University of Southern California Dissertations and Theses

An analysis of traditional birth attendant (TBA) programs in selected third world countries

(USC Thesis Other)

An analysis of traditional birth attendant (TBA) programs in selected third world countries

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content AN ANALYSIS OF TRADITIONAL BIRTH ATTENDANT (TBA) PROGRAMS IN SELECTED THIRD WORLD COUNTRIES by Richard Powell Grieser A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA | I In Partial Fulfillment of the Requirements for the Degree | DOCTOR OF PHILOSOPHY | (Education) : December 1985 UMI Number: DP25048 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. DissertaÈion R ubi shgnq UMI DP25048 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106 - 1346 UNIVERSITY OF SOUTHERN CALIFORNIA T H E G R A D U A T E S C H O O L U N IV E R S IT Y P A R K L O S A N G E L E S , C A U F O R N IA 9 0 0 8 9 T H E G R A D U A T E S C H O O L r s ^ I * * L U N IV E R S IT Y P A R K f c I 5 L Ih, This dissertation, written by Richard Powell Grieser under the direction of h..l9... Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillm ent of re quirements for the degree of D O C T O R O F P H IL O S O P H Y Date November 19, 1985 DISSERTATION COMMITTEE -C:. Chairperson Fh.b- £d ■ ACKNOWLEDGMENTS Especial thanks to Dr. William Rideout and Dr. Alfred Neumann i I for their patience and perseverance, and to my wife, Mona Yazdi I I Grieser, for general support; these three people were invaluable to the study. Thanks also to Anne Maclean Heasty for editing services | and to Joan T. Benefiel for production services. DEDICATION This study is dedicated to Baha*u*llah, Founder of the Baha’i faith; to my parents, Charles Richard Grieser and Margaret Powell Grieser; and to my wife, Mona Yazdi Grieser. I I iii TABLE OF CONTENTS EM ê ACKNOWLEDGMENTS............................................. i i DEDICATION.................................................. iii LIST OF TABLES............................................. vi LIST OF FIGURES....................................... vii Chapter I. INTRODUCTION......................................... 1 The Traditional Birth Attendant Controlling Population Growth Training Traditional Birth Attendants II. STATEMENT OF THE PROBLEM AND METHODOLOGY............ 20 Purpose of the Study Delimitations Assumptions Systems Model Methodology III. REVIEW OF THE LITERATURE ....................... 41 Strategies for Primary Health Care Training of Traditional Birth Attendants IV. IMPLEMENTATION CONSIDERATIONS....................... 106 Preimplementation The Implementation Perspective Implementation Problems in Donor Agencies Adaptive Performance and Incentive Problems in Donor Agencies V. NONFORMAL EDUCATION 141 | Institutionalization ' Characteristics Postadoption Evaluation Supervision and Management I Indonesian Reform Effort ! i VI. CASE STUDIES 179 ’ Background Data Presentation and Analysis ' I V TABLE OF CONTENTS (Continued) Page Chapter VII. FINDINGS, CONCLUSIONS, IMPLICATIONS, AND RECOMMENDATIONS................................ 251 Findings in the Existing Literature Conclusions and Implications Rec ommendat ions REFERENCE NOTES............................................. 272 BIBLIOGRAPHY................................................ 274 APPENDIX— Suggested Questions for Study of Knowledge, Attitudes, and Practices of TBAs................... 290 V I _ _ _ i LIST OF TABLES Table Page 1. Names of Traditional Midwives........................... 5 2. Conflict-of-Values Chart................................ 43 3. Should TBAs Be Trained? 73 | 4. Sample Risk Factors..................................... 93 5. Evaluation of the Program 100 | 6. Primary Health Care Projects: Population.................. 104 7. Evaluation Strategies................................... 161 8. Key Indicator Ratings..................... ............. 228 ; 9. Selected Statistics— January-June, 1983................. 247 I vi j LIST OF FIGURES Figure Page 1. What Is a Midwife?..................................... 7 2. Description of Traditional Birth Attendants I (TBAs) in Developing Countries.......................... 8 I 3. Systems Model for Assessing Traditional Birth I Attendant (TBA) Programs .......................... 24 4. Availability of Doctors and Traditional Midwives........ 67 I I 5. Control of Implementation: Approaches (Part A) and Organizational Structures (Part B)................. 110 6. The Four Factors: Linkages and Performance............. 120 7. Example of Logframe.................................... 128 8. Provisional Model for Planning and Implementing Evaluation............................................ 164 9. Organizational Chart: The Health Worker and the Village Community.................................. 211 10. Sind Dai Training Project............................... 243 V I CHAPTER I INTRODUCTION One of the outstanding problems of Western medicine has been the inability to transfer the benefits of the last decades of tech nical medical development to the large majority of the world’s population who most need these benefits. This failure is attrib utable not only to medicine or health care, but also to the political and socioeconomic situations. In the attempt to provide health serv ices to underserved or unserved (majority) populations in developing countries, the gap between rising expectations and realities heus led to a focus on primary health care (PHC) as a strategy to reach the target populations. There is a wide spectrum of definitions of PHC, from "any care a patient happens to need in hospitals or in the community" to "health care conceived and attempted by various groups of local people for their own benefit." Primary health care can be viewed as part of a response to pressure for public services to the rural or village-level areas of the world. In the health sector, PHC is part of a "from-the-bottom- up" strategy (i.e., a strategy more dependent on input from the village or rural population itself than on the professional health care establishment). The PHC idea was also revolutionary in recognizing that the formal health-care system could not achieve the "health-for-al1" goal of the World Health Organization (WHO), which 2; was formulated at an international conference on PHC at Alma-Ata, | Union of Soviet Socialist Republics, in 1978. ! The Primary Health Care (PHC) approach was a revolutionary strategy which recognized that health for a majority of the world’s peoples could not be achieved through the continued I development of conventional health systems based on sophisticated hospitals and the training of high level workers on the assumptioni that benefits would eventually "trickle down" to reach the poor. , Nor were the mass of vertical style disease programs the answer for the complex interrelated health problems. What was needed I was the involvement of people, both in their own health care and I in managing health services; the integration of health goals and actions as part of overall economic development; and the radical ; changes in the organization and delivery of health services which would involve shifts in allocation of health resources towards meeting the health needs of those whose needs were greatest. It also called for the integration of services so that it is not diseases that are treated but people. (King, 1983, p. 29) i The Declaration of Alma-Ata (WHO & UNICEF, 1978) described PHC as a mixture of concepts and functions embracing the philosophy of rendering to the needs of the family in its home or village environment. Thus, PHC includes all health care provided to families ! outside the hospital. Its goal is to achieve total outreach to pro vide health services for everyone. This focus on PHC was supported by WHO and the United Nations Children’s Fund (UNICEF) together with representatives of the 134 WHO member governments, and in 1978 they ^ launched a campaign to achieve "Health for All by the Year 2000" | through PHC (WHO & UNICEF, 1978). j One of the key implementation strategies of PHC is use of the village or community health worker. This strategy represents a signal I I I I advance in approaches to health care, and it is a key to dealing with | : I I the problems mentioned above (i.e., how to engage a community in I I I health and development matters and how to serve rural areas at a time ' 3 ; of pitifully scarce health-care resources, including finances and i I trained personnel). Thus, use of the community health worker has the I potential for resolving at least three major impediments to the I development of effective PHC programs: I - access of the entire population to the basic elements of I promotive, preventive, and treatment services; I - the cost of covering an entire population with services by the usual professional and para-professional personnel of the health system, which is almost always prohibitive; - social relationships between health workers and the population, which are often a barrier, but which must be close and trust worthy if the population is to be effectively guided and influenced in health-related and development-related behavior. (Ofosu-Amaah, 1983, p. 5) The generic area of community health workers (their recruit ment, training, and utilization) is both vast and complex. At the 1978 Alma-Ata conference, traditional birth attendants (TBAs) were mentioned as an integral part of PHC in the global "health-for-all" strategy. The role of TBAs is more circumscribed than are the roles of other community health workers: it is generally limited to provid ing health care during pregnancy and delivery. The focus of this research is on TBAs as a subset of community health workers. More specifically, this research is concerned with the train ing of TBAs by the principles of nonformal education. Nonformal education is defined as any organized, systematic, educational activity carried on outside the framework of the formal system to provide selected types of learning to particular subgroups in the population, adults as well as children. (Coombs and Ahmed, 1974, p. 8) I This focus on TBAs, and particularly on nonformal training programs for TBAs, remains within the context of the more basic problem: how 4 1 ' to provide PHC where there is a scarcity of personnel, money, and j knowledge about local communities. ; The Traditional Birth Attendant I Definitions I The term TBA refers to a "person who assists the mother at , childbirth and who initially acquired the skill of delivering babies i j independently or by working with other traditional birth attendants" j (WHO, 1979a, p. 9). In light of a review of the literature, this j definition of the TBA (who is usually a woman) is too narrow. The ITBA’s work also includes the provision of basic care and psychologi- ' cal comfort to women throughout their maternity cycle. In addition, the TBA may be involved in promoting family planning, in reducing ! morbidity, mortality, or both, in mothers and in children; in improv- I ing the environment; and, by serving as a bridge between the community I and the formal health-care system, in promoting PHC. Learning midwifery through her mother, relative, or neighbor, ! : the TBA generally is trusted for her skill and kindness. What the TBA does varies from culture to culture. There are many names for traditional midwives ; Table 1 lists the names of traditional midwives { in different parts of the world. Figure 1 gives a brief summary of ! jvarious information about midwives, including the traditional midwife I or TBA, and Figure 2 includes a description of TBAs in developing countries. In general, the TBA does not give much antenatal care, I though she does give dietary, religious/psychological, and herbal- remedy advice for specific problems in pregnancy. . J Table 1 Names of Traditional Midwives Country Country AFRICA Cameroun Chad Dahomey Gabon Ghana Guinea Ivory Coast Kenya Lesotho Liberia Madagascar Mali Mauritius Niger Nigeria Rwanda Senegal Sierra Leone Somalia Sudan Tanzania Togo Upper Volta Zimbabwe matrone, accoucheuse traditionnelle matrone matrone matrone matrone, vigila, miroiiaura. matrone matrone matrone babele xisi empirical midwife, indigenous midwife, traditional midwife matrone Matrone traditionnelle dai matrone granny accoucheuse traditionnelle matrone bunda or bundo omolesso dava el habil traditional accoucheuse traditionnelle, matrone accoucheuse traditionn?lle ambuya ASIA Bangladesh Burma Cambodia Fiji Gilbert Islands India Indonesia Laos Malaysia Maldive Nepal Pakistan Papua New Guinea Philippines Sri Lanka Taiwan Thailand Timor Tonga Tuvalu Viet-Nam MIDDLE EAST Afghanistan Egypt Iran Iraq Israel dai i Ü matrone allewa vuku tia kabung dai dukun ba.ii matrone bidan, kampong bidan fooluma sureni dai, djuddah bidajL. kampong bidan hilot, mananabang gode-vinappa. shan sheng ü lee mon tarn vae wata sitonq ma’uli tofunga matrone, ba do vuon, bâ mu vuon dai dm mama, dada diuddah dm J Table 1— Continued Country Name Country Name ASIA (Cont.) Jordan Lebanon Libya Morocco d m d m d m quabla Haiti Honduras Jamaica Mexico femme sage, homme sage partera empirica nana comadrona, partera empirica Saudi Arabia Syria habbaba, mama d m Montserrat Nicaragua bush midwife comadrona, partera empirica Tunisia quabla Panama partera empirica Turkey ebe, ebe anne Paraguay chae, empirica Yemen Arab Republic dada, dava raua’lida Peru comadrona, partera empirica Yemen, People’s Dem. Republic of diuddah St. Lucia Surinam femme charge granny WESTERN HEMISPHERE Trinidad & Tobago middy Argentina Barbados comadrona nannie Uruguay comadrona, partera empirica Belize bush midwife, nannie Bolivia partera empirica Brazil curiosa Chile partera empirica Colombia partera empirica Costa Rica partera empirica Cuba recoqedora Dominica local midwife Dominican Republic partera empirica Ecuador comadrona El Salvador partera empirica Granada bush midwife, femme shi Guatemala comadrona empirica Note: From Population Reports (1980, p. 439). a> cn > < 0 "O « 3 = 3 "^ < X > "O fo a m fO “O o a O)' c n - O D - ^ D CL) CÜ'-*— • -H <0 • r— » W — ' » - H 0 " O ' — 4 » w cn* » -H 3 0 L > 0 L > <o » o <o a t“o a> v_ c c: . o ) “0 --H cr> a>- -H o ) 0)-c:*»-H 0 <o-c: <o--h :w-» o « <o =3 o <o <o :> o> <o IS; O): V) O) CD O M — < • * — O '*— ( C D **— I 13-5-3:5^ “ O < / ) • cn cn cn cn cn & cXf'-H cr 0 C O <0 c/) X C O <o « o C = O =3 X ^ _C <0 * “ < 1 )0 >» - m H * — 4 I ••-4 C 3 3 t" 0 - cn“0 • <0 0J"O“ O " -t c= 0J“ O 0 4 0 <o**— • <o • O « Q ^ ^ C D » *— I # " " 4 H-4 O C • <o Cr>*-H <o»*»H cr 0Ü o cr CD • ^ <X) V ) ( / ) <X) O 0 <X)“0 < 0 <X) CD 4 — fo c : • * — > o j C D > > »-H I^B • • * —4» f —4 Z OL CD'*— I fO C D C D Z Z — > cn >“ 0-0 O CD <0»*»H»*«H»*«H»*-H 0 0 0 ) ^ 0 . z w — 0 m I o C D z a o> I % fO| c : CD c ;»*-h | ^ I < 0 ^ CD < 0 , -c < o a-olo CD--H 0 | Oi c=| - s t e C = Si- e - o « C L =)— 4 <0 C L ojal T 3 <0 O o> on( 1 < /s <D o> c: =’ :S:Sa •#->D= CD CD CL » 4 - » c= o a O < X ) C O L-- * »H C O Q L .- 4 -» CD <1> « 4 — » O ' *-H . o > a> c L. z •* — » O < X ) O 33 0 > a ^ L U CD -t-4 (/)_C CO " a ^ B » *» H tD i C D “0 C O O) O C L : w — C D L _ C D a D CD L_ O I Z C O o d »4-> "S' B CD"0 • C O ( C D cn Oh c « a > . r-4 c: c= Q D I i ^ .ii « m -4 O ' *”H CD < CD»" I CZ **H CA 4 CO C7>*4-> d * — 4 I a o C O CD CD CD CD CD Q j _ d O C D » * > h » - h L _ I LA CD • —4»r-4 d ^ — 4 O O I f O L 3 » * * H » 4 I * — ; O L. O C D -4-^-d • * — 4 O '*— • -4 -D C D Oh * — 4 “ O < X )'*» H < 4 -D ,-4 -0 — 4 =3 C O -4 -D C D C A 4-A- — l« - » ——4 c K. O =3 C O L_ L m . mH-0 O C O C O C O L- _ d 33 =3D= <0 L- > - C D C3 » 4 -> ' C D o r L_ I C D C D C D D 3 Z C D O CL a * * 0 C A d»4-a <OjO CD LA 0 — 4 d O O » C D Z Z O L A a > - C A C D » 4 -> o •— 4 L_ —4 CDDd • CO_d C O » 4 -> C D C D *— H » — 4 • * — > C D Z O • — 4 CL_d C O jQ o • l 4- c rro -4 -D » *— » C O < X ) C A d d C O " 4 -J _ 0 _ d » d CA »— 4 o a C A »— 4 *« -> & O C D » 4 — » 4-A C D -4 -D d Cr>*-4 C A d L_ C O L A CA-*-4 d Z -4-a C D O C O C D “ 0 • C O » *-H — 4 d L-»*-H L_ d C A LA O W 0> C O <0 » 4 ~ > 0)«4-> C O < X ) d C A D 3Z C D d * o --4 d •— 4 O C O Z > — 4 d C A »4->»— 4 d C A “0 • — 4 •— 4-rkC C D C D C D *— 4 O D » 4 -> C D » 4 -A 0 *0 —4 >• C O <o»w d a CO_d d C D O » 4 ~ > '*-H w. < X )'-*— C D O D fs* z C0*0 Z C O Z CLD33 O 0 L. o g w 4 - ) L, I G O G & Du § U Du C ^- c£ •fU ■ë •H s « 4 - ) I (Ü G S • iH - Cx. T 8 Age Sex ' Status I Education Training Remuneration Antenatal care Delivery care Postnatal care Other care Frequently 60 years old or older; some 40-60; few under 40. Host commonly female but male TBA’s are not uncommon in Latin America and Africa. In much of world, TBAs are highly respected in the community for their knowledge of birth and ritual; in India, Bangladesh, and a few other countries, TBAs are not as respected because they are lower-caste persons. Frequently illiterate but some have several years of education Some TBAs have received formal training from government or private health programs: almost all TBAs have apprentice and practical experience in childbirth. Although usually having many years of experience, TBAs' activity level varies greatly from attending one or two to over 20 deliveries per year. receive some small payment from the family, either in,money or in kind usually receive some small payment from the ramiiy, either in me (often some food); TBAs sometimes give gift to mother and child. TBAs do not usually provide, but do in some parts of the world (e.g., in Mexico and Guatemala beginning in the 4th to 7th month. TBAs see pregnant women, often giving abdominal massages and advice on diet and exercise). Most TBAs provide delivery assistance. Some use massages, herbs, ritual, etc. in normal delivery or in reaction to complications. Many TBAs stay with the mother for a week or longer following birth or visit the mother periodically after birth. Some TBAs treat infertility, many induce abortions (e.g., in Mexico, India, and Greece), and perform female circumcisions (Africa). % Figure 2. Description of Traditional Birth Attendants (TBAs) in , Developing Countries. (From Favin, Bradford, & Cebula, 1984, p. 28.) 9 I Her main duty is to be present and assist at delivery: to comfort and reassure the laboring woman, and to carry out the rituals and tradi- ' tions associated with her culture. I ; Importance ; Often, formally trained health-care personnel look down on i the TBA because these personnel view many of the TBA’s practices as being dangerous to the health of the mother or child. It has been reported that several investigators have suggested the need for structuring the analysis of health related beliefs and practices according to the following categories: beneficial elements that are valuable to health in the local environment and should be encouraged and "adapted to health education"; harmless or neutral practices that have no obvious effect on health should be ignored; harmful pro cedures with detrimental effects and that should be consequently modified; and uncertain or neutral practices whose effects are unknown or that have possible beneficial and harmful effects thus ! being difficult to classify. (Cosminsky, 1978, p. 116) Thus, in training^TBAs, their useful practices should be encouraged, : their neutral ones ignored, and their negative ones strongly discour- I aged. Although there has been some resistance from the medical profession, official acceptance of TBAs generally has been growing in I recent years (Favin, Bradford, & Cebula, 1984; Owen, 1983a). Despite being considered illegal in Egypt in the 1970s, TBAs continued to be active there, even collaborating with governmental health staff (SPAAC, Note 1). According to WHO, 71 percent of the reporting coun tries formally recognize TBAs in some way, but utilization of TBAs may not be part of their health-care strategy. Of 63 countries with I coordinated national PHC efforts that were reported to WHO in 1981, 10 i 52 (82%) reported that they had TBA training programs; in 1983, the comparable figures were 78 of 122 countries (64%). Most countries intend to replace TBAs with more highly trained personnel from the formal health-care system, but financial considerations place realization of these intentions in the future. Traditional birth attendants are community members who already have knowledge of local systems because they are a part of : these systems. The worldwide interest in training of TBAs has resulted at least in part from the nearly universal scope of TBA ' practice: it is commonly estimated that TBAs assist in 60 to 80 percent of births in most developing countries (WHO, 1979a, p. 7). Some experienced observers feel that the actual percentage is I lower because relatives such as mother-in-laws and grandmothers i attend many births. Also, in some areas, such as parts of Mozambique and Indonesia, it is the custom for women to give birth unattended. In these cases, a TBA may arrive after the mother gives birth. (Favin et al., 1984, p. 27) ' Nevertheless, in rural India, well over 90 percent of deliveries are ; conducted by TBAs (Kumer, Note 2), and in parts of Africa and Asia the figure may well be closer to 95 percent (WHO, Note 3). Another ■ general reason for interest in TBAs has been the consideration of I populations in developing countries. In focusing on the needs of the I I rural majority, priorities had to be set and the members of the population most at risk (the people most likely to die) had to be determined. According to morbidity and mortality rates, obviously the most at risk were mothers in pregnancy, nursing women, and chil dren under 5. In the "poverty war," they are the first casualties. Tetanus, malnutrition, postpartum infections, and too many births all ! 11 ' ! ' contribute to the disproportionate risk to women and children. The I TBA can help reduce this risk: I Instruction of TBAs in Western obstetrical skills has been described as an economically and biomedically sound method of I counteracting the high rates of maternal and infant mortality. ! (Davidson, 1984, p. 7) j I I As already mentioned, TBAs have the potential for strengthen- j ing PHC to the large rural populations. Furthermore, although the TBA’s role in family planning is inadequately developed (Favin et al., | 1984), interest in training TBAs to help reduce fertility rates is growing. Increasingly, there is recognition both that rapid popu- ' lation growth is a development problem and that appropriate public j ! I ! policies are needed to reduce fertility. But selective government intervention is necessary, too. Public education, especially elementary education, yields very high eco- ; I nomic returns in developing countries. Government action is also , I needed to improve and extend health and family planning services ; ! more widely. There are an estimated 65 million couples in devel- I oping countries who do not want more children but still do not use ' contraceptives. (Clausen, 1985, p. 9) Controlling Population Growth I ; Importance Recently many multilateral and bilateral donor groups, including the present Reagan administration, set aside substantial sums of money for population and family planning, including monies for TBA training in family planning. Birdsall (1984) stated that rapid population growth slows development by: 1. Forcing difficult choices between high consumption now and investment needed to facilitate investments now, thereby facilitating higher consumption in the future; I ------------ " 12 ' I ! 2. Threatening what is already a precarious balance between i j I scarce resources and people, especially in many populations j still highly dependent on agriculture; and ! 3. Making it difficult to manage the adjustments that are neces- : i sary to motivate economic and social change. | In many countries, fertility declines from the late 1960s into | I the 1980s have been much more closely associated with adult literacy ' and life expectancy than with GNP or with per capita income. Birdsall (1984) observed that differences in income, religion, and culture do , not totally explain differences in rates of fertility; education, I access to family planning services, the status of women, and economic I ; and social policies that bring opportunities to the majority of people | I all make a difference. 1 I Several countries that have supported family planning services have significantly lower fertility than the norm for their income level, including China, Colombia, Egypt, India, Indonesia, Korea, Sri Lanka, and Tunisia. (Countries with relatively high fertility | given their income include Algeria and Morocco, most countries of | Sub-Saharan Africa, and Venezuela.) In most of the low fertility i countries fertility has declined much faster than it did in today’s developed countries— both because of the availability of | more effective modern contraceptives and the rapid expansion of educational opportunities in the postwar period. (Birdsall, 1984, p. 15) Within regions, differences in fertility among countries are ' I due in part to population policy effort. In many developing parts of | the world, TBAs are viewed as prime components of a family planning | j system that will reduce fertility. Increasingly, TBAs are viewed as one of the principal means of delivering family planning services. One cannot improve the status of the world’s poorest popula- j^tion groups merely by improving their fertility control and child I survival through better health care; substantial economic improvement i ' is also necessary. The beneficial effects of economic improvement will, however, be greatly augmented through a parallel improvement of 1 health services and literacy. The general awareness that population ' pressure is a problem is demonstrated by the prodigious recent 1itéra- I I ture on this issue, particularly that from the World Bank and the ; I United States Agency for International Development (U. S. AID). In [ 1 the World Bank’s library are over 15 new documents on the issue of I I population growth and its relation to development. Role of the Traditional Birth Attendant One of the key reasons that TBAs increasingly are being viewed ; as a resource to combat high fertility rates, to upgrade PHC for rural i populations, and to reduce high rates of maternal and infant mortality I is their membership in the rural community. The resulting potential ' for linkage (via the TBA) between the community and the formal health- i I I care system is being recognized as an advan- tage in delivering PHC . services to rural areas. Pertinent to the innovative concept of PHC, and the TBA’s role in it, is the dichotomy between homophily and heterophily. Rogers and , ' Shoemaker (1971, p. 210) defined homophily as "the degree to which : pairs of individuals who interact are similar in certain attributes | I j ' such as beliefs, values, education, social status, and the like" and 1 heterophily as "the degree to which pairs of individuals who interact are different in certain attributes." In their discussion of opinion ; and leadership in the process of diffusion of innovations, these j I authors hypothesized that "change agent" success is positively related j 14 ; I to "homophily with clients" (Rogers and Shoemaker, 1971, p. 242). : This concept of homophily is pertinent to the training of TBAs I because TBAs are homophilous with the rural people so in need of , PHC. Traditional birth attendants are generally illiterate and must I be trained according to their own learning abilities and culture. In the literature on TBA training, virtually all of the findings suggest that the training be based on a thorough grounding in the knowledge, attitudes, and practices of the TBA (Davidson, 1984; Favin et al., 1984; Greenwood, Mann, & McLaughlin, 1975; Harrison, 1983; Neumann, ; Nicholas, Amonooh-Acquah, Pensah, & Boyd, 1983; Walt, 1984). ( j When faced with the dictates of the training programs that aim at ' modifying patient care, TBAs are placed in the position of \ "translator" (see Kleinman, 1980:53). That is to say that the performance of their duties as trained TBAs depends on their ability and willingness to translate the dictates of the program j into the local language of their patients without jeopardizing I the relationship they hold with their patients. (Davidson, 1984, p. 24) i The homophily of TBAs with their patients becomes crucial; a I review of 52 PHC projects (Parlato & Favin, 1982) indicated that many i of these projects failed because of a lack of community input. The same review showed that women were frequently either reluctant or culturally prohibited from contacting male community health workers or young, unmarried, female community health workers. Traditional birth attendants are already members of the community and are gener ally between 40 and 60 years of age (see Figure 2). They also tend to be permanent members of the community: in general, they do not seek career opportunities away from their local communities (as do many other village-level health personnel). The realization that 15 ; I TBAs are part of their respective communities and likely to remain so I is a key reason that a number of developing countries have chosen to j ' ; train TBAs and use them for linkage between rural areas and the formal health-care system. I I Training Traditional Birth Attendants | I I I The practice of traditional midwifery; the local health- ' j service structure; and the political, social, and economic conditions ; I vary among and within countries. Thus, there is a need for socio cultural analysis, political analysis, macroeconomic analysis, and | I managerial analysis as a background to PHC and to the nonformal j I ' j training of TBAs. This quartet of factors (sociocultural, politi- i cal, economic, and memagerial) profoundly influences TBA training ^ I programs— hence the need for such analysis. Even when nonformal educational techniques are utilized effectively in TBA training, preimplementation and postadoption factors must be examined. In | ' short, the successful training of TBAs is more complicated than many i people think, and many of the past failures of TBA programs have been I due to a lack of understanding of the context in which TBA training takes place. I t Difficulties ' I I To have an international model for training TBAs is difficult, ; ^ I I if not impossible, because of variation in the sociocultural, politi- | ' I cal, economic, and managerial context of the training. The suitable use of nonformal education in the training of rural health workers is I a fluid, dynamic process that will change over time. The Chinese j ; "barefoot doctor" program demonstrates such changes. The "barefoot 16 ! I j ! doctor" in China is a village-level worker, and the program worked i ' well for a while. The economic and political evolution in China has, ' I ' however, proceeded to the point where Chinese people in rural areas I have changed their behavior. Farmers who formerly worked in a system ' j of collective welfare now see that if they work harder, they will | I make more money. These same farmers now want to go directly to the ! I more formal, sophisticated health-care system and thus are circum- | venting the "barefoot doctor." In the changing economic and political situation of today’s China, rural people do not want to share risks, and the voluntary insurance schemes are collapsing. The health-care j structure must fit the social and economic conditions of a country; modifying health-care institutions so that they conflict with existing i I economic and social systems is not practical. I Likewise, the tendency to use current rural health-care ' i training manuals (which may well be of high quality) as blueprints , for training auxiliary health-care personnel is not practical. These ' manuals must be used as points of departure and adapted to local con ditions because of the diverse nature of the TBAs, who are mostly | I I I ' illiterate, and because of the context in which the TBAs operate. The ' tendency to "over-blueprint" and be rigid is pervasive, and it is one reason for the failures in actual implementation of rural development ■ programs in general and of TBA programs in particular. Because the j problem of implementation is such an obstacle to the success of devel- ! opment programs, an entire chapter of this dissertation (Chapter IV) ! ! will be devoted to implementation problems and the tendency to regard planning as a separate activity from implementation. J 17 In reviewing the literature on managing induced rural devel- I opment, Moris (1984, p. 34) stated: "Implementation is the major • problem area, but it is both poorly understood and relatively : unglamorous to academic analysts and donors alike." Implementation I and obtaining results has been recognized as one of the problems for 1 I the next development decades. ; Out of the euphoria for the promise of planning the United Nations Planning and First Development Decades of the fifties and sixties emerged strong voices for post-mortems and analyses of what went wrong in development strategies. Even in the late fifties and mid-sixties, serious questions were already raised concerning the efficacy of comprehensive planning in countries where the adminis trative infrastructure to carry out such plans is weak. Some questioned the apparent preoccupation of developing countries for macro and comprehensive planning and the consequent neglect and lack of priority attention given to the development of feasible and viable projects (macroplanning) as a base for comprehensive plans. It has been mentioned that the administrative and manage ment capability for plan implementation often poses serious obstacles for the implementation of comprehensive plans for development. As a result, implementation became a catchword and priority concern in the U.N. development strategy for the Second Development Decade of the seventies; this concern will extend through the eighties as well. (Goodman & Love, 1980, p. 6) One of the weaknesses in TEA studies is a lack of attention to implementation issues and to lessons from other disciplines on these issues. More important than studying and planning the problem, and more important than relying on training manuals, are (1) finding an approach that emphasizes "learning-by-doing," and (2) applying knowledge gained from experience to further improving the health-care system. Chapter III of this dissertation considers not only the approach of learning through doing, but also both PHC in general and an attempt to reorient the formal health-care system more toward PHC. This issue is included in Chapter III ("Review of the Literature," in 18 the section on "TEA Training") because PHC is a crucial concept and TEA training must be seen within the context of providing PHC. Approach The practicality of the "health-for-al1" strategy that recently has become a slogan depends on the realization that its success in developing countries requires innovative approaches not only in the health-care systems in these countries, but also in the donor agencies themselves. Examples of donor agency problems in implementation of Third World development programs will be given for two reasons. First, funds for implementing TEA programs largely come from agencies such as the U. S. AID, WHO, and the World Eank. Second (and equally important), the research literature on development in general fails to acknowledge a bias against implementation that exists in the donor agencies themselves. Although a full considera tion of this bias issue alone would require several dissertations, donor bias is so much a part of the problem of analyzing success vs. failure of TEA programs that it is discussed in this dissertation (primarily in Chapter IV, "Implementation Considerations"). Nonformal education, health-service delivery systems, agri culture, evaluation, and public administration all are disciplines that have much experience in dealing with issues vital to successful implementation of TEA programs. A review of the literature on TEA training reveals very little pertinent cross-fertilization among disciplines; the lessons learned from other disciplines are seldom applied to the problems of health-service delivery in general and TEA J f 19^ I programs in particular. This dissertation is a beginning attempt to I ' ' remedy that serious gap in the literature on TEA training. ; I There are two central themes of this research. The first is I I that in analyzing TEA programs, one must take a systems approach. I The second is that one must look at the whole situation of the TEA I ' I I and take into account her own point of view— not just the points of j I ’ I view of bilateral and international donor-agency members and person- ' nel in ministries of health. Eoth of these themes are crucial to the success of TEA programs. Nonformal education techniques are essential | for optimal training of TBAs because compared with formal techniques, I these nonformal techniques have the potential to be more flexible, to be shorter, and to better address the needs of the TEAs in the train ing situation, A nonformal education approach also has the potential ! to respond to the quartet of factors (sociocultural, political, eco- I I nomic, and managerial) that so profoundly influence the context in i which TEA programs operate. Chapter V will be devoted to nonformal i education and will consider literature on innovation and change > (including their lessons for TEA training), which are crucial because : of the innovative nature of PHC and TEA training in general. The importance and recognition of TEAs is on the rise, and j many countries have ambitious programs for expanding the roles of TBAs. These programs are discussed in WHO (1979a, 1979b, 1982b) and in Walt (1984), as well as in other publications. In a time when resources are scarce, TEA training is considered to be the only feas- \ ible approach to providing some sort of health-care service to the ! ! hitherto unserved majority of rural people. 20 ; CHAPTER II ; STATEMENT OF THE PROBLEM AND METHODOLOGY ! The problem is to analyze traditional birth attendant (TEA) I ■ programs in the context of providing primary health care (PHC) to unserved and underserved populations in rural areas. Very little money is available to provide such PHC, so the usual ministry of health (MOH) mechanisms are not an option for providing PHC to these i rural people. Instead, village-level personnel and resources need to I be mobilized. j Training (or upgrading the training of) TEAs can be useful, because TBAs are accepted and paid by villagers. The literature shows that TEAs are interested in having their skills upgraded and modernized, and their role can be somewhat expanded to include basic education in such elements of health care as nutrition, family plan- , ning, hygiene, immunization, oral rehydration therapy, and the like. Because most TEAs are illiterate or barely literate, techniques for I their initial training and their supervision will have to be based on I nonformal education tailored to the needs of the TEAs and their I supervisors. The training of TEAs should be preceded by a period in which the villagers are organized and educated to familiarize them with ! both the new, expanded role of the TEA and the value of this expanded role. Thus the villagers will serve as a kind of quality-control mechanism, and in response to good service, they will pay the TBAs more than they have in the past. | I ! I It is essential to educate MOH officials so that they will I allocate the resources at their command to properly train, supervise, I I and resupply TBAs and to integrate TBAs into the formal health-care j j system. ■ ; Purpose of the Study The purpose of this study is to test a conceptual systems I model for analyzing TEA programs. This model contains three problem areas : preimplementation, training, and postadoption. It can at least j supplement (and possibly replace) approaches such as "logframe" pro- i I gram assessment (see Chapter IV) because of its broader focus— it , j considers more variables. If this systems model is practical for I assessing programs involving something so fundamental to life and so ! 1 bound to culture (i.e., birth), it has potential for application in j assessing development programs in general. j ; ! I Delimitations Nonformal education’s very lack of structure and its task orientation make it optimally suited to the idiosyncratic environments 1 I and procedures encountered in TEA programs. Its ability to effec- | I I I tively introduce idiosyncratic pedagogy and content into TEA programs j I is important, but a more focused inquiry into these issues is left to ! another study. I In terms of money spent in developing countries, after defense I the largest budgets are generally those for education and health; but I I personnel in the latter ministries are weakest and have the lowest ” 22 I I status. How these circumstances influence TBA programs is another important delimitation of this study. Assumptions I I The use of a systems model to assess TBA programs rests on a I ) number of conceptual and substantive assumptions. Conceptual assump- i j tions are assumptions connected with the theoretical basis of the I systems model being tested, whereas substantive assumptions are j assumptions connected with the design of the study. Such assumptions are made throughout the study, but they are not subjected to any test ing or given any thorough empirical support. The basic conceptual eussumptions of this study are the following: 1. There is a cultural and structural incompatibility between health institutional settings and TBA program characteristics that hampers effective implementation and institutionalization of PHC efforts such as TBA programs. 2. Mere adoption of a TBA program does not necessarily ensure implementation and routinization of the program. The substantive assumptions of the study are the following: 1. Donor agencies (mostly Western) are the principal source of financial and technical assistance for PHC efforts and TBA programs in developing countries; thus, donor agencies play a crucial and influential role in implementation of TBA programs. 2. Although the World Bank, the World Health Organization (WHO), and the United States Agency for International Development (U. S. AID) are the donor agencies from which data were gathered. J 23 other multilateral and bilateral donor agencies probably exhibit the same or similar biases to implementation. 3. In theory, developing countries are coimplementors of TBA programs, but in practice they are closer to being passive recipi ents of prepackaged TBA programs. Many developing countries allocate little monetary and human resources toward substantial PHC efforts. Systems Model It is postulated that TBA programs must be viewed from the ' triple perspective of preimplementation, training, and postadoption. ' These perspectives are the components of the systems model for asses- ■ sing TBA programs. Of the three components, postadoption is likely I ■ to be the most important. The model is diagramed in Figure 3 and outlined as follows: I Stages I. Preimplementation (ministry of health) A. Acceptance of concept of using TBAs B. Legalization/licensing j C. Administrative orders authorizing staff to work with I TBAs I j D. Allocation of resources ■ 1. Staff 2. Supply 3. Transport 24 STAGE I Preimple mentation TBA Program Implementation STAGE II; Training STAGE III: Postadoption Figure 3. Systems Model for Assessing Traditional Birth Attendant (TBA) Programs. A systems approach takes into account a quartet of factors that influence program implementation in each of three stages: preimplementation, training, and postadoption. Nonformal education must be applied throughout the three stages. The approach is "learning-by-doing" and must be flexible. Implementation occurs amidst scarcity and apathy. L - 1 25 II. Training A. Selection of trainers B. Duration C. Location D. Rémunérât ion 1. For training per diem 2. For service E. Evaluation 1. TBAs 2. TBA training programs III. Postadoption A. Supervision 1. By whom 2. Frequency 3. How 4. Training of supervisors 5. Remuneration 6. Content B. Evaluation (effect) 1. Client 2. Participation C. In-service training D. Resupply E. Referral of patients 26 Factors that influence stages I. Sociocultural A. Low status of women B. Minimal community participation and incorporation of traditional practices C. Low level of epidemiological knowledge (biological understanding of populations) and anthropological/ sociological knowledge (behavioral understanding of populations) about local populations II. Political A. Formal health systems not set up for PHC B. Failure to realize that conflict will result in imple mentation C. Insufficient social-science expertise in MOH III. Economic A. Current allocation of three fourths of health-aid bud gets to urban minorities B. Current allocation of three fourths of donor aid to urban (sophisticated) health systems C. Need to reallocate donor aid, getting over three fourths directly to PHC efforts, so that "health- for-all" strategy has a chance to succeed IV. Managerial A. Organizational productivity manifested by effective linkage to rural majority in villages _ J I 27 B. Special emphasis on study of supervisory staff in PHC and TBA programs C. Incentives for bureaucratic efforts of donor agencies, MOHs, and lower level health-care personnel to enhance PHC and supervision of TBAs A suitable supervision, follow-up, in-service training, resupply, and evaluation program for TBAs is essential, or the TBA training will be largely wasted. Eventually, TBAs will return to their traditional ways of practice. Except in "vertical" programs such as immunization, growth monitoring, oral rehydration therapy, and I breast-feeding, supervision is never incorporated into routine visits I : with specific content (Walt, 1984, p. 3). Perhaps as these new roles ^ for TBAs are incorporated into training, supervision will be more ; consistent, of higher quality, and more specific in content. The supervisors seldom understand the administrative or organizational skills of the TBAs they supervise. Promotion of supervisors is 1 ! usually based on seniority, not performance. Historically and chron ologically, the primary focus of TBA programs has been on training. Implicit in the overall problem analysis is a systems approach, which means that more emphasis should be given to pretrain- I ; ing and posttraining considerations. An implementation perspective also accents postadoption behavior as opposed to preadoption behavior. | Whether excellent, good, or bad, training of TBAs is largely irrel- I evant unless there is an organized program of regular follow-up, | supervision, and resupply, as well as an organized program of regular in-service training beginning immediately after the primary training r 28 i I I period (usually within six months, and preferably within two or three ; ; months). Otherwise, the TBAs will probably revert to their former I practices. "Despite improvements in TBA training course over time, I the TBA's impact was limited by lack of support for them afterwards" ; (Favin et al., 1984, commenting on TBA training in Nicaragua). The reality of life in many rural areas of the world is that isolation and lack of transport make it very hard to transfer critically ill mothers and infants to the rural health center or some j part of the formal health-care system. The problem of providing I adequate supervision of large-scale TBA programs is formidable, and I solving it may be the essential element to successful I institutionalization of TBA programs. I Methodology i j The fundamental problem in TBA program assessment is lack of implementation, not bad policy. Personnel in MOHs and donor agencies such as the World Bank, WHO, and U. S. AID do not really know what is I going on at the local level where the TBA functions. I The institutions and roles of TBAs are already established, i A traditional structure (the TBA) is used as a conduit through which i I change is introduced. Nonformal education is the mechanism uniquely I suited to the training of TBAs because of its informal structure and I its flexibility. If the systems model is effective in something as culturally powerful as birth and in TBA training and assessment, then this model probably can be used in base-level rural development as well. j 29 I I I Academic training, field experience, and specialized litera- I ture (especially that on nonformal education and implementation) I relating to rural development activities all contributed to the meth- * odology of this study: testing a systems model by applying it to five ■ case studies. The major sources of project data and donor experience I I I are the World Bank, WHO, and U. S. AID. I I I , I : Testing of Model by Case-Study Analysis To test the systems model, case studies were reyiewed in terms of a series of key indicators to assess the three problem areas. From ' I a careful review of the literature, the key indicators were chosen to ' I assess different factors in the three problem areas. The key indica tors were chosen by consideration of factors that seemed vital to TBA | I programs. These factors ranged from MOH commitment (in terms of ! ' I allocation of supplies and transport) to curriculum considerations ' regarding utilization of techniques to train a largely illiterate TBA population. ! A matrix of factors and key indicators was then drawn up to assess the three problem areas. There were 10 indicators for the i I problem area of postadopt ion, 6 for the problem area of preimplementa- I tion, and 7 for the problem area of training. The data were generated , I by careful review and analysis of case studies of TBA programs in I Bangladesh, Brazil, Peru, Ghana, and Pakistan. Ratings of the differ- j , ent indicators took into account the discrepancies in the literature i on the TBA programs and the shortcomings that were pointed out by I ! various studies on the TBA programs. The result is a more thorough 1 i I 30 I I , and realistic appraisal than would otherwise have been possible. The I I I background for each country helps to put the data in context. I The process of implementing TBA programs, via donor aid and I , local participation, was assessed by analyzing the five case studies ! I I I mentioned above. The criteria used to choose these case studies from | I among TBA programs that reflected the factors of the model were the existence of (1) a considerable body of data and (2) accessible people who were intimately associated with the programs. People involved in the development and implementation of TBA programs were crucial in j evaluating TBA program assessment. Ghana's Danfa project (which ended in 1979) and Brazil's I Fortaleza program (which is still functioning) had excellent conti- i I ! I nuity rates, and their TBAs functioned at the expected level of i I I training. Not only was training effective, but also facets of train- | ing were realized to a higher degree than in other TBA programs where : follow-up, supervision, and management were not consistent. National I programs of Bangladesh, Peru, and Pakistan also were analyzed. t Many encouraging signs of improvements come from Pakistan's Sind dai training project (analyzed by Abbasi, 1984, 1985), which is I I I still going on. These in-depth case studies, the elucidation of I ■ "emic" (insiders') perspectives, and the resulting makeup of key indicators were used to assess the TBA programs. ; The degree of key-indicator attainment was rated on a five- point ordinal scale ranging from very low or absent (rating = 1) to very high or complete (rating = 5). Although the main ratings were 1, 3, or 5 for each factor, ratings of 2 or 4 were also assigned as | 31 needed. For a factor to be assigned a rating of 4, not all conditions warranting a rating of 5 were met; and for a factor to be assigned a rating of 2, not all conditions warranting a rating of 1 were met. The ratings by country and by factor are presented in Chapter VI. It must be kept in mind that in the matrix of factors and key indicators, the factors represent nominal categories, while the key indicators represent ordinal rankings. Thus the matrix scores cannot be used for mathematical or statistical comparisons. Nevertheless, clear patterns emerged that corresponded to findings in the "Review of the Literature" (Chapter III) section on TBAs. The matrix for Stage I of the systems model (preimplementa tion) is as follows: i ( Factor A. Efforts to understand community perspective 1 . 5 3 i i : No effort to make community aware of TBA program and no effort to survey community on health problems. 3 : Some effort to make community aware of TBA program, but understanding of community health problems absent in curriculum, training, and supervision of TBAs. 5 : Genuine effort to understand community (via knowledge, attitudes, and practices [KAP] surveys, for example) and full community awareness of role of TBAs to be newly trained. U1 o . 0)4-» C = f O Factor B. The TBA perspective 1 : Program implemented without any TBA input. 3 : TBAs surveyed; information not connected to TBA program in terms of program design. 5 : TBA perspective taken into consideration as one of main elements of TBA program. I ' 32 Factor C. Government/MOH acceptance and support for using services of trained TBAs Allocation of supplies 1 : No MOH resources put at disposal of project; MOH functioning as if program were outside its domain and support. 3 : Supplies given, but sporadically and not for program improvement. System (MOH) inflexible (not focusing on basic needs of program). 5 : MOH committed and supporting TBAs by supplying TBA program; MOH aware of and responding to implementation needs of TBA program. Allocation of transport 1 : Lack of support in maintenance, gasoline, vehicles for monitoring at program implementation level. 3 = Transportation provided, but with problems in vehicle maintenance, scheduling, and availability of gasoline. 5 : Transportation needs squarely met; MOH or program viewing transportation needs of TBAs and supervisors as major factor in functioning of program. Factor D. Development and pretesting of curriculum 1 : Curriculum either vague (objectives not stated) or borrowed and unsuited for local conditions; no pretesting. 3 : Curriculum developed; little pretesting of results for use in revising curriculum to fit local conditions. 5 : Curriculum developed and modified according to local needs; results from TBA surveys and initial training of TBAs within curriculum used to revise curriculum to fit local conditions. Factor E. Careful selection of trainers (local language and practices) in o C = f O 1 : Trainers selected, but orientation of trainers not known. 3 : Trainers partly oriented to training TBAs; local language ability not part of selection criteria. 5 : Trainers thoroughly oriented to meet special needs of TBAs; trainers use local vernacular. Factor F. Careful selection of TBAs to be trained, with both TBAs and village agreeing in o CD 4-) cz 1 : Selection criteria for TBAs not operable; community unaware of TBA program. 3 : Selection criteria for TBAs partly operable; community still largely unaware of TBA program. 5 : Selection criteria for TBAs operable; community involved in selection and/or agreeing with selection. The matrix for Stage II (training) is as follows; 33 : Factor A . Location o to a s CZ «3 •-H C -> ^ cz cz 3 E = - Z = C 1 : Training location decisions made on ad hoc basis and/or made with only MOH perspective and input. 3 : Local training advocated; TBAs’ training arrangements not fully taken into account. 5 : Training done locally as much as possible; TBAs’ training arrangements taken into account. Factor B. Duration in o c= ra 1 : Duration of training centrally planned; no flexibility. 3 : Duration seen as important factor, but little local knowledge to build on. 5 : Duration seen as important factor, and local needs/variations of trainees taken into consideration. Factor C. Remuneration o to in o cr>*^ fi= <o s ' Z :S-2c 1 : TBAs given per diem, but TBA kit (if given) too costly and items in kit not replenishable locally. 3 : TBAs given per diem, kit, and certification. 5 - TBAs given per diem; kit given and pared down to minimum with needed local items; certificate, diploma given or not given; TBAs important in training program. Factor D. Content ^ 2 • 5 o fO-O CZ cz CO C L ) ac.pg 1 : Content of training program borrowed or not congruent with local needs. 3 : Content thorough but inflexible; limited adoption to local needs. 5 : Content thorough, taught incrementally, geared to local needs; training flexible in being able to add/delete concent as appropriate. Factor E. Methodology to o .5 3 . 5 % 1 = Training not geared to TBAs’ needs. 3 : Training partly geared to TBAs’ needs, but not enough repetition used; too much taught too soon, so consolidation not thorough. 5 : Demonstration, repetition, and incremental training done; TBAs’ participation encouraged. . J 34 Factor F. Evaluation Of training 1 : Methodology of training not geared to TBAs; TBAs usually illiterate; incorrect to assume TBAs illiterate (e.g., in PeruT 3 : Training not thoroughly linked to capabilities of TBAs and not linked to local health-care centers. 5 : Training evaluated on-site; training linked TBAs to local health-care systems. Of TBAs’ training 1 : Learning not evaluated. 3 : Learning evaluated by written exam. 5 : Learning evaluated by written exam and on-site visits when possible; regular follow-up done to have more longitudinal data and to maintain TBAs’ interest and learning. The matrix for Stage III of the systems model (postadoption) is as follows: 35 Factor A. Supervision By whom 1 : Authority for supervision designated, no support. 3 : Authority for supervision designated, partial support. 5 : Authority for supervision designated, full support. Frequency 1 : No schedule for supervision. 3 : Sporadic supervision. 5 : Routinized schedule (at least once every two months). Method 1 : No specification of method. 3 : Specifications, but supervisors overworked or lacking means to do "out" supervision, or TBAs lack incentives to participate in "in" supervision. 5 : "Out" or "in" supervision specified and linked to refresher courses, in- service training, evaluation. Training of supervisors 1 : No training. 3 : Incomplete training, supervisors not fully capable of support. 5 : Thorough training, supervisors know field and can relate to TBAs’ background. Content 1 = Completely non-task-oriented. 3 : Partly task-oriented, but too vague. 5 : Task-oriented, content specific. Remuneration of supervisors 1 : No payment. 3 = Payment augmented, but not much; not much program support or recognition from MOH. 5 : Salaries paid and built on recognition of supervision as essential for success of TBA programs. 36 Factor B. Resupply, in-service training, and patient referrals Resupply and in-service training 1 : No resupply or in-service training. 3 : Sporadic resupply and in-service training. 5 = Regular resupply, and in-service training connected to evaluation and con solidation of lessons learned. Patient referrals 1 = No referral system. 3 = Referral system set up, but linkage to formal health system weak. 5 = Referral system functioning; linkage to formal health system strong; referral one of key points in training and supervision of TBAs and in formal health-care system. Factor C. Evaluation of effect ! ; ; ‘I Targets set/being met 1 - Targets vague; no evaluation based on targets. 3 : Targets set, but vague; evaluation not a priority. 5 : Targets specified; evaluation done in conjunction with supervision and to determine whether targets being met. Linkage to MC ning process in terras of information getting back and aiding health-care plan- 1 = No evidence of feedback. 3 : Information from evaluation fed back, but poor linkage to health-care planners. 5 = Information from evaluation regularly fed back into formal health-care system, and health-care planners modifying their practices accordingly. ! Sequence of Presentation The study relied heavily on a wide range of literature and I disciplines related to implementation of TBA programs. This approach j necessitated a sequence of presentation that takes into account a ; logical interrelationship among various segments of the body of : literature reviewed and among the various components of the systems ! model. 37 1 I Considering the study's special emphasis on a systems approach to assessment of TBA programs, it was necessary to devote an entire I section of the "Review of the Literature" (Chapter III) to PHC. This ! : : section on PHC was intended to provide a backdrop and set the stage I for the review of the literature on TBA programs. i It was equally desirable to provide a broader (theoretical) framework for a study that primarily focuses on the meaningful assess- ; ment of TBA programs. In keeping with emphasis on the systems model, i therefore, greater attention was paid to the concept of implementa- ' I tion, with special accent on several donor agencies' biases toward I the practical aspects of implementation. ■ ! 1 There is an obligation to make sure that TBA programs really | I > I do reach and help the poor. The question of practical means by which ] I ' ; the poor can be given power is important; effective demands by the | rural poor will probably be an elusive exception rather than a real ity. Thus it is imperative to focus on international donor agencies' ; problems regarding implementation as part of a strategy to combat a ! I coalition of indifference to the poor. Even national commitment to : PHC and TBA programs needs international support. There is a need to j I I I go beyond the rhetoric of recently acclaimed solutions for management , I and implementation problems and to acknowledge that these problems ' j are ubiquitous, particularly in donor bureaucracies that lack incen- ' I tives to facilitate implementation of development programs. ! The chapters on nonformal education (Chapter V) and implemen- | ! tation (Chapter IV) are as germane to the analysis of TBA programs | as the chapters on case studies (Chapter VI) and literature review 38 (Chapter III). The reason is that the issues regarding successful j I implementation of TBA programs— especially on a large scale— have not I I been researched extensively, and data on the programs are scanty at I ! best (particularly data on supervision, evaluation, and the PHC 1 context in which many TBA programs function). i I I The literature on TBA training documents no national efforts , to reorient health-care bureaucracies in developing countries toward j I local needs. Now being considered, however, is a WHO (Note 3) project ' request for a partial reorientation in the context of research and ! support of TBAs (including the training and supervision of TBAs in j maternal and child health services and family planning services). j I I I ' I This requested research still will not deal with vital health-care I system issues that a PHC approach entails, especially in reallocation j 1 I of health-care resources. j I I In this dissertation, two examples of national reform efforts I I I (both in Indonesia) are considered, because the success of a national i ! or regional TBA program requires bureaucratic reorientation. One ' I of these Indonesian national reform efforts will be considered in ^ Chapter III of this dissertation (in the section on PHC) and also in | I Chapter V ("Nonformal Education"). These reform efforts were reviewed ' i 1 to determine the lessons learned from national reform experiences. i Indonesia was chosen simply because these two reform efforts were the I 1 best documented and because they confronted issues of reform in j I health-care bureaucracies in developing countries which are pertinent I to TBA program reform needs. Because supervision and support of ! I auxiliary health workers is such a crucial element and entails such 39 i i massive bureaucratie reform efforts, and because there are no studies ! I of "learning-by-doing" in a national formal health system except in I ' Indonesia, this study is a beginning attempt to fill that gap in the ; TBA training literature. The full quartet of factors— sociocultural, political, ! economic, and managerial— come into play in national bureaucratic ! reform efforts. The two Indonesian national reform efforts are also instances of attempts (1) to merge information from the local level j into central bureaucracies and (2) to use the technique of learning- ! by-doing," an approach advocated throughout this dissertation. The issues these reform efforts raise are also part of the context of the problem areas of preimplementation, training, and postadoption. ! One crucial feature of the systems model is that it forces \ issues of TBA training out of the immediate parochial concerns of the | health-care profession into the concerns of a wider range of profes- ! sions. Every profession concerned with development, whether in ! agriculture, medicine, or economics, wants to see its own area of interest as the legitimate boundary for its own activity. Health-care j I ! I professionals are willing to acknowledge the existence of policy prob- j I lems or administrative problems, but not that these problems are in ! their domain. A medical or health-care professional often acts as if i I he or she is in a Third World country only to transfer health techno1- | ogy and assumes that this can be done as it is in Europe or America, * where the technology can be transferred because the health-care delivery system is set up. In the developing countries, however, 1 ' this is very often not the case: in a sense, one has to redevelop the 40 technology. Doing so entails dealing with the quartet of factors in the systems model that are part of the task environment; this is one crucial reason why nonformal education is so critical a concept in TBA training, supervision, and linkage to the formal health-care system. Chapter VI deals with the actual testing of the systems model with case studies of TBA programs in Bangladesh, Brazil, Peru, Ghana, and Pakistan. The final chapter (Chapter VII) deals with findings, conclusions, and recommendations. The recommendations cover a much wider spectrum of concerns than analyses of TBA programs have done up I to this point because this study's conceptual framework for TBA pro- I * gram assessment takes into consideration more variables than other donor models that have been used for the same purpose. n 41 CHAPTER III REVIEW OF LITERATURE j It is essential to discuss primary health care (PHC) because I the training of traditional birth attendants (TBAs) is part of a ! strategy to extend health care to the rural (majority) populations in jdeveloping countries. Using the usual ministry of health (MOH) pat- I terns to do so is not an option, because the required resources— both j funds and trained personnel— are scarce. Logistically and function ally, the formal health-care systems cannot meet the needs of the j rural populations. To focus on TBA training exclusively also is not an option: TBA training must be a part of a multipronged PHC effort. Although this study does not focus on PHC per se, the concept of PHC must be discussed (1) because it is framework that encompasses TBA training programs and (2) because the issues raised by PHC strategies I directly affect (and often are the same as) issues raised by implementation of TBA training programs. I Strategies for Primary Health Care ' The basic strategy for PHC includes wide distribution of a limited number of highly effective preventive and curative technolo gies along with communication and education designed to lead to more I self-reliance in health care. Most existing formal health-care aystems take an opposite approach in almost every aspect of operation. Mosley (Note 4) set up a "conflict-of-values" chart that illustrates 42 the difficulties that face country-wide or regional implementation of PHC. The traditional health-care syst^s referred to in this chart are Western medical systems. The professional elitism, the sophisti cation of services, and the centralized MOHs in many developing countries must be revamped to accommodate PHC and the "health-for- all” strategy that many countries have supposedly accepted but are not committed to practicing. Mosley’s (Note 4) "conflict-of-values" chart (Table 2) illustrates the divergent approaches. The Institutional Approach An institutional approach to PHC is considered in the American Public Health Association’s analysis of 52 projects assisted by the United States Agency for International Development (U. S. AID) I (Parlato & Favin, 1982). The broad goal of these projects was to give basic health-care services to all people by working through and modifying the health-care systems. The projects generally had two components: (1) using the village health committee and village health worker and (2) strengthening the MOH. These were centrally planned efforts to reach rural populations, who were at best marginally served, and they entailed both involvement of local people and exten- I sion of existing health-care systems. These projects accented curative services, because people would and will pay for them, and added a little preventive service. I The projects’ successes were in training people, putting them Iinto the field, and building physical structures. The difficulty was 'supervision— these trained health workers could not be supported in the scattered rural areas. The needs identified were for more 43 Table 2 Conflict-of-Values Chart Standard Primary health care Traditional Western medical system in developing countries Reference for value Judgement Health for all people Self-professionalism | of medical profession i Source of priorities Local needs i Good Western medical practice Educational strategy Brief/pract ical/ behavioral Prolonged/theoretical biological Orientation of practitioners Open/self-critical Closed/defined Quality of goals Coverage of population Sophistication of | technology/faculty i Communicat ion Expected response Two-way/1istening Cooperation Top-down/authoritat ive Obedience Program strategy Population based/ process oriented/ how to do Institution based/ technologically oriented/what to do Evaluation Results produced Procedures/activities done 1 Existing systems Support/expand skills Undermine/limit or i reduce skills ' Impact Strengthen init iat ive/produce autonomy Weaken initiative/ increase dependence Note; From Mosley (Note 4). 