in developing countries, and some anecdotal reports. Several key issues emerge from this combination of sources. There is a lack of high-quality services in developing countries; most centres lack medicines, caring and skilled personnel, and convenient locations and scheduling. Also, the professional medical personnel are rarely sufficiently trained to communicate effectively with patients of different economic and ethnic backgrounds or to manage a health team of different backgrounds. The most successful programmes are those that have effective local leadership, adequate training, some curative potential, and powerful political will. The combined public health, medical, political science, sociological, and anthropological studies suggest no "cookbook" solutions as to what exact set of factors will make any programme successful. They appear to illustrate the effectiveness of flexible regional or local policies as opposed to global ones.
Studies on the use of Western health care services in developing countries have tended to demonstrate that people's acceptance or rejection of so-called modern medicine is based on their perception of its effectiveness  and on how well it meshes with their traditional beliefs and practices concerning health and illness [2; 3]. According to such studies, people participate in primary health care programmes to the extent that they experience the services as offering reliable solutions to what they perceive as health problems, including nutritional ones. The decision whether to use the services at the beginning of an illness, later, or even at all is calculated by weighing the anticipated benefits (relative to other alternatives) against the expected expenditure of time and money, and the personal abuse that these efforts may entail. Understanding how people estimate factors on both sides of the equation relating costs to benefits is crucial to understanding the demand for primary health care services and, as a corollary, the strengths and failings of the programmes
The author is an associate professor of anthropology with the World Hunger Program at Brown University, Providence, Rhode Island, USA.
This paper was presented at the United Nations University Conference on Nutrition in Primary Health Care held in Bellagio, Italy, 1-6 July 1985.
Studies of health care and medical decision making in pluralistic medical situations have been concerned with how people combine medical systems when faced with a variety of health problems, health and illness concepts, medical personnel, and treatment options. A number of studies have outlined the series of steps that people in different cultures follow when evaluating the probable cause(s) of and best line(s) of treatment for particular cases of illness [4-6]. Cultural, social, and economic considerations include perceived type, cause, and severity of complaints; the degree to which a treatment addresses personal (psychological) and social (group) aspects of disease in terms that people can understand; and, most importantly, the cost of treatment in time and material resources.
Other factors that are also basic to recipients' evaluations of a particular therapy but that seem to have been less carefully or systematically considered are the characteristics of the health facilities, personnel, and treatment regime(s). Thus, long waits, cursory examinations, lack of or inappropriate medicines, uncomfortable settings, and inconsiderate or abusive personnel are all well-known reasons why primary health care centres are often under-used or less esteemed than other Western and non-Western medical options. Additional considerations include the logistic costs in time and travel to a health care centre, which may not be open when a client arrives, and lack of medical back-up. Typical of practitioners who are detrimental to primary health care delivery are those who seem to lack interest in patients" problems, who attend clinic irregularly and for shortened hours, who spend little time with each patient, although they may recommend consultation at some other hour for a fee, and who show neither gentleness in their touch nor concern for the possible side effects of their prescribed treatments. Evaluations of health care delivery and effectiveness have less often investigated the quality of medical services and the attitudes of care givers than the simple quantity of services, such as numbers and types of buildings, equipment, and personnel.
Since the Alma Ata declaration in 1978, setting the goal of health for all by the year 2000, a major effort has been made to increase access to health services and to improve community participation in their planning, organization, operation, and control. The aim has been to make health an integral part of more general development efforts . However, advances in the number of people served-judged by the number of people trained and number of facilities built- have come about largely without careful follow-up as to how well the people are being served, and how both they and the health care providers view the appropriateness, adequacy, and effectiveness of the services.
This paper summarizes information on the characteristic flaws in primary health care services, and their geographic patterns of occurrence, with reference to their local, medical, and bureaucratic contexts. The data are derived from published medical ethnographic reports, evaluations of programmes, and surveys of physicians in developing countries. Anecdotal reports from numerous respondents to a request for information on why people in developing countries under-use low-cost public health services are also included.
Several key issues emerge from this combination of sources. First, the factor that most inhibits use of primary care facilities in developing countries is the lack of high-quality services. Most centres lack medicines, caring and skilled personnel, and convenient locations and scheduling. People demand effective curative care first.
Second, beyond such issues of access, the professional medical personnel who deliver services are rarely sufficiently trained in the interpersonal communication skills that would allow them to communicate effectively with patients of different economic and ethnic backgrounds, or to manage a health team of different backgrounds. The relationships of medical doctors to their client populations and traditional healers in their midst seem to vary by region. The greatest antagonism between patients and doctors seems to appear in the primary health care literature from Latin America.
In addition, most studies of programmes that involve health care auxiliaries indicate that most routine tasks can be carried out by individuals with little formal education, if they are properly trained and supervised. Such an allocation of work, however, may be acceptable to neither the client population, who want to receive care from skilled doctors, nor the medical community, who may resent having diagnostic and therapeutic procedures carried out by people with minimum literacy and training.
The most successful programmes are those that have effective local leadership, decent training, some curative potential, and powerful political will. They are also those that combine health with other developmental efforts, so that health, nutrition, food provision, and economics are interrelated and supported by common efforts [8; 9]. In this context, however, it is probably unwise to generalize from one region of the world to the next. A programme that works with volunteer women community health workers in Africa, which has a strong tradition of respecting women as workers and community leaders, probably would be less effective in another area where no such respect exists and where people are sufficiently enmeshed in the cash economy to expect monetary compensation for their time.
