Adapting the Tranng and Visit System Richard Heaver SWP662 WORLD BANK STAFF WORKING PAPERS Number 662 WORLD BANK STAFF WORKING PAPERS Number 662 Adapting the Training and Visit System for Family Planning, Health, and Nutrition Programs Richard Heaver INTERNATIONAL M:ONETARY FUIZD JOIltT LiU2MIY SEP 11'- VE:F.Vs'Xl}XnNn r011.1, o The World Bank Washington, D.C., U.S.A. Copyright (C) 1984 The Intemational Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing August 1984 This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The views and interpretations in this document are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting on their behalf. Any maps used have been prepared solely for the convenience of the readers; the denominations used and the boundaries shown do not imply, on the part of the World Bank and its affiliates, any judgment on the legal status of any territory or any endorsement or acceptance of such boundaries. The full range of World Bank publications, both free and for sale, is described in the Catalog of Publications; the continuing research program is outlined in Abstracts of Current Studies. Both booklets are updated annually; the most recent edition of each is available without charge from the Publications Sales Unit, Department T, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from the European Office of the Bank, 66 avenue d'1ena, 75116 Paris, France. Richard Heaver is an economist in the Population, Health, and Nutrition Department of the World Bank. Library of Congress Cataloging in Publication Data Heaver, Richard, 1952- Adapting the training and visit system for family planning, health, and nutrition programs. (World Bank staff working papers ; no. 662) 1. Rural health services. 2. Health education. 3. Agricultural extension work--Methodology. 4. Rural health services--India--Case studies. 5. Rural health services--Indonesia--Case studies. 6. Rural health services--Philippines--Case studies. 7. Health education --India--Case studies. 8. Health education--Indonesia-- Case studies. 9. Health education--Philippines--Case studies. 10. Birth control--India--Case studies. 11. Birth control--Indonesia--Case studies. 12. Birth control--Philippines--Case studies. 13. Nutrition extension work--India-- Case studies. 14. Nutrition extension work--Indonesia--Case studies. 15. Nutrition extension work--Philippines--Case studies. I. Title. II. Series. RA771.H43 1984 362.1'0425 84-15359 ISBN 0-8213-0393-7 ABSTRACT In agriculture, the Training and Visit System (T&V) of extension has proved a successful managerial approach for dealing with geographically scattered outreach operations; developing and adapting techniques for locally varying conditions; and changing the behavioral patterns of large numbers of poorly-educated field staff and beneficiaries. Mass family planning and health programs present all of these challenges. This paper is a first attempt to explore whether T&V can be adapted to meet the technical, managerial and behavioral needs of population, health and nutrition (PHN) programs, while remaining cheap enough to be widely replicable. The paper falls into three parts. The first part on theory looks at the similarities and differences between managerial tasks and practices in agricultural extension and PHN, and taking into account the differences, suggests a "model" T&V system for PHN. The system that is,put forward is not a blueprint, but a starting point for further adaptations appropriate to the resources and PHN delivery systems of any given country. The second part of the paper moves from theory to practice; it looks at three PHN outreach systems in India, the Philippines and Indonesia. The Indian program already has many of the managerial characteristics of T&V. But while it appears to be more effective than the programs in Indonesia and the Philippines, it is also a more intensive user of human and financial resources. So, while the case studies support the potential usefulness of T&V-type management systems, they also raise the question whether these are necessarily resource-intensive, or whether resources can be diluted without undue loss of effectiveness--a question which can only be settled by experiment with alternative approaches. The third part of the paper therefore proposes some alternative approaches to making outreach programs more cost effective, and outlines some of their theoretical advantages and disadvantages. For PHN specialists unfamiliar with agricultural T&V, a brief outline of the system is given in Annex 1. EXTRACTO En el sector agricola, el sistema de extensi6n mediante capacitaci6n y visitas ha resultado un enfoque administrativo eficaz para realizar las operaciones de extensi6n en zonas ampliamente dispersas, elaborar t6cnicas que se adapten a las diversas condiciones locales y modificar las caracteristicas del comportamiento de grandes cantidades de personal local de extensi6n y de beneficiarios con muy bajo nivel de educaci6n. Los programas de planificaci6n de la familia y de salud en gran escala plantean todas estas dificultades. El presente estudio constituye el primer intento de explorar si el sistema de capacitaci6n y visitas se puede adaptar en la forma necesaria para satisfacer las necesidades t6cnicas, administrativas y en materia de comportamiento de los programas de poblaci6n, salud y nutrici6n, manteni6ndolo al mismo tiempo lo suficientemente barato como para repetirlo ampliamente. El estudio se divide en tres partes. En la primera, de carActer te6rico, se examinan las similitudes y diferencias entre las tareas y practicas administrativas de extensi6n agricola y las de los programas de poblaci6n, salud y nutrici6n, y, teniendo en cuenta las diferencias, se sugiere un "modelo" de sistema de capacitaci6n y visitas para estos uiltimos. El modelo que se propone no es un plan detallado, sino un punto de partida para incorporar luego las adaptaciones que se ajusten a los recursos disponibles y a los sistemas de prestaci6n de servicios de planificaci6n familiar, salud y nutrici6n de un pais dado. En la segunda parte del estudio se pasa de la teoria a la practica; se examinan los sistemas de extensi6n en las esferas de poblaci6n, salud y nutrici6n existentes en la India, Filipinas e Indonesia. El programa de la India ya tiene muchas de las caracteristicas administrativas del sistema de capacitaci6n y visitas. Empero, si bien parece ser mAs eficaz que los programas de Indonesia y Filipinas, tambien exige un uso mas intensivo de recursos humanos y financieros. Por lo tanto, aunque los estudios de casos practicos corroboran la posible utilidad de los sistemas administrativos basados en capacitaci6n y visitas, tambien hacen que se plantee la cuesti6n de si 6stos requieren forzosamente un uso intensivo de recursos, o bien si los recursos se pueden diluir sin una merma indebida de la eficacia, una cuesti6n que s6lo puede dilucidarse mediante el ensayo de los diferentes sistemas. Por consiguiente, en la tercera parte del estudio se proponen otros posibles m6todos para hacer que los programas de extensi6n sean mAs eficaces en funci6n de los costos y se esbozan algunas de las ventajas y desventajas te6ricas. Para informaci6n de los especialistas en programas de poblaci6n, salud y nutrici6n que no esten familiarizados con el sistema de capacitaci6n y visitas en el sector agricola, se ha incluido una breve reseiia del mismo en el Anexo 1. Le systeme de formation et de visites dans le cadre de la vulga- risation agricole a fait ses preuves en tant que m6thode de gestion appli- cable a l'execution de programmes ext6rieurs geographiquement eparpill6s, a l'adaptation de techniques aux conditions locales et a leur diffusion et a la modification du comportement d'un grand nombre d'agents d'execution et de b6neficiaires peu instruits. Les grands programmes de planning familial et de sante posent exactement les memes problemes. Ce document tente pour la premiere fois de determiner dans quelle mesure le systeme de formation et de visites, tout en restant assez bon marche pour etre repro- duit a grande 6chelle, peut etre adapt6 aux besoins des programmes de population, de sante et de nutrition sur les plans de la technique, de la gestion et du comportement. Le document comprend trois parties. La premiere, de caractere theorique, examine les ressemblances et les differences existant entre la vulgarisation agricole et les programmes de population, de sant6 et de nutrition pour ce qui est des taches et des pratiques de gestion. En se fondant sur les differences, elle propose pour ces programmes un systeme "modele" de formation et de visites. Celui-ci doit etre consider6 non pas comme un cadre definitif, mais comme un point de depart pouvant servir a la mise au point de formules adaptees aux ressources des services de popu- lation, de sante et de nutrition de chaque pays. La deuxieme partie du document passe de la th6orie a la pratique en examinant trois programmes exterieurs de population, de sante et de nutrition (en Inde, aux Philippines et en Indonesie). Le programme indien pr6sente deja plusieurs des caracteristiques gestionnelles du systeme de formation et de visites. I1 apparait plus efficace que les programmes indonesien et philippin, mais il met aussi davantage a contribution les ressources humaines et financieres. Ainsi, tout en confirmant l'utilite potentielle des systemes de gestion de type "formation et visites", ces etudes de cas posent la question de savoir si ces systemes consomment necessairement beaucoup de ressources, ou si l'on peut diluer celles-ci sans trop reduire l'efficacite, question a laquelle on ne pourra repondre qu'apres avoir experimente diverses autres methodes. La troisieme partie du document propose donc d'autres facons de rendre les programmes exte- rieurs plus rentables et presente dans leurs grandes lignes certains de leurs avantages et de leurs inconvenients th6oriques. Le systeme de formation et de visites dans le cadre de la vul- garisation agricole est recapitule brievement a l'Annexe 1 a l'intention des specialistes des programmes de population, de sante et de nutrition auxquels il n'est pas familier. ACKNOWLEDGEMENTS The case studies in this paper would not have been possible without the kind assistance of the following in arranging field visits to the project and program areas: Tmt. Susan Mathew, IAS, Coordinator of the Tamil Nadu Nutrition Project, and the staff of agencies implementing the project; Dr. Abinugroho, Indonesia Nutrition Project Coordinator, and the provincial staff of the UPGK and BKKBN in East and Central Java; Dr. Flora Bayan, National Coordinator of Primary Health Care Programs in the Philippines, and the provincial health and development staff in Cebu and Ilo-Ilo. I am also grateful to the following for their comments on the draft: Stephen Denning, Jim Greene, T. J. Ho, Huw Jones and Bernard Liese of the Bank's Population, Health and Nutrition Department, and K. V. Ranganathan of the Economic Development Institute; and in particular to Arturo Israel of the Projects Policy Department, for commissioning and guiding the paper. None of the above bear responsibility for the ideas and opinions in the paper, nor for the impressions of the projects and programs visited. SUMMARY AND CONCLUSION I. Theory Training and Visit (T&V) has been adopted as the extension method in more than forty agricultural projects in about twenty countries, and has influenced many others. It has proved a successful approach for introducing behavioral change on a mass scale in situations where field staff and clients are poorly educated and geographically scattered, and where techniques must be developed and constantly readapted to meet locally varying conditions. The challenges facing mass population, health and nutrition (PHN) programs are similar---though more intense, since innovations in PHN affect the personal rather than the economic life of the family, so that adoption is more threatening to cultural norms; and since the beneficial impact of better practices (especially preventive ones) shows up less clearly in PHN than in agriculture, making adoption and diffusion less automatic. The paper suggests adaptations to the T&V system that would be needed if it were to be applied in the context of PHN programs. The six main elements of a T&V system for PHN are summarized below. Where there are significant differences from T&V in agriculture, these are noted. 1. Concentration on a small number of key tasks. The tasks chosen would a) be few and simple enough for poorly educated village health workers (VHWs) to understand and remember, and b) blend tasks which are of high priority in epidemiological terms with tasks which are of high priority in terms of clients' felt needs. While task concentration is recognized as a need in the PHN literature, it is seldom practiced; VHWs are overloaded and must often attempt to communicate. innovations which clients have little incentive to adopt. 2. A performance reporting system concentrating on key tasks. Reporting systems in current use take up too much of VHWs' time and collect data which are often not processed and used. The primary purpose of a T&V reporting system would be to motivate both VHWs and clients by demonstrating progress. This would be a departure from T&V practice in agriculture, where written reporting has been unnecessary, since the results of adoption are plainly visible in the fields. 3. Regular, frequent home visits. Since staff are scarcer and messages change less frequently than in agriculture, visits might be monthly rather than fortnightly. Predictable routines for home visits are followed in few PHN outreach systems, but have great potential both for building the confidence of clients and for increasing the accountability of VHWs to supervisors. 4. Concentration on selected clients where interventions can have maximum impact on reducing fertility and mortality. Scarcity of field staff and the need for regular follow-up visits to build confidence before adoption and to reassure and reinforce after adoption, argue for concentration of care on pregnant women and mothers with small children, since it is in these groups that mortality is highest and the need for family planning education greatest. This is, of course, a different client group from T&V in agriculture. 5. Regular, frequent field supervision visits focused on support rather than inspection. Regular supervision is missing in most PHN outreach programs. The system would ensure that a supervisor would spend not less than a day each two weeks supporting each VHW individually on the job. 6. Regular in-service training. Monthly training sessions for VHWs, supported by two monthly workshops in which senior staff and researchers would develop new extension recommendations, would have several functions---disseminating new technical messages and IEC techniques; reinforcing existing knowledge; giving VHWs feedback about performance; and sharing and solving problems encountered in the field. Such training is not available in most PHN programs. II. Practice Field visits were made to health and nutrition outreach programs in India, Indonesia and the Philippines. The India program, which had many of the managerial characteristics of T&V, clearly outperformed the others, which relied on volunteers with limited training and infrequent supervision. At the same time, both the Indian program and the model T&V system for PHN put forward in this paper, are more intensive users of financial and human resources than the Indonesian or Philippine programs. The paper does not recommend programs based on paid workers rather than those based on volunteers; what is appropriate will depend on the cultural and administrative context, and the resources available. However, it does note that effective supervision of large numbers of volunteers can be prohibitively expensive; that tightly managed T&V-type programs may be considerably cheaper than well supported programs based on large numbers of volunteers; and that T&V-type programs, while more costly than poorly supported volunteer-based programs, may nevertheless also be much more cost-effective. III. Implications None of the elements of the 'model' T&V system outlined in the paper is new to PHN. However, these elements are seldom introduced as a package or system so that they are mutually reinforcing. This has been the key to T&V's success in agriculture, and would be worth systematic experiment in PHN. In some countries, for example in South Asia, the field staff resources are already available for such an experiment to take place. Other countries cannot afford the large network of full time extension agents on which T&V has traditionally relied in agriculture. But in these countries there may be scope for cheapening the design of the T&V system through employing paraprofessionals as VHWs, and through limiting the task by restricting the full range of services to fewer priority clients in the first years of development, while still retaining the basic managerial principles of the system. CONTENTS Part One: Theory page 1. The Market ........................................... . 1 2. Extension Tasks ........................................... . 2 3. Developing Extension Messages ...... .................. 7 4. Visiting ............................................. 8 5. Training and Supervision . . . 12 6. Summary and Issues for Discussion . . . 14 Part Two: Practice 7. Comparing Performance . . . 18 8. Tamil Nadu, India . . ......................................... 20 9. Cebu and Iloilo, Philippines . . . 26 10. East and Central Java, Indonesia . . .......................... 29 11. Comparing Human Resource Use . . . 35 Part Three: Implications 12. Alternative Approaches to Making Outreach Programs More Cost-Effective ......... ....... 40 Annexes 1 The Training and Visit System of Agricultural Extension ..... 47 2 Message Selection Scale ..................................... 49 3 ORT Behavioral Profile ....... # . ............................. 51 4 Sample Questions for Determining whether an Outreach System has the Prerequisites of Effectiveness .................... 54 PART ONE 1. THE MARKET The market for extension advice in agriculture has by definition been farmers. In practice, it has consisted mostly of adult males who are not in the poorest segment of the population, since the very poor are often not farmers, but landless laborers or providers of marginal services in the informal sector. There is some tendency for T&V contact farmers to be larger than average--a bias which, in agriculture, may be defensible in policy terms. Governments are concerned with production as well as equity, and concentration on larger farmers is likely to yield more output per extension worker because one adoption decision by a large farmer leads to a greater increase in the marketed surplus than the same decision by a smaller farmer. In the PHN sectors, by contrast, policies usually stress the provision of service where the human need is greatest. Health problems are concentrated around pregnant women, babies and small children, so that the major client group is not menl/ but mothers and mothers-to-be. Health problems are also concentrated among the poor, who often cannot afford enough of the right types of food, and whose malnourishment increases susceptibility to disease. The market for government PHN services is characterized by relative lack of demand, especially where the emphasis is on education and prevention rather than cure. Lack of interest in government PHN services often coexists with expenditure of relatively large sums (in cash or kind) for care from non-government sources. However, lack of effective demand for government services can mask significant potential demand in several ways. First, there can be a desire for better PHN care, but insufficient awareness of what services are provided. Second, there can be both desire and awareness, but a client may not have the money to pay for care, or the time to visit a clinic. Third, even where time and money allow access to government PHN services, the services available may not answer felt needs. In each case, demand for care is not lacking, but latent. Instead of an absolute lack of demand, there is a lack of demand for services at the place, time or price currently provided, or in the form currently presented. Put in this way, the "demand problem" in PHN is not unlike that in agriculture where extension services are ineffective. Agricultural advice is seldom in demand where it is dispensed inconveniently in an office, rather than conveniently in the fields; or where recommendations do not make sense in terms of farmers' priorities and perceptions. These are two demand problems that T&V has been able to solve in agriculture. While a latent demand for the right type of PHN service can be assumed, making this demand effective (i.e. presenting PHN-related behavioral changes as worthwhile enough to adopt) is likely to be much harder than in agriculture, for three reasons. First, resistance to advice will be greater, since behavioral change in PHN affects the personal rather than the economic life of the family, and is therefore more threatening to 1/ Although in many cultures womens' attitudes to innovation are strongly influenced by men. - 2 - cultural norms. Second, while effective extension advice must be tuned to clients' perceptions and behavior, clients are less well understood in PHN than in agriculture. Women and the very poor have until recently been at the periphery of development thinking, which long concentrated on men, and on male farmers. Field staff incentives often reinforce this bias. Male beneficiaries are visited because they are better educated and have higher status than women, while the poor, who are uninfluential and often less visible because they live away from village centers and tarred roads, have often been ignored altogether. Third, and most important, the link between better PHN practices and better health is much less obvious than that between better farming practices and higher yields. Extension advice in agriculture (e.g. for fertilizer use) tends to produce clear and positive results visible not only to the contact farmer but to those in neighboring fields. By contrast, extension advice in PHN is typically preventive, and leads to a non-occurence not obviously linked to the advice. Since no one can predict whether the condition which the advice was designed to prevent would actually have occured for any given client, the impact of better practices shows up less clearly and tangibly for the client than it does in agriculture. To the extent that benefits are less obvious, both adoption and diffusion are more difficult to achieve. Each of these factors makes the job of creating market demand in PHN more difficult than in agriculture, and makes PHN more management-intensive. It requires more time and skill devoted to understanding and gaining the trust of the client; more time and skill put into presenting new technology persuasively; and more time and skill spent tracing and communicating results. These needs have important implications for the choice of extension tasks, and for managing the development and delivery of extension messages. 