|Food and Nutrition Bulletin Volume 10, Number 4, 1988 (UNU, 1988, 74 p.)|
|Public health nutrition|
Growth, a positive change in the size of a growing individual, is a dynamic measure of health, the best available indicator of nutrition status, and the only real measure of nutrition adequacy . Deviations from the expected, or predictable, course of growing are not visible at the earliest stage, and such invisibility is a major barrier to preventing and curing health problems. Growth monitoring has gained popularity in the last two to three decades and has been practiced in over 80 countries , with perhaps the earliest report from clinic-based activities in Nigeria .
The most widely promoted method of growth monitoring is weighing and charting growth, since weight gain is believed to be the most sensitive indicator of growth and is universally applicable . This method is favoured by UNICEF. Among other techniques, measuring arm circumference is claimed to be the easiest and cheapest alternative to weighing  and has been recommended for use at the home and village levels whenever regular and frequent weighing is not possible .
While most published papers have concentrated mainly on technical details, more important issues such as the objectives, feasibility, and usefulness of growth monitoring and its relevance and effectiveness in promoting child nutrition and health have not been dealt with adequately. Only a few evaluations to examine its functional utility and effectiveness have been made. At the twelfth session of the Subcommittee on Nutrition (SCN), in April 1986 in Tokyo, the Advisory Group on Nutrition (AGN) of the SCN was requested to make a statement with regard to growth-monitoring techniques and uses. In its statement of February 1987, the AGN recommended that the SCN investigate the literature to identify background papers on the usefulness of growth monitoring, its utility, and conditions of feasibility in relation to different purposes for which it is used. On the basis of such documents more specific recommendations could then be given to the SCN. The present brief review was prepared as one response to this recommendation .
An international workshop was held in Indonesia in 1984 to promote the exchange of experiences in implementing growth monitoring as a primary health care activity . Gopalan and Chatterjee  have critically reviewed global experiences in the use of growth charts, examining the operational problems by referring to published and unpublished reports, consultations, and correspondence. More recently Gopalan again questioned the feasibility of growth monitoring in developing countries  and discussed some basic related issues . The state of the art of routine growth-monitoring activities was examined by Griffiths , who also discussed the unification of growth monitoring and nutrition education [10, 11]. The application of operations-research approaches, methods, and techniques to address the main obstacles in the implementation of ongoing growth-monitoring activities was described by Teller . After an informal consultation at UNICEF headquarters in New York in March 1985, a meeting on growth monitoring was held at the UNICEF regional office for South Central Asia in New Delhi in May 1986.
Yee and Zerfas  reviewed selected projects to search for evidence showing the uses and effectiveness of growth monitoring in certain existing nutrition projects. Whether or not growth-monitoring programmes can reduce diarrhoea morbidity, mortality, or severity was the subject of a review by Ashworth and Feachem . Bhan and Ghosh  and Ghosh , reviewing four UNICEF-sponsored case studies, summarized features of successful growth-monitoring programmes to identify factors contributing to their effectiveness. Hendrata and Rhode  looked for commonly encountered pitfalls. Recently an information packet has been prepared, with articles and summaries on growth monitoring published within the previous five years .
Definitions and objectives
There is, unfortunately, no general agreement in the literature on what growth monitoring actually means. The term appears to mean different things to different people , and varied definitions can be found  . While some imply that growth monitoring is watching over and evaluating a child's growth pattern [1, 10, 19], others emphasize actions to be taken after such monitoring [4, 20-22]. Consequently, the term is considered by some to be inadequate and even misleading, and the terms ,'growth promotion,, and ,'growth monitoring and promotion,, have been suggested [12, 22]. Obviously, clarifying understanding is the key issue, not terminology.
There is considerable confusion about the basic objectives of growth monitoring , and lack of understanding of what growth monitoring is meant to do in specific situations can lead to faulty implementation . Weighing is not growth monitoring and is of little if any value by itself [1, 23]. Measurement without action is pointless and a waste of time and effort, and growth monitoring is not an intervention per se . Where the concepts underlying growth monitoring are not understood and there is no feasible operational strategy, it is not surprising that in many projects all that is left in practice is a superficial ritual of weighing and charting, or growth monitoring is used only as a strategy to help implement other interventions, such as supplementary feedings, more efficiently. This approach has been strongly criticized on practical, economic, and even psychological grounds [6-8, 23]. An important question, however, is whether growth monitoring unconnected to feeding programmes can be sustained and reach poor families on a continuous basis in countries that do not have strong traditions of social discipline and community work .
Since normal growth slows down long before overt malnutrition is apparent, Morley  defined the objective of growth monitoring as preventing growth retardation through timely and early detection of faltering growth. Gopalan  asserted that promoting growth monitoring as an integral part of preventive and promotive health care can be justified only if the objective is to prevent growth retardation. Therefore, using growth monitoring as a screening procedure for a rehabilitation and relief programme, as is done in many current projects (e.g. the Tamil Nadu Integrated Nutrition Project [TNINP] in India), is moving away from the real objective. This is in line with the conclusions that the underlying purpose of growth monitoring is to prevent malnutrition, not to rehabilitate its victims , and that it should be used to detect problems before nutritional status is seriously jeopardized .
A different view, however, was taken by Mukarji: "If the focus is on a nutritional strategy that emphasizes curative or rehabilitative aspects, then there is a place for [growth monitoring]. Should there be a definite shift to preventive and promotive aspects in nutrition programmes, then the strategy would be to emphasize better antenatal and under 2 year child care with a strong health and nutritional education [component]" . In a nutrition programme in Thailand, growth monitoring was used to "eradicate all cases of 3rd degree PEM in under-fives and to decrease the prevalence of 2nd degree PEM" .
The emphasis on growth rather than nutritional status, however, is a key operational communication strategy [17, 29], and it has been recommended: "To provide health care for children, move away from the immediate objective of prevention of malnutrition and monitor adequate growth,, . The monitoring of weight gain oriented toward health promotion is not only more cost-effective than the screening of nutritional status oriented toward treatment and rehabilitation but also more acceptable to both mothers and health workers as it provides more opportunities to observe changes or improvements in nutritional status .
Available data do not give enough indication of the effectiveness of growth monitoring, and studies to demonstrate the usefulness and benefits of such programmes in a community are needed for advocacy purposes. Recent reviews have concluded that many growth-promotion activities are poorly conducted and have discouraging results [31, 32]. In spite of widespread enthusiasm and support, the general feeling is that growth-monitoring programmes have yielded few benefits, and they are often described as failures . However, the reports have shown how a greater focus on growth-promotion strategies and their implications for programme operations can turn these programmes into successes.
Many nutrition and health programme planners in the developing nations consider that growth monitoring is not living up to its potential for contributing to child survival and development . The procedure has been held to be too complex, costly, and time-consuming to be effective in large-scale programmes . The experience of the Rural Unit for Health and Social Affairs (RUHSA) project in India is that appropriate target groups of more vulnerable populations can be identified by other, more general socio-economic factors, baseline surveys, and different procedures; thus individual growth monitoring becomes less relevant and may be unnecessary in a comprehensive primary health care programme . Similarly, it has been suggested that community monitoring of childhood nutrition is more relevant and far more important than individual growth monitoring in the context of community-based nutrition programmes, and that cross-sectional surveys are of greater use than growth charts for this purpose .
Growth-monitoring programmes have been seen as providing a new focus on the nutritional problems of children in developing countries and as having the potential, if conducted on the right lines and with appropriate simple technology, to provide the means to upgrade nutrition from its position of relative unimportance in the health system . The primary issue of the UNICEF conference on growth monitoring in New Delhi was that it was not working because it had not been tried properly . It has been suggested, however, that there have been successes along with the mistakes and that even some of the less successful programmes have shown the potential of growth monitoring to support child survival . The success of growth monitoring using weighing and growth charting in some small-scale projects with dedicated leadership and supervision has been taken as evidence that its full potential and impact is yet to be realized in most places .
Growth monitoring has been advocated worldwide as one of the key elements of child-survival and primary health care strategies  and as an excellent tool for assessing the growth and development of a child in order to detect the earliest changes and bring about appropriate responses to ensure that growth continues uninterrupted .
Reviewing seven projects for evidence of the uses and effectiveness of growth monitoring, Yee and Zerfas  found they all claimed that the technique was useful, but little information was given on how the benefits were measured, and actual data to back up the claims were frequently missing. Generally, the prevalence of malnutrition was reduced in programmes that incorporated growth monitoring; however, the relative contribution of growth monitoring to this reduction cannot be determined easily.