1 i 1 1 1 J - - ■ - 44 ; ' planning and more effective management (i.e., of logistics), and i primary concerns were the solidity of institutional structures and . the reliability of financial support. The cause of the problems was seen as an overeng)hasis on a "top-down" approach, and the solution 1 I proposed was to provide more training and more funds to get the i ! communities to be autonomous. ; I Elements of the institutional (centrally planned) approach to PHC are important. Good supervision and training are vital, but their j I ! ! effectiveness requires genuine local input. Even if knowledge of I I I local systems is scanty, a real linkage between local and institu- ' tional systems is needed to permit the flexibility and adaptations I that are vital if PHC is to be effective. As mentioned previously, i I ! ! the PHC effort in China is being strained because the economic and political situation has changed and no longer accommodates some elements of the PHC effort (e.g., the "barefoot doctor" and the rural | I health insurance). j Primary health care may not be the final answer, but there is I a need to generate health-care systems within the cultural, economic, I ,and political milieus. Cuba provides another example of a welfare ! system with a very efficient PHC program, a program so successful ! I I 1 that infant mortality rates and other indicators of population health ; in Cuba are compared not to indicators for other nations of the 1 I I Caribbean but to those for nations like the United States. "Top-down" approaches and their bureaucracies are perpetuated because they do fit I an economic and social system. The point is that the traditional ' I health-care systems are in conflict with PHC systems more oriented to , 46 1 the rural populations. Thus if a country opts for PHC, it must : realize the constraints to be overcome and commit itself to a PHC approach that entails integrating local inputs into a central j health-care bureaucracy. ' The Technological Approach I In addition to the institutional approach to PHC, the other 1 I approach considered in the study of 52 projects assisted by Ü. S. AID i (Parlato & Favin, 1982) was the technological approach. Basically, this approach involves a powerful (but simple and cheap) technology I to bypass the institutions and deliver health-care services to the ; people. The need for health-care systems is minimal: let doctors and i nurses perform traditionally. In this approach the formal health- i : care system and services back up the PHC projects; the projects can call on the formal system if needed, but such requests are kept to a : minimum. Family planning can serve as an example of this essentially "vertical" approach. The technologies exist; there is an unmet need; the technologies can meet this need; and the PHC project is intended I I , to deliver the technologies directly to the people in need. The ' philosophical thrust of many family-planning projects is to prevent births, and such projects are not fundamentally concerned with | reproductive health care. Operationally, therefore, the thrust is ' not health care. Nonetheless, a simple technology in family planning has multiple benefits. i I Other technologies often chosen in PHC programs are oral i i i ; rehydration therapy, immunization, and growth monitoring. The issue I 46 ! is the use-effectiveness of these technologies, and not just their i effectiveness. Problems include delivery of these programs (most t j pressing), acceptability of the technologies to local populations, I their use once accepted, and their demographic effects. In immuni zation strategies, community organizations are asked to meet extraordinary expectations. With measles vaccine, for example, children between the ages of 7 months and 9 months must be found and vaccinated, although it is difficult to find children even at birth— an event seemingly unique and easy to identify. Growth monitoring also presents a problem, one that is hard for college students and other educated persons to grasp: the difficulty of communicating the concept of grsgphing and other growth-monitoring techniques to an often illiterate target population. The issue is technological appropriateness. The effectiveness and effects of technologies considered for PHC are unresolved. Technologies often advocated may have bad effects (i.e., medical malpractice, misuse of drugs, use of bottle- feeding rather than breast-feeding). Technologies that are generally I neglected (such as those needed for clean water, latrines, and I sanitary habitations) could have multiple effects on other techno1- ' ogies, such as family planning. The recommended strategy of safe I I water and basic sanitation, for example, has several dimensions: it promotes personal and community hygiene, provides a safe drinking- water supply, provides for disposal of excreta, protects water resources and enables surveillance of drinking-water quality, and can be linked to other, related disciplines (WHO, 1981, p. 18). 47 I I Other Approaches j Another e^proach to PHC is a type of population or community 1 strategy (Mosley, Note 4; WHO, 1982a). Understanding this approach I involves the concept that the apparent problem may not be the "real" 1 i problem. In a community, for example, disease may be not a problem | j but a symptom; disease may reflect what is going on in the community. ^ I Questions (fdiat, who, when, where, how, and why) need to be answered, ! I I i and when the "why" is known, a system is developed. This problem- | ! ; j solving approach requires skills involved in getting information that i I I is both vital and usable. The population strategy focuses on two i ; sets of skills: (1) epidemiological, a biological understanding of 1 the population; and (2) sociological or anthropological, a behavioral I I understanding of the population plus the ability to work with the I people composing it. There is a critical shorteige of personnel in both of these skill areas, and a framework is needed for the connec- \ ' tion of epidemiological and sociological skills to PHC programs. | • The new institutional approach (Mosley, Note 4) has been | I likened not to a branching tree but to a seedling— it has growing capacities at the peripheral or grassroots level. An institutional | I framework is essential because of the need for self-sustained efforts j to learn how to change and adapt institutional change. Learning how ! to be effective and efficient and how to expand services is empha sized. Three administrative/institutional principles for the learning i process are to embrace error and learn from it, to link knowledge to I i j action, and to plan with people (Korten, 1980). The same principles : can be applied in the epidemiological/anthropological approach. f . 4S1 ^ The approach is crucial, for there are no blueprints and thus no "top-down" strategies for support, supervision, and community : services in the context of PHC. People in a community can be I j ' utilized, but how, for what, and why are questions that depend on . locality. Some communities may not need village-level workers; they I i I may need different, more specialized skills. | I Problems A major problem with PHC as it is actually implemented in , developing countries relates to health-care systems in the developed I I countries. Prestige, job security, monetary rewards, freedom of I choice, and geographic mobility all are part of the health-care I professional’s environment in the developed countries. The "experts" j I in health care all come from this system and have real difficulty in ! I "changing hats" and focusing in on the health-care needs of rural populations. The health-care cadres in the developed countries have I I little experience in dealing with health care in a context where ! i I reward systems are different. During two trips (January and February j 1984) to the World Health Organization (WHO) headquarters, several ■ I physicians working for WHO said they believe that for the most part, I I WHO physicians are really not convinced about the social approach to ; ' health care that is embodied in PHC. j The following list of international constraints to PHC was j ; 1 gleaned from interviews with WHO personnel, February 23-26, 1984: 1. The scarcity of more international (e.g., WHO) personnel in developing countries; 1 I I î------------------------------ 4 9 ! 1 ( 2. The preponderant influence of medical doctors who are not ; convinced about a "social" approach; and ! 3. The reluctance to raise more funds (resources for PHC are scGurce and WHO’s past efforts to raise money have I succeeded). | i National constraints to PHC include: i 1. The lack of political will; 2. The unfavorable allocation of resources; I 3. The lack of coordination between sectors (i.e., the I I multiplicity of health-care delivery systems that do not communicate with one another); and 4. The lack of incentives for health-care system personnel who , ! I ; work in PHC. ; These constraints operate to lessen the chance that health-ceure . advisors will, if necessary, be motivated to challenge the will of I governments. The technicians chosen as experts often are not experts | ' I ■ in setting the stage for a workable health-care system. Goldman j 1 (Note 5) cited Briscoe’s claim (in Briscoe, J, [1980]. Indications ! of change in developing country mortality trends, the end of an era? 1 ! Population and Development Review. 6[4]) that the PHC approach is | I I I unlikely to be successful, because many governments are not seriously | committed to it. According to Goldman (Note 5), Briscoe argued that ! I comprehensive PHC has been successful in reducing mortality when a ' strong and lasting commitment has been made to it, as is the case in j ' China. He claimed that the standard excuses for failure of PHC i I programs— scarcity of financial resources and lack of political 1 50 t will— confuse the real problem. Briscoe stated that adequate i ! financial resources for PHC are available, but most of the resources I ; for health care will continue to be spent in large urban hospitals I catering to the curative-care needs of the middle and upper classes. I I I He stated that under these conditions, selective PHC will be no more I effective than comprehensive PHC. Furthermore, Briscoe argued that I : I selective PHC will not significantly reduce mortality. | Implementation In light of the above, PHC support should be directed only to | countries that show a serious commitment to the health of the rural population. That the rate of mortality decline in developing countries is leveling off is well known (see, for example, Goldman, | Note 5). This leveling off is regarded as a dual phenomenon | resulting from both (1) a lack of capacity (increasingly viewed as a j lack in the capacity of institutional formal health-care systems to ! deliver health-care services to the rural population) and (2) a I slowdown in health-related economic and social progress. Diarrhea, ! pneumonia, and malnutrition, for example, are common problems in the developing world and are less susceptible to previous public-health interventions. Mosley (1983) claimed that the complete PHC concept frequently is not being implemented and that PHC programs often are i I really no more than workers delivering new technologies. Other i central components of PHC as envisioned at the 1978 Alma-Ata ; conference— such as the use of traditional practices, intersectoral coordination, and community participation— generally are not included j in PHC in developing countries. : 51 ! I These critiques of PHC do not dismiss the PHC concept; rather, ! they highlight the probable failure of PHC programs because of weak ^ political commitment and overly narrow concepts of a cause-of-death I model. Only for a specific disease that has both high relevance and I ‘ I high fatality among otherwise healthy persons (e.g., neonatal tetanus I and falciparum malaria) would a selective PHC approach be effective. • In many cases, unwarranted assumptions are made about the ability to j implement PHC programs. If technicians were given minimal sociologi cal/anthropological information about local communities, efforts often would be more realistically directed toward PHC implementation I I ' I ' ! considerations. Thus, entire programs devoted to preimplementation I ' considerations in PHC may well be needed. Many developing countries ' ; simply are not at the stage where they can consistently support and i supervise health-care workers who are in remote areas or are insuf ficiently committed to PHC. I In an American Public Health Association study (Pyle, Sabin, , : I : & Martin, 1982) of the problem of expanding PHC, four needs were I ! stressed: j ■ 1. Mobilization of political constituency; I 2. Svpply to, and communication with, remote health-care posts; j 3. Realistic, task-oriented training and supervision for periph- I I eral health-care workers; and | I I j 4. Organization and stimulation of communities to participate in ! ! preventive and promotive health-care activities. The report on this study (Pyle et al., 1982) addressed several other important points: r 52 1. Participatory management, supportive supervision, accounta bility, and community participation may or may not be possible in bureaucratic environments (p. 8); 2. In many countries, health authorities are unlikely to spear head programs based on multisectoral coordination (p. 110); 3. Community participation and mobilization is needed where tsnpical medical services are not a large factor; 4. The requirement of PHC for social engineering implies a need for more social-science expertise than is traditionally found in most MOHs; and 5. Any large-scale PHC program has to be supported either exclusively or partly with tax money. The report (Pyle et al., 1982) also stated that: 1. A PHC program will use bureaucratic structure and norms; 2. It will be oriented toward procedures as opposed to results and toward roles as opposed to tasks; 3. It will emphasize financial propriety and not cost effective ness (p. 12); 4. The problem of expanding PHC may involve scaling down (inputs and expectations) as much as scaling up (coverage) (p. 18); and 5. Large, sophisticated evaluations are time-consuming, expen sive, and usually inaccurate in delineating the cause-and- effect relationships in PHC schemes (p. 35). Primary health care in practice is a complex process representing a major departure from traditional health-care systems (i.e.. Western- 53 style health-care systems). The technologies required to reduce i infant and child mortality rates, for example, are clear, simple, and not inordinately expensive. The problem, however, is not so much what services to provide, but how to deliver services to the needy in the community. This problem is primarily organizational; both the ! operational and the organizational implications of the PHC approach | i must be identified and addressed before an effective program can be ' launched (Pyle et al., 1982, p. 39). These operational and organiza tional implications are often ssmonymous with the preimplementation considerations of PHC programs. It is not unusual to find administrative systenm* that accent ! ; , the control of transactions rather than the enhancement of functions. [ 1 Essential support services might include supplies and drugs; transport ! of patients and materials; communications between service sites and I I local health-care centers; hiring and payment of health-care person- 1 nel; maintenance of vehicles, equipment, and buildings; and recording : and reporting of essential statistical data. A reorientation of i formal health-care personnel, including doctors and nurses, is criti- ! cal for the development of realistic PHC strategies. Some built-in 1 measures of program success are necessary as performance indicators. i I Coordination and collaboration should be specified and designed for | j each level of program responsibility. A schedule for accomplishing | ! tasks should be provided. I I Eight essential elements of PHC strategies were suggested by I I WHO: I 54 I 1) Education concerning the prevailing health problems and methods of preventing and controlling them; 2) Promotion of food supply and proper nutrition; 3) Adequate supply of safe water and basic sanitation; I 4) Maternal and child health, including family planning; I 5) Immunization against the major infectious diseases; I 6) Prevention and control of locally endemic diseases. Examples: a) malaria, b) hypertension; I 7) Appropriate treatment of common diseeises and injuries, I Examples: a) diarrhoeal diseases, b) common accidents in the home; j 8) And provision of essential drugs. (WHO, 1981) I Linkage to sectors outside health care is not mentioned in this I document (WHO, 1981), and there is no attanpt to merge the eight ■ elements horizontally at each level of program responsibility. ; Nonetheless, it is necessary to give an overview of the essential elements in a PHC strategy, because the hostile environments in which : people live— the social and biological synergy— militate against the i ! effectiveness of one single intervention. That is one important reason why opting for technologies that have multiple benefits makes the most sense in terms of the effectiveness of PHC program implemen tation. Both the strategy of a PHC program and its fit into the , organizational and political structure of the formal health-care system obviously influence the performance of the PHC delivery systems. Either the strategy can be modified to fit the structure, or vice versa. More realistic is the adjustment of both so that the I structure and strategy are more congruent. Structural modifications I include bureaucratic reorientation which enables the agency respon- j sible for implementing the approach to adopt more of the process I factors responsible for project success (Pyle, 1982). _J r ^ " 551 , One example of an innovative project in PHC attempting the I population approach is Indonesia’s Comprehensive Health Improvement I Program— Program Specific (CHIPPS), which has the goal of helping to I I deliver more effective health-care services to outlying districts I (throughout this dissertation, the term "district" is used to denote | I ; I the lowest level of local government and decentralization; actual I names for this kind of unit vary considerably from country to coun- | try). The basic objective of CHIPPS is the institutionalization of an epidemiological and problem-solving approach. The present Indonesian health-care system is highly centralized in the capital city, Jakarta. Although CHIPPS is trying to change the health system, I it is extremely difficult to change a system if you are restricted ; I to using it exclusively to implement the program; only grant funds ; I permit the flexibility required for innovative programing to I administrate the benefits of a problem-solving approach such as j CHIPPS. (John Snow, Inc., 1984, p. 7) I The above is extremely important because of the need to change the system or reorient it more towards a PHC s^proach. Donor-provided I resources— funds and personnel for PHC efforts— are essential if the I health-care system is to become more responsive to the needs of rural ( I i ' : people. Working within the system is obviously essential if the goal j ! I ' is to devote more resources toward PHC efforts. Furthermore, CHIPPS j ' is attempting to solve an important implementation issue: how to I ; reconcile its conceptual framework (which allows "bottom-up" planning based on problem analysis of the local situation) with national policies and guidelines and the standard budgetary process. Most j Indonesians working in the MOH view CHIPPS as a source of funds in a ! I I time of acute financial scarcity. j 56 ^ Several other findings of the evaluation by John Snow, Inc. (1984) are pertinent: 1. It is important to realize that a crucial shortcoming of the project is the absence of an advocate in the MOH. In retro spect , U. S. AID/Jakarta now believes that a contractor should have been placed in the MOH Bureau of Planning to assume respon sibility for day-to-day advocacy of CHIPPS and to facilitate resolution of the many managerial/administrative difficulties of the project (p. 7). Such a contractor could help strengthen project implementation at the district level, develop monitoring systems, identify appropriate process indicators, and document action in the provinces. Communication of these issues to MOH officials is vital. Although no such provision is in the project, a centrally placed advocate should have one or more counterparts closer to the field work. Well-placed individuals, recon^nsed for extra time spent with the long-term contractor, could be invaluable in getting help for the overall effort from the minority of Indonesians who understand the need for a "bottom-up" strategy. 2. The term "epidemiological approach" must be demystified and seen simply as a problem-identification approach that will lead to problem-solving resolutions (p. 24). Everyone involved in CHIPPS must try to combine community participation with health care activities and try not to emphasize one at the expense of the other. 57 3. How to institutionalize community participation and to determine who is responsible for supervising this activity are questions still unanswered (p. 23). Likewise, the people responsible for CHIPPS implementation have not been designated. Certain village resources such as PKK (a village women’s move ment) are helping in this vital problem area. 4. It is necessary to resolve the problem of centrally planned budget allocations that do not take local considerations into account. 5. Centrally funded packages for PHC need to be augmented, and funding increases should be linked to more effective responses to local needs. 6. A greater understanding of the inactivity of the kader (village-level health-care worker) is needed. 7. More involvement of local medical students in project activities is essential. A greater education and communication effort to explain CHIPPS activities to provincial officials is needed. 8. Provincial data-collection systems need to be better able to identify the nature, extent, and location of health problems. Surveys and studies to identify local health problems have been considered a low priority or a "research" function rather than a management tool (p. 3). 9. Understanding of what happens at the delivery point of health-care services, the puskemas (local health center), is poor. What can the staff personnel be expected to do? An 58 operational analysis of puskemas activity in the provinces is essential. After a recent evaluation of CHIPPS, Pyle (Note 6) concluded that there were three serious problems at the center (i.e., the P^H): 1. Institutionalizing the approach is difficult. The attitude of the Indonesians is to put the CHIPPS approach in the "bluebook"— that is, to put it on the list of foreign donor activities. 2. The provinces cannot get budgets for surveys. 3. The provinces cannot get budgets for training. The central formal system is "cast in concrete." The dimensions of this problem become obvious from the size of the P^H of Indonesia (it employs more than 5,000 people; in contrast, the MOH of China employs about 200 people). Changing the system will require a long-term process of reorientation, taking perhaps 15-20 years. One or more facilitators could accelerate the process, but it would still take a I long time. Additional help could come from the tremendous amount of I money available for local budgets from several sources: (1) the United States Centers for Disease Control (funds only for supervi sion) , (2) the sale of drugs, and (3) the registration fees from puskemas. Using money from these sources for surveys and training closer to local spheres of operation could help in gradually reorienting the "cast-in-concrete" system of the MOH. Many of the problems in the MOH of Indonesia stem from the nature as well as the I ! size of the health-care bureaucracy. The "conf1ict-of-values" chart 1 presented earlier in this chapter is very applicable to the Indonesian ; 59 j I case, and the lines of conflict seem to be more sharply drawn as a : result of CHIPPS activity. Thus an entire cadre of MOH personnel should be undergoing orientation away from the attitude that efforts : to integrate a "top-down" approach with a "bottom-up" approach are in , the domain of the "bludsook" (i.e., the donor agencies). ; ' Conclusions ! I ! ! Understanding and upgrading health-care delivery-service ; I systems in outer (rural) regions is critical if PHC strategies are to | be an effective part of an overall health-care system. Understanding | 1 and integration of international, national, provincial, and local I environments are essential to better management of support services. Any bureaucracy can be considered a power structure. Even when essential resources and perquisites get scarce, workers at the I upper (national and district) levels still get salaries (paraprofes- | sionals can go four months or longer without a salary). Scarcity i I brings out tensions within a system: fuel, vehicles, allowances, and I salaries get cut. Politically, it is not possible to fire lower-level ' workers, but they cannot get (for example) tire loans for the bicycles needed to do extension work, while more centrally located workers can get tire loans for vehicles. No one wants these service positions; there is a lot of turnover and endless retraining. The need to study resource-constrained health-care access bureaucracies is paramount. What happens at the interface between a health-service delivery system and its clients? Information about what really happens in such transactions is vital for several obvious reasons: 60 - Top-down bureaucracies have a tendency to expend most of their resources at higher levels in a hierarchy, so that even apparently well-endowed systems may in fact show unexpected weakness at the base. - In poor countries, it is usually at this level that resource constraints are most evident and yet where, paradoxically, task expectations may be quite high. i - Citizens probably judge the quality of services they receive by the nature of their contact with the access bureaucracy. ! - Environmental difficulties and the ineffectiveness of support systems affect personnel in access positions more than those higher up. - Most official programs embody fairly strong assumptions about ; the likely performance of the access level - and are often | erroneous. I - To the extent that either "top-down" or "bottom-up" communica-| tions is necessary for realizing organizational objectives, ; the capacity of access-level staff becomes a critical con straint. - There is general agreement in the literature that staff at this level are the least prepared and least capable in the j system, while also evidencing low morale and a low task commitment. (Moris, 1984, pp. 59-60) Miatever weaknesses characterize the macrosystem (e.g., lack of funds for transportation, nonpayment of salaries, inaccurate tech nical recommendations, lack of essential inputs) will become evident at the contact (access) level. As Moris (1984) pointed out, access bureaucrats tend to become the objects of structured misperception, because no one reviews the total activity of even a hard-working I employee. Contact staff thus risk taking the blame for what are I I really organizational rather than personal faults. 1 In considering training, it is disconcerting to see how much of the "extension training" being given around the world bears no relationship to the real-life work settings that trainees must 6Ï1 inevitably enter. More detailed information is needed to improve the ; quality of training. There are only a few reasonably complete descriptions of devel oping countries[’] access bureaucracies. Missing from all these are the types of multi-dimensional data on cadre characteristics, , supervisory tactics, staff effectiveness, organizational communi cation, and boundary spanning patron/client relationships. For a given national system where interventions are being planned, we usually do not know whether contact staff are recruited out of their communities or are nationally mobile; we do not know which reference groups field personnel use when evaluating their own comparative benefits and workload; and we can not tell how they use their time or whom they actually serve. Considering such ; critical linkage roles such staff are supposed to be maintaining, these are very large gaps indeed. If the intention is to make j access staff more effective and accountable to clients, these deficiencies in research information must be rectified. (Moris, i I 1984, pp. 62-63) Î j In CHIPPS (in Indonesia), a salient shortcoming is a lack of knowledge | > about the local health centers. The concept of organizational produc- ! I ! tivity needs to be redefined toward effectiveness of linkage with the villages and rural populations in PHC programs. | ! Virtually all the literature on TBAs indicates that super- ' vision is crucial to successful TBA training (Davidson, 1984; Favin | I et al., 1984; Neumann et al., 1983; Walt, 1984), so questions about personnel in the health-care systan, the "access bureaucrats," like- j wise are crucial because these bureaucrats will probably be the ones ' I I who supervise and/or support the TBAs. Hirshman’s (1967) principle | ' of "the hiding hand"— i.e., being tricked into underestimating project i ! i I difficulties (in the case of TBAs, the problem of support and super- i vision) or overestimating project benefits, so that one is tricked j into undertaking a project— will be operating if this question of contact bureaucracies in the formal health-care system is not taken 62; I j into account. The question is amenable to a motivational analysis, a ; I ; ! realistic appraisal of what the rural health-care personnel can and i I cannot do. In an article on traditional health practitioners, Pillsbury ' (1982) mentioned reasons for the limited progress of these practi- j I tioners, even in the context of positive national policy. One reason I is the cultural gap between the modern aaid traditional sectors. ! Another is the overriding personal priorities relating to social and | financial situations: physicians and other civil service employees \ (of the formal health-care system) are more eeiger to advance their j I own careers and their own families’ socioeconomic status than to devote time to the complex task of improving collaboration with j I I ! traditional practitioners. More rewards (incentives) are needed for ' the people working in PHC (particularly doctors and nurses) who take : on the supervisory management of rural and other health-care workers ^ and who successfully facilitate support, supervision, and linkage to I ; I the rural population. ! I To recapitulate, the first key principle in a PHC strategy is ; I I equity. A PHC strategy must redress the imbalance created by the bias of formal health-care systems toward sophisticated, high-technology medicine and urban dwellers. The second key principle is that impor- ! tant health-related activities sometimes exist outside the control of j the MOH (such as activities pertaining to water, education, nutrition, ; and food production). Such interdisciplinary issues require more i than the establishment of ad hoc interdepartmental committees; they require lasting responses at the central and lower levels of the 63 health-care system. Community or local participation is regarded as a third key principle of PHC (WHO, Note 3 p. 9). Obtaining data on population groups or areas is important in focusing attention on the allocation or reallocation of resources as a primary concern of health-care planning. This argument is the crux of a population (or sociological/epidemiological) approach to PHC. I I ' Growing evidence indicates that an important cause for the limited efficiency of the health-care system is often found at the district level. This is the level of government in which basic PHC services : are organized and managed, and primary backup services are situated (WHO, Note 3, p. 31). Any deficiencies at the district level will ' probably lead to poor support and supervision, to shortages of 1 essential supplies, and to lack of facilities for training and I retraining of local health-care workers. Three points are made in a I I ; WHO document (Note 3) on strengthening MOHs for PHC: (1) organization ! j substantially affects performance in MOHs, especially with reference ; to PHC; (2) processes of organizational change can be planned and I j managed to improve performance; and (3) principles of organizational ; change can be applied to a "learning-by-doing" process that develops organizational innovations appropriate to the specific challenges often presented to MOHs by commitments to PHC in a specific context. Planned change processes eure long-term undertakings if seen in the context of improving the performance of the f^H and the formal health-care system. Often, a PHC approach and its actual require ments are not difficult in themselves, but ingrained habits (which are extremely hard to change) militate against the implementation of 64 PIKÎ. Persistence in the face of adversity, willingness to try new experiences, and ability to learn from past failures are all part of innovative efforts, such as PHC, that are linked to bureaucracies (including donor^agency bureaucracies). Innovative efforts such as PHC, and TBA programs as part of the PHC approach, must be geared thoroughly toward the actual demands of implementation. Training of Traditional Birth Attendants Background The practice of midwifery has existed virtually since human kind had its beginning on this planet. ! I Some of the earliest written records from ancient Middle Eastern ' cities refer to midwives. Sumerian clay tablets dating from 2500 BC describe Nintur, the "Midwife of the Gods," along with her I equipment, incantations and rituals. According to a papyrus of i the 19th century BC the birth of Egyptian royal triplets was I attended by three women who understand childbirth. (Population I Reports. 1980, p. J-439) j The term "midwife," however, includes a great variety of birth attend- I ants. The categories of midwifery personnel which are involved in providing basic health care can be classified into three groups: Fully Trained Midwife - The existence of a higher, middle and lower cadre within this group is demonstrated, although this is not applicable to all countries. a) The Trained Midwife is common to most countries of the world. In some countries it is a requiranent to have nursing training before doing midwifery. b) The Nurse/Midwife is fully qualified as a nurse and a midwife. The length of training varies in different countries and also according to the cadre. c) The Community Midwife is a trained midwife with the addition of one year of public health nurse training but without full nurse training (not applicable everywhere). ---------------------------------------------------— ggl d) The Public Health Nurse/Midwife has in addition to nursing and midwifery training a qualification in public health nursing. Recruits for advanced training for supervisory teaching and administrative positions will be found in this . professional group. Her main involvement will be organiza tion and coordination at the district level. e) The Trained Nurse grade A (e.g., Tanzania) has an integrated training, comprised of general nursing, midwifery, public health and psychiatric nursing. j Auxiliary Midwife - She may have been a traditional birth attend- i ant who has been given some training. In some areas where Î education is more advanced, the auxiliary midwife may have had ! more training in prenatal and postnatal care in addition to prenatal care of mother and baby. Traditional Birth Attendant (TBA) - The TBA is usually a highly respected member of a community and has much experience in the profession, or she may be a neighbor or relative who happens to be near at the time of delivery and renders assistance. The contribution of TBAs in the health care delivery program is considerable. (Favin et al., Note 7, pp. 57-58) The TBA is known in different parts of the developing world j as the daya (Middle East), dai (India, Pakistan, and Bangladesh), ! matrone (West Africa), dukun ba.ii (Indonesia), hilot (Philippines), partera empirica (Latin America), or curiosa (Brazil). One common ; i characteristic of TBAs, amid enormous variation in cultural beliefs j and practices, is that they are the ones to whom most poor rural j women turn for support in pregnancy, childbirth, and early child care. | Unlike formal health-care practitioners, traditional midwives are concentrated among the poor and in rural areas. In Egypt over 10,000 dayas serve mainly the rural and urban poor. Of the 38,000 hilots in the Philippines, over 90 percent live in rural areas (Grant, 1985). In India, the goal is to train over 500,000 dais | during the 1980s, and an estimated 600,000 dais attend 80 percent of I all births (Grant, 1985). Kumer (Note 2) estimated that well over 90 66 I percent of births in India are attended by TBAs. These figures are I partly a reflection of the relative numbers of physicians and tradi- ! tional midwives, as shown in Figure 4 for nine developing countries. Stated simply, the TBAs provide the mother-and-chiId health services I for the vast majority of the poor. In many poor communities the TBAs are, therefore, one of the greatest of all potential resources for mothers; they provide knowledge and practical, on-the-spot help in protecting the life and health of the children (Grant, 1985, p. 61). In some countries, interest in training TBAs has existed for quite a while— since the 1920s in Sudan and since the 1940s or 1950s in Peru, Thailand, and Indonesia. In general, the literature points to an increase in training TBAs to attend normal deliveries and refer I complications, to perform in new roles, or both (Burns, Note 8; j Davidson, 1984; Favin et al., 1984; Grant, 1985; Owen, 1983a). Since the 1970s, attitudes toward TBA programs have been changing. In 1972 only about one third of the developing countries offered governmental I or official recognition to TBAs in the form of registration or train- I ing. Today more than three quarters of all developing countries have ' training programs for TBAs (Grant, 1985). Pakistan is basing its ; I ! Accelerated Health Program on training TBAs. Bangladesh will try to I I I have one trained midwife for every village by 1985— a total of 68,000. i I ! j Currently (1985), the World Bank is deciding whether or not to use I ! I monetary incentives for TBAs as an inducement for them to work toward reducing fertility in Bangladesh, where population increase is a serious hindrance to development in that country. In India, over 500,000 dais have been registered and given basic training. These 67 ! India X 0.25_________________ /////////////////////// 9.3 Pakistan Thailand Kenya Sudan XXXXXX 2.43 //////////////////////////////////////// 16.6 Philippines XXXXXXXX 3.22 LLU U JL L L lim U JU l 8 XXXX 1.26 ///////////////// 7.1 X 0.16 /////////////////////////////// 12 i Afghanistan X 0.4 //////////////////////////////./// 13 XX 0.8 u u /iiu u i u m i 7.6 Colombia XXXXXXXXXXX 4.7 /////////////// 6 Haiti XXXX 1.3 /////y//////y////////_////////////////y/y////////// 20 Key: XXXX Number of doctors per 10,000 population //// Number of traditional midwives per 10,000 population Figure 4. Availability of Doctors and Traditional Midwives. (From Grant, 1985, p. 61, as adapted from Population Reports. 1980.) 68 ; programs are Just a few of the many that exist throughout the developing world. The TEA usually is well known to the communities and the poor f I I women she serves. She already enjoys their trust and confidence and I she is not a threat to these people, as is often the case with other I : health-care personnel. Rural women feel comfortable and natural in ' sharing something as intimate and personal as child-birth with a TEA who speaks their own language and knows the subtleties, attitudes, beliefs, and behaviors of their own culture. Even when modern health care is available, women often go to the TEA because of the feelings of trust and confidence she gives them. The attitude and interest of I the formal health-care system— clinical and objective— is a foreign i I I notion of many rural women. I The majority of TBAs concern themselves with the needs of the I mother herself. The TEA often helps with child care and can cope I with the new mother’s emotional life, family relations, rituals, I prayers, and social obligations because the TEA is part of the same culture as the women she serves. Cooking and helping with household ! chores are often part of a TEA’s service and further enhance her role I in the lives of rural women. A midwife in a village in the Nile | I delta has been described as follows; I Zeinab accepts whatever payment is offered her in either money or I kind. She says that she would fear God’s punishment if she did I not respond to the calls of the poor before those of the rich. I (Grant, 1985, p. 63) j The training of TEAs can include low-cost methods of child j ' protection that have the potential to reduce drastically the incidence j I 69] of child morbidity/mortality. One of the most important low-cost | services is oral rehydration therapy. The simple formula of water, ^ salt, and sugar can prevent millions of deaths from diarrheal dehydra tion, which is the world’s number-one killer of infants. The problem with oral rehydration therapy is how to make it known and available to those who need it. Many doctors, nurses, community health workers I I (including TBAs), and pharmacists do not yet know about— or do not accept— the breakthrough of oral rehydration therapy. Although this , lack of knowledge or acceptance makes the task of implementation much ^ more difficult, the training of TBAs (to whom most poor women already : j [ ; turn when bearing children) has the potential to be one of the most I ; ^ cost-effective methods of disseminating innovations to the vast ; I I ! majority of families in the developing world. j 1 Often the formal health-care system (physicians and health- ; care planners in particular) has little regard for traditional medical | practitioners, including TBAs. These physicians and health-care plan- i ; ners argue that modernization entails the elimination of the TEA as i I an anachronism and an impediment to modernity and to science. The : I modern health-care system often regards TEAs as superstitious and ignorant practitioners of unscientific and unhygienic methods relating to pregnancy. Commonly they have been accused of performing dangerous abdominal massages during pregnancy or of advising mothers to t discard colostrum which precedes breast-milk (and which contains valuable immunological properties), or of failing to ! cut and dress the umbilical cord hygienically, or of advising ! mothers to eat less in pregnancy, or of not knowing what to do in the 10 to 20 percent of births when complications arise. (Grant, 1985, p. 63) J ; 70 I It is said that TBAs fail to wash their hands, wear dirty clothes, 1 conduct unnecessary vaginal examinations, pound or push roughly on I the abdomen, fail to recognize breech or other abnormal presentations, ; pull on the umbilical cord to extract the placenta, delay too long in I referring or transporting women with serious complications to medical 1 I facilities, and— above all— use dirty instruments and dressings on ' the umbilical cord, thereby contributing to often fatal infection and tetanus in the infant (Population Reports, 1980). Modern health-care practitioners have in turn been criticized by traditional midwives: Although they do not write articles in scientific journals, traditional midwives have complained to anthropologists and others that physicians and hospital-based obstetric services use dangerous chemicals to induce labor, place women in awkward and immodest positions, prohibit family and friends from coming to provide psychological support, use forceps carelessly, perform too many Caesarean sections, and expose women to all kinds of unnecessary procedures and sicknesses. (Population Reports « 1980, p. J-448) Obviously, cleanliness and aseptic techniques are beneficial, but it is not clear which practices of traditional midwives are dangerous. There is little scientific evidence that hospital delivery is safer than home delivery (Jordan, 1984) or that massage necessarily is dangerous. In the experience of Araujo (Note 9) with training of TBAs, going to hospitals caused TBAs to pick up bad habits that they would not otherwise have learned. Haire (1972), among others, has analyzed the lithotomy position (lying down) and finds that it is not only anti-gravity but permits the fetus to put pressure on the inferior vena cava and aorta, thus affecting the blood supply to the uterus with poten tially negative effects. Some hospitals now have especially designed accouchement beds which can be adjusted so that a mother I 71 can deliver safely in a variety of positions, including a seated I position. (Harrison, 1983, p. 245) : The lying-down position is the medically preferred one, and although I the majority of TBAs support it, they are more permissive in allow- ! ing the mother to lie down, sit down, squat, or whatever she wants to I ' do. External cephalic version, the transabdominal manipulation of a : breech-presenting fetus into a head-down position before birth, is I I common in traditional societies as well as in modem, scientific obstetrics in Europe. As Jordan (1984) pointed out, advocates as well as opponents of external cephalic version hold very strong I opinions about the procedure, and each side marshals equal amounts of I prelimineury and conclusive evidence for its position. In contrast to the procedure’s critics, its advocates tend to be experienced practi- | tioners with low complication rates (complications at births). j I I Jordan (1984, p. 65) concluded that problems such as the above | I may lie where the science of medicine shades into the culture of I doctoring. In the developing world, opting for a Caesarean section | I entails risks such as the inability to breastfeed, the economic burden j I for the family, the disruption of culturally prescribed postpartum j ' I I behavior patterns, the social stigma attached to having a baby by i ' surgery, and other culture-specific factors (Jordan, 1984). Breast- | I feeding is a practice that is not condoned by some modem health-care ! ' practitioners, even though it is known scientifically as a unique and | j effective source of nutrition, and it reduces the risk of infection ; I via dirty bottles and rubber nipples. The main point is that many j ! practices of TBAs are either neutral or actually beneficial, but , 72 nonetheless are criticized because of the bias of formal medical practitioners. The perceived dangers of utilizing TBAs can be used as arguments against training TBAs; Table 3 lists such arguments, and gives an answer to each. Most TBAs are women, and in many Third World countries women have a decidedly inferior status. Some people argue that TBAs thus j have no hope of political power and, further, that the deplorable j state of maternal and child health will persist under circumstances i I of harmful cultural practices; gross social inequalities based on | I social, ethnic, religious, and caste differences; and inadequacies in j basic social services. It is pointless looking to TBAs to reduce the overall maternal i and infant mortality rates when the root causes lie elsewhere - in | socioeconomic deprivation, in top managerial incompetence, and, , most important of all, in mass illiteracy. (Why retrain tradi tional birth attendants. January 1983, p. 224) I Regardless of the beliefs about whether or not to train TBAs, it is ' clear from the literature that the training and upgrading of [ traditional midwives will be an important dimension in most of the health development plans of Third World nations. Stage II Preimplamentation Before the actual training of TBAs commences, important ele ments of TBA programs need to be considered. Rarely is there already a commitment on the part of the formal health-care system. At the university in Fortaleza, Brazil, all students in nursing, medicine, social work, and dentistry receive part of their practical training in a program to train and support TBAs which is an extension of the university’s Health Sciences Department (Araujo, Note 9). 73 I o > -o: C D C O ce es to ce -o: ce it3 to C O to o w on 0 3 O n -ce to _c e on C O - < s : 4-4 40 ce C O O 0 3 o 0 3 & jO CL -ce on 3C to ce on on on es o ce Œ "O to -j= C D to to - o -ce on •*« o 0 3 C O on es to J — . 03 -ce 3 on ce e o e to 4 ^ o -c - o to C O on on 0 3 0 3 -ce -X3 ce - o < e > ce 0 3 0 3 to to 4-4 to 0 3 M c e e to ce 4-4 on es o to to G D > o o o o on ce ce ce C L to - o L- & ce - o o o -ce on - o C D on 0 3 & CL to "£ O to C D e to 0 3 0 3 4-4 "C ce on on c c to o > f— e 4-a C D on 03 on on on - o > jOt C D -ce "C S on to C D on - o 4 ^ on 0 3 O ce ce on on on 0 3 C D CL CL < - > 3 03 C D o o O - o ce ce ce 0 3 to on 03 ce e jO on O on es 0 3 > ■ jO ”0 on O n to C L on C D to sc on •< c ce O o C 3 D ce cn -ce C L •*« to CL Cn T 3 -r-H S c= O L > f O t o .jO C D • * - 4 X o cc o o> Su c e 4 - ) cz to on 0 3 * 4 — rs to . — I o @ 4-4 t _ ) 4-4 on on to CL ce CD O o o to i*- It3 CL 0> -O 3-^ O 03 on 03 CL on on - s a : o • 4 — - r - 4 u a 03 **0 , —I e z 03 to 03 O on o to on o CD -r-4 4-4 ce on 03 to ce C D 4-4 to on on ce on s L . . on * 4 — ce 03 o 0> CL ^ C 3 T > SE "*"4 to > o tn « X I e =3 _C = C= 1 fo <xi =3 cy> s _ e < L > c O D (_) S <xi x a c: (a o t n t n e (o e k *— ■ * - > ez fo - * - » O L i ( X I Z3 • - * — o - j ZZ o ez - * - » C= (D o JZZ C O o = 3 jO _ £ = ; X 3 (XI C/> ^ -C= qj ^ & â to > =3 CL C D O “ O o on 03 3c on 4-4 -o **o on on hm- on f4 to on ""4 > es 4-4 ^ tn 03 on to ca to 4-4 - r - H to c s. —T es 03 CO ce 03 to on r 74 CO f-o o C O On c z L - to o On On c z On “ O to C O -CZ c z O ) on C O On o C O o CD > to O On o 4-4 CD o C O - c OD O -CZ e C O on on to o CO C O c= On " O O C O o & - c C O -c z o On On C O CO o 43 on CD o C L c & on o M — c z C L CD - O O ) o C O On e & C O Z3 “ O CZ On jO to " O CD o “ O -cz to c z O ) On C L CZ C O on CO c z c z c C O On on “ O e o to to o to m & to c= C O “ O c z M — o c z On C O to C O On CD on 3 l C O on on on C O C L C O c z on to 4 0 e to C O o On c z & CD On o o *<c O ) O o C O -c z CZ o o On cn CO o C O On -CZ -CZ CD 40 to to 4-4 On > - On CD to On On 40 -CZ 3 O C O C O 40 C O Z3 > to e C O O on On & o o 4 0 “ O to On to @ o on CO C L C O CD to C O o C O c z c z L - cz Z3 o On On C O c z • * — 4 C O C O m " O Z3 O) o CO -<c On CD on cz On c z o cz C O CO ca M — O to to o C O On Z3 CJf C O C O **o C O C O on C O c z on C O o C O > On o 4 0 e CD to o 4-4 z z # o C L o o on o CO On o & on jO M — CD C O C O " O C O CD o z> o o e CZ to C L to " O to to O ) C L Z3 Z3 On Z3 On C L o On On e “ O <c & C O O o O M — C O -CZ c o to Z3 c z C O On 3 s -CZ M — On CD on C O to <u> - 3 O ) o On On C O C O On C O -o 3EZ C L O On On <o =3 Z3 O o Z3 " O On z> -CZ “ O C O -CZ C L on o C O On Z3 C O -o C = C O <c jO o -c z CD CZ to CD C L CD o C O <o Z3 On C O -c z CD CZ O C O c z “ O C O to Z3 -CZ % on o O CZ 4-4 4 0 3 to to to Z3 On CO -C Z to " O to VO n o c z o -c z On C O " O C O C O on C O c= -CZ on C O C O -CZ 4-4 -c z on fO C L m to o -<c 4-4 C O C L & C O 4-4 CT> 4-4 C= L - C O o “ O C L “ O o> c= t o c = -C C T ) CO *r-4 to C O C O , ; CD to CD CZ C O On CD m C O -CZ o to " - a : & 40 40 CD On L - — -CZ On -CZ on -CZ C O on to “ O on to *<n cz Z3 C L CO CO C O O C O Z3 w e zs: o o “ O On “ O C O -cz c= e to fO o On 3 4-4 to to CD On to on CD to Z3 -cz Z3 On <c o L w “ O C O on cz C O CD On to On ll. c= -CZ Z3 -o 3 O o on & " O o c= On On On o CD > C O Li to o " O Z3 c= to On C O On On on -<c On C L cn to o 1 — Z3 On <=> Z2Z I_____ J ....... 75! I Specially prepared medical personnel train the TBAs and provide some | secondary-level health care at the rural units. It takes about two i years to orient a doctor and one year to orient a nurse to this ; approach to medicine, which includes gaining the confidence and I ; respect of the poor people (TBAs included) the doctor or nurse is j : treating. Real commitment is needed in the formal health-care system. : 1 ! 1 Maintaining respect for all local customs and traditions is | part of the preimplementation stage for any PHC effort but is | I especially important in TBA programs. A strong base of national and | local support is necessary. Tremlett (1983, p. 109) described a j I policy document on PHC in which Zambian formal health-care practi- I tioners exhibited disinterest and contempt toward indigenous health- i care practitioners. One remedy offered was to include the study of I I behavioral science early in the training of nurses and doctors. Such | ! courses would cover the practices, customs, and beliefs of the people among whom the nurses and doctors would be working. Courses struc- ; tured so that no practical work was done before examinations would not allow the constructive consolidation of theory. Almost all literature on TBA programs cites the importance of j interaction and linksige between personnel in the modem formal health- ' i I care system and TBAs. In an article on training TBAs in Tanzania, I I I Pederson (1985) described a major preimplementation step in the TBA i program— a national working/planning group that would first assist in planning the training of trainers and then provide a continuing mechanism for TBA training and support. The participants, who met for five days, were key personnel from the formal health-care system I 76: I ' such as service coordinators for maternal and child health and family i . planning, providers, and supervisors who would be responsible for the j program. I Three major objectives for the working/planning group were the ! following: ^ 1. To discuss the current maternal and child health and family planning system and its policies and goals in terms of the ; roles TBAs could play in relation to that system. 2. To discuss key issues such as the specific tasks the TBA should perform, whether or not to provide midwifery kits, 1 remuneration of TBAs, and distribution of drugs or contraceptives by TBAs. j 3. To develop strategies for training and supporting TBAs in | maternal and child health and family planning, including j objectives, content, materials needed, training schedule, and plans for supervision and evaluation. (Pederson, 1985, p. 45) ! The American College of Nurse-Midwives International has a project I for working with TBAs in Africa. Its objective is to assist MOHs and health associations in initiating, implementing, and evaluating programs for training TBAs in maternal and child health and family planning. Countries involved in the project assess needs and then form a planning group (Pederson, Note 10). This process covers some , of the preimplementation considerations, but a five-day workshop like the one in Tanzania may be too short. ■ A longitudinal evaluation— not only of the TBAs, but also of I commitment on the part of the formal health-care-systern participants— | would be informative. ' I In practice, however, hostility or at the very least communica tions between the educated health professionals - physicians, j nurses, and formally trained midwives - and traditional midwives i is often cited as a major problem in such programs. (Population Reports. 1980, p. J-443) ! J 77l I Personal communication and observation during two 1984 visits to WHO ' in connection with this dissertation reinforced the quoted assertion: even though most WHO personnel supported PHC efforts, some exhibited hostility or indifference to TBAs and to PHC in general. 1 * Another preimplementation consideration is the need to take into account the prevailing customs and culture of the TBAs to be trained. Programs that are designed without sufficient attention to this consideration may fail, as was the case in India and Pakistan in the 1960s (Population Reports. 1980, p. J-443). More focus is needed on social and cultural factors associated with TBA programs, as well as on legal and administrative constraints. The concerns of villagers and intrinsic problems of working in the village setting must be considered and new program planning guided accordingly rather than imposing from above what sincerely I appears to be a good idea, but one that has not been tested at 1 the village level. (Neumann et al., 1983, p. 2) I Testing is also needed at the local district level. In a thesis on I TBA training in Zambia, Tremlett (1983) cited the need in preimplemen- I tation concerns to talk to all village-level women, as a result of j the consultative and tightly knit Zambian culture. The extraordinary 1 variety in the characteristics of traditional midwives, in the work I they perform, and in their position and status within their own com- I munities makes the consideration of culture and local settings crucial j ! for successful TBA program implementation and support. Training that is unidirectional, with an accent on * * upgrading the indigenous prac tices rather than exploring the possibilities of reciprocal teaching and building on the traditional practices’ * (Cosminsky, 1983, p. 143), will be ineffective and is essentially a ’ ’ blueprint” or ’ ’top-down” 78^ I approach. Disregarding the culture of the TBA— her knowledge, atti- ' tudes, and practices— often creates nonlinkage, misunderstanding, and ' tension between TBAs and their clients and the MOH (Greenberg, 1982). Preimplementation or pretraining factors include: 1. Government/Ministry of Health preconditions which must be met prior to initiating an actual training program. a. Acceptance of the concept of utilizing the services of trained TBAs. I b. Where applicable, legalizing the function of TBAs and by extension, legalizing the interaction of Ministry of ! Health and other modern health establishment workers i with TBAs. c. Administrative orders authorizing staff to work with TBAs. d. Designating certain staff as being those responsible for working with TBAs and making the necessary job I description modifications and workload reallocations j and reshuffling. I e. Allocating the necessary resources: # supplies j # support I 2. Prior to undertaking a TBA program in a different given area, I making efforts to understand the community perspective. ^ 3. Same as above for TBAs. I 4. Undertaking the requisite community organization activities j such as the organization of a village health committee, I 5. Carefully selecting TBAs to be trained with both the TBAs and I the village agreeing. I 6. Careful development of a training curriculum which will have j to incorporate of necessity principles of non-formal educa- ; tion. I 7. Careful selection and training of trainers. (Neumann, Note 11) i ! Selection of trainers is an important preimplementation I factor. Usually trainers are nurse/midwives or auxiliary nurse/ mid wives. They know the TBA better and often understand her perspective better than other health-care personnel who are farther removed from local settings. Often these nurse/midwives or auxiliary midwives speak the TBA’s language, thus facilitating the training process. The literature on TBAs indicates the existence of a pervasive attitude j 79 that who is chosen as a teacher-trainer is unimportant. A real shortcoming of many TBA training programs is their failure to select trainers carefully; trainers need to be capable teachers who know the . environment and know how to teach TBAs effectively. During the early ; part of a TBA program, public health nurses (plus physicians, in some I I cases) with experience in community work may be the most appropriate 1 trainers. An expanded program that includes work experience will | utilize midwives and/or TBAs (who are readily available in most rural | areas) as trainers and supervisors (Araujo, Note 12; Danfa Project, j t 1977a). Criteria for selection of trainers would include (1) training in midwifery or (for TBAs) work experience as a midwife in a rural area, (2) ability to speak the local language, and (3) some experience I in (or aptitude for) community organization. The use of local languages in TBA training is contingent on the willingness of district or regional personnel to find qualified trainers who can speak these languages. In more remote regions, I training is often given in the local language; in the nation’s capital ' or nearby, rural people are assumed and expected to understand the ! national language and communicate in it as well as they do in their | I own local language. This assumption is often false and presents an j acute problem if the TBAs are illiterate. The question of providing ' training in the indigenous languages can be considered a management variable. Nonformal education will not be of much use if the teacher/ trainer is weak in the local language. An acknowledged training pro blem is that health-care personnel at all levels often do not know how to teach people who cannot read or write (Population Reports. 80 1980, p. J-475). The usual teaching materials and methods such as books and lectures obviously are inappropriate; even films and i ! flipcharts may be difficult for a TBA to interpret. In summary, 1 1 throughout the preimplementation stage, emphasis must be on respecting I local customs and traditions and on obtaining the confidence of TBAs 1 and communities. These attitudes should be among the criteria for I selecting TBA trainers. Stage II: Training Location of training is an important issue. Taking TBAs to hospitals may be counterproductive (Ofosu-Amaah & Neumann, 1979). Araujo (1984) mentioned that in Breizil, midwives were first trained in a support hospital but this location conflicted with the TBAs* culture and did not allow than enough time with their patients. The training is currently conducted in the community health unit nearest the home of the midwife so she can continue to work in her usual environment and relate to people in her usual way. Training should be provided as close as possible to the TBA’s local village, and not in a central hospital. For in a hospital the TBAs will see sophisticated techniques and procedures which may tempt them to try something they are not able to handle. One must always remember that the TBA, ideally, will handle the normal, low-risk, natural deliveries. Thus, it is not advisable to teach the TBA complicated procedures and techniques. (Araujo, Note 12) Training of TBAs should start with a basic curriculum that includes instruction on performing normal deliveries and referring high-risk mothers to health centers. In Tanzania the working/ r 811 I I planning group decided that the TBA curriculum should focus on the ! following topics: How to perform a clean and safe delivery. I ! I How to identify and refer high-risk maternity cases. I I How to educate, motivate, and refer women to family planning I services. i I The importance of child spacing to the health of mothers, I infants, and children. | How to educate families in essential primary health care. | (Pederson, 1985, p. 45) j Training should be built around local knowledge of existing j technologies, nonharmful practices, and beliefs of the local TBAs. ! I If possible, the trainers and the supervisors should be the same j persons. Instruction must be practical and geared toward poorly edu- ! cated and mostly illiterate women. Trainees should be encouraged to i 1 perform all tasks frequently; the curriculum should include repetition : I and demonstration under conditions similar to those expected during j I childbirth. The method advocated is "sequential diversification"— a j ; I i strategy for starting out with several basic training objectives and I subsequently moving into other objectives. "In general, experience I shows that new skills should not be added to TBA training without ; sufficient funding and well organized systems of supervision and I resupply" (Favin et al., 1984, p. 40). Selected TBAs can be trained for activities other than normal deliveries, such as prenatal care, immunization, oral rehydration therapy, and family planning. Such training would be provided only to TBAs who, in the judgment of the ^ trainers, could use it to benefit the community, and the nonstandard I 82 : activities must be added gradually to their training (Araujo, Note 9). With the recent interest in issues such as lowering fertility rates and oral rehydration therapy, coupled with the knowledge that TBAs are more or less permanent members of rural communities, the tempta tion to overburden the TBAs with too much training must be resisted. j Various TBA programs eure experimenting with using TBAs in new * I roles such as motivators for family planning; distributors of contra- i i ceptives and basic preventive and curative medicines such as iron/ j folate pills, oral rehydration salts, and drugs for intestinal para- j I sites; and assistants in the delivery of PHC to rural populations. | I The experience of TBAs in family-planning programs is mixed. I I Favin et al. (1984, p. 40) reviewed TBA programs that include family j j j planning and listed two responsibilities for TBAs: either to motivate clients to go to a health facility or to actually distribute contra ceptives. The literature shows that in general, successful TBA programs that include family planning are carefully planned and start out with a pilot project and careful surveys before beginning a slow ' expansion. i In one example of a successful program Malaysia trained 200 TBAs between 1972 and 1978, recruited an average of three new acceptors per month and resupplied an average of 60 to 70 women per month with oral contraceptives. This amounted to 3 percent of all acceptors in the government’s program. Experience in training TBAs for work in family planning may be summarized as follows: • Middle East and Africa: There are relatively few programs in this area that train for work in family planning. • Latin America: Many countries have trained TBAs for work in family planning, particularly for referring family planning ' acceptors to clinics. Several countries - including Mexico, El Salvador, Guatemala, Haiti, and Nicaragua - also permit TBAs to distribute contraceptives. ! 83 I I • South Asia: India, Pakistan, and Bangladesh initiated | national programs to use TBAs in family planning in the mid- j 1960*s. Their programs had limited success, however, in part , due to the relatively low social status of TBAs in this part of the world. • Southeast Asia: Indonesia, Malaysia, the Philippines, and Thailand have large and relatively successful projects to utilize TBAs in family planning. (Favin et al., 1984, p. 40) ■ A number of low-cost techniques for reducing mortality/mor bidity in infants are being incorporated into TBA training programs. In particular, four relatively simple and inexpensive methods could enable parents to halve the child death rate and save the lives of | 10,000 children per day according to the last two State of the World’s j Children reports issued by the United Nations Children’s Fund (Grant, | 1985). In brief, these methods are: | • Growth monitoring - which would help mothers to prevent most child malnutrition before it begins. With the help of a 10 cent growth chart and basic advice on weaning, most mothers could maintain their child’s healthy growth - even with their ' limited resources. • Oral rehydration [therapy] - which could save most of the more than four million children who now die each year from diarrhoeal dehydration. ! • Breast feeding - which can ensure that infants have the best \ possible food and a considerable immunity from common infec- j tions the first six months of life. I • A full $5 course of immunization - which can protect a child | against measles, diphtheria, whooping cough, tetanus, tuber- j culosis, and polio. At present these diseases kill an j estimated five million young children a year, leave five million more disabled, emd are a major cause of child malnu trition. (Grant, 1985, p. 3) I Collectively, these four low-cost techniques have become known j as GOBI (G for growth monitoring, 0 for oral rehydration therapy, B | for breast-feeding and improved weaning, and I for immunization). ! I I Grant (1985) listed some programs in which TBAs have been utilized j for GOBI. In Pakistan, where over half a million children have been ! I 84 , dying each year just from diarrheal diseases and diseases preventable ' by immunization, an "Accelerated Health Program" has raised the immu nization rate from 5 percent to almost 50 percent and (in 1984) produced 30 million sachets of oral rehydration salts. This program j has also trained over 12,000 TBAs in low-cost techniques for protecting children’s lives and ensuring their growth. Via the j communications media, coverage of oral rehydration therapy— which is | being called the medical miracle of this century— extends to people in need of it. Oral rehydration therapy is successfully combating I ' diarrheal dehydration, a curse that has stalked humankind from prehistoric times to the present. I I In India, the Integrated Child Development Services program, I which uses similar low-cost techniques, is functioning in one fifth I of India’s development "blocks" and reaching almost 10 million chil- I I dren. Brazil has held national vaccination days on which 2 million I I children under 2 were immunized against measles; 1.5 million were j immunized against diphtheria, whooping cough, and tetanus; and almost I ; all the nation’s children were protected against polio— all through , mobilization of over 500,000 volunteers working at more than 90,000 j vaccination posts. Evidence has accumulated that even within the i serious constraints of world recession, GOBI techniques are proving j both feasible and cost effective, and in all likelihood the training of TBAs to deliver these low-cost techniques will increase. Although the length, location, and nature of TBA training I varies greatly, most TBA training programs last about one week to one I month; are held in some health-care facility, often a health center; ! 85 I and are taught by nurses and doctors (Favin et al., 1984, p. 37), ! The Population Reports (1980) series on traditional midwives and I I family planning also states that TBA courses are usually short, from * I I three to six weeks, and are taught in the clients’ homes or in local : clinics. Training is successful when classes are small— two to five ' students at best, but in any case not more than 10 (Ampofo, Nicholas, | I Amonooh-Acquah, Ofosu-Amaah, & Neumann, 1977; El Tom, Muarek, Wesley, i I I Mathew, & Lauro, 1984; Ofosu-Amaah & Neumann, 1979). Remuneration is a key issue in training TBAs. Most TBA | training programs provide per diem compensation in cash or in kind [ I and other rewards for completing the training. According to the I literature, TBAs continue to be paid by the local ccamnunities, and \ I ! I government officials are reluctant to put trained TBAs on already ! I I ; strained health-care payrolls. The potential new roles for TBAs in I I I I I I family-planning programs and GOBI programs will require attention to ’ i this issue of incentive and remuneration. If other sources of funds are not found, locally as well as from donors, the remuneration will [ be put into the "bluebook" (as weus the CHIPPS project in Indonesia). I The real issue of remuneration lies in the formal health-care system I itself. In general, remuneration of TBAs as part of this system has I not been tried; Malaysia is the only country in which the national j program compensates its TBAs for their family-planning performance, as judged by regular supervision and evaluation (Population Reports. 