In short, the combined public health, medical, political science, sociological, and anthropological studies suggest no "cookbook" solutions as to what exact set of factors will make any programme successful. Moreover, they appear to illustrate the effectiveness of flexible regional or local policies as opposed to global ones.
Social and cultural issues
For many years, anthropologists have commented on the problems in the organization of primary health care programmes from the points of view of folk cultures and medical and bureaucratic cultures. For example: "Medical and public health programmes in developing countries will be successful in design and operation if they take into consideration the social, cultural, and psychological characteristics of the target group" . Another typical comment: "The most successful medical and public health programmes in developing countries require knowledge about the social, cultural, and psychological factors inherent in the innovating organisations and their professional personnel" .
In attempts to understand why people in developing countries under-use Western health services, uncongenial atmosphere in clinics and unfriendly service by clinicians have been documented in the ethnographic literature since the 1950s and 1960s [2; 10]. In addition to the better-known issue that patients and doctors, operating from folk and Western medical perspectives respectively, might not view illness and its treatment alike , anthropologists have noted that the treatment clients receive is often abusive. A dramatic example is doctors' refusing to treat sick children on days set aside for preventive care if their mothers have not registered them on well-baby days. Such an assignment of priorities may appear incomprehensible and outrageous to mothers, especially those who have travelled some distance to a clinic [1; 12; 13]. Another example is doctors' indicating their cultural distance (superiority) by ridiculing patients' health beliefs and practices.
Social distance between foreign physicians and patients (which is often the case in Asia and Africa) or between doctors and their compatriots of lower socioeconomic class (as in Latin America) may cause communication problems on either side. Studies have indicated that some doctors do not even speak the language of those they treat  and that they use insulting and rough language toward patients rather than polite terms of address [13; 15; 16]. Correspondingly, nurses and other clinic workers rudely assemble patients into queues, then rush them through their short procedures. Another problem is that of physicians not understanding or caring to comprehend patients' descriptions of their complaints, which is not restricted to one region of the world. Finally, illness always affects social relations beyond the sick individual, and logistically involves relatives or others to accompany and care for the person. Particularly where nursing care is in short supply, however, rules usually control visiting  or inconveniently restrict who may attend a patient.
Patients often perceive those who work in the clinic as not only unfriendly and insensitive but also incompetent. In Latin America, national rules dictate that first-year graduates of medical schools serve in public clinics, but patients often do not trust these neophyte physicians [6; 18; 19; personal communication, M. Zeitlin and K. Johnson, 1985]. People prefer to consult the more experienced and trustworthy, although more expensive, private physicians. These practitioners have established good reputations, perform reliable therapy, and often have a good understanding of local illnesses or at least of the terms in which people express complaints. In addition, primary care clinics seem to have irregular hours, and patients cannot count on doctors being present with any regularity or having the proper medical supplies when they are there.
In India, it has been noted that doctors in public service sometimes provide poor service to make sure that few people will come to the clinic, thus giving them more time and drugs for their paying clients. On the other hand, they may provide inadequate service so that patients will have to return again and again . Surveys indicate that throughout India primary health care is insufficient and inadequate: doctors lack drugs as well as motivation and training to serve the poor . In addition, medical and housing facilities for doctors are unsatisfactory . In Nepal, physicians may depart from their rural posts for months at a time without being replaced. Problems of lack of facilities and low pay, which shatter the will of the more idealistic young physicians to practice medicine, have been noted in Africa as well, where doctors may even turn to other occupations in order to support a middle-class life-style .
Quality of care
Overall, both medical personnel and clients recognize that inadequate facilities, lack of supplies, and insufficient medical and logistic back-up make effective care difficult or impossible. From the perspective of the patients, the numbers of clinic hours and personnel may be insufficient for them to be cared for without long waits or even at all. From the perspective of health providers, it makes little sense to keep the clinic open if it is not supplied. Once the patient is inside the consultation office, the examination may be brief: in the Dominican Republic examinations were timed as lasting 2 minutes and 50 seconds on the average 19], in Mexico 2 minutes , and in India less than 1 minute , or 2.2 minutes for doctors, 1.8 for pharmacists, and 1.6 for nurses . Such short consultations rarely involve physical examinations or even measuring vital signs .
In Saudi Arabia, baseline data on three health centres showed that the mean length of time physicians spent with each patient was 3.1 minutes; in another case it was 2.3 minutes-21 seconds for history taking, 64 seconds for examination, and 53 seconds for writing a prescription. In the latter instance it was also calculated that, on the basis of the number of patients seen, physicians and staff spent only 1 1/2 to 1, hours every day (15%-18% of their time) with patients. The time spent explaining a patient's illness and medications seemed to be minimal. Fewer than one-third of the patients interviewed after consultations understood their diagnosis or how to take their medicines . In the Sudan, for two clinics and one hospital outpatient clinic, the average consultation time was 2 minutes-0.6 minute for history taking and 1.4 minutes for drug prescription and physical examination .