Few studies have been specifically designed to isolate the effect of growth monitoring on health outcomes by comparing programme villages or individuals with true controls . Given the energy with which it is being promoted, there have still been far too few evaluations of the effects of programmes on health outcome. The most useful estimates for programme planning are those from quasi-experimental evaluations of similar programmes operating in comparable circumstances. Until a body of such results has been built up, there is little basis for a universal statement one way or the other on the effectiveness of growth monitoring. Potential comparison groups may be hard to find, however, as it is already so widely practiced .
Growth monitoring and nutritional status
The role of growth monitoring in bringing about a remarkable improvement in nutritional status in the RUHSA project was said to be minimal , although it also was stated that the project was unsuccessful in implementing regular growth monitoring using growth charts.
Gopalan and Chatterjee noted, "As yet and from available reports there is no evidence that weighing and growth charting operations being promoted at great cost in some countries have in fact resulted in improvements in health and nutritional status of the children being weighed" . This was found to be especially true for large-scale projects. One study found that, although growth monitoring of individual children was not successful because of the community attitude to weighing and the logistical problems in the health care delivery system, there were significant changes in health-status indicators during the period of service, with even some improvement in nutritional status, without any significant contribution from growth monitoring .
To present and analyse evidence that growth-monitoring programmes confer measurable benefits on the children for whom growth charts are kept, Ashworth and Feachem  reviewed the impact of growth monitoring on the reduction of diarrhoea morbidity and mortality or severity either by improving the nutritional status of infants and young children or by increasing their contact with primary health care services. They examined projects in Indonesia , Thailand , Jamaica [39, 40], Ghana and Lesotho , and Malawi . A comparison of clinic-based and village-based weighing programmes in Indonesia revealed that the two were comparable in improving child nutrition . Children participating in the programmes appeared to parallel the Harvard growth standard once they had passed the age of 12 months. The two studies from Jamaica showed a similar reduction in mortality, yet in only one of them  was there a reduction in the prevalence of malnutrition. In Malawi some improvement in nutritional status was found, as judged by a reduction in the percentage of participants with less than 80%, weight for age . However, other services were also provided and their contributions to the improvement have to be taken into account.
According to Yee and Zerfas , the evaluation of the functional utility of growth monitoring by Ashworth and Feachem  tended to view the relationship between growth monitoring and the programmes results in a direct linear manner without assessing whether or not crucial intermediary steps were taken. Yee and Zerfas, reviewing selected projects in India, Ecuador, the Dominican Republic, Tanzania, and Thailand, found that it was difficult to interpret the results because of a usual lack of adequate baseline controls but that in general the prevalence of malnutrition (usually defined as a weight for age under 75% of the reference standard) was reduced over time.
Growth monitoring and mortality
In view of the present emphasis on child survival, Yee and Zerfas  regarded as a key issue the question of whether adding growth monitoring and associated actions to a project contributes to reducing mortality, but they could find no project adequately addressing this question. There is evidence that the prevalence of undernutrition can be reduced in programmes incorporating growth monitoring (e.g., the TNINP in India, the USAID-funded primary health care project in Thailand, and the JNSP project in Iringa, Tanzania). There is far less evidence that mortality has been reduced [25, 40]. In the absence of any systematic studies, the contention that adding growth monitoring to a programme will decrease mortality remains to be proved.
Growth monitoring and the use of primary health care
Growth-monitoring programmes have been shown to increase the use of primary health care services. Village-based programmes in Haiti, for instance, have been found to lead to such an increase, which can be expected to reduce diarrhoea mortality and morbidity or severity [ 14] . Data from Jamaica suggest that growth monitoring had an impact on diarrhoea mortality through the increased use of curative services. Rohde  argued forcefully that regular growth monitoring can increase the demand for primary health care services, pointing out that it is the only recurring activity that brings mothers and children into frequent and predictable contact with health services. At the village level, he says, it has become the basis of primary health care in a number of countries such as Indonesia and Haiti. In ICDS programmes, growth monitoring is an integral part of primary health care. . It has been suggested that weighing has created a forum through which other primary health care activities are being conducted and that their availability in turn has made the weighing programmes more attractive to mothers . Thus, a much-needed breakthrough in problems related to primary health care is provided.
Growth monitoring and nutrition education
Data from Thailand has indicated that growth monitoring per se was ineffective in changing nutritional status but that it was effective when combined with nutrition education .
An examination of the strengths and weaknesses of traditional growth-monitoring and nutrition-education programmes shows that the two are ideally suited to be complementary, according to Griffiths . Growth monitoring makes it possible to give advice appropriate to the individual child's needs at the time it is needed. Thus nutrition education can be made more effective by making it more specific, action-oriented, individualized, and relevant. It will differ from traditional nutrition education in being tailored to the specific child's needs. In Hanover, Jamaica, project workers felt that growth monitoring itself was an intervention, as mothers learned so much about the relationship between diet and health by watching their children's growth patterns that this alone led to dietary improvements that substantially reduced malnutrition and mortality . In the TNINP, growth charts serve as an educational tool, as most mothers can interpret the trends of the growth lines and seem to be able to relate a downward trend with an illness, especially diarrhoea . Sinha  believed that growth charts can be practical and powerful in teaching mothers how to protect children from malnutrition and foster better nutrition through simple messages and discussions.
Hendrata , on the other hand, believed that, while growth monitoring was effective in Indonesia in expanding programme coverage to 30,000 villages in less than five years, its effectiveness in changing mothers, behaviour and the nutritional status of children remained unclear. The RUHSA experience [27, 34] suggests that growth charts have limited use as an educational tool in nutrition education. Moreover, education should be considered a preventive or promotive strategy and not a treatment prescribed at the point of diagnosing faltering growth. It also should be targeted to both men and women in the entire community and not only to mothers whose children show faltering growth. Educating mothers directly about appropriate, practical ways of rearing and feeding children appeared to be far more effective than using that time to teach them the significance of weight measurements and growth charts. Gopalan and Chatterjee  concluded that, despite some enthusiastic claims (although these may be true in a few instances), the evidence that growth charts at present contribute significantly towards educating mothers is not convincing. While experience in many countries indicates that most mothers can readily distinguish the significance of a gain or loss in the weight of their child from one weighing to the next, results obtained in Papua New Guinea suggest that this generalization may be too optimistic .
In response to the need for improved growth-monitoring techniques, a "bubble" chart has recently been developed by Griffiths  with World Bank collaboration, to replace standard charts, which were found to be hard both for health workers to plot and for mothers to understand. The new chart has-been tested successfully in several countries .
Growth monitoring and nutrition surveillance
A very common error, which can be traced in the literature, is a confusion between growth monitoring and nutrition assessment/surveillance. Although to many programme managers the two activities may appear similar, as is evident from this review, they are in fact different strategies based on totally different concepts, operations, and purposes. To make the distinction easier, the main differences have been summarized elsewhere [1, 22, 23, 47]. Some of these differences are as follows: The main strategy of growth monitoring is the preservation of normal growth, while that of nutrition surveillance is the detection of undernutrition. In growth monitoring, therefore, action is based on weight changes of individual children, and in nutrition surveillance it is concentrated on the nutritional status of groups of children. The major approach in growth monitoring is educational and motivational, while in surveillance it is diagnostic and interventional. In growth monitoring, the response to early detected growth faltering is early home intervention based on local knowledge until growth is resumed. In nutrition surveillance, on the other hand, the response can be nutritional rehabilitation, often with supplements continuing until good nutrition is established in the community. Growth monitoring emphasizes the maintenance of good nutrition for individuals and should cover all infants in a community. Nutrition surveillance emphasizes the detection of malnutrition using representative samples of children. Finally, the weight cards used in growth monitoring should be simple, emphasizing growth, while those used in nutrition surveillance must be precise, with emphasis on nutritional status.
It has been suggested that growth-monitoring data could be used for nutrition assessment. If the two activities are combined, the concern for gathering statistically accurate and reliable data would probably be incompatible with the meaningful exercise of growth monitoring based on effective communication and education . Further, it is likely that the confusion between the two would be increased, unless the purpose, structure, and function of each were clearly defined to ensure their precise applications . It should be noted that growth-monitoring data are not randomly based and may not be accurate enough for the purposes of nutrition surveillance. Srilatha  considered that the use of growth-chart data for measuring the proportion of malnourished children in a community is not justified because of problems of representativeness, accuracy, and uniformity. In a comparison to examine the validity of clinic-based nutrition surveillance for estimating the prevalence of undernutrition , the prevalence of low weight for age among first-time clinic attenders in Swaziland was found to be very similar to estimates based on a 1983 national nutrition status survey, but the prevalence of underweight among children attending two or more times a year was less than half that of first-time attenders. The investigators concluded that surveillance data, particularly from repeat attenders, will not necessarily provide a valid estimate of nutritional status within the general population or of differences between regions. In a study of trends in malnutrition in five African countries, using clinic data from the Catholic Relief Services food and nutrition programme, the data, although they were from a non-defined and changing sample of children attending the clinics, were found to give plausible descriptions of trends in the prevalence of malnutrition when the major bias proxied by coverage was controlled for. However, the actual estimates of prevalence may not accurately reflect population prevalences. This can be verified only by representative sample surveys or census data.