1980, p. J-476), The importance of providing incentives to alter behavior is recognized in TBA programs, but not in the larger context. Incentives for altering behavior have to be implemented at every I 86 I level of TBA program activity, including (1) "access bureaucracies" i , of formal health-care personnel in MOHs, who manage, supervise, and I i actively support TBA programs and PHC efforts; and (2) donor agencies ! (where changes are needed to reward the implementation efforts of I professionals). The larger context within which TBA programs operate takes on an added dimension of importance in this issue of remunera- I tion, and its omission in the literature on TBA training is a serious problem. Stage III: Postadoption— Supervision The reality of life in many rural areas of the world is that TBAs are often faced with medical anergeneies. Because of isolation and lack of transport, it is almost impossible to arrange for the transfer of critically ill patients to the rural health center or some part of the formal health-care system. Supervision of TBAs is, therefore, an important consideration. The problems of supervising TBAs are formidable. The typical TBA is middle aged or elderly. She (although male TBAs exist, in Ghana for example, the great majority of TBAs in the world are female) is likely to live and work in areas to which access to a health care facility is difficult, either because the facility is far Eway or because the terrain makes traveling difficult; often it is for both reasons. Traditional midwifery is almost always a spare-time occupation, and in some cases the number of deliveries undertaken may be quite small. TBAs may not speak the same offi cial national language, and they are illiterate. There are also problems about the availability of supervision. Staff from the health system are often reluctant to undertake field supervision owing to the pressure of other work, family demands, transport problems, shortage of personnel, and in some cases professional antipathy to TBAs or lack of conviction of the value of their supervision. Almost all reports on rural health programs, where the training of TBAs has been included, mention the difficulty of instigating and maintaining supervision. If programs fail, it is most often because of the lack of supervision and support. This is true not only for TBA but for other community health workers too. (Walt, 1984, pp. 2-3) 871 ' The importance of supervision and support of TBAs has been < I thoroughly documented in the literature (Davidson, 1984; Favin I et al., 1984; Neumann, Note 11). The supervision of trained TBAs must take into account the objectives and functions, the frequency, | I and the memis of supervision; the personnel who will serve as super- j visors; and the training of the supervisors. The objectives of | supervision are linked to the question of officiousness vs. con structive support, which in turn is related to the need to gain the I confidence of the TBA and the community. Araujo (Note 13) and El Tom ' et al. (1984) are among the authors who have listed as a prime con sideration the need to get the support of TBAs and communities by respecting and taking into account local systems. Police-action supervision that denigrates village workers is not an option for successful TBA programs. With regard to the program in Brazil, i Araujo (Note 13) stated that doctors and nurses have to be trained ; for at least a year in health-care extension programs before they are | , ready to accept and support health-service delivery and are willing | I to facilitate supervision of TBAs. Dr. Galby Araujo (who was the I program director, and provided the impetus, for TBA training in I Ceara*, Brazil), died recently; only the passage of time can reveal I whether his charisma and personality were crucial factors in the institutionalization of support services for TBAs. Countries can opt not to supervise TBAs, but few choose this option. Many TBA programs start out with a system of supervision, but fail over time. Supervision can facilitate in-service training, resupplying of TBA kits, and increased awareness of local systems. 88 ! and it can provide a mechanism for logistic support* Supervision can | also function to enhance PHC in general* Supervision may have several functions, few of which are ever explicitly articulated. One list of these functions, devised for community health workers. . .[,] can be readily adapted for TBAs: | Recognition - help to establish the credibility of the trained TBA in the eyes of the villagers. 1 Maintenance of safe and acceptable standards of care. Motivation - encourage the TBA to practice what has been taught. Education and counselling - reinforce learning and better prac tices as well as provide continuing education for the TBA. Technical assistance - help the TBA to identify and refer women at risk; examine referred cases with the TBA. Linkage - assist the TBA and the community to make use of referral : health services, and to foster cooperation between the formal j health ceure providers (midwives or other health workers) and TBAs. , Monitoring and control - check records and performance of TBAs. Evaluation - gather and analyze appropriate data to assess the effectiveness of TBAs. (Walt, 1984, p. 7) Supervision can be done by the community, by the TBAs them i selves, by other health-care workers, or through state licensing. In I I ' Brazil, some TBAs operate maternity centers themselves. These TBAs j are generally literate, so they can help with record keeping, and ! they serve as an effective link with the illiterate majority of TBAs I (Araujo, 1984). Walt (1984) mentioned a similar scheme in Senegal, I ; where TBAs work cooperatively under the direction of one trained : TBA. In both these programs the supervising TBAs themselves get i regular supervisory visits from formal health-care personnel. By far : the most c<mmon method of supervision is via salaried health-care I personnel who also (ideally) are trainers. Such supervision may be called either "in-supervision" or "out-supervision" (Walt, 1984), though these terms are not mutually exclusive. Under "in-supervision" the trained TBAs go to some point in the formal health-care system (such as a hospital, the nearest health center, or a maternity center) 891 I to receive supplies/stipends or to hand in lists of family-planning I i acceptors or records of deliveries. Under "out-supervision," health- ! care personnel visit TBAs in the field. "Out-supervision” activities may include checking TBA records, accompanying TBAs on home visits, | and convening monthly meetings with TBAs (Mangay-Maglacas & Pizurki, 1981). In-service training and continuing education may be included in supervision, and in some countries trained TBAs may be involved in I team health-ceure activities that enhance their ability to recognize ! high-risk patients and to take various preventive measures. j These methods of supervision are all mentioned in the 1itéra- ; i ture, but seldom do they work over a long period of time. Reasons | t I I I I for their failure include poor planning, clashes between supervisors \ ! and TBAs, logistic weaknesses, and administrative difficulties. These ■ I reasons relate to the planning and implementation stages of the pro- ■ grams (Walt, 1984). i : ; Clashes between young auxiliary nurse/midwives (who may have | I j I little experience in deliveries) and TBAs (who are older and more j i i ! experienced) are mentioned frequently in the literature (Favin et j : I J al., 1984; Population Reports. 1980. Among logistical weaknesses, j I the central problem is incentives. | Exacerbating administrative difficulties of the health care j infrastructure are the attitudes of staff who may be reluctant to leave the health center for a variety of reasons, valid or not. Poor initial training, an overload of patients at the health center, weak administrative links that affect regular delivery of | salaries, drugs, and so on, may result in poorly motivated staff, general frustration, and lack of interest in supervision. (Walt, 1984, p. 12) i 90 The literature mentions that in programs where supervision | was at least relatively successful, the supervision was usually part , of a pilot-research or intervention project with considerable inputs of donor or outside funds and/or personnel. The Danfa project in ; Ghana and the project in Ceara*, Fortaleza, Brazil, are outstanding | examples of successful supervision. An example of successful super- ! I vision without substantial outside input is a program initiated in , I ; I 1982 in Maincaland Province, Zimbabwe (Walt, 1984, p. 20). There, I I I the keys to supervision are the maternity assistants who train TBAs j j ! ' and have detailed knowledge of local conditions. The vanambuya (TBAs j of Zimbabwe) report to the maternity assistants about problems and ! deliveries every month or two. In-service training is conducted at | the clinics, and at present this is the only supervision after initial ' training* There has been no formal evaluation, but 1,500 TBAs had , been trained by mid-1983, and the strong interest evidenced by the | I I I . maternity assistants is noteworthy. The response of TBAs to this ' system and the in-service training courses has been so enthusiastic that too many TBAs have subscribed to the courses. I In summary, constraints on supervision can be placed into five major categories: - weak technical, organizational, and administrative support. - dispersed and isolated communities. - lack of rapport between personnel of the formal health system and TBAs. - failure to involve communities in TBA training programs, limited financial support for training and supervision of TBAs. (Walt, 1984, p. 22) There are many ways of improving maternal and child health services. In deciding between a number of short-term and long-term r ’ " ' ' 911 policies (such as producing well-qualified midwifery staff, intro ducing wide-ranging hea11h-education programs for women, conducting ! immunization campaigns, and training TBAs), policymakers will assess , ' the effect of a specific meeusure in light of some general criteria: j : legitimacy, feasibility, and support. Measures with high levels of ! I I ; legitimacy, feasibility, and support will be most likely to succeed (Walt, 1984, p. 26). Stage III: Postadoption— Resupply. Referral. Risk Assessment I j Resupply, referral, and risk assessment can all be part of a j j supervisory system. Generally, TBAs have a kit; additionally, in | I Zimbabwe all pregnant women attending any prenatal clinic receive ' I ' I before the end of their course a "cord pack" containing cotton wool, | I I a razor blade, string or tape, and surgical spirits. A TBA kit has been prepared by WHO and the British Life Assurance Trust. It is a collection of booklets, cards, and illustrations contained in a com pact carrying case. One of its planners, Ms. Ilfra Lovedee, has some I 25 years* experience in working with TBAs in India, Burma, Nepal, and I I I Bhutan. This kit is like a supermarket of material describing various ! ! I : methods, ideas, and visual aids that could be used in training TBAs I worldwide (Rosser, Note 14). Usually recommended are kits with a ■ I minimum of sophistication and using local technology as far as pos- j sible. In some TBA programs only local items are included, as in the \ "match box kit" from Kenya, which contains only a piece of soap, a j razor blade, and umbilical cord ties. The TBAs put the kit together j themselves during training (PATH, 1984b, p. 13). The contents of kits must be replenished— especially consumable items such as cord J 92 ties, gauze, cotton wool and swabs, razor blades, and medications. Resupply can be the Job of a supervisor. Like supervision procedures, referral procedures are hampered by transportation problems and communication problems. Solving these problems often entails mobilizing the resources of communities and health-care systems. For example, a local private businessman could agree to lend his vehicle for emergency transport of a mother, the newborn child, or both. Most risk factors sure related to a woman * s reproductive his- ! tory or her personal characteristics. Table 4 lists examples of such I risk factors, some of which can be detected through history-taking at j any time before or during pregnancy. ! Other risk factors, including abnormal presentation of the ! fetus (breech or trsuisverse lie), can be detected through physical examination. All TBAs should be trained to conduct external examina tions of the abdominal area by touch (palpation), and they should be taught that it is the size and position of the fetus that is impor tant, not the size of the uterus (PATH, 1984a, p. 3). | ; Stage III: Postadoption— Evaluation i ; In his review of literature on evaluation of training, Hoyle I I ' (1984) makes a distinction between educational evaluation and training I i evaluation. Educational evaluation is almost entirely limited to | evaluation of lengthy educational programs in schools and is amenable to use of the scientific method; performance is subject to objective measurements. Training evaluation, on the other hand, normally is concerned with the evaluation of brief programs for adults, not in Table 4 Sample Risk Factors® Category Risk Factor Age Usually, over 35 or over 30 and primiparous Under 19 Obstetrical history Usually five or more children First child Less than 24 months since last delivery Previous obstetrical complications, especially Caesarean section, perinatal deaths, low-birth- weight baby, etc. Medical history Diabetes, heart disease, renal disease, psychoses, etc. Maternal characteristics Possibility of twins Deformities of the legs and pelvis Short stature Smoking Alcohol consumption Performs heavy manual labor 931 ^The cut-off points or norms for each factor must reflect local conditions and available resources. (Adapted from PATH, 1984a, p. 3) I 94; t ! i I schools, and on subjects that are not amenable to objective measure- I ; ments. , A further look, however, reveals that much of the material is j lightweight and ephemeral and that the range of important con- . tributions to both the philosophy and methodology of evaluation ! is very limited. (Hoyle, 1984, p. 275) i : ! I In developing countries, training evaluation is generally undertaken reluctantly and with the simplest methods. Programs dealing with the : training of TBAs fall into this category, and the absence of clear guidelines as to what constitutes a "good" evaluation only makes the , I analysis of TBA programs more difficult. The literature on evaluation| shows that in programs not incorporating community preparation and | involvement in TBA training, villagers often remained unaware of the | new roles for TBAs and thus were unwilling to increase the TBAs* remu- | neration. The lack of additional remuneration, in turn, reduced the j I I TBAs* motivation to practice their new skills (Neumann et al., 1983; | I Verdose and Turnbill, 1975; WHO, 1979a). j I There is extensive literature on training of TBAs, but little j ! literature on supervision (Davidson, 1984; Walt, 1984), on linking TBAI I training programs to existing health-ceure systems, or on evaluation j j (Davidson, 1984; Favin et al., 1984; Neumann, Note 11; Walt, 1984). It is hoped that TBA training will improve TBA program outputs, for j 1 i ! example by reducing maternal and infant mortality/ morbidity and by I : increasing (1) ability to combat deaths from dehydration (via oral I rehydration therapy), (2) family planning, and (3) environmental I cleanliness. There has been no systematic weighting of training i I components, however, and not much critical evaluation. ! 95 Equally clear is the fact that if decisions to train and use TBAs are based on the hypothesis that trained TBAs will bring about an improvement in the health status of mothers and children, then serious efforts should be made to test that hypothesis in ways that are more meaningful than most of those used so far. Perhaps the greatest benefit derived from the evaluations attempted thus far has been that they point to the need for more adequate instru- ' ments for measuring impact. (WHO, 1979a, p. 42) j One of the weaknesses of training-evaluation designs in ' general is the failure to include longitudinal studies. The usual ; I rationale for not doing such studies is the practicality of following ' up participants (in this case TBAs) to gather attitudinal as well as | behavioral information. The main difficulty in doing longitudinal ' research is gaining access to TBAs who participate in training. Another difficulty in evaluation and follow-up is deciding which TBAs to evaluate. What should one do, for example, when a survey shows that most of î the TBAs attend only 5 to 10 births a year? Findings in Danfa, | Ghana, typical of many areas, showed that TBAs there attended an average of 20 or more deliveries per year. If funding were suf- | ficient and other findings favored training TBAs, it would seem ; that all TBAs should be invited and given the opportunity to be | trained, even if they are old, even if they only attend a few | births a year. If funds are sufficient only for training a por- | tion of TBAs in any area, then it would seem most appropriate to ! train and evaluate those first who attend most deliveries. ’ (Favin et al., 1984, p. 38) | In a paper on the role of TBAs in a PHC strategy in northwest Somalia, Bums (Note 8) recommended training one or two TBAs per settled vil lage. In other types of nomadic settings, training one TBA would not ! I cover the entire population; often, to cover the majority of deliver- i I ies, 10-15 TBAs need to be trained. In such cases, evaluation should ! include consideration of the TBAs who are most active and whether they I are in fixed or nomadic villages. ' " 1 96 I ' Pillsbury (1982) mentioned the paucity of results from eval- I uation of the effectiveness of traditional practitioners (TBAs included), whether in traditional or modern-sector programs. There i ; j has been little systematic evaluation— qualitative or quantitative— of I ; traditional practitioners* performance. j When compared with programs in other sectors (e.g. agriculture | or rural development) health programs are often poor performers. | For example, when a worldwide series of impact evaluations was ! launched in 1980 by the U.S. Agency for International Develop ment, rural health care was one of a half a dozen priority , sectors identified. In each of the other sectors (agricultural j research, potable water, rural electrification, and so on) it proved relatively easy to identify at least six developing coun tries in which a program in the respective sector was judged to j have progressed far enough to merit being a fruitful subject for ; impact evaluation. The health sector, however, lagged far behind ! the rest, for all health project managers approached claimed it ' was still far too early for the project they managed to have begun ; to have an impact: by the end of the year only one project had been identified as a candidate for evaluation. This was a pro ject in Senegal that turned out to be flawed in many ways. (See I Weber, R., et al. Senegal: The Sine Saloum Rural Health Care j Project. U.S. Agency for International Development Project Impact ' Evaluation Report No. 9, Washington D.C. 1980). Also, when , compared with family planning programs, the record of primary health care in having a measurable impact is also poor - in part because much more careful attention has gone into designing and implementing family planning programs to produce short term impacts. (Pillsbury, 1982, p. 1833) Not much is known about the cost effectiveness of traditional practi tioner programs, either. Given that cost-effectiveness analysis is in any case relied upon by many health planners and evaluators, it is meant to signal the fact that we have little evidence at all as to the ultimate cost- effectiveness of any category of health worker improving the health of a rural developing country population. In fact, in the opinion of some veteran analysts of the question . . . , if bene fit cost analysis could be done it would probably show a great financial loss due to the internal brain drain effect of newly- trained primary health workers thinking themselves overqualified for the village and leaving rural health work for larger towns and villages. (Pillsbury, 1982, p. 1834) 97 For TBAs, the problem of internal "brain drain" can for the most part i be disregarded because TBAs tend to stay in their respective villages j I ; or communities. Davidson (1984) cited evidence from the literature that interventions during pregnancy and birth are crucial in determining the future survival of mother and infant. Instructing TBAs in Western obstetrical skills has been described as an economically and medically I I sound method for counteracting high rates of maternal and infant | mortality (Davidson, 1984). Favin et al. (1984, pp. 37-38) gave exam-| pies of TBA programs in which TBA training helped to improve maternal health care, and this issue has been discussed by Araujo (Note 12), I Grant (1985), and Ofosu-Amaah & Neumann (1979), among others. Gen- | I erally, the literature supports the notion that TBA training is | effective in combating maternal and infant morbidity/mortality. j i ' I I j Program designs that permit comparison before and after 1 implementation clearly will permit more meaningful evaluation of the I I ; effectiveness of TBA training programs. Such designs require the \ I collection of baseline data, a detailed assessment of needs, and the I establishment of goals and objectives with a realistic eye to funding, ■ time, and personnel. Because of inadequate funding, these design elements are seriously lacking in most TBA training programs, but I ways must be found to gather data before the TBA program begins I operating. Additional costs are involved, but if the TBA program is a pilot or demonstration project, planners must show the desirability and feasibility of beginning with such a project vs. implementing a limited version throughout the country. A pilot project should I 98 I ; survey the TBAs* knowledge, attitudes, and practices (KAPs) before training. Evaluation of a training program for illiterate TBAs in Sudan I (El Tom et al., 1984) included a before-training, after-training, and I follow-up survey, which is essential but rare among evaluation pro- I grams. The evaluation of the Sudan program concluded that illiterate ■ community-based TBAs could be trained successfully for wider duties 1 through a participatory approach involving the creative reinforcement 1 of a relatively small set of skills. Most dramatic was the change in the percentage of midwives who knew how to prepare oral rehydration j solutions: 3 percent before training, 100 percent just after training, I j and still 98 percent a year and a half later. The content of the j training consisted of oral rehydration therapy for infants, birth I I spacing, nutrition, immunization, and techniques of home visiting and I I health education. Although the TBAs had been trained before, this I ' ; additional training used the sound techniques of participation, j repetition, and creative reinforcement. Further, the program was I reinforced by the Grand Mufti of Sudan, who spoke to the midwives on I j the subject of Islam and family planning, allaying any fears the TBAs I might have had. The program’s opening and closing sessions began I with recitations from the Koran. Evaluation of a TBA should be conducted throughout her train- : ing. In a group, the TBA*s progress can be evaluated by discussion I I of main points and by review of case histories. Individually, the ! TBA*s progress can be evaluated by looking at (1) her record books ^ (if available), supervisor’s assessments, referral records, and i ' ‘ 9 9 1 I I self-reports, and (2) the reports of mothers who have been assisted I ; ' by the TBA. Actual observation of a TBA delivering a baby is dif- , ficult to time, but reviewing lessons and demonstrations can be a ■ substitute. Both evaluation of TBA training and evaluation of the ' overall effect of the program are greatly enhanced by on-site visits, | I I which should be linked to supervision. | Table 5 shows an example of a detailed evaluation of a TBA | training program. The point is that a follow-up survey for comparison requires a similar survey, such as a KAP survey. (See the Appendix j for an example of a KAP survey used in TBA training in Ghana.) j I In summing up evaluation literature on TBA training, a program is far more likely to succeed if it includes elements of national and ; ! : j local support; community preparation, input, and participation; care- ^ ful program planning; evaluation (planning evaluation not divorced i from implementation); and a posttraining support system. A meaning- I ful evaluation requires greater inputs from donors to help facilitate ; i I ; an orientation to PHC. | The question of the need for a KAP survey for supervisors of ■ TBAs has not been mentioned in the literature, but as Moris (1984) I pointed out, more knowledge of health-care personnel at the local ' level would help greatly in designing more effective health-care i I delivery systems. The lack of such knowledge is a serious omission that can lead to ineffectiveness of TBA programs and their evaluation. The five basics of training— (1) training must support the objectives of the organization, (2) training must help improve productivity (in this case, TBA productivity), (3) the effects of training must be Table 5 Evaluation of the Program iôôl Evaluation criterion Before training (%) After training (%) Follow-up survey (%) Percentage of midwives believing that: Breast-milk is the best food for babies 87 97 99 Breeust-fceding should extend for at least years ■ .... ----- 85 Supplementeury feeding should begin by the end of the fourth month postpartum 70 98 100 When a child has diarrhea, he should be given less fluid 71 5 0 When a child heis diarrhea, he should be given no fluid 1 0 0 When a child has diarrhea, he should be given less food 86 23 1 When a child has diarrhea, he should be given no food 13 0 0 When an infant has diarrhea breast feeding should be decreased 77 5 0 When an infant heis diarrhea breast feeding should be stopped 22 0 0 Percentage of midwives who knew: When to administer oral rehydration salts 9 99 99 How to prepare oral rehydration salts 3 100 99 Three or more methods of contraception 41 87 99 Eight or more contraindications to contraceptive pill use ■ ... 89 A benefit of vaccination 61 100 98 That pregnant women should be instructed to eat more and have a varied diet 98 Note: Adapted from El Tom et al. (1984, p. 219). 101 measurable, (4) training must be delivered competently, emd (5) train ing must be evaluated (Petty, 1985, pp. 85-86)— cannot be evaluated systematically without baseline data. Thus, more financial emd human resources must be devoted to preimplementation considerations. | I Cost j The question of health-care financing is enormously impor tant, especially in relation to how PHC can get financed and become self-sustaining. Implementation of TBA programs in a PHC context amidst both scarcity (of human and financial resources) and apathy is ^ difficult, but programs are being implemented around the world. In I 1 I I I I an address on monitoring the "health-for-al1" strategy. Halfdan I ' ' Mahler, WHO director-general, stated that I I I I : I very few countries, including the most economically developed, I were able to assess the amount and flow of resources of Health I for All. In particular, they were unable to distinguish between the allocation of funds for the continuation of old policies on I the one hand, and for the promotion of new ones on the other. ; (Mahler, 1984, p. 156) I ; Trying to find out how much in resources goes toward PHC and TBA I i training is difficult, as attested by attempts made in connection ; with this dissertation: two personal trips (in 1984) to WHO in Geneva eaid numerous inquiries and conversations (in 1984 and 1985) j with World Bank and U. S. AID personnel involved in PHC. I 1 j Lack of interest in cost analysis characterizes the whole range of health activities, but is particularly pronounced in the case I of primary health care, probably because of the diversity of the j activities involved. Shortcomings include poorly defined con- 1 cepts of cost, the use of services irrelevant to the community’s j needs, the performance of costly studies and surveys that are not essential to the actual task of providing primary health care services, and the continuous absorption of technical assistance costs - particularly in demonstration projects - by activities I 102 ' that are often reproducible from one country to another. (Grosse ^ & Plessas, 1984, p. 226) I The poor managerial reputation of the health-care sector \ among bilateral and multilateral development agencies makes the j j donors reluctant to give to health-care systems managed so badly that I they cannot absorb additional funds. Beyond political commitment and coordination of resources, a support consists of adequate funding for personnel and supplies. Auxiliary health personnel cannot function in a vacuum. Joe D. Wray has summarized it well: "No matter how well-intentioned the program, no matter how adequate or inadequate the training, unless auxiliaries are part of a system that is capable of providing them with satisfactory guidance when needed, the necessary supplies and I equipment to carry out the tasks assigned them, and a referral I system to which they can send problems with which they cannot I deal, they are not likely to be able to function effectively." (Storms, 1979, p. 15) ' According to the literature, most PHC efforts and TBA programs are at I least partly funded by inputs from donors. For example, in the TBA program in Ceara*, Fortaleza, Brazil, funds come largely from the 1 I Kellogg Foundation. Although community clinics are largely self- I I I financing, the program provides some vaccines and equipment. More I important, these external funds pay for supervision and training by I the maternity hospital team. Given the importance of supervision and I I support, cessation of funds would constitute a real threat to contin- ' uation of the program (Walt, 1984). Grosse & Plessas (1984) found I that primary health services cost substantially more in demonstration ' projects than in larger scale programs, as would be expected. Three I large-scale programs were compared to four demonstration projects; i j all of the latter have ceased operations. Grosse & Plessas (1984) ! attribute the failure of these demonstration projects especially to I 103 I high recurrent costs that discourage expansion of health service in I * rural areas (see Table 6). ! The literature on cost effectiveness shows that TBA training is inexpensive, but the costs quoted vary considerably and rarely go beyond the direct costs of initial training or take into account the TBAs* roles in family planning and GOBI. Investment costs for PHC projects are one-time outlays to establish a fully operational program. They cover: - research and development; - facilities and major equipment; 1 - initial inventories of supplies and drugs; and - initial training of manpower. (Grosse & Plessas, 1984, p. 228) Operation costs for PHC programs are the recurrent costs of operating ■ and maintaining the program’s level of service. They cover: ! - staff salaries, allowances, maintenance and other benefits; - replacement of equipment and facilities; - replacement training; - drugs, supplies, biologicals, etc.; and - fuel, utilities, etc. (Grosse & Plessas, 1984, p. 228) National authorities need support in determining the cost i effectiveness of training TBAs vs. other means of providing health- , care services (WHO, Note 3) p. 4). When confronted with the costs of : pretraining and posttraining considerations in addition to the costs ; of training per se, are countries such as India and Bangladesh will- I ing to face the financial implications of their programs?— which are I aiming to have a trained TBA in every village by 1985. In both coun tries, such an endeavor entails substantial increases not only in extension training facilities, but also in costs of supervision and support. In demonstration projects such as Danfa in Ghana and Ceara* 1041 • n O eu c r > ■ 4 - > < o o cr .—I - # - » • — I Q . c : !»- r t 3 cu OJ < o o u O l cu i > - o> o- ê- g. c= a> o o > C Ü C Ü C f > Q l O l C O o > > o C O O l C J > = > o c r o c = f O c u e c A 3 Q l. C O C O - l y » O c u o > o c = W W C D 0 » _ _ ^ c ^ o 3 f O C O C k _ C = C U W a i • < C O l w O C O CJ> C O “ O < o C O c = c u - 4 - » f O 3 = C O C O 3 H O _ c = f O o . o -c : ro -a- - so CD CU CO •O CD c s i C S I UD ^ e o cS o sa CO C ? s wo wo — < CD LO W O C D W O - sC CD iy\ s o LO — I ro — 4 • —t - « s f C O c r ^ C M so ro I -a- CO wo LO Cs CO sO CD CD CD CD CD CD Q -O CD CD C O z LO C S I LO SO og ■ s O CD CD CD CD r-s rs. CD C M sO 2^ Cs. so -O JQ rs. rs. O S rs. .-4 C O C M C O C O m cu <o o o O! L» O CU o j O CL <o 3 Q> " C SB CL CL CD CU O C O CO cr "O CU cr OC O CD L4— C O fO S E T cr c r C O C O fO CO C O CD z 3 * ^ o <o j O CU cu <o c r >■ GB cr C7 o O 3 o o > m & -o O fO cu o O - C = CD CL ■c A A C O CD C D CO Dae o s _cr CO o> _cr cu CD >* o L . CO CD CU CO O > O l -o cr CU <o -CD -4D CO cu lO cr -o cu CU CD cu -cr L4— o cu C O -cr cu CD CD > CO fO CU LCD CO O o> -cr -4D CO CO C O -cr > cu o 3 O LCD - O fO fO CU CD cr CU o cr o fO O l LCD C L cu C O cu -o LCD l O cu CO fO C O 4D O CD > • CD ra fO m gd gs I C O Ou 105 in Brazil, is it worth the donor inputs to continue or justify the project? Grosse & Plessas (1984) argued that the high recurrent costs of demonstration projects preclude the possibility of incorporating them into national health-care systems. This argument does not neces sarily apply, however, if these projects are in regions of need and I if they generate data on local systems from which to build workable I PHC strategies. The main point to be made about costs, it is argued, is the j need to reverse some national and international trends. Today, the ■ power of the example set by donor countries is enormous. Making more I money available would aid efforts to reach the poor in the less j developed countries of the world. j Aid programs in particular could be increased to the internation- I ally agreed target of .7% of the donor nations’ GNPs. Today, the figure for the free market industrialized nations stands at .37% j of their GNPs - about the same as in 1974 and considerably less j than in 1964. (Gremt, 1985, p. 56) Mention has already been made that three quarters of donor aid for health goes to benefit sophisticated hospitals, medical schools, and ! urban (minority) populations. This allocation pattern runs counter j to the "health-f or-al1" strategy and the goal of making low-cost ! basic health care available to all. A higher percentage of donor aid must be earmarked only for PHC efforts that are directed toward rural (majority) populations. 106 ’ CHAPTER IV IMPLEMENTATION CONSIDERATIONS j Traditional birth attendant (TBA) programs are faced with j social, institutional, and cultural constraints that work against i effective implementation and continuation. While these TBA programs I are often accepted as part of a primary health care (PHC) strategy I to reach the rural (majority) populations, their mere adoption does I not ensure that they will be implemented and continued effectively. I : The intentions of donor agencies and the receptiveness of developing I countries to TBA programs does not necessarily translate into suc- j cessful, continuing programs. (For a thorough treatment of the ^ problems of practicality in implementing programs with traditional ‘ health-care inputs, even in the context of a favorable national I policy toward the traditional health-care sector, see Pillsbury [1982].) Preimp1ementat ion Pillsbury (1982) recommended that social scientists involve , traditional practitioners in projects or programs as early as pos- I I sible. Participatory planning and evaluation (i.e., the actual I involvement of the community— the "users" or recipients— in the I planning and evaluation process) have been advocated by a number of j theorists (Honadle, 1982; Kindervatter, 1982; Rondinelli, 1983b). I Field experience suggests that the success of institutional [ — - 1071 I , realignment and strengthening depends on working within the system as i much as possible (Korten, 1980). Reification of method models for | planning and project analysis has augmented the tendency to over blue- ■ print and discourage flexibility (Korten, 1980; Rondinelli, 1983b). ; Moreover, there is little correlation between sophistication of plan- i i j I ning and national performance. A consistent plan does not ensure I implementation any more than an inconsistent one does (Paul, 1983). ! j When plans they had prepared were not implemented, planners trained ' as economists assumed that the failure to achieve the targets was I mainly attributable to errors in computing and allocating resources ; or to errors in basic data (Paul, 1983, p. 7). I I ; Davies (1977) presented a model depicting three approaches to j I I I educational planning: "systematic planning," "expedient planning," I ; and "piecemeal planning." Davies’ (1977) analysis indicated that 1 "expedient planning" and "piecemeal planning" avoided prestated | I objectives and large-scale prepackaged projects, and that these two ' approaches produced more effective results by adopting and allowing I I : for phased implementation, in which a number of pilot projects are | i attempted. Rondinelli (1983b) believes that planning and implementa- I I I i I tion must be regarded as mutually reinforcing activities that refine and improve each other over time, rather than as separate functions. In the "top-down" approach of some projects, in which problems are ; defined by drafting solutions, analysts and planners become creators I as well as implementors of policy. | Under conditions of complex interaction and uncertainty, devel- | opment projects can, at best, be designed to isolate small, | manageable aspects of problems and to intervene in what seem to i ' 108 I I I I be strategically important ways to cope with them incrementally. I ! Since the ability of development planners, to predict and control | the outcome of their projects is quite limited, their methods of analysis, planning and management must be better suited to recog nizing and dealing with uncertainty, detecting and correcting I errors, generating and using knowledge as the experiments pro- j gress, and modifying actions as new opportunities and constraints I appear. (Rondinelli, 1983b, p. 321) | ’ Rondinelli (1983b) argued that flexibility is the key to successful I : I j development programs; rather than providing a blueprint for actions, | I I I I projects must facilitate continuous learning and interactions, thereby ! allowing participants the options of adjustment and modification as I I more is learned about the conditions under which development programs ; must operate. Implementation is ultimately the central issue in plan- , ; ning. Feasible, relevant planning, especially for implementation, I I requires the fullest possible awareness of the political environment encompassing it. In fact, implementation depends largely on the I political environment in which it is attempted. The literature on j I implementation strategies is generally descriptive rather than analyt- ; ical. The paucity of documented analytical studies on implementation ! ! I strategies is most pronounced with regard to development projects in j I the Third World. There is, however, adequate literature on innovation | j considerations and educational innovations, and a nearly universal | ! I I recognition that serious deficiencies exist in knowledge of implemen tation issues. It now seems that inadequate attention has been paid to the imple mentation of projects and programs. Decisionmaking is tardy and organization arrangements for managing the projects are vague. | Even when projects were physically completed, too little atten- j tion was paid to getting them to perform well and to maintain the | facilities. Many observers have called this set of issues "man agement problems" and have su~gued that the widening gap between project and program plans and their performance is largely due to log”! management failures. Unfortunately, planners too often view implementation and management as issues outside their purview. The dichotomy between planning and implementation prevailing today in many countries reflects inadequate appreciation of the interaction between the two. Planners are reluctant to recognize feedback on implementation problems as an input to the planning process. Further refinement of planning tools and models cannot possibly make up for the deficiencies in implementation and management. (Paul, 1983, pp. 8-9) I Personal experience with projects (related to the United I States Agency for International Development [U. S. AID]) in Africa bears out the truth of the above statements. One of the more encour aging suspects of the current literature on development projects is that it addresses real, practical problems. The problems are not only real, but have proven intractable, even under the best of circum stances. Planning and implementation should be viewed as being part of an inseparable process. Plans must, if they are to be effective, be I continually adjusted to account for actual conditions. Questions I of project implementabi1ity have sensitized the planning process [ to the need to examine a wide variety of factors and variables i which could impede successful implementation. The variables I include political, economic, administrative, and sociocultural ' factors that relate to the projects. (Goodman & Love, 1980, p. 9) ; The reason that implementation has become a catchword and a priority I ; concern in development literature is simply that so many past efforts to implement development programs have failed. Wells (1976, p. 56) discussed three approaches to control of I , implementation: (1) shared control of implementation, (2) community 1 I input to implementation, and (3) no community influence on implementa- I tion (Figure 5, Part A). He pointed out that the degree of community I I control was critical in the successful implementation of education- I disseminating projects in developing areas, where communities are I _ 110 A. APPROACHES 1. Shared Control of Implementation Communities Other Groups Project Management 2. Community Input to Implementation Communities Project Management 3. No Community Influence on Implementation Communities Project Management B. ORGANIZATIONAL STRIKJTURES I. Separate Agency from the Ministry of Education Controlling Education Planning Group Project Management Ministry of Education 2. Shared Domestic Responsibility for Implementation Domestic Ministry Project Management Ministry of Education 3. Foreign Influence on Implementation Foreign Agency Project Management Ministry of Education Figure 5. Control of Implementation: Approaches (Part A) and 1 Organizational Structures (Part B). (Adapted frdm Wells [1976, pp. 16, 59-62].) I 111 : likely to be rural. The "no community input to implementation" I approach corresponds to a centralized, "top-down" approach. In the I i "community input to implementation" approach, the role of the com- : munity is substantial. The "shared control of implementation" approach is decentralized and allows communities to share the control ; of project management with other groups. 1 I Figure 5 (Part B) represents some organizational structures ' with respect to the control of project implementation (Wells, 1976, pp. 59-62): (1) separate agency from the ministry of education con trolling implementation (the controlling agency is autonomous from ! the ministry of education), (2) shared domestic responsibility for implementation (the ministry of education and another domestic min- , istry share control), and (3) foreign influence on implementation (control is shared between a foreign agency and the ministry of education). Implementation requires an unrelenting effort to rely pri marily on resources from the communities to be served. These ; communities have the traditional knowledge and skills to carry on ,meaningful work in their own areas. The services to be provided in I development projects must not be structured in a rigid, preconceived I plan. The organizational structures that control project implementa- ! tion or that work with and rely on local inputs must be allowed to accept flexible approaches. With their inherent humility and unpre tentiousness, such approaches are necessary for a program to become increasingly comprehensive, yet remain technologically quite simple and relatively inexpensive. These program characteristics help to I 112 ; overcome problems such as (1) an inability to pick up recurrent ■ costs; (2) a tendency to bypass or fragment local communities and institutions, thus neglecting the urgent need for building local capacity; and (3) a truncated time horizon (Honadle, 1982). This project or program approach— in which local inputs are really taken : into consideration and worked with— will enhance a development I program’s ability to make results self-sustaining. I In his discussion of planning and implementation issues, Warwick (1980) regarded the definition of planning as "an exercise in rational decision making" as perhaps the greatest obstacle to effec- , tive linkage between program planning and program implementation. His transactional model, which deals with organizations and bureaucracies, i underlines what Warwick believes to be the essentially interactive ' and political nature of development planning and program implementa- ; tion. He stated that weakness of resolve among implementors is a real j problem in poor implementation and that "factors such as consultation i i • and participation in policy formulation will often carry more weight | at the stage of execution than considerations of technical soundness" (Warwick, 1980, p. 386). Warwick’s approach centers on the notion of ; transaction among individuals, groups, and organizations, and it ; emphasizes the dynamics of the planning and organizational process. I I I This approach (1) assumes consensus via negotiations, (2) is realistic | in stressing that political leaders opt for most pressing problems ; I with strategies expected to produce respectable results in a short time, (3) confirms that constituency building may be more important I I j than technical analysis, and (4) addresses the problem of incentive I 113 I j for implementors by giving them active participation in planning and ' evaluation activities. The process of participation is one potential application of nonformal education and is part of a transitional systems model that I emphasizes both facilitating and impeding conditions. Because of I implementation problems, the stress is more on the processes of i planning than on the product of planning. Effective planning is ! essentially a political activity; planners themselves often form a political interest group and advance the political ends of others. ' Warwick believes that planning should include awareness of the structure of interests and the distribution of power in a society, j Warwick’s approach is essentially a transactional systems I model accenting controversy, compromise, and power within organiza- ' tional and political contexts, and it is especially relevant in bargaining for power in ministries of health within a PHC context. , Improving the implementation of innovative programs, such as PHC j programs, is contingent on analysis of needs in terms of organiza tional structure and characteristics. The implementation perspective cannot opt for the extremes of rational irrelevance or political over control; what is needed is an intermediate position realistically set I in the context of sociocultural structure, bureaucracy, and politics. I The following is a list of myths and realities relevant to 'preimplementation considerations. I I 1. Myth I; The ends of planning are established by political 1 leaders or other authorities; planners confine their activ ities to means. 114 Reality: Effective planning is inherently a political activ ity; planners often form an interest group in themselves and advance the political interests of others. 2. Myth II: In choosing political ends decision makers do not consider means; in recommending means planners do not judge ends. Reality: The choice of political ends is contingent on the availability of means; proposals for means hinge on percep tions of political ends and constraints. 3. Myth III: Planners evaluate all policy alternatives to assess their efficiency, feasibility, costs, and benefits. The primary basis for assessing alternatives is objective information, which is shared with all concerned. Reality: Planners do not evaluate all policy options, nor do they give equal attention to the costs, benefits, and conse quences of those being evaluated. Far from being widely shared, sensitive information is often hidden, husbanded, and managed. 4. Myth IV: A technically sound, clearly formulated, and internally coherent plan contains the essential ingredients for implementation. If the head is clear, the hands will , follow. Reality: From the standpoint of implementation, the process of planning may be as important as the product. Technically sound and internally coherent plans may not be implemented at all while less elegant plans with stronger backing may show a high rate of execution. Much depends on human factors, for I organizations do not function as machines. (Warwick, 1980, pp. 381, 384) - The transactional systems model relates to the fact that donors and academicians dislike implementation because it is unattractive and unglamorous; this dislike is part of the donor-agency bias against 1 implementation, which is discussed later in this chapter. A "learn- i Iing-by-doing" approach deals with elements that are not amenable to 'rational planning models— yet another reason why implementation is not attractive to donor agencies. Actual implementation in Third I World countries is not synonymous with a nice document in English or I French that is amenable to a Western rationalistic approach. r 1 1 5 ; The Implementation Perspective Planning change, legislating change, promulgating change, packaging change, and training for change all fall short of actual ,change (Mann, 1976). Project implementors are in a dilemma: they i cannot know what it is they need to know until project operations are well under way. People at the site or delivery level are crucial; I although site characteristics are extremely powerful and complex, I I probably not all of them are idiosyncratic. Adaptation is the price I paid by the site for accepting the means or goals of any project or I program. i I The demands of implementation (Berman, McLaughlin, Bass, Pauly, & Zellman, 1977) dominate the innovation process and its ,outcomes regardless of the education and treatment being attempted, i the level of resources applied, or the type of funding being used. I Berman & McLaughlin (1975, p. 10) listed three patterns of the implementation process: (1) mutual adaptation— the site and project are each shaped by each other, (2) cooptation— the site is unchanged but captures the project, and (3) nonimplementation— the project breaks down or is ignored by the site. All of these implementation processes, Berman et al. (1977) contended, were contingent on three meijor factors: (1) the extent of the motivations and the circumstances involved in initiating the project, (2) the substance of the project I and the scope of the expected change, and (3) the implementation jstrategy of the project. I Mann (1976) attributed three assumptions to persons who work i in schools: (1) people can be changed— they are malleable and _ - _ J 116 ’ perfectible, (2) people are rational— they orient themselves toward goals or use information and calculation to govern actions, and (3) people in education share social goals on some lofty abstract level (or, if they do not, they can be brought to agree on values and goals). This kind of expectation of consensus on goals that guides our efforts at large in schools is not, however, fulfilled in ; practice. The difficulty for agents of change in education lies in ■ applying methods largely educational to situations fundamentally I political. ! ! Educational innovations are highly variable. The Rand Corporation’s studies on federally supported programs for educational change (Berman & McLaughlin, 1974) rejected a rationalistic, "top- down" view of educational innovations because this view accented ■ adoption, planning, and dissemination while ignoring the crucial issue of implementation or institutional adoption of an innovative ! strategy. In addition, Berman & McLaughlin (1974) stated that on the Î basis of empirical and theoretical evidence, adoption is only one (and usually not the most important) hurdle to overcome in bringing about change in educational practices. ' In sum, the nominal adoption of an innovation cannot be assumed to provide accurate forecast of its actual implementation or use. I The process of implementation in the instance of educational innovation is essentially a 2-way process of adaptation, in which I the innovation strategy is modified to suit the institution; and the institution changes to some degree to accommodate the innova tion. (Berman & McLaughlin, 1974, p. 10) This two-way process, called mutual adaptation, was the primary fea- ; ture of effective implementation. Berman & McLaughlin (1975) defined I mutual adaptation as "a process by which the project is adapted to ! ' 117 1 I its institutional context, and organizational patterns are adapted to meet the demands of the project." The Rand Corporation’s study of educational innovations identified the major factors that affect ; outcomes: 1. Project Characteristics; - Resources - Educational methods - Implementation strategies I 2. Institutional Setting: ‘ - Organizational climate - School characteristics - Teacher characteristics I 3. Federal Policies: ' I - Programs priorities - Management strategies. (Berman & Pauly, 1975, pp. 52-53) ' One of the major conclusions of the Ramd Corporation’s , studies is that implementation strategy is most important in deter- ^ I mining the outcome of innovative projects. Neither the technology, ! i nor the project resources, nor the various federal management strat- ' egies influenced outcomes in major ways. According to Berman & McLaughlin (1975), the successful impie- . mentation strategies were those that promoted mutual adaptation. In the Rand Corporation’s study. Greenwood et al. (1975) found that pro- j jects showing evidence of mutual adaptation were the most likely to | 1 be effectively implemented and to persist. They identified some of i the major features of mutual adaptation as: - Reduction or modification of project goals. I - Amendment or simplification of project treatment. j - Reduction in the amount of behavioral change expected from j participants. - Changed organizational patterns. - Learning new skills or attitudes. (Greenwood et al., 1975, ' p. 29) i r ’ 118 These authors finished by saying: The type and extent of mutual adaptation that is possible in a project depends on the project design, particularly on how complex and specific the methods and goals are, on how flexibly the project can cope with unanticipated implementation problems, and on the motivations of the principal actors. (Greenwood et al., 1975, p. 29) ! I The study concluded that mutual adaptation is best achieved by a i problem-solving orientation, and further that in practice, complex : situations made great demands for flexibility and change— the price for real mutual adaptation. An implementation perspective emphasizes postadoption rather ' than preadoption behavior. Institutional dynamics and characteristics I ; of change strategies affect the outcome of planned educational innova tion. Examining innovation from the perspective of an organizational I model of institutional behavior stresses the resistance to change I that continues after adoption to influence processes of adoption and I implementation. ! Paul (1983, p. 3) differentiated between short-term opera- | tional management problems (e.g., performance evaluation, budgeting, | I and pricing) and strategic management problems, which give a larger | ' I i framework within which the short-term or operational problems operate, j I Most of the consulting and training programs today are largely con cerned with these operational or short-term problems. What is lacking , ' I is a larger framework within which these problems can be solved more j ! readily. Paul (1983) stated that a caveat is in order because stra- j tegic management is not a panacea for all the ills of development , 1 I programs. Repeated personnel shortages, scarce financial resources. P" ■ ■ .. 119 ’ ' and disorganized basic policies will abort development programs. The problem of motivation for achievement was mentioned. Major con straints in many developing countries are the degree to which people ! are oriented toward results or performance and the existence of good > leaders. If leaders and other people are not motivated, their | I responses to programs will be limited. I i I Paul (1983, p. 39) emphasized the need for the four factors— I environment, strategy, process, and structure (Figure 6)— to be I j consistent with one another, very much as seed, water, and fertilizer I ' ' reinforce one another in crop production. This phenomenon of mutual i consistency he called congruence. Economists use the term "production I function" to represent the relationship between output and its under— I I lying input combinations. Paul (1983) used the term synergy to I signify the relationship between performance (output) and its under- I lying combination of environment, strategy, process, and structure 1 I I (inputs). There is a synergy in the combination that no single input j alone can generate— synonymous with the "interaction effects" in Figure 6. j The strategy advocates sequential diversification: that is, j ' starting with a simple service and subsequently moving on into other I services. The concept of strategic management of development programs | j ' j is really a systems model. The management of a development program j I I ! should entail continual sensitivity to the environment and emphasis on I interpretation of the environment. This model of strategic management : ; includes the recognition that inadequate attention has been paid to ' ! I ; implementation of projects and programs. What is advocated is a more I 120’ INTERACTION EFFECTS AMONG THE FOUR FACTORS Accomplishment of Goals PERFORMANCE Structural Forms Decentralisat ion Autonomy STRUCTURE Service-C1ient-Sequence Demand-Supply-Resource Mobilisation STRATEGY Participation Monitoring and Control Human Resource Development Motivation PROCESS Opportunities, Needs Constraints, Threats Scope Uncertainty ENVIRONMENT Figure 6. The Four Factors: Paul, 1983, p* 39.) Linkages and Performance. (From i I thorough analysis of the demands and constraints of the program site ! : and a recognition that a program is an entity with its own organi zational structure, budget, and personnel. Problems of complex organizational structure (e.g., motivation and control of personnel, ; and decentralization) rarely arise in pilot projects but are crucial considerations in major or national programs. Although many development programs focus on supply or delivery of services, few program managers and designers know enough about the implications of environmental factors for their supply function. Mechanistic or standardized approaches to the design and delivery of services work well when the environment is reasonably homo geneous in terms of clientele, physical conditions, logistics, etc. As a program expands in size and geographical coverage, these features usually change. It is extremely important, therefore, for program managers to search for the following features which may have a direct impact on their supply tasks. (1) The degree of diversity in the environment: When there are heterogeneous groups of clients or beneficiaries in the country, the supply system must take this into account if supply and demand are to match properly. Problems and requirements may differ from one group to another and also over time due to cultural or social backgrounds, experience, etc., or to physical and geographical factors which make differences between regions significant. (2) Problems of Logistics: The quantity and quality of transport, storage, and communications vary a great deal from country to country. Environmental analysis helps us assess the adequacy of facilities and skills before planning the supply strategy. (3) Adequacy of Institutions: Programs that do not completely control production and delivery systems must carefully investigate the strengths and weaknesses of the institutions with which they must collaborate. These final choices have strategic implications. The larger the size and coverage of the program, the more critical the analysis of collaborating institutions. (Paul, 1983, pp. 47-48) I Congruence among factors is necessary for program performance and accomplishment of goals (Paul, 1983). If the combination of j factors lacks congruence, excellence in one factor cannot possibly i i 122 I I compensate for weaknesses in others. An innovative program not sup ported by an adequate structure will fail; a well-designed structure lacking proper internal processes will fail; a decentralized structure ; i ; with weak, poorly managed monitoring processes or with untrained or poorly motivated field staff also will fail (Paul, 1983, p. 111). ' The secret of strategic management, then, is in the "orchestration of j I 1 congruence." One of the essential guidelines for strategy formulation is that it is not a mechanical process. The strategy requires imagi- j nation and a sound understanding of the environment. This essentially ; ! interactive process requires the people involved to learn, adapt, and I ' resolve conflicts as they move on. While it is useful to know the : I ; components of a strategy, planning or implementation cannot be reduced j i ■ to a series of simple steps. I ] ■ Effective implanentation of rural development programs depends j i largely on gaining the ability to tap community resources (political, | ; I institutional, economic, and financial). In countries with a strong ' orientation toward central control, strong resistance to the evolution ' I of capable local institutions is inevitable. The centralized politi- j ' I I cal systems of many developing countries are linked with urban elites j I that are heartily pursuing efforts to keep rural people out of the j ; political process (Ruttan, 1984). Implementation, like other aspects j 1 of change processes, has the potential to encourage support and j : I , commitment from the people affected. But for a variety of reasons, \ ! planned change processes do not always succeed. Long-term success in ; I I improving the performance of development programs often is linked to I persistence in adversity, to openness, and to willingness to try new I 123 ! things and learn from past failures. These are difficult attitudes in themselves, and many donor agencies face the same problems. Î The implementation perspective views project development and t , project management in systemic terms. The "horizontal" or system- i oriented program logic of development projects is not, however, i consonant with the "vertical" division of labor that exists in most I governments (Morgan, 1983). In reviewing the literature on decen- ! tralization and development, Conyers (1984) pointed to the well- i established interest of the United Nations in such issues and the j ' relatively poor coordination between the many peripheral United ! Nations agencies working on decentralization. The same is true for I other donor agencies and disciplines working on implementation and on ; development in general. I Nisbet (1969) made the point that social interaction can be | found in social fixity and persistence as well as in social change. | ’ I The greatest single barrier to a scientific understanding of change, I Nisbet believes, lies in our refusal to recognize the sheer power of j Î conservatism in social life— the power of custom, tradition, habit, | F I and mere inertia. Implementation will continue to operate in a polit- ! ical atmosphere where on the one hand, resistance to innovation is I I j high, and on the other hand, demands to innovate are urgent in the face of mounting crises and limited resources. A number of distin guished scholars from various disciplines are currently arguing that not only governments from the developing world, but Western govern- I ments as well, will increasingly be forced to innovate. The success ' of these governments may well depend on their ability to leam to 124 implement innovations. This interplay between the sheer power of con servatism in societies and governments and their ability to innovate in the face of crises will be crucial to the success of development efforts. To sum up an implementation perspective, contemporary plan- I ning must be a multiprofession, multidiscipline team activity; it ' must accentuate the purposes of planning to produce a desired result, i not just a piece of paper called a plan; it is a means to an end, not an end in itself; and, if the development effort is to be effective, planning and implementation must be inseparable. All these points need to be repeated constantly (Conyers & Hill, 1984). Implementation Problems in Donor Agencies Logical Framework Analysis | The construction of a conceptual logical framework is a key | element in project design and evaluation in U. S. AID. As it is used \ I there, it is essentially a good example of a strategy involving a | series of steps or ladder. This logical framework can, however, be a serious hindrance to concern for implementation problems. Nonethe less, if it is used honestly, it can be an invaluable tool for project implementation. The Ü. S. AID logical framework can be portrayed as follows; I I Goal Indicators Means Verification Assumptions Purpose Outputs Inputs J 1251 The belief is that the logical framework, or "logframe," is primarily a device for project planning. The logframe method embodies the con cept of causality. In other words, there is a causal linkage between the project elements: resource inputs are expected to produce project outputs, outputs are expected to result in the achievement of a pro- i ject purpose, and the purpose is expected to contribute substantially ; to a higher goal (U. S. AID, Note 15, p. 37). ^ What the logframe approach often does is to delimit project ‘ I I activities to one logframe column, the inputs-to-goal column. Often j i it is in this column that the foreigners are operating; in the assump-' I i I ' I tions column is where the host-country people are found. In between i I I I these two is the contractor who implements the project. The logframe i I i is often an example of a blueprint or assembly-line approach, which j ' conveniently ignores actual constraints to project implementation by sequestering these constraints in the assumptions column. Although I the logframe can be a useful device, if used dishonestly it can I impair implanentation. Added to it is a contractual model of i I behavior: Congress contracts U. S. AID; U. S. AID contracts to the host country; either U. S. AID or the host country contracts to the : project contractor; and the project contractor contracts to carry the i i j project to specified targets. The project is conducted not to address | I actual problems of implementation, but to fulfill the conditions of a [ j preestablished contract. This logframe/contractual-model system is a j ( kind of production function scheme that specifies inputs to achieve measurable outputs. Often, everything that cannot be measured or | j neatly dealt with is put into the logframe assumptions column. In I 126 I many cases, the issue of incentives or of continuing incentives after the project ends is placed in the eussumptions column. The logframe analysis (or similar approaches) haus influenced most of the larger donor agencies* perspectives on project analysis, and they are comfortable with it. The logframe is amenable to central j . planning, it allows financial implications to be worked out, and it * , seems to be rational. In a strategy like this, judgment logically is j made as far as possible from the "action" (the target area of the I project). Everything is done according to a set of indicators, and | in a central system contingent on indicators, people respond to the indicators and not to the situation itself. It is somewhat like a centrally planned agricultural system: a farmer is told he must plant i 1,000 acres of potatoes; although he knows that the abnormally heavy | I rainfall will cause the potatoes to rot, he plants them anyway because | I 'he has no recourse within the system. If there is no freedom of action, the result is disaster, and this is true in the systems of donor agencies as well eis in the various bureaucracies in the devel- ! oping world. Reviews prepared by U. S. AID address what are called | I policy issues, vastly removed from the demands of the site and from implementation considerations. I ! Although this problem— the lack of practical attention to ! I I implementation by the donors— is beyond the focus of this study, not I I to discuss it would be tantamount to putting it in the assumptions, or ' igiven, column of the logframe. The context within which development programs function is crucial. The entire problem of implementation | i ! is rooted in the failure to focus on incentives and on the rural poor 12?! ■ which characterizes bureaucracies in the developed as well as the developing countries. Certain questions are implicit in logframe analysis. By logframe column, these are: 1. Goal (Narrative Summary of Goal Indicators. Means of Verification, and Assumptions) Why? How? ] 2. Indicators ^ What? How much? When? I Where? 3. Means of Verification— Where are data? 4. Assumptions— What are outside features (i.e., uncertain- [ ties)? A three-step procedure is used for means-ends analysis: 1. List key problems affecting the area in which the project I is intended to operate; 2. Place problems in sequence; and I 3. Invert each problem and put in a solution/objective. The I I elements of U. S. AID project design are hierarchy of means-ends linkages, probable external conditions, description of end-of- project status, and targets and indicator data. I iAn example of a logframe (Figure 7) with a hierarchy of means-ends I 'linkages will accent the point of the analysis. 