2. EXTENSION TASKS Discussion in the PHN literature has paralleled the T&V system's approach to task analysis in most respects. It is widely accepted that mass PHN care needs to concentrate on a few interventions which are simple and beneficial from a client as well as a professional perspective. Some researchers in health care have in fact gone well beyond current practice in T&V in terms of developing systematic criteria for extension message selection--see for example, Annex 2's "Message Selection Scale," developed for AID health projects in Honduras and The Gambia. However, practice lags far behind theory. Despite the rhetoric stressing simplicity, there is no clear consensus about what a primary PHN service should supply. It is common for village health workers (VHWs) to be burdened with more tasks than they can understand and remember, let alone perform; and it is common for VHWs to be asked to promote behavioral change which clients have no incentive to adopt. A T&V system in PHN must therefore be built up from the bottom based on the sorts of tasks that it makes sense to ask a poorly educated VHW to perform. - 3 - 2.1 Types of care: curative, preventive, responsive. T&V stresses the need for initial extension messages to be perceived as clearly profitable, in order to gain the confidence of clients and the credibility of the service. Substantial evidence from the field indicates that in health care, basic curative services must be provided if VHWs are to be seen as credible. This is because clients felt needs are for curative services, where treatment immediately and perceptibly alleviates an immediate and perceptible problem. Curative care must therefore be made available side by side with preventive care, if clients are to build the confidence necessary to adopt preventive measures with less obvious benefits. The need for basic curative services has two implications. First, VHWs must be involved in the distribution of drugs, and, in the increasing number of countries where user charges are levied, the handling of money. This is a managerial complexity not present in agriculture, where T&V extension staff give advice only and are not responsible for input delivery. Second, the demand for curative care raises an incentive problem which must be anticipated in the design of a T&V system for PHN. VHWs will rationally concentrate on the curative services which clients want, and neglect preventive measures, unless the management system builds in some counter-incentive. Both factors reinforce the case for giving VHWs only a few simple curative tasks, so that preoccupation with drugs and curative care does not crowd out preventive care. It also has implications for training, supervision and performance measurement, discussed below. Providing curative care is only one example of responding to clients with services that they want--critical in building the confidence in the extension service which is needed if clients are to adopt the preventive innovations which professionals want for them. While limiting the set of core tasks for VHWs is important (see 2.2), this core must not be rigidly defined, but responsive to local felt needs. For example, the CIMDER program in Colombia was able to respond to villagers priority felt need for clean water supplies by designing a chlorinator for village wells made from locally obtainable plastic bottles pierced with holes in the right size and location to release regular amounts of chemical into the water. Building this sort of responsiveness into the extension system has managerial implications for research, training and supervision, also discussed below. 2.2 Grading and phasing PHN tasks. PHN interventions can be roughly graded into three groups according to the degree to which they answer felt needs and promise results. The following might be a realistic set of tasks for VHWs with limited training: (i) Basic curative care. Curative tasks for VHWs should be strictly limited to first aid and to a very small number (3-4) of prevalent diseases whose symptoms are easily recognizable and which common generic drugs can cure. VHWs should carry a very limited number of drugs. VHWs should also be trained to recognize or suspect the presence of 2 or 3 common diseases which need referral to higher levels of the health service. From a confidence-building perspective, VHWs must know when to refer more complex cases or they will themselves treat them inappropriately and visibly fail. - 4 - (ii) Intermediate curative and preventive care. Three examples of interventions of medium complexity and with high returns are nutrition advice, immunization against neonatal tetanus, and oral rehydration therapy. Nutrition advice, which can be either preventive or curative, has the advantage that significant gains can usually be made with the village's or family's own resources (promoting breast-feeding or nutritious but neglected local foods) or cheap external resources (vitamin A against blindness, for example). A major problem with nutrition education is that malnourishment is not always visible to mothers. Substantial field experience suggests that mothers cannot be expected to adopt nutrition improvements which may go against long-established cultural norms, and may add to the time and money spent on food preparation, unless the benefits are clearly apparent. Nutrition education without a measurement system which clearly shows mothers they are making progress is therefore likely to be ineffective (see 2.3 below). While immunization against neo-natal tetanus is a preventive measure, a number of factors suggest relative ease of acceptance. First, the infection normally leads to rapid death; mothers will be relatively highly motivated to prevent it. Second, the target group (pregnant women) is clearly defined and limited in number: this makes for managerial simplicity. Third, the task need be performed only once or twice per pregnancy, and then by the VHW not the beneficiary, so that the continuous innovative behavior demanded by a nutrition program is not needed. Fourth, immunization and result (child survival) are close in time. Neo-natal tetanus immunization is therefore more likely to be perceived as beneficial than other immunizations--such as measles--which only prove their worth over longer periods. Oral rehydration therapy may be a less straightforward technology than it appears (see 3.1 below). While it is, in fact, curative, it may not be perceived as curative in societies where mothers do not see diarrhea as an illness. However, the relative simplicity of the remedy, the possibilities for using cheap locally available resources, the immediacy of the results and sharp reduction in mortality associated with correct use, make ORT a priority for health extension. (iii) Complex preventive care. There is a case for introducing preventive services such as hygiene education only after widespread adoption of basic curative and intermediate services. Improvements in hygiene are complex to manage, since the behavioral change required is not only alien to tradition, but needs to be continuous and consistent. While in nutrition, one lapse in preparation of a balanced meal will not affect a child's health, in hygiene one failure to boil water or to wash hands may result in infection. Commitment to continuing and consistent behavioral change is unlikely where the link between cause and effect (germ and infection) is invisible and hard to comprehend--unlike the link between more and better food and child growth. For practical purposes, hygiene education may well be unmanageable and ineffective for the first few years of a primary health care campaign. The Alma Ata Conference's definition of - 5 - primary health care, which includes safe water and waste facilities (useless without better hygiene education), is over-ambitious for many developing countries. Giving family planning advice and supplying contraceptives is a logical task for VHWs. Few countries are likely to be able to afford two parallel services--for health care and for family planning--working in the same village: this alone makes a strong case for integration. But in countries where there is little demand for contraception, VHW credibility may be reduced if family planning for purposes of fertility reduction is introduced as a key message early in the relationship with a client. From a behavioral standpoint, family planning advice is more likely to be accepted when clients have built up confidence in VHWs as a result of their effective health care work--a point when, with existing children healthier and more likely to survive, mothers will have an incentive to limit the size of their families. In some countries, therefore, family planning advice should be confined in the initial stages to child spacing seen as a health improvement measure. 2.3 Record-keeping. In PHN, record-keeping will also be a key task for VHWs, despite the fact that a fundamental principle of T&V in agriculture is the absence of written reporting. The only record which village agricultural extension workers (VEWs) keep is a diary which records for each visit the names of contact farmers present, the impact points conveyed, and the main problems raised by farmers. Minimal record-keeping is made possible by the visibility of the results of good extension work to farmers, VEWs and supervisors. As Benor and Harrison (1977) put it, monitoring of progresss, "is relatively simple: farmers can be asked if they know the name of their VEW, the day of his visit and three or four of the recommendations made that fortnight or even during the whole season. If the farmers know these and results can also be seen in the fields (emphasis added), then the extension personnel are doing their main job." In health care, the problem is that results are often not visible in the field, so that not only is it hard for VHWs and supervisors to monitor progress, but, more important, clients have less incentive to adopt innovation. Progress measurement is therefore necessary to motivate both VHWs and their clients. The need to keep basic records is a further argument for minimizing the number of PHN care tasks, so that reporting on a wider range of services given to a few clients does not take the place of actually administering a narrower range of services to more clients. Record keeping for day-to-day management purposes in a home visit system needs to be distinguished both from performance evaluation and from reporting. Impact evaluation is extremely complex and requires scarce statistical and technical skills. Records kept by VHWs will be insufficiently comprehensive and too inaccurate for evaluation purposes; impact assessments should therefore be left to central units doing carefully controlled sample surveys--as is the case in T&V in agriculture. Nor does reporting upwards in the bureaucratic hierarchy serve the key managerial task of motivating VHW and client. Many projects have found that elaborate reporting requirements encourage a focus on the clinic rather than the home; reduce scarce time spent with beneficiaries; and - 6 - produce data that are not processed and used. For a home visit system, the following combination of records might be both useful and manageable: (i) A visit diary for VHWs (as in T&V in agriculture) noting the names of contact clients visited, main messages conveyed, and problems raised. The VHW would also need to note in this diary the quantity of drugs dispensed, and the number of family planning acceptors and cases of malnutrition and dehydration seen. (ii) Client records. These might maintain six types of basic information: on family members (age and sex); on events (births, deaths and pregnancies); on immunization status; on family planning status; on correct use of ORT; and on nutrition status. The first three can be recorded on a single card per family, designed for use by semi-educated VHWs. The second three can.be recorded on a single chart hung on the client's wall and tracking each child's nutritional status (updated for each child for each visit) and the client's family planning and ORT status (whether or not used since last visit). Ideally, both card and chart should be kept by the parents, in order to foster their interest in and responsibility for their family's health, and to provide the incentive of visible progress. Records kept in the home have the advantage of doing away with transportation and filing problems and ensure that VHWs cannot falsely record home visits that they have not made. Measuring progress with preventive care is also an important incentive to ensure that VHWs do not neglect education in favor of curative work. However, asking VHWs to record more information than that outlined above is likely to be counterproductive: for example, attempting to track diarrhea incidence is of limited practical use when VHWs cannot be expected to prevent its occurrence. (iii) Village records. The basic information on malnourishment, dehydration and use and non-use of family planning collected in VHWs' diaries should be collated (see 5.2) and displayed publicly in a central place in each village and updated monthly--a period long enough to show change and frequent enough to maintain interest. The CIMDER program in Colombia has made such information comprehensible to illiterates (most of whom can read numbers, but not letters) through the use of a horizontally striped "health flag" displayed in each village. Each stripe is associated with a different problem (e.g. red for malnutrition) and the number of cases last period and this period are shown each end of the stripe. Costa Rica's system is still more useful, where the skills are available to set it up and maintain it: problem cases are marked house by house with different colored pins on a village map. Public display of PHN information has two major motivational effects. First, greater awareness of PHN problems which were previously invisible fosters community concern about PHN. This reinforces the individual and group concern promoted by home visits. Second, publicity provides a strong incentive for health workers, by underlining both good and bad performance and increasing VHWs' accountability to the community. - 7 - 3. DEVELOPING EXTENSION MESSAGES There is a big step between specifying priority PHN tasks, and turning these tasks into extension messages which make sense to villagers as well as professionals. This section looks at the complications involved in developing effective extension messages, and the implications for research. 3.1 Simplicity is relative. Message development in PHN is simpler to the extent that there are fewer messages than in agriculture, where recommendations must change constantly with the seasons, and must also vary for the wide variety of crops that may be grown in one area in one season. In agriculture, too, extension services are increasingly finding that they have disszeminated the backlog of previously unapplied advice available from research stations. Developing health messages is easier to the extent that there is a tremendous backlog of proven technology not yet disseminated and applied in the villages--innovations which, even if limited to the few interventions outlined above, can counter over half of all mortality and morbidity. Simplicity in PHN extension is, however, relative and largely illusory. While tasks are clear and few in number, extension messages are nevertheless complex and variable. They must vary according to the incidence of disease in different areas, and seasonally. They must vary for children of different ages (e.g., changing food recommendations) and according to the education and financial resources of the family. In some Indonesian villages, for example, better-off families will use a different cooking oil from poorer families--and the differing protein content of the oils can have a significant effect on nutritional status. Message presentation must vary to take account of local beliefs, perceptions and practices; for example, medical or contraceptive practices, which can differ between socioeconomic groups within villages, and between villages in the same region. In different villages on the same Peruvian mountain, for example, different traditional medicines may be used against malaria-- some effective, and some not. Annex 3 illustrates the potential complexity of inducing behavioral charge in PHN, taking the case of ORT. The understanding that a mother must have in order to want to use ORT is considerable, and the list of belief and behavioral changes that she must adopt if she is to use it effectively is a long one. Developing a simple, convincing case for innovation may require complex insights into local beliefs and perceptions. 3.2 Message development: implications for research. The need to alter the content and presentation of PHN extension messages according to clients circumstances, beliefs and needs has important implications for research. Three types of research activity are likely to be relevant at different administrative levels. Certain types of technical research at the international or national level can benefit PHN: developing vaccines that do not depend on a cold chain, for example. National level research also makes sense for the development of drugs and support equipment where there are significant manufacturing economies of scale: quality generic drugs; standardized ORT packets; cheap, lightweight weighing scales, for - 8 - example (although all of these require extensive field testing to ensure acceptability and appropriateness for VHWs and clients). The second--and more important--level for research is the program level, where the development of core messages would take place in the field in a limited number of pilot villages on a trial-and-error basis. A heavy social science input would be needed both to help shape messages that make behavioral sense, and to document the results for training and dissemination on a larger scale. Many primary PHN systems neglect this kind of research, because medical professionals dominate, and research which centers on communication rather than medicine is not what doctors have been trained for. Regional research centers would be the appropriate base for this kind of research. The third--and critical--need is for continuous message development at the local level. Central research and pilot tests are no substitute for an extension system which will routinely adapt recommendations to each village or family situation. Nutrition education provides an example of what is needed. Cooking habits or the types of vegetables grown in home gardens will vary between localities: extension advice must build on local custom. Foods for infants purchased by richer and poorer families in the same village will vary according to what they can afford. Nutrition advice must therefore vary for different families. In any village some poorer families will be better nourished than others with the same assets and demographic characteristics. Only the local extension service can discover and disseminate what the mothers of better nourished families do already, rather than introducing innovations developed by outsiders which may be more alien or less affordable. Adaptive research in health therefore goes hand in hand with extension. In conjunction with her supervisor and subject-matter specialists (see 5) below, each VHW needs to act as an adaptive researcher, responding to the particular constraints and opportunities of her own village. This implies an extension system which can learn as well as teach, and which can disseminate experience upwards and laterally, as well as downwards. These organizational needs have important implications for the regularity and frequency of home visits, and the training and supervision of VHWs, discussed below. 4. VISITING There is a compelling case for home visit as against clinic-based PHN extension. First, there is the simple fact that many clients--especially those who are poor and of low status--seldom visit clinics. Second, clinics are at a strong disadvantage in delivering effective preventive services to clients that do come. As emphasized above, effective education depends on knowledge of client families and their circumstances, building confidence and credibility, and follow-up to see whether and how recommendations need changing. Short, infrequent visits to clinics by clients seeking curative care cannot build the mutual understanding between client and health worker which is the basis for effective extension. - 9 - Much has been written about the use of VHWs for home visits, especially about VHW selection, training, and remuneration. But while a good deal of attention has been given to task analysis and training for VHWs, very little attempt has been made to define how field workers should use their time on a day-to-day basis, and how they should be supported. In particular, the PHN literature neglects such critical details of home visit management as the following: agent/client ratios; frequency, regularity and duration of visits; procedures during visits; supervisor/agent ratios; supervisor routines and behavior.2/ In agriculture, T&V has attempted to fill a similar gap with its structured system of standard routines for VEWs and supervisors. It is this aspect of T&V, coupled with its emphasis on feedback and learning, which may prove particularly valuable for PHN. The next subsections attempt to think through the implications of the PHN extension, task in the practical terms of routines and procedures. 