Feasibility conditions for implementation
For growth monitoring to be useful, two basic conditions must be fulfilled: there must be a clear understanding of the objective, and there must be a health infrastructure capable of using the technology effectively. Without these, even with heavy investment, growth-monitoring programmes will fail, Gopalan [81 points out, and actual field conditions in developing countries should be carefully taken into account. He therefore advises supporting and strengthening the development of integrated programmes of maternal and child health care with growth monitoring as a part, rather than promoting growth-monitoring programmes as such in isolation. Rohde  suggests that, instead of integrating growth monitoring into primary health care, it would be more appropriate to integrate primary health care into growth monitoring. By making measured growth the criterion of health, we provide an important link in the cycle of primary health care for young children which has been missing until now. The paramount importance of adequate backup facilities for growth monitoring, as indeed for all primary health care programmes, has been stressed 
For a growth-monitoring programme to be feasible and have the potential to affect nutritional status and/ or primary health care contacts, the regular attendance of children is of great importance . Srilatha concluded that monthly weighing was not feasible because of inadequate attendance . In certain programmes, food supplements are used predominantly to encourage attendance. It was observed that the number of children coming to a maternal and child health clinic in central Java for regular weighing was extremely small because of other problems, although dried skimmed milk was distributed to those attending . Attempts to improve attendance have so far produced many arguments and conflicting results [6, 23, 51].
A discussion on the effectiveness of any programme would be incomplete without mentioning cost issues, and Yee and Zerfas  point out that cost must be considered in a comprehensive way as a factor in the programme's feasibility. However, this is another controversial point in the literature reviewed. Griffiths [9, 52] considered the cost of adding growth monitoring to an ongoing programme to be extremely low. The expenditure on growth charts and equipment should be minimal, as both last for many years and can be used by all personnel. What increases the cost of growth monitoring is training and supervision and the indispensable education component. The actual materials needed may be basic, unsophisticated, and inexpensive; the essential equipment for an entire programme in Angola cost no more than US$10,000 . No detailed assessment of the costs of transport, personnel, etc. has been carried out, however.
On the other hand, the cost of maintaining growth charts may be very high compared with that of conducting periodic samplings . And it has been estimated that the initial expenditure for scales alone for a growth-monitoring programme for all of India might be US$21 million . Consequently, the estimated cost of some current weighing and growth-charting operations in developing countries constitutes a high proportion of their overall national health budgets. Projections of the costs of operating a growth-monitoring and nutrition-education programme on a national scale similar to an Indonesian pilot project, according to the World Bank , indicate that 0. 1% of the national budget would be required, which would be a sizeable expenditure for many countries.
It has been claimed that, of all the measurements that can be made on children in developing countries, weighing is the most likely to be useful and its cost-benefit value is very high . Others, however, have concluded that the costs for growth monitoring are usually poorly documented and the benefits are difficult to quantify for analysis of cost effectiveness .
Many countries have implemented growth monitoring in various combinations with other nutritional and health care services. However, from this brief review it is evident that considerable confusion exists on its major objectives. This is because, apart from monitoring the growth of a child, the technique and the information obtained on the charts (and sometimes the charts themselves) can be used for many diverse purposes - for example, to evaluate the effectiveness of other nutritional intervention programmes on children's health, to elect the beneficiaries for dietary-supplementation programmes, to estimate prevalence rates of malnutrition and underweight in nutrition-surveillance programmes, to follow up the efficacy of treatment of sick or malnourished children, to trace children not attending or not returning to health centres for immunization, and so on.
Apart from being a longitudinal record of weight changes, the cards can be used to register a child's health or family history: birth weight, weaning time, diarrhoea episodes, dose and date of vitamin A capsules or iodized oil injection received, and the like. It is therefore not surprising to find that in many growth-monitoring programmes, weighing and growth charts are mainly used for these purposes, obscuring the major objective and thus reducing the programmes, effectiveness.
Individual counseling is an important part of any growth-monitoring programme. When combined with appropriate responses for individual children, growth monitoring can no doubt improve nutritional status. Benefits as part of primary health care services are evident through increasing contact and attendance, decreasing the prevalence of malnutrition, and establishing better nutritional practice for child rearing. Even very poor families may have some resources available that can be redirected towards the health of their children if opportunities provided by growth monitoring for face-to-face nutrition education are taken.
Small-scale programmes have repeatedly proved successful, showing the inherent benefits of growth monitoring. What therefore happens in large-scale programmes is improper implementation, with little or virtually no emphasis on the main objectives. It is evident from the literature that not only the mere contact but the frequency and quality of social interactions between health workers and mothers are the important issues on which much of the outcome depends. The potential of growth monitoring for maintaining good nutrition and health in the most vulnerable group of the community can be realized only if techniques and understanding are directed to the issue of proper implemention of the main objectives.
I gratefully acknowledge the support of the ACC/SCN, and in particular of Dr. John B. Mason, its Secretary .
Mode of perinatal transfer of HIV-1
With the spread of the human immune virus I (HIV-1), it is becoming increasingly common for pregnant women to be infected (table 1). As a result of perinatal transmission, the number of infected offspring is consequently also increasing. Three possible routes of perinatal transmission have been considered.
TABLE 1. HIV-1 antibody-positive pregnant women, 1986-1987 (percentages)
|New York City|
Source: A. Nahmias, personal communication.
In most instances the infection is presumably transmitted in utero. This conclusion is supported by a short incubation time after birth in some cases , a high incidence of hepatosplenomegaly and low birth weight , and the presence of foetopathy in children with acquired immune deficiency syndrome (AIDS) and AIDS-related complex (ARC) . The dysmorphic features suggest exposure to the virus in the first trimester of pregnancy . Isolation of HIV-l from a 20-week foetus , detection of HIV-1 in cord blood [1, 4] or within 24 hours of birth , and demonstration of the virus in infants delivered by caesarean section  are also taken as evidence. A second possible route of transmission is the contact of the neonate with infected blood from the mother at delivery.
Finally, consideration is being given to the possibility that breast milk can transfer HIV-1. The virus has been isolated from the cell-free clear middle layer from centrifuged milk from three seropositive mothers . The cells in the milk could not be cultured for the virus because of bacterial contamination. The possible role of antibodies present in milk for infectivity of the virus or for its detection has not been studied. The capacity of the virus to infect by way of the intestinal tract, reaching its receptors on T4 cells, is not clear. However, about 15% of the many intra-epithelial lymphocytes in the gut mucosa are T4 cells.
It should be noted that human milk contains significant numbers of leukocytes. They reach a maximum three to four days after the onset of lactation, with from 500,000 to 10 million cells per millilitre. By four to six weeks after parturition, they are present at less than 100,000 per millilitre. The number of lymphocytes declines sharply within two to three days postpartum, whereas the number of macrophages declines less and remains at detectable levels. With increasing milk volumes they continue to be present . It is likely but not yet demonstrated that such milk cells may be infected with HIV-1.
Only one case of transfer of HIV-1 from mother to baby has been published in which the transmission could have been through breast milk . The mother was infected by a postpartum blood transfusion, and her infant became seropositive either due to the six weeks of breast-feeding or by some other close contact with the mother. Four other cases, from Rwanda and Zaire, not yet published, also suggest breast milk as the mode of transmission.
No definite data exist as to the relative frequency of different modes of transmission of HIV-1 from mother to offspring. Among 104 children in the United States with AIDS who were studied for disease transmission, the risk factors were as follows, in order of frequency: maternal risk factors, paternal risk factors, sexual abuse, and exposure to needles at home . The majority were children of intravenous-drug-abusing mothers and/or fathers or children of bisexual fathers.
Although intrauterine infection from the mother may be most common, Chiodo et al. have described 12 babies from HIV-1 seropositive mothers, of whom 3 of 7 delivered normally became infected, but none of the 5 delivered by caesarean section did . This could be taken to suggest that natural delivery increases the risk of perinatal transfer, but the matter certainly needs further study.
The risk for babies of seropositive mothers of developing ARC or AIDS varies according to different sources, with figures ranging from 15% to 50% for the first child and from 60% to 80% for later siblings [1; A. Nahmias, personal communication]. Only one of a pair of identical twins was infected in one instance ; the healthy twin was followed for 10 months.
The risk of transfer of HIV-1 in breast milk
Few cases are known in which breast milk can have transferred HIV-1. Yet the risk cannot be discounted, and countries such as the United States  and Sweden are advising seropositive mothers not to breast-feed. It is also advised that human-milk banks should pasteurize their milk [12, 13]. Such advice can be followed in places where diagnostic tests for HIV-1 infection are readily available and where pasteurization can be arranged. In such areas it is likely that the risk of contracting various gastro-intestinal and respiratory-tract infections is limited, and decreased frequency of breast-feeding may not have disastrous effects.