128 Foreign "experts" operate here Contractor is in between Host country people are here GOAL— Healthy population INDICATORS MEANS OF VERIFICATION ASSUMPTIONS— Focused diseases are really causes of major health problems PURPOSE— To establish a periodic immunization service Hospital and service will serve popula tion as expected OUTPUT— One hospital; 150 trained nurses, 50 doctors, and one laboratory Government will provide ade quate students; government will maintain hospi tal INPUT— $50 million train ing assistance; 40 training assistants Can find ade quate training assistemts; training assis- sistants will be able to speak in lan guage and know culture; train ing assistants will be avail able in crucial months I Figure 7. Example of Logframe. (From personal work experience with U. S. AID; attendance at a one-week U. S. AID Project and Design seminar; and Grayzel [Note 16].) I 129 The specifics of the logframe here are not important. What is crucial is to understand that if the project is to conform to : reality and address the target-population needs, it must use a kind of circular or spiral strategy rather than a simple "stepladder" - model. The site, where implementation demands will be put on the project, must be known in depth. Assumptions will have to be included I and thoroughly thought out from the beginning. The technical experts simply have to start dealing with the assumptions (or what Heaver I [1982] called the politics of implementation) as well as with the ^ other aspects of the model. In the example (Figure 7), can adequate ' training assistants in fact be found? Will they really be able to communicate their expertise to the government trainers so they can I teach in the language and culture of the area? Are the training : assistants available at the right time? Where will the training be I located? Questions such as these must be addressed, or the project either will be co-opted (i.e., the site will capture the project) or , will simply not be implemented. 1 The above example of logframe analysis of assumptions within the context of a TBA training program is given to demonstrate realities of implementation that often are conveniently overlooked. I Another example will follow. These examples are neither exhaustive i I nor definitive— they merely accent difficulties in inqplementation of TBA programs and help to refocus the cause-and-effeet sequence toward la more realistic consideration in the face of the actual danands of I implementation. Personal experience in Mauritania (1980-1983) in working and talking with U. S. AID personnel has reinforced the point 130 * about assumptions columns in the logframe analysis. The reward structure of the U. S. AID system does not focus enough on positive efforts of project implementation. Compared with personnel from * other major donor agencies, however, U. S. AID personnel are generally more interested in and informed about implementation concerns. (These conclusions have been drawn from personal experience in Africa with U. S. AID personnel and personnel from other donor agencies.) i Example: Analysis of Assumptions for a TBA Program In logframe analysis of a TBA program, the goal assumptions are usually the health and welfare of a population. A TBA program, however, is only one facet of a health-care system, and work cannot I be restricted to that level alone. The goal is to focus on the TBA component within a multipronged (but perhaps critical minimum) PHC I effort. The extent to which different components are mutually reinforcing will be the extent to which the goal (i.e., the health and welfare of the population) is achieved. The input assumptions for a TBA program could be the following: 1. Only rarely will high-ranking, qualified people train TBAs and work in the rural areas, and these highranking trainers probably will not be knowledgeable about the local milieu. Nurses and doctors are trained in the hospital to work in the hospital, but TBAs may be trained in sophisticated environments (such as hospitals) that are not consistent with their working environments. 131 2. Inputs by a foreign donor will never be matched by the host country. Developing countries often will not or cannot keep up the project. 3. The professional and cultural understanding of doctors (e.g., French or American doctors) and the professional health care culture of their trainees will never be the same on all levels, although doctors and nurses from developed and devel oping countries share many attitudes that conflict with PHC efforts. There is a discordance from the start. 4. Traditional birth attendants are already inherently tied into commitments by requirements other than those of the program. Regardless of the individual worker, an understanding of these preexisting requirements should be included in the assumptions column. 5. Nonformal training (training picked up independently of a formal agenda) is one of the few approaches that places training in the context of TBAs* work. For a doctor, hospital training is nonformal (or informal); doctors get training in ; the realities of hospital politics— how to get things done, how j I doctors and nurses relate to one another, and how one specialist ! t relates to another; these lessons are invaluable and may be as , important as formal courses. The location of training is important for TBAs as well. Assumptions have to be grounded in ' practical training considerations. 132 The purpose assumptions for a TBA program could be the following: 1. No pilot program will automatically have all components needed for large-scale programs. 2. No existing national health structure has built into it i either a sufficient constituency or an accepted, established j allocation of resources to support a TBA program. 3. Converting successful outcomes to institutionalization will involve implementing support on a national institutional basis. If targeted outcomes are reached in pilot projects, the | effort must proceed to a phase-two task, which is invariably institutional and in most cases involves national support (i.e., a project does not end with outcomes or purposes but is part of another, larger project phase). The output assumptions for a TBA program could be the following: , 1. Trainees of all kinds will use their training to the best j advantage in their lives (economically and socially). ! 2. Trainees will need some systematic reinforcement on a continuing basis. I 3. Outside reinforcement of TBAs will continue only if their ‘ i supervisors are given the same prerogatives (money, status, or both) that they could get elsewhere. 1 4. The cost of programs can never be borne by populations who ; ( are poorer than city populations. If city populations cannot , do it, how can rural poor populations do it? 5. Resources will be needed to pay for gasoline, vehicle mainte nance, and effective program supervision. 133 j The "World Development Report 1983" An example of an approach with the potential to affect project implementation is the World Bank’s sixth "World Development Report" (World Bank, 1983). It focused on management and development. In beginning the "World Development Report" series in 1978, the World Bank claimed that its broad-based membership, its long experience, i and its daily involvement with the development problems of its ' members placed it in a unique position to analyze the interrelation- I ships between the principal components of the development process (World Bank, 1978, p. iii). The World Bank’s aim has been to acceler ate economic growth and reduce the deprivations of massive poverty. 'Since their inception, the World Bank’s "World Development Reports" have achieved a reputation both as an annual review of the world economy and as informal and practical contributions to discussion about how to tackle particular facets of development. The 1982 Report singled out agricultural development, and, as with previous reports, this one was built on the substantial program of investigation undertaken by the Bank and it included down to earth prescriptions. (Murray, 1983, p. 291) The "World Development Report 1983" was eagerly anticipated ■ because it would be striking out into new territory: it linked concerns about economic development and reduction in poverty-related I deprivation to programs for administrative improvement. The thrust i of the report is a reliance on the market as a highly effective and versatile administrative instrument; and the essential task is to reduce governmental and bureaucratic administrative activities that I interfere with (rather than sustain) the market. ; 134 The state plays a pivotal role; it is government that determines the policy environment in which enterprises and farmers must operate; government that provides the social and physical infra structure that underpins productive activities; and government that frequently contributes to production through state-owned enterprises. In many countries, the expansion of the public sector has stretched its managerial capacity to the point where serious inefficiencies result . . . [there is] a need to reassess priorities, prune what has become unmanageable, and strengthen the effectiveness of the state’s core responsibilities. Less reliance on controls and more on incentives to achieve social and economic objectives would also reduce the administrative burden on the public sector. When governments have tried to control too much economic activity, efficiency has been impaired. (World ' Bank, 1983, p. 46) j The report’s main argument is that the state is too large, and the i public sector is inefficient because of limited personnel in manage rial positions, corruption, political demands, and the absence of competition. The complexities of reality and the limitations of , I ; simple faith in the market are played down. Compared with central bureaucracies, the market— the report argues— can be more effective ' in response to real signals. If, for example, one ministry wants a good staff, this ministry may offer better housing or (on a more : basic level) access to sugar, but these incentives cannot be costed, and control over the system is lost. With many large bureaucracies in developing countries, there is a strong argument that the market | ^ (and that decentralization as a surrogate market) can help adjust the i system toward better functioning. ! I 1 The report, in its preoccupation with a particular theory of ' I economic development, ignores the complex question of whether this i I theory conflicts with basic needs, while barely acknowledging that | I "there will be occasions when the goals of efficiency and distribution | I I I conflict" and will have to be traded off (World Bank 1983, p. 43). j 135 The report conveys the impressions that the authors have a knowledge of a limited number of somewhat arbitrarily selected cases from which generalizations are confidently extended to all. Nevertheless, it runs counter to the prevailing thought in the study of administration that, in the manner of economic analysis, by aggregating a large number of individual instances and then ignoring political, cultural, and other variations, causal admin istrative connections can be isolated and identified and that , these can be translated into prescriptions which can be applied I universally. (Murray, 1983, p. 295) j In pursuing the idea of response to market incentives in developing countries, it is useful to realize that historically this ; ideology comes from British and American contexts. The British | I experience was during the Victorian industrial age, and the American experience has been from that time until the present. In both cases, money was to be made in the country; investment incentives coincided with British or American interests. What characterizes developing I 'countries is precisely the lack of a concrete incentive for invest ment; what incentive there is exists largely in services to urban I elites, foreigners, or both, which once again amounts to direct or ' indirect investment in developed areas of the world. Without the advantage of investment, the marketplace as an engine or mechanism for development does not fit neatly in the developing countries. I I The ideas that (1) there is some technical and ideologically neutral blueprint for administration; (2) political choices are , I : subordinated to universal, objective criteria; and (3) economic j I I theory guides other cultural viewpoints are all offshoots of a "top- down" development theory that characterized earlier development > efforts. A recent "World Development Report" (World Bank, 1984) j develops these themes along the same lines. What is missing is a 136 linkage to implementation problems in the less developed world. World Bank projects are not implemented by the World Bank; thus the bureaucracy is far removed from implementation demands. Personal discussions with World Bank staff members in March 1985 confirmed that implementation issues have not received their due. Instead of continuing the overwhelming emphasis on important macroissues, such as the growth of parastatals (organizations owned by countries, but . run autonomously), more emphasis needs to be directed toward micro- ! I issues, with a special focus on service delivery systems that are ' effective in reaching rural majorities. A more balanced approach would entail more emphasis on project implementation to complement the emphasis on prices and markets. Many of the World Bank staff have rarely spent time in vil lages with rural people, and this omission is a real constraint to implementation; hence the need for reorientation toward delivery of services to rural people. This basic shift in approach would neces sarily involve more knowledge of local systems and more focus on resource-constrained access bureaucracies. It would also confront head-on the demands of politics and cultural diversity and would encourage World Bank staff to consider the implications of their advocacy of macroissues in the context of project implementation. ; For a thorough discussion of problems in implementation of World Bank I educational technology projects in Africa, see Ofori-Ansah (1983). j Adaptive Performance and Incentive Problems in Donor Agencies ' Yet another institutional barrier to implementation is the necessity for adaptive performance. Moris (1984, p. 51) quoted an ! 137 observation that rural development organizations experiencing poor performance can respond in three ways: "a) by refusing to admit a problem exists; b) by projecting responsibility onto the external environment; or c) by learning from their own mistakes" (Korten, 1980, p. 498). Many other implementation specialists also stress the need for adaptive institutional performance (e.g., Rondinelli, 1983a, I ' 1983b). What works in the process of implementation, because it 1 responds to the contextual demands, is due to performance. But in [bureaucracies, people are evaluated via microassessments, and the overview of the implementation process is lost. Innovation requires experimentation, which inherently i involves successes, failures, or both. In many bureaucracies, however, any disturbance is interpreted negatively. The lack of a 1 process orientation (in the bureaucracies of developing as well as j developed countries) is a great impediment to implementation, which is usually innovative and inherently entails many errors, failures, and setbacks. Because there is a bias against learning from error, people in bureaucracies "hedge their bets" and either fail to report the actual situation or help to block implementation of the program. "For creating sustainable capacity, this is the central issue - how jcan existing institutions be encouraged to respond to error by ! improving their own performance" (Moris, 1984, p. 51). Rondinelli (1983a, 1983b) focused on adaptive performance in relation to rural 1 I development needs. What is being argued here is that donor agencies I are very much afflicted by their lack of adaptive performance. 138 At this point in development thinking, the best criterion for 'making Judgments and for finding tools useful in planning, implementa tion, and evaluation is whether a project works or does not work. Unfortunately, donor agencies often make judgments according to whether money was spent, documents were produced, a political riot was prevented, or other such considerations. Institutions and ; bureaucracies in donor agencies are set up for employment. Often, , people do not stay long enough to see implementation through, and j newly hired people can claim that the failure of a project was not i 'their responsibility. Many times, destructive people are rewarded by donor agencies; they become adept at criticizing, realizing that tear- ,ing down a project is easy, but that patience and time are required to see many development efforts produce results. A person working in ; development rarely receives either a reward for aiding project imple- : mentation or any satisfaction from making a personal contribution or effort. More often than not, people who are dedicated, hard working, ^ and in some way responsible for helping rural people by aiding project implementation are not around to receive the praise they deserve. The incentive to do development work must be internal, and the person has I to be extremely dedicated. I "Burnout" of donor-agency personnel is understandable, because of oft-mentioned factors such as working in "hardship" posts, but what I is often overlooked is the way people are judged in donor agencies. A I judgment of project activities should also be a judgment of the pro ject's designers and implementors which directly affects these people. Otherwise, judgment criteria such as indicators will be used, the 139 I success or failure of the project will not be relevant, and social- science expertise (which is sorely needed for implementation to suc ceed) will be looked at askance or— at best— peripherally. In judging people, the emphasis should be on the process, not on microissues; otherwise, the ability to learn from mistakes— a vital ingredient for institutional learning— will be lost. Learning can take place only ; if the focus of judgment is on actual project implementation, how it I occurs, and reasons for its success or failure. It seems likely that I much frustration of donor-agency personnel stems from this institu tional bias against judging performance in terms of ability to help the rural poor via project impl^entation. The following quotations address adaptive performance of incentive problems: j For the rural poor to lose less and gain more requires reversals: spatial reversals in where professionals live and work, and in decentralization of resources and discretion; reversals in professional values and preferences . . . and reversals in specialization, enabling the identification and exploitation by and for the poor of gaps - under-recognized resources, and opportunities often lying between disciplines, professions, and departments. (Chambers, 1983, p. 168) "The motivation of people in Washington is ridiculous," Grace [of the Grace Commission] says. "These people lead lives in fear of being exposed, completely defensive lives. Have you ever seen a hockey game? Then you've seen the team that's playing for a tie. They trail one wing, two defensemen never go down the rink, they never put less than four men on the opponent when he's trying to get into their goal. That's playing for a tie, and that's what these government people are doing. (No trouble. I'll just stay 'till I'm 55, get my indexed pension, stay out of trouble, wait it out.) Tie ball game. Doubtless the tie ball game syndrome is the affliction that debilitates most reform efforts in Washington. (Tice, 1985, p. 57) 140 ^ Although Tice was describing the Grace Commission's view of , bureaucratic culture in Washington, D.C., development bureaucracies all over the world are plagued with the problem of motivating people ' vis—a—vis their performance (the argument in this dissertation is that adaptive performance amd incentives should be linked closely i with achievement in facilitating the implementation of development projects). The literature on development and implementation makes it : seem as though the majority of implementation problems reside in the i developing countries— simply a false and insular assumption. The amount of money going to development (0.37% of the gross national product of the industrialized countries), the amount of money targeted to reach the poorest people in poor countries, and experiences with the problems of bureaucracies in donor agencies all ! refute the above assumption. 141 CHAPTER V NONFORMAL EDUCATION Nonformal education has been defined as any organized, systematic, educational activity carried on outside the framework of the formal system to provide selected types of learning to particular subgroups in the population, adults, as well as children. (Coombs & Ahmed, 1974, p. 8) I The term "informal education" attracted more interest in the mid- I 1960s. The 1970s saw wide acceptance of a three-fold taxonomy of formal, nonformal, and informal education as examples of three modes of learning. Coombs and Ahmed (1974) adopted a functional view of nonformal education, which they equate with learning. Nonformal education is contrasted with informal education, which is defined as the lifelong process by which every person acquires and accumulates knowledge, skills, and attributes from daily experience and exposure ' to the environment. The distinction between the two processes rests primarily on the programmâtic and systematic emphasis present in nonformal education. The distinctions between formal education, ' nonformal education, and informal education are made in the context of the delivery of educational services. Nonformal education is I organized but not fully institutionalized; it is systematic but not routinized. The context of its delivery is out-of-school (Bhola, ; 1983). 1 ! Because of the acute scarcity of resources in developing . countries, the option of expanding formal educational systems is ; 142 ' ! difficult, and the difficulty of expansion is compounded by rising costs applicable to these systems; It is clear that if [formal education systems] continue to grow at the present rate and under the same structural and managerial conditions, they will require funds that - while far below those allocated to education in developed countries - will be beyond the financial capabilities of many developing countries. The financial implications of enrollment by the levels of education projected by UNESCO [the United Nations Educational, Scientific, and Cultural Organization] for 1985 - goals that scarcely provide universal basic education - are that the GNP of developing coun tries must grow at least 7 percent a year just to keep the share I of education in GNP at 1975 levels. This is highly unlikely as judged from trends between 1960 and 1975, when the average annual rate of growth of GNP of developing countries was 2.8 percent in I low income countries and 3.4 percent in middle income countries. ; (World Bank, 1980, p. 70) Thus it was apparent that linear expansion of formal education could not meet demands for both quantitative expansion and qualitative change in education, and proponents of nonformal education believed ' that it had the potential for more efficient use of scarce resources, more effective delivery to previously untouched rural majorities, and more relevance to the demands of rural socioeconomic development. "At this point in history in the developing world, formal education can : not do it alone. Formal education seems to have failed both logis tical ly and functionally" (Bhola, 1983, p. 48). The nonformal education movement may be seen as an attempt to meet the needs of the rural (majority) populations in most developing I countries. Like primary health care (PHC), current nonformal jeducation evolved partly in response to a new social climate. Basic jeducation, PHC, and basic needs are components of a new belief— that I i I 'everyone has the right of access to education and health— resulting I : in part from development thinking of the postcolonial era. Certainly, | I - J 143 I under the colonial system, education and health were not considered a universal right; only the elite had access to them. Within this context a reassessment of education and health took place. The Declaration of Alma-Ata (WHO & UNICEF, 1978) and concepts such as nonformal education were products of this push to deliver services to a large portion of the heretofore underserved or unserved rural . masses. The reassessment of nonformal education became a reassessment j of tasks. Besides the intractable problem of logistics, was a formally trained teacher with a college degree or secondary-schoo1 degree really needed to provide minimal educational services to rural communities? In rural areas, was a physician who had taken 20 years to be trained needed to serve the health-care needs of the majority? From both a functional and a logistic vantage, the formal systems in health and education cannot be the sole answer to development needs of the poorer countries. The literature on nonformal education is full of invaluable lessons and experiences that can be transferred to various domains in the development field, especially with regard to ; the training of village workers. Nonformal education has been used I 1 in attempts to deal with (1) the issue of equal access to services, (2) the basic human needs, (3) the problem of bringing education and training closer to the world of work and life, and (4) the problem of promoting rural development through political participation and social development. Development activities long known under other titles such as literacy, agricultural extension, youth programs, women’s programs, adult education, and training of village-level 144 health-care workers were seen in a new light, because nonformal education was quickly viewed as a complement to the formal system. The frustration with expense, rigidity, and perceived low quality of programs in the formal school system has perhaps been the greatest stimulus to attention to nonformal education programs (Adams, 1972). Nonformal education thus is viewed as being more I flexible in adjusting to changing needs or demands and should result in a more equitable distribution of educational opportunities; I i The strengths of nonformal education lie in its diversity, its vitality, and its ability to respond quickly and creatively to local needs. A decade of experience with attempts to develop nonformal education has sharpened awareness of its limitations and reaffirmed its value in meeting many of the learning needs of developing countries. Nonformal education has demonstrated its capability of carrying out many educational tasks which cannot and should not be attempted in schools. (Evans, 1981b, p. 97) Obviously, for nonformal education programs to succeed, certain con ditions need to be realized; 1. A national commitment to the promotion of mass welfare, one element of which is expansion of an equity in educational opportunities. 2. Decentralization of planning and management in both education and development spheres and meaningful popular participation. 3. A dynamic context for socio-economic change arising from national development policies, priorities, and programs. (Ahmed, 1983, p. 36) I Institutionalization Recent experience has demonstrated a need to incorporate j various nonformal methodological features into formal institutions i and programs in order to cope better with a number of critical prob- , lems faced by educational systems of developing countries (Ahmed, ! j 1983). The question of how to institutionalize nonformal education I is relevant because whatever must be done systematically and with 145 some expectation of continuity needs a more or less institutional ized system (Bhola, 1983). There are two institutional issues in nonformal education that are related to organizational mechanisms for implementation of objectives or delivery of services (Bhola, 1983). The first issue is coordination. The nonformal education function cannot be assigned to one ministry, because ministries or ; departments of health, agriculture, family planning, and other ; fields invariably all have their own nonformal education programs. I The problem of administrative coordination has never been easy to manage, especially when several ministries are involved. On the one hand we have well founded advice that it is dangerous for projects to rely upon several sources of ministerial support (Moris, 1981:122; Johnston and Clark, 1982:180). On the other ' there is a pressing need to improve family nutrition and we can ! not arbitrarily restructure the inter-agency matrix to serve the nutritionists* convenience. The world-wide experience in managing nutrition is reviewed by Austin (1981). His itemization of causes behind weak management (1981:357) have much in common with those in the agricultural sector, suggesting an underlying problem of poor institutional fit. (Moris, 1984, p. 9) Review of the literature indicates that the problem of coordination is pertinent to PHC and traditional birth attendant (TBA) programs. If strategies to improve bureaucratic performance entail the , addition of middle-level management, the strongly hierarchical systems in developing-country bureaucracies generally have no provision for I I consultative contributions from middle-level specialists; the higher- 1 ' level personnel often are impervious to feedback from below in their I own bureaucracies, let alone from other ministerial systems. In the I j context of Third World bureaucracies, nonformal education has a real ' potential to uncover blockages (this topic alone is enough for a ..j I 146 ■ dissertation). Effective implementation strategies for basic and important policy objectives (categorized as socioeconomic goals) have yet to be developed in programs addressing the welfare of women and children. The second issue (Bhola, 1983) is the effect of nonformal education on existing institutions. Can the nature, function, and | ' performance of institutions with social, economic, and political sectors actually be changed through nonformal education? Practicable I TBA training programs reflect a conviction that the answer is "yes" | when nonformal education techniques are applied to TBA training and to linking TBAs with the formal health-care system. Training of TBAs may be seen as part of an effort to make the formal health-care system more responsible and accountable to the rural populations: Non-formal education has a much more important role in educating people in relation to the institutions that surround them and are supposed to serve them. Youth and adults can be taught to under stand institutions, to learn to put appropriate demands on them and to make them both responsive and accountable to the people. ; This appears to be a most promising area, though little seems to | have been done in this direction. (Bhola, 1983, p. 52) | Characteristics Paulston (1972) defined nonformal education as structured, systematic, nonschool educational and training activities of rela- I I tively short duration in which sponsoring agencies seek concrete j behavioral changes in fairly distinct target populations. Brembeck ' (1973) used a similar but broader definition of nonformal education: | i [ ’ nonformal education deals with learning activities outside the form- : I ally organized educational system, aimed toward specific goals, and j sponsored by an identifiable person, group, or organization. I _j 147 Facing the challenging task of planning nonformal education, Evans (1981b) cited two general principles. First, the amount of planning should be limited and the planning responsibility placed as close as possible to the learners. Second, the planning procedures should differ according to the characteristics of the program being planned. Evans (1981b) emphasized a philosophy of minimal planning, a * 'less-is-better" philosophy. Six factors that have brought attention to the potential of j nonformal education are: ' 1. Decreasing resources and expanding school age populations. 2. Equal opportunity and access to education and the resources of society. 3. Need for educational innovation. 4. Supplementing and complementing the benefits of formal education. 5. The need to meet human needs in specific contexts. 6. Possibility of shifting attention from school learning certificates to performance as criteria for achievanent. (Brembeck, 1973) \ The terms "nonformal education programs" and "innovations" are interchangeable. Thus, literature on planned change indicates that nonformal education programs are more likely to be accepted if ! they are perceived by clients as better than alternatives, fitted to . the environment, understandable, amenable to trial, and observable. In a summary of literature on change and nonformal education, several I important points were made: Importance of larger context within which the program exists takes on added dimension. . . . Importance of understanding the J nature of client population. . , , Nature of educational process is a two way process. . . . Nonformal education must go beyond I traditional processes associated with teaching and learning in I schools. . . . Importance of providing incentives for altering I behavior. (LaBelle, 1975, pp. 45-46) 148 A caveat is in order in the literature on change process and diffusion of research. Three difficulties have been noted: 1. Lack of process orientation is a problem. Every textbook definition states or implies process but Arundale (1971) shows in a review of research design that communication almost never allows for over-time aspects of communica tion. [This is crucial in PHC and TBA efforts because the approach is innovative and because change is incremental.] 2. There is a pro-innovâtion bias and associated ignoring of causality. [The argument in this dissertation is that TBA programs generally have this pro-innovation bias (particu larly as the focus is on training and not pretraining and ; posttraining considerations), plus a lack of focus on the ! context of the training.] I 3. Psychological shortcoming structure in which structure and I function disappear and one is left with cell biology. [This ' is why a systems approach is advocated.] (Rogers and Shoemaker, 1971) A rather philosophical consideration of nonformal education includes : the following: j Nonformal education is a new name for an old concern. The need to provide learning opportunities to both the schooled and unschooled throughout their lives, has been expressed before in such terms as basic education, fundamental education, functional literacy, adult education, out-of-school education, second-change education, continuing education, recurrent education, extension education, and life-long education. Behind these words and phrases, rooted in somewhat similar social and educational doctrines, there is a conception of education sus being "an existential continuum as long as life" smd an enterprise larger I than formal schooling. (Faure, 1972) From a different perspective, the characteristics of nonformal education programs are: 1. To complement formal education. 2. To be diverse in terms of organization, sponsorship, and methods of instruction. 3. To be voluntary and appeal to a wide range of ages, backgrounds, and interests. 4. To exist where clientele live and work. (Calloway, 1973) A study sponsored by the United Nations Childrens Fund (UNICEF) (Coombs, 1973) concentrated on nonformal education for rural 149 children and adolescents in order to accelerate rural, social, and economic development. The conclusions of this study suggest a need in rural areas for multiple learning systems (blending formal, non formal, and informal learning into a coherent whole) that rely on community input and that function in terms of the goals of rural development. A study sponsored by the World Bank (Coombs and Ahmed, I ! 1974) similarly concluded that nonformal education can be effective in rural development if it is articulated with complementary efforts. I Literature on nonformal education (Calloway, 1973) suggested that non formal education programs lack (1) resources to carry out intended programs and to admit the number of individuals attempting to matric ulate; (2) organization, resulting in considerable duplication of effort; (3) evaluation of outcomes in terms of effects; and (4) incen- ; tives to enable clients to complete programs (e.g., job placement). These deficiencies exist in TBA programs as well. The hope is that nonformal education programs might be less expensive ways of educating ; and training more people and might enable trainees to improve their quality of life without prior acquisition of formal credentials. Any theory of development or education will, in general, be I inadequate if it is not conceived in terms of constructive sociology 1 aiming at an all-around improvement of the life of the villager and ^ community resident in rural areas. Agricultural change, health change, and educational change must go hand in hand. Without foster ing multiple interventions in the sociocultural system to enhance the possibility of achieving the desired ends, change agents and educators are likely to see their programmâtic efforts swamped or overcome by J ! 150 1 other, more powerful components of the system (LaBelle, 1975). This concept is akin to Paul’s (1983, p. 110) synergy among factors (environment, strategy, structure, process), and Moris*s (Note 17) conception of systemic interlock, which closely parallels LaBelle’s. Mistakes in the agricultural sphere, the educational sphere, the industrial sphere, and other spheres produce a momentum after a cer tain point: the negative effects are so powerful that an isolated 1 intervention gets swamped. Factors contributing to systemic t I : interlock include: 1. In agriculture, spending money on irrigation to solve food problems is extremely expensive. 2. In education, the massive buildup of secondary schools (in the unanticipated absence of cash economies) has created the need to build universities. 3. Corruption is widespread on all levels. 4. Leaders in developing countries will be stimulated to do a lot of things contrary to the nations’ interests. The above is a realistic appraisal of the situation in many developing countries. Academicians without cultural experience in implementation, as well as donors, tend to play down these factors, I although an attempt to deal with them via a new strategy of donor I ' coordination may be starting to take shape. The role of a planner in nonformal education becomes one of coordinating existing efforts and j stimulating new ones. In areas where illiteracy is high and formal I schooling is scanty, planners of nonformal education programs may choose to reach larger groups of people with basic education and to ! 151 opt for participatory strategies that accentuate community or group efforts and "consciousness raising." Less emphasis may be placed on specific skills such as literacy. Rural problems often need polit ical pressure from the general population, but such pressure is difficult to achieve as long as the rural population is largely illiterate and poorly informed of its needs. The root causes of many intractable problems in rural areas are socioeconomic deprivation, ! centralized and local managerial incompetence, and mass illiteracy, j Nonformal education programs are viewed certainly not as a panacea, but rather as part of a not-too-simpie strategy for reaching under- : served or unserved majorities of rural people in developing countries. Postadoption Evaluation Kinsey (1978, 1981) has highlighted many problems with evalu ation of nonformal education. One major problem with many evaluation activities is the tendency to be oriented toward measuring program outcomes rather than assessing inputs and improving continuing processes. In nonformal education, there are more pressing time constraints and thus more concern with improving program effective ness than with devoting energy to outcomes, which often appear to be i t a low priority. The history of evaluation (Kinsey, 1982) is geared i to long-term benefits, outside audiences, and "lip service" to forma tive (process or continuing) evaluation considerations. In many nonformal programs, the inputs and support systems are so weak that any summative evaluation effort is premature. A second problem is that many evaluation strategies and methodologies are not adaptable to the context of nonformal education 152 programs. Often, presuppositions do not exist; many nonformal edu- , cation programs have unstated or confused objectives and even vacillating content, and the programs may be intermittent, without clear beginnings and endings. Assumptions of stable program per sonnel and participants, and some degree of uniformity, may be misleading. An important consideration in evaluating nonformal ! programs with a participatory approach is that teachers are viewed as learning facilitators; the programs have a "bottom-up" approach and I j tend to be responsive to participants’ needs. Unless the philosophy of the evaluator is congruent with that of the program, there will be opposition or outright rejection from both the staff and the partici pants. Another important consideration is that evaluation strategies or methodologies often cannot be used by practitioners as part of the continuing development of nonformal education programs. The necessity for two groups to be involved in evaluation is discussed below. 1. Funders, academicians and program policy makers have to value criteria that stress immediate usefulness to a program and useability by its personnel on an ongoing basis. Beyond valuing, efforts to meet this need have to be given support and reward from these quarters. To balance the typical ' emphasis placed on generalization, macro-funding decisions I and academic soundness per se in evaluation, there also has to be a willingness to sponsor the development of rough but adequate methods which in the short-run may do nothing more than help specific programs improve their inputs. In the case of many nonformal education programs this "less" has the ' promise of amounting to more. I 2. A second group, those concerned with evaluation and training j methods, have to be willing to take criteria for adaptation, I such as those noted above, seriously in developing practical options for use by program personnel. Straddling the worlds ! of theory and practice they need to relate the strengths of j one to the needs and constraints of the other. Once a I variety of options relevant for different levels of field 153 constraints is devised, these need to be field tested and revised. In other words, the developer needs also to be a practitioner in the sense that he follows through at least initially with the practical application, evaluation and improvement of the adapted methods themselves. (Kinsey, 1978, pp. 27-28) Kinsey (1978) went on to say that in the final analysis, the value of the above approach and the resulting methods is measured in their use (and their expected use over time) by program staff. He made an j important practical point. For a given adapted method, is there evi- I dence that it is being adapted and used repeatedly by practitioners i 1 i in nonformal education without external direction or rewards? If a developer were rewarded on the basis of this criterion as much as on ' evidence of publication, more progress might be made. This incentive I I problem arises with respect to universities, bilateral and multi- • ; lateral agencies, and private groups or individuals working in ! development. A people centered approach requires special effort and a willing ness to invest in the development of new methodologies. So long as the ultimate imperative of development bureaucracies is to | generate project documents and move money, the impact of develop ment programs on people will remain an assumed linkage - despite clear evidence that this modus operandi results in substantial leakage of benefits. (Garner & Korten, 1982, p. 21) ' While the literature on development abounds with discussions of the j incentive problan in the developing world, the same problem exists in j I ' [ a totally different context in the developed world— and it is found in all types of bureaucracies and contexts. i I I 1 I Many scholars in various fields are beginning to reject the j I bias in evaluation toward the developed world and its emphasis on aca- i ! demie soundness. Honadle (1982), in discussing rapid reconnaissance J 154 techniques for mapping and molding organizational landscapes in development administration, opted for reconnaissance techniques in design, implementation, and evaluation of rural development schemes. These are data-gathering techniques that generate useful, socially meaningful information, and these rural development estimates are also economical. Honadle (1982) noted three important weaknesses in ; rapid methods of rural appraisal: a lack of attention to (1) the ! situational sensitivity of the indicators used, (2) the organization j and management dimensions of rural development, and (3) the need for such exercises to have as an objective the enhancement of local capacity rather than just the dissemination of information to out siders. If an evaluation is academically and rationally sound, yet irrelevant, its major achievement is to waste money in producing the evaluation report. ; Methods other than rapid reconnaissance have similar concerns. | One of these methods is operations research. Statistical precision is not as great a concern in most of the ' projects PRICOR (Primary Health Care Operations Research Project) j funds as is reducing uncertainty to a reasonable level. [ (Reynolds, Note 18) j I i The following is an example of operations research. , A decision maker might specify, well before a decision has to be i made, the options available and the projected benefits and costs 1 associated with each. With this information, the methodologist I ! could then design an evaluation that would provide the best infor-i mation possible within some set of budget constraints. The | researcher would subsequently carry out the research and report the results to the decision maker. At this point, presumably, a well-informed decision could be made. The short-run benefits are apparent: the best information per dollar could be brought to I bear. In the long run, all parties would learn from one another, I and each group would perhaps be better equipped to pursue both 155 individual and group endeavors. (Berk, 1984 [Decision Analysis for Program Evaluations, p. 7], as quoted by Reynolds, Note 18) Rondinelli (1983b, p. 327) cited Chambers* suggestion (1978) to displace sophisticated techniques of cost-benefit appraisal with simple decision matrices— ranking systems that focus on the effects of alternative project characteristics and approximate cost compari- I sons rather than on elaborate rate-of-return and cost-benefit studies, Rondinelli (1983b) believes that because of the uncertainty and the i paucity and unreliability of data in most developing countries, such I simplified methods of analysis are usually more than adequate. Although the literature abounds with metaevaluation studies, the information is generally useless because it lacks an audience. Such evaluation is done on a megascale by a mandate from on high. Even though the evaluation was done in a developed country (the I United States), the message is the same: the users or the community ! ; I were neither involved nor interested in the evaluation. Not only are i large, sophisticated evaluations prohibitively time-consuming and expensive, but also they rarely delineate with any accuracy the cause-j ' and-effect relationships in PHC schemes (Pyle, Sabin, & Martin, 1982, | p. 5). Franzel (Note 19) mentioned the scarcity of research resources | ; i I in developing countries, and he believes that the formal survey exer- | cise can be considered superfluous if it does not lead to significant I I I improvements in the accuracy of information and the design of experi- ! I I ments appropriate for farmers. The following is his conclusion from i comparing a formal with an informal survey: i_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - , . j 156 In summary, it appears that the contributions of the formal sur vey to developing an understanding of the farming systems and an experimental program for Middle Kirinyaga were rather marginal, relative to its costs. (Franzel, Note 19) Awareness of the importance of noneconomic objectives of small rural-development projects and larger programs has led to more appro priate designs and techniques for evaluating social development programs. A model for project evaluation, developed by Wahidul Haque and his colleagues, identifies criteria within broad areas as follows: The Economic Base: Attitudinal Criteria: - economic benefit - distribution equity - collective accumulation - horizontal accumulation - both expanding the size of participation and of multiplicity in other areas. - developing social and institutional linkages - with other self-reliant efforts. - sense of solidarity - democratic values - spirit of cooperation - collective spirit - collective self-reliance. Self-Administration and Momentum: - experience in economic and social administration. - generation of internal cadres - the "spread agents" who will be responsible for mobilizing the group for different ! economic and administrative tasks. - indigenous momentum - in material, institutional, psychological and leadership terms to develop the self-reliant basis of the project. (Haque, as quoted by Oakley, 1982) With established criteria for evaluation, Oakley (1982) saw the prob lem as identifying, observing, and recording data. The solution may . may lie in description and interpretation rather than in measurement and prediction. L _ ! ' 157 Many scholars have stated that evaluation of social develop ment projects should be a participatory exercise, in which the project group itself contributes to the description and interpreta tion (Feurstein, 1982; Kindervatter, 1982; Kinsey, 1981). It has been concluded that "The only characteristic of an evaluation system associated with utility was the degree of involvement of the user in an evaluation activity" (Waller [1977, p. 11], as quoted by Reynolds I [Note 18]). The value of the emic (insider's) perspective is I emphasized in the following quotation: ! : An emic perspective and ethnoscience methods are needed for organ izational research centered on the native, that is, the insider's or, as anthropologists call it, the "informant’s" view of reality. Thus, the emic approach emphasizes native or respondent categories ' and meanings in general and native rules . . . , or respondent ' behavior, in particular. Etic designates the orientation of | outside researchers, who have their own categories by which the subject's world is organized. (Morey & Luthans, 1984, p. 29) Morey and Luthans (1984) argued that information gathered from an emic perspective by use of ethnoscience techniques would balance the precision of quantitative data with the context and richness of I qualitative data. Knowledge of subject understandings, perceptions, 1 cognitive processes, meanings, and intentions are crucial— especially when working with development programs in rural areas. An anthropo logical perspective implies that social meaning and social order are | ! shaped and created by individuals in the course of social interaction. \ I ' It is this "meaning" system of social life that forms the common frame I ! of reference for individuals. Given this emphasis on the meaning of events to the individual I who experiences, interprets, and reacts to them, the starting point 158 for qualitative analysis in conceptualizing the social world is to develop concepts and theories that are grounded in the data (i.e., concepts and theories that are first derived from the data and then illustrated by characteristic examples within the data) (Morey & Luthans, 1984). These "first-order concepts" are essential to the development of "second-order concepts"— that is, concepts that emerge from the scientific discipline that is attempting to explain a phenomenon. Any scientific understanding of human action, at whatever level of ordering or generality, must begin with and be built on an understanding of the every day life of the members performing those actions. To fail to see this and to act in accord with it is to commit what might be called the fallacy of abstractionism, that is, the fallacy of believing that you can know in a more abstract form what you do not know in a particular form. | (Douglas, 1970, p. 11) | i Basic data-collection activities such as participant observa tions, unstructured interviews, observations, and diaries are used to tap the "meaning" dimension of social life. Questions (What? Who? ^ When? Where? How? Why?) must be asked when looking at the com- ' munity. Answering all these questions and getting to the "why?" | dimension creates a system. The solution, then, is a problem-solving | approach. Both quantitative and qualitative approaches entail participation. In the training of village workers by nonformal education techniques, knowledge of the "meaning" dimension of social ! I life is imperative for a workable training program. Kinsey (1981) I i believes that participatory evaluation is a useful approach in devel- i ; oping nonformal education programs. Participatory evaluation has i 159 1. Pedagogical potential (it can augment learning— individual and group— through evaluation); 2. Programmatic potential (old and new staff members can gain a better understanding of the program through participation in an evaluation, enhancement of communication and trust, reduction of i conflict, and stimulation of motivation and commitment); 3. Potential to increase self-esteem and efficacy (Provus [1971] postulated that self-evaluation may be the essential mech- I anism in changing individual and group behavior); and | 4. Political potential (it can link users to the broader com munity or to outside institutions). All these are possible benefits of participation. Some constraints on participation also merit consideration: 1. Limits of the purpose of the evaluation...a need for quick * results for immediate decision, a complex purpose implying a sophisticated methodology; 2. Limits of the content of the evaluation...the issues of rele vant variables are too distant from experience or awareness of specific participants; 1 3. Limits of the pool of participants...the participants are not I culturally attuned to giving critical feedback or assessments; existence of inhibitions on expression by hierarchical rela tionships or vulnerability; 4. Limits on time...the time required for a given type of évalua- , tion or participation exceeds that available for potential | participants; 1 5. Limits of place...the potential participants are dispersed or ; not in convenient geographical groupings; and 6. Limits of cost...the opportunity cost of pulling staff away from other tasks, or learners away for learning, is higher ! than anticipated benefits, or the implied logistical demands, j financial costs or exposure are deemed too high. (Kinsey, 1981, p. 160) I One problem in nonformal education projects is to get maximum j I input from project-participant perspectives in addition to the client ! 160 or community inputs. In looking at evaluation strategies, one advan tage of the goal-free approach (see Table 7) is that "by ignoring project goals as the focus of inquiry, it elevates the project parti cipants’ perspective to a much higher level of prominence" (Pietro, 1983, p. 74). The disadvantage of a goal-free approach is the dif ficulty of determining the standard against which to assess identified 1 ‘ effects. Scriven (1972), who originally proposed goal-free evaluation ! I and the concepts of formative and summative evaluation, suggested that ^ one of the best ways to conduct a goal-free evaluation is in tandem i ; with a goal-based approach. A much richer description would result, 1 but cost factors would be a constraint. In no case is it proper to use anyone’s goals as the standard unless they can be shown to be both appropriate and morally defensible (Scriven, 1972). The approach uses I open-ended questionnaires and observation schedules in an attempt to "measure" the learning environment. Goal-free approaches lack the vigor of goal-based approaches, but there is less likelihood that I I valuable information will be lost. The goal-free approach to evalu- j ation is often viewed as a "kind of taste" or personal preference j j instead of a theoretical orientation. I Goal-based approaches include the following steps: | ; 1. Determine the goals and objectives of the program; I 2. Translate them into measurable indicators; and 3. Collect data on the indicators for those involved in the I program. i A goal-based approach focuses on assessing the degree to which goals ! (primarily educational) are achieved. The approach is quantitative 161 ( U r —I I •H "S u 4 - > CO § •H I c d g O 44 •S 1 & G G G 44 G X t G 1 G -M G 1 •H 1 O G G G G 44 •H 44 G rH G *H 44 ü O ü rH I —1 G 73 G G 44 G O *H C h G G G A G G 44 G G r4 G G o G G X G cr G > O G A G O O* G G H G 44 1 - 4 G ü G G G G bO bo G <w G G f m m f O A 73 G G G 73 01 O 44 G G 73 G •H G G O •H •H G rO G G 44 > 44 G G G G 73 O bo G O G A G bO > 73 G G G c S *H A O G G 73 G O G > G 44 r 4• r 4 44 G rH O •H G > ü G G G >> G 44 G A • p H G *H G G 44 G G G o c a G O > G 44 40 G 43 G G A r - 4 0 ) A G G G O G bo > A •H G o 44 X «H 4 - > « G 44 44 44 G g rH •H O O ü G T3 1 G O X G G •H G G G ü •H «H 44 G G 'G G 44 G .H a G G 44 "G 4 - > t o G G 44 G 73 G •H •H 01 G G ii o > bo G 44 G 44 G G •H G o 73 73 G 44 s A •H G O ü O G G G G G C h G O A G 44 G ü G ü G 73 a J g G > G rG 'V G G 73 'G G ü rH •H •H > 44 G Q 44 ca 44 o G G G G 44 •H G bO G C O •H G G ü G G G 44 44 ü ü 44 a G G 44 bo G G A G A G 43 •H G G Ü •H ü 1 G • r 4 G G G 44 G 44 ü G G & 0 ) G G G O 73 rX 73 73 G > •H G rG 73 > G 73 L , G G G G G 2 G G o •H G G G G •H •H G A cr x> <5 A G S t —1 G & 44 O •H G M 44 > G c- 1 ç- m 0- G 8 G 1 a TJ > G ç- 43 G o <D G *i4 G G3 73 G 44 44 44 G G •H > 43 44 44 G G •H •H G G 4-i 0 ) 44 ü G 'G o rG G G 44 O Oi G C-- S G G G 44 'H 01 G •H •H 52 43 G C h G TJ O G G 73 G G G :s Q L, C h G G ü 44 r - 4 ♦H G bo O C h G G Of (0 e- G G •H 44 G G G -G G 73 C h G > 3c G *H A G •H > -G G -H *H O m C Q 1 - 4 44 44 *H 0 G G G G 44 73 73 G 0 r - 4 G ü G 73 G G O G G •H Ü t d c ^ - O G O O r —1 *H rH A o & G o o bO *o ü 8 f - 4 G G G G A 44 O rG b bO r - 4 O G O 43 G G G 33 44 44 44 G 44 G 44 O rG G O- r —1 Q G rG A •H 40 G 'G G •H e- 0" rX O G c d 4: G bO G G 01 G G 44 G 44 •H C- 44 G Ü 44 •H ♦H G 'O G 73 •H O G ü «H ü rH G G *H ü G 43 1 - 4 G G 73 G G G rH 73 G A 0 > •H 44 G 44 44 G 44 C h 44 rX A G ü S m «H G O G G 44 G » O o G O O O O G E - t o «H 43 G 43 G G O G G G O G A O :& M 44 1 - 4 CO •H S •H g G3 A A A rH A G ü Oî G 44 > G 1 44 G •H O G G ü S2 44 •H bo G G G S ü 44 G G bO 44 "b ( U < u G G G 44 G G o m > ê G G G G G o <D 4) G O G •H G A A •iH O O rH «H 40 bo 43 C h rH C h G 73 bO 0 Ü td G G G G O G G A G G •H C h G *H «H G 44 44 73 U bo bO bo ü G 73 G G O G G G G G 43 G G G G • r 4rH •H •H rH G 44 0 1 a m G O i O C h G G 0 1 z a > O -H 01 G O > G G G < D bo P * G A G 44 G ü C Q O 01 G G 73 O «H G 01 C h G G G G G < O A < o G S GO A Ü •H a 44 o < U G 01 •H 01 G G 44 •H Ç Q C O O G G rH a m •H bO A 44 G G 3 1 G G 1 G 8 G m r - 4 •H ♦H rH G bo G ( U k , G ü G A 73 44 43 ( Ü o G G O X G G EH A O f = > z O m t - b % £ A i 2 % *H CL, I <2 •s 4 - ) Î < D 4 - > ê J 1621 I and scientifically based, but as Weiss (1972) said, the methods and tools of scientific research are often applied in an action context that is intrinsically inhospitable to them. Popham (in Scriven, 1972) wanted results, not rhetoric; he stated that educators some times become so enamored of writing precise, instructional objectives i ; that the mere act of writing them is not only the beginning but also ! the end of the instructional "ball game." These attitudes are closely I allied to Max Weber’s formulation of bureaucratic authority; Weber I : emphasizes its impersonality, rationality, and expertise, charac teristics especially inappropriate in rural areas of developing I countries. Institutions or bureaucracies are not in ideal situa tions; often they are impoverished and ineffectual. In such situations, nonformal education is invaluable in the evaluation process. ; The CIPP (Context, Input, Process, Product) model of ^ , Stufflebeam (1971) deals with a variety of realities by identifying ’ I (1) the process of decision making, (2) the settings in which it occurs, (3) the decision models most appropriate for each setting, (4) the basic types of decisions that are made, and (5) the types of I I evaluations that serve each best. The essence of the approach is to | I I supply information to decision makers. The notion of goal orientation ' is extended: it is equated with product evaluation. In this approach, I however, the decision-making process is viewed as essentially rational I rather than political. | ' The naturalistic approach or persuasion focuses on a systems ■ ! understanding of program operations. It draws its inspiration from ... J 163 "softer" disciplines such as anthropology. It accents pluralistic values, rejects the notion of one truth, and considers multiple perspectives and conflicts. Besides being process—oriented when it looks at projects and pro grams, naturalistic evaluation tends to be process-oriented in its own implementation as well. More traditional evaluation * approaches tend to be what Stake calls "preordinate." The design i is set at the beginning of the evaluation and implemented accord- . ing to the pre-specifled plan ... in contrast, the design of a naturalistic evaluation is emergent. That is, it evolves through out the period of the evaluation itself in response to the needs I of the stakeholders and to the issues as they develop. Feedback from the evaluator to audience is provided on a continuous basis, often informally. In transactional evaluation, the evaluator becomes part of the operating system as much as possible, pro- , viding information to the program participants as needed. He uncovers sources of conflict in a program and aids the proponents and opponents to develop and impl^ent evaluation plans which will ; help to resolve their differences and improve program performance. (Pietro, 1983, pp. 80-81) ' There is a tradeoff in ability to prove cause and effect and in measurement precision; the approach also may not suit donor agency , requirements. In a provisional model for evaluation (Figure 8), the focus on "clarification of who wants to know what and for what purpose" is j important. The design of the evaluation activity is, in fact, ' i strongly influenced by this clarification dimension. Generally, expectation of relevance is relatively high in nonformal education, so the question of clarification is crucial. There is also an ethni- ■ cal dimension as to who should do evaluation. In TBA programs the j trainers, clients, and local communities should be included in eval- ! uation studies, and the information gathered should be channeled into I the ministry of health (MOH) to refine the program. I I I 164 _ _ I REPORTING OF RESULTS ANALYSIS OF RESULTS DATA-GATHERING, IMPLEMENTATION DESIGN OF EVALUATION ACTIVITY Assessment of CONTEXT, RESOURCES AND CONSTRAINTS PRELIMINARY CONSIDERATIONS AND DECISIONS Clarification of WHO WANTS TO KNOW WHAT AND FOR WHAT PURPOSE? Figure 8. Provisional Model for Planning and Implementing Evaluation. (Adapted from Kinsey, 1981.) ' 165 Supervision and Management Supervision is perhaps the most important element in TBA programs, and the lack of supervision is invariably cited in the literature as one of the salient shortcomings of TBA programs. Centrally oriented formal health-care bureaucracies face uncer tainty and complexity, especially with respect to delivery of PHC services addressing local needs. I I It seems that standard U.S. managerial prescriptions give predic- ! tive results when applied to the internal operations of Third ' I World bureaucracies, but fail to cope with the relationships of a j unit to its environment - a conclusion reached by Kiggundo, Jorganson, and Hafsi in their recent (1983) review of 94 ' studies. (Moris, 1984, p. 35) Insofar as staff productivity is concerned, a ministry in a develop- I ing country can easily get into a situation where its own personnel I have incentives to act in ways that sabotage the realization of organizational goals (Heaver, 1982). It is apparent that assisting ; poorer, disadvantaged groups requires procedures and approaches on * , the part of formal health-care agencies that differ considerably from < the usual norms of the typical public agency and from the norms and values of health-care professionals in general, whether in donor i ! I I agencies or in formal health-care systems in developing countries. | Perhaps the key determinant of TBA program effectiveness is I the degree of task commitment of supervisors to their Jobs. This , ! factor is intimately related to the concept of motivation. Nonformal ! j education is uniquely suited to address this question because: 166 1. In the context of delivery systems, it is environment based (it can take place in the supervisor’s actual work place, where all his or her powers can act); 2. It is community related (generally, supervisors are the personnel who work closest to where clients live and work); ' and 3. Its environment is ' a. functionally related to where the learning (or I ; supervising) takes place, b. flexibly structured (a variety of sequences, structures, and relationships that generate useful data are possible), and c. learner centered (in the supervision context, the | emphasis could be placed on learning what forces are operating on a supervisor who is using a variety of j sources). j Many concepts central to organizational analysis have turned | out to lack intrinsic coherence when subjected to intensive field | investigation. ! ; For example, this was Hertzberg’s fundamental insight about "motivation”: rather than being a unitary organizational property , that can be treated as a variable (i.e., "low" vs. "high”), it is | a resultant outcome reflecting the mixed reactions of personnel to both satisfying and dissatisfying elements in the work environ ment (Hertzberg, 1966; Myers, 1964). Minor (1980:76-105) gives a | j critical update on Hertzberg’s theories. Motivation is possibly i ! the most studied and the most misunderstood topic in organiza- ; tional behavior. (Moris, 1984, p. 22) Coupled to the question of motivation is neglect of the management I ■ factor. 