4.1 Visiting: whom, when, and how often. Since not all clients can be regularly visited, priorities must be assigned. For three reasons, the priority clients for visits should be newly pregnant mothers, followed by mothers with small children. First, from a family planning perspective, these are likely to be women who are highly fertile. Second, mothers' health and nutrition during pregnancy critically affect babies' birth-weights, in turn closely correlated with infant morbidity and mortality. From a medical perspective, this is therefore also a critical entry point in the family cycle. Third, from a behavioral perspective, it is during pregnancy that women are likely to be most concerned about their health and receptive to care and advice. Confidence built up during this period may be the best basis for introducing behavioral changes affecting child health, or for the acceptance of family planning advice. A visit frequency of not less than once a month is likely to be critical for several reasons. First, physical changes occur rapidly during pregnancy and in the first year of life: these are also periods of high risk in terms of mortality. This implies changing extension messages, and a need for regular surveillance: both call for frequent visits. Second, repeated visits are needed to build a relationship of trust between client and VHW, to reinforce extension messages, and to provide encouragement and support for newly adopted behavior - the latter being particularly important for new family planning acceptors. Third, following from the discussion in previous sections, the complexity and uncertainty of PHN extension mean that recommendations may not be adopted, and frequent visits are needed to observe progress, learn more about clients' problems and beliefs, and to develop and test more effective messages. 2/ Most accounts merely contain statements about target and actual agent/client ratios as an indicator of coverage. These statements are seldom enlightening because they tend not to distinguish between the numbers of clients that are in a VHW's circle or area; the numbers that are actually reached; and the numbers that are reached effectively. - 10 - Regular visits are likely to be as important in PHN as in agricultural extension, and for much the same reasons. Clients are busy, poor clients often especially so. VHWs who arrive at random intervals cannot expect to find a ready audience. It is therefore critical, as in agricultural T&V, for VHWs to visit the same village, and the same families within a village on the same day of the week and at the same time of day. Clients can then be expected to be there, and regular visits will promote the image of a reliable, professional service. From a managerial perspective, regular visits introduce two key elements of accountability. Clients who know when their VHW should be with them can hold him/her accountable for missed visits. In addition, supervisors, unlike the case in most other systems, know where their subordinates will be at any given time on any day of the year, providing a further strong incentive for VHWs to make, their rounds. As in agriculture, it would be critical for visits to be scheduled not monthly on a calendar basis, but on the same day of the week every four weeks, so that clients can associate VHWs with a particular weekday. Visit days should be painted on village walls as a reminder. Regular, frequent visits are necessary if PHN messages are to be adapted and continuously adopted. At the same time they imply an intensive use of field staff resources. Extension systems must therefore plan diffusion carefully if they are to be cost-effective. In agriculture, VEWs visit eight different farmers' groups a fortnight, each consisting of about 100 farmers, and each group represented by about ten contact farmers who must attend fortnightly visits. Diffusion relies primarily on the visible effectiveness of extension recommendations. Contact farmers need not act as extension agents themselves, but are expected to spend time showing other farmers new techniques if they express interest. In PHN, there is a strong case for client groups to be smaller than in agriculture, since diffusion among "non-contact" clients will take place less spontaneously, and there will be stronger cultural barriers to adoption even among contact clients. It is suggested that VHWs might concentrate on groups of no more than five "contact mothers" a visit, and that half a day might be spent in the neighborhood in which each group lives. Relatively long visits are necessary to build up a relationship which involves an exploration of client families' PHN status, beliefs and habits. Visiting groups of clients is important not only because individual visits are likely to be too resource-intensive to be affordable, but because group meetings foster communal responsibility for health care, and peer pressure provides a strong incentive to adopt and continue with innovations. Personal attention is likely to be critical if mothers are to adopt advice affecting their personal lives, and for this reason a visit to two groups of five a day is likely to-be more effective than one visit to a group of ten. Since messages will not diffuse by example as in agriculture, each contact mother might be asked to pass on the VHW's advice to two designated other women, forming an "outer group." While meetings would be open to all those living in the neighborhood, outer group mothers would be strongly encouraged to attend regularly, and contact mothers would have to attend meetings regularly, or they would be replaced. Contact mothers would be chosen for their interest - 11 - in health care and willingness to pass on PHN information to others. Contact mothers in a given group should be as far as possible homogenous in socioeconomic terms. This would mean that mothers' problems were similar, ensure that extension messages were relevant to all, and make group involvement and peer pressure more effective. Meeting places might rotate between the houses of contact mothers to ensure equal involvement, to help VHWs learn about individual family's circumstances, and to use differing circumstances and environments to educate the group about the links between MCH habits and health. Counting inner and outer groups, a VHW would be able to reach 30 families a day, or 480 families altogether, of which 160 would be inner group families. This assumes two group meetings a day, and four days a week devoted to group visits (see also 5.1 below). 4.2 Routines during visits. VHWs need clear guidance on how to use their time during home visits, as well as clear schedules of where and when to visit. The routine for a typical visit might be as follows (the order of events is important): i) Inquiry. After the VHW records the names of contact mothers present, he/she would begin by inquiring about deaths and illnesses among the group, their outer group contacts and acquaintances. The VHW would then review whether the group (including each of the outer group members whether present or not) has continuously adopted previous recommendations, assist in weighing and measuring children, and bringing health cards and charts up to date. ii) Teaching. The training session which followed would then center on PHN problems and non-adoptions uncovered through inquiry in the early part of the meeting and on symptoms, e.g., of malnutrition or dehydration, evident in children brought by mothers for weighing. This would underline the relevance and usefulness of PHN advice, and increase client participation. Techniques would be taught by the VHW, and then be demonstrated by one of the group, the others participating with comments and criticism. At the end of a teaching session, particular adoption goals would be set with group agreement for individual members. Peer pressure--backed up by peer review at the next meeting--would be a key incentive. Inquiry and teaching sessions might together take an hour and a half. iii) Curative care. Only after the inquiry and teaching sessions would the VHW dispense curative care, and then only to those who had attended the earlier - 12 - part of the meeting. This might take half an hour. For both VHWs and clients, having curative care sessions which followed and were shorter than promotion/education sessions would be an important and literal reminder that preventive care came first. iv) Home visits. An hour would then be spent making additional home visits to the sick; to outer group members; and to local influentials. Random visits to the homes of outer group members would provide a check on the effectiveness of contact mothers diffusion efforts, and hence an incentive for them to perform. Occasional visits should be made to local influentials in order to keep them informed about progress and problems (and hence involved and supportive) and in particular to draw their attention to the cases of those too poor to do much to improve their health condition. 5. TRAINING AND SUPERVISION 5.1 Training. Training needs in PHN differ from those in agriculture in that extension messages are likely to be less variable from fortnight to fortnight. This does not, however, imply that regular training is not needed, since training sessions have a number of critical functions in addition to teaching new material. First, training sessions reinforce existing knowledge: field staff cannot be expected to learn what to do once in pre-service training and be proficient ever after. Second, regular training sessions are an opportunity to provide VHWs with regular feedback about their performance and allow VHWs to raise problems and to get advice on adapting recommendations to local conditions. Third, regular training in groups has an important motivational impact: regular access to knowledge reinforces VHW confidence; and group problem-solving builds a sense of group solidarity and belonging to a professional cadre. Finally, from the viewpoint of the extension system as a whole, training sessions provide an operational learning capability. In-service trainers are in a unique position to share problems and solutions originating in the field with other trainers and with senior managers. Training might therefore take place monthly, enabling VHWs to discuss problems encountered during any home visit before they return to the same client group. Training groups should consist of the same group of eight VHWs supervised by a single extension officer (see 5.2 and 5.3 below). VHWs would devote a full day each month to training, in order to allow for travel time to training sessions. Assuming a six day working week, this would leave VHWs seven days in each four week period free from home visits and training. These days would be used to make additional visits to families as necessary, and for catching up with administrative tasks--for example, accounting for drugs used, making reports required by the health ministry even though not required by the T&V system. - 13 - Training sessions would also be the logical occasion to distribute drugs to VHWs, for two reasons. First, simultaneous training in curative care would help ensure that no drug was distributed that VHWs did not know how to use. Second, regular replenishment would reduce the stock of drugs that VHWs need carry to a minimum that VHWs could be expected to transport and account for. In agricultural T&V, fortnightly training sessions are supported by monthly workshops, at which trainers meet with researchers, farm problems are raised, and messages are refined and disseminated. In health care, monthly VHW training sessions could be supported by 2-monthly workshops. The key function of these workshops would be to disseminate information about problems and promising approaches both vertically and laterally in the system. The workshops would also provide an opportunity for medical and communications specialists to train trainers and assist in solving problems. However, the contribution of these specialists would be no more important than that of VHW trainers, whose field experience would be a key source of ideas for improving the presentation of extension messages. 5.2 Supervision. Discussing the effectiveness of home visits and extension messages at monthly VHW training sessions is in itself an important form of supervision. However, both T&V experience in agriculture and experience in the PHN sectors indicate that systematic field supervision is a critical complement to in-service training. Successful home visit systems in health care, such as CIMDER's in Colombia, have relied on supervisor/VHW ratios as high as 1:6. In agriculture, experience indicates that performance drops as soon as supervision ratios fall much below 1:8 and there is little evidence to suggest that a lesser density might be tolerable in PHN. Following T&V practice in agriculture, a supervisor would spend four days a week in the field, participating in VHWs' home visits. A supervisor would thus spend two full days in the field in each four weeks with each of his eight VHWs. Like training, supervision has a number of purposes. Most obviously, it provides a check on whether the VHW is on the job--a check only possible in an extension sy tem structured as tightly as T&V because only in such a system is VHWs' whereabouts clearly defined. Regular supervision provides professional support and confidence critical for VHWs with limited training, and moral support and recognition which, in T&V in agriculture, have proved performance incentives more than compensating for VEWs' low salaries. The regular availability of professional support in the field adds to VHWs' credibility in the eyes of clients. Finally, regular first hand experience in the field makes supervisors an important source of feedback in the extension system. As in T&V in agriculture, field supervision would be supportive rather than authoritarian. Although supervisors would inspect visit diaries and client records, their main focus would be on VHWs' preventive and promotional work. They would participate in but not conduct home visit meetings, their role being limited to listening and asking questions of clients (Annex 4). Clients' responses would reveal the VHW's performance without any need to question him directly. Where improvements were needed, the focus would be on joint constructive discussion rather than criticism. - 14 - Supervisors in PHN would have one task additional to those in agriculture. Once a month at training sessions, supervisors would collect PHN data from VHWs, collate the figures for each village, discuss them with village leaders. and post them publicly. 5.3 The Role of Subject Matter Specialists (SMS). In agriculture, the variety of and frequent changes in recommendations mean that VHWs must be trained by specialists in several disciplines--plant pathology, treecrops, etc. In PHN, on the other hand, VHWs do not need training in medical specialties, since referrals to clinics provide the bulk of technical support. There is therefore no need for doctors to be trainers of VHWs, since the more complex aspects of VHWs' tasks are in the area of communications and presentation rather than medicine. In PHN, the main need for SMS support is in the area of IEC, a subject in which doctors are traditionally weak. There is, therefore, a theoretical case for removing the responsibility for VHW training from medical specialists. This might mean more effective training and save scarce doctors' time for referral tasks that only they can handle. Ideally, a sufficient number of communications SMS would be available to train VHWs at each fortnightly meeting. In practice, however, SMS need not have a direct role in training and supporting VHWs if SMS skills are in short supply. As elaborated in Section 12.3 below, SMS skills could be "leveraged" by focusing them on regular in-service training for small groups of supervisors rather than VHWs, as well as on the 2-monthly workshops attended by larger groups of trainer/supervisors. Despite the theoretical case for using non-medical specialists to support VHWs, the political power of doctors in line management positions will often make such a proposal unworkable. The design of the training system would need to be pragmatic, depending on the power base of different cadres, and the resources available. For example, where, as often, an integrated PHC system is supported by an influential cadre of technical SMS from former vertical programs, the opportunity should be taken to incorporate such medical SMS into the system as trainers, and the main design priority would be to build up a cadre of IEC SMS to prevent the domination of training by medical rather than communications skills. 6. SUMMARY AND ISSUES FOR DISCUSSION The six main elements of the "model" T&V system outlined above, can be summarized as follows: 1. Concentration on a small number of key tasks. 2. A performance reporting system concentrating on these tasks. 3. A system of regular, frequent home visits, preferably to groups of clients. - 15 - 4. Concentration on selected clients where interventions can have the maximum impact on reducing fertility and mortality. 5. Regular, frequent field supervision visits focused on support rather than inspection. 6. Regular in-service training. Some of the key differences between agriculture and PHN are summarized in Figure 6.1, together with their implications for the organization of extension work. This figure, and the model T&V system for PHN put forward in Part 1 of this paper, raise many uncertainties and questions for discussion, among them the following: a. Are the priority tasks suggested above the most appropriate ones? b. Is it necessary and feasible for VHWs to keep the three sets of records suggested? c. Is it right to concentrate group meetings and home visits so heavily on pregnant mothers and mothers with infants? d. Are monthly home visits by VHWs frequent enough? e. Should VHWs home visit four days a week, or could they manage five days? f. Will poor, busy clients stay home regularly at VHW visit times? g. Will clients be prepared to attend group meetings with VHWs, when individual home visits are now more common? h. Should VHWs visit two groups of five contact workers a day? Is the idea of inner and outer groups sound? How many outer group members can a contact mother be expected to pass recommendations on to? i. Is the suggested agenda for VHWs visit meetings appropriate? J. Does monthly training make sense? k. Should the VHW training group be the same group of eight under one supervisor? 1. Should a VHW's supervisor also be his trainer? - 16 - DIFPERENCES BETWEEN T&V IN AGRICULTURE & HEALTH CARE FIGURE 6.1 AGRICULTURE HEALTH CARE Implications for Outreach System Design Sectoral Characteristics Implications for Outreach System Design Emphasis on IEC Farmer Women, children, I research clients poor clients Incentives for ncentives for change better 'lchange poorly understood t understood 1 VEW: 80 contact \ Innovation Innovation affects 1 VHW: 160 contact farmers, diffusion affects econo- personal life mothers, diffu- to 800 farmers or mic life sion to 480 less A mothers or more Demand for Little demand for I innovation I innovation Automatic I Benefits of Benefits of T Poor diffusion diffusion' H-- / adoption clear adoption unclear Record-keeping e z Record-keeping unnecessary necessary Emphasis on technical ,,, I Backlog of high High return research return inno- critical/innovations vations remain unapplied I I running out r Fortnightly visits Concentration on Monthly visits and training I constantly A No input Drug distribution I few key tasks and training 1--t .1 varying * distribution necessary P throughout year messages 40C ~ more tasks Several SMS types, A Less scarce Very scarce I Only 2 SMS types, SMS train VEWS t funds and funds and I supervisors skills skills train VHWS With automatic Not all clients diffusion, all may be reachable clients reachable I - implies A - allows - 17 - m. Is the more limited role proposed for doctors and for SMS acceptable? n. Can we afford regular home visits and training, and tight supervision ratios? Or, can we afford not to have them? - 18 - PART TWO 7. INTRODUCTION: COMPARING PERFORMANCE The outreach systems reviewed in the second part of this paper cover a spectrum of managerial approaches. The Indian program relies entirely on paid staff at the village level. Its tight management system, closely resembling T&V, benefits from a relative abundance of money and skilled manpower. At the other end of the spectrum, Indonesia, with scarcer financial and administrative resources and relatively well developed community organizations, relies on volunteers, with no paid workers at the village level. Somewhere in the middle, Philippines backs up volunteer health workers with full time government-employed midwives resident in the village--a system in the early stages of evolution from a curative clinic base to a preventive outreach approach. A sample of three or four programs can yield little in the way of generalization across countries--although the cases do evidence significant differences in performance, in resource use, and in organization, with implications for the usefulness and limitations of T&V in health care. But more fundamentally than the immediate concern with T&V, the cases raise the general methodological issue of how to compare management systems. No common framework for doing this exists--a fact which itself discourages comparative analysis and systematic thinking about outreach program design. Ideally, management systems should be compared on the basis of cost, outputs, and impact on clients' health, since these are what are ultimately of interest. In practice, this is impossible for several reasons. Measurements of impact and output are often not kept, or only kept for certain tasks and services, making it difficult to compare systems which have different tasks and services, and attach different importance to each of them. For any one system, the impact of management is hard to disentangle from the impact of nonmanagerial factors on the health service (such as budgetary constraints), and of changes in the clients' environment (affecting agricultural production and incomes, for example). These problems are compounded when management