The problem is much more serious, however, in areas where such infections are frequent and breast-feeding is very important in protecting against them. A recent estimate of mortality from various diseases in east Africa shows that approximately one person dies of AIDS in large cities for every 10 who die of diarrhoea, but in rural areas the proportion is only one death from AIDS for every 4,000 deaths from diarrhoea , This illustrates the complexity of the problem. It might be added in this connection that the median relative risk of diarrhoea mortality is estimated to be about 25 times as high in non-breast-fed as in exclusively breast-fed infants.
With the limited information presently available concerning the risk of transfer of HIV-1 through breast milk, it seems of prime importance to propose research that can expand our knowledge. Specifically, it is important to determine whether or not those at risk of being infected through milk have not already been infected in utero. Any recommendations concerning breast-feeding in relation to AIDS must be provisional and temporary until additional information appears. It should be realized that today the mortality of infants from AIDS is still minimal compared to that from conditions such as diarrhoea. Diminished breast-feeding would make that difference even more striking .
Maarten D. C. Immink
This paper offers some ideas for the establishment of community-designed, -operated, and -evaluated food and nutrition surveillance systems, in support of food security of the poor and bottom-up planning processes for socio-economic development in Central America. Current social processes are opening up opportunities for the poor to play a greater role in solving poverty problems and to contribute to their own food security. Community-based food and nutrition surveillance systems should be seen as an instrument of socio-economic development.
The current state of food and nutrition surveillance in Central America
The predominant current view in Central America regarding food and nutrition surveillance is that its main purpose is policy and programme advocacy in relation to improving the nutritional status of low-income populations. What have been explicitly identified as food and nutrition surveillance activities have largely been confined to the health sector, using health and nutritional status indicators for the identification of population groups at some degree of risk of undernutrition. Recently, macro-level tools such as food-balance sheets and basic-food-basket analysis have also been considered as relevant instruments of food and nutrition surveillance.
Under the aegis of a regional and EEC-supported food-security programme, efforts are being initiated in some countries to integrate sectoral information systems into a public-sector network in support of food-security activities. Much of the recent technical discussion among public-sector agencies has centred on organizational structures, indicators to be generated, and needed infrastructures for data processing and analysis. What has been largely absent from these discussions is a careful consideration of the use of surveillance data, and of the mull-functional dimensions of food and nutrition surveillance systems as proposed by the FAO and other agencies .
Food and nutrition surveillance systems are seen as supporting public-sector, top-down planning processes and as a basis for sectoral actions in the form of national programmes. Information routinely or periodically generated at the community level is rarely fed back in support of local decision making and action taking. Serious operational bottlenecks at national levels related to data processing and analysis have greatly limited the effective use to which the eventually disseminated information is put. Lack of clarity in the definition of the functions of the surveillance systems and of operational concepts and implementation procedures, as well as "territorial,, concerns among public-sector agencies, have prevented the Central American countries from organizing such a system as a basis for multisectoral decision making and action taking in the short and long run. It is difficult to see how surveillance activities to date have contributed effectively to the food security of the poor in Central America.
Socio-economic development, popular participation, and food and nutrition surveillance
Socio-economic development may be viewed as the full achievement of human potentiality in both material and spiritual terms. Food security at all times is a high-priority component of socio-economic development, in fact an essential pre-condition. Food and nutrition surveillance is an operational instrument to achieve food security for all members of society, and thus an instrument of socioeconomic development.
It has been recognized for some time that popular participation is essential for development, and that development is in fact participation . This idea found formal expression in 1979 in the declaration of the World Conference for Agrarian Reform and Rural Development, which stated in part: "Rural development strategies can realize their full potential only through the motivation, active involvement and organization at the grass-roots level of rural people. . . in conceptualizing and designing policies and programmes,, . This focus has since then guided the development work of UN specialized agencies, and of some international and national non-governmental organizations, though often it has not been put into effective operation.
Popular participation occurs essentially in two ways. In the first, certain groups are organized in order to participate in development activities designed and operated from above, often by a public-sector agency. Participation is seen as making the outcome of the activities more effective in accordance with criteria adopted by central planners, and may consist of community members providing free labour services or material goods for development projects. The second type of participation is a bottom-up process that results in grass-roots organization and allows the poor to take major responsibility for their own development in accordance with their own perceived needs. As part of the process the poor are better equipped to formulate and express their needs and their priorities for actions, and to participate technically in designing and implementing development activities, thus over time decreasing their dependence on external technical and financial sources of development assistance.
It is my contention that there is a role for food and nutrition surveillance in fostering socio-economic development with grass-roots-level participation. This role has not yet been recognized in Central America. Economic growth is often confused with socioeconomic development, and participatory processes are only seen as necessary in order to increase the local impact of national and regional social action and development programmes. However, new social and political currents start to create space for effective community organization at grass-roots levels in support of bottom-up development activities. This is probably even more true for populations living in marginal urban areas than for the rural poor. Nevertheless, food and nutrition surveillance activities, if designed, operated, and evaluated by communities organized at the grass-roots level, can in turn strengthen development-oriented participation processes and contribute significantly to the food security of the poor and their socioeconomic development.
Such activities by the community should result in an understanding of the factors that produce food and nutrition problems and should provide a basis for self-determination and implementation of solutions in accordance with the aspirations collectively held by the community and within the broad context of socioeconomic development.
The implementation process
The following should be the principal functions of community-based food and nutrition surveillance systems:
The community should set priorities as to what basic functions should be emphasized in the initial design of the systems. For example, communities that frequently suffer from acute food shortages may wish to emphasize the system's early-warning function.
No operational blueprint is possible of community-designed, -operated, and -evaluated food and nutrition surveillance systems. Inherently, they are not necessarily duplicable, although they may coincidently have elements in common. This may be partly the result of a demonstration effect among communities that are in contact, especially those that face similar poverty problems or acute hunger conditions.
The role of external (to the community) technical assistance should be to help to create the necessary conditions so that the community can effectively manage, use, and evaluate the system. This may take the form of training community members in data recording, tabulation, and interpretation, establishing certain infrastructure for data processing and analysis, or technical assistance with the design of the system. In general, the system should be simple and flexible in design and should contemplate incorporating qualitative information, not only quantitative data. The former has the advantage of often serving more effectively as a basis for rapid decision making or action taking, especially in situations that do not allow or require in-depth study of underlying causes. The system should be flexible enough so that it can continuously be adjusted in accordance with the current data needs of the community.
The sequence of different stages of the implementation process might be as follows:
During the first stage, the community collectively thinks about and discusses its food and nutrition problems, their immediate causes, and what priorities should be given to solving them. For this process of diagnosis the community may use existing data and decide to obtain other information applying simple data-collection techniques. External technical assistance may facilitate this process. In participatory sessions the community decides on courses of actions and sets priorities. Identifying and assigning priorities to problems and courses of action constitute the bases for identifying information needs, and thus for the design of the information system. At this stage external technical assistance may be required in the forms of training of community members, technical help with validation of the system, infrastructure, etc. Once the system is operational, it should be evaluated continuously by the community. For this purpose, some operational indicators should be included that will allow the system's performance to be monitored as a basis for making adjustments over time.
Community-designed, -operated, and -evaluated food and nutrition surveillance systems should not be seen as the only information systems in support of national food security. Decision making and action taking in relation to food security and socio-economic development take place at different levels and require information at those levels, including macro-level information and data obtained at the community level and integrated at aggregate levels (e.g. , local, regional, national). But in Central America, as in other parts of the developing world, food and nutrition surveillance should not be limited to supporting top-down processes of decision making and action taking related to food security and socio-economic development. In view of the diminishing capacity of the public sectors to provide technical and financial solutions to poverty problems, there is an increasing need for the poor to be able to determine and take responsibility for their course of development.
Once operational, the surveillance systems are likely to be effective because they allow firm integration of information-generating activities with decision making and action taking; they may strengthen community organization at the grass-roots level, which in turn will aid the social development process; they increase the effectiveness of external resources in contributing to the community's food security and socio-economic development; and they mobilize community resources in a more rational way.
The firm integration of information-generating activities with decision making and action taking ensures that the information can be transformed rapidly into decisions, and that it is appropriate and leads to actions that are relevant to local conditions and in accordance with the aspirations and priorities established by the poor themselves.