167 Neglect of the Management Factor Poor public sector management is recognized as a key constraint on poverty-oriented rural development. But it is an elusive and neglected constraint because: (a) It is hard to separate the results of poor management from the many problems in poor countries which are beyond managers * control. (b) Management research has concentrated on the upper levels of bureaucracies, rather than the middle and junior officials who actually implement projects in the field. It thus does not meet the practical needs of managers or advisers. (c) Habits of thought and appraisal techniques have not changed to meet the needs of the 1970s generation of management-intensive rural development projects. I (d) Professional bias encourages the wrong diagnosis of across-the-board organization and management problems as specialized, technical, sectoral problems. (e) Funding agencies have little access to key management processes in implementing organizations, and hence to the causes rather than the symptoms of poor management. (f) The access problem is compounded by the political sensitivity of management issues, especially the controversial question of who has control of scarce ; resources for what ends. (Heaver, 1982, p. iii) The emic perspective could certainly be utilized if super- ■ visor surveys were conducted in the supervisors* work context. Such an option would entail great donor input of money and time, but in rural communities in the developing world, the time spent between the supervisor and the TBAs is very brief, and information useful for j ( I I national TEA programs will not be obtained by one field supervisory | ' I ; visit per year. Use of the nonformal setting over the long periods I ' I of time required for supervision of in-service training needs to be , I studied, particularly with reference to the motivation of supervisory I staff. The need to elucidate key management processes over time ! requires anthropologically oriented research with an emic perspective. J - 168 The issues of incentives, commitment, and compromise need attention. The very idea of influencing incentives has overtones of manipulation and conflict. Given that the aim is to minimize conflict that will impede implementation, the essential prelim inary to considering points of entry at the implementation stage is serious consideration of ’ ’ points of exit” at the planning stage: the use of an incentives assessment to avoid commitment has been detected. However, funding agencies may not welcome the frank appraisal of political commitment which this implies. Aid agencies, in particular, accepters of gradualist change, have not cared to analyze too closely the degree of gradualism they are prepared to accept, the degree of compromise they are prepared to make in pursuit of their official goals. Realistic assessment of those projects which promise to be implementable if incentives can be favorably influenced involves facing this question of compromise squarely. (Heaver, 1982, p. 39) Literature on implementation, evaluation, PHC, and public administra tion in developing countries acknowledges that a dilemma for donor | ' agencies is how to be more adept at playing politics, particularly as j ‘ doing so directs more resources toward rural populations. Personal experiences (in Chad, Mauritania, and Cameroon, working with U. S. AID and observing United Nations and World Bank programs) have demonstrated that donor agencies are unwilling to confront this dilemma. The worsening situation of the rural poor— plus disgust , with ineffective aid (in donor countries as well as in the developing | world)— may, however, force the issue of allocating more resources | toward rural populations. ! i Heaver (1982) emphasized an incentives approach to management | I which would complement structural and training approaches. The ! addition of an incentives approach would add a new dimension to 1 ' I I project design. It is crucial to assess motivation and incentives | within implementing agencies and to pay systematic attention to the 1 169 kinds of incentive patterns that will render the attitude of indif ference to poor beneficiaries no longer politically rational (Heaver, 1982, p. iv). The key assumption (which is borne out by the litera ture search done for this dissertation) is that bureaucrats are rational. Their apathy is not inherent, but the existing incentive systems officially make it in their rational self-interest to be apathetic in the pursuit of development goals. The term "memagement” is used in this dissertation (and in Heaver * s [1982] analysis) to cover the entire organization of imple mentation. Management is viewed not only, or even mainly, as an activity of senior-level people; emphasis is placed on the important management role of district-level and field staff in the context of health-care— a factor that the literature on bureaucracy has tended to ignore. There are two broad approaches to influencing incentives, con cerned respectively with affecting intrinsic and extrinsic motivation. Intrinsic motivation— in the language of this paper the internal priorities that shape attitudes to work and career— is determined largely by financial rewards and the personnel management system. The fact that these are often outside the scope of agency influence does not diminish their importance. As Esman and Montgomery (1980:208) point out, field staff should be rewarded for the extra effort needed to work with the hardest-to- reach, most marginal farmers. Changing financial and benefits systems and the creation of national awards for success in proverty-reducing projects and programs could do much to make the work of agricultural extension or bare-foot doctoring as presti gious and rewarding as careers in research or official medicine, and thus have tremendous effects on performance in the field. At the project level, where direct influence over base salaries is seldom possible, funding agencies may nevertheless be able to influence material incentives in a number of different ways. First, it may be possible to increase cash incentives other than salaries— for example, field and travel allowances. Second, non cash material incentives such as better transport or housing may also be more open to negotiation, and equally effective moti vators. Third, promotion, training, and the payment of annual 170 bonuses cem be linked much more directly to performance. Bottrail (1981:118), in a detailed case study of irrigation project manage ment, illustrates clearly the incentive effect of bonuses payable to field staff at the discretion of project management— an outcome facilitated by the unusual independence of the case project from the government budget, due to revenues collected from farmers in payment for project services. The second broad approach to influencing incentives is concerned with extrinsic motivation, and it is this which has been least systematically considered elsewhere. Extrinsic motivation involves influencing the remaining groups of incentives— official,! informal, and local pressures and incentives. The need is to find ways of reducing counter-productive informal political activity 1 without creating unnecessary conflict, and without overstepping the funding agency’s bounds of legitimacy. In the following , sections, it is suggested that two broad points of entry minimize ■ ! these dangers: approaches through planning and control, and ' approaches through feedback, monitoring and evaluation. (Heaver, 1982, pp. 39-40) The second approach mentioned above— the one concerning extrinsic motivation— is useful, but it will need to be complemented by direct , influence, i.e., through encouragement as well as pressure from national and international donor agencies. Other important dimensions influencing supervision which have been discussed by nonformal theorists, among others, are participa- i tion and decentralization. Once again, these are large topics; but i they need to be discussed briefly as they affect supervision. Par ticipation is a term that includes several meanings. The least j convincing is participation in the form of executing someone else’s i ideas and commands. A more substantial form of participation involves , I I ‘ defining the question or problem for which action might be taken. A j third form of participation is designing solutions to a problem once I ; it has been defined (McGinn & Galetar, 1984, p. 40). An importeint j point about decentralization and regionalization is that they can j i J 171 entail relocation of the authority of central bureaucracy. Decentral ization often means merely that local client groups or elites that benefit from the central system or accept its ideology are allowed to participate in the design and execution of proposals. Managers who ; have a quasi-military, "top-down” concept of individual authority will tend to decentralize tasks but not authority (Heaver, 1982, ' p. 22). If the planning approach is "top-down,” reinforced by the I ' widespread dictum that officials be responsible for all their sub ordinates, and within the context of an already malfunctioning system, attempts to impose control from above, even in the context of decentralization , may increase incentives for intermediate level staff to disguise responsibility; if successful, this can effectively uncouple lower levels from higher level direction. In turn, clients will find it necessary to purchase bureaucratic assistance and cor ruption will flourish. (Moris, 1984, p. 14) . \ Faults will be blamed on incumbents as individuals, further accelerat ing the rotation of staff between positions and the reorganization of agencies as structures. It can be predicted that agencies subjected | to such process-related distortions will become almost impervious to I I organizational learning and to feedback from below. Genuine decentralization and participatory management along | the lines proposed by outside analyses stand little chemce of effec- j tive implementation (Heaver, 1982, pp. 18-22; Rondinelli, Nellis, & I I Cheema, 1983, pp. 52-54). And yet because the aberrations are funda- * mentally questions of process (such as what leaders see as the causes ! j of poor performance; when the finance ministry releases funds; and ^ 172 how agencies delegate responsibility, evaluate staff, and plan), they escape documentation under the usual cross-sectional, questionnaire approach to institutional research (Moris, 1984, pp. 58-59). Three implications in regard to supervision considerations and planning follow: 1. Planners or researchers must enjoy access to inside information (an emic perspective), which means that research and planning must be done within the situation and that all the i political, cultural, eind economic forces of the supervision ' setting must be taken into account. To be relevant to supervision, planning must be collaborative. 2. Attention must be given to means of documenting and analyzing information on institutional process aberrations. 3. Vertical linkages between levels (e.g., directives, incentives, feedback) must be looked at, but in the context of the demands of implementation. Indonesian Reform Effort In Indonesia there is a reform effort in nonformal education : at the community level called the PENMAS Nonformal Education Project. (The acronym PEjNMAS comes from pendidikan masyarakat« which means , community organization.) It is significant in being one of the very i few nonformal-education projects with the potential to reach large numbers of people (Evans, 1981a). The goal of the PENMAS project is i to facilitate learning activities for more than 800,000 people spread ! out over the six most populated provinces of Indonesia. This large- scale reform effort faces the real problem of developing 173' organizational and administrative procedures that meet dual needs: providing resources on a large enough scale and at the same time allowing enough autonomy and flexibility at the community or imple mentation level to respond effectively to diverse local conditions. The same dilemma faces many large-scale development projects, such as , national PHC programs using village-level health-care workers at the | community level. Because of the inherent conflict between the lines of authority and accountability in large administrations and the philosophy of local initiative and control of the learning process j (Evans, 1981a), it is useful to consider the project and its strate gies that are designed to address this conflict. A nonformal directorate within one of the five major sub- j divisions of Indonesia’s Ministry of Education and Culture, PENMAS has over 6,000 employees working in all of Indonesia’s 27 provinces. , It is one of the largest organizations of its kind in the world. j ! The PENMAS Nonformal Education Project, however, is concentrated in the six most populated provinces. There are three partners in the project— PENMAS, the World Bank, and the University of Massachu setts. The training provided by PENMAS is intended to create practical skills and knowledge that will supplement incomes, improve family health, create community infrastructure, foster awareness of government services, and create a receptivity to development efforts. Major learning activities centered around vocational skills, particularly those at the level of cottage and home industry; family life education; community education; and basic education in literacy and numeracy. (Evans, 1981a, p. 237) Although PENMAS has been in existence for over 30 years, its overcentralization and lack of flexibility in responding to local 174' learning needs caused its effectiveness to decline. Because the Indonesian government realized the potential of PENMAS and was faced with a population of 17 million out-of-school youths and (an esti mated) 23 million illiterate adults, the reform effort was launched. The government has given PEÎMAS the following objectives: ' 1. Promotion of basic attitudes and skills for development; ' 2. Education for social leadership; I 3. Cultivation of good reading habits; 4. Education for women; and ' I • I 5. Education and mobilization of youth for community welfare. The key administrative reform for PENMAS is substantial decen tralization to facilitate four types of learning processes— self- ! ; guided study, apprenticeship, courses, and learning groups— which are : used in four learning programs: Packet A (a basic education program), vocational skill training, family life education, and learning funds. , The project focuses on reform while training more than 40 top national and provincial PENMAS officials, who went overseas to receive training , for two to six months. These officials were freed from normal family | I j and work duties and were able to focus on program improvement; thus 1 their training turned out to be crucial. I The model of having short-term and degree training at the same I I institution that was supplying specialist services proved to be a | ' very successful one. Since the specialists were working as an integral part of project activities, they were able to provide | very accurate advice to the University on how to structure short-term training. This training, therefore, was geared not to J 1 the general theory of nonformal education, training, evaluation, and so forth, but to the way in which these areas were being j j applied to real project activities. This helped to give the j short-term training a valuable focus that allowed it to meet its 175 goal of giving the participants a well-rounded basic knowledge of their subject while providing them with concrete possibilities for application when they returned [to Indonesia]. (Center for International Education, 1982, p. 14) The reform also entails massive in-country training every six months for officials and for the penalik (the lowest-ranking field officer working at the local level). There will be a period of in-service training. The penalik works with volunteer facilitators i for village-level groups and hires part-time instructors to teach ' courses in specific vocational skills. The project also will help ! foster nationalism by providing instruction in the national language. In keeping with this aspect of the project. Packet A consists of 100 booklets: 20 are a set of literacy primers in Bahasa, the Indonesian national language, and the remaining 80 form a village encyclopedia , about village health, agriculture, Indonesian culture and history, i and other topical subjects. In the PENMAS structure, the penalik is a crucial component at the subdistrict level. With one penalik serving 20,000 or more learners, there must be a strong commitment at the local level for nonformal education. The penalik’s training and motivation level are critical if the reform effort is to succeed. Penaliks are in a position somewhat parallel to that of TBAs in that they will be , forced to articulate community needs to the district and provincial levels. Penaliks and TBAs are also the least trained, lowest paid, and lowest ranking members of their respective organizations. (If TBA supervisors are considered to be on the lowest rung in the MOH structure, the parallels are stronger between penaliks and TBA I I , 176 supervisors.) Penaliks are generally literate, government-paid officials, however, and will need to reorient themselves away from seeking ’ ’top-down” direction. Generally, TBAs will not be paid by the formal health-care system. A higher proportion of them will be ; illiterate. They will already have more orientation to the community rather than the government system. Nonetheless, parallels do exist. , Both TBAs and penaliks must be able to reconcile local needs with : needs of the larger organization. j A penalik has many diverse skills that must be developed: ; understanding the PENMAS curriculum, adapting the curriculum to local community needs, promoting allocation of scarce resources to non^ formal education, training tutors and pamongs (local facilitators), and developing volunteer support. The background of some penaliks is similar: they are young, well educated, and motivated by previous BUTSI volunteer experience (BUTSI is a volunteer development corps made up of recent college graduates). Other penaliks are older, more : traditional educators. The nonformal education project has put a lot of effort into training penaliks of differing backgrounds. Following are three examples of ways in which a penalik ‘ arranged a learning program. A. The penalik helped a village older woman begin an ’ ’arisan.” An arisan is a traditional village system of saving where each person puts in money or rice each week. Through a drawing, one of the women wins that week’s savings and is j able to make small purchases using local credit. The group I was formed around this activity and the penalik encouraged j them to perform some learning activity. They chose literacy and cottage industry. B. In another village a group of boys needed soccer uniforms and the penalik helped them learn how to make soap and sell it. 177 The group was successful and moved on to other learning activities. C. A penalik arranged for literacy training for a group of unemployed* out of school youths. With help from local leadership, learning groups (based on level of ability) were formed. (Center for International Education, 1982, pp. 35-36) Without local leaders and support, the penalik is doomed to failure. The penalik must develop a local volunteer network and provide significant local resources when needed. An example of a ; volunteer agency that has worked closely with PENMAS in implementing ■ the Packet A program is Pharma Wanita, a national, nongovernmental I volunteer organization composed of wives of Indonesian civil servants. Pharma Wanita has 2.8 million manbers, including wives of cabinet ministers, governors, district and subdistrict officials, and heads ' of villages. This organization has been able to mobilize the local resources in various areas of the country. Because the nonformal education program has a set of materials, an instructional design, and the backing of the government. Pharma Wanita accepted the responsibility of implementing the program in one district. This district now has one of the most successful literacy programs in the country (Center for International Education, 1982, p. 42). The PENMAS project staff has realized that women need special attention, i and the project has been strongly supported with material and human resources from Pharma Wanita. In scope and size, PENMAS is large ; enough for a genuine attempt to reach the poorest of the poor. It is also attempting to implement a client-centered, learner-centered philosophy of nonformal education that encourages local-level par- ; ticipation, decision-making, and control over project activities. 178 These two attempts are related, and PENMAS has built the infrastruc ture in the six provinces to support and forward these attempts. One half of the PENMAS funds were obtained via a loan from the World Bank. The project also provides a source of direct and flexible funding for village or community-level learning activities by establishing local learning funds. Nearly $1 million of the j project budget is being used to supply matching funds to local ! i learning groups. This fund will supply $240 (a nice sum for rural j Indonesia, where the national per capita income was determined by the , World Bank in 1976 to be $240) to a learning group that will match the funds from its own resources or from community funds. Priority will be given to groups that promote productive skills, generate ^ income, help other local development projects, and involve the least ! educated and poorest members of the community (Center for International Education, 1982, p. 10). ' I The PENMAS project is instrumental in pointing out challenges to a PHC orientation if a country opts for a national TBA program. Postadoption considerations— particularly executing follow-up, resup- ‘ ply, and evaluation— are an enormous undertaking for health-care ' systems, and commitment is the key for success of national TBA program efforts. The Indonesian case illustrates that donor funds I are needed to match local inputs. The realization that donor inputs are crucial has to be addressed in designing TBA programs, because : funds for continuing PHC efforts are rarely self-sustaining if the ; host country has to cover these recurrent expenditures. ! 179 CHAPTER VI CASE STUDIES The intent of this chapter is to test the three problem areas in relation to traditional birth attendant (TBA) programs in five j countries. By way of review, the three problem areas are preimplemen- j tation, training and postadoption. In preimplementation, the problem may be stated as: Certain preconditions must be met before under taking actual training and implementation. The preconditions to be . tested are: 1. Efforts to understand the community perspective and to make the community aware of TBA training 2. Efforts to understand the TBA’s perspective 3. Acceptance and support by government/ministry of health (MOH) for using the services of trained TBAs a. Allocation of supplies b. Allocation of transport 4. Development and pretesting of the training curriculum, which will have to incorporate principles of nonformal education I I 5. Careful selection of trainers (knowledgeable in local ' language and practices) 6. Careful selection of TBAs to be trained, with agreement of both TBAs and villagers " ' 180 , In training, the problem may be stated as: Principles of nonformal education are essential to high-quality training of TBAs; the approach must be flexible. The elements of this problem to be tested are: 1. Location of the training 2. Duration of the training 1 3. Remuneration for the TBAs I 4. Content of the training I 5. Methodology of the training 6. Evaluation I ' a. Of the training b. Of the TBAs* learning In postadoption, the problem may be stated as: Suitable supervision, follow-up, in-service training and resupply, and evaluation for trained TBAs are essential; without this support, the training of the I TBAs will be largely wasted, and eventually the TBAs will return to their traditional ways of practice. The elements of this problem to ! be tested are: ■ 1 1. Supervision a. By whom ! b. Frequency c. Method | I d. Training of the supervisors j e. Content task orientation ; f. Remuneration and knowledge of the supervisors 1 ■ I ; I ___________ _ _ _ _ _ J 181 2. Resupply, in-service training, and patient referrals a. Resupply and in-service training b. Patient referrals by TBAs to health centers or hospitals 3. Evaluation of effect a. Establishment and attainment of objectives and targets ! b. Linkage with MOH (new information being fed back to help in the health-care planning process) ' Although all three problem areas are important, postadoption is the J most crucial for successful TBA programs. With the three problem areas in mind, TBA programs in five coun tries (Bangladesh, Brazil, Peru, Ghana, and Pakistan) will be reviewed. The review will include relevant statistics, crucial cul tural beliefs and practices, and managerial aspects of formal health care systems as they relate to implementation of TBA programs. Background Background information on the five countries is given to highlight : important aspects of their TBA programs and to show the variation between and within countries with respect to program considerations. ' Analysis of data in terms of the three problem areas will follow. In ! many of the background discussions, "emic” (insiders’) perspectives were needed to get at the real program issues. Personal communication | . with Dr. and Mrs. Galba Araujo, Dr. Judith Davidson, Ms. Judith j j Greenberg, and Dr. Peter Lamptey proved invaluable. The background information also gives a crucial extra dimension of meaning to the I ■ data. ! 182 Bangladesh Bangladesh is one of the poorest countries in the world and Is designated a "least developed country" by the United Nations. Pop ulation pressure on the agricultural resource base has caused extreme problems. The following information (Grant, 1985; U. S. AID, n. d.) gives some idea of the poverty in Bangladesh. 1. The infant mortality rate is around 135 per 1,000 live births; I 2. The child mortality rate is around 20 per 1,000 children aged ' 3. Life expectancy at birth is about 50 years; 4. Less than a quarter of the people can read; 5. Malnutrition is widespread and hunger is common; and 6. Electricity is available in only about 7 percent of : Bangladesh’s 85,000 villages. With over 96 million people at the start of 1984, Bangladesh is one of the most densely populated countries in the world (about , 1,700 persons per square mile). According to the United States Agency for International Development (U. S. AID) and the World Bank, , the key to helping Bangladesh, with its estimated 2.7 percent annual growth rate, is in reducing fertility through a quantum increase in I birth-control practices. Demographers estimate that for Bangladesh’s 1 ' population to stabilize, the prevalence of contraceptive use will have to increase from the current estimated 16 percent to 61 percent (U. S. AID, n. d. , p. 16). Debilitating and endemic health problems— diarrhea, res- Ipiratory diseases, and malnutrition— add to the suffering of the .J 183 population. The bulk of the population does not know about the ' importance of safe drinking water and proper sanitation. Only about 22 percent of the population over 15 years of age has basic literacy competence. This is the context within which TBA training is being undertaken. ' Based on encouraging results (safe deliveries performed by TBAs) from several pilot projects, a one-year National Training ^ Project, costing $1,311,328, was started in June 1979. The project, I ' which was supported by the United Nations Children’s Fund (UNICEF), planned to train 21,000 TBAs in safe delivery, mother and child care, i and family planning. Each woman received a TBA kit and had training two days per week for three months. After the training period, the TBAs made eight reporting visits to the Union Health and Family Welfare Centers, for which they were compensated (per visit). By June 1980, about 1,250 family welfare visitors (paramedics trained for 18 months in primary health care [PHC] with special emphasis on mother and child care and on family planning) had been trained, and they in turn trained 18,000 TBAs (Claquin, Claquin, Rahman, Razzaque, Shaikh, Chowdhury, & Kanawati, 1982). ! The total Bangladesh TBA program was divided into Phase I (1979-1980), Phase II (1980-1982), and Phase III (1982-1985). Phase i I was completed on schedule, but because of implementation difficul- j ties. Phase II was extended to 1985 and Phase III was postponed ' indefinitely. At the beginning, the program was administered by the deputy director (training) of the Population Control Family Planning I Directorate. The demands of the program for continuing supervision 184 and monitoring, however, required a full-time project director, who was originally appointed in 1981 at the deputy director level, under the director (bio-medical research). The salary of the project director, as well as other project costs, was paid by UNICEF (Greenberg, 1984). The flow of funds was described by Greenberg (1984). Funds were allocated by the project director to the deputy directors I (family planning) at the district level. These deputy directors I dispersed funds to the thana (district— now called upazila) family- planning officers where Union Health and Family Welfare Centers were functioning as facilities for TBA training. Stipends were paid to TBA trainees by the upazila family-planning officer or by a family welfare visitor. The family welfare visitors are responsible for ' training TBAs and are posted at the Union Health and Family Welfare Centers. During Phase I a one-day workshop was held to familiarize deputy directors (family planning) with administration and management of the program. Claquin et al. (1982, p. 5), in their attempt to evaluate the program, stated that only one thana out of the 25 ran domly picked had a continuing TBA training program. This finding had i a definite impact on the methodology of evaluation; obviously, no trainee could be found to provide the needed basis for comparison. The lesson was that the evaluation team could not rely on official statistics. Claquin et al. (1982) also observed that despite the active training of family welfare visitors (supervisors of TBAs), no ; TBA-oriented training or retraining activities took place between 185 November 15, 1980, and February 15, 1981, the dates of the evaluation team’s visits. In a review of an orientation course in TBA training, Greenberg (1984) observed that the training of supervisors— the family welfare visitors (many of whom are men)— did not increase their low level of practical experience in obstetrics. The educational methods used in TBA training were not adequately considered, and problems were found with per diem payments to the TBAs during their training. Furthermore, the TBA training course was borrowed from Pakistan and was too theoretical. Blanchet (Note 20) discussed limitations on the TBA’s role as a healer (these elements were not dealt with in the TBA curriculum); several cultural beliefs strongly influence beneficial, harmful, and neutral birth practices. If birth complications arise, the TBA traditionally calls on the kobira (traditional local village healer). Any attempt to expand TBA roles (such as the World Bank’s proposed family-planning project) must be approached with knowledge of such practices. Certain traditional beliefs in Bangladesh influence maternal i I and child health. Blanchet (Note 20, pp. 24-25) described how the ; belief in bhuts (evil spirits that reside outside the bari, or house hold) circumscribes the social conduct of pregnant women. Bhuts are believed to be especially attracted to pregnant women, lactating women, and infants in utero. As Greenberg (1984) pointed out, I restrictions on women’s movement are commonly associated with purdah, demanded by many men who are Muslims. But religious (Islamic) reasons 186 for purdah are external to most village women; protection from bhuts and other hostile cosmic forces is a major reason why village women do not seek prenatal care, delivery, or services outside their home. Restrictions on food, especially protein-rich foods, exacerbate the women’s already precarious nutritional status (Greenberg, 1984, p. 12). Greenberg (1984) cited several important food restrictions that vary from place to place. Generally, women are told not to eat too I much, so that they will not have a large infant and a difficult , delivery. There are specific food restrictions, largely on protein-rich and vitamin-rich foods such as fish, mutton, duck, eggs, ' shrimp, squash, and sweet pumpkin. In Matlab thana . . . eight types of fish are prohibited; one kind because it will cause epilepsy, and another because it is very large and must be cut into small pieces to eat, thereby angering the spirit of the fish who may harm the child. Some women avoid ginger because of its shape, fearing it would cause a fetus to be crippled, while pumpkin is said to cause coughs, and pineapple induces abortions. (Greenberg, 1984, p. 12) Blanchet, an anthropologist who has done extensive research on TBAs and birth practices in Bangladesh, thinks that "pollution" is the single most important cultural belief that determines birth ! rituals and practices. Women are in a state of "pollution" when they are menstruating or pregnant, and especially when they have just given birth (the postpartum period). The blood of childbirth or menstrua tion is considered especially potent in terms of pollution because it has been stagnant and stored for one or nine months, respectively (as opposed to the fresh blood that flows from a cut). Blanchet (Note ! ■21) makes a distinction between the beliefs of Hindu and Muslim dais 1 I about pollution. The Hindu dai takes on pollution herself, thereby , 187 allowing the mother and household to remain unpolluted. She is always paid for her services and does everything a TBA traditionally does, including cutting the umbilical cord— which is considered the most polluting act. The Muslim dai in Bangladesh does not have a religious function and does not normally receive a fee, but she does receive a sari. Most rural Muslim women have shame if they pollute others, so they are reluctant to seek professional help at birth. Often, rela- ; tives assist at childbirth, as they are less easily polluted than the 1 others. (If relatives do assist at childbirth, they may or may not call themselves dais, depending on their socioeconomic status: richer women do consider themselves dais, as it does not jeopardize their status and they refuse payment, whereas poorer women relatives who help at birth have never been called dais.) One beneficial cultural practice is that because women are considered to be in a state of pollution for 40 days after child birth, they are not allowed to work; thus, they can rest and recuperate during the postpartum period. At this time, a woman is often fed "dry" foods because she is considered to be in a "wet" state. Unfortunately, certain foods may be restricted— such as fish, meat, and eggs, which would provide a new mother with the nutrients : she needs to recover from the blood loss of delivery (Greenberg, 1984, p. 13). Several other traditional birth practices are relevant to con- : sidérations for TBA programs in Bangladesh: 1 - A common practice by birth attendants is to insert a hand into the vagina to determine the progress of labor and often to accelerate it by rupturing the bag of waters and to manually 188 dilate the cervix. Numerous accounts of deliveries mention that birth attendants do not wash their hands before inserting them and use unsterile lubricants (coconut and mustard oil). These practices can be a major cause of uterine infection, and of cervical lacerations. Birth attendants’ belief in cleansing is related to the polluting aspect of the birth process rather than to asepsis. Birth attendants commonly insert their left hand - associated with dirt and feces - into the vagina. It is only after the birth that the dai washes her hands to cleanse herself from pollution. Because of this belief in pollution, women do not deliver on their beds but on the floor, usually on an old chatti mat. A woman continues doing her chores until the onset of labor. At that time, the person assisting her, often a relative, neighbor, or traditional birth attendant (dai), is called. One very positive aspect of traditional birth care is the commitment of the birth attendant. Once she is called, she will not leave the laboring woman until the birth is complete. Another positive aspect is the extensive touching and strok ing the dai employs. This laying on of hands is comforting, relaxing, and helps develop a relationship of caring. Once the bag of waters is broken, the dai encourages the woman to bear down without delay. Expulsion of the baby is often hastened by the dai applying abdominal pressure with her hands or knee. These practices can lead to cervical tears, birth trauma, premature separation of the placenta, and possible death. When the infant’s head emerges, the dai places an old cloth on it, and pulls to deliver the baby. If done too vigorously, this can cause trauma to the infant as well as vaginal and perineal tears. Often the perineum is guarded with an unwashed hand, or foot, which could cause infection. Delivery of the placenta is frequently hastened by pulling on the still uncut umbilical cord, and applying abdominal pres sure. If this is not effective, additional measures are taken including pricking the cord with a sharp object and inducing the woman to vomit by dipping her hair in kerosene and pushing it down her throat. Great importance is attached to the quick delivery of the placenta as the baby is not con sidered fully alive until this happens. Pulling on the umbil ical cord and abdominal pressure can lead to extremely serious complications such as hemorrhage, infection, uterine prolapse, and inversion of the uterus. Failure to resuscitate immedi ately an infant that is not breathing will lead to death within minutes. Allowing the infant to remain wet and uncovered (15-30 minutes in normal cases and hours in the case of retained placenta) can lead to hypothermia and death. The traditional utensils for cutting the umbilical cord are a bamboo sliver or a razor blade. Cord ties are often made from threads of old cloth. In most cases, none of these things are sterile. As the baby is lying on the floor, the L . . . 189 cord, before being cut, is exposed to dirt. After being cut, the umbilical stump is often smeared with cow dung, ghee, or ash. These are prime conditions for exposing the infant to tetanus. - After delivery of the placenta, the mother is told to stand to evacuate her stagnant blood. "When a woman loses a lot of blood after birth there is no panic for this is bad blood that should not remain inside if the woman is to recover her health. It is only when the mother becomes very anemic that help will be sought to stop the blood flow" (Blanchet, 1981). Most women in Bangladesh do not have sufficient hemoglobin levels. Bleeding in a normal delivery, not to mention excessive hemorrhage, can leave her in a debilitated condition. - Breast feeding is usually delayed by three days because the colostrum is considered dirty. Infants are fed honey mixed with diluted cow’s milk, warm water, or mustard oil. This practice deprives the child of nourishment, as well as the benefits of colostrum, which contains substances that help the child ward off infections. Delaying lactation may have an effect on the child’s ability to suckle properly, and the establishment of successful lactation. Lactation in Bangladesh is prolonged and is widely practiced which is positive. Often a woman breast feeds until she becomes pregnant again. Supplementary feeding is introduced on an average by nine months, although nutritionists and health providers recommend it begin earlier, at 4-6 months. (Greenberg, 1984) Even with good management, which was definitely not the case according to evaluations by Claquin et al. (1982) and Greenberg (1984), TBA programs must take local conditions into account. The above descriptions poignantly illustrate the importance of cultural beliefs and practices. Brazil The Ceara’ program is really a comprehensive rural health program. It includes a community education project with teachers, a project in which traditional health-care practitioners (resideiras) are taught to use oral rehydration therapy for children with diarrhea, and a nutrition project. J 190 The overall objective for SAMEAC (maternity hospital in Fortaleza) is to integrate TBAs, traditional healers, teachers and other key community resource people, with university professors, students, and Ministry of Health personnel, so that together they can pro vide the community with essential health care services, health education and an expedient referral system (to existing medical systems at higher technical levels), thus improving the level of the health of individuals, families, and communities. (Develop ment Associates, Inc., 1983, p. 3) The TBA program in Brazil is located in the state of Ceara* along the coast in the northeast, which is the poorest and least j developed part of the country. The state is divided into large and I medium-sized estates (latifundia). About 45 percent of the 4,400,000 ' inhabitants of Ceara* live in urban areas. The capital city, r Fortaleza, has a population of 1.5 million and is under intense pres sure from rural immigration (Araujo and Neto, Note 22). The infant i mortality rate is 142 per 1,000, which is higher than in any other I part of Brazil (Araujo, Note 12). According to a survey covering fourteen municipalities (out of 141 in Ceara*) taken by the secretary of health and the SEPS foundation in 1971, 29.8% of the 1,368 pre-school children and 36% of the babies under six months had normal weight. Approxi mately 70% of the children had malnutrition; 37.6% belonged to the first degree, 21.9% to the second degree, and 10.7% to the third degree of malnutrition (according to the Gomez classifica tion which is based on the percent of average weight for age obtained by a child). First degree malnutrition is the mildest form, third degree malnutrition is the most severe form. At present these percentages have increased considerably, due to the drought which has affected the state for the last five years. ; The principal causes of death in those under the age of five years ■ are: 1. Water and food borne diseases (36.4%) 2. Birth complications and congenital defects (16.4%) 3. Acute respiratory disease (11.2%) 4. Diseases which could be prevented by immunization (7.5%). (Araujo, 1984, p. 1) 191 ! The average income is very low and unevenly distributed, and unemploy ment is high, worse than the national average. The main industry is agriculture, and the main crops are cotton, beans, rice, corn, manioc, coffee, and sugar cane, all chiefly for local consumption. In 1974 Dr. Jose Galba Araujo, professor of gynecology and . obstetrics at the Federal University of Ceara* and director of the j Maternity School "Assis Chateaubriand" (MEAC) of the same university, I started a program to improve health care for pregnant women in poor i and urban areas along the coast and slightly into the interior. This j I ' region was chosen because virtually no health-care services were ' available, it is close to MEAC in Fortaleza, and there is easy rail- ’ road access to the whole area. The program’s overall goal was to save the many desperate women with small pelvises and pregnancy tox emias who were arriving at MEAC. This goal was to be achieved by a , TBA program with referral to the hospital. In the first phase (1974-1980) of Dr. Araujo’s program, I I specific goals were training of TBAs, establishing a good network of support for them, and introducing them to the concepts of high-risk, j antenatal, and postnatal care. During the second phase (1980-1984), ; the program expanded geographically and broadened its scope to include , ! care of newborn babies. The TBAs* functions were expanded to include cancer prevention and family planning, and the TBAs worked closely j with the agentes pediatricos (nursing auxiliaries) who monitor growth, | I help with immunizations, and teach oral rehydration therapy for diar- I i rheal diseases. i L_. I J 192 The program is financed largely by the Kellogg Foundation and with state and federal funds. Although the community clinics are self-financing, the program provides some vaccines and equipment. The supervision and training by the MEAC team is paid for with these external funds (Walt, 1984, p. 19). A report on a program evaluation conducted June 14-July 1, 1983, mentioned that Dr. Araujo had refused ; several donor and government offers of assistance because he wanted to proceed incrementally and because he wanted the project in Ceara* I to have maximum local inputs. There has been criticism of the program by some international agencies in the past about the slow process of training and the equally slow coverage of the area with family planning services. However, there seems to be valid reasons for not going too far too fast. For example, the role of the TBA in the community has for centuries been that of delivering babies and no more. When ' prenatal care and newborn care were introduced by the PAPS (primary health care unit) personnel into the TBA*s job, it was not only necessary for the TBA herself to accept the role and believe in the concept of need for prenatal care, but the com munity also had to change its image of her in the newly acquired role. (Development Associates, 1983, p. 3) The program’s first step was initiated in Pacatuba, some 30 kilometers from Fortaleza, with full community involvement. In Pacatuba a small health clinic had been abandoned, and a group of local leaders, including TBAs, rebuilt the clinic and set up a committee to run and administer it. During this time, groups of six i : TBAs were taken to MEAC to be trained for two months. A year later, the clinic opened with four beds, a delivery room, an admissions office, a kitchen, and a laundry. Four of the best TBAs (who could ! read and write) were chosen to give 24—hour coverage in the clinic. I Other TBAs continued to function as domiciliary TBAs (TBAs working in 193 ' their homes), but they were not paid unless the pregnant woman wished to pay in cash or in kind. Salaries for the clinic TBAs were gathered from several sources, usually state or federal funds, raised by the community. At Pacatuba the mayor’s office employs a driver for the clinic’s ambulance, which is used to transport women who need to be ^ transferred to MEAC. Some of the lessons learned from this first attempt are described below. 1. After the first training course at MEAC, training was short ened and done locally. Going to the hospital had raised TBAs’ expectations - they wanted to do more interventions, to learn to use forceps, to do episiotomies, and so on. Dr. Araujo was disturbed by medical interventions, as evidenced by the | numbers of induced labors, caesarian sections, and episiotomies. The discomfort and tensions of women deliv ering in hospitals was in marked contrast to the calmness of natural deliveries. 2. The kit given to the TBA was also pared down to a minimum after the first experience. For example, Dr. Araujo tells the story of a trained TBA who had received rubber gloves to ; use at delivery. When asked what she did with them she > replied, ”Ah, Doctor, I have kept the gloves ready at the i door so when I leave I can grab them and go." Today, TBAs do ' not use gloves, just soap and water. They do receive a uni form, however, and a few simple items like scissors and cotton. . . . 3. Since those days TBAs are trained locally, as far as possible. They receive four or five lessons over a five week period and spend 24 hours in a clinic observing the work of trained TBAs. , This way the rhythm of their own lives is interrupted as ' little as possible. From the beginning MEAC has underwritten j a system of support. The TBA must accompany a woman in dif- i ficulty to the hospital, and the community must guarantee i transportation for them both. Once in a hospital (usually MEAC) the TBA must be courteously received and well treated. This is a very important aspect of the work of this TBA training program. (Walt, 1984, p. 18) i i In 1981 TBAs had been trained in about eight maternity units (by this time the program was providing prenatal, delivery, and post partum care, as well as family-planning services). Additionally, some 50 domiciliary TBAs were trained with emphasis on prenatal care. - 1 ' 194 delivery, postpartum care, and some newborn care, and with limited information on sex education, family-planning education, and oral- contraceptive distribution. The training was done primarily by Dr. Sylvia Bonfin and Dr. Galba Araujo. Most of the TBAs were illiterate, , so the trainers used and perfected the techniques of extensive repeti tion and demonstration. In addition to training TBAs, MEAC trained ; physicians to work in the maternity units and helped with patients i : ; referred to them by TBAs. j By 1983 the Sociedade de Assistência a* Maternidade Escola Assis Chateaubriand (SAMEAC) had expanded its activities. The SAMEAC | program had 15 new maternity units in 1982 alone, for a total of 30, plus over 1,000 trained domiciliary TBAs. In 1983 Development Associates gave ; I Î financial support for training of TBAs to traditional healers (resideiras), primary school teachers, nursing assistants (most of this group work as auxiliaries to TBAs), and community leaders. PAPS (a system which deals with PHC in the SAMEAC | system) thus expects to gain entrance into the entire community with a "network" of non-formally trained workers. (Development : Associates, 1983, p. 2) ' 1 At the same time, local medical personnel (both physicians and health- ' , department employees) were sensitized and trained to collaborate with I I the nonformally trained health-care workers. By 1984 there were 38 j clinics with beds and 24-hour coverage by TBAs. These clinics may do I up to 70 deliveries per month, although some do as few as six. \ t I 1 Lagoa Redonda clinic, for example, is a 45 minute bus ride from j 1 Fortaleza, and is situated on a tarmac road. It has eight or ten , I beds and four experienced TBAs who cope with around 426 deliveries during each six month period. In the same period, trained TBAs did 85 domiciliary deliveries. The health post is in the same ‘ building, as there is a good relationship between the agentes : I pediatricos and the TBAs. (Walt, 1984, p. 17) : 195 Another service to the communities is offered by trained TBAs who build a delivery room onto their home. Both domiciliary TBAs and TBAs working in clinics refer difficult cases to hospitals (usually MEAC, except in isolated rural areas, where the TBAs refer such cases I to the nearest support hospital). Transport is provided by the local ■ ' authorities; if a private taxi is used, the program pays for it. I The Ceara* program is an extension of the Health Sciences j Department of the Federal University of Ceara’ (Araujo, Note 9). All , I > j university students in nursing, medicine, social work, and dentistry ; : receive part of their practical training in the program. Specially trained medical personnel train the TBAs and provide some secondary- level health care (for example, immunization and prenatal care for ! high-risk patients) at the rural units. They and their students also I give classes in personal hygiene, sanitation, breast-feeding, and child-spacing. These physician/professors or nurse/professors must , (1) learn to present information in very simple language, (2) be | : I imbued with the philosophy of the program, (3) accept the TBAs as colleagues, and (4) gain the confidence and respect of the poor I population they are treating. Otherwise, no one will seek their I services. ; I Southern Brazil is highly developed, industrialized, and ' wealthy; TBAs like those in northeastern Brazil do not exist there. , Although there is no need for a national program, the MOH has begun a | I I regional program on maternal/infant mortality and PHC (based on the | , Ceara* experience) in which the university program has become a i I ^ I training center for professional health-care personnel (Araujo, 196 Note 9). Concern has been voiced in program evaluations that extension of program coverage would entail difficulties resulting from the problems of support and supervision and the remoteness of regions to be covered. Many persons working in the fields of TBA training and evaluation (e.g., see Walt, 1984) regard this program as a good example in that it addresses considerations needed to make TBA I training and supervision effective, to involve communities, and to improve health care at low cost, even though the program may not be j replicable in all its aspects. I The cultural aspects of this program contrast with those of the Bangladesh program. Brazil has a common language understood by everyone. Training, in-service training, supervision, and communi cation are easier in Brazil than in Bangladesh. The predominant religion of northeastern Brazil is Roman Catholicism. Although it has been adapted to and imbued with strong spiritual beliefs and practices, there do not seem to be any very strong rituals or beliefs attached to actual delivery as there are in India or Bangladesh, for instance. This has made training easier and has also allowed the introduction of concepts of ante and post natal care. (Walt, 1984, p. 29) ! Breast-feeding is uncommon among Brazilian women; in the best clinic, ; just over half the women returning 30 days after delivery were still breast-feeding exclusively. I A few of the unique aspects of the Brazilian program are listed below. 1. From the start, there was emphasis on respecting local customs, and TBAs had to be courteously received. The profes- I sional staff was imbued with this attitude, and Dr. Araujo 197 (Note 9) mentioned that it took two years to train a doctor and one year to train a nurse to have this attitude and to interact with and train TBAs effectively. Maintenance of complete respect for local customs and traditions was given priority. 2. The principal responsibilities of the TBA are to provide care at normal delivery and to recognize and refer high-risk patients to the nearest obstetrical unit or support hospital. Selected TBAs can be trained for activities other than care at I normal deliveries, such as prenatal care, immunizations, and ' family planning. This training, however, is confined to the few TBAs who are capable of benefiting from it, and it must be done gradually. 3. The method used to find and begin to train TBAs has involved substantial community input, in contrast to the situation in so many other programs, which recognize such community involvement as a principle but never implement it. The practice of the community—involvonent principle is described below. We go to the community. We talk. We explain. We demonstrate. We try to overcome the disbelief generated by broken promises made during election campaigns. Is it election time? This is commonly heard. We explain. From there we convene a meeting j with local leaders: priests, teachers, TBAs, politicians, all who can participate. And so our work starts. (Araujo & Neto, Note 22) I Peru I ! Peru is often described as a stratified, ethnically hetero geneous state— a plural society. Pre-Hispanic Peru was rigidly ! stratified into three classes: workers, nobles, and rulers. 198 The ethiîo historical basis for the rigid stratification of Peru vian hispanic society is based on evidence gathered during the early stages of contact by Spanish observers and historians who recorded the life patterns and belief systems of the Inca at the time of conquest. In this myth the creator deity, Puchacamac, sent four stars, two males and two females. . . . From one pair of stars descended the rulers and the two classes of noblemen, and from the other descended the common people. (Davidson, 1984, p. 25) Spanish conquest entailed both spritual and economic exploitation, including the forced relocation of peasants from productive lands ' into towns and less productive highland terrain. The inheritance of 1 , dependency created by the Spanish has left its mark on the poor, ' resulting in the experiences of highland indigenous communities and ' "the creation of a suspicious, alienated, and subjected majority, culturally and linguistically separated from the dominant society" (Isabell, 1978, p. 21). There is extreme variation in geography, climate, and ethnic groups. All of these circumstances have led to the importance of kinship and reciprocity in interpersonal relations. They also influence the ways in which TBAs and their patients (who are generally from a lower socioeconomic class) view sanitarios (low est level NK)H personnel, usually in charge of a rural health post, or posta medica. and in charge of supervising TBAs) and other formal health-care personnel who represent higher classes. Peru has a population of 18.2 million (1982), and an infant I mortality rate of 100 per 1,000 nationwide (1982), but in 40 percent I of its population the infant mortality rate is 150 or more per 1,000 (Grant, 1985, p. 131). This difference in infant mortality rates reflects the stratification between classes. L 199 Begun in 1945, TBA training In Peru was given sporadically in the 16 health regions from 1960 to 1979 and then was expanded to a national scale (Favin et al., 1984, p. 38). From her research on TBA programs in Peru, Davidson (1984) stated that the initiation of this national program was a result of the 1978 Alma-Ata conference’s ! exhortations to adopt PHC strategies emphasizing the potential roles I for TBAs. ; The training program has affected the lives of many people. 'About 2,000 people have taken TBA training since its official incep- ' tion as a national program in 1979. That number is increased by ; including the people trained before 1979 in departments such as Ica, Puno, and Cuzco, where training began in the 1940s. Unfortunately, because training was given by such a diverse group of foreign and Peruvian private voluntary organizations, records of the participants are either unavailable or incomplete. The number of patients treated by trained TBAs must range into hundreds of thousands. Nothing is known, however, about the personal effects of TBA training on the lives of the people it touched (Davidson, 1984, p. 130). i The main objectives of the present TBA training program I(Davidson, 1984) are to teach the TBAs to I 1. Use hygienic practices when attending a birth; 2. Recognize danger signals; 3. Know when to call for help from local health-center staff; i 4. Provide systematic prenatal and postnatal maternal care and I infant care; 5. Know how to use a limited set of medications; 200 6. Keep a record of births attended; and 7. Provide basic health-care services to the cononunity. The program’s original written objectives were to modify the methods used by TBAs and to increase the scope of their work. These goals have remained the same; the TBA program has not changed since 1979, although the recent evaluations by Davidson (1984) will probably ; cause some changes in the TBA and PHC programs. One of the salient shortcomings of the TBA program is its I lack of attention to community input. As has already been mentioned, * the American Public Health Association’s study of 52 PHC projects assisted by U. S. AID noted that failed attempts were often those with little community input, and successful attempts were those in i which the TBAs* new duties did not conflict with community under standing of the TBAs* role (Parlato & Favin, 1982). Davidson (1984, p. 197) mentioned two erroneous assumptions by MOH personnel. First, regional health-care planners did not consider TBAs capable of pro- .viding more than maternity care; they generally were unaware of other countries’ programs in which TBAs are already providing a wide range of health-care services to women beyond the maternity cycle and to ' men. Second, the health-care planners assumed that the TBAs were predominantly illiterate. Over 50 percent of the TBAs are literate, and furthermore, illiteracy does not signify the lack of ability to deliver PHC services, including some technologically complex medical procedures. The key to program success is to maintain incentives for TBAs and patients and to maintain participation and commitment in the community (Davidson, 1984, p. 214). 201 The inflexibility of the national health-care policy arose not because national health-care planners were uninterested in community reaction to health programs; rather, it arose because there was no information about the actual effects of the programs (Davidson, 1984, p. 184). Unlike the policies of other Latin American TBA i training programs, which have aimed to decrease traditional practices associated with midwifery (Kelly, 1955, as cited by Davidson, 1984), Peruvian MOH policy has been to refrain from interfering with harm- ' I i I less, or neutral, practices of TBAs. The literature indicates that ; I the central health-care bureaucracy was trying to run a national TBA program (and also a PHC program) with the same level of knowledge and input applicable to a local health center. "Centralized systems fail ' by virtue of inundation when they attempt to manage local systems with the same level of information variety as local systems can handle" (McGinn & Galetar, 1984, p. 49). The result was not catastrophic, but it did contribute to a lessening of the performance of the TBA ^ I program. Specific examples (Davidson, 1983, 1984) follow. * 1. In a training program in Aguaruna, the tribal council representing the entire Aguaruna communities did not want to send any more of their women to the city for training courses. In Aguaruna culture, women do not leave their community except to i collect food, to gather clay or raw materials for handicrafts, or to accompany their husbands or relatives on buying expeditions to , I I I local villages. There is a great fear that women will be mis- I ! ! i treated by mestizos (here, mestizo is synonymous with a Spanish- | speaking person) or that they will be tempted into having illicit J 202 sexual relations. Aguaruna women who initially received training could not put it to use because they were shunned by their local communities. 2. Many indigenous (lowland) training courses failed because— according to surveys— only 2 percent of the population | was aware that a trained TBA was present in the community. (Mani [1980, p. 399] noted that a big factor in program failure in India was that the "program was supply-based rather than demand-based. " ) Traditional birth attendants were trained without creating a feel ing of need for their improved services. Community education should be a major part of the TBA training program, 3. Many local training programs achieved no positive results because of a lack of knowledge about different cultures and a lack , of ability or willingness to be flexible, | 4. Some areas, such as pueblo jevenes or areas with high rates of population movement, lack stable community leadership. In such an area, the sanitario could consult with the local bus- | ! inessmen and/or teachers who participate in the selection of i I candidates for TBA training. It is incorrect to assume that all rural communities in Peru have TBAs, although most do. I 5. In some lowland indigenous communities, there is no | culturally recognized role for TBAs. There the sanitario could | i organize mothers* clubs, preferably like Nigerian mothers* I clubs. (The Nigerian mothers* clubs were based on the idea that a cash incentive was necessary to ensure continued support of a community-based health program. Structuring the clubs as rotating 203 credit unions accomplishes two aims: it provides small amounts of cash to rural women who have no other way of accumulating cash, and it provides a forum for regular contact between rural women— who are potential patients— and trained TBAs. In this way TBAs can give informal instruction on personal hygiene* household sani tation, and family planning.) Training courses for TBAs could be attended by local women who are not birth attendants but have had I previous experience with patient care during the maternity cycle. I These courses should be taught by native speakers and be held ' within the local communities. Shortcomings in supervision and evaluation were also prominent in the TBA program. Among these were: 1. Insufficient input from the local community on TBA super vision; I 2. Lack of feedback on TBA performance to the local community; and 3. No attempt to evaluate the performance of trained TBAs within the community. (Evaluation based on the number of TBAs trained and their performance on a written exam at the end of their training course.) (Favin et al., 1984, p. 38) In Peru, supervision is called "promoter activities." Davidson (1983) gave reasons for promoter desertions. Over 5,000 I health-care promoters have been trained. From 10 to 50 percent of I : these trained personnel (according to region) have already abandoned their work as promoters. Their reasons for doing so, and the per centage of promoters giving each reason, were: ! 1) Lack of salary (32.9%); I 2) Conflict with own work/lack of time (26.1%); 3) Marriage/travel/family problems (12.6%); 4) Too much work (10.1%); 5) Did not like work (6.7%); and 204 6) Being institutionally ignored and lacking necessary materials (6.7%), (Davidson, 1983) Other explanations given for promoter desertion, drawn from case studies, include: * Migration • Illness , # Loss of motivation due to unfulfilled aspirations for a paid position in the MOH • Misunderstandings of the nature of the work • Career plans that require additional education # 111 will expressed by community members who assume that pro moters will attend to their health needs 24 hours a day * Disappointment at not receiving promised equipment (Davidson, , 1983, p. 9) Methods of supervision currently in use are infrequent visits by MOH personnel to the homes of trained TBAs, inspection of the TBAs* records, and infrequent meetings at the local health center. The ' : important point is the lack of any attempt at routine evaluation of 1 the actual performance of TBAs. Reliance on the results of written examinations is unwise because the TBA may learn the course content j ! ! without becoming committed to using the practices he or she was ' taught: For example, a trained TBA can answer that it is important to use j a clean instrument to cut the umbilical cord. The trained TBA i may respond that she will transfer a patient with a complicated j health problem rather than treat the patient herself. Privately, : the TBA may think the importance of cleanliness is exaggerated by professionals and that the appropriate response to complications is prayer rather than referral. The immediate supervisor should , periodically accompany the TBA on patient visits to see how she has treated patients. During these visits patients should be I questioned about their health problems and the TBA*s treatment of | them. Patients should also be asked if the TBAs provided them I with preventive care, counselling on nutrition, infant feeding, and family planning. (Davidson, 1984, p. 200) Task-oriented supervision is the method recommended; the literature | I I on TBA supervision in Peru, however, mentions neither incentives nor j 205 commitment to such a method. (Incentives are discussed in regard to TBA and patient populations; a potentially useful recommendation was to give discounts on drugs to TBAs who are working well.) The sanitarios should evaluate the TBA's actual performance, and that I should be the basis for TBA license renewal (Davidson, 1984). j According to Favin et al. (1984, p. 38) and Davidson (1984), encouraging findings about TBA programs included the following: 1. The quality of care provided by trained vs. untrained TBAs differed markedly, particularly in rural areas. This finding is important because levels of TBA activity and patient use of TBAs are higher in areas (invariably rural) that are not Westernized. , 2. Trained TBAs retained the information taught. i 3. Trained TBAs expressed very few complaints about the i program, despite the fears of evaluators about MOH advocacy of TBA program efforts. Trained TBAs were pleased with their new skills and thought they were in a better position to provide service to their communities. I 4. Patients treated by trained TBAs were satisfied with the j I services they received, and they preferred to use trained rather I than untrained TBAs. 5. Hygienic and nutritional counseling provided by trained TBAs was being used by their patients. j Davidson (1984) described an effective approach to evaluating j a TBA program: examination of the results of PHC programs from the i ------ top as well as from the bottom of the health-care pyramid. Consider ing both patient and TBA perspectives lends support to the assumption , that recommendations for program evaluation are likely to be met with genuine participant compliance. Quantitative and qualitative results were obtained, and emic perspectives were added to knowledge about local conditions. Two methods were used to guarantee the internal validity of the survey results. The examination of pre-tests given to the TBAs before they began the course and interviews with regional health personnel revealed that outside influences were evenly distributed within the survey population. While there were a few cases of , TBAs who had obtained instruction in Western obstetrical tech niques outside their training course, these informants were eliminated from the survey population. The difference in the ^ performance of the TBAs was due to the instructions they received during the training session and were not caused by outside influ- ; ences. The method used to assure the external validity of the survey results was to represent the cultural and ecological diversity of the Peruvian population within the survey sample. ' This strategy proved to be highly successful, since the response of TBAs and their patients is highly influenced by cultural var iables. (Davidson, 1984, p. 192) Ghana Ghana is a typical West African country in that its rural inhabitants, particularly mothers and children, have high rates of I mortality and morbidity. Before the Danfa project there were no i i documented formal TBA training programs, although isolated attempts j by individual physicians in certain areas to work with TBAs were i i noted. 1 The Danfa project was a nine-year (1970-1979) training, j I research, and demonstration project undertaken jointly by the Depart- ; ment of Community Health, University of Ghana Medical School, and the School of Public Health, University of California at Los Angeles. It 20?! was supported both by the government of Ghana and by a U. S. AID grant. Danfa project goals were to: 1. Investigate the state of the rural Ghanian community, its physical, social, cultural, health and economic characteris tics and related factors that determine participation in health programs. 2. Research the most useful and efficient way of utilizing avail able manpower for health and family planning service in rural i areas. 3. Undertake cost analysis to provide data useful to economic and health planners to help maximize the return on investments made in health-related programs. 4. Provide information derived from the project to the Ghana government on an ongoing basis so that the national policy of extending health services and family planning to the rural sector can be best implemented. 5. Help create an institutional infrastructure both capable and willing to adopt, implement, and carry forward the acceptable i findings of the project. (Neumann, Sai, & Dodu, 1974, p. 40) In many ways the Danfa project was a trend setter in terms of ; PHC strategies and represented the "state of the art” at that time. ! It recognized that for PHC to be a national effort, less money must j be allocated to expensive urban-based hospitals and medical schools I and more money allocated to PHC efforts. In 1985 many Third World | countries have at least a beginning PHC infrastructure, but five | years ago very few such countries had any. Three years ago, for i i example, the World Health Organization (WHO) started a PHC health- ! resource group designed to help developing countries map out PHC strategies. In terms of thoroughness, analysis of communities and TBAs, and evaluation, the Danfa project is a good example of a demonstration TBA project. Although it had a large research and I evaluation component, most of the project’s research and evaluation studies, while admittedly costly, are not intended to be continued beyond the present project, although elements of the research phase such as 208; records system procedures, functional analysis research, cost analysis, and other administrative techniques are reproducible and already are beginning to be adopted by staff at the Univer sity of Ghana and the Ghana Ministry of Health. (Neumann et al., 1974, p. 40) The Danfa project operated in a rural district located 15 to 50 kilometers north of Accra, the capital of Ghana, and it included some 500 square kilometers. The district was divided into four areas 1 (three study areas and one comparison area). Data for the period of i ! the project (1974-1978) covered about 60,000 people and 307 villages. ' The fertility rate for women aged 15-44 years was one of the highest in the world; the birth rate was 48 per 1,000 people. The district had a natural population increase of 3.2 percent annually, also very high. In a 1972 registration of TBAs, 237 TBAs and 26 assistants registered in the three study areas. The ratio of TBAs to population was 1:137. During the 1972 registration, it was found that the average age of the TBA was 62, half the practitioners were men, 6 percent of the TBAs were literate, TBAs averaged seven deliveries per - year, and only 6 percent of them delivered 20 or more babies per year (Neumann, Note 11). In a survey conducted to collect information on TBAs* knowledge and practices and their attitudes toward family plan ning, most were found to support family planning and 68 percent were interested in improving their midwifery skills. About 50 percent of the TBAs performed over 90 percent of the deliveries (Ofosu-Amaah & Neumann, 1979). Training of TBAs began in 1973. Surveys indicated that elderly, largely illiterate TBAs would probably have difficulty in assimilating modern midwifery practices. The objectives of the 209 I training program were limited; upon completion of training the TBAs were expected to 1. Recognize pregnancy early and advise and monitor women during the prenatal period; 2. Recognize and refer high-risk women or those with serious j complications of pregnancy or delivery; j 3. Perform all safe deliveries; ' I 4. Practice proper cutting and care of the umbilical cord; and 5. Aid in promoting improved maternal and child health practices ; and family planning in their villages. A community education program, which was an important element of the Danfa project, was conducted before, during, and after the TBA training programs. The progressive involvement of the community in the provision of services can be divided into three phases. Phase I included four programs: # Health Education # Nutrition # Sanitation e Training of Traditional Birth Attendants (Ofosu-Amaah & Neumann, 1979) Phase II was marked by the introduction of two new community-based programs: (1) malaria prophylaxis and (2) immunization (Ofosu-Amaah & Neumann, 1979). Phase III was a village health worker program. j It is obvious that TBA training was part of a comprehensive j approach to health-service delivery. Base-line demographic analysis | I showed that the district had a high chiId-mortality rate. Children 1 were endangered by a number of infectious diseases, especially malaria, pertussis, diarrhea, respiratory disease, worms, skin L ' - 210 ‘ infections, and poliomyelitis. At any one time (the village health surveys were done for pre-school-age and school-age children), 41 percent of the children under 11 harbored malaria parasites in the blood (in 1972 malaria was the most frequent diagnosis at the Danfa health center). The frequency of poliomyelitis has been as high, if ! not higher, than that experienced in temperate countries during epi- ' demie periods of the twentieth century, Guinea worm was a serious problem, causing a work loss in untreated adults that averaged more j than five weeks (usually coinciding with peak agricultural activi- | i I ties). The proximity of a patient’s home to a health-service facility was found to be the system factor having the greatest effect on health. As a result, several more satellite health centers were ; built. I After a survey of TBA knowledge, attitudes, and practices (KAP), TBA training was begun. Follow-up supervision and evaluation, done by a public health nurse and health-center midwife, were crucial to maintaining the interest of the TBAs. One of the salient features of the Danfa project was its focus on health-care delivery systems, supervision, and institutional issues I as they relate to PHC. An organizational chart showing the position I of the TBA within the district (Figure 9) was set up. Within this | I I framework the importance of evaluation was stressed. Evaluation was ; j I i done on two levels: | I : ! 