systems in different countries are being compared, and major cultural and environmental variations are involved. In practice, therefore, management systems must often be compared according to their costs, and the presence or absence of prerequisites for program success. If, for example, nutrition field workers are clear about their tasks and are visiting all of their clients regularly, they may or may not be giving effective extension education and changing clients' behavior. But if field staff are unclear about their role or are not making regular home visits, then we know that they cannot be effective. This method of assessment is obviously crude, since it breaks down when attempts are made to compare relatively well-functioning systems where most prerequisites are present. But in most PHN management systems, enough "sine qua nons" of success are missing, for comparisons on this basis to be useful in practice. Over time, it would be useful to work towards a consistent format for assessing performance and costs. For a brief review (2-3 days in the field for each program visited) of the type that gave rise to the case - 19 - material below, the following starting points might be used as a basis for refinement. First, in the absence of quantitative measures, performance assessment must rely heavily on interviews with VHWs and beneficiaries, which may or may not be consistently approached: a tentative interview format is suggested in Annex 4. Second -- at the risk of overquantifying the subjective -- greater consistency in the presentation of the results of "prerequisites of effectiveness" assessments might follow from a systematic scoring system like that in Table 7.1 below: this is a tentative summary of the case material in Sections 8-10. Third, with regard to costs, the major variable is likely to be human resource use, and a tentative format for comparing manpower use is developed in Tables 11.1-11.4. Table 7.1 below is open to three types of criticism. First, the numbers are subjective ratings. Second, the criteria -- which are effectively a set of weights -- are open to discussion, especially as several are multicollinear. Third, the criteria deal with inputs rather than outputs. Over time, and for programs in which the Bank has a sustained involvement, it should be possible to move towards a more complete set of measures which are not only consistent but focus on outcomes as well as inputs. Similarly, for the resource use figures in Section 11, which focus only on human resources, it should be possible to develop more complete measures including drug, transport and other infrastructure costs, to allocate costs more accurately between clinic and outreach work, and perhaps also to allocate some costs between different outreach services. Table 7.1 Scoring the Prerequisities of Managerial Effectiveness: A sample format Tamil Philippines Philippines Indonesia Nadu National PUSH MOH Normal 1. PHN services preventive 4 2 3 4 rather than curative 2. Pre-service training 5 1 5 1 adequate 3. In-service training 1 1 3 1 adequate 4. VHWs not overloaded with 5 3 4 1 tasks 5. Appropriate records 4 2 4 3 kept up 6. Home visits regular and 3 1 2 2 long enough to be effective 7. Visit coverage equitable 5 1 3 2 8. Messages adapted to 4 1 2 1 family circumstances 9. Supervision effective 5 1 3 1 Average rating 4 1.4 3.2 1.8 Note: 5 is high, 1 is low - 20 - 8. THE NUTRITION PROJECT IN TAMIL NADU, INDIA Health care is provided by two separate services in the field. The key staff are Multipurpose Health Workers (MPHWs), supervised by Lady Health Visitors (LHVs), and working for the Ministry of Health; and Community Nutrition Workers (CNWs), supervised by Community Nutrition Supervisors (CNSs), working for the Ministry of Social Welfare. The health and nutrition services are described separately below. Health 8.1 Multipurpose Health Workers. About 670 MPHWs have been in post for a year or more in the project area. Each covers a population of about 5,000 and is based at a Health Subcenter (HSC) containing living quarters, a clinic room with basic drugs, and a delivery room with one bed. MPHWs are paid about Rs 450 per month (as of 1983). They must be secondary school graduates, and receive 18 months training, 12 in a local hospital, six in the field. The field training is practically oriented: after ten days classroom instruction, MPHWs are assigned to a village and given 100-150 families to care for. A week is spent doing a census, and the MPHW then makes home visits for four months under the trainer's supervision, carrying out all the tasks for which she will later be responsible. A month is then spent working at an HSC, where particular attention is paid to giving students experience with deliveries. A further month's study follows before examination. 8.2 Lady Health Visitors and Doctors. LHVs are also secondary school graduates, trained for 18 months, and responsible for supervising four MPHWs. LHVs in turn are supervised by doctors working out of a Primary Health Center (PHC), which is the referral point for about a dozen HSCs/MPHWs. There should be three doctors per PHC, though many are understaffed. Each doctor is expected to spend two-thirds of his time in the field, visiting HSC areas and their staff, and holding mobile clinics in the villages. Depending largely on the convenience of the PHC's location, out-patient loads vary from 100-300 per day. A combination of high out-patient loads and staff and transport shortages means that doctors' field schedules are often diluted. 8.3 MPHW and LHV tasks and routines. MPHWs have a well defined set of tasks, concentrating on pre- and postnatal services, deliveries, infant and child care, treatment of minor ailments and referrals. They are trained to do immunizations, and carry half a dozen basic drugs, not including antibiotics. In addition to providing basic health services, they are responsible for family planning motivation, and for monthly weighing of pregnant women and infants of 0-5 months, as well as quarterly weighing of children from three to five years. Weight cards are supposed to be kept for each child, as well as family records giving census and vital event data, and noting MPHW visits, treatments and referrals. MPHWs' work routines are also clearly defined. Twenty days a month (Monday to Friday) are spent making home visits to the average of - 21 - 1,000 households in each MPHW's area. MPHWs divide their populations, which consist of three or more villages, into ten geographic areas, and visit each area fortnightly on the same day of the week for a whole day. A written visit schedule is maintained at the HSC. On average, an MPHW must cover about 100 or 120 households a day, travelling by bus and on foot. Every Saturday morning, each LHV meets with her MPHWs to review the week's work, the meeting place rotating between the HSCs. Records are checked, MPHWs discuss problems encountered, and joint plans are made for the coming week (e.g., for immunizations to be carried out in a particular village). Every Saturday afternoon, LHVs go to the PHC for a second review session, led by the doctors. At this weekly meeting, drug supplies are also distributed. Once a month, doctors hold a review session with all LHVs and MPHWs in the PHC area, and on this day staff are paid. LHVs too have clearly defined schedules, and spend Monday to Friday accompanying MPHWs on home visits. One whole day each week is spent with each MPHW, and an extra day with one MPHW. Supervision visits are not made to the same MPHW on a regular day of the week, but LHVs' schedules are known in advance to MPHWs (doctors' visits, on the other hand, are less predictable). LHVs' five days a week in the field and a ratio of four MPHWs per LHV mean that an LHV will return to each MPHW sub-area (i.e., each of the MPHW's ten visit areas) once every eight weeks. Like MPHWs, LHVs travel by bus and on foot. Nutrition 8.4 Community Nutrition Workers. About 1,900 CNWs have been in post for a year or more in the project area. CNWs are full time paraprofessionals, paid Rs 90-115 per month, about a quarter of what MPHWs get. Each CNW is responsible for one village, plus any surrounding hamlets, and is recruited from the village population. CNWs are based at a Community Nutrition Center (CNC), usually a room in an existing village building. CNWs have a primary education, and are chosen by senior nutrition supervisors and village councils, as far as possible from among the most disadvantaged villages. CNWs are trained for two months, of which ten days are joint training with student MPHWs in the field, carrying out all tasks for which CNWs will later be responsible, including coordination with the health service. 8.5 CNW and CNS Tasks and Routines. CNWs' tasks and schedules are clearly defined. Each morning, eligible children come to the CNC for supplementary feeding with a commercially prepared food stored at the CNC. Mothers and grandmothers who bring children are given nutrition education, although many children are brought by siblings. During the last 5-6 days of each month, the CNW weighs all children aged 6-36 months, either in the CNC or in village neighborhoods. Weights are recorded on a chart for each child, and children with consistent weight loss are admitted to the feeding program, and graduated from the program as soon as weight/weight gain is satisfactory. The feeding program is both selective and short-term; children are normally graduated after three months. The monthly surveillance is the key to the system, and participation in this weighing is 80-90% in most villages after a year of operation. Considerable efforts - 22 - are needed in the early stages to make mothers understand why they should continue to bring children for weighing, who do not subsequently get supplementary feeding. CNWs form women's groups of about 20 women in their area, who are encouraged to attend a cooking demonstration by the CNW twice a month, usually on the sixth and the sixteenth. Typically, ten or 15 women come, including 3-4 mothers of malnourished children (attendance is restricted, as the Rs 7.50 cooking allowance limits the number who can sample the recipe being demonstrated). Womens group members are drawn from different areas of the village, and are encouraged to pass on nutrition information and give moral support to mothers of malnourished children in their area. CNWs reserve two other days a month for review sessions with their supervisors (see 8.6 below). The remaining 15-16 afternoons of the month are spent carrying out home visits. CNWs divide their villages into sub-areas, and visit each sub-area monthly on a calendar basis. In practice, this means that a CNW makes 10-20 home visits in an afternoon, depending on the size of her village. During their home visits, CNWs spend most time giving nutrition advice to families containing pregnant women and small children, particularly those with children in or recently graduated from the supplementary feeding program. CNWs other specific tasks consist of 6-monthly administration of Vitamin A concentrate and 4-monthly administration of deworming medicine to all preschool children; and supply of ORT packets to back up education on diarrhea management. Also during home visits, CNWs organize clients for immunization clinics carried out by the health services, and refer cases of fever and other illness to MPHWs. 8.6 Supervision. A Community Nutrition Supervisor (CNS) oversees the work of ten CNWs. CNS receive two months of training, and are either college graduates or promoted from within the service. Like LHVs, they spend 20 days a month on field visits. Their regular schedules allow them to spend one day with each CNW a fortnight; the day of the visit is not known to the CNW in advance. A large part of\CNS supervision activity consists of checking records (weights are sometimes improperly entered on charts), doing sample weighings to verify the accuracy of monthly weighings, and making the final selection of children to be admitted to or discharged from the feeding program. In addition to these individual supervision visits, group meetings of CNWs are held twice a month. One fortnight each CNS meets with all of her CNWs; the alternate fortnight there is a meeting of all nutrition staff in the PHC area (this might involve about 70 CNWs and their supervisors, under the direction of a Community Nutrition Instructress). 8.7 Coordination Between Health and Nutrition. Constant coordination between the two services is needed at the village level. CNWs need to refer children who continue to fail to thrive to MPHWs for treatment or futher referral to the PHC; MPHWs need to pass cases of malnourishment to CNWs for feeding, and diarrhea cases for ORT treatment. Both CNWs and MPHWs are involved in nutrition education during their home visits. - 23 - In the initial stages of the program, cooperation was limited. Two main methods have been used to improve coordination to the point where CNWs and MPHWs can commonly be found making joint home visits. First, the final ten days of pre-service field training for both MPHWs and CNWs are held jointly in the field, in order to familiarize each with the other's tasks, and to stress their cooperative role. Second, workshops are held at 3-4 monthly intervals for staff of all ranks, in both health and nutrition, in a given PHC area. A week before the workshop, staff in different work groups are given different subjects (e.g., diarrhea management, coordination) to prepare, and their presentations are then used as the basis for discussion of field problems. Participation of staff at all levels is apparently enthusiastic, and in addition to providing a forum for sharing field experience, these workshops have had an important role in increasing cooperation on the job. Commentary 8.8 Tasks and Schedules. The nutrition program in particular closely resembles the T&V system in its adherence to the principles of task concentration and visit regularity. Tasks and work routines are clearly defined, and pre-service training is relevant and practical. Visit schedules appear to have been worked out for all staff, and are adhered to in the field. This makes for regular and equitable coverage of populations, and permits effective supervision, since CNS know where their subordinates can be found. MPHWs, on the other hand, have a heavier workload, both in terms of more tasks and more clients to be visited. 8.9 Workload and Routines: MPHWs. The number of homes MPHWs are expected to visit in a day is too great for effective extension. MPHWs tend to make extremely brief visits, simply asking whether there is any health problem in the family and passing on if the answer is no. In an area where over 50% of all pre-school children are malnourished and mothers often do not recognize the signs of malnutrition, this is clearly dangerous. Random home visits commonly reveal morbidity, particularly deficiencies such as anaemia and night blindness, which have gone undetected by MPHWs. Sometimes this reflects poor training of the MPHW, but often MPHWs are able to diagnose the problem correctly when asked, but have not taken the time to examine the child. MPHWs might have more incentive to make thorough checks if they were provided with a four or five-item checklist of key problems to be examined for in each child. Another approach might be to reduce the number of MPHW home visits by targeting them more selectively on families with pregnant women and/or pre-school age children. Although MPHWs visit the same village areas fortnightly on the same day of the week, one departure from T&V practice is that while the MPHW knows her regular schedule, most of her clients do not know either the day of her visit, or often, the MPHW-s name. Writing MPHWs names and visit days on village walls would help ensure that clients are at home when the MPHW comes, and would increase the MPHW's accountability to her clients. - 24 - A second departure from T&V practice is MPHWs' five rather than 4-day visit schedule per week. While this helps them to get round a large number of clients, it can lead to irregular coverage. As part of their family planning duties, MPHWs take women wanting sterilization to the PHC or local hospital and will be absent from their usual rounds on that day, as well as on days when there is an immunization or other campaign in their area. When holidays and illness are added, the interruptions in visit regularity become significant. T&V allows for this by reserving one day a week for catching up on administrative duties and missed visits. However, this arrangement could only be adopted in the MPHW system at the cost of more selectivity in the choice of homes to be visited on the remaining four days of the week. 8.10 Workload and Routines: CNWs. CNC areas vary considerably in population, unlike MPHW areas which are chosen to contain around 5,000 people. Most CNC areas contain 1,000-1,400 people, of whom about 70 are children aged 6-36 months. In a typical area, 18 or 20 children might be receiving supplementary feeding, of whom three might graduate from the feeding program a month, and eight of the children fed might be third or fourth grade malnutrition cases. However, CNC area populations vary from 560 to more than 2,300, with from 30 up to 150 eligible children. Third and fourth degree malnutrition cases vary in number from one to 16 or 18. In the longest established PHC area of the program, relapse cases are significantly concentrated in six or eight of the 68 CNC areas. Villages with high relapse rates tend to have a large harijan community and hence a high incidence of poverty. They also tend to be larger villages with outlying hamlets in which the poorer people live. CNW workloads therefore vary significantly. Those with problem areas tend to have not only larger, but more scattered populations to reach. An obvious but costly response would be to appoint additional CNWs in selected villages with high relapse rates. A cheaper alternative might be to develop a more active role for the mothers' groups in these villages. An important departure from T&V practice is that CNW schedules are on a monthly calendar rather than day-of-the week basis (unlike MPHW schedules). CNWs do not return to the same area on the same day of the week so that visits, while regular, cannot easily be predicted by clients. Since malnutrition cases are concentrated in poorer families where both parents work in the fields and are absent much of the day, predictable visit days would help to increase the chances of CNW and client coinciding. Monthly rather than fortnightly home visit schedules compound the problem of missed clients. While even a monthly visit may be insufficiently frequent to develop an effective relationship with the client for nutrition education purposes (see below), missed CNW visits due to holidays and illness, coupled with frequent clients' absence, must mean that not all clients can be contacted even once a month. As with MPHWs, building free days for catching up on missed visits into the schedule might help, but at the cost of greater selectivity. 8.11 Communication, Supervision and Training. The program again resembles T&V closely in its emphasis on regular and intensive supervision, and the frequent opportunities it provides for review of field experience. Supervision in both health and nutrition services involves regular whole -.25 - day field visits by supervisors, and regular group review sessions. In the case of CNWs and CNS, communication is also improved by the provision of "comment cards" which staff may fill in and post to the center, and a newsletter circulated every two months. However, in both services, supervision and review time appear to be used more for checking performance in terms of field records, than for staff support and training. MPHWs have a clear need for regular in-service training, which is not currently provided. They cannot be expected to recall all topics taught in an 18-month course without frequent refreshment and reinforcement. Time set aside during review sessions for refresher training on specific topics by LHVs and doctors could do much to fill the gap. However, many LHVs themselves would first need in-service training both to refresh their technical skills, and to equip them for a revised supervisory and training role. Many doctors also need reorientation training if they are to be persuaded to give as much time to support for MPHWs' preventive work in the field as to curative care in the PHC and mobile clinics. Recognizing the need to start this reorientation from the top down, program managers have initially proposed five days reorientation training for doctors every six months. With regard to CNWs, random interviews with mothers of malnourished children suggest that not enough time is spent getting to know individual family circumstances, either to adapt advice appropriately or to build the personal relationship needed for effective extension. (It is, however, very difficult to evaluate how far mothers' non-adoption of new food habits is the result of inadequate education and motivation, and how far of poverty). This problem reflects both the relative emphasis of supervision on different CNW tasks, and the varying complexity of the different tasks themselves. CNW supervision appears to concentrate most on the progress of weighing and supplementary feeding, and record-keeping for these activities--an emphasis which is understandable, given that these activities are much more easily measured than is progress with nutrition education. In turn, it is rational for CNWs to concentrate on weighing and feeding, not only because these are the most measured activities, but also because these activities are much more routine and simple than nutrition education, which requires a high degree of initiative in adapting recommendations to individual family circumstances. A heavier emphasis on nutrition education during supervision, with more in-service training on message adaptation given as supervisors accompany CNWs on home visits, may be needed to correct the tendency of CNWs to pass on nutrition messages mechanically. But appropriate message adaptation and effective persuasion also depend on understanding and trust between CNW and mother, and it is unclear whether monthly CNW visits provide a sufficient basis for this. An issue for discussion with regard to later stages of the project is whether the more effective extension that could be provided with fortnightly visits would be worth the cost in terms of more selective client coverage that this would entail. 