Community organization does not take place in a vacuum. It is generated as a need and often constitutes a pre-condition for taking actions. To the extent that these systems serve as a catalyst for strengthening community organizations at the grass-roots level, they contribute also indirectly to the social development process by making other actions at the community level more effective, whether undertaken with external or community-based resources. The community becomes better equipped to define what resources are needed in accordance with the priorities with which actions are to be taken, the timing of resource use, etc., thus increasing their impact. Depending on the process by which they are designed and operated, surveillance systems can be instruments that allow the communities to understand their own reality and thus take greater responsibility in charting their own development course.
Marylou L, J. Mertens and Gretel H. Pelto
The purpose of this statement is to express the need for training programmes explicitly designed to improve the quality of social-science inputs into nutrition and health research and programme activities.
From a variety of sources, including the expanding body of data on the sociaI epidemiology of nutrition and studies of the social correlates of dietary practices as well as the experiences of health-care and nutrition professionals, the role of socio-cultural and behavioural factors in the aetiology of nutrition problems and their solution is becoming increasingly more evident. Nutrition has always been a multidisciplinary field, drawing concepts and techniques from other natural sciences, including physiology, biochemistry, molecular and cellular biology, and medicine. Indeed, the common practice of naming a university unit devoted to nutrition a "Department of Nutritional Sciences,, reflects the multi-sectoral nature of the enterprise. Recognition of the social components of nutrition has led to the need to cast a still wider net, bringing social-science concepts and methods into the investigation of nutrition issues and nutrition-policy implementation.
One important vehicle by which new methods and theoretical perspectives are introduced into a field is through collaborative research activities. The expansion of the multi-disciplinary team of nutrition investigators to include anthropologists and other social scientists is a logical and timely extension of biological investigation. Including social scientists in the development and implementation of nutrition and health programmes would bring a social perspective to bear on direct problem-solving activities.
There are, however, a number of serious barriers to effective collaboration. One of the most important is the scarcity of social scientists with the requisite background, training, and professional situations. Throughout the world, in both industrialized and developing countries, nutrition and health-care projects have found it difficult at best, and often impossible, to develop appropriate working relationships with social scientists. To a large extent, this situation reflects the orientations and traditions of training and research in the social sciences. For example, in anthropology, attention to issues of contemporary health and nutrition, particularly in applied contexts, has Only recently begun to attract serious attention. Moreover, the lack of experience in quantitative methods makes it difficult for traditionally trained anthropologists to collaborate with bio-medicaI researchers.
The professional circumstances and expectations that social scientists face is another barrier. Typically, most anthropologists are employed in universities, where administrators are reluctant to encourage their staff members to participate in projects that take them away from their academic pathways. Promotion and job security are tied to academic publications in the primary discipline, rather than to applied, multidisciplinary, problem-solving research. Positions outside academic settings have only recently become viable alternatives, and they rarely carry the same job security and opportunities for professional development that one finds in a university.
Social organization and attitudes within the biomedical and nutrition community are another source of the problem. Too often, social scientists find themselves in a low status position compared to comparably trained big-medical colleagues. Lack of understanding about the nature of social-science concepts and methods and unfamiliarity with the language and vocabulary lead to misunderstanding, charges of using jargon, and denigration of data quality. At the same time, social scientists often fail to appreciate the complexity of the big-medical issues and are unable to communicate directly about the central issues of concern to biological researchers and clinicians.
Despite their prevalence and historic depth, none of these barriers are irremediable, and ail of them need to be addressed. Education and training programmes are central to the solution. They are critical for ameliorating the "personnel crises,, due to the scarcity of social scientists with appropriate skills and interests. To some extent training programmes will also help to moderate the other barriers as well.
This report outlines several strategies for training, each of which addressees some of the needs that have emerged in recent years.
Long-term professional training
The most intensive and complete training for a professional specialization in nutritional anthropology or social nutrition occurs through a graduate programme leading to an advanced academic degree. Such a programme would draw from both nutrition and anthropology, with the student obtaining a solid educational background in both fields. As is true of any composite field, such as urban planning or social policy, certain features of a "classic education,, in the traditional academic disciplines from which the specialization emerges must necessarily be forgone. However, this does not mean that individuals trained in such programmes must be unprepared for more traditional academic or research positions as a consequence of an educational programme that prepares them to engage in multi-disciplinary research.
Specialized social-nutrition (or social-medicine) professionals can also be trained with a much less lengthy programme if the individuals are already academically prepared. The model of the "postdoctoral year" of specialized study is highly appropriate for both young professionals and older researchers who are seeking new research directions. For example, a one-year post-graduate programme leading to a certificate in nutritional or medical anthropology could be offered for social scientists who wish to become skilled researchers in this area but have little previous experience with medical and nutritional issues.
Integrated curriculum and specialized tracks
The inclusion of social-science components in the training of health-care and nutrition professionals occurs to some extent in many (if not most) professional curriculums. However, too often these subjects are seen as irrelevant, too academic, or dull by students who are eager to learn hands-on, clinical skills.
Integrating social and cultural perspectives directly into technical subject matter has proven to be a successful mechanism for overcoming these attitudes. For example, in courses in malnutrition the social and cultural factors in the aetiology of specific nutritional deficiencies can be presented together with the pathophysiology .
The establishment of a social-nutrition or social-epidemiology track as a specialized option within an MS or MPH programme is another means of increasing the level of social-science skills in big-medically trained professionals.
Short-term training programmes
Awareness-building workshops: nutritional anthropology for big-medical personnel, and nutrition for social scientists
As in all multi-disciplinary situations where individuals must communicate with new vocabulary, new concepts, and different expectations, the involvement of social scientists in nutrition will inevitably engender problems for all the participants. A modicum of familiarity with the modus operandi and basic concepts of the contributing fields is probably a prerequisite for effective collaboration.
Awareness-building workshops are intended to address this need. The purpose of workshops on nutritional anthropology for big-medical personnel is to introduce them to basic concepts and methodological orientations of anthropology and to provide a general overview of some of the key socio-cultural questions in contemporary nutrition. Similarly, workshops directed toward social scientists are for the purpose of introducing them to key methods, concepts, and issues in nutrition and to provide them with some familiarity with the vocabulary of discourse in nutrition.
The purpose of skill-development workshops is to provide an opportunity for individuals to acquire new research techniques and capabilities. Given the short time frame of a workshop, this mode of skill acquisition is probably most useful in connection with specific, well-defined projects. Workshops can be focused on proposal development, on protocol development and data collection. on tools for data analysis, or on aspects of programme implementation.
The specific purpose of a workshop and the types of participants and project needs must dictate format and content. However, it is possible to make some suggestions about general principles to consider with regard to short-term workshops. A general framework to guide the design and organization of workshops might include the following features:
- social and biological interactions in nutrition
- the perspective of social epidemiology,
- the role of culture in relation to other social and biological factors,
- multiple levels of analysis, from the individual to household, community, and society,
- microlevel-macrolevel linkages,
- qualitative and quantitative methodologies and the linkages between these modes of data collection and analysis.
In view of the urgent need to develop multi-disciplinary research and programme teams to work on pressing issues of nutrition and health, and the present shortage of appropriately oriented and trained social scientists, the potential of short-term workshops should be seriously considered. Certainly short-term training alone is insufficient, and longer-term professional development opportunities must also be developed. Several different types of training and education strategies are required, and it is important for various groups - professional organizations, international agencies, academic institutions, and government agencies - to begin to address these needs.
Nevin S. Scrimshaw
I don't quite know how to convey the sense of privilege and pleasure that I experienced when I was invited to give the Ruth Hueneman lecture for 1987. There is no colleague for whom I and my anthropologist wife, who worked with Ruth in Guatemala, have greater respect, admiration, and affection. Our esteem is based on many years of personal association in a variety of activities in several countries. Ruth is an outstanding professional because of her knowledge, experience, judgement, and skill, combined with a conscientious capacity for hard work. She also has special attributes that are even rarer: kindness and generosity to both colleagues and students, and grace under pressure or adversity. She is always a lady. I know that this is an an old-fashioned word, but I cannot very well use the word "gentleman,,; the thrust of both words is the same - she is dignified, courteous, and kind - always.
This lecture focuses on the concept that improving the nutrition of populations is best approached as part of a more comprehensive health effort. This theme is closely related to Ruth Hueneman's lifelong professional interests. With the shift in WHO's approach to primary health care following the Alma Ata Conference in 1978, there was a change in which she participated in the orientation of the WHO nutrition unit. It is also noteworthy that some of the insights presented in this lecture come from research on the actual experiences of peasant women with the health care system in Guatemala, a study in which Ruth Hueneman participated .
In most developing countries with high mortality among infants and young children, Western medicine does not reach the majority of the population, even in a grossly deficient manner. What care is available is too often provided by an auxiliary health worker with limited training, no medicines, and essentially no medical supervision or back-up. Such individuals can do very little to cope with the health problems of the population for which they are supposed to be responsible. Moreover, in developing countries with high morbidity and mortality, health centres, even when well staffed by physicians, usually do not have the activities required to prevent malnutrition and infection.