1. TBA Outputs: this helps the program supervisors ascertain if i the TBAs are meeting program expectations in terms of amount ' I and quality of work. | 2. Program Impact: this assessment shows whether maternal and infant mortality/morbidity are decreasing as a result of the 211 Midwife Sanitarian Health Center Other Workers District Hospital Community Health Nurse Village Health Worker Health Center Superintendent Traditional Birth Attendant Public Health Nurse i Figure 9. Organizational Chart; The Health Worker and the Village Community. (Adapted from Danfa Project [1977a, p. 47].) 212 program’s effort - of interest to program directors, super visor, and TBAs. (Danfa Project, 1977a, p. 35) Another important approach to evaluation was consideration of the purposes of evaluation. Evaluation was necessary not only to ensure that an adequate level of health care was available, but also to provide inputs for designing appropriate training programs for j health-care personnel at all levels, i In the Danfa project, MOH personnel and medical-school person-I I nel set up a research and development unit within the MOH. As an aid < I for collaborative efforts between the MOH and medical schools, the following institutional development recommendations were made: 1. Demonstration Projects of service, teaching and research, such as the Danfa project, should be developed in other countries, and ideally, in each major region of those countries. 2. Teaching staff with the necessary skills to participate effec-| tively in these projects must be developed. ! The skills of particular importance are: - Teaching methods, especially competency-based training , - Epidemiology - Biostatistics ! - Maternal and child health ' - Family Planning I - Control of infectious diseases - Research methods and survey design - Health economics - Management ; - Demography | - Electronic data processing ; 3. Medical schools or other training institutions should aid in ! preparing training manuals. 4. Medical schools should receive adequate support to participatei in these activities. (Ofosu-Amaah & Neumann, 1979, p. CR-10) ' ; The project also thoroughly dealt with health-care delivery ■ within the health-care system. Among the recommendations I ' i 1 (Ofosu-Amaah & Neumann, 1979) were: --------- 213: 1. Rural health-care centers must be made more efficient in the testing of village-based PHC efforts involving village health workers. 2. Health education must be an integral component of every program because it is essential to achieving significant ' I participation of the population in health programs. ■ 3. Health programs must address major causes of mortality/ morbidity. I 4. Training must be made more effective and more relevant to I ' program objectives (e.g., appropriate in-service training exer cises must be developed to help MOH personnel adapt to new job roles). These issues are topical concerns of many PHC efforts today. The difficulty of evaluating a TBA by watching her actually deliver a baby is obvious— the timing. Evaluation during Danfa pro ject training was based on: , I 1. Attendance of at least 75% of the training sessions. 2. Participation in class discussion. 3. Recall during informal oral quizzes of material presented at previous sessions. | 4. Performance of their role in a simulation exercise presented ; to assembled villages. (Neumann et al., 1983, p. 8) i The TBAs who met these criteria were eligible to receive certificates ! of completion. Graduation consisted of presenting, to the entire com- ' ' munity, a skit based on what the TBAs had learned. Supervisory staff determined that the best frequency for ' supervisory visits was every two months. During a visit, supervisors ^ would check the TBA record book (if the TBA was illiterate, a literate | I ! I „ _ I 214 relative or neighbor would help) and examine the TBA kits. Infor mation was collected on the number of prenatal visits, deliveries, postpartum visits, complications seen and referred, and family- planning acceptors recruited. If possible, mothers who had delivered were visited. The mother’s room or delivery room was checked for | cleanliness, the newborn baby was examined, and the care of the umbil ical cord was asked about. Every 6—10 months a refresher course was given, and super visors rated TBA performance according to the following criteria: t 1. Attentiveness in training ! 2. Retention of knowledge 3. Coopérât iveness 4. Referral of problem emergencies . 5. Provision of prenatal care ; 6. Provision of postnatal care 7. Correct use of cord packs (simple pack of gentian violet, clean razor blades, and cord ties). (Neumann et al., 1983) Evaluation (according to the five objectives of TBA training mentioned previously) was thorough because there was a basis for comparison. Supply, in-service training, and referral also were ■ adequate. The TBA’s link to the formal health-care system was the 1 public health nurse. j Under the supervision of the District Medical Officer, the public health nurse coordinates the TBA training program. Her respon sibilities include the supervision of TBA identification and j registration efforts, the selection of appropriate training program for the TBA, and the design of course content, continuing education, and follow-up. (Ofosu-Amaah & Neumann, 1979, p. 6-27) i I Dr. Peter Lamptey, a Ghanian physician who was the medical officer in charge of principal research in Area One from 1976 to 1978, recently j stated (Lamptey, Note 23) that the oil crisis caused a general decline ! in the economy, which led to a breakdown in services, including health: 215! services and hence the Danfa project. This belief is corroborated by a report that profiled Ghana: The major problems affecting children are directly related to the continuing economic problems in the country. Since the late 1970*s the situation has become so serious that there is a severe shortage of drugs and medical equipment; educational facilities suffer from grossly inadequate supplies of teaching and learning i materials; water supply systems all over the country are fre- I quently inactive for lack of power, spare parts, and repair and maintenance; the supply and distribution of local and imported * staple foods is unsatisfactory, owing partly to the poor condition of roads and scarcity of transport vehicles; and the emigration ' of skilled manpower continues to adversely affect the quality of , I social services. (UNESCO, 1982, p. 3) i ! I I Because this crisis took place in the late 1970s, it coincided with Ghana’s inheriting from the ending Danfa project a fleet of 15 vehicles that were crucial for transport of supervisory staff and ; supplies, and for continuing contact with TBAs. Since 1979, the money for supervision has had to come from either the or the I Ghana Medical School in Accra. The sudden transfer from Ü. S. AID support to solely Ghanian support was too drastic. Medical supplies were not even in the health centers, so the TBAs certainly did not j have them. In 1982 Dr. Lamptey went back to Ghana to do follow-up work with illiterate TBAs by using pictorial designs, and he had great j I difficulty finding the TBAs. Dr. Lamptey believed that the Danfa , project had functioned well during its life, but it was too costly i I 1 (even without the oil crisis) in terms of supervision, supplies, and j : constant retraining. He thought that younger village girls (ages I 16-17) should be trained as TBAs because they would learn and j ■ : remember faster than the over-60 TBAs, they would accept modern 216 i health-care practices more readily, and they would require less supervision. The younger TBAs could learn from the older TBAs via apprenticeship (retaining community and cultural links) and from the health center, where obstetrical care is modern. A possible dis- i advantage to training younger TBAs would be high dropout rates j resulting either from higher expectations within the formal health care system or from migration to urban areas; the latter is a common j I problem with the younger village health workers of either sex. I In 1985 a PHC team effort with a TBA component is being , I implemented, but it is having problems because of the lack of funds for the PHC infrastructure, the high rate at which physicians are : leaving the country, and the political events of recent years. | Pakistan : Pakistan had an estimated 1982 population of 93 million (Grant, 1985) and a very high population growth rate (3%), which causes the population to double every 23 years (the population has increased fivefold in this century). Because such a small percentage j of its women have more than just a few years of schooling, Pakistan | has one of the lowest female literacy rates in the world. The national infant mortality rate and the fertility rate are 1 I ! extremely high. Almost all women have some form of anemia during ! their childbearing years; hemoglobin deficiency is more prevalent in pregnant and lactating women than in any other segment of the popula- ; tion. While the Pakistani diet is not iron deficient per se, it is I j based on unfortified wheat, and for complex biochemical reasons, much of the natural iron in the wheat remains unabsorbed. Malaria, 217 tuberculosis, malnutrition, and infectious diseases leading to diar rhea all contribute to the poor health of women and children in rural areas (Pakistan Health Sector Report, 1983). Short intervals between births contribute to high infant mortality. Breast-feeding can reduce fertility, however, and Koranic injunctions state that boys should be I breast fed for 24 months, girls for 22 months. In 1975 a Pakistani j , fertility survey (part of the World Fertility Survey) found that 99 percent of rural children were being breast fed for long periods. I For the past five years, with vigor unique in the recent | \ history of Islam in southern Asia, the Pakistani government has been , trying to Islamize Pakistan. The major Islamic restrictions on women result from "purdah" (the formal separation of women from the world , of men: from the Urdu word for "curtain"). Although not all women observe purdah, most do accept its basic tenets: that women not mix with men and not wear sexually provocative clothes. Thus, if the government’s Islamization effort succeeds, the power and participa tion of women in law, education, and politics will be eroded; to date, modern views have been vying with the conservative, fundamen- ' talist Islamic views. A further complication is that Islamic Jurisprudence differs radically among countries such as Saudi Arabia, Iran, and Pakistan. Rural women must toil in the fields, forage for firewood, carry water, transplant rice, and cut alfalfa for their animals, so their ; condition will not be improved until their burdens are lessened by ^ new technology and by literacy. | L___________ J 218 I In certain areeis of Pakistan, purdah as it relates to rural women is an issue (Hunte & Sultana, 1984). The concept of purdah restricts women’s movements, very much akin to the restriction described by Greenberg (1984) in Bangladesh. Such restrictions were ^ also mentioned by Abbasi (1985, p. 9), who observed that in Pakistan’s Sind province, the TEA ("dad") profession is changing. The majority of dais, who had received training from elderly members of the family, cannot train the younger generation because of the general feeling , that the work is not clean. How all of these issues relate to the ! concept of "pollution" mentioned by Blanche! (Notes 20 and 21) is not ‘ I known, but it seems likely that the same kinds of problems may be operating in the two Muslim countries, Pakistan and Bangladesh. Almost all births in the rural areas of Pakistan are con ducted by dais (Inventory of dai training, 1983, p. 7). Similarly, j in Sind province (Abbasi, 1984, p. i), 91 percent of births are i ' j attended by dais (the discrepancy may lie in the tradition that ■ I : relatives who are dais do not take money for their services and do j not want to be known as dais). In the provinces of Baluchistan, ' Punjab, and Northwest Frontier, TBAs are trained in the public-health ! schools and the maternal and child health (MCH) centers or the MCH ' components of basic health units of the provinces. This training is j the responsibility of the health departments/directorates. It has j ^ been going on since 1947 (Inventory of dai training, 1983, p. 8). ' ■ According to Abbasi (1984, p. 3), training of TBAs in Sind I : j province has been receiving attention from UNICEF and WHO since the ; 1 ! early 1950s. The older TEA programs asked dais for a commitment of r 2191 one year (the dais generally were unwilling to give up family and work obligations) and failed to provide residential accommodations at rural health centers; much of the curriculum included elements not relevant to the dais* work. The lady health visitors were incapable of teaching adult illiterate women. Similar problems arose with TBA | training programs in Baluchistan (Hunte & Sultana, 1984). ! In the Inventory of dai training (1983, pp. 22-32) is a i i summary of TBA programs throughout the country including an analysis 1 of eight programs, both governmental and nongovernmental. Almost all I I I the programs were deficient in supervision, training of trainers, methodology, and collection of base-line data. Only two curricula for TBA programs were available for appraisal; one was from the , : i Pakistan Nursing Council, and the other from the Family Planning Association (both government programs). In these curricula, neither | the role nor the functions of TBAs were outlined, and learning objec- ' tives were not defined in writing. The teaching methods and materials j were generally those used for literate students of midwifery. j Supervision by lady health visitors was a problem, because j i they often practice midwifery and thus were in direct competition with TBAs (this practice is against government rules and regulations, > but it persists in urban areas). In all four provinces, responsible I officials mentioned that lady health visitors in MCH centers were ; using TBAs as their personal maids. The problem is prevalent only I where the lady health visitor lives near or at the MCH center and , when training courses were for one year’s duration. " 2 2 0 ^ Until and unless the LHVs [lady health visitors] succeed in ensuring the TBAs that the case remains hers and that she remains entitled to whatever payment the family makes to the TBA, neither the TBA nor the family will be keen to invite the LHV. As long as the LHV keeps trespassing on the territory of the TBA and keeps stepping on the TBA’s fees, the TBA will not accept either her for supervision or for maintaining her contact with the MCH Center. In which case, both the follow-up and/or supervision will , never become a reality. This has been proved to some extent by ■ those programs where the LHV is not a resident in the area, e.g., the Program of the Directorate of Social Welfare of Sind and some of the MCH/Family Planning Centers of the FPAP [Family Planning Association of Pakistan]. In these programs the TBA maintains a monthly or fortnightly contact with the LHV. (Inventory of dai training, 1983, p. 19) , Personnel involved in planning and implementation were, however, aware ' of the need to revise curricula, improve teaching methods, train lady health visitors to train TBAs, and provide effective linkage between MCH centers and TBAs. , In 1982 the government of Pakistan introduced an Accelerated Health Program (AHP) with three components: (1) training of TBAs; (2) I production, packaging, and distribution of oral rehydration salts for 1 treatment of diarrhea; and (3) expansion of the immunization pro gram. These are the major health elements of the Sixth Five-Year ' Plan (1983-1988). The AHP, a centralized national plan with local j implementation, has already trained 12,000 TBAs (Grant, 1985). This | ' number is, however, far short of the AHP target. Further acceleration ; I or simplification is needed in the TBA training scheme. A pilot, I ' mobile team-taught course lasting two weeks has been introduced in ; Sind. 1 I High maternal and child mortality and a high incidence of neonatal tetanus led to the decision to include strengthening of TBA training as one of the AHP’s main components. Before 1982, TBA 221 : training was offered to any married woman over age 20, and the train ing course lasted one year. The AHP currently has an upgraded curriculum that takes only three months to complete, and trainees receive stipends and UNICEF TBA kits. According to the Pakistan health sector report (1983, p. 16), Pakistan’s responsibility for training TBAs remains with three agen- i cies: MOH Social Welfare, AHP, and Population Division. The training ' I ! activities of all three are under the MOH, but coordination and shar- ^ ing of information are problematical, as we shall see. The TBA training is planned by the central government and implemented by the four provinces: Punjab, Sind, Northwest Frontier, and Baluchistan. Largest of all provinces is Punjab, with 60 percent of the nation’s population; least populated is Baluchistan. Under the Sixth Five-Year Plan, the policy is to train only | TBAs as community health workers. There is the realization that , ; training 50,000 TBAs requires adequate preparation of the villages. ' Undoubtedly, medical technicians will provide the TBAs* major support, : but TBA training will also involve rural health centers, district offices, public-health schools, and hospitals. This involvement in I TBA training has important implications for additional functions for I I . these institutions, implications that have thus far not been covered ; in the literature. i Two components of the AHP are administered by Pakistan’s National Institute of Health (Grieser, Note 24). It is unusual for such responsibility to be given to this agency. The head of the agency (General Burney), however, is known as a very capable, i J 222 effective man; he is the reason the National Institute of Health was chosen to implement these two program elements. The institute facilities— in the palace of an old Moghul king— are very impressive. It is well known that to date, within the AHP, immunization has been the component given priority, although oral rehydration therapy is starting to get attention. The person in charge of the TBA program. Dr. Agha Banoo (a woman in the MOH in Islamabad), has a small cubbyhole of a room. She complained that she has no staff, no resources, and no funds, so she actually does nothing. Fortunately, provinces are training TBAs; Punjab is likely to train 8,000 TBAs in 1985 alone. The low priority given to TBA training within the AHP is not documented: in all the literature, the TBA training component is advocated. Pakistan’s health-care strategy also addresses the rural population through the "Basic Health System": A new health structure, called the Basic Health System, is being developed to provide a systematic link between village communities and hospitals of the modern health system, with services planned to be provided as follows: Villages Basic health units Rural health centers Tensil hospitals District hospitals Teaching hospitals 2 community health workers 1 doctor, 4-6 health auxiliaries 3 doctors, 8 auxil iaries, 10-20 beds 380,000 population Surgical, medical. 1,000 population 10,000 population 100,000 population laboratory, and x-ray facilities 1,160,000 population Main specialties Province All modern facilities Each RHC [rural health center] is designed to serve 4-10 BHUs [basic health units], and should be staffed with three doctors and eight MTs [medical technicians] with functions of providing primary health care for the immediate area; acting as a referral 223 I center from BHUs; providing planning, management and supervision of preventive and promotive programs in the area; and serving as a focal point for information and supplies. (Pakistan health sector report. 1983, p. 13) The basic health units, staffed with medical technicians and support personnel, are designed to provide health-care services for local i areas. At the federal and provincial levels, basic health-service I cells have been established. Although this system is beginning to be | I implemented countrywide, several problems have been identified (Pakistan health sector report, 1983, pp. 12-18): : 1. Front-line health-care workers have little incentive to gather accurate information, because they are remote from those who use it at higher levels in the system; thus information necessary for the effective planning, control, and evaluation of | their work is not gathered. j I 2. Under the basic health scheme, medical technicians are Î I required— to be competent to deal with most problems occurring in | I the basic health units and the rural health centers, and to refer complicated problems to more trained staff. Medical technicians work in relative isolation (they are an example of personnel in resource-constrained access bureaucracies in Pakistan) where | there is no physician. These medical technicians have heavy j responsibilities: performing diagnosis and treatment, supporting j and developing community health workers (TBAs), and linking 1 community-developed activities with established health programs. j There is recognition of the need to train more medical techni- j clans and upgrade their skills; 2,000 had been trained by 1983, 1 224 and the expected total of 1,359 to be trained over the next five years in 26 schools represents (at best) one tenth of the total need. 3. District health officers have the responsibility, on paper, for overall health matters in the district. In practice, however, they lack the authority needed to appoint, dispose of, and discipline staff; to manage budgets; and to plan and coordinate the activities of their district. Moreover, their remuneration and status compare unfavorably with those of district hospital superintendents, a serious disincentive for preventive vs. curative medical work. (The appendix of the Pakistan health sector report [1983] shows the national, provincial, and district structure of the Pakistani health system.) 4. Lady health visitors, who have received a comprehensive theoretical and practical training lasting up to 27 months, provide midwifery and MCH services in rural health centers and MOH centers. At present there are 10 public-health schools that train lady health visitors. Although female medical technicians will assume the functions of lady health visitors, no decision has yet been made on the future of the lady health visitors. They are required to complete at least 20 different forms or registers, including an outdated list of 109 standard disease categories that was adopted before 1900. 5. The problem of equipment and maintenance is ubiquitous. It is easier to purchase equipment or obtain it from donor agencies than to pay for its subsequent maintenance. This 225 ; problem has particular impacts on basic health units, rural health centers, and TBA supervision. Several surveys have shown that a large expenditure on sophisticated health-care equipment is not appropriate, because the equipment is too costly to maintain. 6. There is an oversupply of doctors in Pakistan. During 1 ; the past eight years, the number of medical schools has increased from 3 to 16. Now doctors find it difficult to find work over seas, and the quality of medical education is being severely I criticized. 7. Family welfare workers, trained by and working for Pak istan’s Population Welfare program (under the Population Division in the Ministry of Planning and Development), have been made equivalent to lady health visitors and female medical technicians by the Pakistan Nursing Council. They receive an 18-month course I of basic training in 12 regional training centers. ' The problem of coordinating TBA training looms large on the ! i horizon because different divisions within the government have respon-j , sibility for the training and because of the multiplicity of donor j ' 1 ' efforts involved. It is an important issue because TBA training is a I , prominent part of Pakistan’s PHC strategy. One very positive note is j j the recognition of management problems— recognition by a minority of | MOH officials and in studies done by outside donor agencies. These ; problems are much more thoroughly documented for Pakistan than for I I the other two countries with national TBA efforts (Peru and Bangla- I desh) that are considered in this study. i 226' The traditional beliefs of Baluchistan have been considered ' as they relate to TBA practices. Some of the most common of these beliefs are listed below. 1. Like other traditional medical systems in the world (see, for example, Cosminsky, 1983) health in Baluchistan regardless of ethnicity is perceived as a delicate balance within an indi- vidual’s system of the humoral qualities of hot and cold. Illness is thought to occur when one’s state of humoral equilibrium is upset by eating foods that are too "hot" or too "cold" in humoral quality. This classification is innate and is not connected with physical temperature. This is an ancient doctrine which has its historical roots in Greek medical texts such as the Hippocratic corpus. Indeed, some forms of traditional medicine in Baluchistan are even today , referred to as dewa unani or "Greek medicine." TBAs often prescribe herbal medicines based on this system or give ; nutritional advice accordingly. 2. Belief in .jinn (spirits) is common which is associated with religion. Mentioned in the Koran, some of them are good, some bad. Jinn are believed to cause illnesses, and infants ; and children are especially vulnerable to malevolent attacks ' of these creatures. The related concept of soya (shadow) is associated with illness. Prenatal bleeding, miscarriages, and stillbirths are also frequently attributed to .jinn. Thus, pregnant women along with those who are helping to give birth are prime targets for .jinn. Preventive and curative treatment for .jinn and soya most frequently takes the form of tawiz (amulets containing writings from the Koran) which are pre pared by a Mullah (Muslim priests). [These customs are favored in many parts of the Muslim world, including Chad, Mauritania, and Senegal in Africa.] 3. The belief in chesm-i-bad or "evil eye" serves as an etiolog ical factor. This belief, too, is common to many parts of the developing world; common to all is the fear, envy, or jealousy in the eye of the beholder. Precautionary measures include Koranic purchases, amulets, and herbal preparations. (Hunte & Sultana, 1984) The practices of dais in Pakistan (Grieser, Note 25) appear i to be similar to those of dais in Bangladesh. Of the three national j 1 I ! TBA programs, issues of support and supervision were more clearly i : I dealt with in the case of Pakistan, although these issues are far j ! I from resolved. < 227 Data Presentation and Analysis Each of the six factors in the preimplementation problem area, the six factors in the training problem area, and the three factors in the postadoption problem area was assessed via rating the , ' degree to which a series of key indicators was attained in each TBA j program studied. Key-indicator attainment was rated on a five-point ' I ; ordinal scale. i t I The criteria for the ratings are detailed in Chapter II. ; I ! I Tabulation of the data showed that the scores of pilot TBA programs J and demonstration projects (Ghana and Brazil) were much higher than the scores of national TBA programs, as would be expected. The scores of national TBA programs were clustered in two groups, with ! the programs of Bangladesh and Peru on the low end and the program of Pakistan on the high end. The main reason for this dichotomy is that ; Pakistan is beginning to confront real organizational and management , ; 1 issues, even though serious problems remain. In Pakistan, significant , pilot projects (e.g., those in Sind, Punjab, and Baluchistan prov- I inces) are being implemented, and evidence of coordination is | beginning to emerge. In all three national programs, however, ( 1 ! supervision, support, and follow-up were significant problems. | : Table 8 shows the key-indicator ratings for each TBA program | I I - studied. , Preimplementat ion In Bangladesh, efforts to understand community perspectives are limited. Greenberg (1984, p. 23) suggested the formation of mothers’ clubs, co-ops, or other women’s groups, and strongly 228 Table 8 Key Indicator Ratings Stage and factor Bangla desh Brazil Peru Ghana Paki stan Stage I. Preimplementation Factor A. Efforts to understand community perspective 1 5 2 5 3 j Factor B. The TBA perspective 1 5 1 5 2 j Factor C. Sovernment/HOH acceptance and support for using services of trained TBAs Allocation of supplies 1 5 2 5 I ( 2 I Allocation of transport 1 5 1 4 1 : Factor D. Development and pretesting of curriculum 1 5 1 5 2 , Factor E. Careful selection of trainers (local ' language and practices) 1 5 1 5 2 i Factor F. Careful selection of TBAs to be trained, with both TBAs and villagers agreeing 1 5 1 5 1 TOTAL (maximum = 35): 7 55 9 34 13 Stage II. Training Factor A. Location 2 5 2 5 2 Factor B. Duration 1 4 2 4 3 j Factor C. Remuneration 2 5 3 3 3 Factor D. Content 1 5 1 5 2 Factor E. Methodology 1 5 1 5 2 Factor F. Evaluation ! Of training 1 4 2 4 2 Of TBAs’ learning 1 5 1 5 3 TOTAL (maximum = 35): 9 33 13 31 18 ! Stage III. Postadoption Factor A. Supervision By whom 2 5 2 4 2 , Frequency 1 5 1 5 2 1 Method 1 5 1 4 2 Content 1 5 1 5 2 Training of supervisors 1 5 1 5 2 Remuneration of supervisors 1 5 1 3 2 Factor B. Resupply, in-service training, and patient referrals ! Resupply and in-service training 1 5 1 4 2 1 Patient referrals 1 5 1 5 2 1 Factor C Evaluation of effect ! Targets set/being met 1 Linkage to MOH in terms of information getting 1 back and aiding health-planning process 1 5 1 5 2 1 5 1 5 3 1 TOTAL (maximum = 50): 11 50 11 45 22 GRAND TOTAL (Stages I, II, and III): 27 118 33 110 52 229 advocated taking traditional beliefs into consideration in training programs. Greenberg’s (Note 26) comments would greatly aid the considerations of World Bank personnel who were involved in planning a large family-planning project with TBAs in Bangladesh. Knowing local perspectives is crucial in areas like Bangladesh because so many of the beliefs have profound effects on any PHC effort involving TBAs. No KAP studies were done on TBAs; government allocation of supplies was, however, evaluated in several studies and found to be ^ totally inadequate for TBA programs. The problem of transport, a major constraint in TBA programs in Bangladesh, translates into a lack of support for monitoring at both national and local levels. The training curriculum was not pretested; it was developed in Pakistan, adopted at the International Conference of Midwives at Comilla in 1977, and translated with UNICEF funds into high-level Bengali. Family welfare visitors (supervisors) have too heavy a workload to supervise TBAs and they lack the obstetric knowledge needed for proper training of TBAs (Claquin et al., 1982; Greenberg, 1984). With regard to selection of TBAs, Claquin et al. (1982) found that of 155 TBAs interviewed, 25 had no prior experience as TBAs, 3 . were unmarried, and 12 had no children. If TBAs were chosen by the 1 village’s head man, a lot of favoritism would be inevitable. There : are no official criteria for selecting TBAs to be trained. Offi- ; cially, one TBA is to be selected per village, but Greenberg (1984, 1 ! p. 18) believes that for the program to have a significant impact, 230! two or three TBAs per village should be chosen for training— those who have had the largest number of deliveries. The Ceara* program in Brazil was very strong in obtaining community perspective and incorporating it into training, super vision, and evaluation. From the start, the program included I substantial community input for finding and/or building a local 1 clinic, for paying the salaries of clinic TBAs, and for funding the . transport for referral of patients. Respect for local customs was a j guiding principle throughout the program. [ Many informal and formal surveys were taken to ascertain what the ! TBA knows or feels about certain subjects in order to decide how to teach new subjects. This often was done verbally by personal interview. For example, information about TBAs beliefs, knowledge and attitudes regarding sexuality, conception, contra ception, and abortion was gathered before the professional tried to teach the TBAs anything regarding family planning. It was I discovered that TBAs had a lot of incorrect information, but also 1 many correct ideas. One example is that most TBAs knew that conception is most likely to take place during mid-cycle and that a woman was usually most interested in sex during that time. They | have been successfully instructing clients who were interested in i avoiding pregnancy to abstain from sex during this time. Another I interesting example is that from year one, the methods employed j to control births in Ceara were abortive in nature. When SAMEAC ! introduced the subject of family planning to TBAs, they approached | the subject from the angle that family planning prevented abor- I tions. Only this way were they able to slowly convince TBAs that | , family planning was not synonymous with abortion. The TBAs are i ^ now dealing with the same resistance in their communities. With | ’ information such as this gleaned by surveys, the teacher can build I on knowledge the TBA already has. Grassroots workers or non- I formal agents are proud people, who want to be recognized for \ their accomplishments and abilities. SAMEAC believes that the teacher who understands and utilizes this principle will be bet ter accepted by his other students. The SAMEAC training program is built on frequent surveys or informal research both before and after training. This is one way they determine course content and evaluate whether content has been learned and is being util- ; ized. Surveying is utilized as a form of evaluation and to keep the staff "in touch" with those they teach. The community is also surveyed to see what they know about a particular health question (e.g., family planning), as well as 231 what services and education they want and need. The information gathered in this way also helps to guide SAMEAC in developing course content. (Development Associates, 1983, p. 6) The transportation needs of the Ceara’ program were squarely met. The transportation and communication network in the program area is good; although not all roads are paved, many are, and bus and telephone services are fairly widespread. Also, the referral network is good. First, trsuisport is guaranteed, second, the TBAs always accompany the women to the hospital where they are, in general, well received. Unusually for Brazil, there is good integration between the peripheral clinics and the referral or support hospitals, even though the clinics are administered locally and the hospitals are federal or state institutions. (Walt, 1984, p. 19) One advantage of the program is its geographic limitation. The curriculum was shortened and provided locally. The mate rial covered includes reproductive physiology, pregnancy, labor and delivery, care of the mother, care of the newborn baby, and recogni tion of the high-risk patient (Araujo, Note 12, p. 4). Trainers were MEAC staff, doctors, and nurses, but trained TBAs are now being trained as trainers. The TBAs selected must always be approved by the community, and selection of midwives is the responsibility of a local leader previously chosen to recruit midwives in the area. , Generally, this leader is one of the best-known and most respected midwives in the community (Araujo, 1984, p. 6). Î Davidson (1984, p. 202) highlighted the supply-and-demand problem in the TBA program in Peru by commenting that base-line data t on the communities are needed, and further, that only 2 percent of S the communities surveyed were aware that they had a trained TBA in I * their midst. No KAP studies on TBAs were done, and Davidson (1984) I J 232; r stated that this lack of consideration and knowledge of TBAs’ atti- ' tudes is a major shortcoming of the Peruvian national TBA program. Government support for TBAs in Peru is for expansion of the program, not for evaluation and improvement. Transportation is a real constraint, and lack of knowledge about TBAs and communities ! adds to the inflexibility of the program. The training curriculum was not pretested, but the TBAs said they benefited from the train ing. Evaluation showed an increase in both utilization of and I Î preference for trained TBAs vs. untrained TBAs. Sanitarios were t trainers, and Favin et al. (1984, p. 38) stated that no previously : trained TBAs were on the training staff. This situation should change as a result of suggestions (based on evaluations by Davidson, : ! 1983) to health planners. According to Favin et al. (1984, p. 38), 1 weaknesses in the trainee selection process included insufficient ! community input and lack of attention to creating incentives that would ensure the participation of qualified candidates. | Ghana’s Danfa program used a progressive or incremental j process for community involvement. Surveys and KAP studies were j conducted to find out about community perspectives, which were I 1 I incorporated into training. Similarly, TBAs completed KAP surveys, , and as a result, the curriculum was revised: pretesting revealed that the elderly illiterate TBAs would have a limited capacity for assimi- ; ; 1ating modern midwifery practices. j Formal health-care personnel (doctors, nurses) were involved 1 in and supported the Danfa program from the beginning. This input from the MOH ensured that supplies were forthcoming and that the 233. I training was geared toward perceived needs. Transportation was a ; problem, but with the vehicles provided, schedules were worked out. The selection of trainers was important, and one of the selec tion criteria weis ability to speak in the vernacular. As in the Brazilian program, formal health-care personnel (district public- ; I health nurses with experience in community work) were trainers; later, as the program progressed, village midwives were used for training. I I Pakistan’s beginning efforts to understand community perspec- ' tive have recently been evident in the AHP Sind project (Abbasi, 1985, | p. 17). In monthly meetings, training teams returning from the field report on their experiences. Lady health visitors (trainers and ; . supervisors of TBAs in Sind), social workers, the project director, and UNICEF personnel attend these meetings and discuss the experiences i i of the project personnel in the concluded training (which, for the | I most part, is in local communities). Previous TBA programs simply did i I I ' not take community perspectives into account. Baluchistan and Punjab I provinces are beginning to implement programs, and incorporating the 1 lessons from Sind province into these programs would help nationally. | Dr. Banoo, who is responsible for national coordination of the TBA j i programs, could greatly assist in sharing information, but to date i I I I she has been given no resources. ! Most of the coordination so far consists of sharing the Sind ! province findings with institutions also responsible for dai training, and UNICEF is actively using the Sind information and experience in the implementation considerations of other TBA programs in Pakistem. 234 The KAP surveys on dais were thorough in Sind. From July to December 1984, data were gathered on the age and experience of dais (the aver age age of a trained dai is 46.7 years), the dais’ experiences in conducting childbirths before training, the geographical area of their service, their educational background, how they received their | training as dais, their marital status, and their reasons for becoming 1 a dai. A report on the knowledge, attitudes, and traditional beliefs of Pakistani dais— like the studies of Blanchet (Notes 20, 21) in I Bangladesh— would be invaluable. Government support for dais in Sind includes hardship salary supplements, attempts at linkage with other TBA programs, and educa tion and orientation of physicians working at peripheral units near . trained TBAs’ activities, but the efforts are seminal. The Sind pro ject has transportation problems, but it does have six Land Rovers, 1 six scooters, and a jeep for the project director. i i The training curriculum has been revised as a result of feed back, so a new dai handbook (Abbasi, 1985, p. 15) was to have been ! released in January 1985. The dai handbook was pretested in 32 train- i ing groups (based on the learning objectives of dai training in the ' Sind project) and is now in final form. It is a visual aid useful not only for dais but also for motivating mothers and elderly women. Each dai will get a copy of the book at the end of her training, and she is to discuss the illustrations with other household women and community members. A Dai Curriculum Task Force, constituted by the Planning Commis- I sion of Pakistan in 1982 developed a model curriculum that has j since been adopted by all Federal and Provincial agencies. This . 235 ; shortened curriculum is delivered in fifteen training days, each approximately of six hours duration. The curriculum focuses on safe delivery practices, pre and post natal care, breast feeding, nutrition, oral rehydration and immunizations. Initial training of fifteen days is strengthened by refresher training and a feedback follow up after three months. (Abbasi, 1984, p. 4) The selection of lady health visitors as trainers precipitated 1 the problems of competition (they tried to take business away from the| TBAs) and poor functioning as trainers. In the Sind project, only I those having local language abilities were selected. No other crite- j ria were used in selecting in any category— medical technician, lady ' health visitor, or family welfare visitor— in Pakistan. Again, in the Sind project, elderly woman "ayas" are included in the training . team to bridge the age differential between trainers and trainees. No special criteria for selection of TBAs were mentioned before the Sind project, in which an indicator was used to determine the accuracy rate of dai selection (Abbasi, 1985, p. 2). The indi- cator can be computed as: j I Number of trainees that fulfill criteria x 100 1 Accuracy rate = Total number of trained daii | i , The criteria were: (1) the dai should be rurally based, (2) she should | ; ' I be in good health, (3) she should be well accepted by her community, ' ' and (4) only one dai should be selected from a given village. The j ' indicator can illustrate the differences in the selection process at \ i different times: an accuracy rate of 80 percent was achieved from July to December 1984, whereas it was only 41 percent over the two preceding years. The target is to achieve a selection accuracy rate of 100 percent. I - I 236 Training In Bangladesh, training was conducted in the local health center. The training consisted of 24 sessions, two sessions per week for 12 weeks. The TBAs were given a kit (imported from overseas at a cost of $25 per kit; the contents could not be replenished locally) ! and a stipend during training (although discrepancies were uniformly ; I noted in the amounts recorded as having been paid, vs. the amounts actually paid, to TBAs). The content of the curriculum was borrowed from Pakistan; the subjects to be covered, by session, are listed j I below. 1. Introduction and Purpose of Training 2. Signs and Symptoms of Pregnancy 3. Anatomy and Physiology of Female Reproductive Organs 4. Nutrition during Pregnancy 5. Personal Hygiene/Breastcare 6. Tetanus and Immunization 7. Antenatal Checkups 8. Minor Discomforts of Pregnancy and Relief Measures 9. Signs of Abnormal Pregnancy 10. Preparation for Home Birth 11. Use and Care of Dai Kit 12. Normal Labour and Delivery 13. Breastfeeding 14. Normal Progress of Puerperium 15. Diet While Breastfeeding 16. Proper Infant Feeding and Weaning Practices 17. Complications in Labour, Delivery and Post-Partum Period: Importance of Referral 18. Immunization for Children 19. Prevention and Management of Diarrhoea and Vomiting 20. Reporting of Births to Health Centre 21. Family Planning: Advantages of Child Spacing. (Greenberg, 1984, p. 37) There was no provision for practical demonstration training on ante- ' natal and postnatal care. ! The subjects were introduced too rapidly for illiterate TBAs to comprehend. An invaluable method for teaching illiterate : 237 : people— repetition— was not utilized. Training was mostly by class room lecture. Evaluation studies concluded that (1) the training lacked a practical component, and (2) the trainers were too overloaded to do the training properly and were themselves insufficiently trained in obstetrics to be TBA trainers. The TBAs’ learning was limited by | the lack of repetition and demonstration, and by the superficial teaching of a number of topics— i.e., by doing too much too soon. I The training should last longer than 12 weeks and should consist of [ 16 half-day classroom sessions and visits by the family welfare vis- I itor to the TBA trainees in their villages. The curriculum should be ! redrafted in Bangla (the local language of the family welfare vis- ' itors) in an understandable manner. The curriculum should stress repetition and demonstration as educational methods and should further emphasize safe delivery practices, including care of the newborn and breast-feeding. Traditional birth practices and beliefs ; should be discussed openly; harmful practices should be explained in 1 Î , a way that shows respect for the traditional practitioner (Greenberg, | 1984, p. 22). In Brazil, training started out at the MEAC hospital, but ; soon was being given locally as much as possible. At MEAC, the TBAs ! were observed picking up unnecessary practices (e.g., attempts at ; performing episiotomies and using forceps), so training was done locally with the aim of interrupting the rhythm of the TBAs’ life as I little as possible. Four or five lessons were given over a five-week I period, and 24 hours were spent in a clinic. During training, each I TBA was given a kit (pared down to a minimum), a daily allowance, and 238 a uniform with an emblem showing her village. The TBAs selected to work in clinics were paid a salary, usually by the community. In addition to free materials, TBAs received financial aid during training and in-service training. The methods of repetition, demonstration, and participation were stressed in gearing the train- ' ing toward illiterate TBAs. Various TBA tasks were listed in an evaluation study, but they were contingent on several factors and were taught incrementally to selected TBAs who could learn and benefit from additional training. TBAs are in different stages of performance depending on: - when they initiated training; - where they work and what their client volume is; - how closely they are monitored or supervised; and - how much "recycling" they have had. Tasks include, but are not necessarily limited to: - prenatal exam and screening of high risk patients for nurse and physician management; - monitoring all stages of labor; - delivery; - newborn care; - teaching mother newborn care and breastfeeding; - teaching clients all contraceptive methods, indications and contraindications ; - screening family planning patients for medical problems and referring them as needed; - dispensing birth control pills, referring patients who want lUDs to physician or trained nurse, referring ligations to physicians or to maternity hospitals, referring barrier methods to the BEMFAM program; I - giving follow-up to all family planning patients and referring them to physician or nurse when problem arises; , - giving client and community education in nutrition, with i emphasis on small children, infants and women who are pregnant or 1actating; - helping clients utilize local food to the maximum, giving out I recipes (recipes currently being tested by nutritionist and | I home economics specialist); - handling infants and small children who need oral rehydration, ' teaching mothers to prepare mixture, referring patients who ; i are severely ill or patients who don’t show improvement; | ! - teaching hygiene and cleanliness; and - offering first aid. (Development Associates, 1983, p. 10) 239^ The evaluation of TBA learning was done by MEAC staff. It is more difficult to assess the program quantitatively because of the proximity of the city and the choice of delivery care for the more affluent inhabitants. However, a comparison between hospital (MEAC) deliveries and TBA deliveries in maternity clinics suggests that maternal deaths, stillbirths, and neonatal death rates were higher in hospital deliveries (71% of which were deliv- i ered by physicians or medical students) than in clinic deliveries , by TBAs. This may, however, reflect a difference in the health status of the two groups of women served. The hospital group had poorer health judging by the percentage of low birth weight babies ! [ born in hospital (9.4%) compared to 5.3% in the TBA maternity clinics. (Araujo, Note 12, as cited by Walt, 1984, p. 18) ' Any TBAs judged incapable of adding new roles to their practice were I not trained further. In Peru, the national TBA program generally trained TBAs in training hospitals. In certain areas (e.g., Aguaruna), the location of training was objected to because women were not permitted to leave the home alone. Training lasted two weeks; meals and lodging were paid for, a certificate was given upon completion of the course, and kits were provided. The kits consisted of an apron, a hat, basic ; medicines, gauze, scissors, and simple instruments. The training ! methods used stressed lectures and oral drills, with little discussion j and participation. The social status of the TBAs differed from that | of the sanitarios: interaction with persons higher on the social scale j is difficult in the Peruvian culture. t S ! Evaluation of the Peruvian program did not include a study of j ' ! ' the performance of TBAs in the community (Favin et al., 1984, p. 38); j ; it was based on the number of TBAs trained and their performance on a ' written or oral examination. As was pointed out by Davidson (1984), 240 what TBAs answer in an examination and what they actually do may differ. In Ghana, training for the Danfa project was done locally as much as possible. Its duration varied. TBAs may be trained in residence boarding in a centrally located village or health facility for a variable period of time (usually ' one week), or TBAs can come to a central village or health facil- ; ity for training (intermittent training). (Danfa Project, 1977a, p. 12) , The TBAs were given delivery kits and a certificate of completion for ! completing the course, but they were not paid for attending. Teaching I ! methods stressed lecture, demonstration, and role-playing, with a ' great deal of repetition. The following is a brief outline of a suggested training j programme for TBAs. Session Topic 1 Getting Acquainted; Registration for Course; Discus sion/Demonstration of TBAs* Practices; Explanation of TBA Record-Books 2 Continuation of Session 1 i 3 Antenatal Care ; 4 Nutrition During Pregnancy 5 Disorders and Complications of Pregnancy 6 Identification of High Risk Women 7 Midwifery Kit and Its Use 8 Labour 9 Delivery 10 Immediate Postpartum Care 11 General Pospartum Care 12 Family Planning (Part A) 13 Family Planning (Part B) 14 Infant Feeding and Immunization 15 Continuing Training/Follow-up Procedures, Rehearsal for Graduation 16 Graduation. (Danfa Project, 1977a, p. 12) | Training in Ghana was evaluated via site visits, checking ' with mothers who had recently delivered, looking at TBA record books, ' questioning TBAs about their practice, watching TBAs perform in skits 241 (at the graduation ceremony), and giving formal and informal examinations during and after training. In Pakistan, TBAs were trained in village centers (district) or in hospitals. Now, there is recognition that training TBAs in a hospital may be just as inappropriate as training surgeons in a village. Training courses in Sind province last for 15 days, and a refresher course is given after three months. During training, TBAs receive stipends and kits (which are locally made and as simple as possible), The curriculum content in AHP training programs for TBAs is as follows: Session Topic 1 Introduction 2 Diagnosis of pregnancy 3 Female reproductive system 4 Dietary needs of pregnancy 5 Personal hygiene 6 Tetanus diagnosis and prevention 7 General obstetric examination 8 Discomforts of pregnancy 9 Abnormal pregnancies 10 Preparation for home delivery 11 Dai kit contents, use and care and discussion 12 Labour diagnosis, management and delivery of baby and care of the newborn 13 Importance of breast feeding, technique 14 The puerperium 15 Dieting advice to new mothers 16 Infant feeding, especially breast feeding and weaning 17 Complications of labour, treatment and referral 18 Motivation for immunization against the six EPI [expanded program of immunization] target diseases 19 Diarrhoea, causes, symptoms and management with ORT [oral rehydration therapy] 20 Live birth reporting 21 Child spacing. (Joint Review Team, 1985, p. 67) There is little "hands-on," practical experience, because training is given in centers away from villages. Even though the AHP course is designed to allow supervised deliveries, circumstances frequently ! 242 : prevent them. In Sind province, teaching methods included partic ipatory training techniques, story-telling, and other nonformal techniques geared to the illiterate TEA. ■ Evaluation of the TEA programs and the learning of TEAs in Pakistan was feeble. In Sind province, however, evaluation has yielded data on location of training, dais aged 60 and over, training groups (by size), and number of training courses. Figure 10 shows ' these data presented graphically. I Postadoption Supervision of TEAs in Eangladesh's program is the responsi bility of the family welfare visitor, who is so overworked that she cannot possibly supervise the TEAs in the field. Greenberg (1984) i ! suggested that two family welfare visitors be put in local health , ' centers. The supervision they provide consists of monthly meetings ' i , at the local health center, and there is no documentation that they ' are given any orientation to training (Claquin et al., 1982; I Greenberg, 1984). The supervisors * obstetric skill (Greenberg, 1984, | p. 10) is not commensurate with their task of TEA supervision. The i , training course for family welfare visitors is too traditional, and ^ : supervisory considerations are not in the curriculum. The issue of i remuneration was a fiasco (Greenberg, 1984, p. 19); bureaucratic I , mix-ups were responsible for the failure to pay family welfare I visitors per diem or stipends for their training and orientation. No ! surveys were done to get information on the perspectives or the roles I of the family welfare visitors. 243 SIND DAI TRAINING PROJECT TRAINED DAIS AGE 60 YEARS i 40 LU 30 30 20 O l O A u g , S «« t. O c t. Nbr. 1984 D ec. 3 VILLAGES AS LOCATION OF TRAINING SITES <100- % T O - F so VwlyCcc 82 83 83 84 84 ^too- Û 9q. è 8a to 70. 1 60 u. 50 O 40 U i < 30 t w 20 to- w a Q S 3 0*c Jon Uvty %Wv 0 # C 4o n Ju n « J u ly O ac 82 83 83 84 8 4 DAI TRAINING COURSES JULY 8243EC 64 i n i t i a l t r a i n i n g 50- 30- 20 «*Bty-0«c. Jon-Ouly Jm-Jwrn* JulfOte 82 83 83 84 84 ' Figure 10. Sind Dai Training Project. (From Abbasi, 1985, p. 14.) ; 2441 Kits took up 30 percent of Bangladesh’s TBA budget (Greenberg, 1984, p. 21). In subsequent evaluations these expensive UNICEF kits were found on the black market, men were seen carrying them, and items in the kit were not replaced. There was no evidence ' of in-service training and no patient referral system. Family wel- : ' fare visitors should develop a referral system for high-risk patients ' (one has already been developed in an MOH family-planning project in Matlab, Bangladesh). There were supposed to be 12 monthly meetings j in which TBAs were brought to local health centers for in-service I - training, but the project director returned 300,000 Taka (Bangladesh ' currency) to UNICEF because no one had claimed it for meetings. Evaluation studies also focused on salient shortcomings of : the Bangladesh TBAs* practices. The external evaluation found that 14 out of 155 trained TBAs guarded the perineum with their feet, 47 exerted pressure on the abdomen to deliver the placenta, and 48 recommended that breast feeding should begin as late as the third post partum day. 90% of the respondents cut the cord after expulsion of the placenta. This may be considered acceptable, but only if the infant is attended to promptly. In all but one of the cases cited by the evaluation, the infant was left on the ground, wet and without its mouth cleared of mucus, until the placenta was delivered. In the six deliveries performed by trained TBAs, only 2 boiled the blade. The other 4 TBAs washed the blade with warm or boiled i water. The practice was also found in trained TBAs in Manikgan ; District. (Greenberg, 1984, p. 21) , No evidence was found that evaluation results were communicated to MOH, : and other health-care planners. Supervisors of family health visitors' Î ' often were male, with little knowledge of obstetrics. I I In Brazil, the university PAPS team (a PHC team) is respon- | , I sible for evaluating the TBA program (Development Associates, 1983, I i ! p. 8). The university Health Science Department provides supervision, j ! 245 The team, which comprises university physicians and nurses (Walt, 1984, p. 17), helps in supervising clinic TBAs once or twice a week and domiciliary TBAs once every two weeks. Domiciliary TBAs are some times supervised by trained TBAs who work in clinics. Refresher courses, consultations with trained TBAs, and nonformal education are firmly linked to personnel in the formal I health-care system at the university hospital. Maintaining good relations with TBAs is a prime program goal that continues beyond the j training stage. Training of supervisors, which takes a great deal of i effort, also is at the core of the program in Brazil. The type of supervision varies according to the TBA's level of knowledge. Supervision is done either at clinics or in homes, ! where the emphasis is on a calm, traditional atmosphere. The clinic TBAs are supervised frequently because they serve large segments of the population. These clinic TBAs eventually help supervise domi- 1 ciliary TBAs. The PAPS personnel are paid by federal and state funds supplemented by a grant from the Kellogg Foundation. Much sacrifice and dedication is called for. The clinic TBAs* salaries are paid by the community. ! I In-service training was the last phase of training in the | Brazilian program. It covered subjects taught at regular classes or ' ^ subjects of concern to 72 midwives queried. Also, the contents of | j the TBA kits were replenished (soap, water), although the kits had i ' been pared down to a minimum as a result of project experience. ; Î Janowitz, Wallace, Araujo, & Araujo (1985) reported that the 1 overall level of mortality was lower (2.4 %) in referred patients 246' than in nonreferred public patients (5.9 %) at MEAC, The MEAC | referral system was designed to allow TBAs to attend uncomplicated deliveries but refer women with complications. From recent evalua tion studies (e.g., Janowitz et al., 1985), it is clear that this referral system is working well in saving scarce hospital resources ’ to treat only the most complicated cases, although there is a slight problem of over-referral by trained TBAs. Evaluation of program ! effectiveness was preceded by thorough formal and informal surveys on ' TBAs and local populations. Some survey results are shown in I Table 9, as an example of the evaluation done by MEAC staff. The j same staff who teach and supervise TBAs now also do the evaluation, but in the future nonformal education workers— especially literate ; ' TBAs— will be a part of the process. The Peruvian TBA program used "out" supervision of TBAs— infrequent visits to regional and local health-care personnel. However, the sanitarios were not held responsible for their local TBAs, and supervision suffered. There has been little evidence of training of supervisors, no evidence of content (task-oriented) super- ; vision, and little evidence of remuneration of supervisors (Favin et al., 1984, p. 38), all factors that hindered TBA supervisory acti- | I vities in Peru. Because of the lack of supervision, TBA kits were in ! , disrepair, no in-service training was done, and program regulations j were not complied with (Davidson, 1984). Referral procedures were I hampered by transportation problems. Evaluation activities were weak because of the lacks in (1) community input, (2) feedback of TBA acti- ; vities to communities, and (3) attempts to examine TBA performance in 247 < D r-H ■i E H -P bo » 0 -P 73 0 0 tp rH O lO CO -P 0 to g > L e o -H 0 2 Si: PQ r - t a } o « g i 05 rH L r —{ 0 0 cc •p p 0 <H Ï ? A S 0 >> c- < 0 CL P W « 0 P to CO O *H O CL 0 CO 00 ♦H 05 > i - H 0 >» 0 0 1 (p «H -H O r - H S m to L •H 0 00 C O * Ü > 3 O * r H tH f " 1 1 S • t H 0 4 - > 0 O i •H • t H L •P 0 0 ( 0 > Ü •p •P * i H c / 3 rH g CO 0 -H CM 73 73 rH 0 Ü 4 - > cp O o a 0 •H rH 0 o C / 3 % 73 1 0 0 Pi tl.2 0 0 S o r-H « t H bO 0 r - H 0 CO O p5 lO S C O 00 05 t £ > C M § lO 05 t> O c- s CO 05 00 05 o C M 05 05 O C M O C O 2 •H I O § r- H 0 L 0 C s 4 0 Ü 0 3 Xt 3 ♦ tH 8 •H 3 O L •H P •H O •H Ü bO 0 0 0 0 0 -p 3 L P r-H o O CO E H P4 < 5 2 5 00 A I s 248 communities. Davidson (1984) found that health-care professionals held two erroneous assumptions about TBAs: (1) TBAs are "empty ves sels" that can easily be filled with modern practices, and (2) TBAs are "highly adaptable." The consistent finding, however, is that innovations are not easily updated into traditional practioners* systems. In Ghana, the Danfa project (like the Brazilian TBA program) accented TBA roles within the context of a comprehensive PHC health- service delivery system. Supervision was thorough, done by MOH personnel who were highly interested in research and in delivering and upgrading the practices of TBAs in rural areas of Ghana. The district public-health nurses, the MOH aides, and the local profes sional midwives all were involved in supervision. The public health nurses were crucial; they oversaw the supervisory activities and did supervision themselves. One visit every two months was considered best, but at least one visit was made every three months, and periodic refresher courses (Neumann et al., 1983, p. 8) were linked to supervision. Supervision was task oriented, with specific tasks including: 1. Reestablish friendly rapport with the TBA. 2. Discuss any problems the TBA has experienced in her practice or in her referrals. 3. Discuss any complications or deaths that have occurred since the last visit. 4. Collect and record data on TBA activities since the previous visit. 5. Periodically evaluate quality of TBA work by observing the TBA during prenatal, post partum and family planning visits (using observation forms provided). (Danfa Project, 1977a, p. 33) 249 Supervisors received per diem and certificates upon completing grad uation requirements. Supplies were replenished during in-service training and supervision. The TBAs prepared their kits during training, using very simple, locally available materials. Refresher courses were held every three months in local villages at times con- ! . venient for the TBAs (Danfa Project, 1977a, p. 27). Referral to a health facility was one of the most important lessons given the TBAs; ; t ! different colored cards were used for different complications. (This ' I , practice, used for illiterate village health-care workers, was sim ilar to a system that was supposedly used in Bangladesh, where lack ; of supervision and support of the TBAs kept it from functioning.) Evaluation of Ghana’s TBA programs was thorough and was done ' according to program objectives, which included prenatal monitoring, recognition and referral of high-risk women, safe deliveries and proper umbilical cord management, and family planning. Cost estimates ranged from $4.04 to $6.07 (excluding one-time startup costs) per birth, and the cost of contraception motivation was very reasonable i (Neumann et al., 1983, p. 14). The TBA training was evaluated on the ' , basis of a skit at graduation, formal and informal quizzes during and i I after training, and on-site visits. 1 , Pakistan’s TBA program involves family welfare workers, med- I ical technicians, and lady health visitors. Infrequent, nonroutine visits were the norm, but lessons from the Punjab, Baluchistan, and Sind provincial TBA-training programs are starting to be put into | . supervisory training. Supervision is both "out" (e.g., the Sind I * 250 mobile team) and "in"; all supervisors have a multitude of functions and are overworked. Training is poor but is being upgraded in pilot projects. Even task orientation is starting to be connected to supervision in a Sind project. In this same project, salaries are augmented for supervisory activity, but remuneration is not con- ■ sidered in the literature on TBA supervision in other projects in I Pakistan. Very little follow-up has been done, but AHP programs in Sind and other programs in Punjab are starting to face this issue. I The TBA kit in the AHP program is too elaborate; if lessons I ' from other countries apply, it must be cut down to essentials and I local materials must be used as much as possible. In the Sind project, this has been done, and kit utilization has increased: according to lady health visitors’ questionnaire surveys, nearly 100 percent of the TBAs were using a kit, whereas only 63 percent were using one before training (Abbasi, 1985, p. 3). Lack of transport is a problem in referrals. Revised training has accented referral, but : TBAs have little or no contact with MCH centers of the basic health units, and according to an analysis of the Sind project (Abbasi, 1985, p. 26), 92 percent of trained dais are not referring cases. The focus is now on high-risk pregnancies and problems of referral. Evaluation of Pakistan’s AHP program in Sind province reveals that targets are not being met. Although the issues of poor national I support (already mentioned) and coordination need to be resolved, 1 they are being documented and solutions are currently being proposed; I among these are curriculum revision, supervision, and kit refinement. 