8.12 Summary. The Tamil Nadu management system departs from T&V practice in two main ways--the absence of in-service training, and the fact - 26 - that home visits are to individuals rather than to groups. The elements which it has in common with the T&V approach, are task concentration, regular visit schedules, close supervision, and provision for communication, both vertically and laterally. In terms of having the prerequisites for success, the Tamil Nadu Nutrition Project is extremely impressive. Training programs thoroughly cover not only all the theory field workers need, but give them substantial practical experience in the village. All CNWs, CNS, MPHWs and LHVs interviewed had a clear grasp of their roles and seemed motivated. Drop-out rates were low. All had defined home visit or supervisory schedules and appeared to be keeping to them. All were maintaining records of progress, in terms of treatments in the case of MPHWs, and of entrants to and graduates from the feeding program in the case of CNWs. CNWs and MPHWs showed a ready knowledge of the individual family cases behind the aggregate figures. Also encouraging is the fact that where there are apparent weaknesses--such as in the need for in-service training of MPHWs and greater efforts by CNWs to adapt nutrition messages to clients' circumstances--the design of the system allows these to be corrected without major reform. Intensive supervision and frequent staff meetings provide a ready opportunity for corrective action to be transmitted through the system. 9. VILLAGE HEALTH WORKERS IN THE PHILIPPINES In 1979, the Ministry of Health reorganized its primary care services in order to provide better coverage of rural areas. At the village level, the major innovation was a new cadre of volunteer Barangay (Village) Health Workers (BHWs), supervised by midwives based at local health centers. This case looks first at the new system on the island of Cebu, where it has been longest established. It then discusses a more resource-intensive primary health system developed under the USAID-funded 'PUSH project on the island of Panay. The Barangay (Village) Health System 9.1 BHWs. BHWs are part-time volunteer workers with a primary education, recruited from the village, and receiving no material incentives. About 90,000 are now in post throughout the Philippines, each caring for 15-20 families. BHWs receive 5 days' training, concentrating on village surveys, first aid, pre- and post-natal care, MCH, weighing and referrals. About half a day is spent on IEC work, and there is no practical training in the field. In most villages, one or two BHWs who are particularly enthusiastic and live nearby put in substantial time assisting the midwife in the village clinic. However, most BHWs, at least on the island of Cebu, have part-time paid jobs, and hence limited time to devote to health care. Most say that they devote two hours a day to their health work, but this must often be an overestimate: for example, few BHWs claim to make more than 3 home visits a week. BHWs are generally short of supplies, including basic items such as plasters for first aid. - 27 - 9.2 Midwives. BHWs are supported by midwives based at village health centers, and paid 660 pesos a month (the wage for farm labor is P. 15 a day, or less). On average, midwives are responsible for a population of about 3,500 people, though populations of more than 5,000 are not uncommon. Midwives spend the morning in the health center, dealing with out-patients. Inadequate supplies, from cough medicines and antibiotics to vitamins, are a common problem. In the afternoon, midwives make 3-5 home visits, concentrating on pre-natal cases, and follow-up of patients seen at the clinic, especially those who have failed to report for medicines. Visits last about 30 minutes, and the midwife is sometimes, but not always, accompanied by the relevant BHW. Midwives, assisted by BHWs, are also responsible for weighing all young children once a year. In practice, 75-80% of children under 6 are weighed; it appears that those missed are more likely to have illiterate or poor parents and hence are more likely than average to be malnourished.' Children identified as malnourished are subsequently weighed every three months if they have first or second degree malnutrition, monthly if they are third degree cases. Villages in Cebu will not typically have more than 2-3 cases of the latter. Growth charts are maintained by the midwife for malnourished children, and monthly cooking demonstrations are held, to which 20-30 mothers come. Once a month, each midwife meets with her BHWs (about 75% usually come) to review problems and plans. Typical subjects for discussion are progress in the toilet installation campaign which the Ministry has given priority, supply problems, and plans for upcoming immunizations. BHWs are not trained at these sessions, nor have they received any formal refresher training since recruitment in 1979. 9.3 Higher level supervision. Nurses (paid P. 740 per month) supervise 4-6 midwives each. Visits, using public transport, vary in frequency according to the competency of the midwife: some may be visited once a week, others 2-3 times a week. Nurses spend the morning of their visits at the health center, dealing with 15-30 out-patient referrals. Afternoons are spent home visiting with the midwife. Doctors supervise nurses and midwives, but supervision ratios and visit routines vary widely. One doctor encountered had 12 health centers under her care and visited each once a week on the same morning or afternoon. More commonly, doctors have less health centers to supervise, yet make less regular and less frequent visits. One problem is that inadequate transport allowances mean that doctors must use their own funds if they are to make regular supervision visits. The 5-10 doctors in a given District meet monthly for one day, and there is a quarterly conference for all doctors, nurses and midwives in the District. Panay Unified Services for Health--Project PUSH 9.4 The main organizational differences at the village level between the national barangay health system and project PUSH are that the latter has more paid workers, and offers better training. 9.5 Staff and Incentives. The lowest-level health worker is the volunteer leader of a group of 20 families, as elsewhere in the Philippines. However, in the PUSH project, these workers are known as unit leaders, while the designation BHW is reserved for two full-time paid - 28 - assistants to the midwife. These BHWs are responsible for 100-200 families, i.e., 5-10 unit leaders, and receive a salary plus allowance of P. 315 per month. They spend mornings assisting in the clinic, and average 3 home visits an afternoon. Each unit leader is visited by a BHW on average twice a month, although there is no regular supervision schedule. Under the PUSH project, unit leaders appear to have a clearer and more limited role than elsewhere, concentrating on IEC activities for diarrhea, anaemia and parasite control. BHWs take a more active role in weighing, which is carried out every three months for all children under 6 under the PUSH project. 90-100% of all children are weighed in practice, and the fact that this figure is higher than outside the PUSH project seems to be the result partly of greater staff resources, and partly of better organization: scales are taken to each unit leader's area on a predetermined day, making it easier for mothers to participate. The nutrition surveillance system is complemented by a child day-care system, usually operating in a room next to the village health center. The day-care worker is paid P. 60 a month by the municipality for 4 hours work a day, and her tasks include providing supplementary feeding for malnourished children (parents who can afford it provide food for their child at the center, offsetting part of the cost). Supplementary feeding for malnourished children is available much more sporadically outside the PUSH project, and outside the island of Panay, which has a relatively large per capita income and local government budget. 9.6 Training. The basic training course for BHWs lasts 6 weeks, of which one is spent in the field, learning how to carry out a health survey, and field and clinic work. The classroom instruction has a heavy emphasis on IEC work, and on demonstrations, simulations and role-playing. Most BHWs in Cebu have had one in-service refresher course lasting 15 days, of which 4 or 5 are spent in the field. Students feel that these refresher courses are needed and beneficial. 9.7 Commentary. Maintaining the motivation of the village volunteers is a major practical problem in the nationwide BHW system: about 20% of the volunteers drop out each year. Three factors seem mainly responsible. First, BHWs only receive a few days pre-service training and no in-service refreshment, and hence often do not have a clear idea of their roles. One indicator of this is that many volunteers interviewed did not know the significance of the lines on growth charts indicating different degrees of malnourishment. Second, nonavailability of basic supplies compounds inadequate training in reducing BHWs effectiveness and credibility in the village. Third, supervision is irregular, both in terms of professional help and moral support. Supervision problems can in turn be traced to the workload and priorities of midwives. The fundamental focus of the health care system is still curative. Midwives spend all morning in the clinic, and much of their home visit time following up clinic cases. Preventive health work focuses first on the environmental sanitation campaign. Nutrition education was never mentioned as a priority by the BHWs interviewed, and one symptom of less attention to this area is that, while barangay health centers keep figures for the number of malnutrition cases, most midwives - 29 - and BHWs did not know these figures off the cuff for their areas. Staff estimate that more than half of the barangays do not have an accurate figure for the number of children under 6, making any figures for the percentage of children weighed--one measure of the efficiency of the program--highly questionable. Family planning motivation is given least emphasis of all, as evidenced by the low number of acceptors, and the fact that BHWs mention this activity last or not at all when asked to describe their jobs. The PUSH project area presents a significantly different picture. First, BHWs are more highly motivated because they are better trained, and are paid and full time workers: they therefore feel, are, and are perceived as more professional than the volunteers outside the project. Second, preventive health care can be given more attention because of the greater staff resources. While midwives maintain much the same role as in the national system, BHWs are available to take over much of the nutrition surveillance work and supervision of unit leaders. Hence, weighing can be done more frequently, and a larger percentage of children is weighed. Accurate figures are kept on barangay health center walls for numbers of children and numbers of malnourished, and both BHWs and unit leaders have these figures in their heads, and can describe individual family cases. And because there is more time for supervision of unit leaders, the latter are clearer about their roles, and give greater priority to IEC activities than village volunteers outside the project area. 10. NUTRITION SERVICES IN JAVA, INDONESIA Population, health and nutrition services in Indonesia are particularly complex to analyze because of the variety of channels and levels of service, and because of the multi-purpose roles of most field workers. This review concentrates on nutrition outreach services in East and Central Java. Nutrition education is disseminated through schools, radio and television programs; through the national association of women's groups; and through out-reach services provided by two government agencies: the Nutrition Directorate of the Ministry of Health (MOH), and the BKKBN, the national family planning organization, a separate institution reporting directly to the President. Nutrition services in most villages 4re provided by either MOH or BKKBN, but there is no attempt to allocate Districts (abut 400 villages) or Subdistricts (about 20 villages) to a single agency. The fact that both agencies will be providing nutrition services to different villages in most Subdistricts3/ makes for coordination problems at the Subdistrict, District and Provincial levels. There are some overlaps between the agencies, and some gaps: some 3/ Both MOH and BKKBN will be operating in most villages, but in an MOH village, BKKBN will be providing family planning but not nutrition services, and in a BKKBN village, MOH will concentrate on health, while BKKBN will provide family planning and nutrition services. - 30 - villages, designated "self-help" villages, are without nutrition services from either agency. In order to give an impression of the range of service levels available, four nutrition services are compared below: the "normal" level of service provided by the Ministry of Health; an intensified level of service provided under the Bank-supported Nutrition Intervention Pilot Project (NIPP) in parts of East and Central Java; an intensified service provided by the Health Ministry in the Sapuran area of Central Java; and the BKKBN service. There are many other levels of service provided by the two major agencies, depending on local needs and resource availabilities. 10.1 Cadres. Common to all systems are the lowest level fieldworkers, volunteer villagers known as cadres. In a typical village of about 3,000 people with a "normal" level of MOH or BKKBN service, there will be 15-20 cadres, each serving a designated group of 25-30 families. Most cadres are literate, though this depends on the District: illiterate cadres cause performance problems in areas such as Madura District of E. Java, where the literacy rate in the population is 10%. In most areas, cadres receive no material incentives, although in the NIPP areas they get one new uniform a year. Mostly women, cadres are chosen for their influence in the village, their interest in nutrition, and their previous experience with some form of community development. Cadres are also used as field staff by many other agencies providing services at the village level. The limited pool of effective and enthusiastic volunteers in any village means that cadres will normally be involved in 3-4, and sometimes as mahy as 7 or 8 different voluntary activities ranging from family planning and health promotion through home garden, income-generating activity and religious development. The training received by nutrition cadres varies significantly. For "normal" MOH and BKKBN villages, the aim is 3-5 days pre-service training, though in practice this is often 2-3 days. In the NIPP program, cadres get a week's pre-service training, and in the project's longest-established area (Bojonegoro District in East Java) have had one formal refresher course of about a week in the four years since the project began. The most intensive training is given in the MOH's Sapuran program, where cadres have been getting about two 2-3 day refresher courses in nutrition a year. With a ratio of one cadre to every ten families, there is also a more intensive coverage of the population in this area. The cadres' main tasks are nutrition surveillance and nutrition education; in addition, they are responsible in most MOH areas for administering Vitamin A capsules 6 monthly to children under five, for the supply of ORT packets and for ORT education. The key task is monthly weighing of children under five (usually about 300 per village), carried out at Weighing Centers (converted rooms in existing buildings), of which there are usually 2-3 to a village. Cadres are allocated to a specific center. Weighing takes place on a set day of the month for each Center, and weights are recorded on growth charts kept by the mothers. For each village (and in the NIPP area for each Center) bar charts showing the following information are updated monthly: number of children under five; - 31 - number with growth chart; number weighed; number showing weight gain; number of pregnant and lactating mothers; and, in areas where there is a supplementary feeding program, number entering and number graduated from the program. Nutrition education is given by cadres at weighing time; wherever clients are met by chance in the village; when supplementary food is distributed; and during home visits. Cadres average 4-5 home visits a month, of varying duration, concentrating on mothers of malnourished children. It is difficult to estimate how much time cadres give to nutrition education, since they use their home visits to discuss all of the voluntary activities in which they are involved. While BKKBN cadres are only responsible for weighing and motivational activities, MOH cadres also give cooking demonstrations to mothers' groups. In the NIPP areas, these average one per two months per Nutrition Center; outside the NIPP, a tight budget for cooking materials reduces this to one each 3-5 months. Supplementary feeding is not given by BKKBN cadres. In many "normal" MOH villages, a supplement made from local village produce is given; in the case of the NIPP, the supplement is commercially manufactured locally, and administered from a Rehabilitation Center, one per village. In practice, these Centers apparently open only about two days a week, and most supplementary food is distributed to families at home, where it is not invariably eaten by children; it is not clear how much food is refused by or diverted from children. Entry to the feeding program follows three consecutive months of weight loss/no gain, and supplementary food is not normally given for more than 90 days. In practice in the NIPP area of Bojonegoro, few villages have more than 20 children receiving supplementary food, so one cadre might have 1-4 severe malnutrition cases to oversee. In the NIPP area, the duties of cadres go beyond that of nutrition surveillance, education and supplementary feeding. Their nutrition activities also involve the organization of local cooperatives and health insurance schemes, and home garden development. Various communication support materials are available for the cadres. In "normal" areas, each cadre, or at least each Weighing Center, would have a flip-chart of about 30 pages covering subjects like PCM; treatment of Vitamin A deficiency, goiter and anemia; weighing; growth charts; and home gardens. Posters on the same subjects are generally available. In the MOH intensive area at Sapuran, "action posters" have been in use in each home for two or three years. These picture a nutrition-related activity such as weighing at the top, and have a series of blank squares at the bottom, one of which is to be punched or marked by the mother each time she carries out the relevant action. In addition to weighing, there are action posters for appropriate food and supplements for pregnant and lactating mothers; for feeding infants with both left and right breast (local practice favors one breast); and for appropriate weaning foods for older infants. These posters were also introduced in the NIPP area at Bojonegoro in mid-1982. Tapes and additional written materials are also available in the NIPP area. The tapes focus on such subjects as the importance of weighing, Vitamin A and ORT. They are designed for use at weighing centers and - 32 - village meetings, the messages being interspersed with popular songs. The written material additional to the flip-chart consists of a booklet of about 30 pages on PCM; a book of about 200 pages on nutrition-related subjects; and a further book of about 150 pages on most of the same subjects, but presented in comic picture form. In the NIPP area, one complete "communications kit" is available per village. 10.2 Supervision. In most "normal" MOH villages, informal leaders coordinate the activity of the cadres. In the BKKBN villages, the leader of the cadres is usually the village official in charge of security, who is paid Rs 2,000 a month for superintending the reporting system. The influence of this official is one factor, among others, which have helped to make family planning more successful than most other services. In the NIPP villages, the cadres are led by a Village Assistant Nutrition Program Officer (VANPO), a volunteer who is often a teacher, and has a week's nutrition training. MOH cadre leaders receive no material incentives beyond uniforms, and this only in the NIPP area. A Rs 2,500 per month incentive paid to VANPOs over the last three years has been discontinued due to budgetary pressure, apparently without affecting performance. This is probably because the payment was in any case relatively insignificant (the wage for farm labor is Rs 700/day, plus two meals). Village cadres and their leaders are supervised by full-time government servants at the Subdistrict level and above. In many MOH areas, supervision is carried out by "intersectoral teams" based on the Subdistrict Health Center. In the intensive MOH area at Sapuran in Central Java, a Subdistrict of about 20 villages has six supervisors--two midwives, a sanitarian and three paramedics--who form six teams with their counterparts in agriculture, religious affairs, etc., and in these intersectoral teams supervise multipurpose cadres working for their different programs. It is difficult to estimate how long is spent on nutrition supervision, since supervisors in an intersectoral team have a basic training in each other's work, and will supervise a multipurpose cadre in a number of subjects at one meeting. It appears that on average a cadre will meet a supervisor to talk about nutrition once a month, for about half an hour. In the "normal" MOH areas, a nutrition cadre would not see a supervisor more than once in two months. MOH supervisors get 5-6 days nutrition training, of which one is spent in the field. In the NIPP area, supervision is more intensive, since two members of the Subdistrict Health Center team, designated Training Officer (TO) and Assistant Nutrition Program Officer (ANPO), spend most of their time on nutrition. ANPOs and TOs receive a Rs 7,500 per month bonus for their NIPP work, about 25% of their basic salary. They get two weeks' nutrition training, of which three days are spent in the field. It is hard to estimate how much supervision each NIPP cadre gets, since they receive visits from Health Center staff other than the ANPO and TO, and since the ANPO and TO are involved in supervision of both non-NIPP and NIPP villages--though they give more time to the latter (in a typical Subdistrict of Bojonegoro, there will be about 20 villages of which about four will belong to the NIPP program). Staff estimate that the average cadre will be visited by a supervisor once a month, about the same frequency as in the Sapuran intensive program. The village supervision - 33 - schedule is planned each month for the coming month at the Subdistrict level, but particular villages and groups of cadres are not visited regularly on particular days of the month. At the District level, Senior ANPOs and TOs visit each of their 20 Subdistricts about once in two months. Supervision of the average BKKBN village is more intensive than that of the average MOH village. Subdistrict-level supervisors, called PLKBs, have 2-3 weeks nutrition training, of which five days are spent in the field. On average, there is one PLKB for each 10,000 population, i.e., about six per Subdistrict (although West Java averages only one PLKB to 30,000 population), and on average cadres receive a supervision visit a month. However, supervision patterns vary from area to area and month to month. PLKBs are often divided into teams of three or so, with responsibility for a group of villages. Within that group, individual villages will receive greater or lesser attention according to the performance and motivation of their cadres. Whole teams may spend several consecutive days in a village where this is needed, so that visits can take the form more of mini-campaigns than of regular supervision. Though the frequency of supervision visits varies between services and areas, meetings with individual cadres follow much the same pattern. They last about 30 minutes, much of which is spent checking growth charts and the accuracy of performance reporting. Nutrition education is also given, but the short length of the visits means that this is general training and encouragement rather than specific review of and advice on individual family cases. Also common to all services and areas is a system of monthly cadre meetings at the Subdistrict level. About 100 cadres attend, 4-5 from each village, and the 4-5 hour session is devoted to motivation, training (in nutrition and in other cadre activities), and the drawing of a lottery whose winner rotates among the cadres. Commentary 10.3 Aggregate performance. The BKKBN's well-known rapid monitoring system for progress in family planning has been extended to include nutrition data. Monthly figures are collected for the number of children eligible for weighing, the number weighed, and the number showing weight gain, for both MOH and BKKBN villages (however, no separate calculations have been made for all MOH and all BKKBN villages--these would be simple and might show interesting performance differences). The percentage of children weighed is often in the 50-75% range, dropping to as low as 30%-50% in some areas at periods of peak agricultural activity. It is difficult to estimate how far these relatively low figures are the result of an inadequate management system, or teething troubles in a fast-expanding new program--the number of BKKBN villages where weighing has been introduced has increased exponentially from 0 in 1978 to 20,000 in mid-1983. Stratifying the data for children weighed according to how long villages have been in the weighing program would also be relatively simple and might throw light on this question. Unfortunately, the relatively low aggregate percentage of children weighed makes the second major performance measure--the number of children showing weight gain--almost meaningless, for the time being. This - 34 - is because this figure (which averages about half the children weighed in any one village) will miss out the many children who will be gaining weight in a given month, but cannot be shown as doing so since they were not weighed both that month and the previous month. 10.4 Loose management systems. Judged by T&V standards, the nutrition delivery systems in Java are very loosely managed. Field workers in most areas receive only 2-3 days pre-service training, and formal in-service training is rare. Part-time volunteer cadres with minimal training are often loaded with multiple tasks in different sectors. Supervision is relatively infrequent, irregular and unpredictable. Since cadres usually receive no material rewards, managers have no direct control over results. From a T&V perspective, the surprise is not that, for example, weighing rates are often low or that only 40-60% of cadres understand the growth charts properly (MOH staff estimate)--but that results are as good as they are. In the Indonesian systems, performance appears to be related less to the tightness of the management system than to the amount of encouragement given to the cadres. Tighter management of course makes a difference. The closer supervision in the NIPP area and the in-service training provided in the Sapuran area show up in a higher percentage of children weighed, better kept records, and better trained cadres. But the key to success in Sapuran--where in one area of a few villages almost all mothers keep their action posters for child weighing and feeding fully up to date--seems to be the steady stream of motivation and encouragement coming from an exceptionally enthusiastic staff at the Subdistrict Health Center. Similarly in the NIPP area of Bojonegoro--although it is difficult to draw definite conclusions--it seems that the sharp reduction in malnutrition achieved in the villages initially served by the program is not being matched as the program expands. Staff suggest that the initial results reflect partly the fact that the original villages have higher incomes and less scattered populations than average, but also the special attention and encouragement they received when the program was new and small. 10.5 Replicability. The Sapuran & NIPP nutrition systems prove that nutrition services can be provided by unpaid professionals with only loose managerial support. But the central role of leadership, motivation and encouragement in maintaining effective performance raises two important issues. The first issue is whether an exhortation-based system can be replicated widely within Indonesia and remain effective. It may be that the enthusiasm and support of more senior officials is itself a scarce resource available only in some Districts, or in smaller programs where the 'Hawthorne Effect' can operate. If this is so, then a tighter management system may need to substitute for higher levels of commitment in some areas, or as programs expand, since closer supervision and support may be needed to maintain minimum standards among the cadres (see also 12.5). 10.6 Transferability. The second issue raised by an "exhortation- based" management system is whether it can be transferred to other - 35 - countries. The remarkable efforts made by cadres--given their lack of incentives and support----are deeply rooted in traditions of social responsibility, community participation and compliance. Cultural respect for and compliance with authority in Java strongly assist the implementation of policies firmly supported by key leaders--as the response to support from the President and Provincial Governors for the national family planning program has proved. In addition, traditional community organizations are already in place at all levels of the society, which channel information about policies to communities, and which powerfully promote community solidarity, individual contributions for the common good, and conformity. Village councils which govern local affairs and are led by the village headman meet monthly on the anniversary of Indonesia's independence to discuss how guiding national principles apply to the village. This council is paralleled by a village women's association led by the headman's wife, which strongly influences mothers' attitudes to adoption of health care innovations. These groups at the village level have their counterpart at higher levels of the social hierarchy, so that, for example, the Governor's wife will be the leader of the Provincial Association of Women's Groups. There is therefore already in place a set of community organizations able to transmit policy, influence and support volunteers' efforts, and create demand for the services the volunteers provide. In Java, and in countries such as Korea and China, community organizations and cultural norms lift some of the burden of support and control from official government agencies, and amplify their efforts. Where the tradition of community participation is not as strong, loosely managed official service delivery systems may not be enough, and the case for a T&V type approach becomes stronger. This is at least suggested by the high drop-out rate (over 20%) of health cadres on the island of Lombok in Indonesia, where community organization and participation have traditionally been weaker. 11. COMPARING HUMAN RESOURCE USE The key variable in comparing the costs of different outreach management systems is likely to be human resource use. This is most commonly discussed in terms of dollars spent on manpower per head of population. By this measure, it is fairly straightforward to compare, for example, the cost of training VHWs for six weeks against five days in a given country. It is much less meaningful to compare dollar costs across countries, given exchange rate valuation and other pricing problems. Nor does the dollars per head measure capture different distributions of resources within systems: for example, two systems might distribute the same dollar costs quite differently between high paid supervisors, low-paid full time field workers, and part-time paraprofessionals. How resources are distributed within a delivery system may affect management performance as significantly as differences in the total quantities of resources available. - 36 - The tables which follow present a rough and ready format for comparing resource use in outreach systems, using the Tamil Nadu and Philippines cases. In the case of Indonesia, where both volunteers and paid field staff have a multipurpose role, it is hard to determine the amount of time put into PHN work, and this is not attempted. Table 11.1 summarizes the staffing ratios at the village level. Table 11.1 TAMIL NADU PHILIPPINES RATIOS HEALTH NUTRITION NATIONAL PUSH First. Line Supervisors: 1:5(N:MW) paid Village Workers 1:4 1:10 1:5 1:4(MW:BHW) Paid Village Workers: volunteers _ 1:35 1:8 Volunteers: population - - 1:100 1:100 Paid Village Workers: population 1:5,000 1:1,500 1:3,500 1:800 Note: N = nurse, MW = midwife T&V experience in agriculture suggests that once VHW supervision ratios become much worse than 1:8, effectiveness falls sharply. It is interesting that the single worse ratio (1:35) occurs in the Philippines national system, and that the ratios for the more successful PUSH and Tamil Nadu programs are close to or better than T&V's. Table 11.1 is misleading, however, to the extent that not all staff devote their full time to outreach work, either because they spend part of their time on curative clinic work/supervision (midwives and nurses in the Philippines, MPHWs and LHVs in Tamil Nadu) or because they are only part time health workers (volunteers). Table 11.2 makes a rough estimate of the % of staff time devoted to preventive/outreach work, and of the implications for outreach supervision ratios. - 37 - Table 11.2 TAMIL NADU PHILIPPINES HEALTH NUTRITION NATIONAL PUSH First line supervisors LHV:80 CNS:100 N:25 N:25 % of time on outreach paid village MW:25 workers MPHW:80 CNW:100 MW:25 BHW:100 First line supervisors: 1:5 1:10 1:20 1:20(N:MW) Revised paid village 1:16(MW:BHW) Supervision workers ratio equivalents Paid village workers: _ 1:140 1:8 volunteers Note: N=nurse, MW=midwife Tamil Nadu remains close to T & V's standard ratio; the Philippines national system ratios deteriorate sharply; PUSH still has tight supervision of volunteers, but inadequate back-up of midwives and BHWs by T&V standards. While the above table says something about the distribution of resources by administrative levels, it says nothing about costs. Table 11.3 attempts to compare resource costs without the confusion introduced by currency figures which may mean little. One cost unit is assigned to the salary paid to the lowest level civil servant in each country (midwives and MPHWS), and costs greater or lesser than 1 assigned to supervisors and paraprofessionals according to their rough salary costs relative to the basic civil servant salary in the country--so that a supervisor, for example, will cost 1-1/2 units. Total numbers and costs are shown for each 60,000 population. - 38 - Table 11.3 Staff Numbers and Costs for Each 60,000 Population U N I T C O S T S 1-1/2 1 1/2 1/4 0 TAMIL NADU Numbers of 4 CNS 12 MPHW - 40 CNW _ staff at 3 LHVS given unit cost level Total cost of 10 1/2 12 _ 10 _ staff at each level PHILIPPINES NATIONAL Numbers of 3.4 N 17 MW 75 BHW _ 600 staff at volun- given unit teers cost level Total cost of 5.1 17 _ _ _ staff at each level PHILIPPINES PUSH Numbers of 3.4 N 17 MW 75 BHW _ 600 staff at volun- given unit teers cost level Total cost of 5.1 17 37.5 _ _ staff at each level Total cost for Tamil Nadu is 32.5, for the Philippines national system 22.1, for PUSH 59.6. However, these costs include staff time spent on curative clinic work. Removing the proportion of costs spent on the latter (using the relevant percentages from Table 11.2) gives the following. - 39 - Table 11.4 U N I T C O TS TOTAL COSTS 1 1/2 1 1/2 1/4 Tamil Nadu 10.8 9.6 - 10 30.4 Philippines National 1.3 4.25 _ - 5.5 PUSH 1.3 4.25 37.5 _ 43 The differences in resource use are striking. Total PHN staff costs for the Tamil Nadu program are about 50% greater than for the Philippine national system at the village level (Table 11.3) and more than five times the financial resources are devoted to outreach work (Table 11.4). But the Tamil Nadu program also appears much more effective than the Philippines and Indonesian programs (Table 7.1). The PUSH program is by far the most intensive user of resources, yet appears less effective than Tamil Nadu. This reflects a basic dilemma of systems relying on large numbers of volunteers at the village level: if the volunteers are not well supported, performance drops (Philippines national); if they are well supported, the costs can become prohibitive (PUSH). Conclusions Little comparative analysis of performance and resource use in outreach management systems has taken place. The above format might be used to compare a larger sample of management systems to see whether certain resource use characteristics are associated with above average performance. The very limited assessment made here illustrates, first, one of the paradoxes of volunteer systems. These are often adopted for their cheapness, yet effective support and supervision of large numbers of volunteers can be prohibitively expensive, as the PUSH program shows. Second, the Tamil Nadu case suggests that tightly managed outreach systems using smaller numbers of paid workers may be more expensive than inadequately supported volunteers, but may also be much more cost-effective. Third, this, in turn, raises the question whether tight T&V type management systems are necessarily expensive (i.e., a critical and large mass of resources is needed to make them work) or whether resources can be in some way diluted without undue loss of performance. This question can only be resolved by experiment: the third part of this paper suggests some of the possible alternatives for experiment, and discusses the theoretical advantages and disadvantages of selectively targeting versus partially implementing T&V type systems. - 40 - PART THREE 12. ALTERNATIVE APPROACHES TO MAKING OUTREACH PROGRAMS MORE COST EFFECTIVE Variations on T&V Experience in agriculture suggests that T&V's success is due to simultaneous implementation of all its elements, since these are mutually reinforcing. It is therefore important to explore cheaper ways of implementing the full T&V approach. Two possible (and highly complementary) methods are by making greater use of paraprofessionals, and by targeting the outreach system more selectively on certain clients. 12.1 Using paraprofessionals. T&V in agriculture has always relied on full time civil servants as its village-level workers, arguing that staff must be full time if they are to become professional, and that they must be paid if they are to be expected to keep to the demanding schedules and standards of the system. Yet the Tamil Nadu program demonstrates that paraprofessional staff (CNWs) paid as little as a quarter of a junior health professional's (MPHW) salary can be effective if tasks are limited and close supervision is available. Using paraprofessionals could lower T&V's costs very substantially. The model T&V system developed in Part 1 of this paper would require 24 VHWs and three supervisors at the village level to serve 60,000 clients--totaling 28-1/2 cost units, and comparable with Tamil Nadu's cost of around 30 units (Table 11.4). Replacing the 24 VHWs with paraprofessionals could reduce the cost of the system to 10-1/2 units, assuming paraprofessionals were paid 25% of professional VHW salaries. While medical professionals might not welcome the creation of paraprofessional cadres, these have certain advantages for health administrators. They do not carry the pension and overhead costs of a full time civil servant, nor the right to permanent employment which goes with this status in many countries. The relative cheapness of paraprofessionals and their lack of civil service status may also be preconditions of the ability and willingness of local and municipal governments to share the costs of village health staff with the line ministry. Nevertheless, the effectiveness of paraprofessionals depends critically on close supervision, which may imply substantial additional costs for higher-level staff, unless the T&V system is selectively targeted. 12.2 Selectively targeting T&V. In agriculture, T&V systems have always been designed from the bottom up. Starting with the number of clients, and applying standard staff and supervision ratios, the number of extension staff has been calculated, and institution of the T&V system has been made conditional on the government making available the required full time extension staff. This may simply not be feasible in PHN, where resources are much scarcer. If it is accepted that there is a case for managing a few activities and clients well rather than many badly, then consideration should be given to designing T&V systems from the top down where resources are scarce, i.e., limiting the number of clients visited according to the staff time available, applying T&V staff and supervision ratios. - 41 - The potential home visit coverage could be worked out fairly simply for any given management system. In the case of the Philippines, for example, it might be reasonable to assume that each midwife should deal with her clinic work in the morning and spend all afternoons on home visit extension work (as opposed to curative follow-up visits). Applying the T&V system outlined in Part 1, each VHW would be able regularly to contact 240 mothers/families, or say about one-third of the families in her area of 3,500 population. This implies careful choice of target families, but need not, over time, mean inequitable coverage if the VHW periodically rotated the mothers in her visit groups as outlined below. This is something not practiced in agriculture, where the more automatic nature of diffusion allows ever more advanced technology to be introduced through the same contact farmers year after year. But since the priority for now in PHN is to get the basic innovations across to everyone, and diffusion is less automatic, it may make sense to teach one set of basic technologies and to vary the membership of the contact groups, leaving more advanced technologies to be sought by clients from doctors and hospitals. Unfortunately--and significantly, perhaps, in terms of reflecting lack of attention to the needs of outreach program managers--demographic data is available for the age structure of child populations, and for numbers of children per family, but not for the frequency of mothers with children of particular ages. It would be worthwhile to calculate for different countries at any point in time the percentage of the population who are pregnant mothers, the percentage who are pregnant and/or have a child under 1, the percentage who are pregnant and/or have a child under 2, etc. These groups could then serve as the basis for planning the coverage of direct outreach contacts, with priority going to mothers who are pregnant and/or have infants in the first year of life, since it is these who are likely to benefit most from education. Mothers would then be graduated from the visit program when their children reached one year or more, the length of time spent in the program depending on the outreach staff resources available. Taking the Philippines example and assuming that pregnant women make up 2.3% of the population (PHN Dept. has data for six countries which cluster around this figure), a VHW covering 3,500 people would have about 80 pregnant mothers in her area at any one time. Assuming that a VHW maintained each mother in an inner or outer group until her child was a year old, at the end of a year and nine months, counting in the newly pregnant, she might have at a guess about 200 mothers eligible for care, or anyway a number less than the 240 places in the inner and outer groups. This would leave some slack to allow for bunching of pregnancies at certain times of year, and slow adopters needing longer periods of education. As new mothers became pregnant after year two, they would replace mothers with children of a year or more who had already adopted the core messages and could take on responsibility for their own basic health care. From year two onwards, VHWs would have two responsibilities-- continuing monthly teaching sessions for mothers in the inner and outer groups, and what might be called a maintenance program for graduated mothers. The latter might be handled as follows. First, the requirement that inner group members educate two outer group members would build up the - 42 - promotional skills of mothers interested in passing on PHN information. Then, one or more mothers graduating from the inner group and showing particular interest and aptitude would be designated as informal leader and PHN promoter for some or all mothers graduating thereafter from the neighborhood group. Graduated mothers would then maintain touch with the VHW in three ways. First, they could attend any monthly group meeting and raise any problems encountered. Second, each month the leader of the graduated mothers would be expected to report to the VHW any problems or relapses in adoption among the graduated group. Finally, the VHW would occasionally visit graduated mothers during her home visit period after monthly group meetings, in order to check on informal leaders' reports and promotional work. The basic assumption of this rotational system is that given regular personal attention and teaching over an 18 month-2 year period, mothers would then be responsible for their family's basic PHN care. The group meetings proposed as part of the T&V system would be a key ingredient in developing the sort of informal leadership, group involvement and responsibility for care which might sustain interest in continued adoption of innovations among mothers graduated from the group program. Partially Implementing T&V A great deal of experience in agriculture indicates that it is extremely undesirable to implement only some elements of a system, whose success has depended on the mutual reinforcement among all its elements. Nevertheless, implementing some elements of T&V alone has the advantage of saving on resources and minimizing disruption to existing administrative procedures; both of these are factors which may encourage acceptance of management changes by senior health service administrators who might be hostile to the major reform that introduction of a full T&V system entails. If a partial approach must be adopted, two elements of T&V whose introduction would be relatively cheap and which would be relatively easy to graft onto existing systems, are in-service training and greater task concentration. 12.3 In-service training. Refresher training for village level workers was clearly needed in all three case study programs: most obviously in the Philippines and Indonesia, where volunteers had had only a few days pre-service training and received little in-service feedback on performance, but also in Tamil Nadu, where CNWs had had two months and MPHWs 18 months pre-service training, and in-service support was stronger. Where existing programs do offer refresher training to low level field staff, this conventionally takes place in a training center. There is no obvious justification for this since the basic skills needed by VHWs can best be taught by demonstration in the field rather than in theory in the classroom. The question is what forms of in-service training in the field might be more cost-effective than formal refresher courses. In terms of effectiveness, several theoretical advantages support the choice of regular half or one day training sessions at least monthly along the T&V lines suggested in Part 1. First, training would take place in villages and clinics so that actual problems and cases could be used for - 43 - demonstration purposes and feedback on performance. Second, the trainer, as the VHWs' regular supervisor, would be familiar with local needs, and could tailor training to them. Third, training VHWs in the peer group in which they work would foscer group motivation and problem-solving. Finally, and all-importantly for some programs, regular one or part day training sessions may be the only option for reinforcing the skills of volunteers who may be unwilling to sacrifice earnings while on full time refresher courses, especially if these are away from the village. In many countries, skill shortages would mean the unavailability of sufficient SMS to train VHWs directly. Under these circumstances, as outlined in 5.3 above, VHWs would routinely be trained by their own supervisors. Since most programs already allow for regular meetings with supervisors, which could be redesigned to include training, additional travel expenses for the lowest level staff need not be significant. Additional support would, however, be needed to reinforce the technical and communication/training skills of first line supervisors. Assuming that each supervisor spent a day a month being trained, that a supervisor was trained in a group of 8-10 of his peers, and that a SMS spent 15 days a month training, then one professional trainer could support about 150 supervisors, or about 1,500 field staff, assuming tight supervision ratios. A monthly training system of this type would depend on the mobility of SMS, and hence would increase transport costs relative to current practice.- But monthly training for a given number of supervisors might well be both more effective and cheaper--taking into account SMS staff time, travel and subsistence costs and training center overheads--than 6-monthly or annual refresher courses at a training center for ten times the number of lowest level field staff. The costs of alternative systems could be compared relatively straight-forwardly for any one country. The relative effectiveness of different training systems would need to be tested by experiment. But if, as with the T&V experience in agriculture, regular in-service training proved effective, it might be possible to reduce the resources used elsewhere in the system by shortening pre-service training for some categories of staff. Task concentration. Most outreach programs attempt to limit the number of tasks carried out by VHWs to some combination of the Alma Ata Conference's elements of primary health care. Few programs, however, limit tasks to the degree suggested in Part 1 of this paper, and many overload field staff with more information than they can cope with: the 400 pages of instructional/communications material developed for the volunteer cadres in Indonesia (see 10.1) is a good example. Where less resources are available for in-service training, support, and supervision than in T&V-type systems, there is a strong case for limiting VHW tasks even beyond those suggested in Section 2.2 above (basic curative care, nutrition education, neo-natatal tetanus immunization, ORT, family planning). This case is especially strong for systems using part time volunteers who cannot be expected to be professional at more than a handful of tasks. Where outreach program management is very weak, consideration might be given to the ultimate in task concentration--focusing all VHWs on a single task of high priority. There are clear technical limitations to - 44 - such an approach, since it would be ineffective in dealing with health problems with multiple causes (such as malnutrition), which may need multiple responses (nutrition education, ORT treatment, food and vitamin supplementation, for example). However, for much locally important morbidity--curing TB, preventing xeropthalmia and neonatanal tetanus, for example--concentrating on a single task for a certain period of time makes technical sense and would have important managerial advantages. Supervisors obliged to devote much of their time to clinic work could nevertheless reasonably be expected to see that VHWs performed one outreach task completely professionally, and this might be preferable to poor performance on a larger number of tasks. From a motivational standpoint, VHWs are likely to gain job satisfaction, and villagers confidence in the health service, if they see one task of major local concern effectively completed. Concentration on single tasks would be most effective if instituted together with a T&V-type in-service training system, which would allow VHWs to be retrained for a new task after the initial "campaign" was completed. As VHWs gradually became more professional, and clients more easily accepted recommendations, so more complex tasks could be introduced. The concept of an "incomplete" health care service of this type might be difficult for medical professionals to accept. Yet in practice, it is arguable that the current norm in loosely managed systems of providing a number of services, but of low quality and with inequitable coverage, leads to a more "incomplete" service than a system with more realistic aims. Alternatives to T&V While detailed discussion of alternative approaches to T&V is outside the scope of this paper, it is important to note that tightly managed outreach systems are not the only approach to PHN care, nor necessarily the most cost effective. The two main alternatives briefly discussed below are community organization approaches, and communication through the use of mass media. 12.4 Community organization. Approaches based on community organization may be distinguished from those based on community participation. While the sucess of T&V in agriculture and in PHN depends on the participation of farmers and mothers, the initiating agent of change is a paid worker at the village level who is responsible to the health service. In community organization approaches, however--like that of Indonesia--the key change agents are village volunteers backed by a community organization or committee: while these may be in some way supported by the health service, no one in the village is responsible to or under the control of the health service. It is worth noting that the T&V experience can offer little guidance on the management of such systems. Because of the demands made by the T&V approach, it has always been implemented through paid government servants. There is no reason to expect that parttime voluntary workers would be prepared or able to accept the system's discipline and tight controls. - 45 - USAID's experience with community organization, as summarized by the APHA, is that it has worked mainly where there is a focus on capital investments for improving water availability and quality, and where the major community involvement has been in a one time construction effort. It has been difficult to sustain community involvement in preventive and promotional care beyond an initial enthusiasm for setting up village health committees and selecting VHWs. The implication is not that community organization approaches do not work for preventive care, but that it is a mistake to see them as a cheap substitute for managerial efforts by governments. As the APHA notes, in many so called community projects, the community has had no major role in defining the content of the program, health service staff have little expertise in community development, and there is an almost complete absence of training and follow up support for village health committees. This suggests a case for a middle way between T&V and community organization approaches. Some health services may not be able to afford the permanent reliance on full time government extension workers often possible in agriculture. At the same time, communities need reliable, professional support if they are to become capable of looking after their own basic health needs. Unlike in agriculture, T&V in health must see the building up of self-sustaining community health care groups as a major part of its mission. This argues for combining a group-building focus such as that outlined in 12.4 above with the use of paraprofessionals, some of whom, as non-established civil servants, can be phased out as neighborhood group health care becomes stronger and the outreach program moves into a maintenance phase. 12.5 Mass media approaches. Radio is likely to be the most important mass medium, in view of the cheapness and wide availability of receivers, and its ability to reach illiterates. The medium is still in its infancy in PHN extension. A fair amount is known about radio's reach, and its potential for teaching new information: both are substantial. Less is known about radio's ability to change behavior, as opposed to add to knowledge. However, three facts are encouraging. First, scattered evidence suggests that radio can be an effective means of behavioral change in health care. Second, many current radio programs do not apply state-of-the-art IEC techniques, so that messages may not be presented as effectively as possible. More funds and professional support for radio are therefore certain to lead to performance improvements. Third, radio is extraordinarily cheap compared to face-to-face outreach contracts: it can therefore afford lower performance standards and still be cost-effective. Together, these factors argue strongly for widespread experiment with mass media approaches. In areas where budgets are too limited or populations too scattered for effective face-to-face extension, mass media approaches may have to be implemented on their own. However, it is doubtful whether radio should elsewhere be seen as a substitute for rather than as a complement to person-to-person extension. In broadcasts to individual clients, radio's limitations are that it cannot adapt its messages to individual listeners' circumstances; it cannot answer questions; and it cannot have the persuasive force of face-to-face discussion and demonstration. These are - 46 - important disadvantages where the barriers to adoption are as great as they are in PHN. In broadcasts to VHWs for training purposes, radio is limited for the same reasons and because it cannot demonstrate new techniques, or how to deal with clients, or feed back problems to researchers and administrators. There are three main ways in which radio might support and complement a T&V type extension service. First, at the introductory stage, radio could be used to build community understanding and willingness to participate, by explaining the goals and organization of the extension system, and the benefits offered by preventive services. Second, radio broadcasts to listener groups have been shown to be more effective than to individuals: VHWs could encourage neighborhood groups to gather to listen to evening broadcasts reinforcing extension messages currently being emphasized at monthly meetings. Third, radio's importance as a communicator and motivator might be particularly great in the maintenance phase of a T&V program as face-to-face contacts with VHWs became less regular and intensive for many clients. - 47 - ANNEX 1 page 1 of 2 THE TRAINING AND VISIT SYSTEM OF AGRICULTURAL EXTENSION Training and Visit has been adopted as the extension method in more than forty agricultural projects in about twenty countries, and has influenced many others. It has four mutually reinforcing principles: Singleness of purpose. Village-level workers (VLWs) spend their time learning from and advising farmers; they are not expected to deliver supplies, to organize credit, or to make reports. In India, where T&V has been adopted by thirteen states, moving to a single line of command and responsibility required extensive reform of a system that previously gave VLWs a multipurpose role in community development. Concentration on key tasks. VLWs are trained every two weeks, with emphasis on what is important for farmers to know about the current stage of cultivating their most important crops. This ensures the relevance of VLW work and does not overburden them with more information than they can handle. Regularity and predictability. There are rigid schedules for farm visits, training, and supervision, usually based on a two-week cycle. VLWs spend eight days a fortnight visiting eight separate farmers' groups, returning fortnightly to each group on the same day of the week. Each group consists of about 100 farmers, of whom about 10 are "contact farmers". Contact farmers are chosen to represent a spectrum of the farming community, and for their keenness to adopt innovations. While contact farmers must attend fortnightly meetings and give other farmers access to their fields and knowledge on request, they are not themselves expected to be extension agents. Diffusion occurs spontaneously as other farmers see the profitability of extension recommendations. On this basis, one VLW can provide extension services for about 800 farmers. The regularity of field visits means that farmers know when and where to find extension advice, and can hold the VLW accountable if he does not keep to his schedule. The regular schedule also means that extension supervisors know where all staff will be on any given day. Each supervising extension officer has no more than eight people under him, so that he can visit each of them for a whole day, not less than once a fortnight. Group training for VLWs, also rigidly scheduled, is carried out fortnightly in the field by Subject Matter Specialists and supervisors. In addition, monthly training workshops are held, which are attended by extension specialists, researchers, and supervisors. At these workshops, forthcoming advice is refined and disseminated and farmers' reactions are reviewed. - 48 - ANNEX 1 page 2 of 2 Face-to-face communication and feedback. T&V recognizes the importance of effective communication links between researchers, extension staff, and farmers. Face-to-face meetings, held as far as possible in farmers fields, replace the common emphasis on time consuming and unproductive written reporting. This allows quick and effective feedback and adjustments where necessary. VLWs do no more than keep a simple diary of key messages and farmers reactions for their own reference. Meetings with farmers, in which supervisors participate, are the main means not only of monitoring field staff performance but also of defining priorities for agricultural research. - 49 - ANNEX 2 page 1 of 2 MESSAGE SELECTION SCALE4/ Behavioral Design Principles 1. COMPATIBILITY WITH EXISTING PRACTICES O Totally incompatible 5 Already widely practiced 2. APPROXIMATIONS O Nothing like this now exists 5 Several existing behaviors are similar 3. DIVISIBILITY O Not divisible into discrete parts 5 Easily broken down into segments which can be taught sequentially 4. PERCEPTIBLE POSITIVE CONSEQUENCES FOR THE MOTHERS O No perceptible consequences which the mother would logically associate with behavior 5 Clear, positive, and immediate consequences for correct performance of behavior 5. PERFORMANCE COSTS O Requires some unavailable or expensive outside commodity, or demands some unrealistic level of effort 5 Can be performed with existing resources 6. FREQUENCY O Must be done with great regularity or benefit is marginal 5 May be done infrequently and still have significant value 4/ Produced by the Academy for Educational Development on contract to USAID's Offices of Education and Health for programs in Honduras and the Gambia. - 50 - ANNEX 2 page 2 of 2 7. PERSISTENCE 0 Requires continuous compliance over a long period of time 5 Can be accomplished quickly 8. OBSERVABILITY O The behavior itself is very difficult for an outsider to observe 5 The behavior can be readily perceived and reinforced by outsider 9. SALIENCE 0 Has a modest positive effect on health problem 5 Has a potentially dramatic positive effect on health problem - 51 - ANNEX 3 page 1 of 3 ORT BEHAVIORAL PROFILE5/ Refers to rural mothers with children suffering from diarrhea but no serious signs of dehydration. Motivation Feel need for medication - Knows that fluid loss can be replaced - Recognizes child as being sick with diarrhea - Recognizes sickness is severe enough to require fluid replacement - Believes diarrhea is an illness which can be treated successfully - Believes in her own ability to make/give fluid Feel need for packet - Believes packet works to replace fluid lost and save child's life - Believes packet works better than other alternatives - Feels she can afford packet - Feels getting the packet is worth the bother Feel need to feed child during episode - Knows good food and full diet (including breastfeeding) should be continued - Believes child with diarrhea needs and will accept food Procurement Know about packet - Knows where to go - Knows what to ask for - Knows how much it costs 5/ Produced by the Academy for Educational Development on contract to USAID's Offices of Education and Health for programs in Honduras and the Gambia. - 52 - ANNEX 3 page 2 of 3 Purchase packet - Have the money - Packet must be available where she goes to get it - Have the time/capacity to go and get it Purchase correct amount of packets Acquire appropriate mixing vessel - Vessel must be present in the house - Mother must recognize vessel as being appropriate - Vessel is not being used for something else when needed - Vessel must be clean Mixing Open packet without spilling salts Add salts from packet to vessel - Add all the packet - Add only the contents of the one packet - Add nothing else Add water to same vessel - Add correct volume (fill the vessel without spilling) - Add as clean water as possible (tea) - Add nothing else Stir salts to dissolve - Recognize when the salts are dissolved - A utensil for stirring must exist - Stirring utensil must be relatively clean Do not boil mixture - 53 - ANNEX 3 page 3 of 3 Administration Use a small spoon to give child the one litre mixture over the next 24-hour period along with water, breastmilk, and juices. Continue this regimen for three days or until the diarrhea stops. - Have small spoon or bottle available for bottle-fed babies - Believe that child can consume the recommended volume - Remember to give the child small amounts over time - Recognize what a 24-hour period is - Store solution in place where it will not be disturbed or forgotten - Believe that breastfeeding is all right when child has diarrhea - Have breastmilk, or give water or juices - Give sufficient volume of water and juices - Believe child is capable of taking all these liquids - Go and get another packet, or buy packets at once - Evaluate whether child a) still has diarrhea, but is not becoming dehydrated, b) is becoming dehydrated, or c) has no more diarrhea - Have confidence in her own evaluation of child's status - Feed child regularly and with appropriate diet in spite of diarrhea External help If diarrhea continues after three days, if continuous vomiting occurs, or if child cannot or will not drink, seek help - Recognize that diarrhea continues - Believe that health worker can help - Be able to take child to health worker - Perceive vomiting as continuous - 54 - ANNEX 4 page 1 of 3 SAMPLE QUESTIONS FOR DETERMINING WHETHER AN OUTREACH SYSTEM HAS THE PREREQUISITES OF EFFECTIVENESS Talking to supervisors is less valuable than talking to VHWs. Talking to VHWs is often less valuable than talking to mothers. The higher the level, the more likely the interviewer will hear what should happen, rather than what happens. Talking to VHWs (i) Does she know her role? How long have you been a VHW here? What are your tasks as a VHW? Which are the most important ones? What problems do you have in your work? Can you explain how this growth chart works? When was your initial training and for how long? How much was in the field and how much in the classroom? What were the most useful things you learned? Do you have work problems you didn't learn about? Did you learn things you don't use? What in-service training have you had? What did you learn? (ii) Does she know her clients? How many people are there in your area? How many pregnant mothers? How many children under six? How many 3rd and 4th degree malnutrition cases? What records do you keep? Tell me about the families who have children with 3rd and 4th degree malnutrition. (iii) What does she actually do with her time? How did you spend your time yesterday? Was that a typical day? What do you normally do? Do you spend a fixed time in the clinic and a fixed time on home visits? Do you have a regular schedule for home visits? Is it written down? Do you spend more time in some parts of the village than others? How do you choose which families to visit? How long do you stay when you visit (describe yesterday)? How many families do you visit regularly? (how regularly?) and how many occasionally? (how often?) - 55 - ANNEX 4 page 2 of 3 (iv) Is her extension work professional? What did you talk about with the families you visited yesterday? --Did you give them any advice? Did they accept? 3 What are the most common things you advise when you visit? How often do you check whether your advice is being followed? Which recommendations are mostly accepted and which are hard to get : mothers to adopt? If you can't initially persuade a mother, what do you do? What kind of eating/cooking recommendations have you been giving? Tell me about some mothers who are not adopting these. What is the .;, problem and what do you do about it? (v) Is she adequately supervised and supported? Do you have regular meetings with your supervisor? With other VHWs? How often? .,Tell me what was talked about at the last meeting. Did the supervisor give instructions/training? How long for? . Did she check your records? How long for? Did you or the other VHWs raise any problems or ask any questions? What were they, and what did the supervisor say? Was what happened at that meeting typical, or different from most meetings? When did your supervisor last come with you on a home visit? Tell me what you did, and what the supervisor said and did. How often does your supervisor come with you to visit families? How long for? Does she have a regular schedule for this? When was the time before last that your supervisor came on a home visit with you? Talking to Mothers (i) What contact does she have with the extension service? Is there a health worker who lives in or comes to this neighboorhood? What is her name? Has she visited your house? When was the last time she came? And the time before that? Does she come regularly, and on a particular day? Is there any mother's group in this neighborhood? Do you belong? Does the health worker talk to the group? How often? What types of health problems have you and your children had in the last six months? For each type of problem, who has been most help to you--the government health worker, or someone else in the village? (ii) How useful has the service been? How long did the worker stay when she last came to your house? What did she do and what did you talk about? Did she give you any advice? What? Are you following it? - 56 - ANNEX 4 page 3 of 3 What usually happens/what other advice have you been given when the health worker comes? Have you discussed your family's health problems with the worker? What did she advise you? What did you do? When did one of your children last have diarrhea? What did you do? Why was that? Are any of your children malnourished? How do you know? What did the health worker suggest? Are you doing it? (If applicable) Do you keep a growth chart for them? Can you tell me what these lines and markings are for? The Order of Questions is Important (i) VHWs and mothers often cannot answer general questions, or will talk about what is supposed to happen, rather than what actually happens. So "What did you do during your home visits yesterday afternoon?" rather than, or before "What do you normally do on home visits?" (ii) Open questions should be asked before predisposing ones. So "What happened during you last meeting with your supervisor?" rather than/before "Did you tell her about any problems you have in your work? " before "What advice did she give you on your problems?" W orid Bank value of these findings and provides Demographic Aspects of valuable insights into possibilities for Migration in West Africa- Publications implementing mass programs tor K. C. Zachariah and others needy people in villages throughout of Related the world. Volume 1 Staff Working Paper No. 414. September Interest 1980. 369 pages (including statistical an- Volume 1: Integrated Nutrihion nexes, bibliography). and Health Care Stock No. WP 0414. S15. Arnfried A. Kielmann and others This volume provides detailed data Volume 2 suggesting that synergism between Staff Working Paper No. 415. September malnutrition and infection is probably 1980. 391 pages (including statistical an- the greatest cause of mortality, mor- nexes, bibliography). The African Trypanosomiases: bidity, and retarded growth and devel- Stock No. WP 0415. 315. Methods and Concepts of over a period of four years, vixagers (These Working Papers are background Control and Eradication in received nutrition care, general health studies for Migration in West Afnca: Relation to Development care to control infections, or both. Dra- Demographic Aspects, described in this C. W. Lee and J. M. Maurice matic improvements, including a 40%- section.) Here is a practical cost-benefit ap- 50% decline in mortality, a 20% reduc- Economic Motivation versus proach to an age-old problem affecting creases in heightiando grbihty and in- City Lights: Testing humans and livestock alike, the Afri- tion, detailed information on costs is Hypotheses about Inter- can Trypanosomiases. Describes new to,dtie nomto ncssi tchniues thatnofserts.Detsbes contrl presented that permits the most com- Changwat Migration in technoques that offer tsetse control plete analysis of cost-effectiveness and Thailand withot detroyig gae anials.program relevant costs and benefits Fe rodadSsnH Also summan.zes currelnt research in yet avaLlable in this kind of field re- Fred Anold and Susan H. genetic control, the use of traps and search. The study focuses directly on Cochrane screens, attractants, and pheromones. practical program implications and Staff Working Paper No. 416. September Technical Paper No. 4. 1983. 107 pages. ways in which such integrated services 1980. 41 pages (includingfootnotes, refer- ISBN 0-8213-0191-8. Stock No. BK 0191. can be applied under field conditions. ences). 55. The Johns Hopkins University Press. 1984. Stock No. WP 0416. 33. Analyzing the Impact of 288 pages. Experiments in Family Health Services: Proect LC 82-23915. ISBN 0-8018-3064-8 Stock Planning: Lessons from the Health Services PoetNo. JH 3064. S24.50. Dvlpn ol Experiences from India, Developing World Ghana, and Thailand Roberto Cuca and Catherine S. Rashid Faruqee Pierce Staff Working Paper No. 546. 1982. 44 Volu ae II. Integrated Family A comprehensive review of experi- pages. Planning and Health Care mental efforts in the developing world ISBN 0-8213-0117-9. Stock .o. WP 0546. Carl E. Taylor and others to determine more effective ways of I3S To village people, politicians, and in- providing family planning services. temational health planners, health and The Johns Hopkins University Press, 1978. family planning have always seemed 276 pages (including bibliography, index of NEW to fit naturally together. But in the experiments). early 1960s, when international aware- LC 77-16596. ISBN 0-8018-2013-8, Stock Child and Maternal Health ness of the social and economic conse- No. IH 2013, $19.50 hardcover; ISBN 0- Services in India: The ~quences of surging population growth 8018-2014-6. Stock No. JH 2014. 38.95 Services in India: The moved family planning into a position paperback. Narangwal Experiment of high priority, some international What can primary health care and agencies began to advocate separation Family Planning Programs: An family plannng do for women and of family planning from health serv- Evaluation of Experience children in the poor and deprived ices. In international policy discussions Roberto Cuca areas of the world? Some of the most the question continues to be impor- specific evidence available today to tant. This volume analyzes this ques- Staff Working Paper No. 345. 1979. 146 support the benefits of these services tion and provides arguments and evi- pages (ncludig 2 annexes, references). is contained in these two studies, dence to support integration of health Stock No. WP 0345. S5. which represent the findings of re- care and family planning; it outlines search carried out during 1967-74 in the purposes underlying the research Fertility and Education: What twenty-six villages in Punjab, India. in this area; and it proposes policy Do We Really Know? Member of te resarch taff,whichquestions regarding the effectiveness, Susan H. Cochrane started with 15 people and grew to efficiency, and equity of suchaine- A model identifving the many chan- about 150 by the end of the project, gron. nels through which education might spent manv years working with, and The Johns Hopkins University Press. 1984. sharing the lives of, villagers. The 256 pages. depth of understanding that came LC 83-23915. ISBN 0-8018-2830-9.Stock Prices subject to change without notice from this type of sharing enhances the No. JH 2830. S22.50. and may vary by country. act to determine fertility and a review Health Issues and Policies in Kenya: Population and of the evidence ot the relation between the Developing Countries Development education and the intervening vana- Fredrick Golladav (See descnpton under Countrv bles in the model that affect fertilitv. The Johns Hopkins University Press. l979. Staff Working Paper No. 412 1980 55 Studies listing.) 138 pages (including bibliography, index). pages LC .78-26070. ISBN 0-8018-2140-Z. Stock Stock No WP 0412. 53. No IH 2140, 56.95 paperback. Health, Nutrition, and Family Migration in West-Africa: Planning in India: A Survey of Demographic Aspects Experiments and Special K. C. Zachariah and Julien Conde Fertility and Its Regulation in Projects The first study of the large-scale move- Fertilitys . ment of people in nine West Afncan Bangladesh Rashid Faruqee and Ethna Johnson countries. Discusses the volume and R. Amrhin and Rashid Faruqee Staff Working Paper No. 507. 1982. 108 direction of intemal and external flows Staff Working Paper No 383. 1980. 54 pages (including references) and the economic and'social character- pages (including references). Stock No. WP 0507 55. istics of migrants. Stock No. WP 0383. 53. A joint World Bank-OECD study. Oxford Infant and Child Mortality as a University Press, 1981. 166 pages (includ- Determinant of Fertility: The ing 22 maps, bibliography, index). Health Policy Implications LC 80-21352. ISBN 0-19-520186-8, Stock Fredrick Golladay, coordinating Susan Hill Cochrane and K. C. No OX 520186, 529.95 hardcover; ISBN author Zachariah 0-19-520187-6, Stock No OX 520187. Draws on experience gained from An illustrative analvsls that suggests 58.95 paperback. health components of seventv World infant mortality mav be an important Bank projects in forty-four countries component of a fertility reduction pro- between 1975 and 1978. Emphasizes gram in countries where mortality iS Population and Family the disproportionately high expendi- high and few couples are able to have tures incurred on curative medicine, the number of surviving children they Planning in Bangladesh: A maintenance of expensive hospitals, desire. Study of the Research and sophisticated training of medical World Bank Staff Workzng Paper No. 556. M4ohammad Alauddin and Rashid personnel at the cost of preventive 1983. 44 pages. Faruqee care for the majoritv of the people. Points out that low-cost health care ISBN 0-8213-0147-0. Stock No. WP 0556 Reviews major studies on family plan- svstems are feasible and recommends 53 ning and on fertility trends, profiles, that the Bank begin regular and direct and detemmants. Evaluates results of lending for health, in addition to hav- Integrating Family Planning such studies and critiques their meth- ing health components as part of pro- with Health Services: Does It odology and application. Underscores jengshealother co nentors. of7 need for continued study and suggests jects iother sectors. Help? directions for future research to im- Sector Policy Paper. 1980. 90 pages (in- Rashid Faruqee prove the Bangladesh population cluding 8 annexes, 4 figures, map). Staff Working Paper No. 515. 1982. 47 problem. Stock Nos. BK 9066 (Arabic), BK 9067 pages. World Bank Staff Working Paper No. 557. (English), BK 9068 (French), BK 9069 iSBN 0-8213-0003-2. Stock No. WP 0515. 1983. 176 pages. (Spanish). S5. S3. ISBN 0-8213-0150-0. Stock No. WP 0557. 55. I I II\Visit the World Bank lBookstore Population and Poverty in the i i ~~~KStetwhe youa C.i Developing World I I = | | : Washington, AC. Nancy Birdsall Staff Working Paper No. 404. 1980. 96 - i I Street Xj pages (including 2 appendixes, bibliog- raphy). ASi Ei1 ( \Stock No. WP 0404. S3. H Street H Street 7 | I Lafayette Square Population Policy and Family L J J j Planning Programs: Trends in Avenue Policy and Administration Kandiah Kanagaratnam and G Street The White House Catherine S. Pierce r~ Houe )lStaff Working Paper No. 447. 1981. 80 _____I___I__I__I__._ pages (including bibliography, appendixes). World Bank Bookstore Stock Nlo. WP 0447. S3. NEW published as World Populatzon Projec- falling mortality, and urbanization- tions: Short- and Long-term Projections by factors that tend to lower fertility else- Short-term Population Age and Sex for All Countries, with Re- where. Calls for a viable population Projection, 1980-2020 and Long- lated Demographic Statistics policy and programs appropnate to term Projection, 2000 to 1983. 391 pages. the culture. Stationary Stage by Age and ISBN 0-8213-0355-4. 530 paperback. Staff Working Paper No. 559. 1983. 116 Sex for All, Countries of the pages. Wex forld All' Countries oftheNEW ISBN 0-8223-0152-7. Stock Nos. WP World '' 0559. 55. Mv T. Vu, under the supervision Rapid Population Growth in of K. C. Zachariah Sub-Saharan Africa: Issues and Regional Aspects of Family This repori'~ives detadled population Policies Rgoa set fFml projections by age and sex for each Rashid Planning and Fertility Behavior country at five-vear intervals from 1980 Rashid Faruqee and Ravi Gulhati in Indonesia to 2020 and at twenty-five year inter- No other country has higher fertility Dov Chernichovsky and Oev Astra vals from 2000 to the year in which than Kenya and its neighbonng coun- Meesook population b;ecomes stationary in each tnes in Sub-Saharan Africa. This country. The implied fertility and mor- Working Paper examines the reasons Staff Working Paper No. 462. 1981 62 tality measures'are also given. Revised for fertility rates staying high, even pages (including appendix. references). annually. The 1984 edition will be rising in the face of greater education, Stock No. WP 0462. 53. The World Bank Publications Order Form SEND TO: YOUR LOCAL DISTRIBUTOR OR TO WORLD BANK PUBLICATIONS (Sn the other side of this form.) P.O. BOX 37525 WASHINGTON, D.C. 20013 U.S.A. Date Name Ship to: (Enter if different from purchaser) Title Name Firm Title _ Address Firm City StateP Postal Code. . Address Country Telephone l_I City State Postal Code Purchaser Reference No. Country 'Telephone Check your method of payment. Enclosed is my O Check J International Money Order O Unesco Coupons O International Postal Coupon. Make payable to World Bank Publications for U. S. dollars unless you are ordering from your local distributor. Charge my O VISA O MasterCard O American Express O Choice. (Credit cards accepted only for orders addressed to World Bank Publications.) Credit Card Account N?umber Expiration Date Signature C Invoice me and please reference my Purchase Order No. Please ship me the Items listed below. cutomar InteStl SW INha Autbe/ Tide Routing Cob Quasnliy Unit Pri Tow Amounl I All prices subject to change. Prices may vary by country. Allow 6-8 weeks for delivery. Subtotal Cost S 'btal copies - Air mail surcharge if desired (52.00 each) S Postage and handling for more than two complimentary items (52.00 each) S Tbtal S 10040541 Thank you for your order. Distributors of World INDONESIA Attn: Mr. J.M. Hemandez Bank Publications Attn: Mr. Bambang Wahyudi Madrid JI, Dr. Sam Ratulangi No. 37 SRI LANKA AND THE MALDIVES Jakarta Pusat Lake House Bookshop ARGENTINA IRELAND Attn: Mr. Victor Walatara Carlos Hirsch, SRL, TDC Publishers 41 Wad Ramanayake Mawatha Attn: Ms. Monica Bustos Attn: Mr. James Booth Colombo 2 Florida 165 4° piSO 12 North Frederick Street SWEDEN Galeria Guemes Dublin 1 ABCE Fritzes Kungl, Hovbokhandel Buenos Aires 1307 GUINEA JAPAN Attn: Mr. Eide SeSerback AUSTRALIA, PAPUA NEW GUINEAnRe gata 1T o 65 FI}I, SOLOMON ISLANgS, ~Eastem Book Service Reerngpsgatan12Bo165 FIJI, SOLOMON ISLANDS, Attn: Mr. Terumasa Hirano S-1 2TStockholm WESTERN SAMOA, AND 37-3, Hongo 3-Chome, Bunkyo-ku 113 SWITZERLAND VANUATU Tokyo Librairie Pa ot The Australian Financial Review Tokyo L rairie Pa yot Information Service (AFRIS) KENYA Attn: Mr. Henri de Perrot Attn: Mr. David Jamieson Africa Book Services (E.A.) Ltd. 6, rue Grenus 235-243 Jones Street Attn: Mr. M.B. Dar 1211 Geneva Broadwav P.O. Box 45245 TANZANIA Sydney, aSW 20001 Nairobi Oxford University Press Bydney, KOE,RPBI FAttn: Mr. Anthony Theobold BELGIUM KOREA, REPUBLIC OF Maktaba Road, P.O. Box 5299 Publications des Nations Unies Pan Korea Book Corporation Dar es Salaam Attn: Mr. Jean de Lannoy Attn: Mr. Yoon-Sun Kim av. du Roi 202 P.O. Box 101, Kwanghwamun THAILAND 1060 Brussels Seoul Central Department Store, Head Office 106 Bruss MALAYSIA Attn: Mrs. Ratana CANADA ~ ~ ~ ~ ~ ~~Uiestyo aly opeaie306 Silom Road Le Diffuseur Bookshop LtdM Bangkok Attn: Mrs. Suzanne Vermette Attn: Mr. Mohammed Fahim Htj Thailand Management Association Quebec Yacob Attn: Mrs. Sunan P.O. Box 1127, Jalan Pantai Baru 308 Silom Road COSTA RICA Kuala Lumpur Bangkok Librenra Trejos MEXICO UNITED KINGDOM AND Attn: Mr. Hugo Chamberlain Calle 11-13, Av. Fernandez Guell INFOTEC NORTHERN IRELAND San Jose Attn: Mr. Jore Cepeda Microinfo Ltd. San Lorenzo 153-11, Col. del Valle, Attn: Mr. Roy Selwyn DENMARK Delee. Benito Juarez Newman Lane, P.U .Box 3 Sanfundslitteratur 03100 Mexico, D.F. Alton, Hampshire GU34 2PG Attn: Mr. Wilfried Roloff NETHERLANDS England Rosendems Alle 11 MBE BV UNITED STATES DK-1970 Copenhagen V. Attn: Mr. Gerhard van Bussell The World Bank Book Store EGYPT, Arab Republic of Noorderwal 38, 600 19th Street, N.W. Al Ahram 7241 BL Lochem Washington, D.C. 20433 Attn: Mr. Sayed El-Gabri NORWAY (Postal address: P.O. Box 37525 Al Galaa Street Johan Grundt Tanum A.S. Washington, D.C. 20013, U.S.A.) Cairo Attn: Ms. Randi Mikkelborg Baker and Taylor Company FINLAND P.O. Box 1177 Sentrum 501 South Gladiola Avenue Akateeminen Kirjakauppa Oslo I Momence, Illinois, 60954 Attn: Mr. Kari Litmanen PANAMA 380 Edison Way Keskuskatu 1, SF-00100 Ediciones Libreria Cultural Panamena Reno, Nevada, 89564 Helsinki 10 Attn: Mr. Luis Femandez Fraguela R. 50 Kirby Avenue FRANCE Av. 7, Espana 16 Somerville, New Jersey, 08876 World Bank Publications Panama Zone 1 Commerce, Georgia 30599 66, avenue d'lena PHILIPPINES 75116 Paris National Book Store Beman Associates GERMANY, Federal Republic of Attn: Mrs. Socorro C. Ramos Lanham, Maryland, 2076H y UNO-Verlag 701 Rizal Avenue Lahm ayln,27 Attn: Mr. Joachim Krause Manila Blackwell North America, Inc. Simrockstrasse 23 SAUDI ARABIA 1001 Fries Mill Road D-5300 Bonn I Jarir Book Store Blackwood, New Jersey 08012 HONG KONG, MACAU Attn: Mr. Akram Al-Agil Sidney Kramer Books Asia 2000 Ltd. P.O. Box 3196 1722 H Street, N.W. Attn: Ms. Gretchen Wearing Smith Riyadh Washington, D.C. 20006 6 Fl., 146 Prince Edward Road SINGAPORE, TAIWAN, BURMA United Nations Bookshop Kowloon Information Publications Private, Ltd. United Nations Plaza INDIA Attn: Ms. Janet David New York, N.Y. 10017 UBS Publishers' Distributors Ltd. 02-06 1st Floor, Pei-Fu Industrial VENEZUELA Attn: Mr. D.P. Veer Building Libreria del Este 5 Ansari Road, Post Box 7015 24 New Industrial Road Attn. Mr. Juan Pericas New Delhi 110002 Singapore Avda Francisco de Miranda, no. 52 (Branch offices in Bombay, Bangalore, SPAIN Edificio Galipan, Aptdo. 60.337 Kanpur, Calcutta, and Mvfadras) Mundi-Prensa Libros, S.A. Caracas 1060-A I HG 3881.5 .W57 W67 NO.662 C.3 I HEAVER, RICHARD, 1952- ADAPTING THE TRAINING AND --j T- TT-Sq YS.q T- - r n r A KT_ YV_ The World Bank Headquarters European Office Tokyo Office 1818 H Street, N.W. 66, avenue d'lena Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome Telephone: (202) 477-1234 Telephone: (1) 723-54.21 Chiyoda-ku, Tokyo 100, Japan Telex: WUI 64145 WORLDBANK Telex: 842-620628 Telephone: (03) 214-5001 RCA 248423 WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTONDC ISSN 0253-2115/ISBN 0-8213-0393-7