In addition to a long wait and a brief consultation, drugs may be a problem, as both clinics and mobile units have been documented to be undersupplied in all parts of the world, including Mexico, Central America, Haiti, Peru, India, Bangladesh, Burma, Kenya, Ethiopia, and Tanzania. Aggravating the endemic shortage of supplies may be corrupt use of them. Mexican ethnographic studies report suspected plundering of health and nutritional supplies , and in India drugs were reported to find their way into the private practices of physicians. Contributing to shortfalls in supply is the attitude of those served by public clinics that they "deserve" medicines, which leads to wastage, as people go to the clinic to "get their money's worth" or fake symptoms so that they can obtain strong drugs [19; 28].
Use of pharmacists, local curers, or private physicians in the countries mentioned is based on speedier treatment and better supplies, as well as the usually greater sensitivity of these care providers than patients can expect from primary health care professionals. If a trip to a primary health care facility includes the time and expense of a trip elsewhere to purchase medicine, people will choose to go elsewhere in the first place. In addition people may value the medicines provided by the alternative care givers more highly. Pharmacists usually carry brand-name medicines (with which patients are familiar through the media or prior illness) in small quantities that patients can afford and that better fit their cultural preferences, expectations, and pocketbooks . Pharmacists are usually nearby, are open every day and for long hours, and explain illness and treatments in terms that clients can understand .
Even where drugs are available, however, they may be too expensive. Furthermore, clients may not purchase or take recommended dosages, they may not get better, and thus they may conclude that the health services are worthless [14; 31].
Beyond, but also related to, appropriate drug availability and physician attitudes are issues of the quality of the diagnostic and medicinal treatments. The general conclusions in reports from the Dominican Republic , India [20; 24; 32], Bangladesh , Ghana , and other countries is that the quality of primary health care is poor and that is why people do not use the services. Studies in Tanzania, Saudi Arabia, Ghana, and Bangladesh showed that, when drugs were available, physicians tended to overprescribe or use dangerous agents indiscriminately [25; 33; 35]. Part of the problem may be that patients use medicines incorrectly because drugs often are supplied without sufficient explanation of how often and how long to take them.
In combination, insufficient or improper consultations by clinic or health staff and unavailable, in appropriate, confusing, or overly expensive medicines lead potential clients to evaluate primary health care as useless and to seek medical help elsewhere. This is particularly true when people calculate the time and expense of the visits against the prospective benefits.
Logistics: Calculation of direct and indirect costs
Many studies show that it is not cultural factors but inadequate access that accounts for the under-use of Western-style health services in general and in some instances those that are supposedly free as well. Clinics are not always conveniently located, so that time and expense, including consultation and medication fees for the supposedly free services, may be considerable [6; 18; 36]. Catchment areas seem to vary by practitioner . People in the closest proximity seem to seek health care more frequently, particularly when the services are free and involve no extra cost [28; 37]. Beyond that, people will reckon the likelihood of effective care against the time and cost of seeking it, taking into consideration the probability of being treated by a skilled practitioner (usually a physician) and receiving appropriate therapy.
Problems in the organization of programmes
Medical vs. non-medical personnel and culture
Given the shortage of physicians and nurses in most developing countries, many health programmes have been designed to deliver basic services to the rural and urban poor by means of medical auxiliaries. Such programmes are intended to transfer the management of health from the hands of the medical profession to those of the community. However, political problems, poor medical quality, and logistic failings plague them as well.
As certain classic health projects have shown, projects that are especially sensitive to the socio-cultural context and health and illness beliefs of the recipient population can still fail if they neglect the sensibilities of the medical culture. Definitions of tasks, relationships of authority, and respect among doctors, nurses, and medical auxiliaries were inadequately considered and thus doomed an otherwise culture-sensitive project . The social stratification and the culture of medical society are significant factors contributing to or detracting from programme effectiveness.
Also important are the cultural beliefs of the recipients when health programmes introduce new categories and mixes of health workers without consideration of the clients' own perceptions of needs and their own health practitioners . Several problems arise in programmes staffed by auxiliaries. The role of physicians must be defined vis-a-vis the rest of the health care staff. The literature shows that it is the presence of effective, experienced doctors that draws people to health services: patients are unlikely to take the time to seek services if they feel they are being inadequately cared for [40; 41]. In the view of the health care providers, however, trained medical personnel should be used exclusively to direct the programmes and health teams and to provide curative care. Therefore, as providers define the services and tasks of other personnel, health auxiliaries often feel they are being treated like servants, which results in poor staff morale . Where no team is in place, or where it consists of mobile units, medical backup may be inadequate, resulting in poor care. Such problems are not easily overcome without major changes in structuring.
The relationship of the health care team to the clients must be considered as well. Are health workers to be local villagers, or non-local, often more highly educated personnel? From the recipients' point of view, local health workers are able to speak their language and understand the clients' health concerns, but they may lack credibility since they know little more than the average villager. More highly educated outsiders may command respect; on the other hand, they may be too socially distant and insensitive to culturally expected or appropriate behaviour to encourage the population's participation in health programmes. From the medical providers' point of view, the trade-offs are educational (can villagers, if under-educated, handle the tasks of record keeping?), economic (local villagers may work for less), and programmatic (there may be a desire to train and employ young female workers).
Although some preliminary attention to existing ethnographic literature on the area to be served and additional ethno-medical surveys could alleviate potential problems of understanding the cultural contexts of health and illness, such initiative is apparently rarely taken. Even where information on local cultural conditions hampering health delivery exists, it is often ignored for programmatic reasons [40; 42]. Governments tend to accept the dictates of foreign donor organizations in order to add employment, whether or not it fits their local conditions .
Indigenous health practitioners
Yet another issue is that of how indigenous health practitioners might be incorporated into primary health care efforts. Although directives have tried to improve the effectiveness of reaching the people through consultation with and incorporation of indigenous practitioners, selecting these individuals to work with may be difficult . In addition, "some programmes which aim to incorporate the traditional healer have undermined the healer's status by relegating the healer to the bottom rung of the Western health service and converting him or her from an indigenous expert to a marginal health aide" [39; 42].
An oral rehydration therapy (ORT) programme for children with diarrhoea in Brazil is one example of the successful joining of local beliefs and practices with the provision of modern treatment by traditional healers . The healers skillfully incorporated ORT into their routines, provided the patients with good psychological as well as medical care, and assured follow-up. as the ritual procedures took up to nine days. In this case, national health authorities were willing to include these practitioners in the ORT programme after they had satisfied themselves that the healers did not compromise the level of care. The programme, it should be noted, followed a detailed socio-cultural survey of local beliefs and practices relating to diarrhoea and probably carefully built up respect for and rapport with local healers.
Selecting and training various levels of health workers
We have looked at educational, economic, and cultural issues with respect to local versus non-local personnel. The attitudes of recipients toward the health programme and workers are affected by their perceptions of the workers' knowledge and authority, education level, and social distance. From both recipients' and providers' points of view, a major problem seems to be who recruits the workers. If the key to success is not the technology but the organization and politics, local political cadres should have a say about who works in the programmes. Studies have shown that the effective programmes are those that draw on local political (often factional) support, incorporate highly motivated individuals from the faction, and use existing patron-client relationships to build an effective structure. Those that try to be democratic by constructing community participation in the abstract often fail. In introducing health programmes, organizers might better spend their time studying community structure than trying to organize the residents . Alternatively, one Kenyan example showed a successful programme originating from the inside when a health committee was able to overcome factional differences .
More generally, public health analysts dispute whether the role of community health workers should be mainly medical or political. Those who see the role as a springboard for improving community nutrition, sanitation, hygiene, and disease prevention and for integrating health with other aspects of community development advocate placing more stress on the organizational than the technological skills of the auxiliary . The danger is that the political organizing will go on at the expense of health activities .
Local cultures have definite expectations about the sex, age, and authority of health personnel. Projects that do not pay attention to such information often fail  or are under-utilized . Certain projects have amended their procedures to incorporate women after women failed to use the services when they were staffed by men.
The literature is mixed on how and at what levels primary health care workers should be funded. Some lament the passing of the voluntary village health associations, usually composed of women, in favour of health and sanitation professionals. Once professionals appear, the women's committees are displaced, as is community control. The newcomers are paid to keep more efficient records and handle communications in a lingua franca such as English. Yet they cannot communicate with the people and require intermediate-level native workers to translate for them and facilitate their cultural and linguistic entry into the community .
Many other programmes flounder, studies show, because the workers are not paid sufficiently and therefore are not highly motivated. In certain instances payment in kind, according to traditional custom, has worked out satisfactorily . Regardless of the method of payment, the success of such programmes usually is attributed to the individuals who staff them and their relationship to the community.
How much local people should be paid and from whose coffers is a problem in many instances. Where non-local workers are too expensive, it may be argued that they should be replaced by local people. Still others see paying the local primary health worker at a level the village finds disproportionately high as a source of tension.
Significant difficulties exist in training middle-level workers and trainers. This can result in lack of accountability for care, which leads to poor medical care practices.
Key problems that affect morale are lack of facilities, lack of transport, lack of medical supplies (in some cases, basics like soap), and poor relations between auxiliary personnel and professional staff. Recipients who are aware of these difficulties tend to under-use the services. Community health workers also perceive the problems, which leads to low morale and high turnover rates.
Problems in developing global and country- or region-specific programmes
Although it has been pointed out that health education and training programmes should not disparage local practices and beliefs, conditions, and knowledge and where possible should look at the positive benefits of local traditions for health care and explanations , such advice is not always followed. Additional problems are that existing knowledge may not be used [42; 50] and that information and advice given out by health care workers may be inconsistent or incomprehensible. A general problem is how to control, simplify, and unify these messages, while adjusting their content to local concepts for comprehensibility.
In one primary health care setting in Ghana, information on the length of time to breast-feed, the time at which supplementary solid food should be introduced and how often, and what types of foods to feed was incorrect . In addition, because those who give the advice are young and inexperienced, they may not command the respect of the women they are advising. In some instances, it might be more sensible to enlist older women, who influence the behaviour of younger women within their extended families, to disseminate such information.
Providing correct information seems to be a major problem in relation to growth charts as well. Studies indicate that the data are not always interpreted correctly to the mothers; in fact, even the health worker may not understand their meaning or know how to use them effectively to offer recommendations . In some cases neither lower-level personnel nor supervisors are trained adequately to measure or interpret the charts , which are often quite complicated 53].
Recipients may find it offensive when clinic staff insist on retaining the growth charts and refuse to discuss the findings with them. Instead of being a passport to health care for children, the chart may be seen as a barrier to receiving curative care, since the care may not be made available to those who do not participate in the growth-chart programme. Health providers, on the other hand, often find it contrary to their perceived roles to let the clients maintain control of their own charts. An additional political constraint may be national personnel who do not want to use a chart that they have not personally developed.