Attempts to lower high rates to morbidity and mortality by increasing the availability of curative medical services alone invariably fail. What is required is measures to prevent malnutrition, diarrhoea, and the common communicable diseases. Such efforts would have more than the benefit of preventing single diseases because of the synergistic interaction of malnutrition and infection and of one disease with another. Correction of malnutrition reduces morbidity and mortality from infections, and, conversely, infections are the major precipitating cause of frank nutritional disease. The primary health care approach recognizes these relationships and, as defined by WHO, places more emphasis on the prevention of disease and the promotion of health than on curative medical services. The child-survival strategy of UNICEF focuses on a set of primary health care activities for mothers and young children.
Unfortunately, like other health professionals, neither non-medical nor medical nutritionists have traditionally been trained in this concept. This lecture will try to explain, for those concerned with nutrition' the rationale behind the primary health care approach and what is required to implement it effectively. It will also analyse the place of specific child-survival strategies within the health care system. I will emphasize the need for the integration of nutrition activities into other aspects of health care, and for a change in the way most public health nutritionists are trained and accustomed to think of their responsibilities.
Limitations of curative medicine
I can say with the conviction of personal experience that curative medicine alone at the village level can do very little to lower high child mortality. I maintained weekly clinics in two Guatemalan highland villages for over a year, and the only lives I may have saved were those of two children whom I took to the regional hospital  .
INCAP then established a study in three other villages that compared programmes of medical care in one and supplementary feeding of preschool children in another, and left the third as a control. The treatment village, Santa MarCauque, was staffed daily by a well-trained physician and public health nurse, who had virtually unlimited medicine available but provided no preventive services. Over a period of five years, there was no discernible effect on the high child-mortality in this village . However, in the village of Magdelena Milpas Altas, with only the supplementary feeding of preschool children there was a significant drop in mortality, including that from infectious diseases, despite a lack of additional medical attention.
From these and other experiences I discovered for myself what is well known by health providers who have had such field responsibilities: the largest part of seeing patients involves general support, and very few persons need either highly trained specialists or secondary- and tertiary-care facilities. At all levels of the health care system, time spent on prevention can do far more to improve the health of populations than that spent on seeing patients, especially under the conditions prevailing in most primary care facilities. The most important of the preventive measures are those that directly or indirectly improve nutrition.
Some years ago I had an opportunity to observe a dramatic example of the potential significance of preventive compared with curative services. In 1967 a young public health physician, Warren Berggren, arrived at the Schweitzer Hospital in Deschapelles, Haiti, and found one large, desperately overcrowded ward for children with severe protein-calorie malnutrition and another similar one for neonatal tetanus, These disorders were the leading causes of the very high infant and preschool mortality rates in the region. Most infants admitted with tetanus died. Many of the severely malnourished died in the hospital, and others succumbed at home when they were discharged, only partially recovered, to make room for new admissions.
Dr. Berggren began his work in Haiti by using a jet injector in the neighbouring markets to administer tetanus toxoid in order to immunize all women of child-bearing age, and within a year the neonatal tetanus ward was emptied of patients from the district. He then attacked the more difficult problem of severe malnutrition in young children by establishing so-called Mothercraft Centres, an approach pioneered in Haiti by William Fougere and Kendall King . These centres were similar to the contemporaneous "recuperation centres,, of Colombia and Guatemala  and to those subsequently established in the Philippines . Young children in every village were weighed to find those so malnourished as to be at risk for marasmus or kwashiorkor, as judged by low weight for age. The mothers were then persuaded to bring the children identified in this manner to a special kind of daycare centre in which they received three good meals, and in which the mothers participated in rotation in the purchase, preparation, and serving of the food.
The children did improve; the mothers did learn; and there were few recurrences either in the children treated or in their siblings. The effect on the malnourished children was as startling as it had been for those with neonatal tetanus; soon no more cases of severe malnutrition were coming from the district. No investment in curative medicine could have produced such a reduction in mortality, to say nothing of the benefits to more normal growth and cognitive development. Preventing disease at any age makes a great deal more sense than allowing it to develop unnecessarily and then treating it. Moreover, prevention is far less costly than treatment.
The decline of mortality in industrialized countries
There is a long history of human experience demonstrating that most disease need not occur. In Europe there was a decline in mortality in the latter nineteenth and early twentieth centuries from infectious diseases such as gastroenteritis, measles, whooping cough, diphtheria, tuberculosis, puerperal sepsis, and pneumonia, long before the advent of any effective specific therapy  Figure 1 (see
Figure 3 (see
Fortunately, effective specific methods of prevention, particularly immunization, are now available against many of the more serious infectious diseases, and better nutrition and improved environmental sanitation and personal hygiene can reduce the occurrence of other diseases. Moreover, food enrichment or fortification can eliminate the specific nutritional diseases associated with deficiencies of iodine, vitamin A, and iron. The combination of controlling infection and improving nutrition has had dramatic effects in those developing countries that have applied it conscientiously. Examples are China, Chile, Costa Rica, Cuba, Vietnam, Singapore, Sri Lanka, and Taiwan.
Thus far, my remarks have focused on the integration of nutrition and other health measures in developing countries. I believe strongly, however, that preventive nutritional measures are important for minimizing morbidity and mortality in all countries. We are steadily learning more about the importance of diet in the prevention of mortality from the chronic diseases that are the leading causes of morbidity and mortality in industrialized countries. The nutritionist is at least as important as the cardiologist in the prevention of hypertension and ischaemic heart disease and has a role to play in the prevention of some forms of cancer. In short, improved nutrition has direct benefits, whether in developing or industrialized countries, on morbidity and mortality from a wide variety of acute and chronic diseases.
For this paper, however, I will continue to focus on the integration of nutrition and other health efforts in the developing countries, beginning with the WHO primary health care strategy. Because primary health care has become the major focus and basis for current global efforts to improve health in developing countries, it is essential to understand its origins and what it requires.
The Alma Ata primary health care approach
In the 1970s there was increasing awareness that in the developing countries, where three-quarters of the world's population live, hundreds of millions of people were suffering and dying from malnutrition and diseases that could be prevented. Even in the industrialized countries, many millions of people were not enjoying a level of health which the application of available knowledge could assure.
In September 1978, ministers of health or their representatives from 134 countries met in Alma Ata under the aegis of WHO to discuss the failure of health programmes to meet the needs of their populations. From this meeting emerged the goal of health for all by the year 2000. This would have been just another well-meaning pronouncement if it had not been linked to a new and comprehensive approach: expanded primary health care . Primary health care represents the first level of contact of the community with the health care system. It is provided primarily by multipurpose community health workers who must receive training in health promotion and disease prevention as well as curative medicine. It is a strategy that makes the individual, the family, and the community the basis of the health care system.
TABLE 1. Elements of primary health care (WHO Alma Ata conference, 1978)
Proper food supply and nutrition
Safe water and basic sanitation
Maternal and child health care
Prevention and control of locally endemic diseases
Appropriate treatment of common diseases
Promotion of mental health
Provision of essential drugs
Table 1 gives the elements appropriate for primary health care identified by the Alma Ata conference in the significant order in which they were listed . First on the list is education concerning prevailing health problems and the methods of identifying, preventing, and controlling them. Second comes the promotion of a proper food supply and nutrition. Next is an adequate supply of safe water and basic sanitation, and then maternal and child care, including family planning, followed by immunization against the major infectious diseases, and the prevention and control of locally endemic disease. None of these is the curative medicine for which physicians are trained. The appropriate treatment of common diseases is listed seventh; the promotion of mental health is eighth; and the provision of essential drugs appears last.
As the Director-General of WHO, Halfdan Mahler has explained: "'Health for All, implies the removal of the obstacles to health - that is to say, the elimination of malnutrition, ignorance, contaminated drinking-water, and unhygienic housing - quite as much as it does the solution of purely medical problems,, . It is on primary health care that all health programmes and the health infrastructure should be built, However, this approach entails a thorough reorientation of the existing health systems in most countries. Not surprisingly, this transformation has proved difficult and only a few countries have come close to achieving it.
Targeting mothers and children - the origin of GOBI
At a meeting at UNICEF headquarters in 1983, five years after the Alma Ata conference, individuals with extensive field experience with nutrition and public health programmes in many countries were brought together to discuss approaches to improving maternal and child health that were not sufficiently emphasized in most primary health care programmes. In this meeting, what is now know as the UNICEF-WHO child survival strategy was developed, and the mnemonic GOBI was devised. GOBI stands for growth monitoring, oral rehydration, breast-feeding, and immunization. The full significance of each term represented by the letters of the mnemonic needs to be understood by everyone responsible for its implementation .