251 CHAPTER VII FINDINGS, CONCLUSIONS, IMPLICATIONS, AND RECOMMENDATIONS This study has analyzed traditional birth attendant (TBA) pro grams from the vantage of a systems model. Scrutiny of the existing j empirical and theoretical literature on TBA programs demonstrated that I j analysis of these programs thus far has been too narrowly focused; it is crucial to examine certain issues not previously considered. Among these are the policies and implementation strategies of donor agencies and national health systems. The use of nonformal education for TBAs, taking into account a quartet of factors (sociocultural, political, economic, and managerial) that affect TBA training, is also crucial. ’ A flexible approach is the key. Many bureaucrats tend to regard written documents and well-thought-out plans as sacrosanct, but the literature is replete with refutations of this fallacy. The process of implementation determines the successfulness of TBA programs the world over. "Learning-by-doing" and mutual adap tation in implementation are key concepts in improving the efficiency of efforts to revamp formal health-care systems in developing nations, particularly efforts to deliver health-care services to the rural I (majority) populations. National TBA programs need to involve min- ! istries of health (MOHs) to assist in primary health care (PHC) ; efforts in general. Also needed is study of district or provincial ' formal health-care systems and articulation of their linkages with ^ ‘ " 252 one another— in particular their linkages to the health-care delivery ; systems that are invariably vital to successful TBA programs and PHC efforts. Findings in the Existing Literature Primary Health Care The literature focuses on several issues, one of which is I equity: PHC efforts are an attempt to provide more health-care I resources to rural populations in developing countries. Researchers ^ stress that several essential components of PHC (such as use of traditional practices, intersectoral coordination, and community participation) that were adopted at the Alma-Ata conference in 1978 . are generally missing from current PHC programs in Third World coun- , tries; also, a lack of commitment is said to be a major impediment to [ the success of PHC strategies. In Indonesia the national effort of . the Comprehensive Health Improvement Program— Program Specific (CHIPPS) to integrate information from local communities back into the MOH illustrates attitudinal barriers to PHC implementation; e.g., putting funds for "learning-by-doing" into the domain of the donors. Community participation, a key component, is lacking in most PHC efforts. Data on local population groups are essential for deci- i sions on allocation or reallocation of health-care resources. Traditional Birth Attendant Programs Most TBAs are illiterate or barely literate, so their initial i i training— if it is to be of significant value— will have to be based i on nonformal techniques. Repetition, demonstration, participation, j and a minimum of theoretical and classic techniques are essential r ' "■ 253 components of such training. The literature indicated that often this vital concept of gearing the TBAs* training to their background is virtually ignored. Concomitant with training is the need for careful selection of trainers, a requirement often overlooked even though its importance is almost universally acknowledged— at least on paper. Community edu- I cation before, during, and after training of TBAs is essential if the I TBAs are to really help in delivering better health-care service to ! I their communities* Generally, the literature stresses the importance of organized, systematic TBA programs involving regular follow-up, supervision, resupply, and organized in-service training beginning i within six months after the initial training period. Without such a ' system, the likelihood is strong that TBAs will revert to their , former activities. Classic, anthropologically oriented nonformal- education techniques are needed for follow-up, supervision, and i in-service training. Evaluation of the training programs is generally very weak, primarily because of the failure to do over-time (longitudinal) studies, the failure to assess community involvement, and the need for knowledge, attitudes, and practices (KAP) surveys on TBAs, clients, and communities. Cost evaluations are also weak. Implementation The preimplementation process entails the knowledge and involvement of social scientists, implementors, and members of local communities. Without either KAP surveys on TBAs or base-line data on 254 local populations, it is difficult both to begin to implement development programs and, later, to evaluate their effectiveness. Implementation literature emphasizes postadoption behavior and resistance to change as the key issues in formulating strategies for innovative TBA training programs. Also emphasized is mutual adapta tion, in which both the project and the site are changed as a result : of implementation. Literature on implementation also stresses (1) that conflict is a central political and administrative issue in I development programs and (2) that planning is a means to an end, not t an end in itself. Potential bias in donors includes avoiding the confrontation of previously overlooked factors in assumptions about development programs, when in fact these assumptions are the critical ! bases for successful implementation. Also noted were a lack of incen- * tives for donor-agency personnel to implement development programs, a lack of adaptive performance (the ability to learn from experience, including error) in bureaucracies in general, and a failure to acknowledge bias against really resolving the difficult issues of implementing development programs. Nonformal Education Like the formal health-care system, the formal edqcation system cannot functionally and logistically meet the needs of all I ‘ people in developing countries. The literature demonstrates the I j particular potential of nonformal education to shift the efforts of I formal health-care bureaucracies towards PHC. Often overlooked is I j the anthropological approach (an "emic," or insider’s, perspective on ' national, district, local, and traditional systems) needed to get 255 ' information crucial for strengthening TBA programs. For example, simple, nonformal evaluation strategy is more than sufficient if it generates information useful to implementation of TBA programs; more emphasis needs to be put on description and interpretation rather than on measurement and prediction. This need to shift emphasis is ! one reason why a local informant’s view of reality— the anthropologi cal approach— is so important. Nonformally educated trainers and supervisors of TBAs, the TBAs themselves, and social scientists who 1 sympathize with the local population’s problems and needs can I ' contribute immensely to implementation and institutionalization of TBA programs within the context of a PHC approach. Within the supervision domain lies perhaps the greatest potential contribution of nonformal education. The degree to which ■ TBAs and their supervisors are committed to PHC, and the social dis- ; tance from the supervisor to the TBA and her community (as well as | that from the supervisor to personnel in the formal health-care sys- i I tern), are particularly amenable to analysis based on (nonformally | educated) insiders’ views. Because such nonformal analysis is so , i useful in revealing aberrations in institutional processes, it is | i particularly suited to evaluating decentralization and management i participation with the goal of uncovering blockages and constraints | during the actual implementation of TBA projects. I I The national effort to reform nonformal education in Indonesia - was particularly pertinent because of the finding that crucial donor inputs (incentives) were needed to keep local participation a ' 2561 functioning part of the project, and an enormous effort was required to facilitate local endeavors. Findings from Analysis of Case Studies In the national TBA programs studied (except those in Ghana ' and Brazil), the community surveys and preparation and the KAP ’ ! studies— both for base-line data in the preimplementation stage— have ; major shortcomings. The training and selection of trainers needs I more careful attention because of the difficulty of teaching mostly j illiterate TBAs. Training was too didactic in some programs (those ' of Bangladesh, Peru, and Pakistan), although Pakistan’s TBA training shows evidence of changing. National TBA programs need to focus more on supervisory aspects. Pakistan, although having many problems with supervision of TBAs, is at least confronting the issues, whereas | Bangladesh and Peru really need to focus more attention on supervision j and follow-up in their TBA programs. Ghana and Brazil (both conduct- ■ I ing demonstration projects) have excellent supervision, follow-up, and | support for their TBA programs, and MOH personnel in both countries ; were trained for their roles. The training was task oriented. In all j i TBA programs, management and bureaucratic operations of formal health- ' care systems are major factors in program success vs. failure. Almost | complete neglect of management factors (as in Bangladesh and Peru) is ( ; 1 the norm for the majority of developing countries (Heaver, 1982). j I Pakistan is beginning to focus on implementation of national TBA ! I I ' training in the management domain. In Ghana and especially in Brazil, ' I , I I , the management components of the TBA programs have been thought out. The case-study literature indicated that the management and 257 supervision factors (intimately intertwined) were the most impor tant elements of TBA programs, whether national programs or pilot programs * Conclusions and Implications The foregoing findings were derived mainly from a review of the literature on PHC, TBA training, implementation of development programs and nonformal education strategies, and case-study analysis ; of TBA programs, and from an analysis of donor bias in actual imple- j mentation problems encountered in developing countries. The findings ' with regard to donor bias came from personal experience in working in ‘ development in Africa, review of donor-agency documents, and intimate contact with personnel working in the various donor agencies. Empiri cal and theoretical evidence discussed in Chapter III, "Review of the Literature," provided a useful background for analysis and deriva tion of conclusions. Although no claim is made for scientific validity ’ ; of conclusions and observations, these conclusions and observations stem from a growing body of evidence that is both respectable and ! convincing. This dissertation is a statuent of where we are in the I evolution of research on TBA programs. The systems approach stresses "learning-by-doing" and mutual ! j adaptation within the context of implementation. The evidence points I I I I to (1) a need to tap new sources for coordination to extend PHC to | j villages and (2) a need to force donor agencies and national health- | ( j care systems to increase their commitment, attention, and allocation i : of resources to PHC. The "health-for-al1" goal is far from being I realized. The need for new resources and strategies to reach this 258 goal is outstanding, and the means of doing so continues to be elu sive. One reason for this elusiveness is a fundamental flaw in analysis of TBA programs to date— a flaw in policy formulation, planning, and execution: miscalculation of the true nature of inno- I vations in nonformal education and of the sociocultural, political, j economic, and managerial context in which TBA programs are to | i • ! function. ! i For development programs such as TBA programs, the dilemma is ! i particularly pronounced. Primary health care is suffering under the | t weight of political, cultural, and economic turbulence as well as the conflict between new approaches and the old, unyielding traditions of formal health-care systems. As a result, basic practical policies for health-care innovation have yet to be sharply focused. Policies I and implementation strategies for TBA programs to date have been formulated from outside the developing countries, shielded from the real demands of the site— the realities of the developing countries* circumstances. Now, however, this situation is rapidly changing. The need for change is still urgent, but the necessary strate gies and resources could be within reach. International assistance (donor) agencies such as the United States Agency for International ; Development, the World Bank, and the World Health Organization must I : ’ be diligent in insisting on expansion of the search for resources and 1 strategies for PHC. The effectiveness of future TBA programs depends , , not only on local or national change in formal health-care systems, ' but also on a substantive and substantial improvement in the policies ’ 1 I and implementation strategies of donor agencies. r 259 Recommendat ions The findings and conclusions of this study readily precipi tated a whole spectrum of recommendations. At first, efforts were made to focus exclusively on TBA training. As the study progressed, the futility of this limited approach became evident. The systems model that was used provided a broader approach and resulted in the ! formulation of a wide range of recommendations. Consequently, a 1 deliberate effort has been made to accent the shortcomings of a too j insular focus on TBA program analysis. The following recommendations ■ ' are made in the hope of enriching future research on TBA programs in | I i developing countries. Application of the recommendations will be considered on a country-by-country basis. The need to be flexible is paramount in consideration of TBA programs. Only occasionally is it recognized that the formal health care systems often conflict with PHC strategies. Mechanisms to contain, expose, or deal with such conflicts should be studied in ' relation to the implementation processes of TBA programs and PHC efforts. Need for Donors to Confront Imbalances in Health-Care Systems The overwhelming bias toward smoothing over, rather than settling, conflicts deriving from substantial PHC efforts contributes i I j ; to an aid environment where unaccountability flourishes. In abdicat- j i I I ing international pressure, donors invite irresponsibility: countries j I ! can embrace "health-for-al1" strategies without committing resources , I to PHC. In 1950, two thirds of the world's people lived in the less 1 ! I , developed nations; by 2020, the proportion will be five sixths (Reich, j ! J ! ''' 260^ I 1985, p. 78). The implications for donors and for development poli- | tics are profound. What is being advocated in this dissertation is not the current philosophy of donors, which is a particular low-cost form of charity. All bureaucracies need to change toward efforts to deliver ; services (health care included) to the poorest people in the poor : countries. It is difficult to judge, but this study indicates that ; the need for donor agencies to increase their effective help in ^ actual implementation of TBA programs in developing countries is at , ' least as great as the need for developing countries to upgrade their capabilities in administering and managing these same programs. Previous research on TBA programs focused on the problems of educating rural people about health, training and supervising ' community health-care workers, and other issues connected to PHC • I efforts. Maybe the real challenge, however, is to create commitment > to PHC. For donors, such commitment translates quite tangibly into shifting three fourths of their health aid from allocation for urban | I minorities to allocation for rural majorities. In addition, the I ^ I i termination of PHC projects that really are not receiving any host- I i * country enthusiasm, support, and effort needs to be diligently | enforced. Such diligence in terminating unsuccessful PHC projects ' would entail a real departure from donor practices to date. Pilot projects (even when they "fail") build critical masses of dedicated : people; proponents say "this is the way to do it, just be more patient." Opponents cite the "brain-drain" factor— dedicated people , leaving the country (or at least its rural areas) for more fruitful , r- --- -... 261] pursuits— and the inability of national systems to support substan- ' tial TBA efforts. The evidence indicates that problems of health- , care delivery to rural areas involve a lack of will, not just resources. If one accepts pilot projects, a legitimate question is when to "graduate" to larger programs of routinization. What is being recommended here is not an across-the-board abandonment of ] pilot projects. Pilot projects in many countries have shown what i I they have to show. What is next? , Complexity of TBA Program Implementation The complexity of TBA program implementation derives from the problems of dealing with the innovation (in this case, the TBA pro gram itself), the local setting, and the formal health-care system. ; I ! I The "fit" of these three issues is a current problem for TBA programs ! worldwide. Because of this complexity, health-care professionals in i developing countries, including expatriates working for donor agen- I i cies, must widen their scope of inquiry to include problems of policy, \ administration, and management of health-care systems. In three | national case studies (Bangladesh, Peru, and Pakistan), management | and administration were key constraints to delivery of services to ! rural communities. Legitimate areas of concern for health-care j ! professionals, especially those in PHC and TBA programs, need to | include management, administration, emd policy issues. The review of literature on TBA programs confirms this need. To coordinate national, district, and local personnel in formal health-care systems, and to coordinate efforts toward implementation of PHC and 262 TBA programs, is an extremely challenging task (perhaps the task for the next decade or so) facing MOHs in developing countries. National efforts to reform health care toward PHC (such as CHIPPS in Indonsia) should be implemented. In doing so, however, more effort should be directed toward education and advocacy for PHC | efforts at the central MOH level. An entire division should be ! created to be responsible for seeing that PHC efforts are imple- I mented. In countries like Pakistan, dynamic people with resources 1 ! j and power— not one person— should be responsible for coordination and ■ support of TBA program efforts. Task-Oriented Supervision The failure to specify the work involved in implementation (what people actually are expected to do) is a major weakness of TBA 1 programs, especially with regard to supervision of TBAs. High-level health-care administrators also need training on the specific elements of support required for the lower-level MOH personnel who actually go out and supervise the village or domiciliary TBAs. The supervisors I also must have task-oriented training. I Two-Way Flow of Information i ' I Donor-supported TBA programs should be viewed as a two-way ' j flow of information. Assistance given within the context of ! cooperation fosters active participation by both donor agency and j I i t I , host country. Participating in implementation would force donors to ' I . be more flexible and more cognizant of the actual constraints faced : by implementors in developing countries. A reduction in apathy and passivity of the host country would be a concomitant positive effect r ■ ' ■ 2 6 3 ; of a two-way flow of information. Mutual adaptation— incremental changes in the characteristics of a TBA program to suit the specific site and context into which it is being introduced, with concurrent ; changes in the characteristics of the site or context to suit the specific TBA program being introduced— is a practical offshoot of the two-way information flow. The literature on TBA program implementa- , tion supports the contention that an implementation strategy is more j likely to be effective and to become institutionalized if it allows ( for flexibility and adaptation. t Organizational Arrangements Although it may seem very mundane, poorly conceived organiza tional arrangements (particularly those concerning vehicle maintenance and fuel for TBA activities) contribute to resistance and failure in implementation of TBA programs, even those with initial commitment. In some countries, a "bureau of rural resources" operating on a local level might solve the problems of lacks of transport, adequately maintained vehicles, and fuel— all ubiquitous, intractable problems related to poor supervisory performance, not only in TBA programs but , also in health-service delivery programs in general. Consideration t , should thus be given to creating separate local bureaus responsible for ensuring that local health-care centers have the vehicles, vehicle maintenance, and fuel to permit supervision of village TBAs. , Sharing and coordination of resources from different agencies would . not work because of problems in sharing (of costs, for example) between the ministry of agriculture, ministry of education, and MOH. I I The evidence from past efforts— integrated rural development, for r---------------------- 264 example— demonstrates the futility of such a venture. It can be argued that having separate local bureaus would add to bureacratic confusion, but donor support of these bureaus might be part of the donors’ follow-up to buttress supervision in TBA training programs. Donor support could be linked with a thorough study of local ; communities, TBA supervisors, and local health-center operations. Community Financing Although the Alma-Ata recommendations are sound in their j widespread confidence in self-reliance, which demands a more ! I important role for the community, and the goal of "health—for—all" ! I calls for every financing option, realities require a balanced approach. It must be realized that community financing is, at best, only a partial solution, that it may be more difficult and less effective than the reallocation of direct resources, and that governments have to encourage and facilitate it, not impose it. Researchers must produce more case studies and report not only on income raised but also on the community processes involved in raising , funds and on the subsequent effects of this on the scope and accessibility of services. The time has come to move from i rhetoric to reality, and from small demonstration projects to routine national programs. (Stinson, 1984, p. 125) ’ : I Nonformal Education i ■ The traditional (rural) health-care system should be singled j I i out as a subject for research on change, in the belief that norms, | reward systans, and conflicts are essential and, when understood, will | ■ ; ; greatly enhance implementation of TBA programs. The scarcest resource ; is often the ability to get things done (such as managing and deliv- ; ering PHC in rural areas). Nonformal education can be useful in r ' ~ 265 addressing implementation of TBA programs and solving social problems in that context. The bureaucratic agenda is potentially a particularly fruitful area for the introduction of nonformal education. The literature and case studies on TBA programs in this dissertation highlight the lamen- : table lack of three essential components: (1) knowledge of local ' system, (2) base-line data on TBAs, and (3) specific objectives for TBA programs, along with performance indicators needed for useful 1 evaluation studies. From the literature on TBA programs and ! development programs in general, it is clear that even when these essential components are present, the major determinants of TBA pro gram effectiveness are (1) the degree of task commitment to PHC on the part of TBAs, their supervisors, and local formally trained health-care personnel, and (2) the degree of social distance between local MOH personnel and TBAs, between TBAs and community members, and between community members and local MOH personnel. A nonformal- education approach to analysis of the local health-care structure and personnel could be effective, over time, in differentiating between elements that facilitate supervision and elements that constrain it. Training of TBA Trainers If at all possible, TBA trainers should be TBA supervisors. This dual role will enhance the quality of supervision and should help in evaluating the effect of the program. Trainers must be schooled in methodology appropriate for teaching illiterate persons. In addition to this component of training, a sound knowledge of local conditions and of obstetric problems during pregnancy and delivery is J 266' important to the quality of teaching. Breast-feeding and spacing of ; births are both important components of TBA training, so TBA trainers should be thoroughly versed in these topics. At Princeton University, recent analyses of data from 25 developing countries showed that spac ing of births has a substantial impact on child mortality. If all births were spaced at least two years apart, infant mortality could be reduced by 10 percent and child mortality (ages 1-4) by 16 percent (Huffman, 1984). ; ' I j Several proposals for literacy education of mothers along j with education of TBAs are now being considered. This approach could be significant: most TBA trainers are literate, so they could become I facilitators of literacy programs for women. The World Fertility Survey (1972-1984, 1984 symposium, London, 1984) found that in virtually all 42 countries surveyed between 1972 and 1984, both infant and child mortality decreased with increasing years of maternal education. Overall, the ratios of infant death rates for the highest and lowest education groups range from about one-third in Benin and Costa Rica to more than two-thirds in Bangladesh and Lesotho. (Grant, 1985, p. 98) A study in Kenya attributed 86 percent of the national decline in infant mortality between 1962 and 1979 to the concurrent increase in , I maternal education; the remaining 14 percent can reasonably be ! ! attributed to improvement in the household economic situation | I (Mosley, 1983). Supervisors generally have a higher level of | ! I education than TBAs. In Brazil and Ghana, TBA supervisors were MOH | ! I . staff and generally were highly educated. Likewise, the sanitarios ' I ! in Peru, the family welfare visitors in Bangladesh, the lady health i visitors in Pakistan— all of whom are supervisors— generally are ] literate. These people could become instrumental in educating rural | I I r -------------------------------- 267- women. The education need not bring the women to literacy per se; , indeed, evidence from the literature indicates that in countries like Bangladesh and Pakistan, preliteracy education pertaining to basic hygiene would greatly help the condition of poor women and children. I Training of TBAs , The training of TBAs should be kept simple and geared to the backgrounds of the TBAs, who are mostly illiterate. In the case of ; ! Bangladesh, even though the literature (e.g., Favin et al., 1984, p. I 38) states that TBA illiteracy has been taken into account, closer scru- tiny revealed that, in fact, the TBA training did not take into ' account the needs of the TBAs. This is just one example of many instances in which the actual preparation of trainers for TBAs has ! Î I been poor ("actual" denotes what has really happened in the field, , ' ) not what is documented on paper). Without careful preparation geared to the TBA's background, training will be ineffective. Only simple training is desirable, with the constant realization that the prin- i cipal responsibilities of the TBAs are to facilitate normal delivery and to recognize high-risk patients and refer them to the nearest health center or support hospital. An attempt to transform simple, unlettered local health-care agents (TBAs) into university-level ! • paramedics is a serious error and is not an option unless the TBAs go | I 1 ' through formal education. Younger TBAs would be more amenable than 1 I older TBAs to such an approach. I Community Involvement ! Any village-level training strategy must take community prep- | aration into account. Community involvement and base-line surveys to 268 augment knowledge of local systems are acknowledged as sound princi ples in TBA literature but nonetheless are often overlooked. This must not happen. If these principles have been overlooked and a large TBA project is already in progress, then vigorous efforts should be made to get additional funds for needed surveys and community aware- I ness and participation. In the meantime, every effort should be made ' I I to gather pertinent data on local community characteristics as the i TBA program progresses. Summary t The recommendations include major points about TBA programs in general that either are not found in the literature or are merely ■ implied there and not dealt with. The importance of factors in TBA i I ! training— supervision, community participation, and cultural charac teristics of TBAs and their communities— was confirmed in the | I literature reviews and in the case studies of TBA programs in , Bangladesh, Brazil, Peru, Ghana, and Pakistan. Factors that are crucial to successful implementation and institutionalization of TBA j ' programs, but are often either not dealt with or unrealistically j de-emphasized, include forcing health-care professionals out of | I ( I strictly parochial concerns, reducing donor-agency bias concerning I implementation of TBA programs in developing countries, correcting assumptions about TBA programs, and reallocating health-related donor aid toward PHC efforts. i The new focus of development thinking (economic liberal- | ' ization, increasing administrative capacity, and implementation) is | I I I both timely and urgent. The legitimate concerns of health-care r ' ' 269 professionals must be broadened to include issues of management ■ and administration. In Bangladesh (Greenberg, 1984) many of the recommendations for the national TBA training program entailed administrative considerations. A management professional with experience in administration is needed to direct the current program I in Bangladesh (experience as a physician or nurse is of minor impor- ' tance). A similar focus on management problems in the national TBA program is evident in Pakistani literature on TBAs. Peruvian lit- j erature on TBAs cites management as a constraint; currently, the I recommendations of Davidson (1984) (many of which are management i related) are being considered on a national scale in Peru. The literature on TBA programs in Brazil and Ghana was replete with man agerial and administrative considerations. Review of TBA programs in I the Philippines, Thailand, India, and Zambia indicated that management is a critical constraint. Unless there is a fairly rapid breakthrough in our understanding of how new institutional models might be developed, we can pre dict that polarizations already underway will accelerate. The institutional outcome will then reinforce the most negative aspects of a disembodied, ineffectual bureaucratic culture superimposed upon equally impoverished and ineffectual local organizations. (Moris, 1984, pp. 64-65) In this study, an attempt has been made to determine the rationale, the nature, and the scope of TBA programs in developing I I countries. On the strength of a "human capital" theory many devel- I j oping countries tried, after attaining independence, to improve I education, human resources, and health to support their development ! programs. Many developing countries have suffered from their recur rent expenditures (up to 30 percent on education in some African 270 i countries), The pattern of recurrent expenditures in the health sec tor is still a constraint within these developing countries, but to maximize return on their significant investments in health, they searched for new resources and strategies. Primary health care— a supplement to formal health-care systems in developing countries— was seen as a major component of the "health-for-al1" strategy adopted by | many developing countries after the 1978 conference on PHC in Alma- i Ata. Traditional birth attendants were seen as an important component of the "health-for-all" strategy. Through external aid programs, many developing countries experimented with various roles j for TBAs in helping to deliver PHC, as part of an attempt to upgrade formal health-care systems. The ultimate (equity) goal was to pro- ! vide health care to underserved or unserved rural majorities. This analysis of TBA programs entailed the use of a systems model. Use of this model enabled nonformal education in preimplemen- : tation, training, and postadoption phases of TBA training to be seen in an unusually broad sociocultural, economic political, and manage rial context. The study confirmed what is highlighted in the I I literature on TBA programs; the need for a constant system of evalu- [ 5 ation and monitoring. Nonformal education techniques are optimal for ' most TBA training situations but may well need to be modified. The . core issue of this dissertation is that even though nonformal educa tion has great value in TBA training, it alone is not sufficient. ; The literature on carefully constructed TBA programs as an example of | : I social and nonformal-education experiments fostered the expectation ' that superior outcomes would be dramatically apparent. When no such 271 I i outcomes were observed, it became apparent that these TBA programs did not work out as expected. Instead of giving up on TBA programs, however, concerned parties are trying even harder to deliver health services to the poor. The obduracy of formal health-care systems to PHC efforts has been documented, and attempts to reduce or even eliminate this resistance are recommended. The concept of mutual adaptation as a model for implementa- ' tion may serve as a catalyst to help donor agencies and developing ^ countries realize the need to change in favor of efforts to deliver health services to rural majorities in developing countries. If so, ' this study provides an impetus for further inquiry into how this innovative strategy for programs like the TBA programs can be managed : effectively to deliver services to the poor. In a broader sense, : this study also accents the need for further inquiry into how the , politics of implementation of TBA programs impinge upon real problems encountered in delivery of health-care services in developing countries. 272 : REFERENCE NOTES 1. SPAAC. Basic health services— role and practice of daya in contemporary Egypt. Paper submitted to U. S. AID, Cairo, Egypt, by Social Panning, Analysis, and Administration Consultants, December 1982. 2. Kumer, V. Report of information on TBAs. Personal communication, November 10, 1984. 3. WHO. Research and support to further development of TBAs (including their trainers and supervisors) in MCH and FP services. Unpublished Project Request Document, February 1984. I 4. Mosley, H. New wine in old wineskins: Can primary health care I strategies be accommodated by traditional health systems? Talk given at a seminar on primary health care sponsored by National Council for International Health, January 31, 1985. 5. Goldman, W. Application of a strategy to reduce infant and young child mortality in Asia. Paper, May 1984. ' 6. Pyle, D. Personal communication, June 1985. 7. Favin, M., Bradford, B., & Cebula, D. Improvement in maternal health. Draft of a report (1983) published under a slightly different title (see Favin et al., 1984, in Bibliography). 8. Burns, K. Role of the traditional birth attendant. Background paper for program planning in the Northwest Primary Health Care Program (Somalia), October 1984. 9. Araujo, G. Personal communication, January 1985. 10. Pederson, B. Personal communication, 1984. 11. Neumann, A. K. Traditional birth attendant service programs Paper presented at the Conference on Universities and International Health, Chapel Hill, North Carolina, September 23-26, 1984. 12. Araujo, G. Training and utilization of traditional birth attendants. Paper presented in Cairo, Egypt, December 1983. 13. Araujo, G. Personal communication, December 1984. 273 14. Rosser, J. Unpublished paper on TBA Trainers Kit (UNICEF), International Planned Parenthood Federation, n. d. 15. Ü. S. AID. Design and evaluation of AID assistance projects. Course document for one-week seminar, 1980. 16. Grayzel, J. (Rural development officer for U. S. AID.) Personal communication, 1985. 17. Moris, J. Personal communication, February 1975. 18. Reynolds, J. Beyond evaluation— experience with operations research in international health. Paper presented at the Joint Meeting of the Evaluation Research Society and the Evaluation Network, San Francisco, California, October 13, 1984. 19. Franzel, S. Comparing the results of an informal survey with those of a formal survey: A case study of farming systems research/extension (FSR/E) in Middle Kirinyaga, Kenya. Abstract of a paper submitted to 1984 Farming Systems Research and Extension Symposium, 1984. 20. Blanchet, T. Women, pollution and marginality in rural Bangladesh. Book to be published by University Press, Dhaka, Bangladesh. '21. Blanchet, T. The Muslim dai: Performer of a dirty Job or Allah's meritorious servant. Paper presented at the Women for Women seminar, I.B.A., University of Dhaka, Bangladesh, June 26, 1981. 22. Araujo, G., & Neto, A. Programa de atengao primaria de sande. Fortaleza, Brazil, 1981. 23. Lamptey, P. Personal communication. May 1985. 24. Grieser, M. Unpublished report on a study, conducted in Pakistan, of a national strategy for communications on oral rehydration therapy, 1985. (Study funded by : Management Sciences for Health.) 25. Grieser, M. Personal communication, 1985. 26. Greenberg, J. Personal communication, 1984. 274 BIBLIOGRAPHY Abbasi, Ni (1984). Sind dai training project analysis of two year experience, July 1982-June 1984. Karachi, Pakisteui: UNICEF Sind Program. Abbasi, N. (1985). Analytical progress report on Sind dai training project, July-December 1984. Karachi, Pakistan: UNICEF Sind Program. Adams, D. (1972). Forward. In R. G. Paulston (Ed.), Non-formal education: An annotated international bibliography. New York, NY: Praeger Publishers. Agarnala, R. (1983). Planning in developing countries. World Bank Staff Working Papers No. 576. Ahmed, M. (1983). Critical educational issues and nonformal education. Prospects, ^(1), 35-44. Ahmed, M., & Coombs, P. (Eds.). (1975). Education for rural development— case studies for planners. New York, NY: Praeger Publishers. Ampofo, D. A., Nicholas, D. D., Amonooh-Acquah, M. B., Ofosu-Amaah, S., & Neumann, A. K. (1977). The training of traditional birth attendants in Ghana: Experience of the Danfa Rural Health project. Tropical and Geographical Medicine, 29, 197-203. Araujo, G. (1984). The Ceara' experience, traditional birth attendants and spiritual healers as partners in primary health care. Contact, 79, 1-14. Armenakis, A., Feilld, H., & Mosley, D. (1975). Evaluation guidelines for the OB practitioner. Personnel Journal, February, 99-103. Bannerman, R., Burton, J., & Wen-Chieh, C. (Eds.). (1983). Traditional medicine and health care coverage. Geneva, Switzerland: WHO. Berman, P., & McLaughlin, M. W. (1974). Federal programs supporting educational change: Vol. 1, A model of educational change. Santa Monica, CA: Rand Corporation. (R-1589/1-H.E.W.) Berman, P., & McLaughlin, M. W. (1975). Federal programs supporting educational change: Vol. 4, The findings in review. Santa Monica, CA: Rand Corporation. (R-1589/4-H.E.W.) 275 Berman, P., & Pauly, E. W. (1975). Federal programs supporting educational change: Vol. 2, Factors affecting change agent projects. Santa Monica, CA: Rand Corporation. (R-1589/2-H.E.W) Berman, P., Greenwood, P. W., McLaughlin, M. W., & Pincus, P. (1975). Federal programs supporting educational change: Vol. 5, Executive summary. Santa Monica, CA: Rand Corporation. (R-1589/5-H.E.W.) Berman, P., McLaughlin, M. W, Bass, G., Pauly, E., & Zellman, G. (1977). Federal programs supporting educational change: Vol. 7, Factors affecting implementation and continuation. Santa Monica, CA: Rand Corporation. (R-1589/7-H.E.W.) ! Bhola, H. (1983). Non-formal education in perspective. Prospects, ^(1), 45-54. ; Birdsall, N. (1984). Population growth. Finance and Development, ' September, 10-15. Bock, J., & Papagiannis, G. (1976). The demystification of nonformal education. (Invest in Nonformal Education No. 17). Amherst, MA: Center for International Education. Boland, R., & Young, M. (1983). Putting a price on primary health care. World Health Forum, 4, 131-134. Brembeck, C. (1973). Introduction. In C. Brembeck & T. Thompson (Eds.), New strategies for educational development. Lexington, MA: Lexington Books. Brethower, K., & Rummler, G. (1979). Evaluation training. Training and Development Journal, May, 14-22. Brown, H. E. (1979). The training needs of traditional midwives. Unpublished masters thesis. University of London. Burns, J., & Bialoscienicz, F. (1983). Road-to-birth game. (Technical Note No. 24). Amherst, MA: Amherst Center for International Education, University of Massachusetts. Calloway, A. (1973). Frontiers of out-of-school education. In C. Brembeck & T. Thompson (Eds.), New strategies for educational development. Lexington, MA: Lexington Books. I earner, G., & Korten, D. (1982). People-centered planning: The ' U.S. AID/Phi1ippines experience. (NASPAA Working Paper No. 2). Washington, DC: Association of Schools of Public Affairs and Administration. J 276 Carnoy, M. (1974). Education as cultural imperialism. New York, NY: David McKay Co., Inc. Center for International Education. (1982). Indonesia: Implementation of a large-scale non-formal education project. (Center for International Education, University of Massachusetts). Worcester, MA: Commonwealth Press. Chambers, R. (1983). Rural development: Putting the last first. London, England: Longman. Chenery, H. (1983). The evolution of development planning. (Paper No. 158). Cambridge, MA: Harvard Institute for International Development, Harvard University. I Claquin, P., Claquin, B., Rahman, S., Razzaque, M., Shaikh, K., Chowdhury, T., & Kanawati, N. (1982). An evaluation of the government training programme of traditional birth attendants. (Special Publication No. 18). Dacca, Bangladesh: International Centre for Diarrheal Disease Research, Bangladesh. Clausen, H. W. (1985). Poverty in developing countries. Washington, DC: The World Bank. Coffmann, L. (1980). Successful training program evaluation. Training and Development Journal, October, 84-88. Cohen, S., Grindie, M., & Walker, S. (1984). Policy, space and administrative systems research in donor-led rural development. (Paper No. 166). Cambridge, MA: Harvard Institute for International Development, Harvard University. Colgate, S. (1981). Midwifery, mothering, and behavioral assessments of African neonates. Doctoral dissertation, University of Illinois. Conyers, D. (1983). Decentralization: The latest fashion in I development? Public Administration and Development, 3, 97-109. I Conyers, D. (1984). Decentralization and development— A review of the literature. Public Administration and Development, 4, 187-197. Conyers, D., & Hill, P. (1984). An introduction to development I planning in the Third World. Chichester, England: Wiley. i Coombs, P. (1968). The world education crisis: A systems analysis. New York, NY: Oxford University Press. Coombs, P. (1970). What is educational planning? Paris: UNESCO, j International Institute for Educational Planning. 277 Coombs, P. (with Prosser, R. C., & Ahmed, M.). (1973). New paths to learning for rural children and youth. New York, NY: International Council for Educational Development. Coombs, P., & Ahmed, M. (1974). Attacking rural poverty: How non-formal education can help. Baltimore, MD: Johns Hopkins University Press. Cosminsky, S. (1978). Midwifery and medical anthropology. In B. Velimirovid (Ed.), Modern medicine and anthropology in the United States-Mexico border population. Washington, DC: PAHO. Cosminsky, S. (1981). Knowledge and body concepts of Guatemalan midwives. In McKay (Ed.), Anthropology of births. Philadelphia, PA: F. A. Davis. I Cosminsky, S. (1983). Traditional midwifery and contraception. In R. Bannerman (Ed.), Traditional medicine: Handbook for health practitioners. Geneva, Switzerland: WHO. Danfa Project. (1977a). A program manual for traditional birth attendants: Organization, training, and evaluation. (Joint project of the Department of Community Health, Ghana Medical School, and the School of Public Health, University of California, Los Angeles). i Danfa Project. (1977b). Preliminary manual for trainers of village health workers. (Joint project of the Department of Community Health, Ghana Medical School, and the School of Public Health, University of California, Los Angeles). Davidson, J. (1983). The evaluation of effectiveness of the Peruvian midwife training program. (Agency for International Development Contract No. 527-0129-C-00-20100 and Occidental Petroleum). Davidson, J. (1984). An anthropological evaluation of teaching reproductive medicine in primary health care to traditional birth attendants in Peru. Doctoral dissertation. University of California. Davies, J. K. (1977). Multimedia education. In M. Techervarian (Ed.), Community policy for national development; A comparative perspective. Boston: Routledge and Kegan. ! Davis, R. (Ed.). (1980a). Planning education for development: Vol. I, Issues and problems in the planning of education in developing countries. Cambridge, MA: Center for Studies in Education and Development, Harvard University. 278 Davis, R. (1980b). Planning education for development: Vol. II, Models and methods for systematic planning of education. Cambridge, MA: Center for Studies in Education and Development, Harvard University. Development Associates, Inc, (1983). Evaluation report. Sociedad de Assistência a* Maternidade Escola Assis Chateaubriand (SAMEAC), Fortaleza, Ceara*, Brazil, June 14-July 1, 1983. DJeneh, A. (1984). The evaluation of the training and visit system: Lessons from a recent experience in Ethiopia. (Paper No. 178). Cambridge, MA: Harvard Institute for International Development, Harvard University. Douglas, J. (1970). Understanding everyday life. Chicago: Aldine Publishing Co, El Tom, A. R,, Muarek, N., Wesley, E., Mathew, M., & Lauro, D. (1984). Developing the skills of illiterate health workers. World Health Forum, 5, 216-220. Elzinga, A. (1981). Evaluating the evaluation game: On the methodology of project evaluation with special reference to development cooperation. (R: 1). Swedish Agency for Research Cooperation with Developing Countries. Evans, D. R. (1981a). Ghana and Indonesia: Reforms in non-formal education at the community level. Prospects, 11(2), 225-240, Evans, D. R, (1981b). The planning of non-formal education. Paris, France: UNESCO, International Institute for Educational Planning. Evans, J., Lashman-Hall, K,, & Warford, J. (1981). Health care in the developing world: Problems of scarcity and choice. New England Journal of Medicine, 305, 1117-1127. External Evaluation Unit. (1983). Evaluation of family welfare centers, (EEU Series No. 22). Dhaka, Bangladesh: Planning Commission, Government of Bangladesh. Faure, E. (Ed.). (1972). Learning to be. Paris, France: UNESCO. Favin, M., Bradford, B., & Cebula, D. (1984). Improving maternal health in the developing world. White Plains, NY: World ' Federation of Public Health Associations. Feurstein, M. T. (1982). Participatory evaluation: By, with, and for the people. Idea RRDC Bulletin, March, 18-23. 279 Filstead, N. (1981). Qualitative and quantitative information in health policy decision making. Health Policy Quarterly, Spring, 43-56. Fuller, N., & Jordan, B. (1981). Maya women and the birthing period: Postpartum massage-and-binding in Yucatan, Mexico. Medical Anthropology, 5(1), 35-50. General Accounting Office. (1984). Financial management in developing countries reduce the impact of assistance. (Report to the Administrator, Agency for International Development). Gillis, M., Perkins, H., Roemer, M,, & Snodgrass, D. (1983). Economics of development. New York, NY: W. W. Norton and Co. Goodmem, L., & Love, R. (Eds.). (1980). Project planning and management— an integration approach. New York, NY: Pergamon ■ Press. Government of Pakistan. (n. d.). The functions, resources and management of family welfare centers. Lahore, Pakistan: Population and Welfare Division, Government of Pakistan. Gow, D., & Van Sant, J. (1983). Participation in community development. Community Action, 6, 47-51. Grant, J. (1985). The state of the world's children: 1985. Oxford, England: Oxford University Press. Greenberg, J. (1984). Maternal and child health in Bangladesh: A review of the national TBA training program. Bangladesh: UNICEF. Greenberg, L. (1982). Midwife training programs in highland Guatemala. Social Science Medicine. Greenwood, P. W. , Mann, D. , & McLaughlin, M. W. (1975). Federal ; programs supporting educational change: Vol. 3, The process of change. Santa Monica, CA: Rand Corporation. (R-1589/5-H.E.W.) Grosse, R., & Plessas, D. (1984). Counting the cost of primary I health care. World Health Forum, 5, 226-229. iHall, A. (1982). Brazil: Evaluating an A.T. project— appropriately. Idea RRDC Bulletin, March, 30-31. Hemnah, J. P. (1984). Sustaining rural development: A guide for project planners, managers, evaluators, and trainers. Washington, DC: Development Alternatives, Inc. I J 280 : Harbison, F. A. (1973). Human resources as the wealth of nations. Oxford, England: Oxford University Press. Harrison, P. (1983). To bear a child: Meanings and strategies in rural El Salvador. Doctoral dissertation. Catholic University of America. Heaver, R. (1982). Bureaucratic politics and incentives in the management of rural development. (World Bank Staff Working Papers No. 537). Washington, DC: The World Bank. Heaver, R. (1984). Adopting the training and visit system for family planning, health, and nutrition programs. (World Bank Staff Working Papers No. 662). Washington, DC: The World Bank. Heggenhougan, H. K. (1980). Bomohs, doctors, and sinsehs— Medical I pluralism in Malaysia. Social Science and Medicine, 14B, ’ 235-249. Hirshman, A. 0. (1967). The principle of the hiding hand. The Public Interest, 6, 10-23. Holdcraft, L. (1983). Community development: Rise and fall. Community Action, %(6), 41-46. Honadle, G. (1982). Rapid reconnaisance for development administration: Mapping and molding organizational landscapes. World Development, 10(8), 633-649. ' Honadle, G., & Rusengard, J. (1983). Putting projectized development in perspective. Public Administration and Development, 3(4), 299-305. Hoyle, A. R. (1984). Evaluation of training: A review of the literature. Public Administration and Development, 4, 275-282. Huffman, S. (1984). Child spacing for maternal and child health. Mothers and Children, 4(1). Hunte, P., & Sultana, F. (1984). The dai (traditional midwife) in rural Baluchistan: Tradition and change. (UNICEF Research Report). Hjonan, E., & Corl, T. (1984). An information framework for the monitoring and evaluation of appropriate technology projects. Washington, DC: Appropriate Technology International. 'Inventory of dai training. (1983). (Government of Pakistan, Women's Discussion [sponsored by UNICEF]). Islamabad, Pakistan: Bargsons Printers Limited. 281 Isabell, A. (1978). To defend ourselves: Ecology and ritual in an Andean village. Austin, TX: University of Texas. Janowitz, B., Wallace, S., Araujo, G., & Araujo, L. (1985). Referrals by traditional birth attendants in northeast Brazil. American Journal of Public Health, 75(7), 745-748. John Snow, Inc. (1984). Process review of the comprehensive health improvement program— Province specific. Joint Review Team. (1985). Review of the alternative health program and other primary health care activities in Pakistan. (Report of the joint government of Pakistan/WHO/UNICEF/U.S. AID/ CIDA Review Team, 20 November-6 December, 1984). Geneva, Switzerland: WHO. ! Jones, H. (1982). Project appraisal and evaluation: A review of ’ current practice. Idea RRDC Bulletin. March, 3-11. Jordan, B. (1980). Birth in four cultures. Montreal, Canada: Eden Press. Jordan, B. (1984). External cephalic version as an alternative to breech delivery and Caesarean section. Social Science Medicine. 18(8), 637-651. ' Khanna, S. (1979). A study of midwifery practices in rural ; Chandigarh, U.T.. India. Doctoral dissertation, Columbia ' University Teachers College. Kindervatter, S. (1982). Striving for an ideal: The OEF participatory evaluation system. New York, NY: Overseas Education Fund. King, S. (1979). Approaches to evaluation of maternal and child health in the context of primary health care. (Background paper HSM/79.2). Geneva, Switzerland: WHO. King, S. (1983). Primary health care: A response in support of a health revolution. Development, 1, 29-30. Kinsey, D. (1978). Towards evaluation by practitioners in nonformal education. I Kinsey, D. (1981). Participatory evaluation in adult and I nonformal education. Adult Education, ^(3), 155-168. Kinsey, D. C. (1982). Evaluation in nonformal education. (Issues in Nonformal Education No. 3). Amherst, MA: Center for International Education, University of Massachusetts. 282 ' Kleckkowski, B. M., & Nilsson, N. 0. (1984). Health care facility projects in developing areas: Planning, implementation, and operation. (Public Health Paper No. 49). Geneva, Switzerland: WHO. Korten, D. (1980). Community organization and rural development: A learning process approach. Public Administration Review. 40(5), 481-511. LaBelle, T. (Ed.). (1975). Educational alternatives in Latin America. Los Angeles, CA: UCLA Latin American Center. Lele, V. (1975). The design of rural development. Baltimore, MD: Johns Hopkins University Press. i ' Lewis, G. (1980). Planning education for development: Vol. IV, Data analysis for educational plemning. Cambridge, MA; Center j for Studies in Education and Development, Harvard University. ; Mach, E. P., & Abel-Smith, B. (1983). Planning the finances of the health sector: A manual for developing countries. Geneva, Switzerland: WHO. Mahler, H. (1984). Voices in harmony for health for all. WHO I Chronicle, ^(4), 155-160. Management Sciences for Health. (1984). Evaluation: Health promoter program. Vols. 1 & 2. (Ministry of Health, Peru). Boston, MA: Management Sciences for Health. Mangay-Maglacas, A., & Pizurki, H. (Eds.). The traditional birth I attendant in seven countries: Case studies in utilization and training. (Public Health Papers No. 7). Geneva, Switzerland: WHO. Mani, S. B. (1980). A review of midwife training programs in Tamil Nadu. Studies in Family Planning, 11,(12), 395-400. i Mann, S. (1976). Making change happen. Teacher College Record, 77(3), 313-322. Mather, H. N., Jam, T. D., & Apte, V. (1979). Training of traditional birth attendants: Some observations. Journal of Obstetrics and Gynaecology of India, ^(3), 623-626. I McAnany, E. G. (1973). Radio's role in development: Five : strategies of use. (Information Bulletin No. 4). Washington, DC: Information Center on Institutional Technology, Academy for Education Development. 283 McGinn, N., & Galetar, L. (1984). An introduction to strategic planning in education. (Paper No. 163). Cambridge, MA: Harvard Institute for International Development, Harvard University. Miles, M. B. (Ed.). Innovation in education. New York, NY: Bureau of Publications, Teachers College, Columbia University. Morales, F. (1983). Aspects of financing nonformal education. Prospects, ^(7), 55-60. Morey, N., & Luthans, F. (1984). An emic perspective and ethnoscience methods for organizational research. Academy of Management Review, 19(1), 27-36. Morgan, P. (1983). The project orthodoxy in development: Re-evaluating the cutting edge. Public Administration and Development, 3, 329-339. Moris, J. (1984). Institutional choice for rural development. (Paper No. 179). Cambridge, MA: Harvard Institute for International Development, Harvard University. Morisky, D. E. (1977). Evaluating the effectiveness of the utilization of traditional birth attendants (hilots) as family planning motivators in the Philippines. Unpublished masters thesis, Johns Hopkins University. Mosley, H. (1983). Will primary health care reduce infant and child mortality? A critique of some current strategies with special reference to Africa and Asia. Seminar on Health Policy and Mortality Prospects, Paris, France: Institute National d'Etudes Démographiques. Murlius, 0. (1981). An institutional approach to project analysis. Paris, France: Organization for Economic Cooperation and Development. Murray, D. (1983). The World Bank’s perspective on how to improve administration. Public Administration and Development, 3, 291-297. Neumann, A. K., Sai, F., & Dodu, S. (1974). Danfa comprehensive rural health and family planning project: Ghana research and design. Environmental Child Health, February, 39-53. Neumann, A. K., Nicholas, D., Amonooh-Acquah, M. B., Pensah, M., & Boyd, D. (1983). Evaluation of a traditional birth attendant program in Ghana. J 284 Nicholas, D., Ampofo, D., Ofosu-Amaah, S., Asante, R., & Neumann, A. K. (1976). Attitudes and practices of traditional birth attendants in rural Ghana: Implications for training in Africa. Bulletin of the World Health Organization, 54, 343-348. Nisbet, R. (1969). Social change and history: Aspects of the Western theory of development. New York, NY: Oxford University Press. Oakley, P. (1982). Evaluating social development: *How much* or *how good*? Idea RRDC Bulletin, March, 12-17. Ofori-Ansah, P. (1983). External assistance in educational technology transfer to Africa; The role, policies and implementation strategies of the World Bank. Doctoral dissertation, Columbia University. Ofosu-Amaah, V. (1983). National experience in the use of community health workers: A review of current issues and problems. (Offset Publication No. 71). Geneva, Switzerland: WHO. Ofosu-Amaah, S., & Neumann, A. K. (1979). Danfa project final report. Ghana: University of Ghana Medical School (Department of Community Health) and UCLA School of Public Health (Division of Population, Family, and International Health). Ouedraogo, B. L., & Rideout, W. M., Jr. (1980). Traditional modalities of participation and self development: The utilization of indigenous age-set groups to promote rural development in Upper Volta. In R. Kidd & N. Colletta (Eds.), Tradition for development: Indigenous structures and folk media in non-formal education. German Foundation for International Development and International Council for Adult Education. Owen, M. (1983a). Laws and policies affecting the training and practice of traditional birth attendants. International Digest of Health Legislation, M(3), 441-475. Owen, M. (1983b). The traditional birth attendant and the law. World Health Forum, 4. Pakistan health sector report. (1983). (World Bank Document, September 30). N. p.: The World Bank. Parlato, M., & Favin, M. (1982). Primary health care progress and problems, an analysis of 52 AID assisted countries. Washington, DC: American Public Health Association. 285 PATH (Program for Appropriate Technology in Health Care). (1984a). Technologies for pregnancy care. Health Technology Directions, 4(1), 1-12. PATH (Program for Appropriate Technology in Health Care). (1984b). Technologies for safe birth. Health Technology Directions, 4(2), 1-16. Patton, M. Q. (1980). Qualitative evaluation methods. Beverly Hills, CA: Sage Publications. Paul, S. (1983). Strategic management of development programs. (Management Development Series No. 19). Geneva, Switzerland: International Labor Organization. I Paulston, R. (1972). Nonformal education, an annotated ; bibliography. New York, NY: Praeger Publishers. 1 Pederson, B. (1985). A pilot project for training traditional birth attendants. Journal of Nurse Midwifery, ^(1), 43-47. Petty, M. (1985). Five basics for trainers. Training, February, 85-86. Phillips, J., Simmons, R., Chakraborty, C. A., & Chowdhury, A. (1984). Integrating health services into an MCH-FP program: Lessons from Matlab, Bangladesh. Studies in Family Planning. 15(4), 153-161. Phillips, J., Simmons, R., Simmons, F., & Yonus, M. (1984). Transferring health and family planning service innovations to the public sector: An experiment in organizational development in Bangladesh. Studies in Family Planning. 15(2), 62-73. Pietro, D. S. (1983). Evaluation sourcebook for private and voluntary organizations. New York, NY: American Council of Voluntary Agencies for Foreign Service. Pillsbury, B. (1979). Reaching the rural poor: Indigenous health practitioners are there already. (AID Program Evaluation Discussion Paper No. 1). Washington, DC: Agency for International Development, Office of Evaluation, Bureau for Program Policy and Coordination. jPillsbury, B. (1982). Policy and evaluation perspectives on traditional health practitioners in national health care systems. Social Science Medicine. 16, 1825-1834. Population reports. (1980). Traditional midwives and family planning. (Series J, No. 22). Baltimore, MD: Johns Hopkins University. 286 Provus, M. (1971). Discrepancy evaluation. Berkeley, CA: McCutcheon. Pyle, D. (1982). From projects to programs: Structural constraints associated with expansion. (NASPAA Working Paper No. 3). Washington, DC: National Association of Schools of Public Affairs and Administration. Pyle, D., Sabin, E., & Martin, P. (Eds.). (1982). Expanding primary health care from projects to programs. Washington, DC: American Public Health Association. Reich, R. (1985). Toward a new public philosophy. The Atlantic ' Monthly, May, 68-79. Rogers, E. M. (1962). Diffusion of innovations. New York, NY: j MacMillan Co. I Rogers, E., & Danziger, S. (n. d.). Nonformal education and communication technology: The second dimension of development and the little media. Communication Technology: Little Media, pp. 221-237. Rogers, E. M., & Shoemaker, P. (1971). Communication of innovations: a cross cultural approach. New York, NY: The Free Press. I Rogers, E., & Soloman, D. (1975). Traditional midwives and family planning in Asia. Studies in Family Planning. 6(5), 126-133. Rondinelli, D. (1983a). Implementing decentralization programs in Asia: A comparative analysis. Public Administration and Development, 3, 181-108. Rondinelli, D. (1983b). Projects as instruments of development administration: A qualified defense and suggestions. Public Administration and Development, 3(4), 307-328. Rondinelli, D., Nellis, J., & Cheema, J. (1983). Decentralization in developing countries: A review of recent experience. (World ! Bank Staff Working Papers No. 581). Washington, DC: The World Bank. Ruttan, V. (1984). Integrated rural development programs: A historical perspective. World Development, 12(4), 393-401. i Scriven, M. (1972). Pros and cons about goal-free evaluation. (With comments by D. Stufflebeam, M. C. Alkin, W. J. Popham, & G. F. Kneller). Evaluation Comment, 3(4), 1-7. 287 Simpson-Herbert, M. (1981). Traditional midwives: Controversial community health workers. Development Digest. 19(1), 115-128. Stake, R. (1975). Evaluating the arts in education, a responsive approach. Columbus, OH: Charles E. Merrill. Stinson, W. (1984). Potentials and limitations of community financing. World Health Forum. 5, 123-125. Storms, D. (1979). Training and use of auxiliary health workers: Lessons from developing countries. (International Health Programs Monograph Series No. 3). Washington, DC: American Public Health Association. 8tufflebeam, D. (1971). Educational evaluation and decision making. Iltasca: F. E. Peacock Publishers, Inc. Sukkary, S. (1981). She is no stranger: The traditional midwife in Egypt. Medical Anthropology. 5(1). Taylor, C. E. (1975). The place of indigenous medical practitioners in the modernization of health services. In C. Leslie (Ed.), Asian medical systems: A comparative study. Berkeley, CA: University of California Press. Taylor, C. E. (1984). The uses of health systems research. (Public Health Paper No. 78). Geneva, Switzerland: WHO. Tice, T. J. (1985). Grace under pressure. Ambassador. July, 45-57. Tremlett, J. (1983). The traditional midwife and antenatal services in Zambia. Unpublished masters thesis, University of London. True, S. (1983). The human factor in community development. Community Action. 1(6), 11-20. UNESCO. (1982). Country program profile: Ghana. UNESCO. ■United Nations. (1983). Training of traditional birth attendants I (dais) Pakistan 1983-84 to 1985-86. (Children's Fund Project I Summary). United Nations. (1984). Compiling social indicators for women. ! (Studies in Methods, Series F, No. 32). New York, NY: United ' Nations. Uphoff, N., & Ilchman, N. (Eds.). (1972). The political economy of development. Berkeley, CA: University of California Press. _ J 288 U. S. AID. (n. d.). Country development strategy statement (Bangladesh). FY 1986. Washington, DC: U. S. AID. Vaughan, P., Mills, A., & Smith, D. (1984). The importance of decentralized management. World Health Forum. 5, 27-29. Verdose, M., & Turnbill, L. (1975). The traditional birth attendant in maternal and child health and family planning: A guide to her training and utilization. Geneva, Switzerland: WHO. Walt, G. (1984). The supervision of traditional birth attendants. Geneva, Switzerland: WHO. ■ Warwick, D. (1980). Integrating planning and implementation: A transactional approach. In R. Davis (Ed.), Planning education , for development: Vol. I. Issues and problems in the planning of i ! education in developing countries. Cambridge, MA: Center for i ' Studies in Education and Development, Harvard University. ; Warwick, D. (1983). The KAP survey: Dictates of mission versus demands of science. (Paper No. 150). Cambridge, MA: Harvard Institute for International Development, Harvard University. Weiss, C. (1972). Evaluation research: Methods of asessing program effectiveness. Englewood Cliffs, NJ: Prentice Hall, Inc. Wells, S. (1976). Instructional technology in developing . * countries: Decision-making processes in education. New York, | NY: Praeger Publishers. I WHO. (1979a). Traditional birth attendants. (Offset Publication No. 44). Geneva, Switzerland: WHO. WHO. (1979b). Traditional birth attendants: An annotated ! bibliography on their training, utilization and evaluation. | 1 Geneva, Switzerland: WHO. (Supplement I, 1981; Supplement II, 1982) WHO. (1981). Analysis of the content of eight elements of primary i health care. Geneva, Switzerland: WHO. I WHO. (1982a). Development of guidelines for supervision of TBAs j I and for developing a system of monitoring and evaluation of j j trained TBA performance. (Report of an Inter-Country Workshop, | SEARO, New Delhi, 19-26 February) WHO. (1982b). The extension of health service coverage with traditional birth attendants: A decade of progress. Chronicle. 36(3), 92-96. 289 ’ WHO. (1982c). New achievements with traditional birth attendants (TBAs): A review of trends in the way member countries are collaborating with traditional birth attendants to extend the reach of health services. Geneva, Switzerland: WHO. WHO, (1983). Strengthening ministries of health for primary health care. (Offset Publication No. 82). Geneva, Switzerland: WHO. WHO & UNICEF. (1978). Declaration at Alma-Ata. In Primary health care, report of the International Conference on Primary Health Care, Alma-Ata, USSR, 6-12 September, 1978. Geneva, Switzerland: WHO. ( Why retrain traditional birth attendants. (1983). The Lancet, ' January, 223-224. ' Williams, G. (1982). Indonesia: Choosing the right strategy. Idea RRDC Bulletin. March, 24-30. World Bank. (1978). World development report 1978. Oxford, England: Oxford University Press. World Bank. (1980). Education: Sector policy paper. Washington, DC: The World Bank. World Bank. (1983). World development report 1983. Oxford, England: Oxford University Press. World Bank. (1984). World development report 1984. Oxford, England: Oxford University Press. 290 APPENDIX SUGGESTED QUESTIONS FOR STUDY OF KNOWLEDGE. ATTITUDES AND PRACTICES OF TBAs NAME A. MATERNAL HEALTH SECTION 1. How do you te 11 that a woman Is pregnant? 2. Do you advise a woman to see you In the months before she delivers her baby? 2a. If so, at what month for the first visit? 2b. If so, how often? 3. Do you advise the pregnant woman to eat any foods during her pregnancy? 3a. If so, what foods? 4. Do you advise the pregnant woman not to eat certain foods? * 4a. If so, what foods? 5. Do you prescribe any medicines for the woman to take during her pregnancy? 5a. If so, what medicines and for what purposes? 6. What else do you tell a woman to do or not do during her pregnancy? 7. How long does a pregnancy last? 8. Why do some women bleed from the vagina In the first few months of their pregnancy? 9. What do you do for It? 10. Why do some women bleed from the vagina In the last few months of their pregnancy? 11. What do you do for It? 12. Why do some women vomit excessively during their pregnancy? 13. What do you do for It? | 14. Why do some women have fits during pregnancy? 15. What do you do for It? U6. Why do some women have excessive weakness during pregnancy? , 17. What do you do for It? 18. Why do some women have swelling of the legs during pregnancy? 291 19. What do you do for it? 20. Why do some women get fever during their pregnancies? 21. What do you do for it? 22. How do you tel I that labour has begun? 23. What medicines do you give during labour? 24. How many hours should labour last for a woman delivering her first child? 25. When do you consider such a labour to be abnormally prolonged? 25a. Why do some women have long, difficult labour? 26. What do you do for It? 27. How many hours should labour last for a woman who has delivered a child before? 28. When do you consider such a labour abnormally prolonged? 29. Why do some women have heavy bleeding during labour? 30. What do you do for it? 31. Why do some women have fits during labour? 32. What do you do for it? 33. Why do some women have tearing of the perineum during delivery? 34. What do you do to prevent It? 35. 'What do you do if it occurs? 36. Why do some women get a high fever after del Ivery? 37. What do you do for it? 38. When do you consider bleeding after delivery to be abnormal? 39. What do you do for it? 40. What medicines do you prescribe for the woman after delivery and for what purpose? 41. Do you ever advise a woman to postpone getting pregnant? 41a. If so, under what circumstances? 42. Do you think women should space their pregnancies? 42a. If so, why? 43. Many couples do something to postpone pregnancy so that they can have just the number of children they want and can have them when they want them. How do you feel about this? Would you say that you approve, disapprove or feel uncertain about this kind of family planning? 44. Why do you feel this way? L 292 .45. Do you know about any ways or methods that people might use to avoid or postpone a pregnancy? 45a. (If Yes) - What methods? 46. Here are some methods some couples use to avoid or postpone a pregnancy. Which ones have you heard of before? 1. Abstinence (living apart) 2. Prolonged breast feeding 3. Rhythm (safe period) 4. Withdrawal 5. Condom (sultan) 6. Foam (Emko) 7. Pill 8. lUD (loop, ’ ’rubber”) 9. Sterilization 10. Other 47. Do you think women in this village would be interested in learning I about family planning? > 48. Why do you think that? 49. If women were to begin using family planning, would it affect your work in any way? B. CHILD CARE SECTION I. With what instrument do you cut the cord after birth? t i 2. Is the Instrument treated or cleaned in any way prior to being used I to cut the cord? 2a. If so, how? 3. Do you put any dressing on the cord? 3a. If so, what do you use as the dressing? 4. Why do some babies have a hard time breathing at birth? 5. What do you do for this condition? 6. Why do some newborn babies get fever? 7. What do you do for this fever? 8. Have you ever seen an infant develop a disease in the first two weeks after birth where the following occur: the infant.becomes rigid, the head is bent back all the time and the infant gets f.its. 9. Why do some infants get this condition? 10. What do you do for it? II. Do you continue to care for or advise mothers on the care of infants and children after the period just following birth? 12. If so, up until what age? 13. After six months of age,why do some children become thin, lose weight, or not grow proper Iy? 293 14. What do you advise a mother to do when her child becomes thin, loses weight and does not grow properly? 15. When do you advise a mother to add foods to the infant’s diet in addition to breast-milk? 16. What foods do you advise her to add? 17. How long should a woman breastfeed her child? C. GENERAL INFORMATION SECTION '1. How do women go about seeking services? 2. How do women pay you for your services? 3. Do you know of any government or health centre midwives in your area? 4. Does their work differ In any way from your midwifery work? 4a. If so, how? 5. Would you be interested in seeing the kind of work they do? 6. Would you be Interested in receiving any training in some modern methods of midwifery? 7. Do you think you might be willing to participate with the staff of the health centre in a programme' in helping mothers to have easier deliveries and healthier babies? Questions for the KAP can be added or deleted as required by each training programme. Note: From Danfa Project (1977. A program manual for traditional birth attendants: Organization, training, and evaluation. [Joint project of the Department of Community Health, Ghana Medical School, and the School of Public Health, University of California, Los Angeles]).</u>