Careful ethno-medical study can indicate (predict) possible impediments to implementation and acceptance of ORT measures, although in some instances attention to the culture of the medical community can also anticipate that such information may not be used constructively. In Honduras the cause of diarrhoea was attributed to the local folk illness empacho, but incorporating a special message to encourage use of ORT for the patients went against the culture of the physicians, so the disease went untreated . By contrast, the project we have mentioned in Brazil that incorporated local healers into the programme was largely successful in executing a broad-based campaign for the use of ORT . Problems are that self-sustaining programmes must gain acceptance at both the community and national levels. Thus, efforts in Honduras to promote ORT for diarrhoeal disease may fall short as the government shifts priorities to other communicable diseases.
The success of immunization programmes depends on several factors. Information communicated in the lingua franca may be incomprehensible to mothers who speak mainly a native dialect . In one instance, a programme to combat measles failed because the vaccine was ineffective: 60% had been stored incorrectly. Overworking the community health workers who deliver immunizations can also disrupt effective coverage, as studies in northern India showed: three series of immunizations were piggybacked onto everything else the workers had to do.
An additional problem concerns just how specialized or comprehensive the tasks of the community health workers should be. Nichter , carrying out "community diagnosis research," noted that instead of providing general health and nutrition care, many clinics tried to deliver targeted programmes, which made for questionable relationships among staff and between staff and health care recipients. He reported a situation in India where villagers are aware of target pressures and competitiveness among primary health centre staff, which staff revealed when they attempted to persuade villagers to accept targeted services. Targets and incentives for individual health workers, while designed to focus attention on government priorities, reward initiative, and provide a check on staff fieldwork, have resulted in splitting staff and contributing to poor primary health centre team interaction and community identity.... Primary health centre fieldworkers continue to be cast in the role of "change agents" who are sent to the field with a few insular services, a mandated educational role they have been ill trained to carry out, little backup or constructive supervision, and a sense of mistrust and jealousy of primary health centre peers with whom they compete for targets and incentives. 
Low pay, poor relations with other staff, and a sense of frustration due to lack of supplies, medical support, and interest by other levels in the health system to channel information to their own and other programmes about community conditions, perceptions, and needs generally led to low morale among health workers and disillusionment in their clients.
In general, primary health care programmes, whether public or private, experience problems in information flow between patients and health care workers, and between the various levels of care providers. Some of the problems also may be due to the flow of funds, as well as to difficulty in integrating health care with other development efforts. Most evaluations indicate that both baseline and subsequent data are insufficient. Yet several time studies have indicated that, even where the amount of personnel time devoted to record keeping is substantial. such information is rarely used. In Ethiopia, studies of three rural health centres showed that, where home-visiting nrses supervised registration of all births in communities and kept records, the information went largely unused in maternal and child health programmes. All senior health workers were involved in lecturing on health four to seven hours a week in governments schools, but no efforts were made to provide water or latrines to the schools at the same time .
Researchers in India concluded that, given the volume of records and reports doctors were required to keep, "the quality of information they contain must be suspect, and they are undoubtedly little used even after the laborious recording by hand" . In Tanzania clinics kept growth charts, but workers would not communicate to mothers what they meant (personal communication, A. Fleuret, 1980). In Ghana most of the consultation time was taken up with writing records (sometimes a child had four different records to be completed) rather than with giving advice . In Papua New Guinea follow-up was described as almost non-existent .
Matching expectations and perceptions of medical needs
Critical problems exist in matching or accommodating people's perceptions of their medical needs and their expectations for care with those of health care personnel and administrative planners and executives. The need felt by most people is for curative care that is easily available and effective. Yet additional work addresses how best to match preventive medicine, particularly sanitation, hygiene, and health and nutrition education, with curative therapy, which is what people seem most to value, at least in seeking health services. Recommendations range from fixing single aspects of the health system to instituting sweeping political changes.
One set of problems and solutions concerns the relationship between medical and auxiliary personnel. Jelliffe has suggested that to improve and make optimum use of trained staff, more routine functions such as weighing, taking a basic medical history, directing "traffic," clarifying medical advice to patients, and record keeping could all be done by staff who have minimum technical training . Yet it is unclear whether clients would accept this, or whether they would feel they were being shunted off to inferior and often abusive personnel. It can be argued that people might be better served if governments reduced their stress on high technology and highly trained physicians in favour of more preventive and promotional medicine and the involvement of traditional practitioners in the delivery of various types of care, but this leads to the question of whether people would be satisfied with such care or would still prefer to be cared for by a physician.
Another difficulty concerns political will, at the national and community levels, to provide effective services, particularly for the poor. In India, Banerji  argued that there was no reason why existing services should not be supplying the health needs of the people, but they were failing to do so because of poorly motivated doctors, insufficient staff and supplies, and badly run family-planning programmes. In addition, at all levels, no one was accountable for the effectiveness of any particular aspect of operations, so care was not effective. He offered guidelines for making existing services more efficient. Evaluations such as Pyle's, of what went wrong at all levels in a model community health project, however, cast doubt on the possibility of such recommendations being followed 
A third aspect that in the past was inadequately considered in planning the most effective approach to health care is seasonality:
The importance of seasonality seems to have been largely lost sight of in the planning of modern health services. Attention has concentrated on location rather than timing. Seasonality is recognised mainly in terms of constraints: when rains come and roads become impassable, villages served by mobile clinics are cut off, and mass immunisation programmes in rural areas may have to be suspended. The costs of sickness-in terms of losses of family food and income, of losses of body weight reserves, and of national agricultural production foregone-are both high and very seasonal. Sickness in an agricultural slack season entails suffering, but its social and economic costs may be far less than those of sickness in an agricultural season which directly prevents work to earn income or grow food. There are arguments, on both welfare and economic grounds, for special attention to health care during the agricultural seasons, and to those diseases and complaints which are most likely to incapacitate at that time. There is here a strong but little recognised complementarily between health services and agriculture. 
It has been recommended that clinics in Bangladesh should be seasonally staffed to ensure high coverage in times of greatest need , and that community health workers should be selected so that they will not be diverted to other tasks in seasons of both greatest agricultural labour and greatest illness .
Along these lines, additional understanding of the organization of programmes is essential, from medical as well as cultural points of view. Foster concluded that medical anthropologists should spend at least 50% of their time studying health bureaucracies, saying: "We know more about how communities address health needs than the dynamics of health bureaucracies, the interrelationships between institutional and personal factors that bear on health system planning, the selection and training of health care personnel, and the functioning of health teams in providing primary health care" . While one may question the particular figure, the need for better understanding of ethno-medicine on the part of bureaucracies and medical workers is clear. Only by such understanding can we hope to improve communications, programme implementation, and health at the community level.
Correspondingly, health administrators have emphasized the need for epidemiologists and policy planners to pay more attention to socio-cultural determinants of health in order to make their efforts more cost-effective and health-productive. They have also urged that social scientists should pay more attention to the biological determinants of health:
An enormously ambitious and costly medical intervention research project was undertaken with little attention to and virtually no research on the social determinants of child survival by the biomedical scientists involved. At the same time, in other localities, social scientists were studying the social determinants of infant mortality, but with inadequate attention to the biological intermediate variables that actually operate to influence disease and death rates. Thus it is difficult to derive any specific strategic recommendation for primary health care programs from the results of these projects.
The obvious need is for a multidisciplinary approach into problems of infant and child mortality if a sound base of knowledge is to be developed for health policy and health programs. 
Most of the intensive case studies designed to show why people under-use primary health care indicate a variety of constraints or abuses. Rather than just cost or distance, decisions involve logistic, cognitive, and humanitarian factors. If care is convenient, people seem to use it as a first resort, even if it is perceived as less than satisfactory. If it is inconvenient, they pay close attention to its quality before deciding to invest time and resources in it.
Many of the problems regarding quality of services perceived by recipients are echoed in the complaints of medical personnel. The following are logistic and organizational problems perceived from both sides:
With respect to the relation between the cost of health care and its value, the prevailing attitude among clients is that "you get what you pay for." As they see it, if the services are free, they must be of poor quality. From the standpoint of health care personnel, the free services they are supposed to deliver are usually of lower quality and concern than the fee-for-service consultations with which they supplement their incomes. Long waits followed by cursory examinations are a disadvantage in public services, as are also problems of logistics and quality.
Other aspects of health care may be interpreted differently by providers and clients. Excessive medication often occurs when drugs are available. Patients may complain if they are given no injections or fancy medicines. On the other hand, they may also complain if they receive too many drugs, particularly if they fear overmedication and if they must pay for the drugs. Physicians are frustrated by inadequate supplies, as well as by patient non-compliance with prescribed treatment. Thus they may try to press any available drugs onto patients at the first visit if supplies are abundant and it is doubtful that the patients will keep follow-up appointments.
Preventive services are offered infrequently; for example, clinics often do not give advice on dietary measures. Medical personnel say they do not make such recommendations because they feel that the information is impractical since most patients are too poor to improve their diets. People can know the facts about how to prevent disease, but they do not have the resources to implement them: "Poverty and prevention just do not go together" .
This is the recipients' perception as well. They know that they should eat well and take tonics, or in some cases rest from heavy work schedules to build up their strength. However, they do not have the economic resources to allow these activities. There do not seem to be any simple solutions to these problems.
Coordination of local communities with modern health care delivery
Anthropologists are fond of criticizing physicians and community developers for not learning enough about the communities into which they are introducing programmes. Among other things, health personnel often do not know the local language; they feel superior to the local people; and they do not adequately appreciate the cultural concepts of illness, disease, and treatment. Realistically, health personnel should not be expected to learn local languages. What would be possible within a reasonable time, however, would be to master, with the help of the local population, the prevailing concepts of illness and health, plus polite forms of greeting and vocabulary for dispensing medications. In this way, they would seem less distant and foreign and more interested in their patients, and thus would offend less. (Having to learn local customs and a bit of the language might also have a humbling effect on the superior attitudes medical personnel of higher socio-economic status frequently feel toward their clients, as many anthropologists could tell them.) How to motivate health care workers to enjoy such an exercise in cultural anthropology and to appreciate that it raises their qualifications to practice in local communities and at the same time increases their understanding of health, disease, and people so as to make them better physicians and nurses is, of course, another matter.
The medical concept of preventive primary care must be introduced into the health practices of local people. In particular, use of growth charts must to be integrated more tightly with diet and health observations of the mothers. A UNICEF conference on child survival (Boston, 29-31 May 1985), for example, raised the point that mothers often notice that their children are "ill," that is, that they have stopped eating, before deficits begin to appear in rates of growth. It would be useful if health workers who encourage the use of growth charts to monitor changes in nutrition and health could learn to solicit and react to such signs.
Finally, it must be questioned whether public health officials and others use the best criteria for evaluating the effectiveness of care. Chrisman and Kleinman , following Young , noted three elements that are involved in the healing process:
Habicht and Berman  also stress that more attention should be paid to issues of quality when evaluating the success of the care: not only body counts (i.e., how many lives have been saved) but how the quality of life has been improved through the humanitarian dimension medicine is supposed to have.
Current research continues to document what has gone wrong with the well-intentioned goal of providing health for all by the year 2000 by primary health care, from perspectives of medical, bureaucratic, and community cultures. Although the issues involve operational and implementational analyses, at basis they are cultural and demand the kinds of cultural analysis anthropologists have been trained to do. Problems in the organization and delivery of health care services are intrinsic to both the medical health provider and the folk medical culture. Only by under standing the dimensions how health is determined and dealt with by the home environment from both points of view can we hope to improve health and the impact of primary health care services. As Mosley , among others, concludes, we must proceed with biologically and socio-culturally based studies, or we will end up simply with changes in the causes of mortality rather than absolute reductions in mortality.
1. Erasmus C. Man takes control: cultural development and American aid. Minneapolis, Minn, USA: University of Minnesota Press, 1961.
2. Paul B, ed. Health, culture, and community. New York: Russell Sage Foundation, 1955.
3. Foster G. Medical anthropology and international health planning. Med Anthrop Newsletter 1976;7(3): 12-18.
4. Janzen JM. The quest for therapy in lower Zaire. Berkeley; Calif, USA: University of California Press, 1978.
5. Messer E. Present and future prospects of herbal medicine in a Mexican community. In: Ford Rl, ed. University of Michigan Museum of Anthropology, Anthropological Papers, no. 67. Ann Arbor, Mich, USA: University of Michigan, 1978.
6. Young, JC. Medical choice in a Mexican village. New Brunswick, NJ, USA: Rutgers University Press, 1981.
7. World Health Organization. The Alma Ata conference on primary health care. WHO Chronicle 1978;32:42829.
8. Bossert I, Parker D. The political and administrative context of primary health care in the third world. Soc Sci Med 1984;8:695-702.
9. Pyle D. From project to program: the study of the scaling up/implementation process of a community-level, integrated health, nutrition. population intervention in Maharashtra, India. PhD dissertation, Massachusetts Institute of Technology, Cambridge, Mass, USA: 1981.
10. Read M. Culture, health, and disease. Social and cultural influences on health programs in developing countries. London: Tavistock, 1966.
11. Kleinman A. International health care planning from an ethnomedical perspective: critique and recommendations for change. Med Anthrop 1978;2(2):71-96.
12. Foster G. Relationships between theoretical and applied anthropology: a public health program analysis. Human Org 1952;11:5-16.
13. Frankenberg R. Medical anthropology and development: a theoretical perspective. Soc Sci Med 1981;14B:197-207.
14. Brown M. Una paz incierta. Lima: Centro Amazónico de Antropología y Aplicación Práctica, 1984.
15. Wolff RJ. Modern medicine and traditional culture: confrontation on the Malay peninsula. Human Org 1965;24:339-45.
16. Laderman C. Wives and midwives: childbirth and nutrition in rural Malaysia. Berkeley, Calif, USA: University of California Press, 1983.
17. Skinner E. African urban life: the transformation of Ouagadougou. Princeton, NJ, USA: Princeton University Press, 1974.
18. DeWalt K. The illnesses no longer understand: changing concepts of health and curing in a rural Mexican community. Med Anthrop Newsletter 1977;8(2):5-11.
19. Ugalde A. Where there is a doctor: strategies to increase productivity at lower costs-the economics of rural health care in the Dominican Republic. Soc Sci Med 1984;19(4):441-50.
20. Banerji D. Health behavior of rural populations: impact of rural health services. Econ Pol Weekly 1973;8:226168.
21. Taylor CE, Alter JD, Grover PL, Sangal SP, Andrews S, Takulia HS. Doctors for the villages: study of rural internships with seven Indian medical colleges. Bombay: Asia Publishing House, 1976.
22. Mejido M. Mexico amarga. Mexico City: Siglo XXI, 1973.
23. Taylor CE. The place of indigenous medical practitioners in the modernization of health services. In: Lelie C, ed. Asian medical systems. Berkeley, Calif, USA: University of California Press, 1976:285-99.
24. Johns Hopkins Department of International Health. The functional analysis of health needs and services. Bombay: Asia Publishing House, 1976.
25. Sebaie Z, Miller D, Ba'aqueel H. A study of three health centers in rural Saudi Arabia. Saudi Med J 1980; 1: 197-202.
26. Saeed A. Utilization of primary health services in Port Sudan, Sudan. Trop Geogr Med 1984;36(3):267-72.
27. Romanucci-Ross L. Conflict, violence and morality in a Mexican village. Palo Alto, Calif, USA: National Press Books, 1973.
28. Nichter M. Patterns of resort in the use of therapy systems and their significance for health planning in South Asia. Med Anthrop 1978;2(2):29-58.
29. Ferguson A. Commercial pharmaceutical medicine and medicalization: a case study from El Salvador. Culture Med Psych 1981;5:105-34.
30. Logan K. The role of pharmacists and over the counter medications in the health care system of a Mexican city. Med Anthrop 1983;7(2):68-87.
31. Blustain H. Levels of medicine in a central Nepali village. Contrib Nepalese Study 1976;3:82-105.
32. Djurfeldt G, Lindberg S. Pills against poverty: a study of the introduction of western medicine in a Tamil village. London: Kurzon, 1975.
33. Chowdhury S, Ashraf A, Hai Khan A. Health, disease, care, and cure in rural Bangladesh: a study of three villages in Tangail district. Dacca: June, 1981.
34. Cole-King S, Gorden G, Lovel H. Evaluations of primary health care: a case of Ghana's rural health care system. J Trop Med Hyg 1979;82:214-28.
35. Segall M, White A. Research on primary health care: a multi-disciplinary project in Ghana. World Health Forum 1981;2(3):341-46.
36. Kroger A. Anthropological and sociomedical health care research in developing countries. Soc Sci Med 1983;17:147-61.
37. Walker G, Gish O. Inequality in distribution and differential utilization of health services: a Botswana case study. J Trop Hyg 1977;80:238-43.
38. Adair J, Deuschle KW. The people's health. New York: Appleton-Century-Crofts, 1970.
39. Bloom A, Reid J. Anthropology and primary health care in developing countries. Soc Sci Med 1984; 19(3): 183-84.
40. Cosminsky S. Utilization of a health clinic in a Guatemalan community. Rome: Tilgher-Genova, 1972.
41. Heggenhougen HK. The utilization of traditional medicine: a Malaysian example. Soc Sci Med 1980;14B:3944.
42. Justice J. Can sociocultural information improve health planning? a case study of Nepal's assistant nurse midwife. Soc Sci Med 1984;19(3):193-98.
43. Nations M. Proceedings of the International Conference on Oral Rehydration Therapy. Washington, DC: USAID, 1983.
44. Paul B, Demarest W. Citizen participation overplanned: the case of a health project in the Guatemalan community of San Pedro La Laguna. Soc Sci Med 1984; 19(3): 185-92.
45. Willms D. Community-based health care programs in Kenya: three case studies in planning and implementation. In: Morgan JH, ed. Third world medicine and social change. Lanham, Md, USA: University Press of America, 1983:299-312.
46. Rifkin S. Politics of barefoot medicine. Lancet 1978; 1:34.
47. Marchione T. Evaluating primary health care and nutrition programs in the context of national development. Soc Sci Med 1984;19(3):225-35.
48. Schoeffel P. Dilemmas of modernization in primary health care in western Samoa. Soc Sci Med 1984; 19(3):209-16.
49. Werner D, Bower B. Helping health workers learn: a book of methods, aids, and ideas for instructors at the village level. Palo Alto, Calif, USA: Hesperian Foundation, 1982.
50. Kendall C, Foote D, Martorell R. Ethnomedicine and oral rehydration therapy: a case study of ethnomedical investigation and program planning. Soc Sci Med 1984; 19(3):253-60.
51. Reid J. The role of maternal and child health clinics in education and prevention: a case study from Papua New Guinea. Soc Sci Med 1984;19(3):291-303.
52. Wilcox J, Teller C. Towards effective nutrition action in primary care: findings from sub-Saharan Africa. Presented at the 110th annual meeting of the American Public Health Association, Montreal, Canada. 14-18 Nov 1982.
53. Tremlett G, Lovel H, Morley D. Guidelines for the design of national weight for age growth charts. Assignment Children 1983;61/62:143-75.
54. Brown J. Low immunization coverage in Yaounde, Cameroon: finding the problems. Med Anthrop 1983; 7(2) :9- 18.
55. Nichter M. Project community diagnosis: participatory research as a first step toward community involvement in primary health care. Soc Sci Med 1984;19(3):237-52.
56. Spruyt D, Elder FB, Messing SD et al. Ethiopia's health center programs: its impact on community health. Ethiopian Med J 1967;5:4-87.
57. Jelliffe DB. Organization of MCH services in developing regions: VII. Young child clinic-basic problems. J Trop Pediatr 1967;13(4):182-84.
58. Chambers R, Maxwell S. Practical implications. In: Chambers R, Longhurst R, Pacey A, eds. Seasonal dimensions to rural poverty. Totowa, NJ, USA: Allanheld, Osmun, 1981:226-38.
59. Chen L, Chowdhury AKM, Huffman S. Seasonal dimensions of energy protein malnutrition in rural Bangladesh: the role of agriculture, dietary practices, and infection. Ecol Food Nutr 1979;8: 175-87.
60. Foster G. Anthropological research perspectives on health problems in developing countries. Soc Sci Med 1984;18(10):847-54.
61. Mosely H. Will primary health care reduce infant and child mortality: a critique of some current strategies with special reference to Africa and Asia. Rome: International Conference on Population, Expert Group on Mortality and Health Policy, 1984.
62. Chrisman NJ, Kleinman A. Popular health care, social networks, and cultural meanings: the orientation of medical anthropology. In: Mechanic D, ed. Handbook of health, health care, and health professions. New York: Free Press, 1983:569-90.
63. Young A. Some implications of medical beliefs and practices for social anthropology. Am Anthrop 1976; 78:5-24.
64. Habicht J-P, Berman PA. Planning primary health services from a body count? Soc Sci Med 1980;14C: 129-36.