Clearly, promotion of breast-feeding can do little for child morbidity in populations in which breast-feeding is universal. Under these circumstances, timely and appropriate complementary feeding of breast-fed infants must be promoted. Similarly, growth monitoring is a meaningless ritual unless the mother is enabled to understand the significance of growth trends as they influence the feeding of her child during the critical weaning period. It is essential that health personnel and mothers comprehend this and ensure that growth monitoring becomes a guide to appropriate remedial actions.
It must also be recognized that, even though it can save lives, oral rehydration does not prevent diarrhoeal disease. Moreover, it does little for the more frequent mild-to-moderate episodes that do not require oral rehydration for survival and that have the greatest cumulative adverse effect on nutritional status. Prevention of diarrhoea! disease requires environmental sanitation and personal hygiene, programme elements that are not captured by the mnemonic. To clarify matters further, the suffix FFF is sometimes added to GOBI to stand for feeding the preschool child, family planning, and female literacy, although these may or may not receive any emphasis, depending on local programmes . Still missing from the GOBI mnemonic is a reminder that effective health education of the mother, the community, and health-care providers is as necessary as technology to success.
The GOBI approach, dramatized as a potential revolution in child health, has been enthusiastically adopted by a large number of countries on the basis of strong UNICEF and WHO support. It has been more successful in stimulating specific preventive measures than the general promotion of primary health care. The genius of this approach to primary health care for mothers and children is the appealing message that appropriate technology is now available to bring about a revolution in child mortality in developing countries. Its major weakness is that most of the measures proposed require genuine understanding and effective application by both mothers and peripheral health personnel. Unfortunately, there is no easy way of achieving this.
The genesis of RAP
In addition to the development of GOBI there was another important outcome of the 1983 meeting at UNICEF headquarters, in which both WHO and the United Nations University participated. This was approval of the UNU-UNICEF joint research programme to use anthropologists and sociologists for the evaluation of the impact of programmes of nutrition and primary health care. The first phase of this study examined the impact of government nutrition and health-care programmes on the knowledge. and, more importantly, on the health-related behaviour of family members. The methodology was an adaptation of classical anthropological techniques of participant observation, direct observation, and guided questioning, all designed for obtaining limited, specific information. "Focus groups" in which individuals meet to discuss a specific issue were also found useful. When these technologies are limited to nutrition and health-related issues, they permit more rapid appraisal of conditions than is possible with traditional surveys.
The guidelines themselves, which represent anthropological approaches to improving the effectiveness of programmes of nutrition and primary health care, have since been extended to the evaluation of the behaviour of health providers. Rapid assessment procedures, identified by the acronym RAP, have been developed for this project and have been published in English  and will appear also in Spanish and French. Not only should the information obtained from these investigations contribute to improving the effectiveness of primary health care and GOBI programmes, but the techniques themselves can be used for continuing programme evaluation.
Current problems with programmes of nutrition and primary health care
This multi-country RAP research has revealed a series of common problems that in most countries have prevented the primary health care programmes from achieving the effectiveness anticipated. They include the lack of proper training and motivation of health personnel, the lack of interest by recipients in preventive programmes, the failure of health care providers to consider the beliefs and attitudes of their patients, and the lack of understanding of the financial and time constraints of recipients.
Health personnel are not properly trained
With rare exceptions, physicians are generally trained and interested only in curative medicine. The result is that, when they are assigned to peripheral health services, they soon become immersed in curative tasks to the neglect of the Alma Ata concept of primary health care. Unfortunately, the rest of the health personnel follow their example. Although they are responsible for providing individuals and the community with the knowledge to make their own health decisions, this task is either totally neglected or done so badly as to be useless.
Thus, not only the physicians but also the nurses and auxiliary nurses in the public health systems of most developing countries spend most of their time and effort dealing superficially and generally ineffectually with complaints of illness. The second problem can only be fully appreciated from actual field experience with the promotion of preventive medicine at the primary level. This is the fact that individuals, unless well-informed, come to the health worker only when they are sick and want to be cured. They are not at all interested in the kind of preventive measures called for in the Alma Ada declaration.
I have been in country after country in which there are antenatal clinics to which mothers do not come, well-baby clinics to which babies are not brought, and immunizations that are offered and not accepted. In order to implement the nutrition and preventive medicine components of primary health care, the resistance of both providers and recipients must be overcome .
The first step in applying the principles of Alma Ada is a reorientation of the training of health workers toward preventive roles that include communication of health promotion concepts to people and gaining the confidence of communities. The public health nutritionist must provide the appropriate training and supervision for the nutrition component of this task.
In Guatemala an ambitious system of rural health posts staffed by doctors has proved too costly to maintain when health expenditures were cut 60% in the past decade as the result of economic adjustment policies. As I observed the functioning of the system, however, this will make less difference than conventionally assumed, The doctors manning these posts were mainly recent medical graduates forced into a period of rural service without adequate training or logistic support. Most were interested in obtaining money from private practice on the side and returning to a major city as soon as permitted. They were frequently absent from their posts and rarely did more than respond superficially to patient complaints. No time was devoted to preventing disease. As inadequate as their behaviour may have been, they were only reflecting their training and the priorities inherent in the system. Moreover, the demand from individuals and the community was for medical treatment, with no felt need for preventive health measures.
Twenty-five years ago an American physician, Dr. Caroll Behrhorst, came to Guatemala to open a clinic in the highland area of Chimaltenango . It soon became apparent to him that the sophisticated training and skill of a physician are neither affordable nor essential for primary health care. (In fact, from my own experience, they are virtually useless .) He began recruiting and training assistants from local communities and gradually incorporated preventive services, including the encouragement of breastfeeding and proper supplementary feeding, family planning, tuberculosis control, and sanitation. The one or two physicians assisting him were used as organizers and teachers, not as primary health care providers.
Because of its assistance to the indigenous population, the programme has encountered suspicion and hostility from recent governments in Guatemala, and one of its physicians and several of its village workers have been killed or disappeared. Nevertheless, more than 100 village workers have been trained, and most are still working successfully in the area. The village programme is self-sustaining. It covers only a small proportion of the Mayan Indian population, but it does demonstrate what is possible.
A somewhat similar programme with minute monthly payments by villagers was established in Solo, and later in other areas of Indonesia by a dedicated physician, Gunawan Nugroho, after participating in the INCAP summer field course in clinical and public-health nutrition and observing the Behrhorst programme. The current highly acclaimed national nutrition and primary health care programme in Indonesia benefited from this demonstration.
The Indonesia programme is also similar in principle to the outstanding national primary health care programme in Thailand initiated by Amorn Nondasuta as Director of Public Health, which is serving as a model for other developing countries. The original "barefoot doctor,, programme of mainland China was established on similar principles and dramatically improved China's health statistics.
Constraints and beliefs of the recipients are ignored
Health workers cannot hope to influence health practices without understanding both the constraints on their clients and their beliefs . Social scientists observing them are often shocked by the extent to which the economic and time constraints of the people are not understood or are ignored. All too often, health personnel assume that the greater their need, the more time poor people will have available to wait in clinics, bring children to health centres, make repeated visits, or attend lectures and demonstrations.
Health programmes are often ineffective because the populations they are supposed to serve cannot afford the economic losses and additional cost of using them or feel unable to neglect other crushing family responsibilities to seek the services, For example, a woman on a Guatemalan coastal plantation begins her day at five o'clock in the morning by taking lime soaked corn to the village mill to be ground and bringing it back home to prepare tortillas for breakfast and lunch. She must then get the children off to school, wash clothes and diapers in a nearby stream and leave them to dry, and then go to the field to pick coffee, taking her preschool children with her.
After carrying her heavy sack of coffee to be weighed, she may return to feed the children, chickens, and a pig; clean the house and wash dishes; take corn to be ground for supper; boil the next day's supply in lime-water; and make several trips for water carrying a heavy jar on her head. She has also been breast-feeding her infant on demand and changing innumerable diapers (especially when the child has diarrhoea), and doing other necessary chores, possibly including some work in her home garden.
If she breaks this unending routine to go to the health centre an hour away by bus, she will not only lose a day's wages but spend an equal amount on bus fare. At the clinic she may have to wait for hours or may not be seen at all. Even if she is seen at the clinic, it is likely that they will be out of medicine and will give her a prescription which she has no money to fill. Moreover, the hasty diagnosis and treatment she receives may be of little value. She may take a desperately sick child to the clinic, but she is not likely to go back for purely preventive services on still another day.
Lack of knowledge and consideration of the beliefs and constraints of health-care recipients engenders much misunderstanding and distrust. All populations have their own system of indigenous beliefs and practices pertaining to illness, and most make use of a variety of indigenous practitioners, often using the government health system as only one resort. Cosminsky and Scrimshaw  have described the way in which a mother with a sick child on a Guatemalan coastal plantation may try a local healer (curandero), or a spiritualist, or an injectionist, or ask for medicine at the pharmacy and usually will take the child to the government clinic only if the child fails to improve.
Such medical pluralism has also been described by anthropologists Pamela Hunte and Farhat Sultana for Baluchistan, Pakistan , and David Nyamwaya for northern Kenya , to mention only two more examples. In fact, medical pluralism is characteristic of developing-country populations, who often go back and forth between traditional and Western medical practitioners. Primary health care personnel must understand this behaviour and compete by showing the same concern for the feelings of their clients as the successful traditional practitioners. Health workers also need to understand and to respect folk medical beliefs and treatments and to recognize the gap between Western and folk medicine that primary health care must bridge. This obligation should be included at appropriate levels in the training of all health workers [17, 18, 19].
Evaluation as a guide to programme improvement
One of the most important ways of judging the effectiveness of programmes and improving them is formal evaluation. However, conventional nutrition and health surveys are expensive and time-consuming and collect large amounts of data requiring lengthy analysis. Furthermore, they are often not acceptable to the responsible health authorities or to the population to be studied because their scope and invasiveness are so threatening. The RAP methodology is, in contrast, able to provide information in a few weeks for a few thousand dollars without necessarily involving the health authorities at all.
Conventional nutrition surveys may reveal dietary habits and nutrition and health status, but they usually do not indicate whether or not a programme is having any significant effect, unless there have been good baseline studies or an appropriate control population. The RAP approach does not attempt to determine nutrition or health status. However, at a comparatively low cost and in a short period of time, it can provide information on the influence a programme is having, if any, on the knowledge and behaviour of the intended beneficiaries. Even more important, it can tell a great deal about why a programme is succeeding or failing. This information can serve as an immediate guide to programme improvement. Moreover, the RAP approach can be conducted at the local level without elaborate top-down planning.
In the UNU/UNICEF-sponsored RAP evaluations of primary health care in sixteen developing countries, it was apparent that the factors that made for relative success or failure were remarkably similar in all these different countries and cultures. Among the major problems most often identified by health-care recipients as reasons for not going to the health centre were the lack of adequate supplies and medicine, inconvenient scheduling of services, and abusive, rude, arrogant, or inconsiderate treatment by health providers. It was apparent that routines were most often established for the benefit of the physician and health personnel. The patients were expected to be able to leave their daily responsibilities at the convenience of the health centre, and no effort was made to schedule different services in the same visit.
Mothers often were asked to come on different days for the well-baby clinic and for immunizations, with long waits each time. Separate days for sick children and well children may prevent the mother with more than one preschool child from bringing them both on the same day. Some patients felt that no one listened adequately to their complaints or examined them properly. A review by Messer cites numerous additional examples . Such flawed health services, with few or no preventive services, are a poor use of a nation's health resources. The RAP methodology offers a practical and affordable means of determining the effectiveness of existing primary health care programmes and how they can be improved.
It had been assumed that the RAP methodology required social scientists with training and experience in anthropological methodology, but it turns out that this is not necessarily the case. Peripheral health workers in Panama acquired a new sense of understanding and control of programme effectiveness when they used the RAP methodology, and it is already being utilized for the evaluation of specific health activities, such as those for the control of epilepsy. It is now clear that nutritionists who are already experienced in field work can take advantage of the RAP methodology for the qualitative evaluation of nutrition intervention programmes.
There will always be a need for nutrition surveys to determine nutritional status, dietary habits, and nutrient intakes and to measure directly the biological impact of nutrition programmes, but nutritionists will benefit from adding the RAP methodology to their skills.
The role of nutrition in primary health care
Finally, it should be mentioned again that there is no single element of primary health care that is as important as nutrition, because it is such a critical factor in the high morbidity and mortality of the underprivileged of developing countries. In their studies of the causes of child mortality in Latin America, Puffer and Serrano found that nutrition was responsible for 57% of all deaths of children under five years of age as either a primary or underlying cause .
The health of an infant depends to a great extent on the adequacy of the mother's nutrition during pregnancy, on the initial breast-feeding of the infant, and on timely and appropriate complementary feeding. The burden of diarrhoea! and respiratory diseases will be lessened if good nutrition is maintained. Although it will not prevent the common communicable diseases of childhood, good nutrition will practically eliminate mortality from them.
The most effective way of determining the adequacy of a child's nutrition is to monitor growth. However, as already emphasized, growth monitoring is useful only if it enables the mother to know when her child is failing to gain weight and what is the proper action to take. She must then be motivated to improve the child's food intake and, if the trend is not reversed, to bring the child to the health centre. The rapport necessary to communicate this information and to motivate the mother is also important for the promotion of family planning and prenatal care.
It is highly appropriate for this lecture to be given before the Society for Nutrition Education, because nutrition education of the community, particularly the mother, is essential to the success of the nutrition and primary health care strategy. The Alma Ata conference considered community and individual involvement to be fundamental to programmes of nutrition and primary health care. For people to be effectively engaged in caring for their own nutritional and health status, they must have the opportunity to understand what leads to health and what endangers it.
While this understanding requires education, conventional nutrition and health education in health programmes has generally been a failure. Lecturing at people without interacting with them or understanding their doubts and the problems of implementation does not work. The same is true of educational materials that fail to take into account beliefs and constraints. Health educators need to use the findings of medical and nutritional anthropologists. They also need to adopt some of the methods of anthropology in order better to interact with individual households.
It is not fair to blame the lack of appreciation of the importance of preventing disease on the victims. Preventive measures introduced by sympathetic and well-oriented health personnel who have the patience and motivation to explain and promote them are generally accepted. The problem, therefore, lies not so much with the lack of formal education and the perversity of the underprivileged, but with failures of current nutrition- and health-education practices. This is the challenge.
Prospects for the future
I would like to end on an encouraging note, because real progress is being made. Most countries in the world have signed the WHO resolution on "Health for all by the year 2000,,, although not many have yet demonstrated the political will to provide the necessary and appropriate health services. However, some elements of the UNICEF/WHO child-survival strategy, particularly immunization and oral rehydration, are being widely adopted. Seventy-seven developing countries have declared their intention to immunize the majority of their children UNICEF estimates that one million children's lives are being saved annually by immunization alone and that, without the introduction of oral rehydration in 1984 , three million more children would have died. Growth monitoring is more difficult to introduce effectively, but here too, good progress is being made in an increasing number of countries.
The rapid improvement in nutrition and health achieved by two small Latin American countries with very different political systems but a common decision to shift resources to education and health, Costa Rica and Cuba, is particularly encouraging . In both countries there was a rapid fall in infant and preschool mortality once the decision was put into effect. Although both are poor countries, their current mortality and life-expectancy figures are closer to those of Europe and the United States than to other developing countries of the region except Chile [24, 23]. Chile has a long tradition of effective health care and supplementary-feeding programmes, which no recent government has dared to reverse. These three countries in the Western hemisphere and others in Asia are striking proof that political priorities and decisions determine health statistics, not gross national product .
There is, thus, ample additional evidence that developing countries that are strongly committed to promoting health as an integral part of social equity and provide for community participation in preventive services can achieve rapid improvement in infant, preschool, and overall mortality rates. Unfortunately, the continuing high death rates in most developing countries bear evidence to the sad fact that the health services are ineffective for the great majority. Nevertheless, more and more countries are implementing child-survival strategies. Moreover, the concepts of Alma Ata are making their way into the training of developing-country physicians and other health personnel, often in programmes identified as community medicine, rural medicine, or family medicine.
Wherever a new generation of health workers has been oriented toward preventive as well as curative medicine and this approach is supported by strong national health policies, as in the examples cited, rapid declines in morbidity and mortality have followed. Nutritionists must not allow themselves to be left out of this health revolution.
Summary and conclusions
The separation of nutrition programmes from other health activities is no longer acceptable. Since the WHO Alma Ata Conference on Primary Health Care in 1978, WHO and UNICEF have led a global effort to make nutrition a major component of primary health care. However, most nutritionists, whether medical or non-medical, have not been trained in this concept. Neither have most physicians and other workers who are responsible for health promotion and health maintenance at the community level. Yet community understanding, motivation, and involvement are each essential for the success of primary health care. As an approach to health for all it is neither an empty slogan nor an unrealistic ideal, but it does face many obstacles. The most important of these are the lack of a strong national commitment to health equity and the lack of training and commitment of health workers in the primary health care approach. As public health nutritionists, we must adjust our thinking and actions and the training of future nutrition and health workers to view nutrition education as an interactive community process and to recognize that the prevention of malnutrition must involve interaction with other public health disciplines as well.
The substantive and editorial contributions of Karen Mitzner and Mary Scrimshaw are gratefully acknowledged .