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Allied Health Professions Admission Test : its role in selection for physical therapy programs

PDF

Self-perceived job satisfaction and dissatisfaction of urban elementary classroom teachers

PDF

The 'Widmar' (1981) decision and the status of religious liberty on the public college and university campus: Antecedents, effects, and prospects

PDF

Studies of two naturally occurring compounds which effect release of acetylcholine from synaptosomes

PDF

Numerical analysis of ecological survey data

PDF

The interaction of learners' spatial ability aptitudes with filmic schematic operations

PDF

Resocialization of older adults to work roles through worklife autobiography

PDF

The innervation of the teeth and periodontium of the rat

PDF

Cation localization in growth hormone and prolactin cells of the rat pars distalis

PDF

A study of Japanese students at the University of Southern California, 1946-1980: vocational impact of American academic experience on Japanese students after returning to Japan

PDF

An analysis of the need and opportunity for the continuing liberal education of doctors of medicine

PDF

The role of microbes in the formation of the tubeworm fouling community in Fish Harbor, Los Angeles Harbor, California

PDF

An analysis of a humanistic and existential oriented doctoral counselor training program

PDF

The relationship between cultural-value orientation and achievement in three learning environments: Lecture, small-group, and self-study

PDF

The orientation toward employment of Taiwanese graduate students in the United States

PDF

Out of many...: A social history of the homosexual rights movement as originated and continued in Los Angeles, California

PDF

Nude mouse grown human nephroblastomas

PDF

Studies of frog retinal ganglion cell responses to temporal, spatial and chromatic stimuli

PDF

Chemical and immunological studies on a heterophile antigen extracted from bovine erythrocyte stroma

PDF

Studies on the aerobic oxidation of fatty acids by bacteria

Asset Metadata

Creator Grieser, Richard Powell (author)

Core Title An analysis of traditional birth attendant (TBA) programs in selected third world countries

School Graduate School

Degree Doctor of Philosophy

Degree Program Education

Degree Conferral Date 1985-12

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Education,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c30-282169

Unique identifier UC11226811

Identifier DP25048.pdf (filename),usctheses-c30-282169 (legacy record id)

Legacy Identifier DP25048.pdf

Dmrecord 282169

Document Type Dissertation

Rights Grieser, Richard Powell

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA