This s e r i e s of unreferred dis~:ussion pap 1 s is intended to provide a? i n i t i a l circulation of work prepared under the auspices of the Poptilation and Human Resources Division. Commen a r e therefore very welccme. Ifie papers should not be cited without clearance by the authors. The views expressed are not necessarily those of the Vorld Bank Group. Population and Human Resources Division t Discuseion Paper So. 81-62 BENEFITS OF RJTEGMTIYG FA.FULY PLANNING WITH HEALTH SERVICES: TLE NARXNGWAL E!!BRDENT December 1981 Prepared by: , I Carl Taylor, i7.S .3. Sarna, Robert park^ i and a 'Jillian Beir..ie (Johns Hopkins Gnivbrsity) R s h i d Faruqee (Tie 'A'orld Bank) Prepared f o r: Xesearch Project 671-38, Narazgdal P ~ ~ u l s t i aend Xutriticn Contents PagecNo. The Narangwal Teams v i i Preface x i PART I: THE W G W A L EXPERIMENT . Chapter 1 Sackground, Design, and Pnlicy I s s u e s C a r l E. Taylor Chapter 2 ,Hain Research Findings on Policy I s s u e s C ~ r E. l Taylor, W i l l i a mA.Seinke Rashid Faruqee, Robert L. Parker and R.S.S. Sarna PART 11: THE DETAILED FINDINGS Chapter 3 The Use of Family Planning R.S. S. Sarma and Rashid Faruqee Chapter 4 The Use of Health Services Robert L. Parker and William A. Reinke Chapter 5 RcLations between t h e Use of Health S e r v i c e s and Family Planning William A'. Reinke and Robert L. Parker Chapter 6 Determinants of the Use of Family Planning t Robert L. Parker, William A. Reinke R.S.S. Sarna and Narindar U. Kelly Chapter 7 Z Determinants of F e r t i l i t y and Its Decline 's -- i Rashid Faruqee and R.S.S. Sarna Chapter 8 ) E f f i c i e n c y and Equity o f Services - Robert L. Parker and R.S.S. Sarma A p p e n d i ~A Data C o l l e c t i o n Methodology Appendix B 3 e s c r i p t i o n of Services Prcvided i n the Narangwal Population Study Table 1.1 Selected Demographic, Socioeconomic and Other C h a r a c t e r i s t i c s of t h e Experimental Groups of Villages, 1971 27 Table 1.2 The L i t e r a t e Populatiozi i n Narangwal Villages: Indian Government Census 1961 and 1971 28 Table 1.3 Cross- Sectional Surveys, Longitudinal Data Collection, and Service Input Information Table 1.4 Summary o f Service I n p u t s Table 3.1 Percentage D i s t r i b u t i o n of Current Users of Family Planning by Nethod Currently Being Used on March 31 of Each Calendar Year by Experimental G r ~ u p Table 3.2 Pregnancy Rate and Relative Effectiveness of Contraceptives Table 3.3 Knowledge and P r a c t i c e o f Family Planclng before the P r o j e c t Table 3.4 T r a n s i t i o n from P r i o r Use t o Project- U s e by Type of Contra- c e p t i v e Methods i n the FP-Service V i l l a g e s Table 3.5 Sociodemographic 3 i f ferences i n T r a n s i t i o n from Preproject t o P r o j e c t P r a c t i c e of Contraception Table 3.6 D i s t r i b u t i o n c f Women by Their Preproject Response Regarding A t t i t u d e s and B e l i e f s Towards Family Planning and About Child Mortality and Their Contraceptive Behavior d u r i ~ : ~ , the P r o j e c t Table 4.1 Relationship between Volume of Children's Services and Change i n A t t i t u d e s and B e l i e f s I t ~ a b l e ~ 4 . 2 E f f e c t s of Children's Services on Mortality Rates, by Age, Caste, and Experimental Group, 1970-73 Table. 4.3 E f f e c t s of Children's S e e i c e s on Morbidity Levels, 1970-73 '2 Table 5.1 Relationship between ~ r o j k Use of Health Services and t Acceptance of Family Plazning * Table 5.2 a e l a t i o n s h i p between Famiry P l a n n i ~ gP r a c t i c e and Average Number of i i ~ a l t hService Contacts during E a r l i e r and L a t e r Portions of t h e Project Table 5.3 Relationship o f Family Planning P r a c t i c e before t h e P r o j e c t to Use of Health Services during t h e P r o j e c t 152 Table 5.4 Combined E f f e c t of P r i o r and Project Family Planning P r a c t i c e on Use of Health Services i n 1971 155 Page No. Table 5.5 Relationship of Project Family Planning Practice t o P r i o r Family Planning P r a c t i c e and Use of Health Services, 1971 157 Table 5.6 Estimated Net Effect of Health Services on Project Use of Family Planning 162 Table 6.1 Results of the Multiple Regression Analysis Examining Service U t i l i z a t i o 3 i n 1969-1971 and its E f f e c t on Use o r Nonuse o f Family Planning at Anytime during the Project, 1969-73 Table 6.2 Adjustment Factors f o r S i g n i f i c a n t Variables i n t h e Regression Analysia i n Table 6.1 Table 6.3 Summary of E f f e c t s on Contraceptive Use Table 6.4 Results of the K u l t i p l e Regression Analysis Examining the Volume of Service Contacts i n 1969-73 and its E f f e c t on Use or Nonuse of Family Planning a t Any Time during the P r o j e c t , 1969-73 Table 6.5 Mean Number of Services and Associated Regression C o e f f i c i e n t s from the Regression Analysis i n Table 6.4 Table 6.6 Results of the Mu1t i p l e Regression Analysis Examining the Volume of Service Contacts i n 1969-73 and its Effect on the Duration of Family Planning Use Among A l l Project Users Table 6.7 Regression Analyses f o r Experimental Groups of Villages Table 7.1 Age- Specific 5-Year Period Marital F e r t i l i t y Rates by Major Sikh Caste Groups, Narangwal Study Villages, 1949-1968 Table 7.2 Women's Age at Marriage and Age a t Cohabitation by Harriage Cohort, Narangwal Study Villages Table 7.3 Hean Kumber of Live Births Per Married Woman, F e t a l Death Rate, and Child-boss Rate, by Age Table 7.4 islean Number of Live- Births Per Married Woman by Age -- and Religion- Caste Table 7.5 Mean tiumber of Live- Births Per Married Women by Age'i - and Education of Husband -- Table 7.6 Sonparison of Average Number of Surviving Children +th Average Number of Children Desired by Continuously ,Married Women a t Original and Reinterview by Sex of Child and Age Cohort of Wife, Narangwal, India Table 7.7 Regression Coefficients and Their Levels of Significance Table 7.6 Percentage of Women who had a Birth i n a Calendar Year, bjr P r a c t i c e of Contraception the Preceding Year Page No. Table 7.9 Percentage of Waen who had a Birth, by Practice of Family Planning i n Preceding Year and Religioa-Caate 225 Table 7.10 Percentage of Women who had a Birth, by Practice of Family Planning i n Preceding Year and Education of Husband Table 7.11 Percentage of Womln vho had a Live Birth, by Practice of Family Planning i n Preceding Year and Occupation of Husband Table 7.12 Relation of PtobabiLity of Birth i n a Calendar Year to Selected Variables Table 7.13 Annual HarLtal F e r t i l i t y Rates, by Experimental Group and Village, 1970-73 Table 7.14 Crude Birth h i e s , by Experimental Group, 1970-73 Table 7.15 Percentage of Womerl who had a Birth, by Practice of Family Planning i n Precetiing Year and Experimental Group Table 7.16 Regression Results with the Dependent Variable: Percentage Cecline i n Average h n u a l Probability of Birth Table 7.17 Hean Number of Service Contacts by Type and Percentage Decline in Fertt1.it y Table 8.1 Distribution of Costs by Specific Categories of Expenditures i n Each Experimental Group Based on Average Annual Costs Table 8.2 Averags h n u a i Cost per Capita f o r Different Services i n Each Experhental Gror~p Table 8.3 Average Cost per Service Contact f o r Different Services i n Each Experimental Group Table 8.4 Estimated Annual Per Capita Expenditure on Xealth by Source of Care f o r Each Experimental Group =;A ControA Villages ' (rupees) Table 8.5 P e r c e ~ t a g eof Families with Various Socioeconomic Character- i s t i c s Using Curirtive Services f o r Women o r Children during the Project, 1969-73 Table 8.6 Percontagc of Families with Various Socioeconomic Character- i s t l c s Using Other Services f@rWomen o r Children during the Project, 1969-73 . Comparison of Preproject Use (Modern Temporary and Permanent Hethods) and Project Use Rates, by Sociodemographic Group Table 8.8 Sociodemographic Differences i n Recruitment to Project of Previous Nonusers of Fawily Planning Page No. Table B.l Effective Starcing Dates of the Major Service Components i n Study Villages and Experimental Groups 292 Table 8.2 Coordination of Routine Children's and Women's Services a f t e r the Neonatal Period 296 FIGURES Figure 1.1 Map Showing Experimental Groups of Villages 19 Figure 1.2 Experimental Design f o r the Narangwal Population and Nutrition Projects 22 Figure 1.3 Outputs (Behavioral Factors) 35 Figure 2.1 Sequence of Organizing Integrated Prlmary Care fo Health, : Population and Nutrition 64 Figure 2.2 Fourteen Health-Service Entry Points f o r Family Planning & t i - ~ a t i o n a lActivities 67 Figure 3.1 Percentage of Currently Married Women Aged 15-49 who were Acceptors of Family Planning a t Specified Points i n Time by Experimental Group 76 Figure 3.2 Cumulative Rates of New Acceptors i n Experimental Groups 77 Figure 3.3 Percentage of Currently Married Women Aged 15-49 who were Practicing Family Planning by Experimental Group Figure 3.4 Trends i n Contraceptive Practice Rates i n Experimental Groups Figure 3.5' ' Percent Currently Harried'Women Aged 15-49 who were "Effective Users" of Family Planning a t Specified Points i n Time by Experimental Group 85 - +- Figure 3.6 Trends i n Effective- User Rates i n Experimental Groups 87 ' - i Figure 3.7 Preprogram Use of Permanent, Modern-Temporary and Traditional Methods among Experimental Groups s I Figure 4.1 Average Number of Recorded Service Contacts Made by A l l Staff Working i n Each Experimental Group Figure 4.2 Percent of Ill Individuals Receiving Treatment from Different Sources of Care i n Each Experimental Group, 1973-74 Page No. Figure 4.3 Effect of Children's Services on Average Weight, Adjusted f o r Sex and Caste, Expreesed a s a Percentage of the Hamarc? Median Weight Standard, 1970-73 134 Figure 4,4 Effect of Children's Services on Average Height, Adjusted f o r Sex and Caste, Expresaed a s a Percentage of the Hamard Median Eeight Standard, 197073 136 Figure 5-1 Health Sarvices aud Family Planning Use by Experimental Group Over Time 142 Figure 5.2 Graph of Relatlonships between Program Family Planning, Prior Family Planning and Health Service Use 160 Figure 6..1 Bivariate Relationship between Contraceptive Use and Selected Explanatory Variables 179 Figure 6.2 Bivariate Relationship betveen Contraceptive Use and the Volume of Service Contacts i k Figure 6.3 Bivariate Relationship between Duration of Contraceptive I Use and the Volume of Service Contacts 189 Figure 7.1 Percentage of Wives, Aged 15-39 Years, Wanting More Children by Number of Living Daughters and Sons, Narangwal, India Figure 7.2 Percent Women who Had Birth i n a Calendar Year by Previous Parity and Practice of Family Planning i n Previous Year 221 Figure 7.3 F e i t i l i t y Decliae and Its Relationship with Services 246 Figure 8.1 Average Time Spent Providing Direct Services to A l l Staff Working Ln Each Experimental Group, 1971 251 . I Figure 8.2 Average Time Spent Providing Direct Services by Family Health Workers i n FPWSCC, FFWS, FPCC Villages, and F a t l y Planning Ed-~catorsi n the FPED Villages, 1974 - - 254 '? i - v i i - - The Narangwal Teams e a r l E. Taylor and R.D. Singh -- Principal Investigators and . Senyukta Vohra, Inder S. Uberoi, Susheila B. Takulia, Eiarbans S. Takulia, R. S. Tak, G. Subbulakehmi, K. Sivaram, R.S.S. S a m , Kailash S h a m , William A. Reinke, Robert L. Parker, A.K.S. Hurthy, Colin W. HcCord, Norah Masih, Dolores I.aliberte, Ranjit Krmrar, N. Sengupta Kielmann, Arnfried A. i Kielmann, Nandini N. Khosla, D. N. Kakar, Ruth K. J u l i u s , Kamlesh S. J a i n , Jeanne A. G r i f f i t h , Alice H. Forman, E l s i e Ferguson, Donald C. Ferguson, t Shridhar Duivedi, Cecile DeSwemer, R. Bruce Conyngham, Sush*m Bhatia, 5, Thomas E.C. Barns, and Dee Raj Arora. I i Analysis and Report Preparation E Carl E. Taylor, R.S.S. Sa~ma,William A. Keinke, Robert C. Parker, and Ra~shidFaruqec- Editors and D. Storms, K. Sivaram, Jeanne S. Newman, Frederic .4. Nass, A.K.S. Murthy, Ranjit Kumar, Arnfried A. Kielmam, Narindar U. Kelly, P.L. Grover, Cecile DeSveemer, Sushum Bhatia. assisted by Hartha S. L i s t , A. .Ula Crawfox:d, Carol A. Buckley, and Peggy E. Bremer. 9' :'" i.p- i.,. Si rk$ P $@ S q k Famflv Health Workers Shashi K. Wasti Hadalsa H. Khera Veena Vema Satnun Kaur Usha Trivedi Halkit Kaur Sudershan Trikha J a g j i t Kaur Kanchan S o d Gurmit Kaur Bij a y Solomon Gurcharan Kaur Krishna G. Singla Chitranjan Kaur Kamla Sikka Amar Kauru Rajinder Sharma Sudesh h n t a S a r l a Sehgal Amarjit h k a r i a d Kamlesh Sahnl Mavis Jatiram Daljit K. Sachdeva Surinder K. Grewal Mary P. Peter armail K. Dhillon S h e i l a P a t i l S a r l a Devi Harjinder Narang Pushpa Dawar Nirmal S. Moudgil '&mini W. Dass Inder Nohini Roselyne R. Daniel Bayant K. Mayo Veena Chopra Zarina D. Masih Unnila Chopra Veena Masih Raj K. Chib Shanti B. Masih Virbala Chhabra Grace W. Masih Sheela Chand h r j i t Mangat Svaran Chadha Vidya W. Malik Shakti B. Chadha V i j a y Malho tra Raj D. Babbar P e t e r Ganda Ma1 Surinder K. Arora Surinder hakkar Sampuras b o r a .. Surinder Kwatra Harinder K. Arora Vijay Khera Family Planning Educators Laboratory Technicians Sukhdev Kaur Prem P r a s s e r Naohhatar Kaur Fazal Masih Harbinder Kaur Dhrshana Kumari -Family ~ e a f t hSupervisors Rupinder Brar Savita S e t h i I Family Planning Workers Kamlesh Sahni N i r m a l P u r l S u r j i t Singh Promilla M. Nanda Sukhdev Singh J e e t Kaur Kushdev Singh A m a r j i t E. Gandhi Sukhdev S. Greual Swarn Dogra G a j jan S. Grewal Surinder Bala Kuldip S. Ghumarl Ramesh C. Kaushal Family Planning Supervisor Neena K. Narula F i e l d Interviewers and I a v e e t i g a t o r s Data Processing S t a f f .Wadyal Arpana Wattae Sukhwsnt K. Walia Kailash Vem Surinder S Darshan S. Surah J a t i n d e r K. Wadyal Harn?k S. Sundra Jagdish Verma Sunita Sood Sushma Uberoi Manjula Sood Verna Singh S u r j i t Singh Rulda Singh Meua Singh Mukhtiar Singh Jaswant Singh k h a n Singh Harnek Singh Malkiat Singh Gobind Singh Curcharan Singh Darshan Singh S.B. La1 Sherry Balwant Singh A.K. Seth B?.ldev Singh Nilu Sapra S.D. S i L r i R.K.V. Rao U s h a Sharma Kulwant Rai R.D. Sharma Neela Yurthy Kusum L. S h a m Shashi Hennan J.P. Sharna Shivcharan K. Mangat Bindu Saxeaa V.K. Malhotra Surinder S a i n i Xanohar Khosla Shashi Sabharwal Kulwant Kaur Bela R a n i Ramesh C. Kashyap Nirmal Puthia Indra Kalrar Ramesh Monqia D.C. J o s h i Ta jinder kianocha S u r j i t S. Greual Prem Malik R.C. Gupta Manohar L. Sharma A.K. Gupta Madan Khoslz Usha Goyal Sukhdev L. Kaushal D.C. Garg Sur jit Kaur Vinod Dhir Sukhwant Kaur P h i l i p Dayal Manjit Kaur Madan Chakravarty Gumat P. Kaur Indu Bhatia Balwinder Kahlon Krishna Bhambri Cameron James Iftikhar Baig Jagmohan S. Hans S u s h a n Gupta Nirmal Gupta -F i ~ l dSupervisors Manjit Grewal Jaswant S. Grewal Amrik M. Seerha G u m a i l K. Grewal Satinder P r a q a k a r Xarveen Goindir Manjit A. Zumar R i t a K. Gawari B a ljit J. Haqs Reena Gandhi B. B. Garg Rajeshwari Dwivedi Suman Dhir Shobha Chhabra S a l i l c ~ hDarshan Parminder Cheema Hardeep Chandha Kanta Bhatia Xdarsh Bhatia Biochemists- Nutritionists -c S o i t l S c i e n t i s t s G. Scbbulakshmi Harbaas S. Takulia N. Sengupta Kielmann R. S. Tak B i d a D. Arora Alan L. Sorkin R.D. Singh Public Health Nurses Jeanaz S. Newman Ran jit Kumar Indet jit Walia D.N. Kskar Senjukta Vbhra Swaran L. Gupta Norah Maaih Prakaeh L. Grover Harbhajan K. Malhotra Donald C. Ferguson Dolores L a l i b e r t e Raahid Faruqee Ruth K. J u l i u s Shridhar Dwivedi Alice Forman Jagdish Bhatia S h i r l e ) 23.Bohnert Des Raj Arora Physicians Statisticians- Programmers Inder S. Uberoi K. Sivaram Carl E. Taylor R.S.S. Sarma Sushiela B. Takulia Satya P. Sangal N i r m a l Sharma D. A l l i e t e t Robertson Kailash S h a m W i i l i a m A. Reinke Aari Shankar Be:t y Parker Robert L. Parker Thomas Palmer Charlotte Neumann Fredetic A. N-ss Alfred K. Neumann A.K.S. Murthy Colin W. McCord Kamlesh S. J a i n Arnfried A. --delmann George Immerwahr Nandini N. Khcdla E l s i e Ferguson J a s b i r Kaur Larry Curcj.0 Har jinder Kaur William J. Blot Alfred M. Haynes Kamlesh Gupta Administrators Jeanne A. G r i f f i t h t Cecile D e Sweemer Shamsher Singh R. Bruce Conyngham Jamison Mzredith Sushum Bhatia-' Ramesh LYannan Thomas E. C. .Barns J u l i a Detato Joseph D. Altar Colvin Dayal Sam A. David - K. G. Bhambri Although many o t h e r administrative and support s t a f f a r e not l i s t e d , we would l i k e t o recognize t h e i r important coutribution t o t h e project. They i n c l c d e secretaries, clerical and account? staff, construction and maintenance staff, watchmen, v i l l a g e attendants, d r i v e r s , and mechanics. - x i - Preface Narangwal is more than the name of a village idpunjab. It alco is the name for a new vay of ~ ~ l v i noldg probleme of village health. The Narangwal . research projects have explored the frontiers of what primary health care can do for poor and deprived people throughout the vorld. The research produced sone of the most inportant evidence so f a r on how t o organize effective pri- mary health care and cn the need f o r integrating service6 at the periphery. It showed how auxiliary-based health care can be provided for an annual expenditure of l e s s than $2 a person. The research was adapted and devi~opedi n a design that shows what can be done i n controlled f i e l d t r i a l & . By measuricg the inputs of services and the outputs and outcomee of improved health and family planning, it was possible to calculate the effect and cost- effectiveness of different intervcrr tions. The findings should help national health p l a ~ e r ai n t h e i r judgments of probable costs and benefits. Our experience indicqtea, however, that t r i a l s a r e needed i n each nev area--to adapt what was learn=. a t Narangwal to local . conditions. The Narangwal population project grew out of long invoivement i n the l i f e and problems of village India. Members of the s t a f f spent many years working i n the villages and living i n village homes with simple hygienic 'C improvements that would demonstrate h e a l t h f h living. - The depth of understand- ing that ccmes from sharing the l i f e of vi-gers - cannot be matched by sophie- ticated data gathering. - x i i - The more formal origins of the population project derive from the research a c t i v i t i e s of the Naran~walRural Health Research Center. I n two adjacent villages--Bara (big Narangwal) with 1,800 people; and Chota ( l i t t l e Narangwal) with 800 people--the research center was s e t up i n 1961 i n collaboration with t.5e Indian Council of Medical Research (ICMR). I t is near a teaching health center, started by the Ludhiana Christian Medical College r i n 1955 t o provide a base f o r r u r a l internship training. The f i r s t project of the Rural Health Research Center was a study of the r u r a l orientation of physicians (Taylor and others 1976). That led t o a s e r i e s of studies on indigenous practitioners and on the b e l i e f s of v i l l a g e people about d i e t and disease. A research project w a s then undertaken to develop a method f o r measuring hzalth needs and resources f o r primary c a r e (Johns Hopkins 1970). With t h i s background information as the basis, two research projects were started on the two health problems that seemed most important i n village India. One was the Narzngual population study reported here. The other was a similar controlled experiment on the interactions between malnutrition and infection i n children of weaning age. The nutrition project included a l l children under three i n four groups of villages: one group received nutrition- a l surveillance and supplements; another was provided health care emphasizing infection control through immunizations and earl.y diagnosis and treatment; the . t third received both n u t r i t i o n and health care measures; and the fourth conti- nued to receive routine government services and served a s a control. -- The use of p$ary-care a u x i l i a r i e s led to dramatic improvements i n growth, development, - mortazity, and mcrbidity. The n u t r i t i o n and population projects were developd . e i n pa?allel, and the combined nutritit--. and healrh care group of v i l l a g e s f o r the n u t r i t i o n project was also tne c k ~ care and family planning group of d v i l l a g e s f o r the population project.. The methods that evolved f o r providing services a r e summarized i n two f i e l d manuals f b r v i l l a g e s i l i a r i e s on child care and child n u t r i t i o n (Uberoi and others 1974; De Sweuner and others The i n i t i a t i v e for t h i s reaearch came f r m the Indian Ministry of Health and Family Planning. National leaders knew that balanced and continuing long- ten. development of services needed f i e l d research on how t o make family planning a c t i v i t i e s an effec-tive part of routine health services. The Ministry of Health and Family Planning delegated responsibility f o r the oversight and surveillance of the Narangwal projects t o the ICXR so that the research ..odd become part of the national research e f f o r t . Annual reports were made t o the ICUR t o keep research i n l i n e v i t h national objectives. Direct comm~micationsvere also maintained with o f f i c i a l s i n the M n i s t r y of Health and Family Planning, and there were numerous s i t e v i s i t s and con- ferences a t Narangwal throughout the research. The f i r s t f e a s i b i l i t y funding w a s from a PL-480 grant by the U. S. Department of Health, Education, and Welfare. The project vas greatly expanded with long-term financing from the U.S. Agency f o r International Development. Funds were also contributed by the World Health Organization (891181122 and H9/181/23). Most important, , I the ICMR provided d i r e c t research grants a f t e r the PL-480 grant ended, indi- cating ..he commitment of the Indian government to t h i s research. - - .dhen the ~ a r a n ~ w @studies began, o f f i c i a l s in the Ministry of Health and the research woTkers agreed that the second stage would be t o have 4 demonstration projects that-apply Narangval methods and findings, adapted to local conditions, i n other parts of India. Several such projects have - x i v - succeeded i n the y e a r s s i n c e the Narangwal f i e l d work stopped. They show t h e value of community p a r t i c i p a t i o n and the ways a h e a l t h team can help people solve t h e i r own h e a l t h problems. As the n a t i o n a l program of t r a i n i n g community h e a l t h w r k e r s develops i n India, there w i l l have t o be a continuing stimulus from demonstration and research projects, And the focus w i l l have t o be sharpened i f i n t e g r a t e d packages of primary h e a l t h care, family planning, and n u t r i t i o n a r e t o begin t o improve the q u a l i t y of l i f e of the r u r a l poor. The findings of the Narangval research w i l l be important i n supporting t h i s continuing e f f o r t . I n expressing appreciation t o those who made the Narangwal population project possible, we must f i r s t recognize a long and distinguished sequence of Indian government o f f i c i a l s . The decisions t.. do the research were by the M n i s t r y of Health and the ICMR. The c r e d i t f o r sponsoring the research and bringing i t to f r u i t i o n must go t o the directors- general of the ICMR-- s t a r t e d under Professor Wahi, continued under D r . Gopalan, and e s p e c i a l l y helped by Prof.=ssor Ramalingaswami. With t h e i r colleagues a t the IC.W, t h y - provide1 i n t e l l e c t u a l and administr=cive support t h a t went f a r beyond thz o f f i c i a l requirements of l i a i s o n with the research team. ' Many o f f i c i % l s of the Ministry of Health helped i n times of' uncertainty. A t t h e Narangwal conferences, l e a d e r s i n academic i n s t i t u t i o n s and i n c e n t r a l and state governments provided i n s i g h t s &out what should and s i coul-d be done and about how t o i n t e r p r e t t h e findings. S p e c i a l mention must - be F d e of o f f i c i a l s i n the Tunjab Health services--incgding t h e h e a l t h s e c r e t a r i e s , d i r e c t o r s of h e a l t h s e r v i c e s , s t a t e o f f i c i a l s i n Chandigarh, X V - and d i s t r i c t o f f i c i a l s i n Ludhiana-*ho helped i n ways ranging f- -m selecting study areas and seconding s t a f f t o discussing a t length how the research could be made relevant. Many colleagues i n academic and research i n s t i t u t i o n s participated i n the f i e l d research, especially those from the Ludhiana Christian Medical College, A l l India I n s t i t u t e of Medical Sciences, and the Chandigarh Pocc Graduate I n s t i t u t e . We also thank the many agencies that supported the research. We got ~*.ctrneededfinancial support a t a c r i t i c a l stage from the ICMR. A t the U.S. Agency f o r International Developent, the Asia Bureau made the decision t o pay f o r the f i e i d w r k , and the Office of Program Planning and Coordination l a t e r supported the analysicr. We received gravts from the special program of the WHO f o r research on human reproduction and from the Merrill Fund. The U.S. National I n s t i t u t e of Health assisted i n the analysis. And the Develop- ment Policy Staff of the World Bark, especially Timothy King and Ravi Gulhati, supported the l a s t stage of analysis that led to t h i s book. Those of us who have had the privilege of producing t h i s analysis and report f e e l a special debt of gratitude to our many colleagues who shared the f i e l d work. Listing t h e i r names and roles cannot give enough recognition of the pride and e f f o r t that made the f i e l d experience so productive. But the 16ng list a t the front of t h i s book indicates theeimportance of the contribu- tions by a large number of researchers and f i e l d s t a f f over the l i f e of the project. Starting with f i f t e e n or so, the s t a f f grew t o about 150 by the end of the project; a t any time during the project an average of two professionals from six countri-es s t h e r than $dia were 'on the f i e l d s t a f f i n Narangwal. The excitsuent of l i v i n g and working together i n the villages strengthened the t i e s of collaboration. And a s individuals moved t o new positions during o r a f t e r the project, they carried an imprint of t h i s experience. A l l i n the Narangwal family know t h a t l i f e brings few opportunities t o m r k with so congenial and dedicated a group. Our deepest thanks go to our mnderful hosts i n the villages. Their hospitality made us f e e l that the v i l l a g e s were our homes. We often ,iden- t i f i e d so completely v i t h t h e i r joys and concerns t h a t we worried about ~ e i e n - t i f i c objectivity; t u t because ccnmunity participation c a l l s f o r seeing things through the eyes of l o c a l people, t h i s identification became an asset. With patience and goodwill they welco~aedour research a c t i v i t i e s . Their under- standing of the reasons f o r the research led to t h e i r pleasure i n knowing that lessons from t h i s work might help programs i n v i l l a g e s elsewhere i n India and the world. To our v i l l a g e friends, we express our deepest gratitude-for the shared hope that t h i s book can help improve the quality of l i f e of the world's neediest people. Carl E. Taylor R.S.S. Sarma William A. Reinke Robert L. Parker Rashid Faruqee PART I THE W C ; W A L EXPERIMENT Chapter 1 - Carl E. Taylor Background, Design, and Policy Issues The f i r s t thing people do when t h e i r living conditions h p r o v e is to seek better health care f o r t h e t r children. National decisionmakers thus r e j e c t the occasional suggestions by international experts t h a t e f f o r t s should be focused on family planning while child health services remain undeveloped. Because the services f o r both health and family planning need to be developed, it makes sense to promote the Favorable interactdons between them, not to view them 3s competitive. i /P : The arguments against integrat,'.on usually turn on the ineffective- 1 ness, inefficiency, and low s t a t u s of health ministries. It sometimes is 3:* 1, said that family planning cannot wait f o r health services, which seldom a r e readily available to v i l l a g e Families. O r i t is claimed that services f o r e 1 family planning can be provided %are effectively and e f f i c i e n t l y a s a separate t: activity. The proponents of single-purpose programs of ten justify separate 6 3 services by saying t h a t an intensive, , ~ncentratede f f o r t w i l l have a more $ $ ~4 hmediate effect- ad that these services can be integrated l a t e r . The reason XI 1?i r a f o r such focused e f f o r t is clear i f there is a prospect of a categorical cam- k! . r 1; paign against4 an infectious disease t h a t can be eradicated. The saving from eradicating smallpox has been tremendous because vaccination progrsms a r e no - - longer needed. But there has been a massive re-gence s f malaria because i there was no health infrastructure i n which to @corporate the maintenance I * 1; f phzse of the control program. With present control measures, the sane w i l l be S true of almost a l l programs to control major diseases. :1 i f t h e case f o r single- purpose programs is tenuous f o r the prin- c i p a l diseases, it is even more tenuous f o r separate family planning ser- vices. I n most c o u n t r i e s t h e demand f o r contraception-- by parents whose f a u i l i e s already exceed t h e i r desired family s i z e o r t h e i r a b i l i t y t o c a r e f o r more children- - has not y e t been met. A t t h e s t a r t , it seemed t o make sense t o g e t some family planning s e r v i c e s out t o them as f a s t a s possible. The experience, however, has been t h a t when such a s e r v i c e is s e t up sepa- r a t e l y from health s e r v i c e s , long-term negative e f f e c t s have followed. P a r t i c u l a r l y damaging have been t h e personnel a t t i t u d e s of jealousy and competition created by providing s p e c i a l incentives f o r one group, because t h i s has reduced subsequent chances f o r cooperation with o t h e r h e a l t h and development workers. WHY INTEGRATE HEALTH AND FAMILY PLANNING? Six arguments suppqrt t h e benefits of i n t e g r a t i n g programs f o r health and family planning. The f i r s t is efficiency and effectiveness. An i n i t i a l but one- step increase i n family planning can be produced j u s t by providing ser- v i c e s to meet e x i s t i n g demand. For continuing e f f e c t s on f e r t i l i t y , however, demand must be increased by influencing motigation f o r E m i l y planning by t h e nost d i r e c t means t h a t f i t each l o c a l s i t u a t i o n . Of these, one of t h e most - universally a v a i l a b l e is the organizational b e n e f i t of combining health and .. -- family planning services. -Rather than have four t o six vorkers carry out one -- * t a s k each, a s i n g l e m ~ l t i p u ~ p o sworker can carry o u t s e v e r a l tasks, i f t h e e - - package is simple enough f o r consisten: performance. This kind of integration can improve the s t r u c t u r e of the organization, t h e t r a i n i n g of personnel, and the use of supplies, equipment, and f a c i l i t i e s . - 5 - Second, integrated programs offer more d i v e r s i t y of responsibility and therefore more challenge t o field-workers. The routine of doing only one task can sap the motivation that leads t o service of good quality. It also helps t o have variation i n tasks, so that imrkers can g e t a favorable response from people f o r a t lee.st some of what they do. A major danger of integration, however, is the overloading of tasks, which then permits workers t o concen- t r a t e on what they arc most comfortable with or on what brings the greatest reward and social response--because they obviously cannot do everything they have been asked t o do. I f the population covered and the package of services a r e kept small enough, it is possible ta avoid such overloading i n integrated services. Because health workers usually a r e most interested i n curing patients, it is important that curative a c t i v i t i e s should not crowd out both preventive health services and family planning. There must be established routines, good management, and supportive supervision t o ensure continuing attention to family planning i n integrated services. Third, integrated services make sense t o families. When a health o r family planning vorker v i s i t s homes, patients l i k e t o have a l l t h e i r problems cared f o r , rather than have some of t h e i r problens referred t o other personnel f o r reasons the people cannot understand. They want to g e t care f o r r t & themselves and t h e i r children from one person i n one c l i n i c , and they do not l i k e to go t o various c l i n i c s a t different times. Fourth, public relations can be improved by attaching a c t i v i t i e s B that have equivocal acceptance to programs that ha;e spontaneous and contin- uous demand. Even though some parents a r e eager f d r family planning because B 3 they have more children than they can care f o r , there generally is a much h - l a r g e r number of demographically important low- parity parents who have mixed t B f e e l i n g s about family planning- For them the case f o r ear1y spacing o r llmita- t i o n is more con i n c i n g vhen it comes from a health vorker they have learned t o t r u s t because of continued help f o r t h e i r h e a l t h problems. F i f t h , family planning f o r women has been shown ( i n h o s p i t a l post- partum studies) t o be especially e f f e c t i v e vhen provided a s a routine p a r t of maternal care. The reason i s the high motivation associated v i t h pregnancy. When education s t a r t s i n the prenatal period, the mother accepts as n a t u r a l t h e proposition t h a t family planning should begin immediately a f t e r d e l i v e r y t o protect her h e a l t h and the vell-being of the c h i l d . I n the normal routine of maternal care, e n t r y points f o r family planning can be i d e n t i f i e d and r e a d i l y introduced. The high motivation a t t h i s time has s e v e r a l sources. I n a d d i t i o n t o health, t h e r e a r e considerations of t i m e and money: parents would have d i f f i c u l t y caring f o r the c h i l d r e n i f another pregnancy s u p e r vened. There a l s o seems t o be a n important psychological consideration: having j u s t passed through one pregnancy, mothers tend t o want a r e s p i t e before another . C u l t u r d patterns and prolonged l a c t a t 2on support such spacing, but a r e l e s s e f f e c t i v e i n s o c i e t i e s that a r e rapidly changing. Sixih, it seems reasonable t h a t the experience o r expectation of reduced i n f a n t and c h i l d mortality w i l l increase tGe p r a c t i c e of family , L planning. As long as the proportion of children who d i e i n zhildhood remains substantJa1, a major psychological o b s t a c l e must be overcome i n promoting - t h e l i m a a t i o n of family size. Bealth s e r v i c e s t h a t lead t o s i g n i f i c a n t - * reductions i n c h i l d mortality may thus be an important, though not e s s e n t i a l , - means of encouraging lower f e r t i l i t y i n a r e a s where f e r t f l i t y and m o r t a l i t y a r e high. E f f o r t s t o reduce c h i l d l o s s and especially t o increase the p e r ception of higher c h i l d survival, although n o t guaranteeing a . d e c l i n e i n f e r t i l i t y , wodd be expected t o shorten the demographically Lmportcnt lag betveen t' iecline of mortal.ity r a t e s and f e r t i l i t y rates. Dr.. jite t h e m argun~entsf o r integration, it has been d i f f i c u l t t o integrate hervices once separate services have been s e t up. Competitiveness a t the periphery is strong, and workers have t o change basic a t t i t u d e s i f they a r e to work effectively i n an integrated services. Our experience has been that vorkers from special programs proved more r e s i s t a n t to taking on expanded a c t i v i t i e s than workers i n comprehensive services, who seemed from the begin- ning to be more ready t o adapt to changed emphasis and responsibilities. A t the top and middle administraitive levels, the p r o b l w s of separate services have been just a s great a s those a t the periphery. Some categorical programs have v i r t u a l l y been paralyzed because the people running them spend so much of t h e i r e f f o r t justifying t h e i r separate existence. In one country on the Indian subcontinent, integrated services had to be designated as a categorical i program outside regular health services to get the attention and p r i o r i t y t o cornpet. v i t h the continuing categorical programs. I n India, rudimentary services f o r maternd and child health were 1 I essentially destroyed when tht? v e r t i c a l family planning program vas s t a r t e d i n 1 the 1960s. It is important nov i n the cc-verse situation that, as integration 8 , * i is carried out, there should be special provisio& to ensure that family C 8 I planning services a r e not downgraded. A rational approach t o integration - - . requirts getting avay from l o o k i 4 a t the problem purely i n either- or terms. B i I Any integrated program +uld have program p r i o r i t i e s c l e a r l y f: i: Jc focused i n a general framevork t o achieve benefits similar to those i n ver- t i c a l programs. As much attention should be devoted to effective management i i n integrated programs a s i n categorical programs, In the past, the best man- agement capability was usually absorbed by v e r t i c a l programs, which have had special glamour and resources. This pattern made i t almost impossible t o develop the general health infrastructure with vhich the special rrograms were eventually t o be integrated. Proponents of separate v e r t i c a l services genei-ally agree that such services vill eventually have t o be integrated. If two o r more separate services a r e combined a f t e r each has become strong, the r i v a l r i e s and adjust- ments vill be traunaatic. The alternative usually recommended is t o add services progressively t o one service a f t e r it ha; achieved good coverage. The problem is t h a t each categorical service then sees i t s e l f a s being the one to which other a c t i v i t i e s should be added, and the problems of eventual integration remain. UBAT IS TEE EVIDENCE? It is generally recognized that high fertility and short birth intervals cause extensive danage t o the health of mothers and children i n most developing countries (Omran 1971 and 1976). The converse influence of health on f e r t i l i t y and birth intervals (and thus on population grovth) is much more complicated . , and remains the focus of a djcrr policy confrontation i n many countries and international agencies. The obvious d i r e c t effect of better health is t o - reduce mortality. But i f f e r t i l i t y remains high, the populat~oI)increases. L Of more interest f o r its long-term e f f e c t is a group of i n t e r a c a n g factors * i? by which h e e t h services promote family planning and thus r e d u c e f e r t i l i t y . Attempts t o unravel the many determinants of population growth have mainly relied on increasingly complicated s t a t i s t i c a l analyses of cross- sectional scrveys t o determine associations betveen and among variables (Norman and Hofstatter 1978; Kendall 1979; UNFPA 1980). Such interpretations of the spontaneous influences on population growth and national developent do not always lead to an understanding of what happens when different kinds of program interventions o r social changes a r e deliberate. This kind of i n f o r nation is best obtained i-om prospective f i e l d research, such a s t h a t reported here. Recent analysis i n several developing countries has confirmtd the d i r e c t e f f e c t of Zamily planning services i n reducing population growth (Xauldin and Berelson 1978), but more all have to be done to maintain the decline i n f e r t i l i t y . Largely because of the apparent succesa of programs i n such countzies a s China, the projection of world populh~ioni n 2000 has come . down from more than 7 b i l l i o n ten years ago and between 6 and 7 b i l l i o n f i v e years ago ( b r l d Bank 1974) t o about 6 billion today (World Bank 1980). There is growing information on imw program interventions from other development sectors can influence the Fractice of family planning with special attention to education and the r o l e of women (Ridker 1976, but see World Bank 1980). The Narangwal study was focused on health because it has many natural program links to family planning. 4 Ekperience i n a growing number a£ c o u n t r ~ e shas led to a rejection of the trickle-down theory of economic development and t o a correspond*ng acceptance of the need for allocating resources t o s o c i a l developent--with w the expectation t h a t t h i s w i l l c o n t r v u t e to a rapid decline i n f e r t i l i t y - (Rolfson 1975). Despite the rhetoric about social justice, however, much national and international decisionmaking contint2s to f i l t e r out legitimate claims t o equity, so that programs still disproportionately benefit the e l i t e . Skepticiss e x i s t s about whether the new emphasis on equity can be put i n t o action. Economic planners have been c ~ s ccmfortable with studies t h a t meas- t I ure economic e f f e c t s , which seem to be precisely quartifiable. S ~ c i a ldevel- f 5 opment variables have seemed vague, m;stlcal, and diffusa with few c l e a r l y d definable indicators. Despite the problems of measurement, experience i n several countries suggests that one of the best ways of bringing mortality and f e r t i l i t y i n t o balance a t low l e v e l s is through d i r e c t measures t o improve social equity and d i s t r i b u t i v e justice. It would be wonderfiil i f such measures, which carry a strong moral and p o l i t i c a l imperative, could also be shown by s c i e n t i f i c data to be important i n solving the world's population problem. Other thau by p o l i t i c a l revolution, the most d i r e c t way of improving equity is co g;ve the poorest people better access to services t h i t can meet the basic requirements of l i f e . Because s p e n d i ~ gon health care and n u t r i t i o n uses a large p a r t of the income of poor people, health interventions a r e particularly important. Three indicators make up the Physical Quality of Life Index: infant mortality, l i f e expeciancy, and l i t e r a c y (Grant 1978). That tvo a r e health i r l i c e s further indicates t h e importance of health i n development. Of special i n t e r e s t is the: ekperience of S r i I.snka and the s t a t e of Yerala i n South India (Ratcliffe 1978) where p o l i t i c a l decisions create3 - situa+ions that permitithe separation of social ,nd economic influences on - L population growth. Both governments provided rearly universal cweraga of - h health cars, family plaanicg, education, and basic nutrition, but each s t a t e $% experienced almost zero growth i n GNP. Birth r a t e s declined more than ten %-'A? points i n ten years. The reasonable conclusion is t h a t i t is not necessary t o wait u n t i l people have enough money to pay f o r social services as economic benefits t r i c k l e down t o them. In the long run, of course, economic develop- ment is essential t o support the whole development process. From analyses of national data, the need has become evident for studies t h a t dissect i n a more \ precise way the dynamics and causal relation8 among social development var- iables a t the l o c a l l e v e l t o guide general policy decisions and program plan- ning. Feu longitudinal, controlled studies have attempted t o measvre the strength of program variables i n influencing the practice of family planning. One of the best-knom f i e l d projects i n India is the Khanna project, which w a s conducted betveen 1953 and 1969 In the Ludhiana D i s t r i c t of Punjab, j u s t twenty-six miles from Narangval (Wyon and Gordon 'FT1). It provided data on the e p i d e m i o l o ~of~ f e r t i l i t y and a wealth of information on methods of research. The e f f e c t of e f f o r t s to promote family planning was minimal partly because foam t a b l e t s were the only contraceptive used. The fieldwork was con- ducted in a way t h a t was especially sensitive to what people i n the v i l l a g e s were thinking, and the reports stressed sociocultural and economic considera- tions i n the epproach of village people t o problems of f e r t i l i t y and the value of children. Much was added to the understanding of why v i l l a g e people a r e C cautious about family planning. The study identified fi-re principles of popu- l a t i o n control t h a t e r e important for future planning: reducing child mortal--.. '2 - i t y , promoting comu11ity education, encouraging s o c i a l and material progress, -- - providiag induceuients f o r dhlayed m r r i a g e and small families, and ensuring .t supplies of suitable and e f f i c i e n t contraceptives. As f a r as we how, Naranqwal was the f i r s t f i e l d project to measure i n a prospec-lve and controlled experimental design the interactions between various combinations of health services and family planning. After the Narangwal p r o j e c t s t a r t e d , a study v i t h s i m i l a r o b j e c t i v e s but a very d i f - f e r e n t experimental design was :onducted in Danfa, Ghana (University of Ghana 1979). According t o that study, fankiiy planning acceptance increased from 11 percent i n 1972 to 34 percent i n 1977 i n the area v i t h the most- intensive s e t v i c e s . I n an a r e a v i t h f e v e r services-nly those f o r h e a l t education and family planning- the increase was from 7 percent t o 2 1 percent; i n an area with family planning alone, from 2 percent t o 8 percent. The p r o j e c t showed t h a t acceptable s e r v i c e s can be developed at reasonable c o s t under t h e condi- t i o n s i n Africa. A l a r g e r demonstration p r o j e c t vas conducted from 1974 t o 1979 i n Lampang Province, Thailand i n a n e f f o r t t o improve t h e coverage of primary h e a l t h c a r e f o r more than 600,000 persons ( I n t e r n a t i o n a l Council f o r Educa- t i o n a l Development 1979). Although not s t r i c t l y a research p r o j e c t , its value a s a prototype soon l e d to the designation of c o n t r o l d i s t r i c t s and t h e we of a complex system of gathering data. Special t r a i n i n g o f ninety- six medical a s s i s t a n t s and 901 commuaity h e a l t h volunteers produced s u b s t a n t i a l i n c r e a s e s i n the use of services, but it vas d i f f i c u l t t o supervise the l a r g e , dispersed cadre of volunteers. Evaluation, adapted from the Karangwal analysisL, is under way. Meanwhile, many p a r t s o f t h e program a r e being r e p l i c a t e d throughout Thailand v i t h World Bank support. These seem t o be the only controlled comparisons of the e i f e c t of w h e a l t h i n t e r v e n t i o n s on family planning and f e r t i l i v . Related information * can b e derived from demonstration p r o j e c t s t h a t did not have c o n t r o l groups but t h a t meas:-sb ;ksffect of innovative packages of integrated services over Elme. Examples a r e the Companiganj P r o j e c t i n dangladesh; the p r o j e c t s O F the Population Council i n Turkey, Nigeria, Indonesia, and the Philippines; the projects developed by the Antia (1979), Coyoji (1379), P. M. Shah (1977), and the Aroles (1975) i n Haharashtra; and an increasing number i n a l l p a r t s of India. Finally, there is the open question, Do changes i n child survival influence motivation f o r family planning? I f they do, they probably operate differently a t var'aus stages of development and iii '-he context of other forces influencing a t t i t u d e s about f e r t i l i t y . From our e a r l i e r m r k i n Punjab villages, it was evident that the e f f e c t is not automatic. There is certainly not a one-to-one replacement of children who a r e l o s t . Two types of motive- tional forces need to be distinguished: "replacement" of children wi-lo d i e and "insurance" births because of the expectation that some might die. A confer- ence report summarizing some of the evidence on replacement (Preston 1978) showed that the main reason for the well-documented shortening of interpreg- nancy intervals a f t e r a child death was the cessation of lactational amenorrhea, which provides biological protection from pregnancy. Separate but smaller reductions i n b i r t h intervals a f t e r a child death, presumably due t o replacement motivation, were demonstrated i n situations where l a c t a t i o n could not have been the explanatiod (Taylor, Newman, and Kelly 1976). The consensus is that motivational influences may be associated with the replacm-ent of a t h i r d to a - halfiof the children l o s t (Preston 1978). -i A recent analysis of Korean data showed that before 1965 most of th; .? C !c mortality-related reduction i n b i r t h r a t e s was due to lactational amenorrhea; a f t e r 1965, when the National Family Planuing Program became effective, the motivation f o r replacement became more important. I n any case, a f t e r mortali- t y had already declined, the frequency of child deaths became so low t h a t it could have contributed only about three percentage points t o the reduction in t h e number of b i r t h s (Park, Eian, and Choe 1970). This supports the general experience (Predton 1978) that t h e motivation t o r e p l a c e c h i l d r e n who d i e can ve demonstrated mainly i n such p l a c e s a s France and Taiwan, where development has been general and where parents a r e already l i m i t i n g t h e i r family s i z e . Replacing a c h i l d i m p l i e s a n uncompleted i d e a l family s i z e and t h e r e f o r e vould be r e l e v a n t only up t o l o c d family-size norms. As long as reproductive performance is n o t c o n t r o l l e d and people f a t a l i s t i c a l l y o r e n t h u s i a s t i c a l l y accept whatever c h i l d r e n cme, t h e r e is l i t t l e r e s t r a i n t on f e r t i l i t y t h a t can be removed i n respcnse t o the replacement motivation. Therefore, replacement probabiy is not a n important f o r c e i n maintaining high b i r t h r a t e s i n less delreloped c o u n t r i e s . From t h e point o f view o f p o l i c y , however, t h i s conclu- s i o n does not j u s t i f y t h e f u r t h e r assumption t h a t i t d i s p r o v e s t h e child- s u r v i v a l hypo t h e s i s . Under c o n d i t i o n s i n which c h i l d m o r t a l i t y remains high, such a s i n the Punjab, t h e v a r i a b l e t h a t is more r e l e v a n t but more d i f f i c u l t t o measure is i n s u r a n c e motivation. Early s t u d i e s assumed t h a t d e c i s i o n s t o have e x t r a c h i l d r e n because some might d i e would be made d e l i b e r a t e l y and r a t i o n a l l y , but w e have found t h a t t h e s e a t t i t u d e s seem t o a r i s e from subconscious expec- , L 0 t a t i o n s . Early i n ;his research, w e showed t h a t o n l y 10 t o 15 percent o f parents s a i d yes when asked i f they were having more c h i l d r e n because of -- t h e f e a r t h a t some might d i e ; 45'fercent s a i d they had never thought of t h e p o s s i b i l i t y (Taylor and Takulia 1070). Cross- tabulation showed s t a t i s t i c a l l y !e s i g n i f i c a n t a s s o c i a t i o n s between $rents' expectations of a c h i l d death and t h e i r readiness t o p r a c t i c e family planning, suggesting t h a t t h e subconscious motivation of insurance needed t o be s t u d i e d to d e f i n e t h e s t r e n g t h of t h i s a s s o c i a t i o n and whether it might have policy i m p l i c a t i o n s . The research had two goals. For general international, and s c i e n t i f i c inter- e a t , a complex s e t of research hypotheses and models m a tested i n a con- - trolled experimental design. Different packag?s of seirvicr were the inputs leading to quantification of the relations betveen inputs, outputs, and outcomes. A p a r a l l e l goal, i n response to the ceeds of program developent i n India and other developing countries, was to develop low-cost packages of integrated services f o r rural areas. These packages then could be adapted i n demonstration projects to the needs of g o v e m e n t services i n the other s t a t e s and countries. The Indian government's m a i d objective i n promoting t h i s % research, then, was to find out hov t o develop integrafed services t h a t com- I bine family planning, women's services, child care, anb nutrition. Officials a t the Ministry of Health had already d e c i d d that thep needed to go beyond the intensive singlepurpose national program for famiJy planning. Massive investment was being concentrated on s t e r f l i z a t i o n camps f o r vasectomies and tubectomies. The mass approach of inserting intrauterine devices (IUDs), a l s o . t i n camps, had already produ*ced a serious backlash. Intreasing use of pressure and incentives was leading to negative reactions among the population. - The research approach rejected the notion th&t it reasonable to i w a i t for s o c i a l and economic forces to reduce b i r t h rates. We s e t out t o find !P the most a i r e c t and culturally acceptable ways of promoting f a r i l y planning through integrated services. It was obviously important to separate the i t e f f e c t s of general social change from those of the integrated program, and K i : t h i s called for the simultaneous analysis of many variables. Two groups of f a c t o r s were i d e n t i f i e d . The first-- comprising many v a r i a b l e s s t r o n g l y associated with a d e c l i n e i n f e r t i l i t y - a r e n o t e a s y t o manipulate. E i t h e r they are endogenous and not s u s c e p t i b l e t o change, such as c a s t e , o r they a r e the r e s u l t of slowlnoving s o c i a l t r e n d s , such a s changes i n t h e s t a t u s of women and i n preferences f o r sons. Policymakers and progran managers must understand such v a r i a b l e s i n a r d e r t o i d z n t i f y t a r g e t groups f o r s p e c i a l services. Our r e s u l t s include data on these v a r i a b l e s and show how s e r v i c e s can be focused on groups with s p e c i a l p a t t e r n s of response- The second group of v a r i a b l e s t h a t con he manipulated through program i n t e r v e n t i o n s is much smaller. Throughout t h i s r e p o r t it w i l l be apparent t h a t we were frankly i n t e r v e n t i o n i s t i n seeking a c t i v i t i e s t h a t can be r e a d i l y changed. The set of v a r i a b l e s most r e a d i l y manipulated is i n t e g r a t i o n of h e a l t h s e r v i c e s with family planning. Surprisingly, t h i s a r e a had received almost no systematic research a t t e n t i o n . The Narangwal research n o t only measured the s t r e n g t h of a s s o c i a t i o n s between various combinations of s e r v i c e s ; i t also shoved hou these combinations could be implemented. A l l s e r v i c e s were designed to be c a r r i e d out by normal g o v e m e n t s e r v i c e s using auxiliaries-- at an annuale c o s t of less than $2 a person. a S e t t i n g A study of t h e ecology of population dynamics I n v i l l a g e communities must s t a r t with an understanding of v i l l a g e c u l t u r e . The Punjab has always t tended t o b e progressive. As beneficiakies of the Green Revolution, these communities enjoyed socioeconomic development and improvement i n access t o basic s o c i a l s e r v i c e s before and during t h i s p r o j e c t period. For those who have not been i n the Punjab r e c e n t i y , it nay be hard t o appreciate the trans- formation s i n c e the Green Xevolution doubled a g r i c u l t u r a l production and brought considerable economic affluence, even though i t went dispropor- t i o n a t e l y to the r i c h . Money has changed t h e viliages- - as b r i c k replaced mud, and e l e c t r i f i c a t i o n brought labor- saving devices and t h e radio, s h a t t e r i n g the q u i e t of v t l l a g e l i f e . The paving of v i l l a g e streets eliminated t h e need f o r a g i l i t y i n maneuvering past mud holes. Mechanized a g r i c u l t u r a l implements were brought i n t o ease the work pressure of handling l a r g e r crops. Education expanded s o r a p i d l y t h e r e soon were t h a t t h e r e soon were primary schools f o r g i r l s and bcys i n almost every v i l l a g e , a high school w i t h i n a few miles, and a n increasing number of r u r a l colleges. Paved roads have extended putPic and p r i v a t e transport by bus and bicycle. The spreaC of l i t e r a c y has increased the supply of newspapers, magazines, and o t h e r w r i t t e n materials. The Punjab has always been p o l i t i c a l , but v i l l a g e p o l i t i c s have become even more i n t e n s e because of the i n f l u e n c e of panchayats (elected v i l l a g e counsels) on t h e e l e c t i o n of l o c a l government o f f i c i a l s . One g r e a t advantage of working i n Punjab v i l l a g e s was thac zhe progressive a t t i t u d e s helped i n g e t t i n g c o o p ~ r a t i o nand feedback from communi- t i e s as we j o i n t l y worked o u t b a t t e r ways of ~ d n gthings. Free and open discussion with t h e aggressive and vigorous Funjabi v i l l a g e r s about any * 1 1 a c t i v i t y t h a t might b e n e f i t them and t h e i r children showed a healthy balance between eagerness and reluc tance t o change. The cooperation was g r a t i f y i n g , a s l o c a l people took p a r t i n t h e research by t e l l i n g us what would o r vould a o t work and by suggesting b e t t e r approaches. . Punjab v i l l a g e s a r e strongly i n t e r a c t i n g s o c i a l u n i t s , each of vhich has i t s own h i s t o r y and personality. Groups of v i l l a g e s tend t o be geographically pocketed, and each one has two o r t h r e e strong f a c t i o n s based on c l a n s ( o r p a t t i s ) of the dominant Jat-Sikh caste. This c l u s t e r i n g of characteristics and a f f i l i a t i o n s produced unexpected problems i n getting comparable experimental groups. The v i l l a g e of Narangwal is ir* the Pakhowal block- in the south- i ? central section of Ludhiana District. A careful selection process was used t o G identify twenty-six experimental villages i n eight c l u s t e r s with a population of about 35,000 i n the Pakhowal, Sudhar, and Dehlon blocks (see figure 1.1). 8 Government health services i n the area a r e based on primary health centers, with each center serving one community development block of bezween 80,000 and 100,000 peopls. A l l the usual luauti Eiable parameters were taken into consideration i n selecting villages, but there still were major differences i n important variables, differences outlined l a t e r i n t h i s chapter. The lack of homogeneity i n more subtle but ultimately more important characteristics, such a s willing- ness to change, obviously were impossible to balance i n village selectica. Only a f t e r the fieldwork began did we realize how d i f f e r e n t the Kesponses were d between villages and between groups of villages- According to several indica- i If tors, the villages receiving comprehensive integrated se-vices happened to be the most conservative i n the study, partly because we chose them for their * t isolation from existing services. This gave us even more assurance that the f i n a l differences between experimental groups were important. - - ii A cultural 'Bharacteristic that complicated data gathering is the pattern of v i l l a g e exzgamy. All g i r l s marry outside the village, and for the b f i r s t several deliverfes women go home to their parents. More than a third of cz % the births i n our population were i n villages where we had no personnel, so we ry had to s e t up elaborate mechanisms f o r recording incoming and outgoing births. I n addition, women typically go to t h e i r parents' homes two or three months - A IIC CLnaL --*-· ' @ C O N ~ N LI~IWnlr& I , ! before d e l i v e r y and s t a y f o r three o r four months a f t e r delivery. That made i t d i f f i c u l t t o g e t complete reporting of pregnancy wastage, o f neonatal and i n f a n t m o r t a l i t y , and of such s p e c i f i c measures aa b i r t h weights. That a l s o made i t d i f f i c u l t t o provide cont:lnuity o f maternity s e r v i c e s and family planning. L General f e r t i l i t y trends i n r u r a l Punjab needed t o be c a r e f u l l y E accounted f o r i n t h e analysis. There was a steady d e c l i n e i n f e r t i l i t y during the decade before the project s t a r t e d i n 1969. Data from Punjab generally and from Ludhiana D i s t r i c t show t h a t b i r t h r a t e s had f a l l e n from more than 40 p e r thousand t o t h e mid-303, something l e s s than 1 percent a year. Our d e t a i l e d d a t a from pregnancy h i s t o r i e s suggest t h a t i n the c o n t r o l v i l l a g e s t h e r e was no general a c c e l e r a t i o n of the r a t e of decline before o r during the p r o j e c t , d e s p i t e a massive n a t i o n a l p r o g r h f o r family planning t h a t s t a r t e d i n 1965, f i r s t with IUDs and then with s t e r i l i z a t i o n s . An important v a r i a b l e defined i n thz Khanna study (Wyon and Gordon 1971) was the increasing age a t marriage f o r women. I n our population t h e mean age a t f i r s t marriage increased from 12.8 years f o r those married before 1940 t o 19.3 years f o r those married a f t e r 1970. I n Puajab a second ceremony, muklawa, t a k e s p l a c e vhen cohabitation starts, and the mean age f o r muklawa f o r t h e same groups of women increased from 15 years t o 19.8 years. -- ' ? The f e r t i l i t y patterns of all experimental groups appear t o have -- - been f a i r l y uniform during- the years before the p r o j e c t began. This is * E indicated by t h e s m a l l differences i n mean age f o r given p a r i t i e s o r f o r oregnancy order. For all women i n the study population, t h e mea, p a r i t y was Design and d a t a base The experimental design g r e v o u t of t h e general hypothesis that the a t t i t u d e s toward a~ the p r a c t i c e of family planning would improve i f family planning were integrated with h e a l t h services. lko subhypotheses were t h a t each of the s e p a r a t e components of services would have measurable e f f e c t s and t h a t t h e dynamics of t h e i r i n t e r a c t i o n a could be traced t o d e f i n e options , f o r policy and program decisions. An a d d i t i o n a l hypothesis was that a d e c l i n e i n i n f a n t and c h i l d m o r t a l i t y would lead, a f t e r a l a g , t o increased contracep- t i v e p r a c t i c e because of a t t i t u d e changes associated v i t h expectations t h a t more children would survive. We a r b i t r a r i l y estimated at the o u t s e t t h a t a t l e a s t f i v e years vould be needed t o denonstrate motivational changes result- ing from improved c h i l d survival. A major problem i n t h e a n a l y s i s was t h a t a change i n relations betveen India and the United States at the t i m e of the Bangladesh war l e d to the termination of the p r o j e c t a f t e r four and a half years of s e r v i c e implenentation i n some experimental groups, agd a f t e r only two years i n others. This me'ant t h a t we d i d not have enough t i m e t o measure the long-term e f f e c t of motivational changes r e l a t e d t o t h e child- survival hypothesis. b Four experimentd groups of v i l l a g e s received d i f f e r e n t s e r v i c e packages; a f i f t h served a s a c o n t r o l ( f i g u r e 1.2). Throughout t h i s r e p o r t they w i l l be i d e n t i f i e d a s follows: 1/ - FEJSCC: Fami1y planning, women's se-rvices, anh c h i l d care s e r v i c e s , E including n u t r i t i o n c a r e (mid-1969 th;ough 1973). -1/ The dates show when full-scale family plailning services were provided. I n some cases, e s p e c i a l l y i n FPCC v i l l a g e s , some s e r v i c e s had been s t a r t e d e a r l i e r . - . Figure 1.2 I I Experimental Design for the Narangwal Population and Nutrition Projects FPWS: Family planning and wmen's s e w i c e s (mid-1969 through 1973). FPCC: Family planning and child care services, including n u t r i t i o n c a r e (1971 through 1973). FPED: Family planning education (1972 and 1973). . CONT-P: Control for population project The concurrent n u t r i t i o n project had four experimental groups of villages, vhich received the following service inputs: NUTHC: Child care services: nutrition care and health care, mainly infectious disease control and health educaticn. NUT: Nutrition care. HC: Health care, mainly infectious disease control a~ health education. C0tiT-N: Control for uutrition project. One experimental group vas used for both projects: the family planning and child care group i n the population study (FPCC) was also the child care group i n the n u t r i t i o n project (NIJTHC). Designing a f i e l d project of t h i s scope introduced major conceptual and methodological problems. One conceptual option a! the beginning was to L match the inputs of a l l services i n each experimental c e l l or to match inputs according to each serviie. To compare the effects of each service uould - 'C - obviously have been desirable for equaaizing each component. This would have meant, however, t h a t the FPtlSCC v k l a g e s vould have received the additive * inputs of three services. Because we were interested not i n the separate e f f e c t of each service but i n the e f f e c t of integrated services, we decided to t r y the other approach and equalize the s v e r a l l inputs, which meant that e f f o r t I that improved efficiency and other benefits would compensate f o r t h i s reduction I of e f f o r t i n any one service. In any case, we developed methods of measuracnt that would give us detailed evidence on inputs by a l l categories of workers i n minutes of e f f o r t per week. The inputs f o r each e x p e r h e n t a l c e l l were viewed a s service packages b u i l t around the various combinations of family planning, women's services, and children's services. Racher than j u ~ adding a c t i v i t i e s together, g r e a t t e f f o r t went into working out the best combinations and sequences under f i e l d conditions--so that a c t i v i t i e s were t r u j Integrated, not merely juxtaposed. Becsuse our most important objectives were to develop services f o r the primary health centers, we decided not t o follow the usual principle of holding inputs uniform throughout the experiment. What we f i r s t decided would be best vas frequently shown to be unworkable o r ineffective a f t e r a few weeks o r months. Interaction and feedback from family health workers and v i l l a g e people was encouraged and proved most constructive and innovative. Holding rigidly t o a prescribed service input, even though we had learned a b e t t e r way of providing services, would have been unethical. * C o n v i c t ~ ~ nabout the ethics of research i n v i l l a g e communities s obliged us t o ensure t h a t the rights of v i l l a g e people took precedence sver the objectives of our research. The f i r s t of these r i g h t s was t o involve them i n p l a n n i ~ gand implaenting the fieldwork.'Prsliminary negotiations x r r e E conducted with panchayats, which approved a l l a c t i v i t i e s . It proved relatively easy to get understanding and concurrence with the n o t i ~ nof an experimental design i n which d i f f e r e n t packages of services were to be offered to various v i l l a g e s . I n any case, our s e r v i c e s supplemented t h e l i m i t e d care provided by ;~overmtentp r a - r y h e a l t h centers. There was a d i f f e r e n c e , however, i n t h a t primary health c e n t e r s provided i n t e n s i v e c a r e t o 5,000 people i n immediately adjacent villages. For e t h i c a l reasons, we decided t h a t control v i l l a g e s , family planning education v i l l e g e s , and the family planning and women's s e r v i c e s v i l l a g e s t h a t did not receive 2hild care from us should be the i a t e n s i v e zoDe o r a s c l o s e a s possible to the government primary h e a l t h c e c t e r s . This d i l u t e d considerably t h e comparability o f our controls. It a l s o nade f o r s i g n i f i c a n t differences i n p r i o r use of contraception, because the maia a c t i v i t y o f t h e prinary h e a l t h c e n t e r s had been family planning: The 3 e t efEect cC these differences, however, is t o make our r e s u l t s more convinc- ing because they had t o start by compensating f o r these i n i t i a l advantagea. To have a bcse a g a i n s t which t h e influence o f health s e r v i c e s could be measured, it was necessary to oeasure the e f f e c t of family planning services alone. Acceptance always increases when a new program is s t a r t e d because e x i s t i n g dananci f o r family planning leads people t o t r y modern methods a s a s u b s t i t u t e f o r t r a d i t i o n a l methods. W e expected t h a t p r a c t i c e curves i n fam- ily-planning-orly v i l l a g e o would plateau i n about two years, a f t e r e x i s t i p g t C dzrns~df o r family planning had been met. These v i l l z g e s presented a ? r o b l a i n oegotiations because our goyernment advisors considered the e x i s t i n g situa- - .C ti01 i n control villages semi- by t h e national family planniilg program t o * - have been e s s e n t i a l l y a f a m i l y ~ l a n n i n g - o n l y approach, and they were i n t e r e s t e d ~ a i n l yi n cesting o t h e r combinations of services. It was finally agreed that the family- planning-only group should become a family planning education group to see how much an intensive educational program could increase the use of family planuing. Rather than using family health workers a t t h e v i l l a g e l e v e l , the basic f i e l d personnel f o r t h i s group of v i l l a g e s were family plan- ning educators, who had been t r a i n e d as junior b a s i c teachers f o r v i l l a g e L schools. These negotiations and arrnngements took time, and s e r v i c e s i n t h i s group of v i l l a g e s d i d not g e t s t a r t e d u n t i l more t h a n two years a f t e r those i n t h e f i r s t groups of v i l l a g e s . Differences between v i l l a g e s Experiment a 1 f i e l d s t u d i e s t e s t i n g s e v e r a l intervent ions and using c o n t r o l s i n v a r i a b l y confront t h e r e a l i t y t h a t v i l l a g e s a r e unique and complex s o c i a l units. Me attempted t o reduce differences by se!ecting and grouping B v i l l a g e s to provide comparability i n major socioeconomic c h a r a c t e r i s tics. S e l e c t i o n c r i t e r i a a l s o included c l u s t e r i n g v i l l a g e s i n each group and kesping t h e c l u s t e r s at a reasonable d i s t a n c e from other c l u s t e r s t o reduce intergroup contact. Year-round a c c e s s i b i l i t y t o four-whee?. d r i v e t r a v e l within one hour of the project headquarters i n Narangwal was the main geographical c o n s t r a i n t . Despite these e f f o r t s , some important d i f f e r e n c e s between experimental groups were found i n subsequent analyses ( t a b l e 1.1). FPED v i l l a g e s and t o a lesser extent c o n t r o l v i l l a g e s had the lowest dependeucy r a t i o s , t h e highest p r o p o r , I L t i o n of higher c a s t e s , the highest l i t e r a c y , t h e most landowners, and t h e - highest average income of the various groups of villages: Additional evidence from the 1961 and 1971 p4~vernmentcensuses of these v i l i k e s ( t a b l e 1.2) r e v e a l s t h a t FPED v i l l a g e s had a h i s t o r y of higher l i t e r g c y a s well a s more * e p rapid improvement i n the education of women between 1961 Land 1971. The d i f - &% e -* ferences between experimental groups indicate t h a t t h e b e t t e r r e s u l t s from i n t e g r a t i n g s e r v i c e s f o r health c a r e and family planning probably would have been even stronger i f these d i f f e r e n c e s had not e x i s t e d . Analyses a t the in- d i v i d u a l l e v e l were a b l e to ad j u s t s t a t i st i c a l l y f o r zany of these d i f f e r e n c e ~ . ~ u c a nLftarata t 33.1 25.7 34.0 43.2 34.3 . , 38.3 -- Orrt;lprtionof LLird of Borrrlhatdt W P - ( ~ ) a . 7 rda 3 3 ~ . 5s.z 43.0 46.2 t w n , 34.5 33.L - 3s-7 2S.3 29.8 25.2 oathas 23-8 20.9 30.7 - - L9.S 27.2 28.6 -2 8 A l L t U d 3L2 34.7 '-, . U - O - 3s.6 33-9 23.6 8 ' m . r d - 88.9 90.2 93.3. . - 94.4 0S.O 40.0 )WDraFPr 44.2 44.0 55.a- ' U.3 38.8 54-7 0- 42-2 '24.7 ! 42.8 2 4 3 .23.6 35.8 . . . -x - . . - -... . . -- -. ~ d h ~x ~ m f ' m - 5-7 8.5 s.a 4.9 7.8 5.4 a'EraEtorusas 0-2 1 . ~ 8 17-7--- - 27.5 22-8 a.7 t T ' & v d l ~ 27-2 32.1 ' 2L.3 21.9 33.3 32.1 -kf&LdS (p-y ard cbrrrlr*ilb m c d thmur3h ~ M o ofn- s a d by arauu of ottras surreys, not -aa+ad.oathe ca~lslu#rtrlde a a s e 8. +hr to- FQpuLatfoP w aiJoat5-6 fn -+srpai-ntal gmup =am= - - =a t n t psopb ulrr hbmdfLe&, &brinq,+hr totalr- p p l a - tian up tu a b u t 35,0013. .,-'Phis . ,2 +--of hmsrholdtarminQla3d KQ~Bdi,stancu in w t e r s v d g h f d by vilIaqa poealatian t Data for PPEd is fur 1972- Summary of v a r i a b l e s A l a r g e number of input- output-outcome v a r i : ~ l l e swere measured i n h i t h e many surveys and s t u d i e s of the project. Table 1.3 summarizes the data sources and the daces when the data c o l l e c t i o n was c a r r i e d out. A d e s c r i p t i o n i of the methods of d a t a c o l l e c t i o n is i n appendix A. Field p r o j e c t s usually C do not measure i n p u t s i n d e t a i l but attempt instead to control f o r t h i s by equalizing personnel and resources between groups. Even though we attempted t o standardize formal inputs by having one family h e a l t h worker o r family planning educator f o r each v i l l a g e , we realized t h a t t h e d i f f e r e n t packages of services would involve d i f f e r e n t r e s p o n s i b i l i t i e s and investments of time. Fortunately, t h e method of functional a n a l y s i s t h a t we had developed e a r l i e r provided a n excellent means of g e t t i n g d e t a i l e d d a t a on s e r v i c e s , including counts of s e r v i c e contacts, time investments f o r each function, and d e t a i l e d c o s t s by a c t i v i t y (Johns Hopklns 1970). This provided a b a s i s f o r using q u a n t i t a t i v e measures of services a s independent v a r i a b l e s i n input- output- outcome a n a l y s i s and f o r performing d e t ~ i l e ds t u d i e s of cost- effectiveness. To d e f i n e the background f a c t o r s t h a t influenced program r e s u l t s , * 9 I many socioeconomic, demographic, and a t t i t u d i n a l v a r i a b l e s vere measured. These included indices of some of the v a r i a b l e s considered most l i k e l y to - - influence population growth: ede&ational l e v e l , communication betveen husband and wife, a t t i t u d e s toward child&, and a t t i t u d e s toward planning f o r the S f u t u r e ( E a s t e r l i n 1974; Ridker 1 n 6 ) . Analyses reported i n t h i s moubgraph have focused on the v a r i a b l e s t h a t have the g r e a t e s t policy relevance. A b r i e f descripton of these v a r i a b l e s follows. Variations i n the data sources w i l l be described with the analyses f- and sordirr u u e carded ouf sy3r-tically fn all cxr s- s ~- t s & ~ ~ ~ ~ ~ ~ ~ ~ ~ * t d b an-seanaL 3- aucUnrC-=-Qf- 1467-68 22- s~ooad rsn 22 v i ~ ~ g - ~urtinrcensus of uf FapPlatfaz - @ 9972 4 crfllagas(E'FEd) ThFrd Causus 1973 26 villaqes -==J'w - " f9fi8-69 nvfllaiqrr mat+ W t y P 1 MrprpY 197L-72 26 w u . 3rd SUIP.Y. lS74 4 ~ - ( & c c ] ~ttfhlrIlcarrd 1 .. msrtr~all 1968-69 22 About Family ZIuminq 1 -. 2nd- 1971-72 26 oillaw and-&- * . 3rd 1974 4 vFUages (FFWSCC] rnouldqm a d P m t 1 . -.survey 196849 22 vfllaqea of Pancfly Ptlnnia~ 1 '* asd m y 1971-72 26 vFLlages 3rd s u r a y 1974 4 villaqes (FPWSCC) w c S m v p k- 1S7-68 22 villag- 2nd =kt- 1971 22 w a s -: Irtsun787 i 1972 4 vFUaqes(FPPd) \ Ungitndiaal Data Callaaforr F e y P I i u m h s !?raeh= l969-74 8 vFllagu 19n-74 3 villages(FPCCI 14fZ-74 4 vCUqes(PPEd) - P U t p Sta- EUgihle uo~men 1969-74 U wiUaq& 1971-73 7- villages . Pita Statk- - Bfrthr and Outfis.1 1969-74 22 vfll%m - I 3972-74 s &agasIpp~d) MarrLaqes . 1972-74 26 villaqes o Demographic: Age, sex, marital s t a t u s , p a r i t y , number of l i v i n g children, number of c h i l d r e n who had died, and dates of entrance and exit from the study population were used as important c o n t r o l v a r i a b l e s as v e l l as t o de-belop s p e c i f i c population denominators. o Socioeconomic: Religion and c a s t e group, husband's education, husband's occupation, household income, housetiold land ownership, and possession of sdlected household items were used t o measure individual and household cnarac t e r i s t i c s . o A t t i t u d e s and beliefs: Approval of family planning f o r themselves o r f o r newlyveds, i n t e n t t o have another c h i l d , and b e l i e f s about changes i n c h i l d mortality, chances of c h i l d r e n surviving, o r a combination v a r i a b l e vere obtained from responses t o a questtonnaire by married wonen aged 15-49. An important index proved t o be whether husbands and v i v e s talked about family planning, a v a r i a b l e t h a t we a s s m e t o be an i n d i c a t o r of a voman's f e e l i n g of indepen- dence and of her a b i l i t y t o p a r t i c i p a t e i n family decisions. o Knovledge and practice: Awareness of methods of contraception and use of contraception before the p r o j e c t were a v a i l a b l e f o r a l l married women aged 15-49. II o F e r t i l i t y and mortality: A l l b i r t h s and deaths (including f e t a l , i n f a n t , and c h i l d ) before the project were recorded i n a d e t a i l e d pregnancy h i s t o r y f o r marrie$ women aged 15-49. A l l b i r t h s and deaths ( a l l ages) during th? project were recorded as they occurred, using many sources of repcrting. o Morbidity and growth: Longitudinal periodic measures of days of selected i l l n e s s symptoms and weight and height of c h i l d r e n under t h r e e were collected i n c h i l d c a r e v i l l a g e s . o Project services: Services t o individuals were categorized as: women's i l l n e s s care, which includes v i s i t s o r contacts related t o women's i l l n e s s e s usually i n i t i a t e d by ill women o r t h e i r families and most often occurring i n the v i l l a g e c l i n i c ; women's other care, which includes contacts i n i t i a t e d by project 5ealth workers through routine home v i s i t s t o wwaen f o r f e r t i l i t y sur- veillance, health education, family planning motivation, and preventive care; children's i l l n e s s care, which includes v i s i t s o r contacts f o r curative purposes to children of women i n the study villages, usually at. family i n i t i a t i v e and i n the village c l i n i c ; children's other care, which includes routine contacts i n i t i a t e d by project health workers f o r prevention and surveillance (but including early diagnosis and treatment i n the home) f o r children of women i n the study villages; family planning services and followup, which includes contacts with both women and men involving the provision of modern family planning methods o r the follovup of i 5 users; and family planning motivation, vhich includes contacts by r s i. m l e family planning workers with husbands of vomen i n the study F yillages (these contacts invoolved general rapport generating activi- I ties, health related advice, specific family planning and population education, and motivation t o use contraceptives). The f i r s t two constitute constitute women's services (WS), the second two children's - . services (CC), and the third t w s family planning (FP) and ramily planning education (ED). o Work sampling and cost analysis: Detailed recording of aczivity times and costs by functional zategories permitted calculation i n minutes per week o r c o s t s attributable t o specific functions, a c t i v i t i e s , o r service;. Aggregate time and cost variables were used i n analyses a t the experimental-group level. o Use of nonproject health services: Sample household surveys monitored the use of government and private services and related out-of-pocket expenditures i n study villages. o Project f a a i l y planning practice: Use of modern methods of family p l a ~ i n gby anyone i n the family planniag villages during the project (1969-73) was recorded by type of method and date of use. A s the services evolved during the course of the study, care was taken to ensure t h ~ reach a c t i v i t y remained within the program category prescribed f o r t h a t experimental group. The project services a r e summarized i n table 1.4 and described i n more d e t a i l i n appendix B. Services were never s t a t i c but cottinued t o evolve, so that particular a c t i v i t i e s wculd become more relevant and r e a l i s t i c . The analysis was further complicated because we did not s t a r t services i n all v i l l a g e grodpa a t the same time. Analytic model To systematize analysis of policy and program variables t h a t can promote family planning and reduce f e r t i l i t y requires a complex model. We used the model shown i n figure 1.3 t o help s e l e c t the interactions between variables to be tested gost intensively i n the analyses. Conceptually we .r -- found that we f i r s t had'to separate inputs, outputs, and outcomes. inputs i n c l u d ~two groups of variables: those taken a s given i n our study, and those t h a t we considered appropriate t o manipulate experimentally. The f i r s t group can be visualized best a s a cluster, the influence of which can either be aggregate o r individual. Identified by c i r c l e s i n the model, these a r e the variables that most family planning research has focused on because these data a r e gathered readily i n surveys and have obvious inportant - P ~ y O l a r m p r r a ~ t r r n Y t Z r a ~ ~ ~ ~ ~ k r t h Tby u u a village taadmrs~ubasrt-1~ and saperpFsto~p u -3u lrvd and incmlitp oe -FEW'S- associations v i t h f e r t i l i t y . They did not receive primary a t t e n t i o n i n our ~ t u d y ,however: the reason is t h a t policpmakers can do little t o change them, except t o be aware of t h e i r influence i n selecting and targeting i n t e r ventions. One exception is education, which we nevertheless placed i n the category of givens because it vaa outside our competence and constraints. We emphasize t h a t we do not discount the importance of these variables and that broad policy must consider them i n balancing papulation grovth with development. Our i n t e r e s t , however, was i n f a c t o r s that could be d i r e c t l y influenced i n a short time with resources available t o health and family planning services i n India. The main attention i n our analysis, therefore, was on controllable inputs, shorn i n t h e model a s squares. These variables include health and family planning services- identified by type, cost, provider, location, and time req,uired (both f o r the service provider and the people served). We have used the term, outputs, i n a limited way-referririg specifi- c a l l y t o readily quantifiable behavioral changes o r use patterns t h a t resulted from the inputs. Fo--1r categories of information were used, of which two overlap with inputs (shovn by t h e dotted line). Both prior use of family planning and talking v i t h the husband about family planning obviously were carried i n t o the project period a s . p ~ e d i s p o s i n gfactors. For analytic purposes they proved so important t h a t it seemed appropriate t o study t h e i r patterns of continuity during thezproject a s well a s before. The outputs most d i r e c t l y .r influenced by controliable inputs were the use of health and family planning -- services and the use bf specific contraceptive methods during the project. I The term, outcomes, is used here t o r e f e r t o substantive changes that resulted from outputs- - that is, variables that changed a s a r e s u l t of ~ o l i c y and program decisions. F e r t i l i t y and mortality obviously a r e the ultimate outcomes. We a l s o have included knowledge, a t t i t u d e s , and beliefs about family planning and child survival. Although these i n a sense a r e intervening variables, they represented so much of the e f f e c t we t r i e d to achieve and rceosure that they needed c l e a r identification. The greatest emphasis was placed ..n family planning practice. Because of the early termination of the project, ws did not have a s many years a s we needed to show vhether f e r t i l i t y was changing. F e r t i l i t y d a t a a r e therefore presented mainly t o support family planning data. Similarly, our analysis thus f a r indicates t h a t only a feu of the a t t i t u d i n a l variables changed i n the study period. This applies especially t o the child- surviva' variables. I n the arrangement of part two, therefore, ve have placed the practice of family planning f i r s t , the use cE a l l services next, and the outcomes i n f e r t i l i t y l a s t . The arrows indicate the subanalyses t h a t eluci- date the composite interactions. The solid l i n e s were the main focus of the analysis reported i n t h i s book; those indicated by dotted l i n e s need more analysis. POLICY ISSUES STUDIED Integration is a rather d i f f u s e term with many interpretations. The need f o r specification led us t o convert the s c i e n t i f i c hypotheses t h a t we used i n originally getting up the e x p e r h e n t a l design i n t o a series.0; s p e c i f i c policy issues. In chapter 2 the findings a r e surr-arized according t o these Issues an3 grouped around three interrelated themes. :The f i r s t of these themes is - effectiveness: - o Do integrated services increase the y s e of family planning? o What is the e f f e c t of integrated services on health and on f e r t i l i t y ? o What services should be provided, and who should provide the.,? o What manazerial and organizational changes a r e needed f o r integration? The second theme is efficiency: o How cost- effective a r e integrated services? o Whar a r e the tradeoffs between services a d sociodanographic variables i n influencing family planning? f+ % The third is equity: i+ $ o h:.zt considerations a r e involved i n targeting services? $. $ o Can integrated services do nore to reach disadvartaged groups,? o What is needed t o ;:':sin and then rnaintain community support :4 f o r serving disadvantaged groups? @ 8-b : - Obviously there a r e tradeaffs among the policy objectives implicit i n these L <3 t-2 issues. Special care is needed i n balancing considerations of efficiency and those of equity. For example, exclusive atrcntion t o cost- effectiveness i.1 promoting family planning could lead t~ a focus on upper socioeconomic groups rather than on the poor, who would receive most attention i f the a t t e n t i o n were equity. The r e l a t i v e strezyths of motivational constraints and f a c i l i t a t i n g factors thus influence both efficiency and equity- but often i n opposite directions. -. % C s* $7: - =, 1% p 1 TT1- 5;': * ax p-: 3. . ?'% zE t*< < g L $%, w" L C a r l E. Taylor, William A. Reinke, Chapter 2 .. - Rashid Faruqee, Robert L. Parker, and R.S.S. Sarma Main ae3.b.. rah Findings on Policy I s s u e s W i l l t h e p r a c t i c e of family planning improve i f s e ~ s i c e sf o r family planning e r e ~ n t e g r a t e dwith thoae f o r health? That I s the .ct:~:ral question the Narangwal popl.~lat!.on p r o j e c t examined, and the ansvcr is yes. The f i n d i n g s show t h ~ tthe benefits o f pdckages o f family ctanning and maternal and c h i l d c a r e can be great- - in e f f e c t i v e n e s s , i n er'fjcjency, and i n equity. o Effectiveness. Combined c u r a t i v e and preventive h e a l t h s e r v i c e s f o r w r e n and f o r c h i l d r e n increhsed t h e use of family placning and reduced f e r t i l l r y without s a c r i f i c i n g h e a l t h benecits. -- E Efficiency. I n t e g r a t e d s e r v i c e s were two t o t h r e e t i n e s more cost- 1 e f f e c t i v e ihan s i n g l e s e r v i c e s ic promoting family planning- again B, C \ without s a c r i f i c i n g h e a l t h benef 1:s. 1 o Equity. The u s u a l d i s p a r i:ies between socioeconomic groups i n t h e $ 1 d i s t r i b u t i o n of s e r v i c e s were reduced f o r both h e a l t h s e r v i c e s and i family planring. This was done by continued i n t e n s i v e s u r v e i l l a n c e ia of tlie underservad population (buch a s low- caste v i l l a g e r s ) and by i' 5" combinins s e r v i c e s , such as by introducing family planning i n t o - f - r o u t i n e maternal and c h l l d care. f " -- i 3. The data from tne study can be used i n designing s e r v i c e s chat w i l l reaeh i r *undercerved - .f [ peaple i n v i l l a g e s around t h e world. I r the f i r s t c k p c e r , it w a s suggested t h a t h e a l t h s e r v i c e s might P increase t h przictice of family planning i n s i x vays. ~ Two involve the h e a l t h system end the f a c t o r s t h a t influence t h e e f f e c t of s e r v i c e s , the e f f i c i e n c y (or cost- effectiveness) of programs, and the a t t i t u d e s of personnel. Arguments advanced i n support of v e r t i c a l programs have typically been based cn the assumption that focusing on a single type of intervention promotes effectiveness and efficiency. We shm, hoveter, that integrated services can i ; be more effective, more cost- effective, and =-,re organizationally e f f i c i e n t 1 than family planning service6 alone. E-c~er'e-ce i n the single- purpose family planning villagzs a l s o shows that a f t e r two years of doing only family plan- ning, village workers were i n s i s t i n g on having something nore t o talk about with village women, who becane bored by repeated discussions of the same topics. Expanding the t e r r i t o r y of responsc b i l i t y of village workers would pot have reduced the monotony of t h e i r a c t i v i t i e s and wocld have reduced eff'ciency because of the greater time needed t o travel between villages. Two other ways that heaith services might increase the practice of family planning involve the people being served and the better rapport and patterns of use t h a t integration can i n s t i l l . Combined services were more convexient for the people because they were designed t o d e a l with problems as seen i n the village home. Two important features of these services were surveillance to reach those i n greatest need and the use of entry points that . r linked naturally relcted services i n t o sequences accommdating t o the convenience of families-. - Such entry points make it possible t o use the rapport inherent i n su& services as maternity care, f o r vhich there is continuing demand, t o iGprove the use of family planning and preventive h services, f o r which spoEtaneous demand is less. The l a s t t w 3 ways that health services might increase the gractice of family planning involve the prc;umption that fuadamental changes i n attitudes and beliefs can be ercouraged by integration, especially i n relation t o the child- survival hypothesis. It had k e n postulated a t the beginning of the project that the r e s u l t s would be most quickly evident i n d i r e c t project interactions and cost- effectiveness. Next would be d i r e c t influences on patterns of use and behavior. We assumed that basic changes i n a t t i t u d e s and beliefs would take more than f i v e years :o demonstrate. Because the project was terminated e a r l y , the findings on changes i n a t t i t u d e s and beliefs were consistent with the postulated e f f e c t s but l e s s definitive. There can be no argument, then, that i f population policy and health policy can be made congruent rather than competitive, both w i l l benefit. When t h i s research s t a r t e d , the separation of family planning from health serv:ces i n many countries had led t o open competition between the two hierarchies, even though they were often ostensibly under one ministry, a s i n India. But since the Bucharest Conference on Population and the Alma Xta Conference on Primary Health Care, most countries and international agencies have made the basic policy decision t o integrate the delivery of services f o r health and family planning. There a l s o is growing agreement that population planning and primary health care must be considered i n the broader context of i n t e r s e c t o r a l development and t h a t local implementation must rely increasingly on finding ways t o prompte c o m n i t y participation and self- reliance. As ,deliver-. sys- tems f o r services a r e integrated, there w i l l be continuing need t o give spe- c i a l attention t o the broad implications of popplation growth and t o the means '4 of coordinating the policies of several ministries. -- The detailed findings in chapters 3 tg8 a r e brought together i n this - chapter t o provide decisionmakers with a base f o r judgnents about t h e i r policy options. Findings related t o each policy issue l i s t e d a t the end of chapter 1 a r e scmmarized here around th2 brosder themes of the effectiveness, e f f i - ciency, and equity of integrated services. EFFECTIVENESS The f i r s t question posed i n developing the project design was whether family planning and health services should be integrated t o increase the effective- ness of family planning services. Discussion of t h i s question w i l l be brief, however, because t h e worldwide evolution of public policy seems t o have already answered t h i s question i n the affirmative. For largely p o l i t i c a l rea- sons, almost a l l countries and international agencies have decided t h a t only integrated services a r e acceptable. The main questions now a r e w h c t services should be integrated and hov? A t Naracgwal the effectiveness of health and family planning services was measured by output, defined a s the use of specific services, an3 by out- come, the results i n health and f e r t i l i t y . The time f o r f i e l d observations was relatively short, so our most convincing evidence about the e f f e c t s on f e r t i l i t y is i n output measurements of the greater practice of family plan- ning. Heasures of f e r t i l i t y outcome were used mainly t c confirm r e s u l t s about family planning. Health indices seem t o respond more quickly t o services: both output and outcome r e s u l t s are equally convincing. The effectiveness of , e services i n reaching health objectives i: clearly borne out by the r e s u l t s of the Narangwal experiment. - .- - .* - Effect of integrated services on the use of family planning i - New acceptors pZ modern methods of contraception were recruited at* - I essentially the same r a t e i n a l l experimental groups during the f i r s t year. Then a d i s t i n c t slowdown ia acceptance races occurred i n FPED and FPCC v i l - lages during the second year: i n both groups the sloxdown presumably r e f l e c t e d t h e s a t i s f y i n g of d-nand and the f u r t h e r need t o increase motiva- tion. 11 - According t o our o r i g i n a l hypothesis, it would have taken a t l e a s t f i v e years t o demonstrate t h e e f f e c t s of improved c h i l d s u r v i v a l on family planning motivation. But t h e p r o j e c t was terminated when family planning had been provided f o r only t h r e e years i n FPCC v i l l a g e s . I n both FPWSCC and FPWS v i l l a g e s , f o u r y e a r s of family planning s e r v i c e s produced steady and contin- uing r a t e s of i n c r e a s e of new acceptors of family planning. Current- user r a t e s r e f l e c t e d s i m i l a r p a t t e r n s i n the various groups of v i l l a g e s . Three- fifths of t h e p r i o r users of modern methods of family planning became p r o j e c t u s e r s , a s d i d two- fifths of the p r i o r u s e r s of t r a d i t i o n a l methods. Fewer than a t h i r d of t h e people who had never used family planning accepted f a a i l y planning during the project. P r i o r u s e r s of modern methods were more l i k e l y t o be of high c a s t e and from f a m i l i e s i n which the husband was more highly educated and i n a s e r v i c e occupation and the wife was i n the 25-34 age group and of high parity. Subsequent discussions of e q u i t y show t h a t integrated s e r v i c e s eliminated most of these d i f f e r e n c e s i n the use of family planning. Tho r e l a t i v e e f f e c t of t h e various types of h e a l t h s e r v i c e 8 on the p r a c t i c e of family p1annir.g can be distinguished best by reviewing r e s u l t s i n terms of p r i o r p r a c t i c e of family planning. Because p r i o r nonusers. of family -1/ FPED v i l l a g e s received family planning and f a a l y planning education under the p r o j e c t ; FPCC v i l l a g e s , family planning ana c h i l d c a r e ; FPWS v i l l a g e s , family planning and women's s e r v i c e s ; FPWSCC v i l l a g e s , family planning, women's s e r v i c e s , and c h i l d care. Women's s e r v i c e s and c h i l d c a r e both comprised i l l n e s s s e r v i c e s , e s s e n t i a l l y those i n i t i a t e d by t h e p a t i e n t and delivered i n c l i n i c s , and o t h e r c a r e , e s s e n t i a l l y t h a t i n i t i a t e d by t h e project and delivered i n homes. The t e r m , i n t e g r a t e d s e r v i c e s , r e f e r s t o services i n FPWSCC v i l l a g e s . For d e t a i l s on s e r v i c e s , s e e c h a p t e r 1 and appecdix B. - 44 - plannirg s t a r t e d v i t h the lowest practice r a t e s , they tended t o shov t h e greatest increases. Children's other services t r i p l e d the use of family plan- ning by prior nonusers and almost t r i p l e d the use of modern methods by prior t r a d i t i o n a l methods. No significant e f f e c t was observed among prior users of modern methods. Similar but l e s s dramatic e f f e c t s vere observed f o r vorcen's other services. %ildren0s i l l n e s s care more than doubled family planning practice r a t e s among prior nonusers. Effects l e s s dramatic but still signifi- cant were observed f o r both children's and women's i l l n e s s care f o r other categories of use. I n contrast with other care, i l l n e s s care for both women and children produced a significant increase i n the practice of family plan- ning by prior users of modern methods. Because the provision of other ser- vices was largely i n i t i a t e d under the project, the r e s u l t s indicate t h a t they have the potential of being targeted t o those who have never used family plan- ning or who have used on17 t r a d i t i o n a l methods. People with prior experience of family planning tended t o respond spontaneously t o the a v a i l a b i l i t y of fam- i l y planning and h e a u h services; therefore, project i n i t i a t i v e seemed t o be l e s s important, and i l l n e s s care had the greater impact. I n t e g r a t i ~ n ~ f a c i l i t a t econsistent move,ment from l e s s e f f e c t i v e t o d more effective contraceptive methods- Among couples who had never used family planngng, a third became project users; am- those who had previously used '5 -- - t r a d i t i o n a l methods, fewer than half became project users; among those who had .. - p r e v i p s l y used modern methods, nearly three- fourths became project users. Eventually more than a quarter of a l l project users accepted permanent methods. Among t h i s group, 36 percent had been s t e r i l i z e d before the project s t a r t e d , 33 percent shifted from modern t o permanent methods, 16 percent shifted from t r a d i t i o n a l t o permanent methods, and 15 percent went d i r e c t l y from no use t o permanent methods. How does a l l t h i s compare with the experience of a single- purpose family planning program? Even though a preliminary review of our data suggested t h a t concentrated family planning education and services (FPED vil- lages) produced greater acceptance over the f i r s t one o r two years of imple- mentation, more careful analysis shows that these impressions were misleld- ing. The apparent rapid i n i t i a l increase i n acceptance and practice of family planning was largely due t o substitution: that is, t o the use of project sources of supply by people who were prior users of family planning. A further indication that acceptance i n FPED v i l l a g e s was l e s s e f f e c t i v e than i n other groups was i n the balance of modern methods a t the end of the project: 48 percent were condom users, compared with 29 t o 35 percent i n other v i l l a g e groups. Effect of sociodemographic and a t t i t u d i n a l variables on t h e use of family planning The sociodemographic variables, the focus of most previous research * on family planning, showed the expected relatfonbhips i n our study popula- . tion. 11 - But they were much l e s s significant than health services o r p r i o r use of family planning. Their e.?fect tended t o be additive t o the e f f e c t of ? L - health services use. The stronggst a t t i t u d i n a l influence was general expres- sion of approval of family plann%g, and t h i s a l s o had a clear additive rela- tionship. When compared with couples who were nonusers of i l l n e s s services -11 See table 1.1 i n chapter 1 f o r a list of the sociodemographic variables and the discussion on page 31 f o r a description of the a t t i t u d i n a l var- iables. and who expressed disapproval of family planning, women who used i l l n e s s ser- v i c e s and approved of family planning had twice the prc2ect use of family planning. The d i f f e r e n c e was even more marked i n comparison with those who were uncertain a b u t f a d l y planning, a group t h a t showed its n e g a t i v i t y i n many ways: f o r vomen uncertain about family planning, t h e use of h e a l t h s e r v i c e s had twice a s much e f f e c t on family planning a s d i d t h e a t t i t u d i n a l variable. Other h e a l t h s e r v i c e s had even more e f f e c t when compared with a t t i - t u d i n a l variables. Women's o t h e r s e r v i c e s had an e f f e c t three t o twelve times more than the approval v a r i a b l e , children's i l l n e s s c a r e s i x t o seven times more, and children's o t h e r c a r e f o u r t o e i g h t times more. Most clear- cut was t h e finding t h a t children's s e r v i c e s c o n s i s t e n t l y influenced t h e acceptance of family planning regardless of t h e women's i n i t i a l a t t i t u d e s and beliefs. The use of c h i l d c a r e was highest among women who disapproved of family planning and were u n c e r t a i n a b u t whether more children d i e now than t h i r t y years ago, y e t t h e family planning e f f e c t of these s e r v i c e s on those women w a s j u s t a s strong as t h a t on wcmen with more p o s i t i v e a t t i t u d e s . Child c a r e s e r v i c e s can thus be a u s e f u l v e h i c l e $or h e a l t h and family planning education. They ram change negative a t t i t u d e s and encourage family planning use. Throughout t h e a n a l y s i s an a t t i t u d i n a l v a r i a b l e thak emerged a s an '5 important i n d i c a t o r of behavior was whether wonen s a i d they had t a l k e d with - -- i t h e i r h a b a n d s about family planning. . We consider t h i s a s @proxy f o r women's I p a r t i c i p a t i o n i n family decisionmaking. It had a stronger association with t h e p r a c t i c e of family planning than any o t h e r a t t i t u d i n a l v a r i a b l e except a general expressioq of approval f o r family planning, and i t was almost a s sig- n i f i c a n t a s the use o f h e a l t h services. The strongest a s s o c i a t i o n was with both i l l n e s s and o t h e r services f o r children, with sevenfold d i f f e r e n c e s - between women who d i d not use children's s e r v i c e s and d i d not t a l k with t h e i r husbands (10 percent) and those who used services and talked with t h e i r 5 husbands (70 percent). The policy s i g n i f i c a n c e of :his finding is t h a t i t should be possible t o promote family communication about family planning by using children's h e a l t h c a r e a s a channel f o r such discussions. A demographic v a r i a b l e t h a t was consistently s i g n i f i c a n t i n its asso- c i a t i o n with contraceptive p r a c t i c e was the number of a woman's c h i l d r e n who r i had died. The r e l a t i o n &tween c h i l d deaths and r e s i s t a n c e t o family planning i k was one of the strongest i n d i c a t o r s t h a t the child- survival hypothesis works $ 1 ' i n t h e way w e postulated and t h a t it is susceptible t o manipulation by a I P d e l i b e r a t e program t o increase awareness of c h i l d s u r v i v a l t o promote the use of family planning. The r e l a t i o n is f u r t h e r defined by responses t o the question of 5 $ 1 whether c h i l d deaths a r e more frequent today than t h i r t y years ago. This f belief was s i g n i f i c a n t l y associated with use of family plann4,g? Among % nonusers of children's s e r v i c e s , t h e use of family planning was a k u t 45 8 $ 8 percent higher by those who believed t h a t fewer childreri a r e dying today than .. v by those who believed t h a t more o r t h e same number are dying. The use of k $ c h i l d c a r e s e r v i c e s overcame these d i f f e r e n c e 8 s o t h a t t h e p r a c t i c e of f a n il y s; 8 % planning was e s s e n t i a l l y the same regardless of e a r l i e r b e l i e f s about c h i l d f "- $ survival. The e f f e c t was somewhat g r e a t e r f o r those who made most use of i children's other services. I n regression a n a l y s i s , t h i s v a r i a b l e of b e l i e f s t about c h i l d deaths w a s not s i g n i f i c a n t , perhaps because of its a s s o c i a t i o n with other variables. Other socioeconomic variables of caste, income, land ownership, mate- r i a l possessions, and husband's education were much less strongly associated with the use of family planning, especially when compared with health ser- vices, which seemed t o overshadow any differences that had been present. This was especially true of children's services. Special e f f o r t s t o motivate hus- bands showed a similar (highly significant) additive e f f e c t i n association with approval of family planning and talking with wives. Further analysis checked t o see i f project services d g h r have directly modified a t t i t u d i n a l factors. The only measurable change wzs an indication that both women's services and children's services l l m ited any ten- tendency t o move toward disapproval of family planning, but they did not increase movement toward approval. Children's services nevertheless f a c i l i - tated movement toward the belief that fewer children die (p < .02) but did not specifically strengthen the belief that more survive- perhaps because of the differences between responding about something immediate and responding about something i n the long run. These basic a t t i t u d e changes would presumably have been greater i f the project could have continued longer. It was also observed that, thoughtthe consistency of responsg between the f i r s t and second surveys was only about 50 percent on most sociocultural variables, it was about 75 - - percent for the question dealing with expectations about insurance births. ** - - f And when a t t i t u d e s about insurance did change, more than three i n four changed i n the direction of needing no insurance births. t Q Effect of integr,\ted services on f e r t i l i t y A measured decline i n f e r t i l i t y a t Narangwal has t o be considered i n the context of the general decline over time i n the Punjab. Our experimental groups were kept small because of the need t o concentrate service resources . and t o get d e f i n i t i v e data. Recognition of the f i l l a b i l i t y of f e r t i l i t y measurements i n any small population led us from the beginning t o deemphasize the importance of f e r t i l i t y a s an outcome measure. O f f i c i a l government figures, analyses of our pregnancy history data, and the r e s u l t s of the ten- year follow-up of the Khanna project a l l indicate that the general secular i r a t e of f e r t i l i t y decline i n the Punjab i n the 1960s and the f i r s t half of the b f B 1970s was about 1 percent a year, compared with about 5 percent a year i n our 1e integrated service villages. Specific analyses related f e r t i l i t y t o whether a couple practiced family planning during t h e previous year. The probability of a birth among contraceptive users was found t o be half that among nonusers. Although we achieved an equitable d i s t r i b u t i o n i n the use of family planning and health services, caste differences i n f e r t i l i t y remained. I n a l l experimental groups, there were d i s t i n c t differences in' f b r t i l i t y between users and Aon- users of family planning: these differences were greatgst (p < .005) i n FPWS Fd - ! and FPWSCC villages, s i g n i f i c s n t only i n one year i n FPCC villages, and insig- - s; ni-ficant i n FPED villages. -- L A regression a n a l p 5 of variables influencing the probability of a $ birth i n any year showed that three variables were most significant: previous O i parity, the time since a previous l i v e birth, and whether the couple had prac- $ Y I ticed family planning i n the preceding year. Caste was barely significant; L. f $ education and occupation were not significant. It was estimated that i n the R f i n a l f u l l year of project a c t i v i t i e s (1973) fewer than 20 percent of women i n a l l experimental v i l l a g e s had a birth. Without contraceptive use, t h i s f i g u r e would have been at l e a s t 12 percent h i g h e r - w i t h a f e r t i l i t y rate of 224 per thousand women, n a t 198. Age- epecific f e r t i l i t y r a t e s showed t h a t d e c l i n e s were nainly f o r women over t h i r t y but t h a t f e r t i l i t y among t h e young remained · high. E f f e c t of i n t e g r a t e d s e r v i c e s on h e a i t h The number of s t i l l b i r t h s , the i n f a n t m o r t a l i t y r a t e , and the death r a t e f o r children aged one t o t h r e e were 50 percent lower i n FPCC v i l l a g e s than i n control v i l l a g e s . I n FPWSCC v i l l a g e s , r e s u l t s were more caste- s p e c i f i c , with reductions i n s t i l l b i r t h s centered on low- caste women, which seems reasonable because they were the t a r g e t population f o r p r o j e c t - i n i t i a t e d interventions. Reductions i n t h e i n f a n t mortality r a t e were centered on higher c a s t e s , probably because c u l d c a r e services depended more on parent i n i t i a t i v e i n bringing c h i l d r e n t o c l i n i c s . Child morbidity improved i n v i l l a g e s with h e a l t h care, as evidenced by t h e s h o r t e r duration of t h e most conmon i l l n e s s e s , which were reduced 14 t o 33 percent. This amounted t o a 20 percent reduction i n r o t z l i l l n e s s f o r each , C 1 c h i l d , o r an average of twenty-two days a year. Growth increased dramatically, with average w e i g h t d i f f e r e n c e s of 0.5 - kilograms and height d i f f e r e n c e s of 2 centimeters b e t w e e 2 a l l c h i l d r e n i n - - - study v i l l a g e s and c h i l d r e n i n c o n t r o l v i l l a g e s . .E Ti.? ef*t of s e x and c a s t e on growth w a s a d d i t i v e t o these p r o j e c t e f f e c t s : a high- caste male c h i l d from a study v i l l a g e was 2 kilograms heavier and 6 centimeters t a l l e r a t age t h r e e than a low-caste female c h i l d from a c o n t r o l v i l l a g e . The number of s i b l i n g s - 51 - living also had an additive effect: a child with two male and t v female s i b ~ lings was Gn average 0.5 kilograms l i g h t e r than a child with one living brother o r none. This is a powerful argument f o r the health benefits of fanr 8 i l y planning. L Considerations i n integrating services Analyses of thz timing and sequence of relation Letveen the use of health services and t h e practice of family planning showed that the associa- tion was strongest w h e ~services were concurrent. For wonen's i l l n e s s care, the relations with the number of v i s i t s i n 1969-71 was strongest for those who practiced family planning i n 1969-71, l e s s for those who practiced family planning i n 1969-73, l e s s still f o r those who practiced family planning only i n 1971-73, and least f o r those who did not practice family planning. For chi,.!renOs i l l n e s s care, the same order was found, but the range was greater. The project- initiated other services for surveillance and prevention had much more universal coverage and showed l e s s of a d i r e c t concurrent relation with the use of family planning than did i l l n e s s s e r v k e s . The only apBarent association was that new recruits to planning l a t e i n t h e project seem t o have received somewhat more children's other services during the period of successful recruitment to family planning. A converse relation - -- -early aceptance of family planning leading to greater use of health - services- -was not demonstrated. - 1 These findings strongly support the general policy developed t~-pro- mote the use of entry points to introduce family planning i n t o health service routines. A systematic process of defining a c t i v i t i e s i n the care of b t h women and children that could be used to promote concurrent practice of family - 52- planning was shown t o be r e a d i l y incorporated with t h e day-to-day work of family h e a l t h workers. The Narangwal experience a l s o showed thht d e l e g a t i o n of a c t i ~ i t i e sa s . f a r t o the periphery as possible improv s coverage and e f feceiveuess. In::dr- vention:: should be a s simple a s possible, limited i n number, kcpt within l o c a l . resources, and organized t o encouraze p a r t i c i p a t i o n of che i n d i v i d u a i ~ involved, e s p e c i a l l y t h e mothers. Mothers, t h e most important h e a l t h workers i n t h e v o r l d , provide more h e a l t h c a r e than any o t h e r type of h e a i t h worker. Their regular home respon- s i b i l i t i e s include a wide range o f h e a l t h a c t i v i t i e s t o n a i n t a i n n u t r i t i o n , t o improve hone s a n i t a t i o n and personal h e a l t h h a b i t s , and t o monitor t h e i r c h i l - dren's h e a l t h and i n s t i t u t e e a r l y treatment when needed. The Narangwal proj- e c t supports the conviction t h a t t h e most important changes nezded a r e i n r o u t i n e health p r a c t i c e s i n the home, backed up by acceptable village- level a u x i l i a r y health cave and by refer- a1 t o prcfessionnls. Kealth and family planning p r a c t i c e s i n homes were d i r e c t l y modified by systematic s e r v i c e s , with t h e r e s u l t t h a t t h e y became p a r t of a new p a t t e r n of s o c i a l behsvior. - 53 - EFFICIENCY T'hc kt tcr the balance of rervices ir an e ~ e ~ i m a n tgroup, the more s f f i - a l cient they becane. 11 By combini~gvarious e c t i v i t i e s in a s i n g l e v.isit, - economies were significant i n b!th t i m e and ffioney. I n addition, vhen the com- binations made sense i n the village home, tkere appeared tc. be a naeuial synergis;.lc i..iternctio~in creatlng rapport ~ n dcha;cging Sehavior patterns. These synergistic benefits ere possib2e i f services are developed 1.1 response t a parents' concerns and i n a mamcr designed t o promote cxuaurdty involve- ment-two throgs that canno: 3: done vitt ~ t e g o r i c a lprograms. Time i n relation t o output - Next t o ~ o t h e r s ,the most important provieera of health care a t Xarangwa,l vere fsm'ly hedtn workers. They were responsible f o r 90-95 perceut of a l f health s e m . i ~ #c o n t x t u . ~ A l l o t 2r a c t i v i t i e s i l d services were designed t o auppart: t h e i r a c t i v i t i e s . Faally health workers pli:in rhrse- fourths of the service t i v = i ~th.? projrzct. \ Yale f a d l y planning workers put i n 13 percent of the ssrvlce time; fmily health supervisors, 4 percent; and phys;cl~ns, 6 percenc f o r Ln-servize s u p e ~ v i s i o cand consultation. Ib - 1 ' "ficiency r l l a t e s outnut t o input a u l is a measure of cost-efft-ive- ness. 9utpuf can ke ciefiaed i n many ways, such a s by the nwber Br-d duratioa of s e r d c e contacts; input is d:. xed here a3 the time ~f personnzl acd t h e ccsc of resources used. 10 attempt is made t o rtisti3guish tha quality of ou?pnt on the basis of whether servi.ces a r e provided separately o r In combinatibn with others. Xor is the q u a l i t y of a \:nit f,f output distf-lguisi.?d on the basis of who uses it. I f t h e family h e a l t h workers had limit^^ .heir a c t i v i t i e s t o s e r v i c e a c t i v i t i e s , i t is estimated t h a t the populztion coverage would have been one worker f o r 2,000 t o 3,000 people. Further s i m p l i f i c a t i o n of s e r v i c e s and g r e a t e r involvement o f cornmunit] he:Lth workers would be neeled t o i n c r e a s e coverage t o 5,000 people, as is planned i n t h e n a t i o n a l program. I n i n t e g r a t e d womcn's and children's (illnesti and o t h e r ) s e r v i c e s , t h e r e was a 20-34 percent savlng i n t i m e spent on family planning compared wtth e i t h e r women's o r children's s e r v i c e s provided alone with family plan- ning. A 35 percent saving i n t i n e was observed f o r women's c t h e r s e r v i c e s provided with c h i l d c a r e ; : 2; perzent saving f o r children's o t h e r s e r v i c e s provided with women's s e r v i c e s .It is apparent that careful planning of s e r v i c e s i n i e la t e d by 9.he p r o j e c t gave considerable opportunity f o r e f f i c i e n t l y scheduling and combining s e r r i c e s . I l l n e s s c a r e tended t o t a k e about twice as nuch time ~s o t h e r s e r v i c e s i n each exoerimeltal group, mainly because s e v e r a l preventive s e r v r c e s could be c a r r i e d o u t 'I9 t h e same s e r v l c e contact. By 1374 t h e i n p u t of family h e a l t h workers and f a a i l y planning educa- t o r s t o che various s e r v i c e packages, a f t e r considerable streamlining of t h e i r s e r v i c e s , avezaged between s i x t o o i g h t hovcs of d i r e c t s e r v i c e time a week. For f a d l y plcnning a c t i v i t i e s t h e z e e k l y d i s t r i b u t i o n of time f ~ each group r was 37 m i n ~ t e si n FPWSCC v i l l a g e s ; 57 minutes i n FPCC v i l l a g e s ; 83 n i n u t e s i n - FPXS villages; and 249 n i n u t e s i n FPED v i l l a g e s . The time taken t o m a i n t ~ ? ~ f r i e n d l y r e l a t i o n s and rapport took 113 minwes i n FPED v i l l a g e s , compared . with 43-49 minutes by family h e a l t h vorkers i n t h e o t h e r experimental v i l - lages. Further evidence on the r e l a t i v e i n c r e a s e i n e f f i c i e n c y is provided by the time per average service contact Ln 1973-74, which was shortest where integration was greatest: 4.4 micutes i n FPWSCC vtllages; 5.7 d n u t e s i n FPWS villages and FPCC villages; and 14.3 minutes i n FPE:D villages. Cost i n relation t o butput The cost of integrated s e r ~ l c e sa t Narangwal ( i n HPWSCC villages) was $2.20 (%I 16.7) per capita per year. By comparison, the cost of services in FPWS villages was $1.80 (Rs 13.5) per capita per year, that in FPCC villages was $2.80 (Rs 21.3), and that In FPEII v i l l a g e was $1.20 (9s 8. 7). Because of the marked differences i n output, however, these costs should be related t o partirular benefits t o obtain a more accurate picture of efficiency. A l l such calculations unequivocally showed thc! greater cost- effectiveness of integrated services (table 2.1). These analyses permitted ca:lculations of r e l a t i v e cost- effective- ness. The cost per new family planning acceptor 14~s$12.?7 (Rs 92) i n FPWSCCvillages, twice that mount i n FPWS village3,, and three times that amount in FPCC and FPED villagee. The cost per couple- year of family planning was $10.27 (Rs 77) i n FPWSCC villages, i.7 times chat amount i n FPWS villages, 2.5 times that amount i n FPCC villages, and 3 times that amount i n FT?EI, v i l - lages. . L Detailed calculatio ; separately a l l x a t e d costs t o reluctions in mortality and t o improvements i n morbiaity and n c t r i t i o n i n FPCC villages. The cost per perinatal death averted was $9.87 (ILs 74), that for an infant Table 2.1. Comparative ef ficiency of Narangwal services * (U.S. d o l l a r s = 7.5 rupees) Control FPED FPWS FPCC F'PWSCC Government S ~ r v i c e v i l l a g e s villages v i l l a g e s villages v i l l a g e s services Annual cost per capita Child c a r e -- -- O.OSa 1.07 0.56 Nutrition care - 0.87 0.60 Women's services - 0 . 0 7 a 0.88 0.'- ' 0.60 0.08 Maternity care 0.01 a 0.44 0.11 a 0.23 - Family planning -- 1.08 0.44 0.67 0.24 0.08 Cost Der service conrac t -- Women's services - 0.36 - 0.29 0.29 0.21 C Child care -- 0.23 0.19 0.13 Maternity care 1.31 1.21 0.79 Family Planning -- 1.45 0.77 1.92 0.51 0.33 Expenditure per capita - Private services 2 12 3.05 2.09 2.20 1.76 - Government services 0.83 0.45 0.39 0.20 0.20 0.25 a. These r e f l e c t cost f o r supportive services (related t o e i t h e r the main services o r incurred t o maintain rapport). b. This figure is for curative and maternal child health care combined. c. This figure is f o r care of illaess. d. 'Ehis figure is f o r care of maternal ,and child health. e. This is the average cost of government health services within a primary health center a r e a (Johns Hopkins 1976). - Not available o r not applicable. '2 .i death (aged 0-1) averted was $37.33 (Rs 280), and t h a t f o r a child death (aged 1-3) averted was $101.47 (ls 761). The portion of the c o s t s a t t r i b u t e d t o reductions i n morbidity was calculated a s the cost of a day of i l l n e s s averted: t h a t cost was $0.53 (Rs 4) f o r an infant and $0.40 (Rs 3 ) f o r a chj' aged 1-3. Finally, by using the portion of c o s t s a t t r i b u t a b l e t o I , 'ion, we could calculate the cost per additional centimeter of growth a t - ~ i r - y - s i xmonths of age: it wan $26.27 (Rs 197). I n the integrated (FPWSCC) package, the cost component for drugs was 10 t o 15 percent, and t h a t for other su2plies was 6 percent. This supports the conclusion that these e s s e n t i a l expenses can be readily funded and indi- cates the inefficiency of continuing present practices i n government services, where the lack of drugs and supplies is the greatest s i n g l e obstacle t o effec- tive services. Similarly, 7 t o 11 percent of the costs were f o r v2hicles and transport. These expenses a r e also essential, because the supervision, refer- r a l , and support t h a t make services effective depend on the r e a l i s t i c provi- sion of transport. Finally, the allocation t o amortize building costs was only 1 t o 2.5 percent of t o t a l cosrs, o r a b u t $25 (Rs 200) per subcenter per year. The figure is lou &cause we worked with panchayats t o f i x existing village f a c i l i t i e s %rovided by the village. The people took pride i n ' providing the f a c i l i t i e s , and some healthy competition developed between villages. Most important, people f e l t comfortable u s g g the f a c i l i t y because .r i - it was. compatible with the local culture. = .I Our experience raises questions a b u t c u r r e n c p a t t e r n s of funding, i n which investments i n buildings and the pretense of providing drugs a r e mainly to get p o l i t i c a l c r e d i t f o r f r e e medical care. P o l i t i c a l leaders always want to provide items f o r wt.i:h there is exlsting demand. But people would gladly pay rrany expenses themselves. When t h e p r o j e c t was being terminated and . S e ~ i c e Swere being t r a n s f e r r e d t o t h e teaching h e a l t h c e n t e r of t h e Ludhiana C h r i s t i a n Medical College o r t o government primary h e a l t h c e n t e r s , d i s c u s s i o n s with v i l l a g e people i n d i c a t e d t h a t they were eager t o pay d i r e c t l y f o r drugs. They s a i d t h a t they had more ccnfidence t h a t t h e q u a l i t y would be good i f they paid. I f people were permitted t o choose what they c o n t r i b u t e , t h i s would permit t h e government t o u s e money f o r things t h a t people do not want t o pay f o r , such as t r a n s p o r t , s a l a r i e s , supervision, and s u p p l i e s f o r preventive services. EQUITY The t h r e e p r i n c i p a l parameters t h a t w e have used t o measure performance a r e e f f e c t i v e n e s s , e f f i c i e n c y , and now equity, which is used h e r e i n the g e n e r a l sense of providing s e r v i c e s t o a t a r g e t group t h a t is underserved o r disadvan- taged. I f adequate coverage of h e a l t h s e r v i c e s , education, and n u t r i t i o n can be provided, t h e poor seem t o be as w i l l i n g t o l i m i t t h e i r f e r t l l l t y a s t h e rich. Collaboration between t h e h e a l t h system and the community is needed t o measure c u r r e n t d i s p a r i t i e s i n need, t o adapt i n t e r v e n t i o n s t o those i n great- @stwed, and t o take the i n i t i a t i v e i n prefereneially meeting those needs. ifeasurements of e q u i t y should be based on concepts of coverage and s u r v e i l l a n c e . A l l messures of coverage r e q u i r e c a l c u l a t i o n s based on popula- I : t i o n dencminators. Surveillant-e requires identifying those who a r e most a t r i s k and monitoring key i n d i c e 3 t o i d e n t i f y problems and apply a p p r o p r i a t e interventions e a r l y . lJonen and c h i l d r e n i n v i l l a g e s , who i n t h e p a s t have been most discriminated a g a i n s t , continue t o hsve g r e a t d i f f i c u l t y i n t r a v e l i n g t o formal f a c i l i t i e s and do not f e e l comfortable when they g e t there. Therefore, it is e s s e n t i a l t h a t a l l a c t i v i t i e s be brought a s c l o s e t o the homes of the poor as possible. Because a l l e f f o r t s should be within t h e scope of l o c a l resources, these s e r v i c e s should be r e g u l a r l y provided by t h e most simply t r a i r e d and peripheral workers. . Considerations i n t a r g e t i n g services Health and fez t i l i t y surveillance- done by using lists of a l l married reproductive age women (couples) and c h i l d r e n under three- proved t o be an outstandingly important a c t i v i t y t h a t can be b u i l t i n t o routine services. A t l e a s t half t h e s e r v i c e time a t Narangwal was i n preventive services, with s u r v e i l l a n c e being the maio instrument of e f f e c t i v e outreach. Surveillance was not limited t o d a t a gathering but included appropriate responses t o p r o b lens i d e n t i f i e d f o r individuals o r f o r groups. Care was taken t o ensure t h a t outreach services were not subordinated t o c l i n i c a l emergencies, because frequent canceling of outreach s e r v i c e s t o c a r e f o r c l i n i c a l emergencies can quickly destroy t h e s u r v e i l l a n c e system. A balance must be achieved e i t h e r by reducing the s i z e of t h e population t o be covered o r by l i m i t i n g the range of c l i n i c a l r e s p o n s i b i l i t i e s . . t To obtain a l o c a l l y appropriate mix of s e r v i c e s , a basic management * question is deciding how t o t a r g e t s p e c i f i c a c t i v i t i e s t o reach p a r t i c u l a r populations. Some population c h a r a c t e r i s t i c s emerged from these s t u d i e s and can be used a s i n d i c a t o r s f o r s e l e c t i n g t a r g e t groups. F i r s t , p r i o r u s e r s of - family planning tend a l s o t o be spontaneous users of hea*h services and I therefore require l e s s concentrated outreach e f f o r t s . They w i l l come f o r ser- vices anyhow, Second, nonusers of family planning require much more focused outreach. Our evidence shows t h a t both women's and children's other s e r v i c e s were important i n reaching individuals who had not previously used o r did not approve of family planning. Behavior w i l l not automatically change with a l l outreach services, but change can be promoted by using entry points t o l i n k health and family planning services. Some demonstrated associations v i t h a t t i t u d i n a l f a c t o r s can a l s o help define target populations f o r service a c t i v i t i e s . Attitudinal variables were studied mainly f o r t h e i r relation t o family planning, but there were a l s o d i s t i n c t differences i n t h e i r associations with the use of health services. An understanding of these associations can help show how integrated services might be packaged t o reach particular groups. Women's i l l n e s s care was used most bg those who approved of family planning and l e a s t by those who were uncertain, a pattern t h a t recurred throughout t h i s analysis, showing that those who were uncertain a b u t family planning were consistently negative a b u t other services. People who s a i d that fewer children d i e now thdn t h i r t y years ago uss,d women's i l l n e s s ser- vices most. Women xho talked with t h e i r husbands about family planning, a presumable indication of a greater r o l e i n family decisionmaking, a l s o made more use of women's services. With children's i l l n e s s care, a similar pattern was observed--with the 'least use of services by famiries i n which women were uncertain about the approval of family planning and thought t h a t more children - die now t$n t h i r t y years ago. L The use of women's other services was not related t o other variables, - - !e showing that e s s e n t i a l l y complete coverage had been achieved. The only sig- .E nificant difference was that f o r women under thirty- five i n FPWS villages: 95 percent of the low-caste women received care, compared with 87 percent of high- caste wonen. Other care f o r children under three a l s o showed a good distribution, with more than 90 percent coverage and no significant differences by s u b groups. The use of children's services seems t o be almost universally accept- able, regardless of i n i t i a l a t t i t u d e s o r socioeconomic groups, and therefore proved t o be a good vay of gaining access t o resistant families f o r health and family planning education. Effect of targeting integrated s e n - I c e s The most evident s h i f t toward equity i n our findings was i n the per- centages of new r e c r u i t s t o family planning. By concentrating integrated services on those i n greatest need, d i s p a r i t i e s vere eliminated so that no differences vere significant i n the practice of family planning under the project. n e use of curative services a l s o was essentially equalized, with the only significant differences being the s l i g h t i y greater use of women's i l l n e s s services by landowing families and br f a d l i e s v i t h educatad husbands. For children's services the s l i g h t l y greater use of services by low-casre and low- income families was consistent but not signiticant. I n addftion, there was a dastinct difference i n that high- caste families conqinued t o use private ser- vices while low-caste families tended t o s h i f t largely t o w i n g project ser- - vices. - 's Project- initiated preventiv; -- and surveillance a c t i v i t i e s demonstrated the a b i l i t y of the project to move byond equality of access t o achieve equity - through preferential allocation based on need. Fewer than 10 percent of the families were not covered. For children's other services, differences were shifted i n favoi' of l o r c a s t e children ( p < .001), low-income families ( p < .Ol), landless fawilies ( p < .001), and those with the fewest material possessions ( p < .01). The d i f f e r e ~ c e si n favor of more-educated families remained, however ( p < -05)s Gaining and maintaining community support for targeted seryices --- People i n greatest need do not come spontaneoutly f o r care because of I a long tradition of psychological, geographical, and s o c i a l barriers- - barriers that make the pretense of equal access l i t t l e more than a deception when services are supposed t o be available and free. Outreach is needed t o overcome the long- standing reluctance of people t o open themselvco t o the p o s s i b i l i t y of being rebuffed i f they ask f o r help. A t Narangwal, the achievement of equitable coverage f olloved naturally when v i l l a g e leaders began to see how serious the health needs of the poor a r e and why it is i n t h e i r own best i n t e r e s t s t o have a general improvement i n health conditions. Cormunity participation grew naturally i n 2 sense of partnership between the p-cnject and v i l l a g e leaders. A simplified system of gathering and feeding b c k information he!oed i n identifying the problems i n every home and showed the groups f o r which p r o b l v s were most severe. These findings helped v i l l a g e a 1 leaders t o understand why those i n greatest need must have focused services. A natural r e s u l t was that v i l l a g e leaders then spontaneougly undertook t o - - ? convince reluctant families to cooperate i n preventive s e t u i c ~ . ' - .Wrveillance - w i l l gain public cooperation only i f i t a c l u d e s mechanisms f o r rapid response when needs are identified. When family planning entry points produced a response and a couple indicated that they were ready t o s t a r t a family Flaming method, the response vas prompt. I f complicati ns vere reported, they vere rapidly cared for. I f the regular monitoring of ill- ness and growth indicated an e a r l y disease problem o r a lack of weight gain, established routines provided appropriate interventions. Equity can be achieved most readily vhen programs are e f f i c i e n t and effective. FRAMEWORK FOR MPLEMENTING INTEGRATED SERVICES Integrated services have been c r i t i c i z e d f o r being vague and conceptually dif- fuse because they seem t o promise everything. The greatest need i n planning, therefore, is t o s e t p r i o r i t i e s and to focus limited resources i n ways that have the most effect. Froa the f i e l d experience a t Narangwal ve developed two frayneworks f o r focusing a c t i v i t i e s , one general, one specific. The general framework is a s e r i e s of eight steps that vere defined t o integrate services. The specific framevorX is a s e t of fourteea entry points used t o introduce family planning t o routine health services. Practical steps i a developing integrated services 4 sequence of eight steps can guide the development of community- baaed primary health care (figure 2.1). The f i r s t i s t e p is t o define l o c a l health problem and p r i o r i t i e s , balancing two sources of information. One source is a profe3sionally determined definition of need, using epidemiologi- - c a l methods t o concentrate on problems that a r e common, serious, and prevent- able. This should be balanced h a parallel process of defining cornunity . Figure 2.1. Sequence of Organizing Integrated Primary Care For Health, Population, and Nutrition 1. Clearly defining p r i o r i t y problems i n the l o c a l combunity R . Epidemiologically detereined needs- b. Community demand (perceived need) 2. Selecting s p e c i f i c interventions that a r e most cost- effective i n meeting problems 3. Functional analysis a. Reallocating tasks i n health team delegating responsibility t o the periphery, including compunity workers b. Shifting a c t i v i t i e s a s close t o homes a s possible 4. Retraining and supervising a l l personnel 5. Designing programs of surveillance f o r equitable coverage and defining high- risk groups 6. Establishing management systems f o r support and r e f e r r a l 7. Balancing responsibility and authority between the health system and the community 8. ' ~ d n i t o r i nand evaluating prdgrams to develop improved services ~ demand, so t h a t p r i o r i t y s e t t i n g grows out of l o c a l piaferences and con- cerns. By doing what the people want, it is possible t o educate them r b u t why a t t e n t i o n must be given t o needs that have k e n determined epidemiof~gi- cally. It helps to have members of the commuclity work on data gathering dnd interpretation t o generate more understanding of long-term e f f e c t s and the possi bll i t i e s of prevention. The second s t e p is t o s e l e c t f o r each 3 L the p r i o r i t y problem, the most appropriacc interventions t h a t can be applied under local constraints. If possible, these should be sufficiently safe, cheap, simple, and locally maintainable t o be taken over by community health uorlers, family members, o r peripheral health workers. Judgment nust be balanced i n c h o o s i ~ gbetwen methods that can be carried out i n the home and methods raquiring technical i expertise and s p e c i a l f a c i l i t i e s reached through ref ezral. The t h i r d step is t o decide who should perform the selected interven- tions. The main c r i t e r i o n f o r determining a reallocation of tasks is t:his: tasks t h a t can be readily routinized should usually be! delegated t o the p~-iphery, i f they a r e s u f f i c i e n t l y comon t o j u s t i f y the focused training and l o g i s t i c input. A t the local level, a package of perhaps a half dozen simple services can be choser. f o r rdutine implementation, service& t h a t f i t together naturally according t o what m k e s sense i n the home and how people view t h e i r - problems. T'his*reallocation of tasks requires c l e a r specification in stsnding - L orders. !B The fotuth s t e p is training (or retraining) f o r the new package of tasks. One of the most e s s e n t i a l features of a smoothly functicning team is a system 0.'supervision that is more supportive. than puni:ive and that uses professionals f o r continuing education. The f i f t h s t e p is t o have routines o.E surveillance t o monitor the t o t a l p o p d a t i o n and ensure complete coverage and preferential a t t e n t i o n t o those at greatest r i s k o r with e a r l y problems. The s i x t h s t e p i: t o provide a management system t h a t s3pports a l l s t a f f mesbers. Rather than becoming l e s s available a t t h e periphery. drugs, supplii?:,, aud foms-as v e l l a s transport, ho-ing, and s a l a r y payments Pov thcsc i u the mest: d i f f i c u l t sitt~ations-should be available even more regc- l a r l y . This support should include the r e f e r f a 1 of c l i n i c a l and public health problems t h a t c a m o t be handled locally. The se~renths t e p is t o balance respobaibility and authority. berveen what the health system and the comm*inity w i l l do. Ths eighth s t e p is t o s e t up a simplb system f o r reportiag informa- t i o n t h a t lea back -0 problen d e f i n i t 'm, p ~ r ~ g r aevaluation, and a raturu m t o the f i r s t s t e p i n a cyclic repetition. The sequcnce of steps should cnsare prompr feedback t o identify and resolve problems e a r l y with careful moni',aring t o ensure equity f o r those a t mot:: risk. Use of faml planning entry points i n routiqe service2 Much of the f i e l d e f h r t a t Narangwai wen: i n t o working aut cntr-1 Ib 1 L points t h a t b i i l d natural links between services and the d a i l y routines of family health vorkers. Figure 2.2-is a list of the fourtzen eatry points t h a t - .? - ver- defined f o r family planning services. Zle c r i g i n a l list had more than t h i r t y entry points, but f i e l d t e s b made b t p o s s i b l e t o streamline the entry I points t o those that seemed most effective. These entry points .{ere b u i l t i n t o the work records, especially the ones f o r care during ths pregnancq ALP Z f use 2.2. Fourteen Eeslth-Se.rvice Entry Points ft~rFanily Planning Mo tivatlonal A c t i v i t i e s 1. D U P I E ~t h e r o ~ r i n eferti1it.y survey of nonpregnant menstruating ellgj.t l e woEen 2. At. th? time of coafirnatiol; sf pregmrrcy 3. A t the time of p ~ 3 t a - h r . C i 0?are, If a progntncy ended i n abcrtion ~ 4. A t about the t k i ~ ~ y - s i xweek of ?regmncy, during the a n t e r a t a l t t ~ viuj. t 5. A t the fou?:eenth-day post parturn arAd neonata3. 2xarnination 6. A t the sixth- sck postparttun ex;-rmination, combLc-J ~ 5 t htlie well-- h b y checku, 7. Dtlring the f i f t h co s i x t h oontn h f t e r delivery- - before i ~ u g h t-:s f of :he vil.l-ige l e f ~t h e i r mother's home and when daughte:s-in-law retrlnled Prom the:: ut-rna: homes (this was combined with the well-k',?jy clzeckut,, immunization, ar?d weighing of t h e child. ) . Duxiag the seventh month of lactation, combined wtth the v e k - baby checkup 9. Duzing the ninth and 7ccCb month of lactation, coiibiaed witb tbl. we;. !.-ha by 2heckc? la. A a l l r c u t i a e health checkup of chil?rcn rlnd-.- chree 11. At the rout;,- ~ a i g h i n gof c!rildren under three a I;!, A t the tb?- diet.-ry adv:'.ce f n r a child was given 13. After completion of a child's basic immuni~ations 1L. After iclentir'yi;l.z o r t r e a t iug health problems, includiog malnutrition, pre9tcr; ty, anemia, coltgenital d i -ease, acciJents, and severe i l l n e s s of children .F cycre, with a box t o be checked t o indicate t h a t the family plarlning motiva- t i o n a l message had been delivered. As part of t h k s i c training, peripheral workerc rvadily learned work pattzrns designed t o help them remember routine statements of motivational messages. It was much harder to augment with integrated functions the routines of a u x i l i a r i e s who had previously been indoctrinated i n A categozical syRtem of work, but our evidence on t h i s 4 s anecdotrl rather than ~ystematic. WHAT SPECIAL PROJECTS CAN DO Ally f i e l d research makes a contribution t o human welfare only i f the f itldings- ' a r e puc i n t o practice- that is, only i f they a r e replicated. But an unfor- tunate u.3e of the term, replication, has evolved to d i s t o r t t h e meaning of the important procevs of moving s y s t e ~ a t i c a l l yfrom projects t o general use. It is unfortunate because it implies that: a project's methods can be replicated f n rccagnizable form i n general services, which rarely happens. Adaptation rather than replication s h i - I d be t h e goal. A t Yarang~ralit was recognized from the beginning that no package of servicos would ever be f u l l y impleeented elsewhere. Even though much e f f o r t weft i n t o trying t o ensure t h a t a l l intervenzions tested were inexpensive and simple enough f o r J a s s use, it was clear t h a t t o have the moat impact, - the l e s s o m from t h i s kind of research should unobtrusively permeate t h e e n t i r e h e a l t h system. The r e s e a r c h had many i n p o r t a n t influences on government s e r v i c e s , i n p a r t becausc t h e r e was a healthy and continuing interchange between o f f i c i a l s at a l l Levels during each phase of f i e l d a c t i v i t i e s . Our periodic conferences a t Narangwal produced numerous examples of s i t u a t i o n s i n which f i n d i n g s and observations v e r e rapidly implemented i n general s e r v l e e s when they met s p e c i a l needs. I n addition, t h e findings on family h e a l t h workers contributed t o f u r t h e r adaprations i n o t h e r projects i n India and t o t h e thdnking of p o l i c y groups a b l t new r o l e s f o r multipurpose and community h e a l t h workers i n n a t i o n a l services. The lorrgrange p a t t e r n of implementation planned a t Narangwal was t h a t findings from t h i s research would be applied next i n demonstration and t r a i n i n g p r o j e c t s i n various regioos and s t a t e s . I n adapting procedures and methods, t h e emphasis would then b ~ v et o be even more d i r e c t l y on placing a c t i v i t i e s i n t h e framework of regular services. Services cculd be s t r e a n r l?..ied and r e d i s t r i b u t e d , with c o m n i t y h e a l t h workers taking on much of t h e r e s p o n s i b i l i t y t h a t family h e a l t h vorkers had a t Narangwal. I n t h e years & t s i n c e the Narangwal f i e l d wo:k stopped, s e d e r a l dozen good demonstration -p r o j e c t s have applied and expanded ouz findings. - '.i This research has l e f t us with a sense of optimism a b u t the possi- i z b i l i t i e s of providing both family planning a ~ hde a l t h c a r e t o v i l l a g e peo- h E p l e . Methods a r e a v a i l a b l e t o a c c e l e r a t e the :urrent d e c l i n e i n both f e t t i l - i:y and mortality. They a r e f e a s i b l e and c o s t - z f f e c t i v e , but they f a c e t h e long, hard :ask of implementing much of the r h e t o r i c of r e c e n t years. Of special significance is the potential, demonstrated a t Narangwal, f o r equitable distribution: the most-needed services can reach the most-deprived peo;)le i n villages. j r That the Narangval project was conducted i n an area of rapid socio- economic development naturally raises some question about the relevance of our f [ findings t o other parts of India and t o other countries. The r e a l i t y , 1 I however, is t h a t development is now rapid i n a t l e a s t half the s t a t e s of India and i n many developing countries. With the increasing pace of development i n a l l areas and i n more countries, our r e s u l t s should be applicable soon, i f i they a r e not already. The e f f e c t s w e demonstrated i n l e s s than four years can be achieved elsewhere, even i f the time t o produce change has t o be longer. General principles derived from t h i s analysis of interactions between the com- ponents of health and family planning services w i l l , of csursc, require appro- priate adaptation i n l o c a l implementation. Our prediction, however, is that the quantification we have pi-oduced-of such things a s r e l a t i v e cost- effec- tiveness and the indices of the e f f e c t s of specifi: a c t i v i t i e s - -w i l l be f a i r l y representative of average conditions i n most developing countries f o r the next two decades. As ire used t o say a t the Nar~a@dlConferences, our objective has been t o provide a c a f e t e r i a of specific health measures with suggections - f o r implementation s ~ with pr$$e tags, a c a f e t e r i a from which planners and d - i administrators Cali selecL the bst combination t o meet the needs of t h e i r - !c area. - PART TI THE DETAILED FINDINGS The main question examined i n the Nerangwal population research is whether integrated services increase the use of family planning. In chapter 3 find- ings pire presented t o show that they do. Differences i n the use of family planning are then analyzed to determine how much of that use can be attributed to different combinationd of praject services. I n addition, the chapter investigates intervening and exogenous factors associated -with effectiveness i n family planning. Because integrated services could divert attention from heelth ser- vices, i t was a l s o important to show that there was no s a c r i f i c e i n health services and benefits. I n chapter 4 it is shown that there was no such sac- r i f i c e and that i n a n y instsnces there was an hprovernent i n health benefits. Given the emphasis of t h i s research on f a m ~ l yplanning, we naturally wanted to explore its determinants and those of f e r t i l i t y . I n chapters 5, 6, and 7, we analyze the e f f e c t s of services on family plannjng and f e r t i l i t y ; the effects of attitudes, beliefs, and s~ciodemographiccharacteristics on family planning and f e r t i l i t y ; and the effects of these determinants on each other . Because the greater effectiveneba of integrated services would be l i t t l e cause for elation -.f they were not more cost- effective a s well--and i f they did not reach the groups disadvantaged mcst by poor access t o ser- vices--we also examined the efficiency and equity of Narangwal services. The detailed findings on efficiency and equity a r e reported i n chapter 8, the f i n a l chapter. Chapter 3 R.S.S. S a m and Rashid Faruqee . The Use of Family Planning m e practice of family planning was the most important outcome measure i n the Narangwal experiment. In t h i s chapter trends i n the acceptance and practice of family planning i n the four experimental groups of v i l l a g e s a-e compared. -11 Differences i n achievements a r e analyzed t o determine how much of the use of contraception can be attributed to the various combinations of project ser- vices. The influence of prior use of t r a d i t i o m l or modern methods of family ! planning is carefully assessed, a s is the s h i f t from these methods :o those supplied under the project. This chapter also investigates intervening and ? exogenous factors associated with effectiveness t n family planning, factors t h a t f a l l under four headings: a t t i t u d e s and beliefs; knowledge and prior _ practice of family planning; socioeconomic status; and demographic factors. 1 USE OF CONTRACEPTIVES In the Narangwal experiment several modern methods of family planning were offered, some temporary, some permanent: condom, intrauterine device (IUD), o r a l p i l l , injectable Depo-provsra, vasectomy, and tubectomy. These w i l l .I C I t be referred t o a s project methbds. Shifts i n the use of contraception were , -I recorded when coup!es switched from one method t o another, even i f they waited f o r some time without protection before again using the same method o r another ., method. About two-thirds of a l l project users confined t h e i r contraceptive practice to one segment of use. On the average, a contracepting couple had , - 11 A s is discussed i n chapter 1 and i n greater d e t a i l i n appendix B , the service packages included different combinatious of services: family planning (FP), wonen's services (WS), c h i l d care services (CC), ~ n d f mily planring education (ED). The i n i t i a l s are combined i n identifying . le four experimental groups by the services provided: FPWSCC, FPkS, ,'CC, and FPED. 1.5 segments of use during the project, v i t h 1.6 i n FPWSCC villages, 1.5 i n FPWS a d FPCC villages, and 1.4 i n FPED villages. Two simple indices of contraceptive use a r e the current- user rate and the e v e r u s e r rate. The current- user r a t e is the prevalence of couples I Y 1 practicing contraception a t a specified t i m e . The e v e r u s e r r a t e includes i prior users a s w e l l and r e f l e c t s cumulative acceptance of family planning. E The e v e r u s e r r a t e can apply t o those who used contraception a f t e r the a t a r t of the projcct or it can include contraception before the project. Acceptance of the project's contraceptive methods included only those who used project sources f o r sup plies and services. Since project methods were only modern, the e v e r u s e r r a t e f o r the project included only modern methods. I11 estimating the preproject e v e r u s e r rate, the use of t r a d i t i o n a l methads was considered and the transition to project (modern) methods was analyzed. D a t a w:re systematically collected on c o u p l e s - 4 t h vives in a current married s t a t e and i n the 15-49 age group--ho practiced modern contraception. Acceptance of project contraception Family planning services were s t a r t e d a t different times i n the four experimental groups: September 1969 i n FPWSCC and FPWS villages, December 1970 i n FPCC villages, and June 1972 i n FPED villages. The services . I were phased i n gradually, with methods f o r integrat4-ng services i n a package being worked out i n one v i l l a g e and then extended t o the others i n that group. - - I n 1969, when the 'project started, it took four t o six months t o get services - t o a l l v i l l a g e s inzan experimental group. It took somewhat less time to phase !e i n the third g r o u p l n 1970 and the fourth group i n 1972. Considerink only modem methods of fami?;- planning, acceptance rates a t the beginning of the project were between 12 2nd 26 perc .nt i n the expzri- (figure 3.1). The FPWCCC and FPWS groups, a f t e r more than four years of family planning services, had acceptance r a t e s of 51 t o 54 percent. The FPCC group, with more than three years of family planning services, reached 46 percent. A l i t t l e l e s s than two years of project operation i n the FPED group produced an acceptance r a t e of 37 percent. Data w i l l be presented t o show t h a t much of the acceptance i n FPED villages was associated with a combination of substitution from uniquely high preproject use and the higher levels 06 development i n these villages. I n the control villages, a repeat cross-sec- L tional survey midway through the project i n 1972 showed no change from the i current- user r.lte of 9 rercent before the project. L G The acceptance curves In figure 3.1 a r e f o r a l l acceptors; no dis- b r t tinction was made between preproject users and nonusers of modern methods. i, A s w i l l be shown l a t e r , the mix of preproject users and nonusers of modern methods varied i n the four experimental groups. Since the e f f o r t needed t o r e z r u i t acceptors is clearly greater f o r those who had not previously used family planning, we separated the two categories of project acceptors. Figure 3.2 shows acceptor r a t e s for couples t h a t used modern methods f o r the f i r s t time o r s t a r t e d t o use again a s a r e s u l t of project a c t i v i t i e s . They may have used traditional o r indigenous methods e a r l i e r . Cumuiative project acceptor r a t e s a t specified times, computed f o r the four experimental groups * . I and i n comparison with figure 3.1, show t h a t although the overall acceptor r a t e s were higher i n FPWS villages than i n FPWSCC villages, t h e project - *s acceptor r a t e s were higher i n FPWSCC villages; &he gap between the curves for - - .e FPCC and FPED villages, which closed toward the'end of the project period i n !P I) figure 3.1, remained essentially the same i n figure 3.2. The much greater source- substitution by prior users i n FPWS and FPED villages accounts f o r the d i f f e r e n c e s between f i g u r e s 3.1 and 3.2. We f e e l t h a t f i g u r e 3.2 i s a more accurate portrayal of the project results. - F i g u r e 3.2 CUMULATIVE RATES OF PROJECT ACCEPTORS I N EXPERIMENTAL GROUPS The results a t Naraagval compare favorably with the achievements i n family planning by other experiments and special projects i n India. For I example, family planning acceptance r a t e s i n the integrated health services project a t Jamkhed rose from 2.5 percent of e l i g i b l e couples i n the preproject period t o 50 percent over a period of f i v e years. Project M r a j i n Maharashtra reported a threefold increase i n family planning acceptance a f t e r three years. The e f f e c t s of the nationvide family planning campaign had a major influence on the responses i n each project, but the s i m i l a r i t i e s i n r e s u l t s indicate vhat can be achieved by special effort. Practice of project contraception The acceptAnce curves, e s s e n t i a l l y p a r a l l e l i n the f i r s t two years of project a c t i v i t i e s , indicate t h a t the project succeeded i n getting couples t o s t a r t contraception more o r l e s s a t the same r a t e i n all the experimental grcups. More importart, however, is a comparison of trends i n continuing- user rates. The proportions of women practicing project contraception a t specified times a r e shown ia figure 3.3 f o r the four experimental groups. The curves f o r tkAe FPWSCC and FPWS groups had more o r l e s s l i n e a r increases; t h e curves f o r the FPCC and FPED grocps had an i n i t i a l p a r a l l e l r i s e and then tended to plateau i n the second year. These visual impressions of the family- planning- practice curves qere checked by f i t t i n g quadracic equzitions t o obtain s t a t i s t i c a l verifica- '5 -- i i o n of the onset of plateauing. The significacce of quadratis terms i n the - - &lynoudals fittec' to practice r a t e s coufirm t h e p l a t e a ~ i n gi n the FPCC a t 8 . FPED groups. The quadratic terms i n the regression equations f o r the FPWSCC PEdCENT m P HARRIED AGED 15-49 Mi0 k i PUCTICLYG FAMILY PLANNING BY EilPERIENTA3, GROW acid FPWS groups were not significant. The smootbed cbrves ai? s h o w ia figure 3 . 4 . Similar tests of evcru-ser r a t e s ; :wed uo s i g e l f i c a 2 t differences. These findings f i t v e l l with the pattern of curves we ariginally p ~ s t u l a t e d . We h.:d c x p c t e d plateauicg at various revels a f t e r an i n i t i d rapld increase to meet existing demand. In the FPCC group be expected a lag period, a f t e r whicn the child- survive1 influence i - ~ u l dbecome evidznt and the carve would begin to r i s e again. Adjustments far differences i n the effectiveness of contraceptive methods - The demographic k.:LC of thd project was greatly influen-ed by th: ccntraceprive d x i n experimeucal giOUpS- To calculate practice r a t e s that r e f l e c t the expecled demograpnic imp-ct from these different MXeS of fanily planning methods, we developed a new measuze t h a t adjusted f o r differences t i n the e f f e c t i v e ~ e s aof contraceptive methods. W e called it the effective- user rate. The mix cf :ontraceptivss differed greatly f o r the four experi- mental groups and changed from t h e to time i n each grmp. Table 3.1 shows that 'oward the ecJ of tho project almost half those practicdn3 conbrncep- t i o n i n the FPED group ware still olzly usicg condoms. A t the other extreme, nmre than 40 percept of chose practicing ccntraceptim a t the end of *the prqject i n the FPCC group =ere s t e r i l i z e d . So the expectations f o r dmo- graphic impact would be reduced f o r the FPED group ana increased f o r the FPCC --- '9 - group. Particularly impressive were the high general demand f o r i n jectionu a@ the rapid response t o the a v a i l a b i l i t y of Cepo-prsbera, ew=n though it I produced a high r a t e of amenorrhea and intermit.tent bleedirrg. The variation in use seemed t o be g r e a t l s t f o r LLVs. Our impression is t h a t the use of Figure 3.4 ,* I ( b & 1 1 Table 3.1 Peroentage Diekr:bhcion o f Current Userv o f Family Planning by Method Currently Beiny Ueed a8 on maroh 31 o f &oh Calendar Year by Experimental Croup I U D s was affected by l o c a l tumors about complications ( a f t e r the government's mass program some f i v e years e a r l i e r ) and by the differing a t t i t u d e s of s t a f f members. The highest r a t e of using IUDs was in- FPWS villages. That may have been the r e s u l t of s t a f f cooperation with nearby government primary health 2 i centers t h a t had effective programs. Table 3.2 shows the pregnancy r a t e and r e l a t i v e effectiveness f o r each contraceptive oethod f o r all users a t Narangwal. The pregnancy r a t e is the number of pregnancies per hundred woman-years of use; the r e l a t i v e effec- tiveness is equal t o 1 - [pregnancy rate/pregnancy r a t e (40.7) f o r no contra- ception]. Condoms and o r a l p i l l s vere only about 30 percent effective; IUDs and Depo-provera vere about 90 percent ecfective; s t e r i l i z a t i o n was 97 p e r cent effective. I n estimating t h e e f f e c t of differences i n contraceptive mix, prac- t i c e r a t e s vere adjusted on the basis of the r e l a t i v e effectiveness of each method used. -L/ This yielded an estimated r a t e reflecting the proportion of married women protected from pregnancy. The curves of effective- user r a t e s i n f i g u r e 3.5 a r e both lower and f l a t t e r than those i n figure 3.3. F i t t i n g quadratic equations t a these data again showed d i s t i n c t differences. Effec- tJve-user r a t e s had s t a r t e d t o plateau i n the FPCC and FPED groups. There was. no evidence of claixsuing i n the FPWSCC and FPWS groups, even though they had -1/ An adjustment factor was const-cted f o r each experimental group a t each point i n time. The r e l a t h p effectiveness r a t e f o r each method, obtained e a r l i e r , was applied te the number of users of that method t o get the number of effective users f o r t h a t method. The number of e f f e c t i v e users was obtained by summing effective users of a l l methods. The adjustment factor was then obtained a s the r a t i o of effective users to tne t o t a l users. - - 84 Table 3.2 PREGXANCY RATE AND RELATIVE EFFECTIVENESS OF CONTRACEPTIVES Contraceptive method Pregnancy rate -a/ Relative effectivenees-b/ No contraceptives Condom Oral p i l l IUD Depo-provera Vasectomy Tubectomy a. Pregnancies per 1CO women-years of use of contraceptive method. b. Equal to 1 - [pregnancy rate/pregnancy rate (40.7) for no contra- ceptives]. - Figure 3.5 P g R m -Y JLAaaaLD WOIYQI AGE l H 9 WHO WERE " X F F E ~ U S E R S n 09 PAMaY PCMNDtG AT SPSQRED PO= BY -AL GROUP J S D M J S D M J - S D M J S D M J S D M 1969 1970 1971 1972 1973 1974 t been exposed t o project services f o r a considerably longer time and bad reached considerably higher effective- user r a t e s (figure 3.6). Adjustment f o r differences i n duration of proyect Because the four groups were not exposed t o services f o r equal periods, the achievements i n family planning cannot be directly compared: project performance i n March 1974 reflected the differences i n duration of project services. About twenty-one months of project operation produced ever- user r a t e s (of modern methods) of 36 percent and 37 percent i n the F?CC and FPED groups, compared with 31.5 percent and 33.1 percent i n the FPWSCC and FPWS groups. Continuing-user r a t e s i n the FPCC and FPED groups, which had a l a t e r s t a r t , were a l s o higher (27.3 percent and 29.8 percent) than those i n the FPWSCC and FPWS groups (21.6 percent and 25.2 percent). Because of the slower start- up i n the f i r s t two groups, the differences between the experi- mental groups a t twenty-one months probably aye unimportant. This ranking is supported by the effective- user rates, which take into account the differ- ences I n effectiveness of contraceptive methods, differences that were l e s s i n groups having a l a t e r s t a r t . The higher r q t e s of acceptance and practice were evidently achieved by more aggressively promoting the l e s s effective contraceptive methods. s h e data presented thus f a r have not taken contraception before the project '5 Pnto account. Acceptors of project contraception f a l l i n t o three categories: C * B .. #ose who had used modem contraceptives, those who had used traditional , methods of contraception, and those vho had not used contraception. The distinctions a r e important because they show how great an cffect the project had. For example, the biggest s h i f t i n use is from nonuse before the project Figure 3.6 S D M $ S D M J S D M J S D M J S D d - 1969 1971 1972 1973 1974 --- l % Q 11- Shrch J J~lln 9 sbptcmkr D I k a m b a t o the use of modern methods under the project; t h e smallest, from p r i o r use of modern methods t o modem acc'ods under t h e project. The second is merely a s h i f t i n the source of supply, but it may also include a s h i f t t o more e f f e c t i v e methods. Of t h e twenty-six v i l l a g e s covered by t h e Narangwal study, f i f t e e n i i received family planning services, eleven did not. Continuing information on contraceptive practices was-obtained i n t h e f i f t e e n v i l l a g e s wfth services. Cross- sectional surveys were a l s o used t o obtain information on family planning practices i n a l l v i l l a g e s i n 1968-69 and i n 1971-72. To measure preproject practice of family planning, t h e 1968-69 cross- sectional survey vas conducted i n the eleven v i l l a g e s t h a t received no family planning add i n the eleven v i l l a g e s i n which services s t a r t e d i n 1969 and 1970. For t h e four FPED vil- lages i n which project a c t i v i t i e s s t a r t e d i n 1972, t h e 1971-72 s-urvey was used t o measure preproject contraception. Information was obtained on whether, when, and what method of family planning had been used. About half the women responded t h a t they o r t h e i r husbands had already practiced some kind of family planning. Of these couples, about a t h i r d had used modern methods, two- thirds t r a d i t i o n a l o r indigenous methods. Table 3.3 shows the d i s t r i b u t i o n and use r a t e s of indigenous methods i n 1968-69, before project a c t i v i t i e s started. The most commonly reported . , C methods i n current use were withdrawal, abstinence, and s a f s period (based on l o c a l c u l t u r a l b e l i e f s that the uterus a t t e r menstruation b l i k e a freshly - .* plowed f i e l d , waiting f o r seed and then slowly closing by d d p e r i o d , when it - - is thoGght to be no longer fertile). Only a few people of &he many who knew I of indCgencus medicines and herbs said they used them. A variety of home methods, such as washing, douching, sponging, and simply squatting, were considered useful but seldom used. Many people knew about abortion, but on t h i s f i r s t survey we did not f e e l f r e e t o ask about its use. Many people C C I a I also knew about condoms from army experience, but condoms were considered something t o be used more f o r preventing veneral disease than f o r contracep- tion. The national campaigns f o r IUDs a d s t e r i l i z a t i o n had produced a high awareness of these methods, but few people admitted t h a t they had used than. We l a t e r found that prior use was greater than i n i t i a l l y reported. A comparison of preproject use of permanent, temporary, and tradi - tional methods of family planning is shorn f o r the four experimental groups i n figure 3.7. Considering both modern and indigenous methods, the prior-use r a t e s were similar i n the villages where family planning services were offered (49.4 percent) and i n those where no family planning servic s were offered (52 percent). The prior- use r a t e s i n the four experimental groups ranged from 46.4 t o 51.2 percent, but the differences were not significant. The overall preproject use of modern methods was 17.6 percent but varied greacly among the four groups (reflecting t o some extent the t i m e difference i n the s t a r t of services i n the d i f f e r e n t groups): the r a t e was 26 percent i n the FPED group, 19.1 percent i n the PPWS group, 18.1 percent i n the control group, 11.9 percent i n the FPCC group, and 11.7 percent i n the FPWSCC group. Modern contraceptives accounted f o r about half the prior contraception i n the FPED group and more than a t h i r d i n the PPWS and control groups. These villages . I were selected t o be as close a s possible to primary health centers because of an e t h i c a l consideration: - w e did not want to have study v i l l a g e s without - access to child care. Siqce family planning w a s the main outreach a c t i v i t y of C primary health centers, tlpir impact w a s greatest i n these nearby villages. b In the FPWSCC and FPCC grmps, modem contraception w a s used by only a fourth of those who had practiced family planning, and the use of t r a d i t i o n a l methods w a s correspondingly greater. A feature of preproject contraception is the frequent combination of aethods. Thirty-two percent of the couples had used o i l y traditional methods; another 11 percent had used boeh modern and eraditional methods. Where family plan'aiq services were offered, modem-temporary methods were used by 4 p e r cent of couples, but these methods were used i n combination wieh other methods by 10.6 percent. Similarly, permanant methods were used by 1.8 percent of > . couples, and another 2.3 percent used permanent methods a f t e r using other methods. SOCIODEIYOWHIC DIFFERENCZS I N CONTRACEPTION BEFORE THE PROJECT Because of growing attention to the equitable distribution of services, meas- urements of coverage before and under the project a r e of special interest. Five sociodemographic variables were andyzed: religion- caste, education of husband, occupation of husband, age of wife, and number of living children. Religion- caste Before the project, contraceptive practice w a s l e a s t (45.7 percent) f o r scheduled Sikhs (Sch-Sikh, the lowest castes), next highest (50.9 percent) for the Jat- Sikhs, and highest (54.4 percent) for other c a s t e groups. The II main reason f o r these differences were highly significant differences i n the use of modern-temporary and permanent me;-;hods. The use of traditional methods - - was not significantly different i n the three groups. *a .S Education of husband The husband's education wao also related to ti?e pr3ctice of family plaming, uith nonusers raxrging from 51 percent with no formal education to 30 pt rcent f o r those with eleven or more grades of eaucation. These highly significant differences werc e n t i r e l y i n the use of modem contraception. For both traditional and permanent methods, practice r a t e s vere not s'gnificantly d i f f e r e n t f o r educational groups. Occupation cf husband Preproject contraception alsc differed by occupational g;oup, ranging from 55 percent f o r services and others t o 50 percent f o r farmers and 46 percent for laborers. These categories paralleled the educational and c a s t e distributions. The we of traditional methods was almost equhl i n the four groups. Age of wife --- The age of the wife a t the t h e of the baselino survey was an extremely powerful indicator of prior use. Preproject contraceptive practice was lowest by women under 25 (41.5 percent), highest by women aged 25-34 (56.9 percent). Half the women over 35 had been prior users. Again, modern-temporary methods showed the greatest variation ( 1 8 percent i n the 25-34 group, 12 percent i n the other two groups). The use of permanent methods increased progressivel, with age, but no differences were foc-d i n the use of t r a d i t i c n a l methods. t Number of living children The d i r e c t association between the number c f children a t the time of the preproject surrey and prior practice of family planning was more signifi- cant than any of the other variables. Thirty- eight percebt of couples had E fewer than thrze children, 45 percent had three to f i v e , 5nd the remaining 16.7 percent had more than fiue. Preproject contraception (of modern and traditional methods combined) was l e a s t (38 percent) among women who had fewer than three children; it was 58 percent f o r the other two groups. The dif- ference i n practice of traditiotial methods was significant, increasing from 26 percent (for those with fewer than three) t o 37 and 34 percent (three t o five) (nor2 than five). The use of permanent methods increa3ed from 1 ,krcent f o r couples with fewer t h ~ qthree children t o 5 pey-ent f o r those with three t o five, and t o 10 percent f a r those with six ur more. Prior use of modern- C temporary methods, which presumably includes spacing, was highest f o r couples with three t o f i v e children, lowest for those with fewer than three. TRANSITION FROM PREPRaTECT TO PROJECT CONTRACEPTION Information on prior contraceptive practice w a s available for 2,603 couples i n the f i f t e e n v i l l a g e s where family plannixig services were offered and f o r 1,339 couples i n the eleven villages where no services were offered. I n the f i r s t s e t of villages, 49.9 percent of the couples reported t h a t they practiced one o r more methods of contraception (modern or traditional) before the project; i n the second 52 percent. For general analysis, preproject contraceptive use was grouped i n t o four categories: permanout methods, modern-temporary methods, traditional methods, and no method. Because no change was possible f o r the use of permanent methods (vasectomy and tubectomy), only the remaining catego- a . I r i e s have been analyzed. I n the four experimental groups, 53.9 percent of preproject users of - - . temporary methods (traditional oreodern) practiced contraception during the i project. The t r a n s i t i o n from preeroject t o project use was highest i n the !B FEWS group (61.1 percent) and lowest i n the FPED group (49.5 percent), probably because of the length of exposure t o project a c :.vities. ~ The most important finding is the r e l a t i o n between the probability of becoming a project contraceptor and the type of contraception used before the project. I n all s e m i c e villages only 35.3 percent of couples who had e a r l i e r used no contr.septioo became contraceptors during the project, compared with 46.2 percent of :ouples who had used- traditional methods and 72.7 percent of couples who had used modern temporary methods. Transition from preproject to project use showed progressive s h i f t s t o mors effective methods. Table 3.4 shows the pattern of transition f o r categories of project contraception: nonuse, modern-temporary, permanent, and modern-temporary followed by permanent. To the last three categories of project contraception, transitions were greatest when preproject contrace2tion included modern methods, and transitions were l e a s t from those with no preprogram contracep- tion. Patterns of t r a n s i t i o n t o permanent methods a r e especially important. 8 Twenty-four percent of the women eventually used permanent methods. Of these, 36 percent had s t e r i l i z a t i o n s before the project started, 33 percent shifted from modern methods, 16 percent shifted from t r a d i t i o n a l methods, and 15 percent had not previously used family planning. Of the project contraceptors, 61.6 percent v t r e prior users. The FPED group had the highest transition r a t e (70.5 percent), t o be expected because of the high r a t e of prior use of modern methods. The transition r a t e s f c r the FPWSCC, FPWS, s.%dFPCC groups were between 59.3 percent and 57.4 a percent. Included'in these figures is the r a t e of substitution from moderr meth~ds,mostly a change i n the source of supply of contraceptives. Overall, t h i s r a t e was 29.5 percent, but the differences between v i l l a g e groups were dramatic: 48.1 percent f o r the FPED group, 31.4 -percent for the FPWS group, t 21.5 percent £01. the FPCC group, and 17.9 percenr for the FPWSCC group. But when the calculations a r e base&on the proportion of a l l preproject users who started project use, the transit ,,n from preproject t o project use was highest f o r the FPWS group, lowest f o r the FPED group. . Table 3.4 - - --- . Transition fremr ? d o t U u to m 8 m Csa by Typm of Omttacmptixm "'Ritliod(s) Fn th. PP-S.rPicr ViUag.9 Wn-Use of Use of Pmgranr Wth& P,=gram m+bodr* mdern and :.Warn per-=anent Permanent (Temporary) !- None 1 852 401 44 19 I316 I ' (64.7) (30.5) (3.3) (1.4) (100.0) 450 317 47 73 837 _ '(53.6) (37.9) (5.61 (2-7) (100.0) * Includes use of traditional and fndiganotu mathods as wall as non-use of any imtbd.8- The e f f e c t of the project on f e r t i l i t y presumably was greater where the r a t e of substitution, e3pecially of source- substitution, was lower. Orr t h i s presumption, the project i n the PPUSCC v i l l a g e s had the biggest 4 e f f e c t i n recruiting new cases; that i n the FPCC came next, followed by t h a t i n the FPUS villages. The project i n the FPED v i l l a g e s had considerably l e s s effect. What seemed a t f i r s t t o be a remarkably rapid acceptance of family planning i n FPED villagee vas therefore largely due t o high r a t e s before the project. Sociodemographic differences associatcd with the t r a n s i t i o n from preproject use of traditional and modern tanporary mezhods a r e presented i n table 3.5. The important finding is that patterns of transition were much the same f o r the various social and demographic subgroups of the population. Transition from the use of modern methods before the project occurred e q m l l y , regardless of religion- caste, education of husband, occupation of husband, age of wife, o r number of l i v i n g children. Transition from the use of traditional methods also did not show a significant association with any variables, except the age of the wife. Ccuples with wives over thirty-fivk had a lower transi- t i o n r a t e from no use t o project use. A major purpose of analyzing differences i n fam$ly p l a m i n ~p a c t i c e before and during the project was-to identify subgroups on which project e f f o r t s could be concentrated. These data give a quantitative base for judging probable patterns of impact. I f the main objective is t o increase 's acceptance rapidly, ?t is e a s i e s t t o reach couples who previously used modern -- methods of contracepHon, more d i f f i c u l t t o reach preproject users of tradi- - tional methods, and most d i f f i c u l t to reach nonusers. But i f the objective is =o equalize contraceptive practice among subgroups of the population, inte- grated services can effectively cover subgroups underserved before the p r o j ~ c t , such as scheduled castes, lower educational levels, laborers, younger women, and women with fever living children. Table 3.5 Socio-Damographfc D i f f u e n t t a l s in Tranairion fraa Pra-Program to Progranr Practffo of Contraception Traditional ivethods .WenTemporary.%athods 1 No. Pro- No. griftad No. F m - No. Shifted soci~-~snaographi~ Program taproqcam Progr- '=-a= Groups U s s r s Hod.H.thod.. Psrcaat U s s r s S0uc.s Psrcont 'ReUgion-Cas ta: C . .t : I Jat-Sikh 386 174 45.1 199 144 72.4 Seh-Sikh 294 i - 139 - - 47.3 92 70. 76.1 Others 151 70 46.4 57 41 72.0 1 Xt2) 0.33 p .3504 x;~) 0.51 p = .7760 I Education of Husband: 0 473 210 44.4 1-5 95 46 48.4 6-10 204 105 51.5 11+ 22 11 50.0 X:3) = 3.07 p = .3916 Occupation of Husband: Farming 317 149 47.0 Labor 221 109 49.3 Service 102 - 48 47.1 Other '157 68 43.3 x:~) 1.33 p = -7224 !Age of Wife: i <25 !1 157 90 57.3 63 46 73.O 25-34 297 161 54.2 153 122 79.7 35+ 298 98 32.9 109 73 67.0 I ~:2) = 36.70 p < -010 v2 *(2) -5.47 p = .0688 ! , II ATTITUGES, BELIEFS, AM) PRACTICE The research objectives included specific attention to a t t i t u d i n a l influences on the practice of family planning. Associations have been s h o w between c e r t a l n a t t i t u d e s about family planning o r b e l i e f s about child survival and contraceptive behavior. Our results do not f i t the general opinion t h a t a t t i t u d e change is a prerequisite f o r practice. An e f f o r t has been made t o t e s t the possible causal relations through ~ e q u e n t i a lanalysis. Attitudes and beliefs were measured by verbal responses t o two cross- sectional surveys i n a l l the villages. (FPED v i l l a g e s were excluded from t h i s analysis because of the short period of observation.) Analysis has been focused on the responses of 1,327 women whose a t t i t u d e s and b e l i e f s were recorded i n the f i r s t cross- sectional survey; who were currently married a t l e a s t u n t i l the second cross- sectional survey about two years l a t e r ; who were from villages where family planning services were offered during the project; and who had not practiced contraception up t o the time of t h s f i r s t survey of a t t i t u d e s and beliefs. They were asked whether they approved of family planning and whether the7 approved of tho use of family planning by newlywed couples: 53.1 percent of the women i;?r;-oved of family planning, 29.5 percent , disapproved, and 17.3 percent were uncertain; 39.1 percent of the women @ approved of family planning for newlywed couples, 31.6 percent disapprcved, and 27.7 percent were not sure. Of the 1,327 couples i n the analysis, 5.4 percept accepted a p e r manent method through the project and 41.5 percent p r a c t e e d a temporary method a t some time during the four years of the project. The women were a l s o asked whether they thought the chances of children surviving were b e t t e r than t h i r t y years ago and whether they thought more children were dying: 32.3 percent of the women said t h a t the chances had improved, 19.2 percent t h a t the chances had deteriorated, and 47.9 percent were uncertain o r thought there was no change. When the question was reversed, 29.4 percent of the wouen said fewer children were dying, 25.6 percent saia more, and the remaining 45 percent were e i t h e r uncertain o r thought there was no change. I n response t o a question about whether the wife had talked with her husband about family planning, 51 percent of the women said t h a t they had, 49 percent t h a t they had not. Approval of f arnily planning The effect of approval of family p l a ~ i n gon subsequent contracep- t i o n was s l i g h t ( t a b l e 3.6). Among those who approved of family planning, 52.1 percent started t o practice family planning during the project. Among those who i n i t i a l l y disapproved of family planning, 45.4 percent subsequent- l y s t a r t e d practicing. Of those who were uncertain about t h e i r a t t i t u d e t o family planning, only 33.9 percent s t a r t e d contraception. The r e l a t i o n . 8 between approval and practice was especially strong among those who accepted permanent methods of family planning. - - Of women who approved of family plan- ning f o r newl+d - couples, 54.2 percent started practicing contraception during t h e proBct; of those who disapproved, 45.6 percent; of those who !e expressed uncertainty, only 38.7 percent. I i Table 3.6 8 . ~ i s t r i b u t i o nof Women by Thg#r;,pre-prograa Husponse Regarding Attitudes and Beliefe Towards Family P1~1mingand About Child Mortality and Their contracgpt!ive Behqvior During Program I t . t i : I , 1 ,a 1 8 , ; ' I Aacepted During Program Did I b t Accept Pre-Program Respanee Regarding Any Hethod Attitudes and Beliefe During Prograa Total ! 1 1 Tubeatolay) Only * IJ:!.~!~ Approval of Approve 310 (44.0) 338 (47.9) 705 (100.0) Family Planning 167 (42.6) 214 (54.6) 392 (100.0) 74 (32.1) 152 (66.1) 230 (100.0) -' Approval of Pamily Approve . 44 (8.5) 237 (45.7) 238 (45.9) 519 (100.0) Planning for Newly-Wed Disa-I rove i j.; , 16 (3.8) 175 (41.8) 228 (54.4) 419 (100.0) ' I- Couplee ,* Uncertain 367 (100.0) , ' 1 4 (3.9) 131 (15.7) 225 (61.3) 0 l t j b b iI t NO Re6pnse i' 1 (4.5) 0 (36.3) 13 (59.1) 22 (100.0) i.:l1.,! ' I CL~ancesof Chiid-Survival More dmnaee !:j, 3$ (7.2) . 105 (43.1) 213 (49.7) 429 (100.0) as Compared t o Leae ahsnaeo .':; ; a 11 (4.3) 115 (45.1) 129 (50.6) 255 (100.0) 30 Years Ago Same chanaea . . . .!' i 3p (4.7) 248 (39.0) 350 (56.3) 636 (100.0) o t uncertair) l. lf ; . N o R e s p n s e ! ! ; I ! ~ 6 (0.0) 3 (42.9) 4 (57.1) 7 (100.0 , r r ' * I !; I 390 (100.0) 0 Loas/More Children Dle Now 4 ' Less d i e I : 29 (7.4) 169 (43.3) 192 (49.2) I I than Before nore die .. " . 1 4 (4.1) 132 (38.0) 194 (57.1) 340 (100.0) same as befor;. i , 29 (4.81 250 (41.8) 318 (53.3) 597 (100.0) o r uncertain j ' ii , . Corarnunication with Talked /w huaGn4 . 51 (7.5) 325 (48.0) 301 (44.5) 677 (100.0) llusband About Did not talw. ,;,;. I ~amilyPlanning /w hueband. : 20 (3.1) 224 (34.0) 339 (62.0) 643 (100.0) NoReeponsei: ' I 1114.3) 2 (28.6) 4 (57.1) 7 (100.0) . ; I I i I I 1 ; ' , I ( The three categories of approve, uncertain, and disapprove a r e usually treated a s f a l l i n g on an ordinal scale v i t h "uncertain" i n the middle. with great consistency, however, the uncertain group has proved t o be more negative than those who said they disapprmed. These differences were signifi- cant. The differences i n contraceptive practice r a t e s among those who ap- proved and those who disapproved vere marginally significant, but so many women who originally disapproved started to practice t h a t the differences do not seem particularl: important. Beliefs about child survival The contraceptive practice r a t e among those who believed that child mortality had gone down uas 50.8 percent, followed by 46.7 percent among those who believed there was no change, and 42.9 percent among those who believed it had increased. These differences were significant but perhaps unimportar:t because more than 40 percent of people who believed t h a t child mortality had increased still began to practice family planning. When tlie q u e s t i o ~was turned around, however, and asked i n terms of whether more o r fewer children survive, the r e s u l t s were more equivocal. The contraceptive practice r a t e was 50-3 percent f o r women who expressed the belief t h a t the ch-es of survival foz children had increased over the preceding t h i r t y years; 49.4 percent for those who said that the chances had decreased. Among those who were uncertain o r who said t& chances .r i - vere the same, the contraceptive practice r a t e was 43.7 percent. The- relation - seemed especiall* strong among those who accepted permanent methods family planning. Communication with husband about family planning A question that was a particularly good predictor of family planning practice was whether a woman had talked with her husband about not getting pregnant. Even though behavioral, it is included v i t h the a t t i t u d i n a l varia- bles because it presumably is important i n joint decisionmaking. We also consider t h i s question to be our best indicator of the r e l a t i v e l i b e r a t i o n o r independence of women. Almost half the women said they had talked with t h e i r husbands about family planoing; 55.5 percent of them started contraception during the project. Of those who did not t a l k about family planning v i t h t h e i r husbands, only 38 percent started contraception. Associations between a t t i t u d i n a l variables and practice A wife's communication with her husband about family planning had the strongest association with subsequent contraception, her a t t i t u d e toward the practice of family planning by newlywed couples the next strongest. The difference i n contraceptive practice between those who approved o r disapproved of family planning was barely significant, but the uncertain group practiced considerably l e s s contraception than those who disapproved of family planning (and t h i s difference b+shighly significant). Couples who believed t h a t the \ chances of children dying were l e s s than before were more l i k e l y t o practice C family planning (and t h i s association uas significant). But the association was, not significant when women were asked ahether more children were surviving. Among those who accepted permanent methods, the most significant differences were fouxd between those who Ealked with t h e i r husbands about E family planning and those who did not, befween those who approved o r disapproved of family planning, and between those who approved o r disapproved of family planning f o r newlywed couples. The difference i n acceptance of s t e r i l i z a t i o n $ by those who believed child mortality had improved and by those who believed it had gotten worcz was barely significant; the difference was not significant when the question vas asked i n terms of survival. I n Narangwal v i l l a g e s before the project, about half the couples had used some kind of contraception, but only about a sixth had used modern methods. Village groups differed significantly i n the use of modern contraception: The FPED group had the Nghesr preproject use, the FPWSCC group the lowest. That explains the rapid i n i t i a l r i s e i n project acceptance by those i n the FPED group. Ever-user (acceptance) r a t e s a t the beginning of the project were between 7 and 14 percent i n the experimentid. groups. m e acceptance r a t e s i n FPWSCC and FPWS villages, a f t e r more than four years of family planning services, rose t o between 51 and 54 percent. Those i n FPCC v i l l a g e s rose t o 37 percent a f t e r about two years. These increases r e l a t e to all acceptors. A more relevant indlcatcr c?f the effect of the p r ~ j ~ ist the new-acceptor c r a t e , f o r those t h a t used modern methods f o r the f i r s t time a s a result of project a c t i v i t i e s . Cumulative n e r a c c e p t o r rates i n the FPWSCC, FPWS, FPCC 4 and FPED viliages respectively were 45, 42, 39, and 22 percent by the end of t h e project. * - .r The continuing- user rates a l s o increased subst&ntially. I n a l l L groups, the continuing-user r a t e s were l e s s than 10 percent at the begfnning- * E of the project; these r a t e s increased t o 40 percent i n FPWS villages, 35 percent i n FPWSCC villages, 29 percent i n FPED villages and 27 percent i n FPCC villages by the end of the project. These r a t e s of contraceptive use indicate the e f f e c t of the project impact. but a more appropriate indicator vould allo' br the differences i n the effectiveness of the methods because t h e mix of CCi, caceptives differed videly f o r the four experimental groups. An allowasi.., Jol. t h e effectiveuess of the c o n t r a c e r ~ i v emixes i n the four groups iucreases the value of the result obtained i n FPCC v i l l a g e s (high r a t e of using condom). More than half the preproject users became project contraceptors. The t r a n s i t i o n from preproject t o project use was highest i n the FPWS villages, lovest i n the FPED villages. Chapter 4 Robert L. Parker and William A. Reinke The Use of Health Services The Harangwal pop-dation and n u t r i t i o n projects provided an exceptional opportunity t o measure the effect of different packages of inputs. In contrast t o most studies concerned priaarily with results of interventions, w e obtained data f o r detailed input-output- outcome analysea t o establfsh the quantitative relations between t h e inputs of services, t h e practice of famtly planning, t h e use of health services, t h e decline i n f e r t i l i t y , and the improvement i n health status. This d a t a has also - y m i t t e d analysis of the content, cost, and e f f e c t of the different packages of services, so that health planners using these data can s e l e c t the components most appropriate f o r meeting t h e i r objectives. I n t h i s chapter w e describe service inputs i n each experimental group by t h e service contacts provided by t h e project and by the care re- ceived from outside the project The use of project and nonproject health services is then related t o socioeconomic characteristics of individuals and t h e i r families and t o a t t i t u d e s and beliefs already shown t c be impor- t a n t determinants of family planning. Last, the e f f e c t of health services on , t 4 selected attitudes, beliefs, and health indicators I s examined. I n chapters 5, 6, and 7 the service contacts are related to measures pf outccae to find - .a out the effect of thes'e services on the use of family pla&ning and on f e r t i l i t y . - I n cKapter 8 measures of the cost and time of service inpGts a r e presented t o !e 8 I demodstrate the efficiency of the different servi:e combinations, especially i n t h e i r cost- effectiveness. Examination of t h e evidence f o r t h e equitable d i s t r i b u t i o n of services among population subgroups is a l s o discussed within the considerations of equity i n chapter 8. SERVICE CONTACTS netailed records on patients and services gave infomation about services that spanned the l i f e of the project. The service contacts indicated the home or c l i n i c visits provided weekly per 1,000 population. Figure 4.1 summarizes the contacts by a l l s t a f f i n FPWSCC, F'PWS, and PPCC villagea from 1969 t o 1973 and i n FFED villages from mid-1972 t o early 1914. Bealth services were being b u i l t up i n 1969 and reached a peak i n 1970-71; except f o r FPWS v i l l l g e s , service contacts declined i n 1972-73. It FPED villages the recorded service contacts per week were still expanding a t the end of the project. Explanations f o r some s h i f t s i n the volume of s e r ~ i c e scan be found by examining specific services. After 1970 t h e weekly service contacts per 1,003 population f o r women's s e m i c e s were f a i r l y constant i n each acperi- mental group, ranging from 56 t o 6 3 i n FPWS villages, 49 t o 52 i n FPYSCC villages, and 11 t o 14 i n PPCC villages (mainly pregnancy surveillance v i s i t s i n the FPCC villages). Ir addition, family planning contacts steadily increased i n FPWSCC villages (from 4 a week i n 1969 t o 1 3 i n 1973). i n FPWS villages (from 4 a week i n 1969 t o 17 i n 1972), i n FPCC villages (from 6 i n 1971 t o 9 i n 1972), and i n FPED villages (from 11 i n 1972 t o 19 i n 1975). The only declines 0 t i n family planning contacts were i n ,973 i n FPWS and FPCC villages, whe~ethey came down to 14 and 5 a week. Big s h i f t s i n service contacts were best entirely due to changes i n children's s e r ~ i c e s . I n FPWSCC villages a significant decline i n children's I C service contacts occurred betueen 1971 asd 1972, when they dropped from 73 a week t o 47. This decline was maintained in 1973, when there were 36 such contacts. After children's services had been developed i n these villages, a deliberate decision was taken to simplify them f o r replication i n a national program. Service contacts that were empirically found to be l e s s essential were grsdually eliminated a f t e r mid-1972. The main reduction was i n the f r e q u r x y of surveillance contacts during the child's f i r s t year of l i f e . I n FPCC villages there was a s m a l l decline i n children's services betwesin 1970 and 1972 (109 contccts a week t o 94), but a big decline i n 1973 (to 66). This change came a f t e r t h e completion of tile overlapping nutrition project i n 1973, when the decision uas made t o adjust the frequency of children's service v i s i t s i n these villages toward those i n FPWSCC villages. But some differ- ences between t h ~frequency of children's v i s i t s i n PPCC and FPWSCC villages were deliberately maintained to examine the relation of surveillance v i s i t s to infant mnrtality (Taylor and others forthcoming). Host service contacts were provided by family health workers and family planning educators. Family health workers were responsible f o r 90 t o 95 percent of all children's and women's contacts. Almost a l l the remaining contacts were r e f e r r a l s t o physicians f o r complicated or emergency cases. Similarly, f o r family planning services, doctors provided only 5 to 10 a percent of the contacts. The distribution of other family planning contacts varied considerably, with male family planning workers averaging about 20 percent of the contacts i n FPWSCC villages, 35 percent i n FPWS villages, 60 percent i n FPCC villages, and 5-10 percent i n FPED villages. Family health I workers and family planning educators pf ovided the remainder of the family planntng contacts, ranging from an average of 30-35 percent i n FPCC villages to 80 percc-t i n FPED villages. - 110 - USE OF NONPROJECT SERVICES Smple b u s e h o l d surveys i n 1973 and 1974 estimated the prevalence o f ill- nesses on the k i s of recollections of the two weeks before the interview. During ebis tw-ek period, 56 percent of the children d e r three, 30 - percent of chil&ren three t o fourteen, 50 percent of the -men. and 27 percent of tbe vere reported t o have been sick. These tworeek prevalence rates dZti not sZgnificantly d i f f e r among the experimental groups. '1Zle s m e y s a l s o determined vhat proportion of individuals received some health c a r e f o r their i l l n e s s e s and identified the source of care (figure 4.2). Almost 60 percent of the ill children ',n FPWSCC villages and about 50 perceat in FPCC villagss received some care; only 30 to 40 percent of ill children received care i n control villages or in project villages t b t had no childrem's servfccs. Although project services e r e concentrated op children under t h r e e fi all villages, FPWSCC villages provided more services than E'PCC villages to c h i l a e n three to fourteen. The proportion o f ill women vho vere created ranged 26 percent i n control villages to 35 percent i n FPED villages, 28 percent 5n PPCC villages, 42 percent i n FPWSOC villages, and 47 percent Zn F d S d l l a g e s . Forty percent of the men received some treatment. Project services vere primarily responsible f o r the b e t t e r coverage o f chii&en in m S C C and FPCC v f l G g e s , vhere our staff accounted f o r more rhan tuo-thirds a5 the care. But other sources o f care, primarily indigenous private praccitiamers, continued t o be used by a third of the children- an illusprrathn c8 the continuing influence of traditional care de'spite the b e f f e c t i v e ard easy accessible project services. When compared vlth the use of health care in control and FPED villages, project services for children i n child care villages replaced about half the care usually received from other sources. The e f f e c t s of project services on the use of services by vomen in FPWSCC a t d FPWS v i l l a g e s were similar t o the e f f e c t s of children's semices. About 30 percent of ill women received care from the project i n both these i experimental groups, o r between 60 and 70 percent of a l l care. Again, the use i, I of other sources of care vas about half that i n control and FPED villages. The project provided care t o about 10 percent of ill vomen i n FPCC villages (mostly symptomatic treatments t o maintain rapport), but the small amount of project child care services i n FPWS villages did not a f f e c t the use of other sources of care. Ih villages without access to project care, more aen obtained senvices outside the project than did vomen. This pattern was strongest i n control villages and was probably related to the greater travel by men t o nearby market t o m s o r c i t i e s , where there were many private practitioners. Because of the higher prevalence of i l l n e s s e s among vomen, however, the numbers of consultations vere similar for men and women. Where women's services were provided by the project, the use r a t e s f o r vomen vere s l i g h t l y higher than those f o r men. Government health centers did not provide much care i n any of the villages. TICgreatest amount of treatment by government sources was i n P I C c o ~ t r o lvillages, where 5 percent of ill children and 7 percent of ill women used the primary health center, located within three kilometers of these - - villages. In contrast, private practitioners, most of whom had no fo+al training, provided 80 t o 90 percent of the care of i l l n e s s i n villages'without E - *- project services. RELATIONS BETWEEN INDIVIDUAL OR HOUSEHOLD LWCTERISTICS WITE THE USE QF HEALTH SERVICES In chapter 2 several predisposing factors were shown to be associated v i t h the practice of family planning. Some of these, such a s caste and the woman's age, probably a f f e c t the use of health services, too. Demo. -aphic and socio- economic characteristics a r e known t o influence strongly a vide range of behaviors. But a second s e t of predisposing factors- - related t o the use of contraceptives and including a t t i t u d e s toward family planning, b e l i e f s about child death, and communication with husbands about family p l a ~ i n g - - W o ~ lnot d appear -to be d i r e c t determinants of the use of health services. It might be hypothesized, however, that they could have indirect links t o the use of services through an association v i t h some general characteristics, such a s modernization o r a villingness t o change. In any case, a s the f i r s t step i n clarifying the r e l a t i o n between health services and the practice of family planning, we f e l t it important t o identify the extent t o which variables that influence family planning also affect use of health services. These factors would then have t o be controlled f o r when analyzing links between services and family planning. Two cross- sectional surveys measuring a ttitudes, beliefs, and socioeconomic c h a r a c t e r * I ~ t i c of women permitted a two-stage analysis. s The f i r s t , i n 1969, measured attitudes, b e l i e f s , and socioeconomic s t a t u s before exposure t o project services; the second, i n 1971, meesured attitudes and beliefs a f t e r two to - three years of project services. .E Women's i l l n e s s services A review of the many tabulations relating predisposing variables to care of women i n 1969 and 1970 revealed that: o The analysis had meaning only i n the FPWSCC and FPWS villages which had all the women's services (see table 1.4 i n chapter 1 f o r a des- cription of senrices). o There was no evidence of complex interaction effects between the different predisposing factors. o illthough the use of senrices increased between 1969 and 1970, the basic relations of predisposing factors to services stayed the same. Analysis of service information f o r 1969 showed that 54 percent of married wonen aged 15-49 used i l l n e s s care services a t l e a s t once during the year. In FPWS villages the use r a t e was 55 percent; i n FPWSCC villages it vas 54 percent. For those uuder thirty- five, the use r a t e was 55 percent; f o r those over thirty- five it was 52 percent. Neither difference vas significaut. This finding, coupled with the evident absence of interaction effects, l e d to a series of one-variable-at-a-time analyses of attitudes, beliefs, and socio- economic characteristics t o provide many observations f o r analysis. rill v,ariables except religion- caste proved to be significant. Although not sigaificant, there was an indication that non-scheduled (high) ,castes had a somevhat higher r a t e of use than scheduled (low) castes. - (In .w :this analysis religion- caste was a dichotomous variable with a l l high-caste - w S i k h s and Hindus lumped together a d all low castes combined without regard to religion.) Other differences were i n the expected direction. For example, those who expressed approval of family planning had a relatively higher rat= of service use than those who expressed disapproval. But persons uncertain about family planning used services the least, indicating that they may be a special group whose vague response hides a more negative attitude to use of services than overt disapproval of family planning. Beliefs about child deaths were l i n e a r l y related to use: those who believe that more d i e nov used services l e s s than those who believe fever d i e nov; those who vere uncertain o r said the same number d i e now a8 before used senrices between the extremes. ina ally, women who talked more vith their husbands about family planning, vomen who had more possessions, and vomen whose families owned land vere more l i k e l y t o use services. In 1970 the use r a t e rose to 69 percent, a n increase of 16 percent- age points over 1969. The-increase vas remarkably similar i n all population subgroups--none of the differences i n change was significant. The net r e s u l t - . of the uniform increases i n use was that differences i n attitudes, beliefs, and socioeconomic characteristics i n 1969 vere much the same i n 1970. Women's other services Three-quarters of married vomen aged 15-49 received other care i n 1969. Unlike the pattern f o r i l l n e s s care, differences were significant by cohort and age. Younger vomen and those i n FPWSCC villages had higher r a t e s of service use. Subsequent analysis of differences i n attitudes, beliefs, and socio-economic characteristdcs was accordingly conducted 'sepa- a rately by age group and cohort. It yielded three findings of significance: o Younger vomen i n TPWS villages vho were uncertain about differ- 's ences between the6present and the past i n deaths of children had a -- higher r a t e of s e y i c e use than was generally associated with that m cohort and age group; otheruise t h i s belief was unrelated to use of women's other care services. o In contrast, uncertainty about approval of family planning w a s associated with a relatively low rate of s e n i c e use among almost all groups of vomen, but vas sfgnificant only for older vornen i n FPWSCC villages. o A voman's communication with her husband about avoiding pregnancy vss associated with an especially high r a t e of use among older vomen i n FPWSCC villages. One negative finding w a s important: socioecono- mic differences, unlike t h e i r effect on patient- initiated i l l n e s s care, did not have any significant effect on the use of other services i n 1969. It is possible that project- initiated services effectively overcame ar.y i n i t i a l socioeconomic barriers t o use. I n other vords, considerable equity vas achieved i n the provision of project- initiated vomen's services, a finding explored i n greater depth i n chapter 8. Use of vomen's other care increased i n 1970 by a scant three percentage points. But the increase vas much larger among younger vomen, especially i n FPWS villages vhere coverage reached 90 percent i n 1970. The analysis f o r 1970 shovs only one significant socioeconomic difference i n s e m i c e s that vas not seen i n 1969: In FPWS villages, nearly 95 percent of low-caste vomen I L under thirty- five received semices, compared with 87 percent of high-caste ..- vomen under thirty- five. This finding probably r e f l e c t s a ~ o n c e r t e deffort t o reach low-caste vomen with these services, an e f f o r t o r i g i e t i n g i n the - assumption that they were i n greater need of health care an8 family planning. e E - - In general, age differerces affected the use of services more than differences in attitudes, beliefs, and socioeconomic characteristics. This was particularly true f o r other care services, which showed a striking reduction from 1969 t o 1970 i n the use of services by older =men i n FPUSCC villages i n contrast *Ath the increase durinq that period i n service levels f o r women m d e r thirty- five i n FPWS villages. These changes must have been r e s u l t s of the groving enphasis between 1969 and 1970 on children's services and family planning i n FPWS villages, deliberately shifting the focus away from older women. This decision was made when it became evident that the family health workers were overloaded v i t h tasks, especially i n FPWSCC villages, and a deliberate reallocation of time was needed. -Average number of v i s i t s by users of women's services In addition to determining the percentage of vomen using i l l n e s s f care and other care services, the volume of such use was examined. The average number of v i s i t s f o r i l l n e s s care i n 1969 was 4 per mnran i n FPWSCC villages and 5.1 i n FPWS villages. The difference betveen the two experimental groups was significant and became even greater as the average v i s i t s increased more rapidly i n 1970 i n FPUS villages than i n FWSCC villages. The average number of contacts per mman receiving other care was almost the same i n both experi- mental groups i n 1969. By 1970 vomen in FPWS villages averaged 5.9 contacts, compared vith 5.4 i n FPWSCC villages, a difference that was significant. Age did not have any significant e f f e c t on the averaze number of v i s i t s per woman, except i n 1970 f o r other care i n FPWS villages: women under thirty- five av=raged 6.1 contacts, those thirty- five and 0ve.r 5.5. - Children's i l l n e s s care This analysis was limited t o experimental groups t h a t provided child care--the FPWSCC and FPCC villages- - and mainly involved family- initiated c l i n i c contacts. In examining the 1969 data, we found that about 67 percent of the women v i t h children of any age had children who received curative services i n the v i l l a g e c l i n i c s . The difference between the experimental groups i n coverage of a l l children was s i g n i f i c a n t (75 percent f o r FPUSCC v i l l a g e s and 58 percent f o r FPCC villages). This difference was mainly due t o the design of services i n FPCC villages, vhere there was a concentration on care of children under three, using frequent home v i s i t s . Of women with a child under three, 79 percent took t h e i r children t o a c l i n i c f o r care; of women with children three and older, 53 percent took t h e i r children t o a c l i n i c i n 1969. Unlike the e f f e c t on women's i l l n e s s c a r e services, the variables of owning land o r having more modem p o s s e s s i ~ n swere not associated with s i g n i f i c a n t differences i n the use of children's services. Nor was c a s t e related t o variations ir service use. Among the measures of a t t i t u d e s and beliefs, approval of family planning and b e l i e f s about c h i l d deaths were agao,ciated with significant differences i n the use of services. interestingly,' although the lowest users of services were children of women uncertain about family planning, t h e children of disapprovers used services more than those of approvers, possibly indicating t h a t disapprovers had more children. Women who believed t h a t more children d i e cow than before had children +o used servlces significantly l e s s than women who were uncertain o r said fewer dL- now. A wife's talking with her husband about family planning seemed to make no difference i n the use of children's services. Despite big differences i n service patterns, analyses f o r each age group i n the separate experimental groups showed similar associations between these predisposing variables and the use of c l i n i c se mices . Use of children's illnese cate services i n 1970 rose only slightly- to 69 percent. Differences i n use by age of the children rgmained basically unchanged, but the di.ference between experimental groups diminished, with r a t e s of 72 percent i n FPWSCC villages and 65 percent i n FPCC villages. The differences i n use related t o a;.titudes, beliefs, and socioeconomic characte- r i s t i c s were much the same a s i n 1969. Children's other services Because these services were for children under three, only women with children i n t h i s age group were included i n t h i s part of the analysis. The services were worker- initiated, and more than 91 percent of these women had children who received chis type of care. Coverage i n the two experimental groups was about the same-90 percent i n FPWSCC villages and 93 percent ill FPCC villages. Attitudes, beliefs, and socioeconomic characteristics had r t a t best only small effects on r a t e s of use. In 1970 use was almost the same--at 92 percent--as were its selations with attitudes, b e l i e f s , and - - socioeconomic ~ B a r a c t e r i s t i c s - . Service contacts with users of children's services The average annual number of i l l n e s s care v i s i t s was about ten i n FPWSCC villages and seven i n FPCC villages; t h a t of other care (both home and _ c l i n i c ) v i s i t s w a s s l i g h t l y l e s s than twenty i n PPUSCC v i l l a g e s and forty i n FPCC villages. These findings, and especially the figures on service contacts, further documect the different patterns of child care i n the two experimental groups--FPCC villages received much more intensive inputs of child care than did FPWSCC villages. aecause of the large variance i n contacts per family. the differences i n attitudes and b e l i e f s a r e not significant. But the general patterns are very similar to comparisons involving use o r nonuse of services i n any year. For example, those uncertain about fainily planning used fewer services than did approvers o r disapprovers. The use of children's services seems to have only a limited relation to attitudes and beliefs, malciag these services more acceptable t o families regardless of t h e i r i n i t i a l a t t i t u d e s or beliefs. These services may thus be useful f ~ education i n health and family planning. r EVIDENCE FOR CE,wC;ES IN ATTITUDES AND BELIEFS FROM USE OF HEALTH SERVICES Analyses,in subsequent chapters demonstrate the strong direct r e l a t i o n between the use of health services and the practice of family planning, holding a t t i t u d e s and beliefs constant. 1tLis also important t o identify .r whether d i r e c t e f f e c t s on a t t i t u d e s and beliefs mediated some of the effect -- i of health services on family planning. !P - Women interviewed i n both cross- sectional surveys were included i n t h i s ana1ys;-. Their a t t i t u d e s toward family planning and their beliefs about child death i n the two surveys were compared and classified as follows: - -1971 1969 - Positive change i n a t t i t u d e toward family planning M sapprove Uncertain Disapprove /tpprove Uncertain Approve Negative change i n a t t i t u d e toward family planning Approve Uncertain Approve Disapprove Uncertain Disapprove Positive change i n belfef about child deaths More d i e Uncertain/same More d i e Fewer d i e Uncertain/ Fewer d i e same Negative change i n belief about child deaths Fewer d i e Uncertain/same Fewer d i e More die Uncertain/ More d i e same ! ! b Earlier analysis indicated that "uncertain" responses vere a mixed category, t - often representing probably the most negative a t t i t u d e s a d beliefs. A a. b i second classification therefore involved only women who had clrir- cut changes . I i n attitude: f o r example, froa di&pprove t o aporove, from approve to d i s a p prove, from more to Lever, and from fever t o more. - In addition t o identifying changes i n a t t i t u d e s and beliefs,.; i women o r t h e i r children vere grouped i n three ways: those having used s i v i c e s - - h i n a l l three yearsw(l.969-71) . between the cross- sectional surveys; thosg k ; having used services i n one o r two of the years; and those not having used services. For each category the analysis compare$ the percentage of -men who had a positive o r negative change i n a t t i t u d e s only i n r e l a t i o n t o the women I who could change their a t t i t u d e s i n that direction. For example, women already ap?roving family planning did not appear i n the denomizlator represent- . I ing women with a potential for positive change. . Both classifications produced similar results, but only a few I were significant. Some r e s u l t s nevertheless suggested that sarvices may induce a change i n a t t i t u d r s and beliefs. Changes i n a t t i t u d e s and beliefs related t o women's services Users of i l l n e s s care services f o r women had a greater positive change and smaller negative change i n a t t i t u d e s toward family planning than did nonusers. The effect on negative change comes closes to being significant, $ t and both e f f e c t s a r e strongest for women over thirty- five. The proportions of women who did not use services but changed t h e i r a t t i t u d e s i n a positive o r 1 negative direction were 60 percent snd 28 percent respectively. Of women who used services i n all three years, the proportian with posicive change was I 68 percent, and that with negative change 23 percent. Users f o r services f o r ona or two years showed fewer favorable changes than those who used semices f o r three years. In t h i s analysid we combined two similar r&estions sbout women's b e l i e f s related to the death and survival of children. There was no - consistent patterq'relating r use of women's i l l n e s s care services to changes i i n beliefs about child death. - Other c a!E services for women did not consistently follow the w pattern of i l l n e s s car? services i n their effect on attitudes and beliefs. g Significant effects occurred only i n the negative change of B the variable for the approval of family $laming. Users of services f o r one or two years shoved t h e l e a s t change (18 percent); user6 of semices f o r three year8 shoved about the silse negative change a s nonusers. It this case users for three years may have included more older vomen. Positive change i n the variable f o r helief about c h i l d deaths was related t o other c a r e services i n a manner similar t o i l l n e s s c a r e services. But users of other care services f o r one t o # b two years showed almost twice the amount of negu:.ive change i n beliefs about chiid ueatirs as users of these senricea f o r three years (31 percent compared with :8 percent). The change by nonusers vas i n betveen. The main e f f e c t of servicac, *en identified, seems t o have been t o prt vent a negative movement i n a t t i t u d e s and beliefs. Lxangea i n a t t i t u d e s and b e l i e f s related t o children's services Changes in the approval of family planning were not significantly related t o ct.ildrenOs i l l n e s s care, but changes i n the positive directioa followed a pattern c o ~ s i s t e n twith a service e f f e c t similar t o women's i l l n e ~ s care. I n t h i s case 57 percent of nonusers showed a positive change, a s did 63 percent of three-yeezr users. This' pattern w a s predominantly f o r families with ail children over three. Tinree-year users of children's i l l n e s s care had st&niflcantly l e s s negative change i n the9.r b e l i e f s abdut child deaths '15 percent) than nonusers (27 percent) and users f o r one t o two years (28 percent). - Because of the small number of n ~ n u s e r sof children's ocher care, ' 'i testiug f o r signilicsncg could not be f r u i t f u l . Basically, however, users of * childreil's ocher care hfd l e s s negative change i n the variables for approval? snd child deaths, a finding t h a t parallels t h e e f f e c t of women's services. The contacts or visits by users of children's services f r o s 1969 through 1971 were compared vdth changes i n approval of family planning and i n b e l i e f s about child death-- for children of vomen with d e f i n i t e changes (approve t o disapprove, disapprove t o approve; more t o less, l e s s t o more). Although n c t significant, t h e results tended t o support the patterns of change established when cor~pariagusers with nonusers and a r e therefore f e l t t o have practical Implications ( t a b l e 4.1). I n t h e FPWSCC group, positive change both i n t h e approval of family planning and i n t h e b e l l e f s about child deaths ware related t o a higher average number of contacts than were negative changes i n these a t t i t u d e s o r beliefs. This was t r u e of both i l l n e s s care and other care. I n the FPCC group, t h e same pattern was shown f o r t h e child death variable, but the average use of satvices was higher f o r those who had a negst.ive change i n a t t i t u d e s about family planning, a finding t h a t is not surprising because family planning education and setvices were not incorporated with t h e services of t h i s group of villages mtil 1971. I n summary, there is evidence t h a t health services may have induced .;some s m a l l positive changes i n a t t i t u d e s o r b e l i e f s about family planning. But the predominant detectable e f f e c t appears t o be preventive, supporting the continuation of positive a t t i t u d e s and b e l i e f s and limiting the extent L of negative change. I n t h e short period of the project, therefore, the e f f e c t s of services on the practice of family planning w e r e most l i k e l y d i r e c t and only weakly mediated, i f a t a l l , by changes i n a t t i t u d e s and beliefs. Such mediation may, however, have become more important over YE - Table 4.1 --. A- Y&.r of Sarrricu -0 1969-1971 Pat F e y by Expmdmntal Croup and f ~ p r of Service - m c c - rpcc C - Fastor 1 l . h Cara ~ ~ 0ch.r Cai. 1-8 Car. 0th.r Car. - *Dafinicfve chugrs only, 8.g. tpprova--c dbappratt. or diszpprwe- appro-; mar. lus Or 18~3- Wma EFFECT OF SERVICES ON HEALTH Measuring the health benefits of project services was important for tvo reasons: f i r s t , t o test the effect of various experimental inputs on health indices; second, t o t e s t the child- survival hypothesis, which required a program of health care that improved child health enough t o make benefits apparent t o families and the village community. The services constituting that program had t o be replicable under the f i r w ~ c i a land manpower constraints i n India and other developing countries. i Data gathering on changes i n health s t a t u s was focused on children G under three. This focus r e f l e c t s the emphasis 01: care of young children i n FPCC villages, which were also a part of the nutrition project. Data were collected on all deaths through the special surveys and from registers of v i t a l s t a t i s t i c s i n all study and control villages. Although data were available f o r all age groups, the numbers of deaths occurring i n the study population were large enough t o r analysis only f o r children under three. An important measure of the effect of women's services would have been changes i n maternal mortality r a t e s , but the number of maternal deaths i n the study villages and control villages was too small f o r analysis. * I t Special morbidity and anthropometric data were collected f o r a l l children under three i n villages i n the nutrition project. - These data . were used to demonstrate the eff&t of services f o r child health and nutri- L tion on measures other than mortelity i n FPCC villages. Special studies were b not carried out to measure changes i n women's morbidity, but perinatal mor- Mortality Mortality r a t e s i n the control villages of the population aud n u t r i t i o n studies during 1970-73 a r e the base against which the e f f e c t s of services i n the two child- care experimental groups a r e compared. The percentage differences i n r a t e s were calculated by caste and by age 0 a t death. but only those f o r the tvo predominant castes a r e shown i n table t 4.4. Races a r e also shorn ( i n parentheses) f o r the t o t a l population, which includes all other castes and mixes high- caste and low-caste groups. Table 4.2 a l s o includes figures from the Khanna project, con- ducted nearby i n Ludhiana D i s t r i c t i n the early 1960s. During the decade a f t e r the Khanna project, Punjab's economy developed dramatically because of the Green Revolution. Comparisons of mortality r a t e s i n Khanna villages and i n our control v i l l a g e s show an increase i n s t i l l b i r t h r a t e s (possibly related t o improved reporting), no decline i n neonatal death r a t e s , a decline $ of 39 percent i n postneonatal death r a t e s , and a decline of 60 percent i n the : ! mortality r a t e of children aged one to three. The decline i n the l a s t two i r a t e s presuaably was primarfly a t t r i b u t z b l e to socioeconomic changes during I the intervening decade. F L The s t i l l b i r t h mate per 1,000 l i v e and still b i r t h s was 57 i n t - I the control villages, 44 i n FPWSCC villages, and 37 i n FPCC villages. This r a t e presumably measures the probable e f f e c t on the f e t k of prenatal '5 D - '% f care f o r the mother, especially the provision of i r o n and f o l i c acid ( f o r -- - L anemia) &o a l l mothers and the provision of nutritional s u p p h e n t s t o poorly . nourished mothers. Effects were comparable n both FPWSCC and FPCC v i l l a g e s i n the low-caste group; only i n FPCC v i l l a g e s was there any e f f e c t measured among high-caste families. The more intensive services i n FPCC v i l l a g e s - 128- . . Table 4.2 Igir30Jm casts were tba Sat Sikhsr the landovnfng far~rrs,rad 'ImT" c u t e wsra the . schrtiulad S-, prdcdaatdy ~~r lakeratl. These castes E.da up . bator* 75-85 parcsnt of tbe population. ~otdlsorttty ratus includkrg other castes a r m shown jn wcnthes~?r. / Rates are per thaPsand Live and s-! corebixcd. Rates are -per t b u t a m i Live 5irh Rates are par 1000 a l d r a n 1-3 years of age. b(or -ax I 100 = t-ai,ciarsnca. a 4 cacbined zates from the control villa'ges of the Population ard Nutrition Studies, .Wyon and Godon. The Xhanna St.~dy .P 5 villages appear to have encompassed a l l caste groups. Workers i n FPWSCC villages had more to do and may have concentrated t h e i r prenatal care, especially supplanentation, on low-caste mothers. Infant mortality r a t e s were s p l i t i n t o neonatal death r a t e s ( f o r the f i r s t month of l i f e ) and postneonatal death r a t e s ( f o r the next eleven months). i i' The neonatal death r a t e i n control villages was 78 per 1,000 l i v e b i r t h s , that i n FPWSCC v i l l a g e s -8 64, and that i n FPCC villages was 47, reductions of 18 percent and 40 percent. The high-caste children i n FPWSCC v i l l a g e s benefited from the services, but the low-caste children did not. This was i n contrast t o FPCC villages, where the low-caste groups appeared to benefit most. Apparently, i n FPWSCC villages, where services were l e s s intensive, high- caste families sought care from the project o r from other sources and moderr.tely reduced neonatal mortality. But low-caste c h i l d r ~ nhenefited only i n FPCC villages, vhere the outreach into homes was much more intensive. During the postpartum period, home v i s i t s were weekly i n FPCC villages, monthly i n FPWSCC villages. The basic pattern was the same f o r postneonatal children: high- caste children i n FPWSCC v i l l a g e s and a l l children i n FPCC villages received significant benefits from the child care. The obvious prog,raq implications of these findings a r e not only that frequent surveillance r of infants is important i n reducing mortality, but t h a t such care must be focused i n special outreach programs on the lcwer socioeconomic groups to compensate f o r t h e i r greater needs and the constraints affecting t h e i r use - of services. B I Although there were fewer: d e a t t s of children aged one to three, the effect of services on reducing such deaths per 1,000 childrsn was a s great a s t h a t on reducing deaths a t an e a r l i e r age. The mortality r a t e s per 1,000 children sged one t o three vere 7 i n FPWSCC villages and 1 3 i n FPCC villages, I compared v i t h 19 i n control pillages. For t h i s age group the e f f e c t was 5 C greater i n FPWSCC villages (63 percent lover than controls), where services had an equal impact on a l l castes. Interestingly, services f o r older children i n FPCC v i l l a g e s had no e f f e c t on high- caste childr=n. This finding is d i f f i c u l t t o explain, but it is possible t h a t high- caste children i n these villages may not have used nutrition supplements. Results from the nutrition project shoved t h a t health care hzd a major e f f e c t both on the mortality r a t e s of infanzs (under a year) and of children aged one to three, while nutrition care vas most effective only f o r children aged one to three. I n general, the differences i n the number of deaths a r e significant only i f c a s t e s are combined. But caste differences a r e important i n gauging the r e l a t i v e e f f e c t of child health services OL subgroups of the population. On the whole, the combined services i n FPWSCC villages had a moderate and caste- selective e f f e c t on infant mortality r a t e s and more e f f e c t on child death r a t e s of all caste groups. The more intensive child care i n FPCC villages had a more consistent effect on age- specific mortality rates, with caste differences showing up only a f t e r one year of age. Interviews of mothers 'of' children who died (verbal iutopsies) provided additional insights that possibly explain some of the differences i n death r a t e s be&en castes. Almost all children received treatment 'Z L f o r i l l n e s s e s t h a t j e d t o death. But high-caste families tended t o seek * care e a r l i e r than Bw-caste families: - 68 percent of high- caste children who died received care i n the f i r s t twenty-four hours of t h e i r terminal i l l n e s s , compared with 49 percent of low-caste children. That nay help explain the improvement i n infant mortality r a t e s among high- caste children and the absence of a drop i n death r a t e i n low-caste infants when services were accessible, a s i n FPWSCC villages, but not intensively delivered i n the homes, a s i n FPCC villages. To get information on the influence of child deaths on the motiva- tion f o r family planning, parents were asked i n the verbal autopsies whether the! Intended to have another child. No association was found between yes respumes and the sex, age, caste, o r nutritional status of the deceased chiJ.d. The stated i n t e n t t o have another child was strongly associated, however, with the number of living siblings. Ninety-five percent of mothers who had ?o living child o r only one said they intended t o have more children, compared with 54 percent of those who had two o r three living children, and 11 percent of those who had four or nore. Morbidity Weekly surveillance of morbidity was part of the nutrition study. Data therefore a r e avaiiablu to measure the effect of child care on the i l l n e s s e s of children i n the FPCC experimental group and i n the nutrition Froject's control group (table 4.3). Conditions were selected for their frequency and importance f o r children aged 0-3. 'Ihe average duration for each . e I condition was l e s s i n FPCC villages: differences ranged from 14 t o 33 percent - . - l e s s than i n the controls. By using the axerage durations and an average incidence r a t e f o r each condition, the tot's annual days of i l l n e s s due t o - these conditions per child could be c a l c u l z e d for infants (under a year) and .8 * for children aged one to three. 11 ~ e m i c &i n FPCC villages effectively - - 1/ Because the incidence of disease was underreported i n the control villages, the average incidence across a l l villages was used. Because within-group analyses showed a reduction of incidence i n some conditions f o r children with better nutritional status, tho- r e s u l t s probably understate the e f f e c t of children's services on morbidity. Table 4.3 reduced the amount of i l l n e s s by 22 days a year i n each age grorp, a reduction of 16 percent f o r infants and 2 1 percent f o r children aged one to three. This reduction r e f l e c t s the changes already demonstrated i n mortality. In addition, reducing the number of days of i l l n e s s by more than three weeks a year reduced the metabolic drain on nutrients i n children, and presumably contributed t o r improved n u t r i t i o n a l status. I E Growth The e f f e c t of services on the growth of children was analyzed only f o r the FPCC group of villages. As f o r the analysis of morbidity, the data come from the n u t r i t i o n project and can be compared with the n u t r t t i o n control data. At ages seventeen through thirty- six months, children i n FPCC v i l l a g e s had significantly Ligher average weights and heights than children i n control villages. This pattern was true f o r males and females and f o r high and low castes. Average weights, adjusted f o r unequal sex-caste :omposi- t i o n in the groups and expressed a s a percentage of the Harvard standard veight, a r e shown i n figure 4.3 f o r ages above one year. Differences between the service and control villages were about three to four p e r c e n t a e points, t corresponding t o 0.4 and 0.6 kilograms, beyond two years of age. Differences i n average veight a t ages below thirteen months were significant only a t - birth, when the control averages were higher (probably due t o underreporting of b i r t h e i g h t s of low-birth-weight babies). . For children aged tuenty-one months and above, average n?Cght uas significantly greater i n the service villages than i n control villages ( f i g u r e 4.4). Differences i n heights i n the FPCC and control groups opened just before the sgcond year of l i f e and averaged about 1.5 t o 2.5 percentage . points, o r 1 to 2 centimeters. Sex and caste shoved a highly sigcificant e f f e c t on weight a t every age. Males on average weighed 0.6 to 1 kilogram more than females, v i t h the l a r g e s t differences f o r those aged f i v e to seventeen months. Eeyond the age of one year, average w i g h t s of high-caste children were about 0.75 kilogram more thaa those of low-caste children. Sex and c a s t e also had a strong influence on height. Males on average were about 2.7 centimet2rs t a l l e r than females aged nine t o seventeen months; differences then decreased to s l i g h t l y Less than 2 centimeters at tuenty-seven months and above. Differences i n average height between high and low castes increaaed from 1.4 centimeters at nine months t o about 2.5 centimeters a f t e r two years of age. In sunmary, these r e l a t i o n s between services, sex, and c a s t e were additive: a high- caste male i n a service village on average veighed 2 kilograms more and w a s 6 centimeters t a l l e r than a low- caste female i n a control village. The number of siblings a l i v e a t th,e time of the child's birth, had a significant influence on mead veight and height. Children v i t h two o r more male siblings had the lowest-average weights and heights, regardless of the - .a number of female siblings. -ao s e with two female siblings and two or more - male siblings were p a r t i c u l a ~ l ylow i n weight and height, about 0.4 kilograms X l i g h t e r and 1.3 centimeters shorter than those with one o r no living; brothers. This finding demonstrates the e f f e c t t h a t family planning might have i n improving the nutritional status of children. me use of health services i n the study villages ha5 been described a s s a w i c e s cf the project and services outside the project, primarily private ~ta:titioaets. The project provided 50 t o 60 coutacts with wonten's services per 1,009 population per week i n villages prcviding such services. I n contrast, the contacts with children's aerpices varied f'r~xcabout 40 t o 110 per 1,000 population per week, depending on the year of the project and the experimental group. Discreet fami?.y planning servicer increased from about 4 t o 6 contacts per 1,000 population per week early i n tae project to about 13 t u 19 l a t e i n the project. About 90 t o 95 percent of all contacts w e r e provided by the village- level family health workr. Outside services i n villages with project services were used fn abput 14 t o 18 psvceot of illneaoaa, compared with 30 t o 40 percent i n villages without project services. Attitudes, beliefs, and sosioeconomic factors had aone impact on the use of health services. There w69 significnatly greater use of services for women's iilazsses semicee b> women who ayproved of family planning, who believed t h a t fewer children d i e nov than i n e a r l i e r years, who talked with tho-ir husbands ?bout family planning, and vho came from households with more possesviclns cr with pore land. t Any effect that health services may have had on attitudes and - t e l i e f s was small.-The pradcainant effect was probabiy supportive, favoring co?tinuation of pd5itive atticudes and beliefs. There w s e well-defined effects sn child health through the provi- * sion cf children0s'sei~icesU Infant mortality was reduced more than 35 percea? Zn FPCC villages, child mortality mora than 50 percent i n FPWSCC villages. The average aaaual nxnnber o t days of i l l n e s s per child was reduced 16 percent for infants under a year and 21 percent for children one to three in FPCC villages. Children i n villages vith nutritional supplenen- tation services on the average were 2 centimeY.rs taller and 0.5 kilograms heavier by age three than childrin i n c u o r y , ~ illages. Chapter 5 William A. Reinke and Robert 2. Parker Relations Between the Use of Health Services and Family Planning t I- The main hypothesis of t h i s study was that t h e use of health services contri- butes f i r s t t o the practice of familg planning and later t o the reduction of f e r t i l i t y . The f i r s t e f f e c t was f e l t t o be mediated i n tuo ways: by b e t t e r rapport of health and family planning workers with the community, and by introducing f a n i l y planning education and reinforcing the motivation t o use contraception a t spontaneous o r predetermined "entry points" i n the health services, a t occasions when health workers could reasonably introduce family planning t o discussions with c l i e n t s (see table 2.2 i n chapter 2). For example, a worker, when caring f o r and showing concern f o r the health of a mother o r child, could readily convey arguments f o r longer spacing between children. It was also postulated that improvixig the general health of children and reducing infant and child mortality could indirectly influence couples t o change b e l i e f s and behavior associated with the expectation that by having many children some would survive. It was hoped that the period needed for t h i s i n d i r e c t e f f e c t uould be shortened by pointing out t o parents the im- proved health of the children. The profect used entry points t o increase awareness of the survival of more children. A c l e a r r e l a t i o n between health - services and the use of family planning emerged and is documented i n the - 'r following pages. - -- i w Q- t - 140 - That r e l a t i o n gives rise t o corollary questions relevant t o policy. m a t is the sequence of usiap services? Is there a long lag betveen receiving health s e m i c e s and accepting family planning? Is continuing, intensive provision of health s e m i c e s needed t o sustain contraceptive use? Do users of health s e m i c e s tend t o be prior users of family planning? Doee health care lead t o the recruitment of new acceptors of family planning o r merely induce prior usera t o change from traditional t o modern methods? These are the questions addressed i n t h i s chapter, and the discussion is further elaborated i n the next. Variables i n the analyses include the following: o Project family planning practice, which r e f e r s t o the use o r nonuse * of contraception o r s t e r i l i z a t i o n by women (or t h e i r husbands) a t any t i m e during the project. o Prior family planning practice, which r e f e r s t o the use o r nonuse of traditional o r modern methods of family planning methods before the start of the project. o Women's i l l n e s s care (WILL). o Women's ether care (WOTH). o Children's i l l n e s s care (CILL). L o Children's other care (COTH). V i s i t s o r service contacts tinder these headings were measured during a speci- - fied period by the presence o r absence of one o r more v i s i t s o r contacts- -or by .b the average number of v i s i t s o r contacts per person who received any services. Where appropriate, the years i n which services were provided were separately analyzed. The analysis of the e f f e c t s of health services was done f o r three of the experimental groups of villages: FPWSCC (family planning plus women's services plus children's services); FPWS (family planning plus women's s e r vices); and FPCC (family planning plus children's services). Villages t h a t did not receive women's o r children's services- - the conti01 and FPED vil- lages--9ere excluded because the analysis vas focus+. on t h e r e l a t i o n of health services to the use of family planning. The r e l a t i o n between health services and family planning vas f i r s t com~aredf o r the three experimental groups ( f i g u r e 5.1). The use of family planning, measured by the percentage of current users, tended t o increase. For a l l types of health service, the project succeeded early i n reaching a majority of the target population, and the coverage was sustained. A s vas described i n chapter 4, the contacts per 1,000 population peaked near the middle of the project and then tended t o decline, s l i g h t l y f o r vomen's ser- vices, more markedly f o r children's services. Community-level analysis of the d i f f e r e n t packages of health ser- vices provided a crude f i r s t approximation of the r e l a t i o n between the use ' of health services and t h a t of family planning. No time-phased association between the use of' h i a l t h parvices and family plgnning is c l e a r l y discernible when relations are viewed i n aggregate experimental groups. The r e l a t i o n - d e p e 4 s more d i r e c t l y on the use of services i n the community. - Subsequent i analyses a r e based, therefore, on the sequence and continuity of use of speci-' - !& E f i c t e e s of service by the family and its members, regardless of t h e i r experiz mental group. In other words, we w i l l distinguish between use and nonuse of individual services, regardless of whether nonuse was due to the absence of -orlackof exposure to--services. Disregard f o r t h i s d i s t i n c t i o n i n avail- a b i l i t y d i l u t e s the e f f e c t s derived i n subsequent analyses, but it had to be done t o provide enough observations f o r analysis. I n eech of four categories of health services, e l i g i b l e women - a r e distinguished by whether they used a specified service f o r themselves o r t h e i r children a t any time during the program ( t a b l e 5.1). The r e s u l t s reveal consistently higher r a t e s of f a u i l y planning practice among users of health services than among nonusers. Except f o r woaen's i l l n e s s care, acceptance r a t e s among users of services were about three times the r a t e s 1 i among nonusers: more than half the users became family planning acceptors, i whereas fewer than a f i f t h of the nonusers did. Although l e s s striking, the difference i n acceptance r a t e s between users and nonusers of women's i l l n e s s c a r e still was significant. The main feature of t h i s service was the somewhat higher acceptance r a t e among nonusers of health services, suggesting that many women were reached through other service channels. CONTINUITY OF USE What is the importance of the timing and continuity of health services? . t To explore the question of continuity, the project was divided into two periods: 1969-71 and 1972-73. Health service users who became family a a n n i n g acceptors .r were distinguished by whether they received services i n one p e ~ i o do r both. - * In each s e t v i c e subgroup, the percentage of acceptors who prac!?iced . - family planning i n both periods was compared with t h a t of acceptors whose practice was i n one period. 4 . Table 5.1 U s e NOn-Use Children's Othar ~~) Use 828 Except f o r the women's other care, differences i n family planning practice between part-time and continuing users of health services were not significant. The r e s c l t s suggest that although some use of health services is closely associated with the acceptance of family planning, continuing contracep- t i v e practice depends more on factors other than repeated exposure t o health services. The exceptioq is nottiworthy because it suggests, perhaps surprising- l y , that part- time recipients of women's other services were more likely to be continuing users of family planning than full- time recipients: 86 percent of the part-time service users were continuing users of family planning. But the exception is not a s surprising a s it may a t f i r s t appear: those s t e r i l i z e d became full- time users of family planning, while their continuing need f o r women's other services diminished. SEQUENCE OF USE I f the simple use of health services, not the continuity of use, is more c r i t i c a l for the acceptance of family planning, timing becomes a concern. Is the receipt of health services associated with nearly concurrent acceptance of family planning, o r is there a long lag? Did the acceptance of family planning precede the use of health services, o r vice vers?" . I W e examined the relation between the volume and timing of health services received by a woman or her children and her practice of family plan- ning (table 5.2). The investigation excluded those who received no health - services during the project. Average numbers of health service contacts i n B I either 1969-71 or 1972-73 were related to four exclusive categories of family planning practice: none, practice only i n 1969-71, practice only i n 1972-73, or practice i n both 1969-71 and 1972-73. The average number of i l l n e s s care v i s i t s by women w a s twenty-four for those who practiced family planning only i n 1969-71, twenty-one for those who practiced family planning i n both periods, eighteen f o r 1972-73 practi- tioners, and sixteen for nonpractitioners. In contrast, more v i s i t s for women's i l l n e s s care i n 1972-73 were by those practicing family planning a ~ t that l a t e r time, especially i f they vere e a r l i e r users a s well. The pattern of other care was more uniform f o r women under thirty- five. The contacts with children's i l l n e s s cqre vere more varizble and showed patterns about the same a s those for women's i l l n e s s care. The average number of child v i s i t s during 1969-71 ranged from eighteen f o r nonacceptors to thirty- nine f o r those who practiced family planning o d y during 1969-71. (T5e corresponding range of v i s i t s f o r women's i l l n e s s care was sixteen to twenty- four.) The analysis of children's other serpices was limited to children under three and done separately f o r the two experimental groups vfth child care (FPCC and FPWSCC). The reason is that the scheduled home v i s i t s differed substantially between them (surveillance v i s i t s were monthly i n FPWSCC villages and weekly i n FPCC villages). I n both experimental groups the volume of children's other services was higher'thhn that f o r other types of hervice. For example, children of nonacceptors of family planning i n the FPWSCC villages received, on the averag;, s i x t y children's other care v i s i t s i n 1969-71; the '1 L average number of women? other care vSsits was thirteen. Despite the higher - w volume, the receipt of mildren's other services was f a i r l y unifora for the family planning groups, a s was the receipt of women's other services. In general, then, variations i n i l l n e s s care vere found to be more closely associated v i t h the use of family planning than was the receipt of other services. Presumably, the reason is that i l l n e s s care was i n i t i a t e d by patients and reflected the natural predilections of families. Many of the foregoing findings, although striking, barely bordereJ on significance because of large variances and the small number i n each category. Further appraisal produced noteworthy patterns. o Acceptors of family planning a t any time during the program averaged 20.5 contacts with women's i l l n e s s services during 1969-71, non- acceplors 16.1. Women who never practiced family planning invariably received fewer services than other vonen, and their children received fewer services than other children. o Family planning users who concurrently received health services i n 1969-71 received more health services of all types than women whose practice of family planning began l a t e r . o Women who practiced family planning i n 1972-73 tended t o be greater concurrent users of health services than e a r l i e r f a n i l y planning users who had discontinued by 1972-73. These three findings support the conclusion that., use of family planning '. untier the project is positively associated with the volume of health services. -. o The use of health services i n 1969-71 improved the chances f o r family Z - rS planning acceptance in that same period. But did heavy early use of - * health services helpsto sustain family planning practlce i n 1972-73? !? - * For reasons that a r e uncleer, the continuing users, i f anything, received fewer health services eerly i n the program than those who discontimed family planning (column lL of table 5.2). - 149 - o Did new r e c r u i t s t o new family planning exhibit a d i f f e r e n t pattern from previous users? The r e s u l t s a r e inconclusive, except to suggest t h a t the new recruits perhaps received laore children's other services during the period of successful recruitment t o family p l a n n i a (column 12). While not significaut, the r e s u l t s tend t o reinforce the importance of concurrence i n the receipt of health services -ncl the practice of family planning. . o The issue so f a r has been: How does the use of health services lead to the practice of family planning? We now ask the obverse: ! Does eaxiy contact with the systes through the acceptance of family planning lead t o increased subsequent use of health services? I n other words, was the use of health services i n 1972-73 greater among e a r l i e r acceptors of family planning than among those who f i r s t accepted i n 1972-73? There is l i t t l e evidence to support t h i s proposition. The analysis is confounded, however, by the concurrent association between health services and the use of family planning i n 1972-73. The findings ( i n c-lumn 12) removed t h l s factor since they were limited t o users of both health and family planning services i n Nevertheless, considering columns 12 and 13 together, L we . 1 r I 1972-73. see no suggestion that the practice of family plannisg leads t o the use of health services. - E I USE OF FAMILY PLANNING I N RELATION TO PERIOD OF USING HEALTH SERVICES Tne use of h e a l t h services a t t h e midpoint of the project (1971) w a s conpared with e a r l i e r (1969-71) and l a t e r (1972-73) use of family planning, and with nonuse of family planning. As i n earl.ier analyses, nonacc=;ptors c l e a r l y differed from acceptors i n t h e i r use o i healch services" I n ao case, hobever, a r e prior and subsequent users of i a n i l y pianning s t a 2 i s t i c a l l y d i f f e r e n t i n t h e i r receipt of health services i n 1971. For example, 59 percent of women who di.d not practice contraception during the project used i i l n e s s care services--compared with 76 percent of women d o practiced family planning e a r l y i n the project and 78 percent l a t e i n the project. The difference between 76 percent and 78 percent is not significant. Because s t e r i l i z e d women ( o r women whose husbands -were s t e r i l i z e d ) received other c a r e at a much reduced r a t e a f t e r 1970, they vere excluded from the analysis of other care. The remaining family planning users were almost completely covereti by these ~thr.:'health services (91 t o 93 percent) i n con'srast t o nonusers of family planning (62 percent). Use of children's services followed similar patterns: children of family planning accei~torshad higher coverage by health services i n 197! than nonacceptors, especlEally by other . t L services, but coverage was not distinguishable by t h e time of practice of - - family planning. - The most striking finding condnues t o be the importance of con- P - current use of health and family planni services. The implications of r h i s finding f o r program inanagement a r e clear: I f health services a r e to sustain t h e i r effecr: on tne practice of fami.1~plarning, those servicss rnuet be well coordiuatea ar 1 combined v it h contraceptive service8 . The finding gives cxedex2 t o t'le ~ s t u l a t e dimportance of the d i r e c t linkit-4 of integrated health and f8mri.y plamtng servi2os. USE Ou HEALTH SERVICES: L I. CONTROLLING FOR PRlOR PAHILY PLANNING USE e i-- i. Pr:.or usxs of family planning a r e more I :':ely t o become progriun users. That I has been established. Eut thette prior users night have been (;c!nerally inclined k to respond t o a variety o i nodern services, inciuding health c:.re, vhen it was convenient. IF t h a t vere so, acceptance of Fanily planning unc1e.r the project might have merely been a response to e prior f e l t need f o r family planning-- and the causal r o l e of health services use w ~ u l dhave Seen spuricus. We therefoxe investigated the three-way r e l a t i o n between prior use o'f family planniqq and 6!~e usr 42 health services and f m l l y pTanning undler' the project. A ~ r i ~ : history of f a a i i ) plan&+-ng pract'ce was indee:d a:-iociated with greater use 02 praject health services f o r women, t:lo.;gh riot. with tbwt of those f o r childrerr (table 3.2). Wfimen who had not used f w i l y plsnning befare the project hod a m e r a t e of illncsv care of about 63 percent i n both 1970 and 1971,. The r a t e s f o r users of traditional family pl&nin3 methods were , 74 percett -2 70 percent: f o r prior users of modern contraceptives, they were 75 percent a d 76 percent. Alchough mar?. wosen received other care than i l l o e ~ ocare, t h e pattern waa similar f o r prior users and nonusers of family - plannl:,g. excepc f o r ?rior users of nodern 8ethods i n 1971. Only 64 percent of I *he prior users ~f modern methods used services f o r ott.er care, c:l7>mpared with 76 percent of ~ z i o rw e r s cf t r a d i t i o n d sthnds. Because t h i s c ~ i l snot seen i n Table 5.5 RZLXZIONSMP OP ?JLKILY P-L\'G P- P a O R TO 'LBe - P - T O U S Z O ~ ~ S P N I C E S ~ T E O R O J E C " A Typa of R08lth S.rcrica u . s s CaRS ( a w r C ) O T B G R m- Y a u ud Ptior ~ e c f p i a n t m US. a 'JW xZ * 0- x NO USE 649 63.6 16.1 7 73.8 - 153 - guidelines of the project. Host prlor users of modern methods had been steri- lized and therefore did not need v i ~ i t st o assess t h e i r fertility. Although women of different ages used health services a t d i f f e r e n t rates, t h e i r use of family planning produced such similiir e f f e c t s on use of health services t h a t a l l ages have been combined i n table 5.3 (and i n subsequent tables, unless specified otherwise). t In the analysis of children's services, women v i t h children under i three were analyzed separately beca~.seother care was targeted exclusively on t h i s group of children. There was l i t t l e difference i n the use of children's sr,-vices between 1970 and 1971. So only services i n the second year a r e shown i n the table. Although the children of women who had never practiced f m i l y planning used children's services s l i g h t l y less than the children of prior users of family planning (whether traditional o r modern), the differences were not significant. The use zates f o r i l l n e s s care services and other care were almost identical. Findings were similar f o r women v i t h children over three, but the use r a t e s f o r these services were lower. In summary, the main d i f f e r e n c ~ si n the use of health services were between nonusers and prtor user,s of family planning. D i f f ~ r e n c e sbetween prior users of traditional modern methods were significant i n only one case (women's other care i n 19 1). - Women who used family plannin;: before the .- .a - project mURt have been more open to using health services as w e l l a s family planning under the p@ject (see chapter 3). The nonusers of family planning m before the project may represent a more resistan2 population that c a l l s f o r USE OF HEALTH SERVICES: CONTROLLING FOR PRIOR AND PROJECT FAMILY PLANNING USE I n the combined relation between prior and project family planning use and use of health services, t h e largest and most significant variations were f o r women's services (table 5.4). (Use of family planning a t any tine during the project ues considered to be project use.) For i l l n e s s care, f o r example, the use i n 1971 ranged from 55 percent for women who were neither prior nor project acceptors of family planning t o 82 percent f o r women who us& modern methods i n both periods. Among women whose practice oc . a i l y planning was confined to one period, project users had a somewhat higher r a t e of using health s e r vices (75 percent) than prior users (64 percent). Prior users of traditional [ metho& were combined with prior nonusers i n t h i s analysis because they were comparable i n t h e i r use of health services. Age had no e f f e c t on the differ- ences noted above. Womr rrcategorized by their family planning experience used other care much a s they used i l l n e s s care. Project family p l a ~ i n gueers who had used modern methods before the project were found t o use other care a t a r a t e of 63 percent, but t h a t r a t e r i s e s to more than 90 percent i f women who used a permanent method (fubectomy o r vasectomy) a r e excluded from the anal3sis. Women thirty- five and older used other care less than younger womeU, but t h e - trend was i n the same direction. i '1 - More than 76 percent of the children (under three) of women who had - i vt F used family planning used either i l l n e s s care or *her . - care. Use of these services was between 83 and 87 percent f o r children whose mothers were prior users of family planning but not project users and between 90 and 100 percent i f t h e i r mothers were project users. The use of i l l n e s s care by older children Table 5.4 cam= ErFEc- QF FRIOR m D P.%RAH FiL'lItY PGtnuxG Ps&cTxcE a apuTsS m a S IS 1971 Type of E U t h s e e . 1Un.s~Car. (Clinic) 0th.r Car. Redpieat 02 S d c . aad Ty3. o f FP US8 n U ~ b g x2 utkut x2 EnmiCcrY <3 tiam 106 76.4 U.06 76.4 32.49 (p a .OOS) (p < .oo1) Prior, No Pgm . . 69 82.6 87 .O ra Prior, Pgm . had 1 Prior. P9m +a Prior, Pgnr 28 89.3 100.0 QfaDm23 C ~ona 178 50.0 22-84 (p-g.001) - Prior, No Pgm U8 59.5 - - N o Prior, P g a3 134 Trad Prior, Pgm . 8 :bd Prior, Pgm 0 35 - 74.5 *Ewer in parenthesis indicates percent using services when sterilized wonen are excluded from the analysis. (over three) varied i n much the same vay-from 50 percent by those whose mothers did not use family planning t o 75 percent by those whose mothers were project family planning users. Among nonusers of project family planning, there was a significant difference betveen prior users and nonusers i n the use of women's health b services. For example, the use of i l l n e s s services by 64 percent of the p r i o r users is significantly greater than the figure of 55 percent for nonusers. But corresponding differences with respect t o child care a r e not significant. Among project usefs of family planning, there e r e no significant differences i n the use of any category of health services according t o prior family -". planning use. To summarize the findings, prior use of family planning contribv.ted t o the use of project health services, especially those f o r women. Even stronger, however, was the associatron between project use of family planning and health services. In f a c t , the use of project family planning essentially obscured the e f f e c t t h a t prior use of family planning had on the use of health services. USE OF FAMILY PJXWING UNDER THE PROJECT: L CONTROLLING FOR PRIOR PRACTICE AND FOR USE OF HEALTH SERVICES The much more important question of i n t e r v e n t i ~ ndeals with the possible influence of the use of health services on the use of p ~ o j e c tfamily planning, controlling f o r prior use of family planning. It was e*ined by using the percentage of women practicing contraception a s the dependent variable ( t a b l e A strong association be-2n the use of services and of project I family planning was again shown, but t h i s time the e f f e c t of prior use of I family planning was more clear. Among nonusers of women's i l l n e s s services i n 1971, f o r example, the acceptance r a t e f o r project family planning was 57 percent among prior users of modem methods and 28 percent among those who were not prior'users. Use of health services raised the acceptance rate to 79 percent f o r prior users of modern methods t o 48 percent f o r those who were not prior users. Results f o r prior use of traditional methods were i n between. Generally, patterns were similar i n otner categories of nealth service. The r e s u l t s tend t o reconfirm the e f f e c t s of both the prior use of family planning and the project use of health services on the project use of family planning. But a r e these e f f e c t s separate and independent? The chi- square values of table 5.5 suggest that they may not be, f o r the use of health services invariably produced more significant chi- square values among those who were not prior users of family planning than among prior users of modern methods. I n f a c t , t h e use of other health services produced a negligible, o r a t least noneignificant, increase i n acceptors of project famiiy planning among prior users of modern methods. The question of independence is examined morelclosely i n figure C 5.2. If the e f f e c t of using health services vere constant, regardless of prior experience of family planning, ea* pair of curves i n the figure would be '5 parallel. That is, the solid l i n e f o r users of health s e n i c e s would l i e a t a - I fixed distance above the broken l i n e f o ~ o n u s e r s . Such a pattern is not - observed f o r users of women's other health services. Deviations h 1 expzcted results, hypo thesizing independence, have been tested to p r o d ~ c ethe chi- square values shown i n the figure. Consider the r e s u l t s for use of women's i l l n e s s services: For nonusers of women's i l l n e s s services, irrespective of prior use of family planning, the r a t e of project use of family planning was 35 percent. For users of those services, the r a t e vas 55 percent. I f the effect of using health services vere hypothesized t o be constant a t 20 percent ( 5 5 percent minus 35 percent) the acceptance r a t e among those who were not prior users of family platning would be expected to go from 28 t o 48 percent with the receipt of health services. Similarly, the acceptance r a t e s use of health services shou1.d be 60 percent f o r prior traditional users of family planning and 77 percent f o r modem users. The comparison of r e s u l t s v i t h these expectations yields a chi-square of 5.40. Application af the procedure to a11 categories of health services ccnfirms-dramatically-the difference i n e f f e c t between services f o r i l l n e s s care and other care. Other services should definitely be directed to those with prior family planning experience. Although the chi- square values associated with i l l n e s s care a r e small, they suggest a characterisgic of p r i o r users of traditional methods of . . family planning. Without i l l n e s s care, such vomen vere more l i k e l y than nonusers t o become project acceptors. Among health service users, hawever, ' .s - there vas l i t t l e difference between nonprior users and traditional uzers i n the - * . acceptance of prefect family planning. Thus the incremental e f f e c t @fusing health services on the acceptancz of family planning was small f o r those with previous experience of traditional methods. This is shown i n figure 5.2 by steeper slopes f o r nonusers of i l l n e s s care than f o r users. - 160- Figure 5.2 G3AP3 OP -S BETUEEt4 PLEOGBAH P A Z Y PIX-C, PRIOR FMcLY PL?mlRlG A310 mTS SERVICE USE %=. PO SOT Mod. m mad. Hod. Prior PP Us. m. Prior .FP Use Prior FP Us8 Prior experience of family planning increased the r a t e of use e of project health setvices. The use of health setvices i n turn increased the likelihood of acceptance of project family planning. What was the double- I I - barreled effect of t h i s prior experience? Results of the investigation of & t h i s question a r e shown i n table 5.6. Consider women's i l l n e s e care. Of 649 women who were not prior users of family planning, 409 (63 percent) were recipients of women's i l l n e s s care i n 1971 (see table 5.5). Users of illness care had a practice rate f o r project family planning t h a t was 19.8 percentage points (47.7 - 27.9) higher than nonusers of such health services. Thus f o r every 1,000 women without prior experience of family planning, 630 vere reached by i l l n e s s care. These 630 contributed 125 family planuing acceptors over and above those who could have been expe-ted t o become project acceptors without t h i s i l l n e s s care. The r e s u l t s of similar calculations f o r all categories identified i n t a b l e 5.5 a r e summarized i n tab12 5.6. They a r e not significant; i n p a r t i c u l a r they do not consider interactive e f f e c t s of d i f f e r e n t combinations of services. Nevertheless, t h e findings suggest possible guidelines f o r emphasis i n programs. The e f f e c t of i l l n e s s care on the acceptance of family * planning is f'airly uniform, regardless ('1prior experience of family planning. --Although prior practice of family planning reduces the incremental e f f e c t of the use of health services, the increased likelihood of health services use i n 'I - t h i s population group provides a larger subgroup of users. This is i l a s t r a t e d f by the following example of offering wonten's i l l n e s s services t o prior honusers of family planning and t o prior users who used traditional methods. Table 5.6 !Wne SO0 236 Traditional 375 2!56 %umbers reflect rahtiva maqnitudes only. Thoy are not additive across serPices -use separataarralyses did not examine s d c e inter-relationships. .. Prior family Users of Incremental planning Hypothetical women's i l l n e s s e f f e c t on use of experience population senices family planning Gain None LO0 x .630 x .198 a 12 Traditional 100 x .740 x ,136 = 10 Haking children's other care available t o the children of prior users- of modern methods of family planning produced an ef?ect on project use of family planning that was cdly about half a s large a s that from making children's ' i l l n e s s care available. But children's other care had about three times more e f f e c t among those without prior family pLanning experience -ban among former modern users. Similar relations of effektiveness a r e obsemed for vomen's semices. These variations i n the e f f e c t of services have implications f o r targeting semices so as to mad.mize the practice of family planning. I n integrated programs of health and family planning, these e f f o r t s would have to * I r be balanced by possibly competing demands f o r a d i f f e r e n t distribution of the services to maximize t h e i r e f f e c t on health. - But a s we-found at Narangwal, prior nonusers of family planning tend to be those who 'seed health care the -- mosl . .8 s 1 The use of health services a t Narangwal was consistently associated with increased practice of family planning. Women who received i l l n e s s services had a family-planning-use rate 22 percentage paints higher than those who did not receive i l l n e s s services. Women who received women's other oenrices a o r whose children received i l l n e s s o r other services had a family-plannfng- use r a t e about 35 percentage poin.s higher than aonusers of these health services. When the patterns .f use of health services and family planning were examined over time, it was shown that concurrent periods of use of health c a r e were more closely associated with fanily planning than e i t h e r prior o r subsequent use t f health semices. I f women used modern family planning wethods before the project and i f they o r +heir children nsed any type of h e e l t t service during the project, they had the highest fdly-plauning-practice r a t e d u r h g the project (between 70 t o 80 percent). I n contrast, i f women neither practiced family p l a n n i q i n t h e past nor ccrrently -.sad healch servicer, they had a family- planning- practice r a t e of only 20 t o 30 percent d u r i q t h e project. ' T1 - use of ;llness services increased the use of fautly planning among e l i g i b l e women during t h e pzoject no matter what t h e i r prfor experience I --. of -fanL;Ly--pl'anning had been. .. Ths use of other s ~ r v i c e s ,however, had more .# - e f f e c t h women who had never practiced family planning o r who used traditional - - . methods-before the project than on those who were prior users of mofiern i ~ ~ , t h o d s S e So, a l l health services were effective i n recruiting new acceptors of ~noGem methods (from those who never used family p1,nniug before), bct other services did better than i l l n e s s services. Robert L. Parker, Xil.licun A. Rrinke, Chapter 6 R.S.S. Sama, and Warindar U. Y e l ? ~ .-a Determlnents of the Use o w Famfly P l a ~ n i n g Thu preceding chapters have zstablished tm strong aesoci,*+.iona: that between tha use of health senices a~ the p r a c l ~ c eof family planning and that of the influenca of such prerllspo~iagfactors aa attitudes, helief 8, and sosiecono;nic chazactsristice: on the 3se s f hea1rl.r services or :.he practice of family plan- ning. %cause health services =re. the primary iaterventiou variables fin the qtuiy, ve had to determine the strength of their essociation with famiAy pbauiing practice while c m t r o l l i n g for the h o r n ?ffect of the predispoa5ng f a c t ~ r e . For cacar?le, i: is iuiportant to knaw vhether services inducz s b i l a r or different chu *, . 'n tne use : r> family plarining mong women who h z ~ e d i f f e r m t attitudes toward fami2y planoing. The results could be vsed to ldentffy :#oaen vha w u l d be more responsive o r lesa responqive to oe?aice intzrvenc.tons and to determine p r i o r i t i e s f o r target groups t h s t ::ava specific p r d l s p s i n g c h a r a c t e r i s t j . ~ ~ .In addltioa, the findf rg that a prdispcsirig chr-ractezistic had vc influence 0.1the impact of s a v i c e s could ~ntiicatet h a t i 2 vaf: unimportazit or trxtt the sarrices were de1.ivsred i n s vay that . r t uentralized its influence. The f i r s t enalpsf s i n this chapter therefore exarrf*:es the thrse-- - vap association betwecn health services, the practice -f f a x i . 9 planring, and attitudes, belief, and sociecoaomic characteristics. Allsrariables of I attitudes; belfefs, and soc.ie.:ono:pic characteristics in t h i s fia2ysis are from the midproject crosa-sectioml surveys (1971). Variables of health :e..--~icrs and family piknnin~aye frgm the crxperienze of indivi.i~:.ala ?l~roughout the project :i969-73;. To obtain a more compreheasive view of the r e l a t i v e importance of the many factors examined, a l i n e a r multiple regression model was developed t o r e l a t e the practice of f m i l y planning (the dependent variable) t o attitudes, beliefs, knowledge, practice, socioeconomic characteristics, demographic factors, and the use of services (as independent var: 51cs). WOMEN'S AND CHILDREN'S KEA.LT!i SERVICES > The gerceutaqe of women who practiced family planning a t some time during the I projef:t has been relaced t o t h e i r use (or t h e i r ckildren's use) of health services. dsers of wamea and children's services were categorized by no use, some use (ia e i t h e r 1963-71 o r 1972-73), and f u l l use ( i n both periodsj. Yhe;e categories were then used t o subdivide women with s p e c i f i c response! categories f o r each predfsposing variable. I n the most favorable ciccumstances--when women approved of famlly plamiag and used servlces f o r i l l n e s s care throughout the project--the r a t e of prscciciag family ~ l a n n i n gwas more than 60 percent. This r a t e Is twice t b i t cZ the disapproving nonuser and three times that of the ~ m c e r t d i nnonuser* As m ~ n t i m e d.in chapters 3 and 4 , wc.len uncerzain about family planning appeared more negative to project servitks than disapprovers of famtly planning. The e f f e c t s of the use of services on the practice of family planning, ho1dia.g a t t i t - d s s about family p l a n n i q co?atant , were significant f o r approvers and 'uncertain women-but not significant for disapprovers, even though the direc- tion of cllange associated c l t h use of services 23s positive and on the order - o I seventeen percentage points. The e f f l c t of women's other services a s well . a s both curative and other services f o r children w a s consistently strong on che practice of f a q i l y planc .ng f a r a l l women, including disapprovers. Other care services for children (preventive and surveiilsnce v i s i t s ) produced the wldest spread i n r a t e s of p r a c ~ i c i n gfainily planning- Moving from disapproval t o approval within a use category o r from one use category t o the next higher category, with the a t t i t u d e held constant, i r ~ o l v e dan approxisate iucrease of twencp t o t h i r t y percentage points. * It appears that the effects of attitudes towazd family planning and of-child- ren's other care, o r any of the women's o r children's services a r e additive and st least partially independent. It also appears that the effect of services was simi!.ar regardlese of the women's a t t i t u d e s , except f o r vomen's i l l n e s s care among disapprovers of family planning. The additive e f f e c t of a voman's communication with hes husband about family planning and the use of health services was equally dramatic. I The use of children's other care again produced the most consistent findings. For example, family planning practice r a t e s f o r nonusers of services ranged from 12 percent f o r vomen who did nQt t a l k with t h e i r husbands t o 35 percent f o r those who did. I n contrast, women who did not t a l k with t h e i r husbands but whose children received other care part of the time had a practice r a t e of 34 percent. In both instances a change i n behavior was related to an increase i n the family planning practice r a t e of about twenty percentage points. A l l these changes a r e highly significant- and j u s t a s noteworthy, changes of such magni- t t x a should have great practical importance. Hcre than any variable of a t t i - 4 t i e s : be?iefs, and socioeconomii characteristics, the variable of a woman's talking with ker husband abouc family plandng contributed t o differences i n I family planning p a c t i c e about a s much a s services. For women's i l l n e s s !ervices, the independent contribution of talking wfth the huslrand is s l i g h t l y P greater than the independent contribution of t i a t service. T h e e findings a l s o support the importance of such communicati~nbeiween a woman and her husband abcut family planning. Differences i r beliefs about changes Fn the frequency of child dearhs over the previous t h i r t y years produced only small v a r i a ~ i o n si n family planning practice-rates i n any category of health service use. The e f f e c t was l e a s t among women (or their cMldren) who used services throughout the prcject ( f u l l use). In contrast, i n any belief category, practice r a t e s of family planning were consistently and significantl: related to the use of health services. What is particularly important, services had a strong effect on the practice of family planning, no matter what women believed about changes i n the survival of children. The effect was most dramatic for women provided with services throughout the project. This finding supports the hypotkeses that services could modify the beliefs and actions of women by changing their perception of t h e i r children's chances of surviving. If the socioeconomic characteristics of women--caste, jncome, hus- band's education, and ownership of land or selected modern possessions--were held constant, associations between the use of women's and children's ser- vices and the practice of family planning remain highly significant. Generally a change from no use of services to f u l l use was associated with an increase of twenty to forty percentage points i n the proportion of women practicing i d l y planning. The only times that serv4ces did not a f f e c t family pl.anning practice a s strongly were with women's i l l n e s s care among women with no land, with medium income, o r of low caste. Tn contrast, i n only a few 4nstances did these socioeconomic characteristics have an e,ffect on the practice of family planning i f services were held constant. - Host of the exceptions involved the .r - use of children's s e n h e s . Fcr example, among women whose children had i l l n e s s care during o n g p a r t of the project (some use), r a t e s of practicing family planning f o r wonen vkiose husbands had fewer than s i x years of education were twenty psrcentage points lower than those of wo.ten whose husbands had six i o r more years of ed~cation--29 percent compared with 49 percent. I n summary, t h e use of h e d t h services was much more strongly associated with the practice of family planning than v s r e the socioeconomic variables, and it probably overcame most of the preexisting differences i n the use of family planning related t o these characteristics. These services almost invariably vere women's i l l n e s s care i n the fev instances i n vhich services vere not strongly associated v i t h increases i n the practice of family planning MOTIVATION OF HUSBANDS FOR FAMILY PLANNING General and motivational contacts with husbands by male family planning work- e r s were aLso analyzed f o r possible associations with the practice of family planning and with the variables or attitudes, beliefs, and socioeconomic -, characteristics. For these analyses, women were identified a s having husbands who e i t h e r had o r did not have such contacts. Ths r e s u l t s w e r - j u s t as dramatic as those with women's and children-9 health services- practice rates ranged from 24 percent for women whose husbands had not been contacted by the family planning workers and who did not t a l k with them about family planning t o 69 percent f o r women whose husbands had been contacted and vho talked about family planning. Women v i t h husbands who did not ~z.7.e such c m t a c t s but who talked with them about family planning- -or nontdkers who had - f anily plannipg sontacts-practiced family planqfng a t almost the same interme- li d.,ate rates of 47 percent and 49 percent. Although t h e association between - ? * motivational contacts with the husband and the p w c t i c e of family planting is strong, the contacts could either be stimulating practice o r be a secondary result of i n t e r e s t i n practice i n i t i a t e d by the mail receiving the contacts- - The direction of cause and eZfect has not been established. The variable of beliefs about the frequency of child deaths and that of special socioeconomic factors had l i t t l e influence on practice r a t e s when general and motivational contacts with husbands were held constant. But these contacts still were closely associated with the practice of family p l a ~ i n gwhen the other variables were held constant. I n surrmary, the threeway analyses of men'sr, women's, and children's services indicate that the associations between the use of services and the practice of family planning often add to and a r e independent of the e f f e c t s of attitudes, beliefs, and socioeconomic characteristics, thus strongly supporting the importance of the relation between providing services and practicing family planning i n the Narangwal setting. The data also suggest that the associations between attitudes, beliefs, and socioeconomic charac- t e r i s t i c s and the practice of family planning--associations that were seen before the project but t h a t became l e s s evident during the project-may have diminished because of the high use of project services. Health services were i n general used equally by a l l groups i n the population, and the influence of t h i s use on increasing the practice of frmily planning generally was also similar i n each group, no matter what rhe i n i t i a l use was i n that group. L REGRESSION ANAi,YSES OF THE USE OF HEALTH SERVICES AND THE PRACTICE OF FMILY PLANNING .r Three regression analyses w e r e conducted. In the f i r s t , the independent -- .s - variables ceasuring specified services were simply defined a s use o r nonuse E * i n one. two, o r three year5 of the period 1969-71. I n the second acd third analyses the number of s e r v i l e contacts i n 1969-73 was used. As i n the preceding section, the services included women's il.lness and other care, children's i l l n e s s and other care, and male motivatl@nalcontacts. These variables vere considered t o have d i r e c t program implications. Measurements of contraceptive practice a f t e r the baseline cross- sectional surveys were restricted to methods offered by the project: condom, o r a l p i l l , vasectomy, tubectomy, intrauterine device, and injectabls Depo- provera. Use or nonuse of one or more of these methods during the project was the dependent variable i n the f i r s t tw regression analyses. The dura- tion of the practice of f d l y planning i n days of use during the project vas the dependent variable i n the third. Couples who had a vasectomy cdr tubectomy before the project did not have an opportunity t o accept a method through project sources and were excluded from the analysis. And because outmigration and dissolution of marriage would deter the acceptance of con- traception, only couples who were i n experimental villages and currently married a t the time (1969 and 1971) of the two cross- sectional surveys meas- uring attitudes, b e l i e f , and socioeconomic characteristics were included. The analyses were therefore conducted on 1,187 wcmen aged 15-49 who f u l f i l l e d these c r i t e r i a and who provided complete information on a l l the variables. @ The variables of attitudes, beliefs, knowledge, practice, and socioecononic and sociodemographic characteristics were considered t o be exogenous (nonmanipulable) but potentially important determinants of the use of family planning and of interest i n targeting services. - The exogenous variables were generally limited to those that s h o a d some promise i n preli- I minary bivariate analysis (see the left- hand column of table 6.1). Results of the f i r s t regression analysis ~ssingthese exogenous variables and the 7 . e ~ r i c eintervention variables mentioned e a r l i e r a r e i n that table. Table 6.1 UF TBE r m m L E -STON ANALXSIS EXMXXIX SMVICP, GTZLIZATIQN IN 1969-1971 iLW ITS P T t E ON USE OR SON-USE OF FASLY PLWNXSG AT i L t % D m TXE PRWECT (l969-1973) mqrarsion Coefficients Md th& Levels of Siqdficaace Education of tnuband .0543 1.9 *.05 c Oc=upation of Muband: Labor p-g OthsI Religion-caste: Jat Sikh -.Of61 1.6 nos. lunalsia and other low c u - 4 -.0579 1.3 a.s. other L i v h q C h i l d r a n and Wee's Age: 3 and < 35 .I894 3.6 c-001 5 3 a a d L 3 5 a -. > 3 a a d < 3 5 .2788 4.7 e.001 > 3 m d 3 3 5 -0997 1.9 *.05 . Srrrusr of m e n D.rd -.0305 2.1 <.05 Aw=ar.ess of Wrn ,Contraceptian Prior Use of Contzacsption Tal!! with Husband .0893 3.2 e.005 ~ezvice~ti-lizati~n~omtn's1ll1~4ss -- : .0481 4.1 <.001 WoPun's War .0032 0.2 n.3. QF1dren's Illness -.a199 n.s. 'J 1.1 a -- - L Children's 0eh.r .lo49 6-1 <.001 Hale .Motivation .lo22 5.3 <.W1 - * - . rr2 I 0.24 suppressed ca-ry; effect includad in constant term The regression analyses helped i n assessing the separate and com- bined importance of interesting variables, but three limitations of the analy- sis must be borne i n mind. F i r s t , only l i n e a r effect3 were considered. Thus, f o r example, the second year of use of health services was assumed to be a s effective as the f i r s t . Second, interactioss were not i n c l u d d i n the analy- sis. It is possible that the effect of joint use of services was, i n fact, not precisely the sum of the separate effects. Third, the regression analysis does not give explicit recognitioa to the f a c t that predicted values of the dependent variable (such a s the percentage accepting family planning) cannot exceed LOO. Despite these possible limitations, regression r e s u l t s conformed closely t o observed conditions and led to the conclusion that the simple regression models used were adequate. Socioeconomic variables Religion- caste and occupation were not significant , reflecting the success i n the equitable distribution of contraceptive use in the population. Education of husband was slighrly significant i n the expected direction: the higher the husband's education, the greater the contraceptive practice. Demographic variables . I There was more contraception among young women (under thirty- five) than older uomen. This w a s clarffied by the number of surviving children. 'J Among younger women, the contrace?ption r a t e f o r those who had more than three - * children w a s higher than t h a t f o i t h o s e who nad fewer children. - This has obvious implications f o r deLining a particularly receptive target group f o r family planning. A significant negative association vas found between wmen's contraceptive practice and the number of live- born children who died: the fewer the children who died, the greater the contraceptive practice, even f o r women with the same number of surviving children. Attitudes, beliefs, knowledge, and practice Approval of family planning, belief about child mortality, communica- t i o n with husbands about family planning, awareness of modem contraception, and prior practice of reversible methods of family planning constituted t h i s 1 group of variables. Awareness of modem contraceptive methods was found f o r 93 percent of study women and obviously was not crucial i n subsequent contraceptive behavior. Approval of family planning and communication with husbands did help subsequent contraceptive practice. Beliefs ?borlt child mortality were i n the expected direction, with greater contraceptive practice r a t e s among those who believed child mortality t o have been l e s s today than t h i r t y years ago. That t h i s variable vas not significant i n the regression analysis- - as i t had been i n the bivariate analysis- is a t l e a s t partly due t o a small but significant I correlation with experience of child loss. Prior practice of reversible r I methods of family planning :as significantly related to project contraception. - Service inputs i ' ? -Use of a l l f i v e service input variablez during the f i r s t three -L * years of y e project was related to the practice of family p l m i n g . . - Of special significance was the use of women's i l l n e s s services, children's other - 175 - services, and male motivational services. Note i n table 6.1 that eight of the twenty variables were significant: age of mother and number of living children, number of children who died before the project, prior use of contraception, communication with husband, approval of family plaaning, and use of women's i l l n e s s services,, children's other services, a& male ~ o t i v a t i o n a lservices. To identify the balance of influences, we s t a r t with a hypothetical baseline population that is average i n social status, i n knowledge of modern contraception, i n beliefs about child mortality, and i n the use of women's other services and children's i l l n e s s services. Further assume that the wives a r e under thirty- five, have no more than three living children, have had no child deaths, and a r e average i n previous contraceptive use. The regression equation predicts :hat 9.8 percent of these baseline couples w i l l become project contraceptive users without further project iaputs. Because the l a s t three factors mentioned i n the preceding paragraph were significant, different assumptions would a l t e r the forecast of project contraceptive use indicated i n table 6.2. For example, i f i n t e r e s t is cen- tered on women under thirty- five with more than three living children and no previous experience of family planning, the table indicates an expected project use r a t e of 13.6 percent (9.8 + 8.9 - 5. I ) , s o t 9.8 percent. The results of the regression analysis summarized i n table 6.1 suggest that i f , i n addition t o the foregoing baselipe conditions, husbands and wives conum~nicateabout fainily _planning through project interventions o r otherwise, project use of famil? planning can bc expeceed t o increase 8.9 percent. Ghen couples express approval of family planning, the use raze can be expected to increase another 12.1 percent. These and the service input e f f e c t s a r e ~ummarizedi n table 6.3. Table 6.2 Variable L a ~ a U C a t s g o ~ ~ Adjuatmnc (t) > 3 and <3:5 > > 3 and -35 Numbar of CxUd D e a t h s -P r i o r Family Planning P t a c t i c o mne - 5.1 Traditional only + 3-3 .bdarrr +U.8 Table 6.3 SUMMARY OF EFFECTS ON CONTRACEPTIVE IISE Percentage e f f e c t oa contraceptive Pst-centage Rate, o r variable use r a t e of use Baseline use r a t e 9.8 percent Talk with husband about family planning t 8.9 Approval of fasnlly planning .!-12.1 Women's i l l n e s s services i n + 4.8 one year Children's other service0 i n +lo. 5 one year Male motivation services i n +lo.2 one year Modified use r a t e f o r one year of servicea 56.3 percent Modified use r a t e f o r two years of services, assuming l i n e a r e f f e c t s (+25.5) 81.8 percent I f a l l the l i s t e d effecrs befL present, the r a t e of 'amily planning use cqr',er the project would be expected '. reach 56.2 percect, an illcrease 6 cif 46.5 percentage points from baseline corditions. Of the s e w i c e vartables, children's other services and male motivation s e ~ r f c e sappear to be most importznt. Assuming l i n e a r effects, use of the three lypes of scrvtce i n two years not one, would ad3 25.5 pel entagc points t o the r a t e of us'ing family planning, b-inqiq* it t o 8L.d percent. The varlables i n t h i s aaalysis explain about 24 percent of the variation i n contraceptive p- i c tice. R e unexplained variation could be attzibuted t o variables not inciuded o r t o nonlinear components or variables already included. Part of the unexplained variation could also be due to interaction betwe ?n variables not considered. Figure 6.1 shovs the relacions betseen contraceptive p r - zt i c e and a fev of the explanatory variables. These relations are aggregate: nc adjustments have been made for the influence of other variable: The three variables--number of children who died, L.< of mala motivation services, and elucation of husband--seem t o hava f e i r l y l i n e a r relations with ccutra- ceptive practice. The relations of vomen's i l l n e s s s e n i c e s and children's other services a r e l e s s li&ar; those of women's other s e r r i c e s and chLldren'b i l l n e s s services a r e f ~ from linear, vhich s a y have influence^ their not. being r s.i.gnificant i n the regression . - - *- REGESSION ANAL'fSIS OF SERVICE CONTACTS * S AND USE OF FAMILY PLAhWING I A second regression analysis incorporated the s=:e var.i.ables a s the Z i r s t , but services were ozpresse;. a s che number of contacts during the project. Contraceptive use o r nonuse during the project was again thc dependent variable. The relations of variables of a t t i t u d e s and b e l i e f s , knowledge and practice, socioeconomic characteristics, and demographic f a c t o r s t o contraceptive use remained e s s e n t i a l l y the same a s i n t h e f i r s t regressiolr- with the s i g n and s i z e of the regression coefficients and the significance basically unchanged. Wo variables l o s t t h e i r significance: t h e number of live- born children who died and the women who were over thirty- five and had more than three children. The biggest differences between the two analyses were i n the ser- vice input variables ( t a b l e 6.4). I n t h i s second analysis, the measure of services was changed from the f i r s t i n tuo ways: the number of service contacts was exam'ned, not simply whether services were used i n one o r more years of the project; the use of services was considered during the whole project, not j u s t t h e f i r s t three years. A l l women's and children's services had a significant and positive association with contraceptive use, but U G L ~ motivation was not significantly related (and it changed signs). Other changes from che findings of the f i r s t regression i n t a b l e 6.1 are the significance and change i n s i g n f o r .ch,ildrenOs i l l n e s s services apd the highly significant association of xomen's other services i n the second regression. - 'i - The c o e f f i c i h t s i n the second regression a r e much smaller because - service inputs were e x k e s s e d by individual contacts, which were much more numerous. But the coefficients can still be interpreted as representing E(ESfJLTS OFTHE ~ D L E ReCFIE5SXON ANALYSIS M G TSE VOLtEE OP SERVICS M 1969-1973 i\bB ITS =CT Otf USE OX WN-USE W FiUULY P L A W E S AMY TIHE DO- Tf3Z PRaren (1969-1973) Itdigion-Cart.2 J r t 9Uh -.0800 1.8 n.s. almGIia rrrd ot!!a lov cutas -.0474 1.1 n-s. Othrr II e Prhax Us. of Caatzac.PtlbP . .0704 3.S <.(DO1 F L .Oll2 t maf abutChild-+p 0.7 nor. n.a. = not significant (p>.05) approximate increases i n the percentage of coatraceptive users i n the popu- l a t i o n given an increased input of one unit (contact o r v i s i t ) of service per woman. In t h i s context one contact with women's other services is seen to have a h o s t nine times t h e impact of one contact with children's other services; a v i s i t f o r vomen's i l l n e s s services is about twice a s effective a s a v i s i t f o r child's i l l n e e s services. But when the e f f e c t of services is examined in r e l a t i o n t o the mean number of services received, the picture is somewhat different (table 6.5). I f the coefficient for each service variable is multiplied by the mean number of service contacts, the e f f e c t of children's other services i n the project is shorn to be as strong a s women's i l l n e s s services, but still not a s strong a s women's other services. Women's other services were associated with a 14.4 percent increase i n the probability of accepting famiLy planning, compared with a n increase of some- what l e s s than 5 percent associated with women's i l l n e s s and children's other services. I n t h i s case the r e l a t i v e importance of the variables i n explaining differences i n contraceptive use i a about the same a s the ranking of the "t" values (table 6.4) and the beta coef f icents, normalized coefficients that take into account differences i n standard deviations of the variables (see also table 6.5). r Here a r e f i v e possible explanations f o r the differences between the - - two regressions: o F i r s t , because most women 'received women's other services, '5 - the r e l a i i v e l ~simple indication of the use o r nonuse of E k I these s e k i c e s i n the f i r s t regression obscured a n e f f e c t that showed up strongly i n the second. Apparently the Tabla 6.5 HEAbl MMBlSR 03 S E R V I a AND ZEGiOSSXON -1- --- m TEz. -1- ;\wLPsxs rst TABLE 5.12 (a)l(Z B a t 8 s r r ~ i ~ . ' vuirbl. o f (x) Cortfiduru (a)_ x 130. C o c i f l c i m u '-'I Ills- l0-1 .OO26 4.7 0.1472 Woma's Othu 20.3 . W 7 l 14.4 0.1729 Childrm'~0th- 60.3 .0008 4.t 0.0973 . s I . frequency of contacts f o r women's other services (which included aotivation to accept family planning) was impor- t a n t i n i n i t i a t i n g contraceptive use. o Second, the use of women's i l l n e s s services w a s important no matter how these services were measured. Because i l l n e s s is a random event and a v i s i t f o r i l l n e s s care is usually i n i t i a t e d by the patient, the a v a i l a b i l i t y and use of these services vhen needed seems t o be a s important a s the frequency of t h e i r use. assumed that a similar argument holds f o r e emergence of children's i l l n e s s services a s an important explanatory variable i n the second regression may be related t o the longer dura- tion in the second regression (1969-73) than i n the f i r s t (1969-71). The association of children's i l l n e s s services with the practice of family planning may have been greater i n the l a t e r stages of the project, possibly because of improved service linkages. There may a l s o have been a cumul?tive e f f e c t of these services o r a lag between the use of these s e n i c e s and t h e i r e f f e c t ( a s postulated i n our statement of the child- survival hypothesis). - Fourth, although still importamt i n the second regression, .t children's other services had a relatively smaller e f f e c t than i n the f i r s t regression. This difference is probably related to the very high frequency of cortacts i n the FPCC villeges, a pattern s t r i c t l y prescribed by tha require- ments of the overlapping nutrition study. The nmber of prescribed contacts was rrignificantly smaller i n FPUSCC villages, but coverage of children i n any year by prescribed services was similar. Simple use o r nonuse of services therefore was probably a more appropriate measure across all experimental groups providing child care. o That male motivation services l o s t significance i n the second regressicn MY also be related to timing. Motiva- tional contacts may be effective i n the early stages, when the number of possible acceptors is large. But as the number of candidates for ready recruitment falls, the marginal effort t o get a new acceptor may greatly increase the motivational contacts needed to recruit new acceptors. This seems t o be borne out by the bivariate analyses i n figure 6.2 (e), vhich shows a curvilinear relation with male motivation up to a a a x h of five t o ten contacts. The percentage of couples with eleven o r more contacts who accepted family planning for the f i r s t time f a l l s off. I t A l l the bivariate analyses of service variables used i n the second regression analysis are summarized in figure 6.2. - - The analyses are all highly signifi- cant, but a s i n the firseregression, the assumption of linearity holds only 5 for both kinds of vomen'gservice. * Of special interest are the curves for children's services, vN& are remarkably similar to those of figure 6.1. Zero use of c'nildren's services was associated with a rate of contraceptive use t h a t vas a r t i f i c i a l l y high. The: d i s t o r t i o n i n curves resulted from in- cluding vomen from v i l l a g e s not r e c % l v f ~children's care. This introduced an upward bias i n the zero rates of contraceptive use--as demonstrated i n bivariate analyses limited to data from v i l l a g e s with children's services, analyses t h a t produced e s s e n t i a l l y l i n e a r curves. REGRESSION ANALYSIS OF SERVICE CONTACTS AN3 D W T I O N OF CONTRACEPTIVE USE The third regression analysis exmined the relations of the tventy independent variables t o the duration of use of w d e r n contraceptives i n the e n t i r e project period by those who accepted contraception (nonacceptors were excluded). - The data vere expressed as t o t a l days of use, whether continuous o r interrupted by periods of nonuse, which provides a r e l a t i v e l y broadly defined dependent variable. As i n the second regression, services were defined as the number b of service contacts received by a voman o r her husband or children. The analysis included 559 women who practiced contraception during the project i n villages receiving both health and family planning services, o r twenty-nine fewer than were i d e n t i f i e d as users i n the second regression ( t h e definiti.on of duration excluded women ubose duration of use was not known). Table 6.6 summarizes the results of t h i s analysis; figuqe 6.3 gives the r e s u l t s of bivariate analyses of the independent service variables and the dependent - - variables. - - Among the predisposing variables found t o be s i g n i f i c a n t l y associated -- i - with any use of family plan+ng i n the f i r s t two regressions, only prior use of * - contraceptives was found t o r e l a t e significantly (and positively) t o days of I contraceptive use during the project. This suggests t h a t the following f a c t o r s Table 6.6 :a-n.TS OF TEE MnTI?Lz = ~ S I O H ATrilGYSIS ~~ vox.mE OP SERVICE L~?IT~CTSIN 1969-1973 XND ITS mSCT ON TEE DUICF\TION OF fP USE ALL ??axmf USERS xacrrassion Coafficiutm and Tha- L m l of sLQnFf- kvar ot Corificiaat +-+am QIgaific.sce Xmliqion Caste: Zaf Sikh 39.87 Ramdash and othrt lcn elutes -52.46 othrt < Uvbg Children and Age: ;3 and ;3S 0.5 aos. -3 aad -35 t a3 aad :35 0.9 nos- >3 aad -3s 1.5 n-s. . Prior Us8 o f Contrac- 98.55 3.1 e.005 3eliaf Xbaut C S J d m t y -15.92 0.5 nos. Aftltoda Toward , P a y Planning: Approval 124.U 1.7 n.s. ULlcUtain * . t Dhapprwal 121.22 8.6 nos. . Se-?rice Ccntacts: Waman's Illness 1.83 2.2 .. < -05 (1563-1973) Wornfin's O t h e r -3.88 1.7 n.s. Childrenms Illnam 0.51 0.8 n-s. Childten's Other -1.03 3 -9 <. 001 .Hale Motivation -21.82 4.4 e.001 :..-,,?ressed .:. category; effect included i n constant tern -\ * * =-*:ai-c or tolerance- level insuffizicnt :o enter rqression :. 3. = n3z significant (~2.35) age, parity, communication vith husband, and approval of family pl&ng. The only predisposing variable that vao significant i n t h i s regression- but was not significant in the f i r s t tw regressions- is occupatior? of husband. Occupations as farmers and (basically agricultural) laborers were positively associated with longer use of contraception. Three of the five service variables vere significantly associated with days of contraceptive use. But these results must be interpreted with care because the relations are not entirely linear (see figure 6.3). For women's illness care, the effect on days of use vas positive, but use of children's illness care and wa~men's other care had no effect. The other services vere negatively associated with days of contraceptive use, a fact confirmed by the slope of the curves i n figures 6.3(d) end 6.3(e). The follow- ing explanations for these relations are possible: o The tendency among women vho used contraceptives over longer periods is to have used services for women's lllness care a t least once. But the frequency of this use does not have an important effect on duration of use. o Among women who were sterilized, the umber of contacts for women's other care was reduced because we did not monitor their menstrual . , cycles a f t e r they vere sterilized. Womed sterilized early i n the a - project vould have high number of days of contraceptive use but fewer contacts uithPvomen0s other care. This may be the principal explanation for theZnonsignificant and negative association (and for Q the "u" shape of th; curve) i n figure 6.3(b). If women v i t h fewer than ten contacts are taken out, the relation between volume of other care and days of family planning use becomes positive and esseutially linear. o Children's i l l n e s s contacts- hich are not significautly related t o t h e dependent variable i n t h i s regression--would also become more l i n e a r and probably sigcificantly and positively related i f we excluded from the analysis women whose children had no access to child care (say, women i n the FPWS villages.) o The cabse of the negative relation betweeu children's other care and days of family planning use may be intentional o r unintentional pregnancies that interrupt contraceptive use by women who therefore had higher use of other care services f o r t h e i r additional children. o The strong negative association between male motivation contacts and days of family planning use can be e x p l a i ~ e dby the same phenomenon postulated i n the second regression: the acceptance of family planning was low among couples who had a large number of contacts due e i t h c r t o inherent greater resistance o r because they delayed t h e i r acceptance =ti1 the end of the project. The finding that the number of contacts goes up a s the average duration of use gets * . L shorter may r e f l e c t the poorer chance that Late acceptors would have i n using contraceptives for extended periods. It may also be that - - male motivation workers tended t o promote the use q£ condoms, which - - had an average duration of use a f t e r f i r s t acceptagce that was E h shorter than that of other temporary methods (twelm months compared with nineteen). Regression a n a l y s i s within experimental groups Ln planning the regressions we recognized that the absence of a i s e r v i c e i n an experimental group vould tend t o make the r e l a t i o n bctween s e r v i c e s ( e s p e c i a l l y children's services) and family planning p r a c t i c e non- l i n e a r . For example, a l l users and nonusers of family planning i n FPWS v i l l a g e s would have received e s s e n t i a l l y no c h i l d c a r e services f o r tlheir children from project sources. To eliminate t h i s e f f e c t of the a v a i l a b i l i t y of services on the use of these senrices, we repeated the regression analyses separately f o r each zxperimental group. - The two regressiocs having use o r nonuse of family planning a s the dependent v a r i a b l e gave r e l a t i v e l y s i m i l a r r e s u l t s , except tha: the find- ings were somewhat more s i g n i f i c a n t f o r s e r v i c e contact data from the e n t i r e - p r o j e c t period (1969-73) than from 1969-71 alone. The regression resuilts based on the f i r s t s e t of d a t a a r e therefore presented t3 i l l u s t r a t e the intergrgup d i f f e r e n c e s ( t a b l e 6.7). Predisposing f a c t o r s In the e a r l i e r regressions, edccation was s h o w . o be a moderately s i g n i l i c a n t determinant of family planning practice. I n the regressions shown i n table 6.7, t h i s influence was e s s e n t i a l l y confined to the FPCC vil- , . c lages, possibly suggesting that education had been o f f s e t a s a predisposing f a c t o r i n FPWSCC and FPWS v i l l a g e s by attempts t o r e c r u i t acceptors from a l l - educational s t r a t a , but chat i n FPCC v i l l a g e s the s h o r t e r period of family planning e f f o r t s had not achieved the same degree o f e q u i t q by the end of the I project. Table b.7 PEGlKSSIdN ANAI.YSBS POLL E%PEKiHUTAL CROUPS OY VILI.AbES (Exoxlning ttie E f f e c t of t!,r Volume of S e r v i c e Contnctn i n 1969-1973 011 Uae o r Non-Uee of Family Planning At Any Time Durlna t h e P r o j e c r ) R . -- C o e f f l c i e n t a , F-Values and Levels of Significance Lor Each Experimental Croup A lndcpcnden; PPUSCC P H s PPCC vilrlvblcn Cocff. F- Value S i g n i f . Coeff ; F-Value S i g n i f . Coaf f. P-Value Sigr.if. Educat Ion of hunlund Occupn t i o n of husband Labor Parulny Ot her K c l l g l o n - h u t e J a t Slkh Knsdasid and o t t ~ e rlow cu8Les Othur L f i -:.. c l ~ l l d r e l lnrrd uif.o'r a80 -r J m11d < 35 -> < 3 and 2, 35 3 and < 35 > 3 and 2 35 Nuaber of c h i l d r e n dead Awtlrcneen o f pqeth c ~ d 8 l c 0 p t l 0 1 i P r i c r use of c o n t r a c e p t i o n Dclicf aboct c h i l d mortality Talk w i ~ hlrushund A t t i tudc toward family p1annir:g Approval U~icertaln Dlsdpprovdl S e r v i c e Contacts Voracn'n I l l n e s s Uomcn's Othar Children's I l l n e s s C t ~ l l d r c n - sO t l ~ e r klale ktotivation K~ 2 H c o n t r i b u t e d by n e r v l c e c o n t a c t s 0.24 Co~ist n n t * = Suppressed category: e f f e c t included i n c o n s t a n t t e r m s . Mi = Did not e n t e r tlie regression. n.s. = not s i g n i f i c a n t ( p > .05). It appa-;a fim the results that there A Ss i g n i f i c a n t (synergistic) interaction between the husband's education and the use of services f o r child care i n influencing acceptance of family planning. Effecti., use of c h i l d care I i i is alv-ys facilitated by parents' education, and i n FPCU villages, the husbands' i b education seems t o have interacted with the provision of c h i l d care t o produce a significant e f f e c t on family planning. Younger women (under thirty- five) with more than three children were more l i k e l y t o be family planning ueers i n both FPWSCC and FPWS v i l l a g e s , but these r e l a t i o n s were significant only i n FPIJS villages. Age and p a r i t y had l i t t l e influence, i f any, on family planning practice i n FPCC villages. ':appears that thereissomeinteractionbetween thekjndof ser- vices and t h i age and parity of the c l i e n t . Women's services, i f provided a s intensively as i n FPUS villages, encourage young women t o accept family ;:lanning irrespective of the number of children they have. Women services, i f less intensive and combined v i t h child care, a r e e f f e c t i v e f o r family planning acceptance f o r women under thirty- five having three o r more children. A very i n t e r e s t i n g finding t h a t lends support t o the child- survival hypo t h e s i s and to the inclusion of children's services i n integrated health and family planniag programs i s the significant association between the number , t of a voman's children who have died and the practice of family planning i n P W S v i l l a g e s bct n o t i n child care villages. Apparently child ~ a r e ~ s e r v i c e s '2 -- reduce the offect of prior child deaths on a woman's - uillingness t o h ~ c t i c e - mily p l a m i ~ . * t Prior ;se of contraceptives uas a s i g n i f i c a n t predisposing factor only i n FPWSCC villages. The greater i n t e n s i t y of woman's o r children's services i n the other villag:s nay have made t h i s p r i o r use less im?ortant. Talking with husbands v a s s i g n i f i c a n t only i n v i l l a g e s with women's s e r v i c e s (FPWSCC and FPWS), suggesting a possible i n t e r a c t i o n betveen t h i s v a r i a b l e and the use of women's services. Womn's illness services were important vhen women's s e r v i c e s =re provided alone ia FPWS v i l l q e s . (A s i m i l a r a s s o c i a t i o n i n FPCC v i l l a g e s was due t o the use o f these i l l n e s s services as a way t o maintain rapport with family planning u s e r s , s i n c e such s e r v i c e s were not generally a v a i l a b l e t o a l l women a s a matter of project design.) Women's o t h e r services were important i n v i l l a g e s where they were provided along v i t h c h i l d care (FPWSCC and FPCC). Children's i l l n e s s s e r v i c e s vere important i n FPWECC v i l l a g e s , where they vere provided at ti.e iuitia'ive of families. Most of these i l l n e s s contacts i n FPCC v i l l a g e s were combined v i t h o t h e r services, so the children's i l l n e s s services by theniselves a r e not seen to be important i n the FPCC v i l - E lages. Children's o t h e r aervices a r e one of the most important s e r v i c e s ii t available, as s h o m by t h e i r significance i n both FPWSCC and FPCC v i l l a g e s . B These individual regressions do not change the conclusions from 1 the e a r l i e r aggregate regressions. Nor do these results show a n important i B i u t e r a c t i o n between individual services. But t h e r e is some evidence of in- C t e r a c t i o n of such services as vomen's c a r e and c h i l d c a r e with such predis- f posiag variables a s a uoman's age and p a r i t y and a husbard's education. k fhree-way analyses examining t h e r e l a t i o n s between t h e use of h e a l t h services, the p r a c t i c e of family planning, and a t t i t u d e s , b e l i e f s , and socioeconomic f a c t o r s suggest t h a t h e a l t h services had a strong and p o s i t i v e impact on family I , t 1 planning, i r r e s p e c t i v e of the a t t i t u d e s , b e l i e f s , and socioeconotnic characteris- I tics of women. Wheh these other f a c t o r s a f f e c t t h e p r a c t i c e of family planning, 1 t h e i r e f f e c t s and those of health s e r v i c e s generally a r e additive. rS:! '.tple regressiott analyses also demonstrate that the a ~ s o c i a t i o n betveen t h e p r a c t i c e of family planning and t h e use of h e a l t h services remains strong even vhen predisposing f a c t o r s a r e included i n t h e analyses. Children's o t h e r c a r e tended t o produce t h e strongest and most c s z s l s t e n t e f f e c t on t h e use of family planning. But vomen's i l l n e s s c a r e a l s o had a strong e f f e c t when women's s e r v i c e s a r e provided vithout children's services. Women's o t h e r c a r e tended t o have t h e i r g r e a t e s t e f f e c t vhen combined v i t h children's o t h e r care. The duration of family planning use was less c l e a r l y associated with h e a l t h s e r v i c e s than was the simple use o r nonuse of family planning. Chapter 7 Rashid Faruqee and R.S.S. Sarma Dsterminants of F e r t i l i t y and Its Decline The goal of i n t e g r a t e d health and family planning program8 is t o influence mortality and f e r t i l i t y . A t Narangwal, however, the s h o r t period of project a c t i v i t i e s and the f l u c t u a t i o n i n f e r t i l i t y r a t e s made it impossible t o estab- l i s h the l i n k between project components and a decl-ne i n f e r t i l i t y . Some i d e a s nevertheless emerged about possible i n t e r a c t i o n s of f e r t i l i t y with interventions, socioeconomic f a c t o r s , and demographic and biological variables. This chapter presents the p r o f i l e of f e r t i l i t y i n Narangwal v i l l a g e s a t the beginning of the project and the r e l a t i o n s between f e r t i l i t y and socioeconomic c h a r a c t e r i s t i c s of households. It next presents data on how the fertility p r o f i l e changed during the project because of the i n t e r a c t i o n of contraceptive. use and socioeconomic, demographic, and biological. v a r i a b l e s , such a s p a r i t y and the i n t e r v a l s i n c e the l a s t b i r t h . It then describes the apparent effect of d i f f e r e n t s e r v i c e packages on f e r t i l i t y . FERTILITY I N PUNJAB Crude b i r t h r a t e s and age- specific m a r i t a l l f e r t i l i t y have been d e c l i n i s f o r some time i n rural. Punjab a ~ dt t c Narangwal study area. According t:o o f f i c i a l v i t a l statistics, which arq generally assumed t o be underreported, the crude '9 - b i r t h r a t e i n Pucjab f e l l firom 45.5 per 1,000 i n 1959 t o 32.1 per 1,000 i n - 1969. I n Ludhiana ~ i s t r i c t & t h e f a l l was from 50 per 1,300 t o 34.6 i n the same period. 1/ - -1/ These estimates a r e based on two censuses-- in 1961 and 1971--and on v i t a l r e g i s t r a t i o n f o r the area. A d e t a i l e d pregnancy h i s t o r y o f each married woman i n the Narangwal study a r e a was compiled from a cross- sectional survey i n 1968-69 i n twenty-two v i l l a g e s . The survey recorded pregnancies o f women then aged 15-49, t h e d a t e of termination of each pregnancy, the outcome o f each pregnancy, t h e sex of the c h i l d , and, i f t h e c h i l d died, t h e d a t e o f death. Analysis of age- specific m a r i t a l f e r t i l i t y from r e t r o s p e c t i v e preg- nancy h i s t o r i e s shows t h a t from 1949 t o 1968 there was a progrnssive d e c l i n e i n f e r t i l i t y among women over twenty- five. That d e c l i n e was r e l a t e d te in- creasing p r a c t i c e of family planning aimed at l i m i t i n g r a t h e r thaq spacing b i r t h s . But t h e r e a l s o w a s an increase i n f e r t i l i t y f o r the 15-24 age group during t h i s period. X possible explanation may l i e i n a "catch-up" e f f e c t : women who marry l a t e have more c h i l d r e n i n the f i r s t f i v e years of marriage than those who marry e a r l y . Such i n c r e a s e s i n Narangval study v i l l a g e s may a l s o be explained by a combination of o t h e r factors. For the 15-29 age group the increase is due i n p a r t t o a s h i f t i n the d i s t r i b u t i o n of women who are married towbrd older and more fecund ages. In addition, the younger women i n t h e 15-19 and 20-24 age g r o u p i n the l a t e r cohorts, e s p e c i a l l y i n 1964-68, may a l s o have had b e t t e r r e c a l l of t h e i r l i v e S i r t h s (which, chances a r e , survived b e t t e r than those of MQ@nwho were 15-19 and 20-24 during 1949-53). Anal$ses by c a s t e showed t h a t t h i s p a t t e r n of decreased f e r t i l i t y at older ages and - increased f e r t i l i t y at young ages was l i m i t e d to (high- casts) Jat- Sikhs - ( t a b l e 7.1). - -- i Table 7.1 AGE-SPECIFIC %YEAR PERIOD MARITAL FERTILIN RATES BY hlAJOR SIKH[CASTE GROUPS NARANGWAL STUDY VILLAGES, 1949-1968 5 Year Time Age-Group of hlarried Women Period 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Jat Sikhs Scheduled Sikhs . 1949-53 34.6 229.0 341.7 326.7 282.1 - - - F. j: F Live births per 1000 married women in 5 year age-groups between 10-49 years for whom pregnancy histories (PHSurvey Round Ii) are available (N = 5126). 'J Trends i n vomen's age at marriage and the subsequent ceremony at t h e sikrt of cohabitation (muklawa) p a r a l l e l e d the d e c l i n e s i n crude b i r t h r a t e s and age- specific marital f e r t i l i t y . Much of t h e d e c l i n e i n the Punjab b i r t h rate has been a t t r i b u t e d t o a r i s i n g age at cohabitation a f t e r marriage (Wyon 1971). The d i s t i n c t i o n between the ago a t marriage and the age a t muklawa is important i n Punjab. Marriage has locg been at.a very young age f o r boys and g i r l s , p a r t i c u l a r l y f o r g i r l s . But t h e p r a c t i c e was tempered by another common obsemance: a delay between marriage and its consummation u n t i l c l o s e r to the s t a r t of menstruation. Generally, t h e e a r l i e r the mar- r i a g e f o r a g i r l , t h e longer the waiting period would be between marriage and muklawa. Table 7.2 presents d a t a o n the trend of t h e mean age a t marriage and the age a t consummation. The mean age at marriage rose from nearly f w r - teen years i n the e a r l y 1940s t o nineteen years i n the e a r l y 1970s. There v a s a corresponding change i n the age at consummation from nearly s i x t e e n years t o nineteen years. I n a population such a s t h a t at Narangwal, where childbearing begins soon a f t e r s t a r t of cohabitation, a delay of three years probably had a considerable e f f e c t on t h e b i r t h rate. The a n a l y s i s w a s based on 3,280 married women from the f i v e groups of ex- I. - perimental villages. A socioeconomic survey, done a t t h e same time a s che pregnancy h i s t o r y , provided inf ormation on income, education, oEcupation, I religion- caste, and current age o f v l f e and husband. Because there was l i t t l e v a r i a t i o n i n the vlfe's education (generally none) and occupation (household Table 7.2 WOMENS ACE AT MARRIAGE AND ACE AT COHABITATION BY MARRIAGE COHORT, NARANCWAL STUDY VILLACS Marriage Cohrt Num'bec Mean Age Mean Age at (Yex of Marriage) of Women at Marriage Cohabitation 1969-1973 517 19.0 19.1 Note: The age data relate to only "e1igil;le" women: that is, to women married once and between 15-49 years of age during the Narangwal Population Project for whom pregnancy histories and other relevant data are available (N = 4,752). I L - '5 - i * h - work) we used husband's education and occupation f o r t h i s analysis. Detailed c h a r a c t e r i s t i c s of the Zive c l u s t e r s of v i l l a g e s were e s s e n t i a l l y similar except f o r the f~ l i l y planning education group (FPED), which was older, more educated, of higher caste, and had a lower r a t i o of vonen t o men. The number of children born a l i v e was a measure of f e r t i l i t y . Tvo, other demographic variables measured ware f e t a l wastage and child loss. , Fetal wastage was measured a s the proportion of pregnancies that ended i n f e t a l deaths; child l o s s a s the proportion of children who were born a l i v e but were dead at the time of survey. Table 7.3 presents the aggregate picture of f e r t i l i t y , f e t a l wastage, and cuild l o s s among women i n the Narangwal study area. Fetal wastage was probably underreported: 5.5 percent of dl pregnancies is much l e s s than the 10-15 percent usually reported. Lapses of memory probably explain t h i s and the decrease of reported f e t a l wastage with age: from 13.6 percent f o r women aged 15-19 t o 3.6 percent f o r women aged 45-49. The invariance of the child- loss r a t i o across age groups, around 20 percent, may mean t h a t the underreporting caused by problems of r e c a l l was more than o f f s e t by the rapid decrease i n child mortality. . L Religion dnd c a s t e The most important social- status variable i n rural Punjab Is religion- - caste. Eighty- four percent of the populatio$ were landowntng Jat- Sikhs (high - i caste) and scheduled Sikhs (among the lowest s a s t e s ) - . The remaining 16 percent !e of the population were non-Jat-Sikhs of high Saste, Hindus of a l l c a s t e s , and those of other religions. They represented a diverse range of socioeconomic levels. Table 7.3 k ! ~ nXO* Um NO. of Fecal of &a XaryIedVoorn XunXo.of P i t 3 LiVFBorn Child G- m. 2 . L i Y 8 4 f r t h . D a u b xat to.= at& The aggregate f e r t i l i t y of scheduled Sikhs was 4.34 l i v e b i r t h s ; t h a t of .?at-Sikhs 3.92 l i v e births. For women aged 15-24, the differences i n l i v e b i r t h s between scheduled Sikhs and Jat- Sikhs were l e s s marked than a t l a t e r ages. Presumably. the differences i n f e r t i l i t y between the two groups vere not due t o differences i n age a t u r r i a g e , because marriage is common for women aged 15-24. The completed f e r t i l i t y oE Jat- Sikhs w a s lower than t h a t of 8 scheduled Sikhs. The aggregate and completed f e r t i l i t y of others was i n between. Because the age d i s t r i b u t i o n of women i n the three caste groups differed, the mean number of l i v e b i r t h s was adjusted by using the age distiibu- t i o n of a l l women a s the standard: t o 3.74 f o r Jat- Sikhs, 4.65 f o r scheduled Sikhs, and 4.10 f o r others. Differences i n f e t a l wastage were small. The higher f e r t i l i t y of scheduled Sikhs i s balanced by higher mortality so t h a t t h e r e a r e equal numbers of surviving children. I n each age group the mean number of child deaths was higher f o r scheduled Sikhs than f o r Jat- Sikhs, v i t h others i n between ( t a b l e 7.4). Scheduled Sikhs l o s t twice a s many children a s Jat- Sikhs. Zducation of husband About 59 percent of the husbands had no schooling; 11.6 percent had 1-5 years; 25.6 percent had 6-10 years; 3.6 percent higher e d ~ a t f o n . .Asthe education of husbands increased, the f e r t i l i t y of married women i n almost a l l age groups declined. The mean number of l i v e b i r t h s f o r women whose husbands had no education was 4.7. - d i t h 1-5 years of educa- - tion t h i s came d u n t o 4.1 l i v e births, with 6-1% years of education t o 3.3, and w i t h more education to 3. The r e l a t i o n holds when adjustment is made for age differences, but the differences a r e narrower ( t a b l e 7.5). Table 7.4 MEAN NUMBER OF LIVE-BIRTHS PER MARRIED WOMAN BY AGE ANi)RKLIGION-CASTE - Relison-Caste Age Group Jat-Sikh Sch-Sikh Other -- - 15-19 0.75 0.84 0.57 Total- Adjusted far age-differences Table 7.5 MEAN NUMBER OF LIVE-BMTHS PER MARRIED WOMEN BY AGE AND EDUCATIONOF HUSRAND Education of Husband (Grade Passed) Total- Adjusted for age-difterences 4.34 4-15 3.81 -3. X I I For the purpose of age-adjustment mean number for age PO-44 was treated as t h e mean for ages 40 and over. ? - - 1 - t *- The l i n k between education of husband and f e t a l wastage is l e s s clear-cur:. For women over thirty- five, t h e r e was a problem i n recall, except f o r those whose husbands had higher education. Child l o s s , on the o t h e r hand, varied g r e a t l y with education. The r a t e of c h i l d l o s s I n the lowest education group was about tuo and a half times t h a t i n the highest educational group. 9ifferences i n c h i l d l o s s o f f s e t differences i n l i v e b i r t h s , s o t h a t the mean n ~ b e of survivfng children i n the d i f f e r e n t ?ducational groups rarige~from r 3.4 t o 3.0. The higher f e r t i l i t y of lower educational groups thus conpensated f o r the higher n o r t a l i t y o f t h e i r children. Occupation of husband I n the study v i l l a g e s , 41.4 percent of husbands were land-owning farmers, 25 percent ..orked as farm l a b o r e r s (which includes teriding c a t t l e ) , and 15 percent were i n s e r v i c e s , working f o r government o r i n p r i v a t e busi- ness. Farmers had high Income and some education; those i n s e r v i c e s had more education but l e s s income; laborers had low income and l i t t l e education. ihe fewest l i v e b i r t h s were reported f o r women whose husbands were i n services, perhaps indicating t h a t education is more importanr: than income i n affecting f e r t i l i t y . The highest f e r t i l i t y was among laborers; farmers and other groups were i n between. The highest f e t a l wastage was reported i n the s e r v i c e group, presum-- - a ~ l ybecause r e c a l l was b e t t e r i n the more educated group. Child l o s s was '2 - highest among laborers' families. AE with husband's education, groups- that - I h ~ dhigher c h i l d mortality had higher f e r t i l i t y , s o t h a t the surviving a m b e r ) of children was about the same i n the d i f f e r e n t groups. - Income As i n other surveys, it vas d i f f i c u l t t o g e t information on income: d a t a were obtained f o r only about half the families. The average nunber of l i v e b i r t h s f o r married women decreased with increasing incoae: fran 4.9 t o 4.1. The difference between the highest income group and the lowest inzome group was significant; other differsncea were not. This f i t s with other d a t a t that show income t o be less important as a c o r r e l a t e of f e r t i l i t y than educa- t i o n and some other s o c i a l variables. Fertility - related a t t i t u d e s of Narangwal wives For the oany variables that were simultaneously measured at Narangwal, a special e f f o r t vas made t o ascertain attitudes about family planning and beliefs about child survival. The general assumption has been t h a t a t t i t u d e surveys should provide measurable fodices of the notivational determinants underlying f e r t i i i t y . Considerable e f f o r t went i n t o measuring a t t i t u d e change through cross- sectional surveys of some 2,000 women i n study v i l l a g e s i n 1969 and again i n 1971. h f i n a l sarvey i n 1974 had j u s t s t a r t e d when the project ended. Preliminary analyses showed t h a t the variables which seemed most promising f o r detailed analysis we,re,ideal family s i z e , i d e a l number of sons 6nd daughters, and a d e s i r e fcr more children. Ideals of family size and composition were comeared with achievements and reproductive intentions. The .r - preferred family siz& of Narangwal women w a s measured i n responses to the - question: "How many ghildren (sons and daughters) should a family l i k e yclurs haye?" Bine ty percent of the respondents wanted between two and four children, and haif considered two sons and one daughter t o be i d e a l . Evidence of the importance of sons was exhibited i n s e v e r a l ways. C % For women aged 25-29, the number of l i v i n g daughters generally exceeded t h e i r f t s t a t e d i d e a l number of daughters ( t a b l e 7.6). For those aged 30-34, :he number of l i v i n g children exceeded the stated i d e a l family size. It waz nnct u n t i l woeen were 35-39 t h a t the average number of l i v i n g sons exceeded the s t a t e d ideal. Considering sons and daughters together, half of a l l women d i d not schieve t h e i r s t a t e d i d e a l number of sons daughters u n t i l they had four l i v i n g children. With f i v e l i v i n g children, four- fifths of t h e women had achieved a t least the i d e a l of both sexes. When views about i d e a l family composition changed, t h e r e was a s l i g h t tendency t o reduce the i d e a l nlnnber of sons and to increase the i d e a l number of daughters. Because two sons and one daughter v e r e t y p i c a l l y considered i d e a l , t h e r e vas a reaso~nable likelihood of achieving an excess of daughters i n r e l a t i o n to sons, thereby producing a r e v i s i o n it expectations. Analysis of data from t h e f i r s t a t t i t u d e and belief survey has suggested several influences on the individual's d e s i r e t o have children. I n p a r t i c u l a r , a ooman's age, number of l i v i n g children, nrrmber of ]Living sons, and number of sons r e l a t i v e t o umber of daughters appear to play a n important r o l e i n a wife's decision about the d e s i r e f o r additional children. e , I Wives' responses to the question: "Do you vant to have more chil- dren?" were contrasted with information o a the number of l i v i n g children .. - r e l a t i v e to the perceived i d e a l number. Thz results showed a poor c:orre- spondence between i d e a l family s i z e an3 rep?oductive intentions. * Tkere were always nore women who s a i d t h a t they wanted %o more children than those who had reached t h e i r i d e a l family s i z e o r i d e a l number of surviving sons and Tabla 7.6 CGMPARISON O F AVERAGE NUMBER O F SURVIVING CHILDREN WITH AVERAGE NUMaER OF CMLDREii DESIRED BY CONTINUOUSLY MARRIED WOMEN AT ORIGINAL AND RENTERVIEW BY SEX O F CHILD AND AGE COHORT O F WIFE NARANGWAL, INDIA First Interview (c.1969). Reinterview (c. 1971) Age at Sons Daughters Children Sons Daughters Children F i s t Interview 15-19 Achieved Desired 20-24 Achieved 0.83 0.71 1.51 1.24 1.00 2.24 Desired 1.94 1.09 3.00 1.99 1.19 3.15 25-29 Achieved Desired 30-34 Achieved Desired 35-39 Achieved Desired 40-44 Achieved Desired 45-49 Achieved 2.62 2.34 5.17 2.85 2.32 5.18 Czsired 2.18 1.30 3.48 2.14 1.26 3x36 ' J - daughters. Aggregate inconsistencies between i n t e n t i o n s and i d e a l family s i z e were g r e a t e s t a t p a r i t y three: only 10 percent of parity- three women had reached i d e a l family s i z e , y e t 7.6 percent of them wanted no more children. When the r e l a t i o n betveen i d e a l family s i z e and the d e s i r e t o have more children was examined by the sex of surviving children, i t was found t h a t 93 percent of the wmen who had two l i v i n g sons s a i d tt y wanted no more childreu. The number of l i v i n g daughters i n the family d i d not have nearly the same influence on the s t a t e d d e s i r e f o r a d d i t i o n a l children. Figure 7.1 shows the percentage of women wanting more c h i l d r e n i n r e l a t i o n to the number of l i v i n g sons and daughters i n the family at the time of the f i r s t interview. Women with no sons, regardless of the n'umber of daughters i n the family, wanted more children. About nine of ten families v i t h no sons wanted more children, regardless of the number of daughters. In contrast, only about one family i n twenty with two sons wanted more children, i f t h e r e was a t l e a s t one daughter. Depending on the number of l i v i n g daughters, the u t i l i t y of having additional b i r t h s dropped considerably when there was a t l e a s t one l i v i n g son. i For example, f i g u r e 7.1 shows t h a t women v i t h one son and several daughters k tended t o favor family l i m i t a t i o n s , sven though the i d e a l second son had not been born. In other words, the r i s k of y e t another daughter oucqeighed t tne uncertain prospect of having a desired son instead. In summary, a wife's decision about c h i l d r e n were r e l a t e d , t o t h e s i z e and sex composition of families. The most c o n s i s t e n t finding i n t h i s - a n a l y s i s i s the importance of the number of sons daughters i n the fer- I CiLity decisions of t h i s sample of r u r a l Punjabi wives. Of l e s s importance is the relation of the desired number and sex of surviving children to the achieved number. Figure 7.1 Percentage of Wives, Aged 15-39 Years, Wanting More Children by Number of Living Daughters and Sons, Narangwal, India ..Humber of Uving Daughters Hultivariate analysis A l i n e a r regression model vas used t o study the r e l a t i v e importance of socioeconomic, demographic, and a t t i t u d i n a l factors i n explaining variation i n the aggregate number of l i v e b i r t h s a t the beginning of the project. In this analysis, based on 1,721 women for whom there was information on income, there a r e eleven independent variables, with the cumulative number of l i v e births a s the dependent variable ( t a b l e 7.7). Since age was most important i n explaining variation, we subdivided the sample i n t o three broad categories, under 25, 25-34, and 35 and over. The variables explained only a small amount of the variation in l i v e births: 9.1 percent f o r those under 25, 1 8 percent 2 f o r those 25-34, and 12.5 percent f o r those 35 and over. The low il is compa- rable to that found i n other studies that do not consider physical factors, such a s duration of marriage and biological a b i l i t y t o reproduce, a s deter- minants of f e r t i l i t y . Other important variables a r e c h i l ~loss, membership i n Jat- Sikh and farming groups, and approval oz disapproval of family planning. 'IF2 positive association between f e r t i l i t y and child l o s s can be explained i n part by the general finding that mortality increases d i r e c t l y with f e r t i l i t y . Higher mortality may also lead to higher f e r t i l i t y by, say, shortening lacta- . I lu tional amenorrhea. >fembership i n the Jat-Sikh group w a s strongly associated with lower f e r t i l i t y . This is significant only a t higher ages, which implies -- t h a t t h i s caste group w i l l have '4 lower f e r t i l i t y r a t e a f t e r balancing child loss. The e a r i i e r association vfth occupation seemed to be washed out i n a b multivariate framework, perhaps b'kcause of multicollinearity. .. a o m * QI 0 I ? ? The tvr, variables f o r a t t i t u d e s toward family planning a r e p o s i t i v e l y associated with f e r t i l i t y , probably because both groups were compared v i t h the uncerta' A group, which consistently was the most negative. The r e s u l t s a l s o i n d i c a t e that approval a d disapproval of family planning vere both p o s i t i v e l y associated v i t h t h e depeudent variable. Disapproval led t o l e s s contracep- t i v e practice. I n the case o f approval the a s s o c i a t i o n may have r e f l e c t e d a response to f e r t i l i t y : vomen vere pore l i k e l y t o be motivated t o p r a c t i c e contraception once the? reached t h e i r desired number of children. Although c h i l d l o s s was a s i g n i f i c a n t v a r i a b l e i n almost a l l age groups, the b e l i e f about child survival was not. I n o t h e r words, once the e f f e c t of m o r t a l i t y was controlled, there uas l i t t l e additional e f f e c t from perceptions about mortality i n explaining v a r i a t i o n s i n aggregate f e r t i l i t y . Knen other v a r i a b l e s were controlled, awareness of modern contraception explained none of t h e v a r i a t i o n i n aggregate f e r t i l i t y . Education of husband, which had been a s i g n i f i c a n t c o r r e l a t e of fertility, c l s o was not s i g n i f i c a n t i n this regression. CHANGES I N FERTILITY DURING THE PROJECT F e r t i l i t y can be measured by tche number of l i v e b i r t h s , by the pregnancy r a t e & by closed and open b i r t h i n t e r v a l s , and by changes i n the probability of b i r t h during the program. This analysis w e d a dicho tornous variable- -whether a couple had a l i v ? b i r t h during a calendar year- as the dependent variable. - A dichotomous v a r l a b l e was a l s o used to indicate whether theBcouple had used . contraception i n the previous year. The variables included i n t h i s analysis a r e as follows: Dependent variable: Occurrence o r nonoccurrence of l i v e b i r t h s t o a couple i n a calendar year. Explanatory variables: (a) Use o r nonuse of family planning i n previous calendar year. (b) Time since l a s t l i v e b i r t h before beginning of c a l e ~ d a ryear ( o r since marriage). (c) Number of l i v e b i r t h s a t the end of the preceding calendar year (previous parity) . (d) Religion- caste (two dummy variables designating membership o r otherwise i n the Jat-Sikh and scheduied Sikh groups). (e) Education of husband (grade passed). (f) Occup~tionof husband (three dummy variables designating membership o r otherwise i n three groups: farming, labor, and services). Tie cases i n the analysis are currently married women aged 15-44 who ]Lived i n experimental v i l l a g e s u n t i l 1973 and for whom pregnancy h i s t o r i e s were available. The villages a r e those that offhred family planning serviceas: eight i n 1970, -leven i n 1971, and f i f t e e n i n 1972. We analyzed b i r t h s i n 1971, 1972, and - 1973 separately:' Corresponding to these births, we examined the pract:ice of i i contraception in- the preceding years, 1970, 1971, and 1972. - Age groups of women were i n t r q u c e d separately f o r 1971, 1972, and 1973. The numbers of uonen i n the analysis f o r the three years were 2,398, 2,330, 2nd ~ , 2 6 2 . The focus of t h i s a n a l y s i s is t o observe t h e e f f e c t of contracep- t i v e p r a c t i c e on t h e probability of b i r t h s during the project. Since t h i s ef f e e t is bound t o interact vit h socioeconomic, demographic, and b i o l o g i c a l v a r i a b l e s , w e have included those i n o u t analysis. The l i n k between f e r t i l i t y and family planning use helps i n understanding the e f f e c t of the project- an e f f e c t defined a s d i f f e r e c c e s i n b i r t h s t h a t can be ascribed t o the use o f contraceptivea from project services. To s e p a r a t e the e f f e c t of contracep- t i v e use from p r o j e c t s e r v i c e s , we regarded couples w i n g t r a d i t i o n a l and indigenous methods during the p r o j e c t as noncontraceptors. It is important, however, to emphasize the limitations of the main v a r i a b l e s chosen t o cramine the l i n k between f e r t i l i t y and family planning. Examining b i r t h s and contraceptive p r a c t i c e i n a calendar year is a r b i t r a r y and imprecise. It presumably diluted the relation between them. A b i r t h i n a calendar year, preceded by nine months of pregnancy t h a t may have s t a r t e d i n the preceding calendar year, vill not be d i r e c t l y influenced by the p r a c t i c e of family planning u n l e s s it was practiced continuously and e f f e c t i v e l y f o r the e n t i r e period of s u s c e p t i b i l i t y t o pregnancy. The contra- c e p t i v e use v a r i a b l e s i n t h i s dichotomous form may i n d i c a t e only a behavioral predispos&tion to family planning, not practice. Findings that the r e l a t i v e effectiveness o f protection provided by continuous use of s p e c i f i c methods o f - contraception were summarized i n chapter 3. '9 - I n general, b i r t h s among women who_$racticed family planning were less frequent than among those vho did riot ( 6 b l e 7.8). About 17 percent of t h e couples practiced family planning i n t h e f i r s t year, about 27 percent i n the second pear, and about 36 percent i n the t h i r d year. The proportions of Tabla -7.8 - -- USUS of PP w i t h BFrchSkr Soo-Uamtl of FP i.n Pnem eraoioru ~ a l ~Isr b ~ a h s l c ~ t ~skr x P-V~JU. so* as Vorwa So. of W m m n W c m d ~XO- of vith LB Ln J o eof w i t h La Ln Iaar U- Reamc I u r Percat$ Umau Present Year Percent women who had a b i r t h i n the three years were 11 percent, 14 percent, and 14.5 percent f o r those who practiced family p l a ~ i n gi n the preceding year, ca,apared with 23.5 percent, 25.1 percent, and 21.2 percent f o r those who did not. The [, The effectiveneee of contraception i n a v e r t i n g birth8 is i n f l u e n c d ' ii by the age of usrrs: recruitment to project contraception was much Easter L f o r women under twenty- five than f o r those over tventy- five, d e s p i t e the lover motivation f o r continuous use by younger women. For wosen under twenty- five, the proportion giving b i r t h who had ueed contraception during the preceding year was not s i g n i f i c a n t l y lover than t h a t f o r those who did not use zsntra- ception. Perhaps young couples used the nethods i n t e r m i t t e n t l y f o r spacing. For older couples, contraception was more l i k e l y t o have been c o n t i n u ~ u a and e f f e c t i v e , s o more b i r t h s were averted. Even i f t 3 e b i r t h performance of younger women was not dramatically influenced by cnntraceptive use, it presumably w a s important i n r e c r u i t i n g women who would l a t e r continue family planning more e f f e c t i v e l y . Previous p a r i t y t * F The probabilitp of giving b i r t h i n a year decreased with increas- % 8 ing p a r i t y , because of t h e p r a c t i c e of family planning and the impairment of fecundity. The proportion of women who practiced contraception during - k+:r !lit the previous year a l s o increased with p a r i t y . Excepk f o r the f i r s t group with @, * k;: a p a r i t y of zero to one, f e r t i l i t y w a s lower among contraceptors than noncon- $+$ t r x e p t o r s . There was a l s o a decline i n f e r t i l i t y with increasing p a r i t y f* dt L" ' --220 -- ma% both contraceptors and noncontraceptnr ( f i g u r e 7.2). The progressive reduction i n the rate of d e c l i u e i n f e r t i l i t y by p a r i t y among noncoatraceptors confirms the e f f e c t of o t h e r f a c t o r s on t h a t decline. These f a c t o r s include secondary s t e r i l i t y , separation of marriage p a r t n e r s , and nouproject aud tradi- t i o n a l methods of family planning. This d e c l i n e can be quantified rocghly by t determining the proportion of women - d t h open b i r t h i n t e r v a l e ,?rericer than f i v e years. This proportion increased rapidly up t o a p a r i t y o f t h r e e aod then more slowly, perhaps becruse of the use o f nonproject methods of contra- ception and o t h e r m-Pasures. -In t e r v a l s i n c e last b i r t h Another f a c t o r influencing a b i r t h i n a year is t h e time s i n c e t h e last birth. This is a n open interval measured up t o the beginning of the calendar year under consideration. If there was no e a r l i e r b i r t h , t h e i n t e ~ a wl a s meas-. .ad from t h e time of maxiage. Decreased f e r t i l i t y during l a c t a t i o n is ,,?->owed by a higher p r o b a b i l i t y of b i r t h a f t e r l a c t a - t i o n a l amenorrhea. :ressed duration is associated with lower p r o b a b i l i t y of b i r t h , s i n c e longer durations r e s u l t from secondary s t e r i l i t y o r t h e p r a c t i c e of family pl.anning. i f contraception is practiced t o space b i r t h s , open i n t e r v a l s longer than the desired spacing .wotuld be associated with a 0 higher p r o b a b i l i t y of b i r t h . Presumably, the t h e s i n c e the last b i r t h at f i r s t has a p o s i t i v e a s s o c i a t i o n M t h the p r o b a b i l i t y of b i r t h , but then '5 turns negative when t h e time is g;eater -- than desired spacing. The proportion of womenwho gave b i r t h i n a calendar year increased - up to an open i n t e r v a l ol' twenty- four months and then s t a r t e d t o decline?. Of p a r t i c u l a r i n t e r e s t are the year-by-year d e c l i n e s i n the proporti.on o f - -- - 221 - Figure 7.2 Psrcent Wcime-i Who Had Birth in a "ileadar Year by Previous Parity and Practice of Family Planning in Previous Yea- - 222 - women who had a b i r t h and the increases i n the proportion who practiced family planning. The proportion of women who had a b i r t h was considerably lower among contraceptive u s e r s than nonusers f o r all i n t e r v a l s . The d i f f e r e n c e s were greater, however, after an interval of about eighteen months than before. Up t o t h a t time, some women probably-were i n postpart- amenorrhea, and contraception had little e f f e c t on t h e i r b i r t h performscce. Only a f t e r the period of postpartum amenorrhea could the benefit of contraception be f u l l y demonstrated. a open i n t e r v a l of zero to six months would normally be covered by postpartum amenorrhea, except f o r Tero p a r i t y and f o r uomcn who did not l a c t a t e because t h e i r last baby died soon a f t e r birth. We presumed that the r a t e of family planning p r a c t i c e among women of zero p a r i t y and soon a f t e r the dea;.h of a baby would be negligible. We found, however, t h a t the r a t e was s u r p r i s i n g l y high during the open i n t e r v a l of zero to six months: 10 percent i n 1971, 22.1 percent 1, 1972, and 28 percent i n 1973. The demographic e f f e c t of contraception during amenorrhea would presumably be lower than a f t e r it. In general, these r e s u l t s confirm the impression chat family plan- ning was used mostly f o r terminating childbearing r a t h e r than f o r spacing. Among f & i i y planning users, the birth' i n t e r v a l curve was highest at 19-24 months i n 197i but declined ?o 13-18 months i n 1972 and 1973. Among nonusers the i n t e r v a l was 19-24 month; throughout the project. Information about the l a s t b i r t h i n t e r v a l observed during t h e proj- b e c t period provided s e v e r a l other i n t e r e s t i n g findings about spaclng'births during the project. The shortening of i n t e r v a l s kollowing a child detirh vas l e s s than anticipated from previous studies, only s l i g h t l y more than four months. Parity strongly influenced b i r t h intemals. Intervals were shortest a f t e r the f i r s c b i r t h whether the child survived (thirty- one months) o r died - (twenty- five months). The intervals increased up t o a parity of five. The shortening a f t e r c h i l d death was greatest a f t e r the f i r s t b i r t h ( s i x mont:hs) buc differences were progressively reduced to five months, three months, two months and one month a t parities up to five. t Child deaths presunably shorten the length of the interval by in- - creasing fecundity and by madifying desired f e r t i l i t y through varying dynamics 'epending on the age of the child that dies. Ths subsequent 'birth-tu-birth interval lengthens steadily a s the age a t death becomes higher. The e a r l i e r the infant death, however, the longer the median time before a new coccep- ticn. The low probability of conception i n the f i r s t twelve months a f t e r a l i v e birth i n t h i s .population is probably mainly attributable to lactatioa, but lactation seems to play l i t t l e o r no role a f t e r twelve months. Even though lactation averaged seventeen months i n t h i s population, we did not find any trace of a release of lactatL.~nalprotection on conception a f t e r deaths of children over one year. When children survived, the difference i n birth intervals between nonusers and e v e r u s e r s of family planning was . r more than four months during the project period. When children died, those differences were almost six months. - Child deaths produced a shortening of intervals of more t$n f i v e months for family planning users; 3.7 months i f o r non-~sers. The e n s i s t e n t differences : :he length of the birth interval b length are remarkable because they were f o ~~ t hwhen children survived ,? $"A * < and when they died, and also a t each parity. These were the couples, then, t h a t used family planning to stop childbearing rather Lhan to promote spacing. A t each p a r i t y , i n t e r v a l s were longer a f t e r a male b i r t h than a female b i r t h i f the c h i l d survived. The d i f f e r e n c e was small, however, aver- aging 2.6 months. No d i f f e r e n c e s i n i n t e r v a l were observed a f t e r a c h i l d death, regardless of whether the c h i l d vas male o r female (30.5 months i n both). During the p r o j e c t period, b i r t h i n t e r v a l s a f t e r . the deathe of male c h i l d r e n were about f i v e months s h o r t e r ; those a f t e r the deathe of female c h i l d r e n were only about three months shorter. The reason f o r t h i s was the d i f f e r e n c e i n i n t e r n a l 3 when children survived (males 35.6 months and females 33 months). Differences between c a s t e s were similar. If children survived, Jar,-Sikhs had b i r t h i n t e r v a l s 2.7 months longer than scheduled Sikhs (35.6 compared with 32.9 months). I f the c h i l d died, however, scheduled Sikhs had i n t e r v a l s 1.7 months longer than Jat-Sikhs. S o c i a l s t a t u s The e f f e c t of family planning p r a c t i c e would not be expected t o be the same i n all socioeconomic groups. Table 7.9 shows proportiolm of u s e r s and nonusers of contraceptives by c a s t e groups who had a b i r t h a year l a t e r . Among the religion- caste categories, the g r e a t e s t impact was i n the highest c a s t e group, the Jat- Sikhs, among whom contraceptors had s i g n i f r :. n t l y f e v e r b i r t h s than noncontraceptors. Scheduled Sikhs a l s o exhibitted the s i g n i f i c a n t e f f e c t of family planning on births- - except i n 1973, when the e f f e c t of the p r o j e c t should have been g r e a t e s t . =There were no s i g n i f i c a n t .E d i f f e r e n c e s i n b i r t h performance among contracept6rs i n the group of o t h e r s , - - . - presumably because f e r t i l i t y already w a s low f o r *ose not using contraception from project sources. Table 7.9 Parcmt Vcnnen Who BPd Birth - The e f f e c t of the husband's education on contraception and b i r t h s was examined by exploring differences between those v i t h l e s s than a sixth- k grade edrtcation and those v i t h a sixth- grade e d u c a t ~ oo~r more. Births among I users of family planning were significantly L:wer than among nonusers i n both education groups ( t a b l e 7.10). During 1971 and 1972 t h e e f f e c t of family planning on b i r t h s was greater f o r the less- educated group. It appears, however, t h a t t h e differences i n b i r t h s between users and nonusers of family planning narrowed f o r the less-educated group from one year t o the next, perhaps because of the inclusion of more young women i n l a t e r gears. When examined by occupation groups, the e f f e c t of family planning on b i r t h s is greatest among farmers (table 7-11., followed by those i n ser- vices. I n the other two groups the e f f e c t of family plauning on birth occur-. rence was significant i n only one of the three years under study. Multivariate analysis I n a m d t i v a r i a t e analysis, a dichotomous dependent variable indi- cating whether the woman had a birth during a calendar year was regressed on q several explanatory variables: previous parity, i n t e r n a l since l a s t birth, contraceptive use i n previous year, religion-c=i.?te, eduration of husband, and occupation of husband. Separate regressions were done f o r 1971, 1972, and , r a 1973 (table 7.12). Table 7.10 Lduatiun Ot P a t c a ~ tftommt Who Ead Bitrh ( G a Crlmuhr Jonilectr of F d y Vsem of P a u p Pused) Y u r -S Pkanins 2 P-VLtu . . a . 1: vdll8S h p.~STAthuu chr mrrrb.ZS of V- vhi~hQUmtqU uerm based. a t e 2: Sm-usars md us- of FP in above Table add u d y co 1903 tad 405 kr 1971, 164s md 595 in 1972, 1370 +nd 803 in 1973. w t a ~ ~ i m .was not haown amcmg nua-ruers and taser~for 77 md 13 in 1971, .Mtod 26 kr ign, 76 md 13 kr 1973. Table 7.11 Percent Y o w n Vho Bad Utt. Birth - Occtrpatfoo of caladu 8oo-0sars of F d y Usan of P d y EItub8.d Y u r p-8 p-g 32 P-VPlru Xota 1: Val- in porao&asas are -he d a r of w- QOvhich puccntages vr+. b u d . 'i ? L I Rbto 2: SQI-natr and unara of 7P Ih & w e Table add d y to 1926 and 412 in 1971, 1655 md 615 Fn 11672, 1394 and 809 Fn 1973. Oscupotioo w a s ooc k;unm -8 ootr-rum and users for 56 and 6 in 1971, 54 and 6 Fn 1972, 52 and 7 in 1973. . (Y m d I (Y a 4 d . I C1 m . 'a (el d 1 (Y m Previous parity w a s most significant i n explaining variation i n the probability of birth. Also highly significant was the finding that the longer the open interval since l a s t birth, the lower the probability of b i r t h during the year. For every additional ten months of interval, the probability of birth uas reduced by about 1 percent. Since the t h e since the l a s t b i r t h includes lactational amenorrhea, the effect of contraception may be partly captured by t h i s variable. Contraception i n the preceding year had a significant effect on births, reducing it 7 percent to 11 percent, depending on the year. This estimate of the effect of contraception is low for the following reasons. Anyone who practiced contraception i n a preceding year was couniod a s e con- traceptor irrespective of the length of use, vhich voulii d i l u t e the effect. Persons uho used contraception other than that provided by the project were regarded as nonusers. The methods varied greatly i n t h e i r effectiveness. The effect of s t e r i l i z a t i o n is especially reflected i n the interval variable. Even among those using family planning methods provided by the project, only the preceding year's contraceptors vere considered i n the model. Births i n a year would a l s o be i n f l u e x e d by contraception during the same year. In 1973, when the delivery of family planning services reached a peak, the. proportion of a women b-ho gave birth was only 19 percent, and the proportion of ccclples who used contraception i n the preceding year was nearly 29 percent. If none of the women had used family planning during the preceding year, the proportion giving birth i n 1973 ~rouldhave been nearly 12 percent higher (the marital f e r t i l i t y E -. r a t e would have been 22& per 1,000 women, not 19g). Relations were significant between parity and births i n a gear L and, except for 1972, between r e l i g i o w c a s t e and births. Being a Jat-Sikh was strongly associated with a lower probability of birth. The probabi1:Lty of giving b i r t h i n a year did not, however, seem to be associated v i t h the education of the husband. EFFECT OF SERVICES OIq FERTILITP When the Narangwal study was designed, it was assulled that integrated services would produce an increase i n contraceptive practice and a decline i n f e r t i l - i t y . Because the project did not continue long enough t o show sigaificant differences i n the e f f e c t of services on f e r t i l i t y , we have relied mostlyr on !# E contraceptive practice to measure that effect. In the preceding sections we shoved how contraceptive use, interacting v i t h other variables, affected b i r t h s i n the study villages. In t h i s section we trace f e r t i l i t y i n the experimental groups and examine changes i n f e r t i l i t y i n relation t o the use of services. Analysis by experimental group Analysis of f e r t i l i t y i n the experimental groups was useful f o r t confirming effects on the practice of family planning. But limitations i n -financial and l o g i s t i c a l resources made i t impossible to include a large enough - - .gopulation i n each experimental group t o obtain d e f i n i t i v e f e r t i l i t y infor- a a t i o n i n the short time we had for f i e l d observation. - The f a l l i b i l i t y OF h E annual f e r t i l i t y measurements, which fluctuate regardless of population slze, means that we should have observed f e r t i l i t y trends over a few more years. In our small population, we expected to have a standard error that permitted plus-or-minus fluctuations of two percentage points i n annual b i r t h r a t e s . There were six different sources of b i r t h data, but i n general we s t a r t e d with the two pregnancy-history surveys and then used other sources t o update and correct b i r t h reporting A b i r t h missed hy the f i r s t o r second pregnancy history, picked up by special v i t a l s t a t i s t i c s investigators o r throozh service records, vas added t o the pregnancy history of the mother. Service records varied somewhat from one experimental group t o another because of different patterns of coverage and varying intensity of services. We computed annual age- specific f e r t i l i t y r a t e s f o r the study pope- lation. For the experimental groups we calculated b i r t h rates and annual. marital f e r t i l i t y r a t e s f o r 1970, 1971, 1972, and 1973. We were surprised by the 5 percent average annual decline i n f e r t i l i t y i n villages with integrated ~ e r v i c e s ,a decline much f a s t e r than that of s l i g h t l y more than 1 percent a year i n Punjab. 11 But we lack conviction i n reporting these reaults because - of fluctuations i n annual rates and because f e r t i l i t y i n the control group also f e l l 3 percent a year. F e r t i l i t y of the study population. Peak f e r t i l i t y for the study population for 1970-73 was mostly i n the 25-29 age group. The youngest age group, 15-19, continued t o have similar f e r t i l i t y throughout the four years. The overall decline i n fertility vas small and occurred mainly i n age- specific f e r t i l i t y r a t e s f o r women over thirty. Our other data showed that these women . t 4 used contraceptives most frequently. This indicates t h a t Narangwal women used contraceptives more f o r limiting b i r t h s than f o r spacing. - - The pattern of fer@lity indicates an acceleration of the secular i I decline i n Punjabi f e r t i l i t y = c a l c u l a t u i a t s l i g h t l y more than 1 percent a h I -11 A s noted before, t h e decline i n f e r t i l i t y i n r u r a l Punjab and i n the Narangwal study area started long before the project: the b i r t h r a L = in Punjab f e l l from 45.5 i n 1959 t o 32.1 i n 1969; i n Ludhiana D i s t r i c t the f a l l was from 50 t o 34.6 during the same period. year during the 1960s. The marital f e r t i l i t y r a t e f o r all groups combiced dropped from 189 i n 1970 t o 172 i n 1973, f o r a d e c l i n e of 9 percent i n four years. F e r t i l i t y d a t a by experimental group. Annual marital f e r t t l i t y rates by experimental group f o r 1970-73 are i n table 7.13. The f l u c t u a t i o n i n f e r t i l i t y r n t e e from year t o year is zonsiderable, p a r t l y because-of the .mall i . sample s i z e f o r experimental groups. The l a r g e s t decline was i n FPWS v i l l a g e s 1 (21 percent), followed by FPWSCC v i l l a g e s (a5 percent). The decline i n FPED v i l l a g e s was 7 percent, t h a t i - the c o n t r o l v i l l a g e s 12 percent. The FPCC k 1 v i l l a g e s registered a rise of 2 percent. Thus two a f the groups receiving F i n t e g r a t e d services (FPWS and FHJSCC) had declines s i g n i f i c a n t l y g r e a t e r chan t h e c o n t r o l group. The crude b i r t h rates r e f l e c t the same patterns ( t a b l e 7.14). The f e r t i l i t y data disaggregated by experimental groups do not permit f i r n inferences about trends i n annual f e r t i l i t y rates o r about the e f f e c t of services on these rates. As already mentioned, the period was too s h o r t to observe a s e c u l a r trend o r t o c a l c u l a t e three- year moving averages t o c o r r e c t fluctuations. I n addition, the b i r t h s i n a year, on which the f e r t i l i t y r a t e s were based, were few i n number, and f l u c t u a t i o n s were ac-en- . t tuated because the birthas were ,o d i f f e r e n t groups of women. There a l s o t was variation i n the quality of birth data gathered i n the various groups: - - v i l l a g e s tha: had the m Q s t extensive services had the leasg cnderreporting. i @ - Family planning use produced its g r e s t e s t e f f e c t s n FpWS v i l l a g e s : F b d i f f e r e n c e s were highly s i g n i f i c a n t i n all three years (tab3e 7.15). Reduc- $ T t i o n s i n b i r t h s were also s i g n i f i c a n t i n EPWSCC v i l l a g e s , but the e f f e c t was J r t l e s s than i n FPWS v i l l a g e s . These r e s u l t s e s t a b l i s h t h e l i n k between the F + use of services and the reductions i n f e r t i l i t y through contraceptive use. + f $ !d. Table 7 . U mu*'3atLtal Fartfllty -6er Sy a&xi3lantal Groups and vu.Lsgas 1970-1973 ibnuka nunbar of livebirtbdfram a.U sourees) p&r 1000 martied --en (Lrm census 3 rounds, w i t h U.ar extraplation Setxeen =cnmds> ~ X S C C 36.7 34.4 3z.e as. o * AVeraga ntrmber of Livebirths per 1000 & N a t i o n INumer-.t6rrfm g study sources pooled. Dekdnators fran CI'.MUS Rounds L (19681, 2 (1971 excepr =Ed in 19721, 2 (19'13)with lime extrapolations for years 1973 and 1977; . s Table 7.15 CaOrJo r- . -a Plrrming XL P-Val". No. mSa. Table 7.13 showed t h a t t h s f e r t i l i t y decline was highest i n FPWS v i l l a g e s (23 percent) folloved by FPWSCC v i l l a g e s (21 percent). The r e s u l t s indicate t h a t such declines were achieved by contraceptive use from program sources. I n FPCC and FPED villages, the reductions i n b i r t h s were not significant, partly because of smaller numbers of contraceptive csers. Analysis of individuals t Because of limitations i n the group data, we a l s o focused on the individual data. Such an analysis requires specification of the use and outcome of services f o r individuals. This analysts was confined to the e f f e c t of services on f e r t i l i t y , primarily measured through changes i n the f e r t i l i t y of individual Komen a s related to the use of service inputs. 11 - The exgecta- t i o n that outcomes may be significant seems reasonable because ve have shown t h a t b i r t h r a t e s f e l l about 20 percent i n some v i l l a g e s during the time that Br the number of women using family planning methods increased significantly. Measuring reductions i n f e r t i l i t y requires, f i r s t , selecting a ref- -- < erence mark, a lwel of f e r t i l i t y from which the reduction is t o be estimated; and second, r e l a t i n g the reduction to a period during vhich the intement:ions were operating. These measures should distinguish changes i n f e r t i l i t y due t o and independent of project efforts. L I I The change i n f e r t i l i t y from p::eproject t o project levels was defined a s follows: by using the information on duration of marriage (Dl) and - - '2 i -- 1. This analysis is similar to that of a produeion- function relationship i n econometrics, but does not use an input-outputLframevork because of the problem of disentangling cause-and-effect sequencing i n the use of services and the decline of f e r t i l i t y , a s is explained l a t e r i n the section. the t o t a l number of l i v e births ( B ~ UP t o the end of 1969, the average - ) annual probability of b i r t h (PI S1/D1) was computed f o r each woman to indicate preoroject f e r t i l i t y . Similarly, by using the duration of the project (D2) and t o t a l l i v e births (B2) during the project up t o the time the women withdrew from project exposure (outmigration, dissolution of marriage, o r end of project, whichever happened f i r s t ) , another average - annual probability of b i r t h (P2 B2/32) was computed. The absolute - i k decline (PL P2) i n the probability of b i r t h uas then expressed a s a ;r b t percentage of the i n i t i a l level (PI). The percentage decline i n ferti1it:y was LOO(P1 - P2)/P1- 1/ - A* )r The analysis included only women who had demonstrated preprojeet f e r t i l i t y o r women who had a t least one l i v e b i r t h a t the beginning of t h e -I project- and only wmen from the four groups of villages (FPWSCC, FPWS, FPCC, and FPED) that received family planning services up t o 1974. Measurement of preproject f e r t i l i t y of these women w a s based e n t i r e l y on pregnancy h i s t o r i e s obtained i n the cross- sectional survey at the beginning of the project. These - h i s t o r i e s were updated up t o the end of the project by using longitudinal sources of b i r t h data. Information was available on 2,298 women, though the exact duration of exposure to the project was not known for forty-two of them. L Socioeconomic data were available f o r only 1,818 women, so the multivariate .* 1. This measure has limitations. -Yarriage duration a s a measure af exposura to sexual union is not homogenous; a duration o f , say, f i v e years uhen a rwomap - is twenty is different frSm the same duration when she is thlrty-five or fort 4i In addition, D2 always r e t e r s to a period when the mman is older, and con- sequently when ( f o r most Qomen) her fecundity is lower. Thus :he amount and direction of change of individual f e r t i l i t y measured by t h i s formula is a s much a function of age a s of the use of family planning. We therefore dis- aggregated the sample by age i n the multivariate analysis. analysis was performed on that number. These data show t h a t f e r t i l i t y can be effected not only by d i r e c t nonproject inputs, but a l s o by introducing varia- tion i n the effectiveness of project inputs ( f o r example, education helps - i n the effective use of services) . I Health and family planning services were originally classified i n * nine categories, but for t h i s analysis ue reduced 'the classifications to s i x c \ broad categories: women's i l l n e s s care, women's other care, child care for prevention, child care for cure, general and motivational family planning, and family planning services vith follov-up. Maternity services f o r women were dropped from the analysis. In t h i s categorization, the type of personnel who delivered services was not taken i n t o account--in each category, services offered by a l l personnel were grouped together. The dependent variable w a s the decline i n the probability of b i r t h during the project. Of the 2,298 cases examined, about 18 percent had an increase i n 'he probability of b i r t h , and about 4 percent had no change. But 78 percent of women i n the sample had a decline, though low. More than half had a decline i n probability of l e s s than 20 percent; about a f i f t h more than 20 percent. A s would be expected, age is significantly related to the dependent variable. The proportion of women who did not exhibit a decline i n f e q t i l i t y decreased with age because these wnen had already completed child bearing. The s h i f t is shown by the wmen who had an increase or no change i n f e r t i l i t y : - 44.3 percent of those under 25, 28 percent of those aged 25-34, and only 6.2 . percent of those 35 and over. Conversely, the proporjion of women who expe- rienced a moderate (up to 20 percent) deciine i n f e r t i l i t y i ~ c r e a s e dwith age: - 240 - 20.9 percent of those under 25, 58.3 percent of those aged 25-34, and 92 percent of those i n 35 and over. High declines i n f e r t i l i t y (more than 20 percent) decreased with age: 34.8 percent of vomer under 25, 33.7 percent of those aged 25-:?.4, and 1.7 percent of those 35 and over. Variation i n the decline of f e r t i l i t y was most for women under 25, l e a s t f o r those 35 and over, presumably because the f e r t i l i t y of the older women was already lov when the study started. Of socioeconomic variables t5at influenced changes i n f e r t i l i t y , 1 family income was considered f i r s t because it can a c t through several links. b i* i F i r s t , private inputs supplemented project inputs influencing f e r t i l i t y and 5 I mortality. Second, income is an indicator of living standards and is presumably /j related to modernizing influences on members of the family. Higher family income should indicate better access to information about family planning and greater awareness of the desirability and f e a s i b i l i t y of regulating f e r t i l i t y . The proportion of women who experienced either an increase o r no change i n f e r t i l i t y decreased with income, but a greater proportion of women i n higher income groups had declines i n f e r t i l i t y . The association is significant. Caste is considered the most important indicator of social s t a t u s i n Punjab. For Jat- Sikhs the proportion of women who experienced an increase . t i n f e r t i l i t y was lowest and the proportion who experienced reductions was highest. The findings were reversed for scheduled Sikhs. - - The others were :Ln between. '5 - THe husband's education was associated differently with declines I Q I i n f e r t i l i t y . The better educated group (sixtn grade o r hizher) had an equal d i s t r i b u t i o n of those whose f e r t i l i t y increased and those whose f e r t i l i t y declined more than 20 percent during the program. A t both these extrtaes the proportion of women was greater i n the better educated group. The leus educated group had a higher proportion v i t h a moderate (up t o 20 percent) decline i n f e r t i l i t y . These findizigs did not indicate a c l e a r associii- t i o n of declines i n f e r t i l i t y v i t h education. Similarly, there seemed t o > be no association with occupation. Thus f a r the analysis has nhovn r e l a t i o n s between the dependent variable (the percentage decline i n the average probability of b i r t h per year of exposure t o the project) and such conditioning variables a s age and socio- economic s t a t u s (income, caste, and so on). To evaluate the e f f e c t s of service inputs on the dependent variable, it is important t o control f o r these variables. A multiple regression model was therefore used t o p a r t i a l out the e f f e c t s of the s e m i c e inputs on the dependent variable ( t a b l e 7.16). With the per- centage decline i n the average probability of b i r t h per year of exposure a s the dependent variable, t h e r e s u l t s show t h a t the e f f e c t s of the six tze-vice inputs, when controlling f o r various socioeconomic variables, a r e not i n the direction expected. Family planning motivation services a r e significantly but negatively associated with a decline i n f e r t i l i t y . Family planning -emices L with follow-up have a consistent and positive ('bu; not significant) associa- ' t i o n with a decline i n f e r t i l i t y . Preventive s e m i c e s f o r children a r e signifi- - . cant, but the association is nega.ave. Among t h e conditioning variables, being i of high c a s t e (Jat- Sikh) and b e i n e i n farming were significant. h The negarive a s s o c i a t i o ~ v i t hsome of the service inputs may be caused by t h e f a c t t h a t the use of some services- -such a s health sem:tces f o r women and curative services for children- are related t o births, which Table 7.16 UCRESSIUH RESU1.TS Dependent Variable8 Porcer,tage Decline i n Averafia Annual Probability of Birth . A l l A ~ e e <25 25-34 3% t Lgreeeion t Pofireaeion t Pegreesion t r u 1 Ibfireesion Vurlablo Coofficiant value Coafficient value Coefficient value Cocltticiant value IJomcn I11 (~1~1,) - ,0095 .3 ,0601 .4 - .0339 8 - - Women Other (WO'l'll) - .0228 @ 3859 1.7 04247 2. 3 l3 .0363 1 4 C h l l d Other (COTII)- ,0289 3.6 - .0323 1.4 - .0232 1.9 -.0213 3.1 Ctrlld 111 (CILL) - ,0022 0 - .0351 . . , -4 00052 0 Jat-Siklr 5.3998 6.2 14.7814 2.2 4.9112 1.5 - - Sch-Sikh -1.6055 .9 -5.1730 .9 - 07339 o3 -05438 .7 Education - ,3181 - - 2.U .7706 1.5 1576 l6 9.1027 1.0 Incomo ,0002 1.5 - .1)002 l 3 l0004 1 l6 .002 207 Farming -5.0605 2.6 -11.5980 1.2 -7.3550 2.2 a4857 .5 - Scrvico 7442 .3 3.5560 .5 -4.4415 1.3 .4967 .G generate many of service contacts. Use of sezvices then represents a response I I t o need p::c]l.ced by b i r t h s rather than measuring a f a c t o r that could contribute to a decline i n f e r t i l i t y . The negative association between a decline i n I f e r t i l i t y and male family planoing motivation be the r e s u l t of intensive e f f o r t s by motivators t o influence high f e r t i l i t y couples. This analysis did I not permit a c l e a r definition of cause and e f f e c t between the use of services and the decline i n f e r t i l i t y . So, conclusions about the impact of services a r e tentative, though suggestive, i n explaining the dynamics of the interactions. Most women experienced a mild decline i n f e r t i l i t y , which i s con- s i s t e n t with the trend. I n a l l instances, the l e a s t use of services was related t o a moderate decline i n f e r t i l i t y ( t a b l e 7.17). Those who had no delcine, o r even a negative decline, were heavy users of services, a s were those who had high declines i n f e r r i l i t y . These r e s u l t s suggest t h a t some heavy users were families to whlch services were rendered i n response t o health needs generated by high f e r t i l i t y . The most relevant finding is t h a t the remaining group of heavy users of services had a substantial decline i n f e r t i l i t y (16 percent and more), and these couples may be those f o r whom the f e r t i l i t y impact of services is beginning to be evident. The r e l a t i o n between service inputs and the dependent variable ( the percent,ibe decline i n the I average probability of b i r t h per year of exposure t o the project) is nonlinear. This nonlinearity reduces the value of the regression r e s u l t s i n table! 7.16, and further analysis would require e f f o r t s t o separate service use associated with an increase i n f e r t i l i t y from tha; associated with a decline. E I Tabla 7.17 - Pexcsut Haan -at of Contacts - D.c.Lh. ia Avuage &nu W m ' s W a s Qljldren's Childrenas TP RP PwbabiUtp rr 0th- Xllnus O t h u Xotitratioa Sarviea h of Birth S.rPie'u SaNFccq Sarpicu S e e - S e e - Foilo-M? (m'Xli) (CILL) (COPI) (FPHX) (F?JER) The suggestion that the use of Narangwal services =as partly a response to specific health needs arising from a pregnancy o r b i r t h is indi- cated by figure 7.3, which shows the relations between mean numbers of service contacts and the decline i n f e r t i l i t y , a s measured by changes i n the average probability of birth. The f i r s t segments of the graphs show t h a t high use for some women was associated with no decline i n f e r t i l i t y , o r even a rise. In these cases services were presumably provided i n response to births o r pregnan- ! ties. In other cases field-workers, such as male motivation workers, d g h t have worked more intensively on high parity cases. Fr: women whose use l of services i n response to high f e r t i l i t y did not show a decline i n f e r t i l i t y , i an effect might have been seen i f the program had been extended. The l a s t segments of the graphs probably r e f l e c t the effect alf services on f e r t i l i t y , b-ith a heavy use of services being associated with a substantial decline i n f e r t i l i t y (16 percent and more). The low points i n the middle parts of the six graphs represent the cases of l e a s t s e n i c e use. This s l i g h t decline i n f e r t i l i t y ( l e s s than 20 percent) was found t o be consistent with the trend among women i n Narangwal. There is no clear indication that a significant decline i n f e r t i l i t y was related to reachling a thresh014 of some miniam volume of services. . * SUMMARY -- 'g - F e r t i l i t y patterns before the project show hat higher caste groups had .E lower f e r t i l i t y than lower caste groups. - higher f e r t i l i t y of the low- caste group was, however, balanced by higher mortality, so that there a r e Figure 7.3 Illlacior-Jup of of tuzzlirf 2oclir.a and Ckildren'a 0a.c S ~ t w l c a a / - krcanc 3eelI.n. in tertility - a I " lo * 20 JO 4; equal numbers of sumivjng children. I n almost all age groups, the f e r t i l i t y of marrizd women declined a s the education of t h e i r husbands increased. Of the three major occupations i n the study villages- fanners, services, znd laborers-- laborers had the highest f e r t i l i t y , those i n services had the lowest, and f a m e r s and other groups vere i n betveen. The decline i n aggregate f e r t i l i t y was shared by a l l the experhen- tal groups. It vas largely due to declines in age-speciflc fertility a t the beginning (under twenty years) and tovard the ~ n d(more than 'hirty years) of the childbearing ages. Caste and parity differences vere found significant i n b i r t h inter- vals, which became shorter a f t e r a child death. The use af contraception i n averting births was generally effective, but since the delivery of familly pianning services l i d not reach a peak u n t i l 1973, t h e period of observation vas too short t o describe the e n t i r e effect of f d l y planning on b i r t h occurrences i n experimental groups. Previous parity vas most signfficant i n explaining variation in t h e probability of b i r t h i n a given yea:- Also highly significant vas the finding that the longer the internal since the l a s t b i r t h , the lover th~e probability of b i r t h during the year. L - What is the effect of s e r v i c ~packages on f e r t i l i t y ? The largest decline was i n FPWS villages (21 percent), followed by FPWSCC villages (15 percent). The decline i n FPED villages was 7 percent, that i n the t control group 12 percent. The group receiving-integrated service-. thus had a substantially larger decline than control groups. Individu~l-level analysis (rather than village- level) is moie relevant t o evaluate t h e impact of serpices. The use of servlces by indivi- duals can be a response t o a need produced by a birth, o r it can be a contribut- iq flctor to a decline in fertility. I n all instances, the h a s t use of serpices was related t o a moderate deciine i n f e r t i l i t y . Those who had .13 a.?cline, o r even a r i s e , were heavy users cf servlces, a s were those who had declines i n f e r t i l i t y . This implies t h a t some heavy users wcre families t o t which services were rendered i n response t o health r led8 genlerated by high i f e r t i l i t y . The r a i n i n g group of heavy users of services had a substaztia!. decline i n f e r t i l i t y . Chapter 8 - Robert. L. Parker and R.S.S. Sarma Efficiency and i-qui t y of S e n i c e s Some of the most d i f f i c u l t issues that policywakers and planners a f health and family planning ~ e r v i ~ must resolve are those of balancing tradeof f s betveea e s . cost, equity, and effectlvcnesa. Gccause of limitec cesaurces, there ~muetbe e f f o r t s t o maximize the effect o: services through increases i n effici~ency. Atte*.~ptric;t equitablle distribution, houo,ver, may d i l u t e services so much that chey h,~ve Little ef'Pect. A t tLranjptaL. vc found t h a t focusing services on the group most neglected o r a t '.i.g&~estr i s k (see appcodix B) bdlancetl the distribu- tion of services a d incrcnscd the efficiency (and effect) of services. In t h i s chapter we suppJrt the case for integrating health arid family planning services by ~ r ~ s a n t i descriptions of the inputs of ti.me hnd ~ l g cost and calculatic~nsof efficiency expressed as time req-Are3 per service t contact, cost pi contact, and tne cost-affect~veness of dffferent I :;~TC;LC 5 t service ~ m b i ~ i i c : The coaeiderat .oas of a f f i c i e :cy a r e rhea b a l ~ d c e dby ~ ~ . 1e an examinstion of pr.- ject achievements iti prrmoting equity ic the distribu- g t i o n of health senrkces znd In the use sf f a d l y plz6cing- In addiiion t o e ! ccnfirming the previ9ucly d~crurnentedeffect of services on family planning acd health, the &r:a here demonstrate the eqr-;tyr and efficiency of integrated Is f services it' the h'arangual pieject. - 3sed Sample ebscr-vationti of work were t o estimate the amount a£ time project k 1*-:.'f spent i n c a r i ~ i n gout the various ccmponents oi services i n each of 6 L L t h e four experlmec~:al groups of villages: osc receiting family planning, C ": gomen's s e r ~ i c e s ,;and child cars (FFWSCC); these, receiving family plannJ.>g A and vomen"s servic12s (PTYS); those receiving f a d l y plannicg 2nd child care & (PPCC); and those receiving f d l y planning and family planning education (FPET,). I n the original experimental desigr decision was to have equiv- alent personnel-to-population ratios, even though the i o t k load differed from group to group. Some variations i n the size of villages and i n stafftog made. i t necessary to standardize the results of the work sampling as time inputs per 1,000 population per veek. Figurz 8.1 show f o r each of three experimental groups of villages the time spent on direct scrvices a t the midpoint of the project i n 1971. (Services did not s t a r t i n PPED villages u n t i l 1972.) The t o t a l d i r e c t service time :hat a l l project s t a f f provLded during an average week varied from 700 minutes per 1,COO people i n FPW9 vil.lagee t o almost 1,100 ninutes per 1,000 i n FF'CC villages. Tne FPPWSCC villages had a staff input of about 860 minutes per 1,000. The graate: inputs i n FPCC villages were related t o the higher staff- to- population ratios called f o r 'uy rhe research a c t i v i t i e s of the nutrition project- The intensive services f o r child care consumed twice as much staff time (673 minutes) a s similar services i n PPWSCC villages (335 d n u t e s j . Although the content of services wap s w l a r , the larger volsme of child care i n EPCC villages was associated with much more frequent surveil- I lance contacts t o collect idformation oa r.utritiona1 s t a t u s axd infection (Taylor and others forthcodng). Althougn the design called f o r ao child care -- i n PPWS villages o r services for L 1 1 women i n FPCC villages, itgproved neces- L snry t o spend a l.'ttle - time i n such a c t i v i t i e s t o ~ . i n t a i nrappgrt with b vLllage fandtlies, i n cases of emergency, or when these types of .care could not . be separated from other services. Examples of the tt.ird requirtaent inc11:de Figure 8.1 - ;\vsraqe spsnt Pmvfdiag D L z t Se.~ceuby A l l Staff - working in Each Group 1971 .Yi=nrtes per Weak per 1000 PcpuLatfon essential care of the newborn i n the immediate postpartum period i n FPWS villages, o r care of a wornan during pregnancy i n FPCC villages a s part of prenatal care of the child. Unlike these differences i n the design of child care, vomen"s semices i n FPWSCC and PPWS villages were planned t o be identical i n content and frequency of routine contacts. I n practice, t h e s t a f f i n FPWS villages spent about 25 percent more time on women's services (397 minutes) than the s t a f f i n FPWSCC villages (321 minutes). The t h e spent i n care of ill women was almost the same, suggesting t h a t demand f o r c l i n i c a l services produced a comparable response even though time constraints differed. The main dif- ferences between the two experimental groups were i n routine surveillance and -* maternity care. As noted i n chapter 4, t h i s difference resulted from the increased frequency of contacts and, a s vill be noted l a t e r i n t h i s chapter, from the gxcatet time spent per contact i n FPWS villages. The difference also suggests that fonL7.y health workers tended to use d i r e c t service time more efficiently i n FYWSCC villages because child care and women's services could be co~binedi n one home v i s i t . Care of women required l e s s time i n FPCC villages t h a ~i n ?PWSCC villages (77 minutes compared with 103). That care was devoted t o surveillance t o identify pregnant women early i n pregnancy L , * and t o provide them with "prenatal child care," focusing primarily on services needed t o produce a healthy baby. -- '5 - Family planning services during an average week took up 97 minutes -- i per 1,003 people i n FPCC villages, 81 minutes i n FPWS villages, and * .44 rninutes - i n FPWSCC villages. This difference helps t o explain how combined services achieved the greater efficiency shown by the cost- effectiveness calculations l a t e r i n t h i s chapter. Finally, time spent on contacts whose purpose was mainly to maintain good relations v i t h individuals and communites was higher i n FPWS villages (200 minutes) and FPCC villages (194 minutes) than i n FPWSCC villages (137 minutes). This presumably r e f l e c t s the additional time needed o r available t o maintaJt rapport i n villages that did not receive complete Seventy-six percent of d i r e c t service time i n study v i l l a g e s was r-ovided by the family health worker residing i n each village (averaging 'ho.: one family health worker per 1,600 people i n FPWSCC and FPWS vil- 1,ges and one 5amily health worker per 1,000 i n FPCC villages). Abcut 70-75 percent of the uorker's a c t i v i t y obsened during work sampling was clasuified as service, the r e s t as research. These findings mean that i f Narangwal services were to be replicated i n other similar locations and i f no time were a l h c a t e d to research, the worker-to-population r a t i o s could be be.tween 1:2,000 and 1:3,000. I n addition to resident family health wcrkers, other s t a f f were involved i n d l r e c t services i n each village on regularly scheduled v i s i t s . The proportions of d i r e c t service time provided by these s t a f f were 13 percent by male family planning workers, 6 percent by physicians, and 4 percent by family health supervisors and pzblic health nurses. A second work sampling, i n '1973-74, provided more data oh changes i n the distribution of direct service t h e , including information about the FPED villages, uherk services started in mid-1972. I n those villag~es 'S i family planning educator= rather than family health uorkers were used a s - pe resident workers (one 1,400 people). Because of a smaller sample size i n 1973-74, data were sufficient o u y t o describe the d i r e c t service time of family health workers and family planning educators (figure 8.2). In general, the d i r e c t service time of family health workers followed a pattern of Figure 8.2 Avarage Tiam Spent Providing D h e c t Sentices 3y Family Efealth *&rksr?r fa ,FPwsCC, m,FPCC V U a g u , and Family PLanning Educators ur (21. -FPfd Villaqes - 1974 .xinutea par week o u 1000 POD. d i s t r i b u t i o n similar to all s t a f f i n 1971. The important finding w a s t h a t the average time put i n by family h e a l t h workers i n 1973-74 v a s only 70 percent of t h a t i n 1971. Much of t h i s reduction was i n c h i l d care, which was progressively streamlined and simplified i n FPWSCC and FPCC v i l l a g e s . The t b e family h e a l t h workers spent i n FPWS v i l l a g e s did not change much: almost 500 minutes per wezk per 1,000 people on d i r e c t s e r v i c e s i n FPCC v i l l a g e s , 440 minutes i n FPWSCC v i l l a g e s , and about 415 minutes i n FWS v i l l a g e s . The d i r e c t s e r v i c e time of family planning educators was 400 minutes per week per 1,000, c l o s e to the time of family health vorkers i n FPWSCC and FPWS v i l l a g e s . I n 1973-74, a s i n 1971, c h i l d care ( e s p e c i a l l y f o r ill children) vas higher i n FPCC v i l l a g e s than i n FmSCC v i l l a g e s (254 minutes compared with 92 minutes); care of w e l l women, including materdity c a r e , w a s higher i n FPWS v i l l a g e s than i n FPWSCC v i l l a g e s (130 minutes compared with 54 minutes). That some a c t i v i t i e s were c l a s s i f i e d a s children's and women's services i n FPED v i l l a g e s (37 minutes) indicated t h a t family planning educators spent some time discussing these topics during t h e i r home v i s i t s . No care vas provided, but advice was given about what they could do o r m e r e they could go f o r care, usually t o the nearest government health center. A s might be expected, f a n i l y planning educators provided much more time s p e c i f i c a l l y f o r family planning (249 minutes) than did any of the family h e a l t h workers c i n t h e i r v i l l a g e s (83 minutes i n FPWS v i l l a g e s ; 57 minutes i n FPCC v i l l a g e s ; 37 minutes i n FFWSCC v i l l a g e s ) . Finally, the t i m e spent maintaining f r i e n d l y r e l a t i o n s was a l s o ~ u c hhigher i n FPED v i l l a g e s (113 minutes) than i n the P other experimental groups (43 t o 49 minutes). The " friendly" t i n e a n t h e health s e r v i c e v i l l a g e s may i n d i c a t e that rapport had s t a b i l i z e d and t h a t family health workers were accepted, whereas family planning educators had been a c t i v e for l e s s t i m e and had l e s s to o f f e r . The findings from t h e 1973-74 work sampling confirmed t h a t workers providing a l i m i t e d range o f services spent much more t i m e o n those services than workers who had t o provide s i m i l a r services i n combination with o t h e r services. This may seem obvious, but it was important t o i d e n t i f y these q u a n t i t a t i v e differences between experimental groups t o determine whether v a r i a t i o n s i n amounts of a service were associated with d i f f e r e n c e s i n outcomes. TIME PER SERVICE CONTACT The comparison of s e r v i c e c o n t a c t s i n chapter 4 showed t h a t t h e combined services i n FPidSCC v i l l a g e s generated more contacts than those i n FPCC v i l l a g e s , even though s t a f f spent more time on c o n t a c t s i n FPCC v i l l a g e s . I n addition, people i n FPWSCC v i l l a g e s had almost t w i c e as many service c o n t a c t s a s people i n FPWS v i l l a g e s i n 1971 and about a t h i r d more in 1973. The d i f f e r e n t p a t t e r n s of t i m e i n p u t s and s e r v i c e contacts can be compared by c a l c u l a t i n g t h e average t i m e p e r contact. In 1971 t h e average t i m e spent: per contact i n FPWSCC v i l l a g e s was 5.4 minutes. This f i g u r e combines a l l s e r v i c e s i n the home and c l i n i c i n t o one average. The corresponding averagss were 6.9 minutes i n FPWS v i l l a g e s and 7.6 minutes i n FPCC v i l l a g e s . The greater efficiency o f the combined s e r v i c s s is indicared by t h e averages f o r c h i l d care: 4.6 minutes per contact i n FPWSCC v i l l a g e s , where they were combined with women's s e r v i c e s , but 6.7 minutes i n FPCC v i l l a g e s . Another f a c t o r contributing t o t h e d i f f e r e n c e s i n c h i l d c a r e s e r v i c e s between FFWSCC and FPCC - '5 v i l l a g e s w s s the g r e a t e r use sf home v i s i t i n g , which required more t i m e per - contact i n FPCC v i l l a g e s . L I n 1973-74 t h e contacts i n FPWSCC v i l l a g e s were more e f f i c i e n t i n the use of d i r e c t service time (4.4 minutes) than i n other v i l l a g e s . I n FPED v i l l a g e s each contact took more than 14 minutes, almost t h r e e times t h a t :n the other experime. ' I groups. When time per family planning contact was calculated separately, the averages were found to be about 4 minutes i n FPWSCC villages, 8 minutes i n FPWS villages, 19 minutes i n FPCC villages, and 16 minutes i n FPED villages. Why were the contacts i n FPCC villages l e a s t e f f i c i e n t ? One explana- tion is t h a t family placning services were started several years a f t e r child care services had been well established, thus requiring a big readjustrrent i n the work patterns ~f family health workers. Just a s important, i f not more so, child care i n these villages was supervised by a male pediatrician, family planning by a female physician. As a result, family planning workers had d i f f i c u l t y integrating the two a c t t v i t i e s i n t h e i r thinking and i n t h e i r work. From our experience, child care services w i l l always be somewhat more d i f f i c u l t to integrate with family planning serivces than with women's services. But when family planning was combined k i t h both wome?'s and children's services, family health workers made much more e f f i c i e n t use of t h e i r work t i m e . COST OF SERVICES Detailed accounting of expenditures i n the Narangwal p p u l a t i o n project permitted calculation of cost related to specific functions, a c t i v i t i e s , and * I services. Work 6ampling and other data were used t o estimate the distribu- tion of s a l a r i e s and other expenditures f o r the different experimental groups - and types of services. The f i e l d s t a f f established.2riteria to define the i - project a c t i v i t i e s that were classified a s research- In estimating cost t h i s - h Eresearch component was excluded so that findings wodd r e f l e c t the service I cost of each component of the service packages. The c o s t a n a l y s i s was a s comprehensive a s possible i n i d e n t i f y i n g the components. For example, a l l donated drugs, food, buildings, and land were given a n estimated value i n current market prices. Capital expenditure f o r bu'.ldings `led large equipment was amortized at annual r a t e s of 2 percent and of 5 percent; t h a t f o r small equipment and v e h i c l e s at 10 percent. The components of average annual c o s t i n each experimental group are listed i n table 8.1. The share of the c o s t r e l a t e d t a persannel- - the s e r v i c e component of salaries- - ranged from 45 percent i n FPWSCC v i l l a g e s t o 71 percent i n the FPED v i l l a g e s . FPWS and FPCC v i l l a g e s were i n between at 63 percent and 51 percent. I f the c o s t of the n u t r i t i o n supplementation prcgram (food, supplies, s a l a r i e s , and so on) is subtracted from a l l c o s t s , the pro- portion a t t r i b u t a b l e to s a l a r i e s r i s e s to about 54 percent i n FPWSCC v i l l a g e s and 65 percent i n FPCC v i l l a g e s . On the o t h e r hand, the amount s p r n t on the s e r v i c e compocent of s a l a r i e s was l e a s t i n FPED v i l l a g e s ($823 o r Rs 6,151), 5 most i n FPCC v i l l a g e s ($1,446 o r Rs 10,842,, and i n between i n FPWSCC v i l l a g e s ($995 o r R s 7,465) and FPWS v i l l a g e s ($1,132 o r R s 8,491). ! Three other d i f f e r e n c e s werc major: the l a r g a proportion of the cost of food i n child c a r e v i l l a g e s (18-20 percent), the g r e a t e r c o s t of supplies and drugs i n the c h i l d c a r e v i l l a g e s , and the marked v a r i a t i o n i n the cos.: of r e f e r r a l . TIe r e f e r r a l c o s t i n FPWSCC v i l l a g e s was almost -the same as , the combined referral costs i n FPWS and FPCC v k l ~ ~ ~ eFor. the other recur- s r i n g c o s t of consumable material, e s p e c i a l l y f o r d m g s and supplies, the -- efficiency of combined s e r v i c e s i n FPWSCC v i l l a g e s is evident: they a r e ' ?- lower i n FPWSCC v i l l a g e s than_ i n FPWS and FPCC v i l l a g e s combined. - 259 - Table 8.1 DISTRIBUTION OF COSTS BY SPECIFIC CATEGORIES OF EXPENDITURES I N EACH EXPERIMENTAL CROUP BASED ON AVERAGE ANNUAL COSTS PER 1000 POPULATION* (Rupees) Experimental Group Cost Categories FPWSCC FPWS FPCC FPEd Buildings - ** 2% Maintenance Equipment - 5 o r 10% ** Supplies Vehicles -10% ** Vehicle Running and Haintenance Food Drugs Referrals 1 Salaries TOTAL R s 16700 Rs 13500 R s 21300 Rs 8700 * E 1970-73 f o r FPWSSCC, FPWS, FPCC (1971-73 f o r FPCG, family planning component); 1972-73 f o r FPEd. ** Capital expenditures amortized a s indicated t o provide annual costs. (Column percentages i n parentheses.) Note: US$1 = RS 7.5- - 260 - The amortized c a p i t a l cost was only 3.5 t o 6 percent of the t o t a l c o s t i n any of the groups and i n rupees was almost the same. This shovs t h e need t o plan f o r the proportionately g r e a t e r d r a i n on resources r e l a t e d t o r e c u r r i n g c o s t s wheu developing primary c a r e s e r v i c e s and t h e importance of not becoming overcommit t e d i n c a p i t a l investnents . For example, drugs are important i n making programs of primary h e a l t h c a r e e f f e c t i v e . I n the experi- a e n t a l groups with h e a l t h c a r e s e r v i c e s , t h e proportion of expenditure f o r drugs ranged from 10 t o 15 percent. The annual drug c o s t per c a p i t a i n the FPWSCC v i l l a g e s was about $0.33 (Rs 2.5). The s h a ~ eof the c o s t of transport i n t o t a l c o s t ranged from about 7 percent t o 11 percent. The major d i f f e r - ence i n a comparison with t h e d i s t r i b u t i o n of expenditure i n government h e a l t h c e n t e r s is t h a t s a l a r i e s consume about 75 percent of a l l funds i n government s e r v i c e s , proportionately reducing t h e amount t o spend on drugs, supplies, and transport. These findings show the g r e a t need f o r reversing p a t t e r n s of expendi- ture. Rather than spend s o much on buildings, it would be much more p r a c t i c a l t o provide adequate drugs, supplies, and transport. I n any attempt t o plan s e r v i c e s , it is important t o know what proportion of t h e cost is fixed (and w i l l produce a constant demand on resources) and what proportion w i l l vary depending on use of the s e r v i c e s . Costs t h a t w e c o n s i d ~ r e dt o be fixed i n t h e t s h o r t run were primarily c a p i t a l c o s t s , s a l a r i e s , and p a r t of the building and vehicle maingenance. - Fixed c o s t s averaged about 53 percent i n FPWSCC v i l l a g e s , 58 percent iH FPCC v i l l a g e s , 73 percent i n FPWS v i l l a g e s , and 83 percent i n - - FPED v i l l a s e S . These findings show t h a t servi.ces i n t h e l a s t two experimental !e groups would be much more prone t o changes i n e f f i c i e n c y depending on v a r i a t i o n s i n use. T a b h 8.2 gives t h e average annual cost per c a p i t a f o r each type of service i n each experimel~talgroup. Chiidren's services c o s t $0.56 per capita i n FPWSCC v i l l a g e s and $1.07 i n FPCC villages. This difference is another indication of t h e differeace i n intensity of surveillance i n the two experimental groups. Nutritional supplementation, primarily f o r children aged one t o three, a l s o cost more i n FPCC vJ.llagcs ($0.87 per capita) than i n FPWSCC v i l l a g e s ($0.60 per capita). The cost of women's services other than maternity care was $0.60 i n FPWSCC v i l l a g e s a r 3 $0.88 i n FPWS villages; t h a t of maternity care was $0.23 and $0.43 f o r these two experimental groups. The cost per capita of limited services ro women i n FPCC v i l l a g e s was much less. For both women's services and child case, t h e costs per capita i n FPWSCC v i l l a g e s were consistently lower than i n villages receiving e i t h e r service lone-indicating the greater efficiency of combined services. Efficiency vas a l s o increased f o r f d l y planning, which showed a dramatic progression of cost per capita from $0.24 f o r PPWSZC villages, $0.44 f o r FPWS villages, $8.67 f o r FPCC villages, aad $1.08 f o r FPED viilages. The cost per capita of each service package, obtained by adding t h e costs of component services, was $1.16 f o r FPED villages, $1.80 f o r EPWS villages, $2.23 f o r FPWSCC villages, and $2.84 f o r FPCC villages. The estimated expendi- r t u r e per capita was between $0.27 and $0.40 for~selrvicest h a t government primary health centers provided i n Punjab i n 1968-69 (Johns Hopkins 1976). But it was estimated t h a t i f Nara wal s t a f f had received t h e same lower 3 fringe benefits of the govermnent personnel, i f medicine aad supplies had beec -- i bought from government stores, andhP,f a major share of the food f o r nutrition - supplementation had been donated by outside sources o r collected from the villages, the per capita cost f o r the FPWSCC services could have been reduced - Table S.2 91- AlamAL COST PER CXPm FQR DZFFEZZerP s&RvxeES m EA(B WtlP ~RCIPEES) Child rare mtrition hy about h d f '19 $1.07 (Rs 8.0), which is still about four times the cost t n sslancing were ~oor1.yintegrated b e c ~ u ~cfe the overlapping aceivit leo o f the nuLrition proj1.2t for t t ~ group. * 'The cost i n FFdSCC v l l l a g e s apprcached the 1965-69 cost pcr family planning contz--.': - ? - 264- Table 8.3 - Child Cua 1.4 - 1.7 NutritFon* - W e n ' s Services 2.2 2.7 2.2 Family Planning 3.8 5 . d 14.4 10.9 *&st pcr ru$ple!aeatal fading i n primary h e a l t h centers ($0.33 o r Rs 2.5). The cost per supplemental feeding was $0.03 (Ks 0.25) i n TPWSCC villages and $0.04 (Rs 0.32) i n FPCC villages. J u s t a s with c o s t per c a p i t a , FPWSCC villages. c o n s i s t e n t l y had t h e lowest c o s t per u n i t of services- - again indicating the g r e a t e r efficiency of com- bined services. Expenditure on nonproject s e r v i c e s Despite government and project expenditures on health c a r e services, t h e v i l l a g e r s still spent a l a r g e amount of money on health care. Information on such OLL-of-pocket opending was c o l l e c t e d dui-ing the sample household survey and combined with data on project and government expenditure t o etsti- mate t o t a l expenditure on health c a r e per c a p i r a i n p r o j e c t v i l l a g e s , includ- ing control areas. The findings i n t a b l e 3.4 show t h a t provision of p r o j e c t services permitted a reduction ia the amount. of money spent o n h e a l t h c a r e by t h e v i l l a g e r s a d by t h e government. Nonproject h e a l t h c a r e expenditure was $1.96 (Rs 14.7) i n FPWSCC v i l l a g e s , which had :he most project services, compared with $2.96 (Rs 22.2) i n c o ~ : r o l v i l l a g e s and $3.51 (Rs 26.3) i n FPED v i l l a g e s , which had no ~ r o j e c thealth care services. When the secortd t u b amo~t-tsa r e compared with the expenditure per c a p i t a ( p r o j e c t acd nonr project) ir, FPWZCC v i l l a g e s , it is c l e a r thaz introducing the comprehensive package of women's and children's ~ e r v i c e s( e x c l d i n g n u t r i t i o n ) i n Punjab v i l l a g e s cost only s l i g h t l y more than t h e q i s t i n g p a t t e r n of government and '5 -- p r i v a t z ' services- $3-59 (Rs 26.9) per capit% a year, compared with $2.96 (Rs .. 22.2) and $3.51 (Rs 26.3). In c o n t r a s t , tad expeniituze on each of the other s e r v i c e packages--FFWS, FPCC, and FPED--st.ow t hSJ to he somewhat more ey-pensiv-3 alternatives. Table 8.4 ESTlXM"l'DANNUAL PER CAPITA EXPENDITURE ON HEALTH BY SOLJRCE OF CARE FOR EACH EXPER?XENTAT,GROUP AND CONTROL VILLAGES (RUPEES) FPWSCC FPWS FPCC FTEd CONTROL Covermnent and Other Services Private Practitioners 13.2 15.7 16.5 22.9 15.9 Subtotal 14.7 18.6 18.0 26.3 22,2 - Project Services* 12.2 i3.5 14.8 8.7 I-I.- TOTAL 26.9 32.1 32.8 35.0 22.2 . f * Excluding nutrition supplementation. C I Note: Cost for government, ocher services, and private practitioners were $. estimated using data from the furctional analysis project and the g 1973-74 sample house:lold sitrvey (Yellow Book). ;{ Note: GS$1 = Rs 7.5. COBt-ef f ectiveness Estimates of cost- effectiveness were made from the data previously summarized oa outcome and cost. The average family planning cost. per new acceptor of a modern family planning nethod during the second year of services i n FPWSCC v i l l a g e s w a s half t h a t ir. FPUS villages and a third that i n FPCC and FPED villsges. Next, i f i t is ar.sumed that the average number of coup~les practicing family planning duriq; the second year of services represents a f u l l year of use by that many couples ( a couple-year of use), the cost of using family planning was $10.27 (Rs 77) per couple i n FPWSCC villages. The * cost per couple-year of use was 1.7 times more i n FPWS villages, 2.4 times more i n FPCC villages, and 2.9 times more i n FPED villages. The r a t i o s confirm the much greater efficiency cf family planning services i n the FPUSCC villages. I f a l l health and family planning costs a r e i n the cost- effective- ness calculations, the picture changes considezably--with FPED services l e *st costly, because of there being no associated health costs. This compar:Lson is only valid i f one is no^ concerned about an output other than family planning. T.3 measure of health outcome used for the cost- effectiveness . calculations was deaths of chiidren under three. This was considered a proxy f o r effect cf child care. Average death r a t e s f o r children of t h i s ?ge i n h FPWSCC and FPCC t i l l a g e s were conpared with similar Fates i n the cor.tro1 villages to estimate the number of child deaths probably prevented by project s e n ; es - . The cost of child care, including nutritional supplementation, was then related to t h i s outcome measure to give the cast- effectiveness .E # 8 $$ P2 V E ratios. The t o t a l cost attributable to each death averted was $749 (Rs 5,616) i n FFWSCC villages and a l i t t l e more i n FPCC villages. The closeness of the cost- effectiveness r a t i o s f o r these two experimental groups was a r k surprise because services i n FPWSCC villages had l i t t l e effect on deaths of infants (under a year) but those i n FPCC v i l l a g e s had a much greater effect. A t these levels of services, i t would seem that reductions i n infant mortality a r e directly related to the increased input, mainly a difference i n the frequency of surveillance. A more careful analysis that used data from the nutrition study to identify the costs of services that most likely produced such outcomes a s reducing morbidity, improving growth, and preventing death. These costs could be determined by allocating part of the costs of n u t r i t i o n and health care t o the prevention of deaths according to the proportion of children dying i n control villages (age- specific mortality rates). The r e s t of the health care costs were then attributed to rcducing morbidity, the r e s t of the nutrition costs t o improving physical growth. For FPCC villages, which were also a part of the nutrition study, the costs per death averted a r e much I--er than i n the previous analysis because only a part of costs was counted as contriLuting to prevented deaths. Prevention of perinhtal deaths required the l e a s t ($9.87--Rs 74-per death averted); prevention of infant deaths cost four times t h i s amount; prevention - of qiaths of children aged one to three cost ten times t h i s amount. ?c It i . appesrs that as age- specific mortality r a t e s declined over the f i r s t three - I y e a 3 of l i f e , the cost of averting death increased. The cost per day of of i l l n e s s averted was f a i r l y similar for i n f a n t s and f o r children aged one to three--53 (Rs 4) and $0.40 (Rs 3). Nutrition costs per additicnal centimeter of growth by age three were just under $27 (Rs 200) per child. I n summary, these findings of cost- effectiveness support conclusione about the efficiency of the integrated services i n FPWSCC villages. EQUITY I N THE USE OF PROJECT HEALTH SERVICES Before the start of project health services, surveys were conducted to collect data on the use of government and private sources of care kn s i x villages to be included i n the population and nutrition studies. These surveys demonstrated a significant d i r e c t relation between income and the use of health services. Lower income people used health services l e s s than higher income people (Johns Hopkins 1976, p. 134). Project services were designed, however, for a l l people i n the study villages, with special concern for those with the greatest need (see appendix B). It was therefore important to find out whether we overcame preexisting d i s p a r i t i e s and improved equity i n the use of project services. Curative services Table 8.5 presents the coverage of,wopen and children by services B f o r i l l n e s s care throughout the project. These services required a v i s i t t o the fanily health worker's village c l i n i c and were mainly i n i t i a t e d by patients. - '..- Use of services was categorized'in three ways: no use during the project, use early (1969-71) o r l a t e (194-73) i n the project, and use throughout the (.. n n n n n m *m woa dm I - - w r - * m e project. The families were characterized by socioeconomic measures, includ- ing caste, income, ownership of land, education of husband or father, and the possession of radios, bicycles. o r seving machines. The project services attracted a l l types of women and children a l s o s t equally. Women from land- owning families and those whose husbands were b e t t e r educated used women's i l l n e s s services significantly more than other wouen. Other than that, there were lo significant differences i n use between different socioeconomic groups. But there was a small, consistent trend for wmen of higher socio- economic s t a t u s and children of lower socioeconmic status (except f o r educa- tion) t o rise services more. Cross- sectional household surveys measuring the use of different sources of care mentioned i n chapter 4 also confirmed the equitable distribu- t i o n of project curative services and the removal of d i s p a r i t i e s between women and children of high caste and those of low caste. The data from these surveys showed that 38 percent of illnesses of high-caste children were treated by project personnel, compared with 39 percent of low-caste ci-iildren. b o n g women the proportion of illnesses receiving project services was 28 percent f o r high- caste families and 29 percent f o r low-caste families. Although project services were equally used by d i f f e r e n t castes, the use of b outside sources of services by children i n nonservice villages was similar by caste group: 35 percent for high castes, 34 percent f o r low. 3ut i n project villages, high-caste families (20 percent) continued to use outside sources of services m ~ c hmore than low-caste famiLles (7 per cent), producing a use r a t e E significantly higher for high-caste cHildren (56 percent) than f o r low-caste L children (45 percent). Thus project services i n many cases tended to supple- ment outside sources of care f o r high-caste children, but: they substituted for most outside sources of care for l o r c a s t e children. Tho pattern of displacement of other sources of care by project services among high-caste women and lo-, -caste mmen was similar to that observed i n their children. Ln nonservice villages, however, high-caste women used outside sources of services significantly more (34 percent) than did low-caste women (26 per- cent). The same was true f o r men who had access only to outside sources of services i n a l l the project villages. The data therefore indicate that the project delivered services equitably, but that the demand for continueti use of other sources of care varied significantly by socioeconomic level. Other services Most services other than curative services were initiated by the project. They include prevention and surveillance, usually i n the home. Except f o r differences by caste, the proportion of women using these ser- vices varied slightly but not significantly by socioeconomic classification (table 8.6). Generally, 10 percent or fewer never used the services. In addition, a greater proportion of low-caste women used them, unlike curative services, which higher socioeconomic groups used more. The opposite was true f o r educational level: women whose husbands were better educated used other services more than women whose husbands were less educated. Use of children's other services must be interpreted with care because the services were prima- I $ 5 r i l y designed f o r children under three. Because thz two time intervals i n g .? k -- the project were short, it was d i f f i c u l t to f i t observation interval9 with the & * ? - periods i n which children were e l i g i b l e to receive services. So thewnumbers $ b did not match precisely for children who passed their third birthday during P - Table 8.6 PERCENT OP PAHILIES WITll VARIGUS GOCIOECOHOHIG CIIAMCTERISTICS USING OTllER GERVICES FOR w a w OR alrmReN DURING nje owr;cr (1969-1973) Voaen88 Servicse Chlldren8s Barvicee no . Boepa Pull 140 6-e Pull N "usr 'use ('~ea N h e '066 'UBI Caetttr P a w ' 477 7.1 16.0 , 76.9 1 476 16.0 21.6 62.4 1 Income; ,a r t 6IRd 1894 450 11.1 16.2 72.7 ) 357 171 . 15.5 57.4 ) . ~ducntlon, 0-5 years 979 9.0 17.7 73,3 1 i27 6 yaara 334 8.4 14.2 77.4 ) ""' 438 Noter SomO.Uas - Ueo in 1969-71 or 1972-73 - Pull U6r Uar i n 1969-71 1972-73 -- Chl-aquare taatac u... not rrignlticant - 6' ~ 4 . 0 5 ** "' PC.01 P(.OOl Table 8.7 Coupariaon of Pra-Progrm Urie (Hodern Tuuporary and P e m n o n t Hothoda) and Progrnm Use Rates m n g Socio-Demographic Croupa #der -ProgramUsere Pro Proyram Users (Hodern,Temprary, G Percent Parma~~entHetliOd) Percoat Religion-CaatoL Jat-Sikh Sch-S tk11 Others Education of Iiuoband~ 0 1-5 6-10 11t 0ccupat i o d Iklhua%&"d' Par~uiny Labor Service Otl~er Aga o f Wife8 (25 25-34 ?5t NO. of Livir~gChildren; (3 890 105 11.8 369 41.5 3-5 1047 227 21.5 536 51.2 6+ 388 94 24.2 x2 = 42.25 193 49.7 x2 = 19.50 (2) (2) . -- P = (.om p = (.001 - - - - I . the project o r were born i n the second helf of the project. It bra shown i n chapter 4 t h a t i n a given year more than 90 percent of the children under three received other setvices. But because t h i s table includes some families with children over three, the coverage of a l l familie8 by these services at any time during the project is about 80 percent. &spite these lfmications, it is apparent that Eaailies from lower socioeconomic groups benefited most from the project's preventive and surveillance servlc ~ fso r children. The exception, ss with curative services, is again related t:, the father's educa- tion: b e t t e r educaced hcuseholds used these services nost. Althcugh some of these differences can be explained by the higher number of young children i n lower socioeconomic households, the main explauat i o n is t h a t these targeted services identified high- risk children and concen~ratedservices on them. Family planning services Differences i n preproject r a t e s of ever-use of modem cont~aceptives vere significant among religion- caste groups, arredg education groups, :nong occupation grmps, among age groups, and among fcnily-size groups (table 8.7). The general pattern is t h a t higher caste, higher education, higher occupati,n, and larger family size groups had greater r a t e s of contraceptjve use. The L group aged 25-34 had the highest rate, the group under twenty- five the lowest. The differences i n r a t e s of contraceptive use during the project - - were l e s s significant than the differences i n preproject use. The exceptiot- was the age of wife: there was a reversal of the difference between the = ? b youngest (under twentyafive) and the oldest (over thirty- five). The reversal, vhich was demographically $ore advantageous, was so great t h a t the age dl:£ er- ence b e c a e even more significant than the preproject difference. Couples reca-ited by the project were f r m thrre prior-lse cate- gorier.: s s e r s of traditional xethods, user; of modern nathods, and nonusers. It was show. in chapter 3 that there uera no socioeconomic dir'ferencos i n acceptors recruited from preproject users of traditional s e t h ~ d sand of modern methods. More knportant than e l b i n a t i q g differences i n the recruitment of prior users was aarrowing diffeyencee aong new r e c r u i t s r The sociodcaographic c h r a c t e r i s t i c s of nev recruite Here d i s t i n c t l y different from the preproject users mainly because of considerable success i n eq~ulizil?gt h e i r distribution to r e f l e c t general comnunlty characterf s t i c s . This is indicated by comparing preproJect differeuces i n table 8.7 with those of new r e c r u i t s i n -.able 8.8. Table 8.8 shows that overall differznees a r e not zignificant, wl.ich is i n marked contrast t o the great differences (p < .001) moag preproject users of modern methods. It sep-ms extremely imporcant for equity that the previous p a t t r r n of distribution tas been reversed artd the low-caste group had a recruitment rate considerably greazer thafi Jld the high-caste group (x2 5.0, p < .025). Dirfereoces by :he occupation of (1 >- th-. husbad, higMy s..gnificant for preproject users, were eliminated, Education of husbad maintained the same pattern of distribution a s that f o r preproject users, with higher education being associated with higher family L p l a n i n g r a t e s but a t cot7iderably l o ~ r e rlevels of significance. In contraat t o preproject users, who mainly were older woeen, the project w a s able i. get fresh acceptors a t a considerabfy higher race from donp, women under thirty- five. Only about 23 percent of the women over thirty- five iecame users 02 family B planning f o r the f i r s t tine; but aboxt 26 percent of-women under thirty- five Table 8.8 Socio-Demogrsphic Differentiale i n Recruitment into Program o f Previoue rn . Norr-Usere of Yemily Planning Socbo-Demographic No. of Non-Usere 3rcent Groups . No. of Non-Users Wlm Became Program Fresh Before Proqram usere Recruit. Religion-Caste r Jat-Sikh 626 203 32.4 - Sch-Sikh 486 189 38.9 Othors 191 68 35.6 x2 5.01 (2) ( a9 4 PO* If Occupation o f Husband; Fannlncj Labor Service Other Age of Wife: - < 25 25-34 35+ Lfving Children; < 3 551 3-5 435 6+ 1 6 1 , did. The relation betveen preproject use and number of living children was significant. By contrast, fresh recruitment ranged £?om 33.5 t o 38.4 percent, but these differences were not significant. SUMMARY Integrating services proved e f f i c i e n t from several points of view. F i r s t , there was substantial saving i n vorkers' time when services were combined. For integrated women's and children's services there k-as a saving of between , :b. 20 and 34 percent i n time spent on family planning services when compared rzith women's or children's services provided alone v i t h family planning; a 35 percent saving i n time for wrnen's other services when provided with child care; and a 24 pe:,ent saving i n time for children's other services when provided with women's services. Second, integrated services were cost- effective. The cost per :s& '-Fkn..+ new family planning acceptor was $12.27 (Rs 92) i n FPWSCC villages, tvice :r '- t h a t amount i n FPWS villages, and three times that amount i n FPCC and FPED villages. E-e cost per year of family planning up was $10.27 (Rs 77) i n FPWSCC villages, 1.7 times that ;aount i n FPWS villages, 2.5 times i n FPCC villages, and times i n FPSD villages. When a l l health costsawere a t t r i - buted to reduceti mortality, improved morbidity, and better nutrition, numerous cost advantages were observed i n integrated services. - An important achievement of the Narangwal project was that it . reduced preexisting C i s p a r i t i e s and improved squity i n the use of project sen-ices. By providing integrated services t o those i n greatest need, d i s p a r i t i e s i n the use of health and family planning services by d i f f e r e n t castes, occupation groups, and educational levels were reduced. For example, 39 percent of the nonusers i n t::r lowest castes became users of modern family planning methods during the project; the corresponding figures f o r other castes and f o r J a t Sikhs a r e 36 percent and 32 percent. Before the project, use of modern methods of family planning vas primarily a high- caste activity. In health, the uee of curative services vas essentially equalized f o r the different socioeconomic groups. It appears that the provision of curative services by the Narangwal project made it possible f o r the low-caste group t o s h i f t t o project services, while the high-caste group continued t o use private services, but the disparity i n t o t a l use of the services by caste was reduced. Project- initiated preventive and surveillance a c t i v i t i e s promoted equity because allocation vas based on need. For children's other services, differences i n use vere shifted i n favor of low-caste children, low-income families, landless families, and those with the fewest material possessions. The coverage by women's other services was similar. Method of Data Collection CROSS-SECTIONAL SURVEYS Census and socioec~nomicd a t a Demographic and socioeconomic d a t a were c q l l e c t e d i n three censuses of the study v i l l a g e s i n 1967-68, 1971-72, and 1973 and i n two socioeconomic household surveys at the t h e of the f i r s t two censuses. k i l l a g e s i n the family planning. education group (RED), were not , however, included i n the f i r s t round of e i t h e r survey. A team of from six t o ten intentiewers was responsible f o r these two types of survey. Young men and women with master's o r bachelor's degrees were recruited as interviewers and given intensive training i n interviewing I techniques and i n the use of the intervieid schedulesl. They were supervised by one of the s o c i a l s c i e n t i s t s . Censuses. The f i r s t census was c a r r i e d out a f t e r the study v i l l a g e s were c a r e f u l l y mapped and households numbered. h r i b g the census interviews, a l l families i n each household were i d e n t i f i e d and a l l resident members recorded. To be considered a redident, t h e member hhd to haGe been l i v i n g i n the v i l l a g e s i x months o r have expressed the intentiOn to s t a y of six months o r more. ~pecial.'care was taken to ask about manber4 who were temporarily .r L away but were norm-ally r e s i d e n t s of the v i l l a g e . Syqtematic probin,g t o * determine t h e a g e 5 f each individual was standardize4 by using v i l l a g e events calendars containing Indian months and a l l major national and l o c a l h i s t o r i c a l events. Reinterviews and cross- checks with o t h e r d a t a sources showed t h a t t h i s method of determining age was q u i t e r e l i a b l e . Other information c o l l e c t e d a t the time of the censuses included marital status, education, occupation, and dates of migration, marriage, and b i r t h s o r deaths occurring between censuses. Socioeconomic surveys. Social, economic, and other household data were collected i n a l l households. Items i n the survey included caste, religion, housing, sanitation, selected possessions (including ;.and) , nodand income (such 8s that from labor, service, and tvade) , and land (agricultural) income (available only on the second survey). When possible, socioeconomic intervie48 were with the senior man I of the houaehold, because of the need f c r accurqcy i n getting d e t a i l s of the type of crops, yield, and market prices used i n estimating agricultural income. Attitudes and beliefs Attitudes toward family planning, specYfic contraceptive methods, survival of children, and family size a t the beginning of the project prcvided an important pare of the baseline data collected, Change was measured by repetition of t h i s cross- sectioral &mey a f t e r two years of services. I n FPWSCC villages a third round was carricd out a t the end of the project. Because of the inclusion of F?ED villages midway \through the projecr, only one round of cross- sectional surveys wasiavailable for t h i s experimental group. Development of survey forms. The vide4y recog2ized d i f f i c u l t y of - measuring attitudes and beliefs led 0s to adopt a cautious approach. q e r e - I 6 were no standard procedures that could be applied to the questions that-we - I w We were especially conscious of the delicate nature a f wriy of !e wanted to ask. I important issues and of the possibility of icadvertently antagonizing v i l l a g e people, such a s by questions related t o the death of a son. I n addition t o long-standing cultural blocks, ve realized the rieed f o r special s e n e i t j v i t y t o recent developments, such as the backlash i n Punjab from rumors about the hazards of the IUD. This led t o a sequential development of survey forms by our s t a f f of Indian social s c i e n t i s t s . The f i r s t stage was conceptualizing the qucs- tions t o be asked. This vas folloved by an esseptially anthrnpol.ogical expioration of the issues through discussions with a limited number of respondents. We then developed a structured f o m a t t h a t was revised i n several pretest f i e l d t r i a l s . The form was most+y prestructured with only a few i t e n s permitting open-ended conditional responses related to reasons f o r particular opinions. A number of modifications +ere a l s o made before the second round of surveys because of problems and gaps encountered i n the f i r s t . survey . Data collection team. The major respoqsibility f o r conducting thIs part of the research was assumed by the Indian social s c i e n t i s t s on our s t a f f . Six to eight f i e l d investigators were selected tarough nevspaper advertisements and interviews. They were a l l young wonen and recent master's graduates i n one of the social science fields--such a s socioljgy, anthropology: geography, o r social work. A l l vere Punjabis, although on19 a few had come from a village. Their f i e l d orientation included d i s c u ~ s i o n sof the conceptual and i n t e l l e c t u a l bases for the research. They were diven special orientation i n - 283 - the interviewing nethods of village workers, using such techniques as r o l e playing. As mentioned e a r l i e r , intensive f i e l d practice was under the e i r e c t * supervision of the social s c i e n t i s t responsible. Data gathering. The data collection team went out a s a group t o 8 villages. Each village had previously been carefully mapped, and all houses had been given an identification number. A complete census had been done, and each v l l l a g e resident was given an identification nwber based on the house number. From the census, complete lists were ? r a m up of a l l women aged 15 t o 49. They were interviexed i n t h e i r homes, and the tean of f i e l d investi- gators sweeping systematically throqgh the village, In the surve,y of a t t i t u d e s and beiiefs, each investigator was expected to jntenpiew four familietr a - Knowledge and practice Ilifonnatian on the knowledgz and use of the various methods of family planning was obtained front a l l mnrrried women aged 15-49 at the same t i m e as data w a s collected a t t i t u d e s and beliefs and by the same survey te?a. Questions about nethods of contraception were i n f i v e categories: (1) never heard about the family planning method specified; (2) heard about the method but did r o t h o w how t o use it; (3) knew how t o use the method but , 1 never used it; (4) used the nethod e a r l i e r but not currently using i t ; and (5) currently using the method. - 1 - .a I -- pregnancy history and present f e r t i l i t y *- The same team of social s c i e r t i s t s and interviewers that carried out :ile surveys of attitudes, beliefs, knowledge, and practice were also involved i n interviewing a l l married women aged 15-49 about t h e i r pregnancy h i s t o r y . This w a s done 11cocjunction u i t h t h e o t h e r cross- sectional surveys, e i t h e r on the same v i s i t O r i l i t y as possible would :9 8elegated to those with l i t t l e training, e ~ p e c i a l l yt o family health workers, the illdigen- ous midwives, v i l l a g e attendants, and the f a m i l i e s t h e ~ s e l v e s . o Medical diagnosis and c a r e would be moved as f a r a s possible t o the periphery i n t h e s e r v i c e system and i n a s e t t i n g appropriate to the v i l l a g e environment. An e s s e n t i a l aspect of t h i s w a s t o have family health workers l i v e i n t h e v i l l a g e s where they worked. o The s e r v i c e system would provide good canmunications and medical back-up f o r r e f e r r a l . o Rather than wait f o r serious i l l n e s s e s t o develop, t h e emphasis would be on preventive services, home contacts, and e a r l y diagnosis and care. o A major investment would be made to g e t complete coverage through surveillance, not t o g e t compliance with m a s s ,. acedures f o r control. For example, regular weighing and measuring o! children would be a b a s i s f o r identifying f a l t e r i n g growth, and n u t r i t i o n a l supplementa- t i a n would the be focused on those i n need. - Similarly, i n t e n s i v e pzatal screening would be a basis for identifying high-risk - - p r a n a n c i e s , so t h a t normal d e l i v e r i e s could continue t o be handled by d a i s (indigenous midwives). o Aoutine supervisory checks and support would maintain the motivation of workers and :he q u a l i t y of services. o P a r t i c u l a r l v important would be continuing education of a l l members of the health team through r e g u l a r l y scheduled programs of in- service t r a i n i n g and t h e d e v e l o p e n t of f i e l d manuals. o Cornunity p a r t i c i p a t i o n would be encourgaged i n planning and impl.=enting services. The d e v e l o p e n t of the services followed t h e c y c l e of s t a r t i n g from what was simple, next being w i l l i n g t o test many a l t e r n a t i v e s using complex procedures of research when appropriate, and then returning to the simple i n streamlining f o r implementation. As s e r v i c e s were developing e a r l y i n the project, new .. components and programs w-re gradually added, enlarging the scope and comple- x i t y of serv!.ces. About midway through t h e project, a consci.ous effort: was made t o begin t o trim away unnecessary p a r t s of services- - with the ultimate goal of a simple model i n l i n e with r e a l i s t i c c o s t and manpower constraints. ThSs process was still i n progress at t h e end of t h e project. It thus was impossible to plan and impl-ment a l l the experimental service inputs at the same time. Children's s e r v i c e s were t r i e d out and applied f i r s t , followed by women's s e r v i c e s and family planning. Table B . l l i s t n t h e approximate startin&d a t e s f o r all the major s e r i i c e components by ' v i l l a g e and experimental group. These d a t e s r e f e r to the time when the preliminary development of the i n i t j a l service packages was: considered f a r '2 - enough along to have f a i r l y complete services i n t h e v i l l a g e s . - i u t the * s e r v i c e s cohtinued to evolve a f t e r these dates. !e - w Table B.l EFFEC'TiVE STARTING DATES OF THE HATOR SERVICE COWONENTS I N EACH OF TEE STUDY VILLAGES AND EXPERIMENTAL GROUPS Women's Family Children Services Planning a . FPWSCC Ballowal Sep. 1968* Jan. 1969* Sep. 1969* ~Chaminda Sep. 1968* Mar. 1969* Sep. 1969* Dhaipee Sep. 1968 Feb. L969* Sep. 1969* Dolon Kalan Xar. 1969 O c t .1969 Oct. 1969 - FPWS Sekha n.a. Feb 1969* Sep. 1969* Uksi-Dudhal n.a. Apr 1969* Sep. 1969* Chomoir n.a. Pfay 1969 Sep. 1969* Kishanpura n.a. Mar 1970 Mar. 1970 - FPCC Mansurac x p . 1968* n.a. Dec. 1970 Ratt a n ~ e p .1968* n.a. Dec. 1970 Saya Sep. 1968 n.a. Dec. 1970 - FPED Rajoana Kalan n.a. n.a. June 1972 R a j o na Khurd ~ n.a. n.a. June 1972 Bami n a. n.a. June 1972 L i ttar n-a. n.a. June 1972 * Villages i n bhich preliminary developmeit of the d i f f e r e n t s e r v i c e components had been uorked out and t r i e d before t h e d a t e shown. CHILDREN'S SERVICES In rural Punjab and i n most areas of India, children under three bear the brunt of mortality, morbidity, and malnutrition. This include? the neonate, whose immediate weli-being may be determined by the health and nutrition of the expectant mother. The program of child care provided intensive care t o children under three, and episodic coverage of children aged three t o fifteen. Intensive care included routine surveillance of health and developent, immunizations, n u t r i t i c n a l supplementation, and curative care. Episodic coverage f o r older children was primarily curative service i n s u b c e ~ t e r c l i n i c s a t the i n i t i a t i v e of the family when the chlld was ill. A s data from ongoing surveys were analyzed durjng the project, a pattern was developed f o r defining a high- risk child. This knowledge was used to teach fainily health uorkers t o identify these children and maintain closer surveillance f o r them than f o r other children. The characteristics predisposing the high r i s k were: o Low socioeconomic s t a t u s of the family o Low b i r t h weight o r prematurity o A female chilfl, especially wher there was more than one . L o i n the family o Age range between eight and eighteen months. = '2 - o A high b i r t h order. -6 O f chQdren's . services, the components t h a t varied betuee&villages according I) to the ex2erinental design were a s follows: Prenatal care Tetanus immunizations, iron tablets, and f o l i c acid ' e r e provided t o pregnant mothers a s an i n t e g r a l part of child care. Underweight, poorly nourished mothers received supplemental f eedings . Neonatal care c The newborn was checked and weighed by the family hea1t:h worker a s soon a f t e r b i r t h a s possible. Repeat postnatal vi;sits a t twenty-four hours and a t two, s i x , nine, fourteen, and f o r t y days were made on the same schedule a s f o r postpartum care. All neonates with problws o r those weig;hing l e s s than 2,500 grams were referred t o the physician fuc evaluation. Routine care The underlying principle of the whole health program was surveil- lance, i n that much e f f o r t was devoted t o screenirlg f o r morbid it:^ and f a l t e r i n g growth. Care could be early an6 highly focused. Each child was checked a t home i n a systematic sequence. This involved a history and review of symptoms, examination of the child, and checking f o r milestones i n psychomotor develop- ment. Health education was given to the mother a s needed. These v i s i t s were monthly during the f i r s t year of l i f e and then gradually reduced i n frequency * I I t o quarterly v i s i t s i n the thikd year. No routine checks were made a f t e r thirty- six months, unless there was a specific prpblem t o follow up. I n the FWSCC villages, v i s i t s during ihfancy were monthly, but midway through the study t h i s was changed to seven v i s i t s sp?ced a f t e r the I l a s t postnatal v i s i t . (See table B.2 f o r t1.e timing of the+ v i s i t s , t h e i r content, and t h e i r r e l a t i o n to women's services.) I n the FPCC villages, however, a weekly morbidity survey was part of the data collection for the parallel nutrition and infection study. This took the place of the health check. During the l a s t year of the study, a f t e r the nutrition project was completed ia May 1973, the frequency of v i s i t s i n these v i l l a g e s was reduced to a biweekly v i e i t during the f i r s t year of l i f e ahd then according to the pattern of the FPWSCC v i l l a g e s f o r children over two years. Surfeillance vaa also the reason f o r regullarly measuring weight and height a s a device to monitor health and n u t r i t i c ? . Children were! weighed and measured a e part of the health check. Mothers of dlder children were encouraged t o bring them to the c l i n i c f o r their measuremente. Children who had been identified a s faltering below the expected growth cjurves-especially i f they f e l l i n t o any of the three degrees of malnutritiondrere referred t o the physician and started on regular supplementation iq the feeding center. 11 - Smallpox and three DPT immunizations werd routinely given before the I ninth month. Polio, BCG, and measles immunization4 were provid'ed during special campaigns. A t various times, smallpox vacdinations were given I by the family health worker, the family planning w rkerer, o r the government vaccinator. DPT immunizations were f i r s t given b y a team--to get quick I coverage of a l l the children- and then the programlwas shifted to the mainten- l ance routine of the family health worker. -11 Weight-for-age charts based on those develope4 by Dr. David Morley were used to monitor growth of each child. Curves were printed on weight charts indiczting the ranges f o r malnutrition a s follows: .r o F i r s t degree: -- 5 between 70 and 60 percenk of the f i f t i e t h percentile of the Harvard Standard Weight !@ f o r age (HS-50). o Second'degree: between 60 and 50 percent of HS-50. o Third degree: below 50 percent of HS-50. By plotting the child's weight a t each v i s i t on t h i s chart, it was easy to identify whether children were attaining the expected increments i n weight for their age, a s v e l l a s whether they were i n one of the malnourished groups. Table 8.2 COORDINATION OF ROUTINE CHILDREN'S AND WOMEN'S SERVICES AFTER THE NEONATAL PERIOD Child's Children's Services women.'^ Services age (Months; Health Diet survey Immuni- F e r t i l i t y Post check Weight and advice zation Milestones check natal Nutrition services A t the time of t h e health check, the dietary history of the child was determined and ths mother was given appropriate advice about feelding. Early weaning was discouraged both i n d i r e c t advice t o the mother and i n special group meetings organized a s special educational sessions. After four to six months a f t e r birth, supplementation of breast feeding w a s encouraged with appropriate, l o c a l l y adapted weaning foods i n gradually increasing Feeding centers were organized a s a part o f child care t o provide supplementary Feeding t o children identified a s f a l t e r i n g i n growth. Such children were given food supplements i n the feeding center twice a day. If they did not attend, a v i l l a g e attendant took the food t o the home and super- vised the feeding. Siblings who took responsibility f o r bringing younger children were fed, too, i f they wanted some of the supplement. The supplement a t F i r s t consisted of a drink of skimmed milk powder, Fortified with sugar and o i l to increase its calories, o r a gruel of cracked wheat, crude sugar, o i l , and skimmed milk- or both. During the l a s t two years of the project, combina- tions of corn-soya-wheat o r corn-soya-milk were mixed with o i l and crude sugar i n the form of various traditional dis' e s t o replace t h e gruel. All. these Foods were accepted by most children. Iron, f o l i c adid, and vitamins were routinely added to the feedings . 4 Day care centers During the harvest season, day care centers were organized on a 2 - - *- experimental basis i n ~ o m evillages f o r children unddr three. These lasted - four t o s i x weeks. While the mothers from poorer f a d i l i e s were working a l l day i n the i i e l d s , the children were cared for and f=d by v i l l a g e attendants under the supervision of the family health worker. Curative Care The v i l l a g e family h e a l t h worker i d e n t i f i e d i l l n e s s e s e i t h e r during a home v i s i t o r when the child was brought t o the clinic. H i s t o r i e s were e l i c i t e d , examinations performed, and treatments provided-or r e f e r r a l s recom- mended, according t o very s p e c i f i c g u i d e l i n e s i n manuals and standisng orders. The standing orders were modified as f m i l y h e a l t h workers demonstr,ated increasing a b i l i t y t o handle most of the medical problems t h a t coultl be t r e a t e d on an ambulatory ' -sis. Children with s e r i o u s i l l n e s s e s o r with symptoms c l e a r l y defined i n standing orders were referred t o t h e physician a t the time of weekly v i s i t . If necessary, and i f the parent could not take the c h i l d to p r o j e c t headquarters, t h e phys1,lan made a n emergency house call t o see t h e c h i l d i n t h e v i l l a g e . Hospitalization depended on the physician's recommendation, and p a t i e n t s were s e n t e i t h e r t o the teaching h e a l t h center of t h e medicai c o l l e g e next t o t h e project headquarters i n Narangwal, o r i f necessary t o the medical college h o s p i t a l i n Ludhiana, twenty miles away. This system was s e t up t o r e p l i c a t e t h e government's regionalized p a t t e r n of care i n subcenters, primary health centers, s u b d i s t r i c t and d i s t r i c t hospitals. A s t h e project progressed, it was discovered that more and more cases that would originally have been hospitalized were more adequately t r e a t e d at home i n the village. This was e s p e c i a l l y t r u e f o r children with t h i r d degree malnutrition who did not respond rapidly t o supplementation. I n the long run they did a s v e l i with village- level n u t r i t i o n s e r v i c e s a s k-ith i n t e n s i v e - r e h a b i l i t a t i o n i n a hospital. Children's s e r v i c e s were c a r e f u l l y integrated with other s e r v i c e s provided i n t h e villages. A l l women and children i n t h e same family were treated simultaneously i n the home o r the ciinic. Family q l a ~ i n gadvice was given according t o the entry points outlined a t the enl of chapter 2. Careful planning, coordination, and timing of a l l these services was required t o assure the most e f f i c i e n t use of precious home- visiting time. Special reminder cards were developed for each family t o show when each v i s i t was due and what a c t i v i t i e s were appropriate. The reminder cards were a i n t e g r a l part of the family folder, which included all records nf women and children i n each family. An index card f o r each child was f i l e d separately t o help locate the appropriate famlly folder. The folders included the e i g h t chart, routine v i s i t cards f o r the health check, and patient cards f o r curative services. Research d s t a were collected on special record forms f o r morbidity, anthropometry, d i e t , and feeding. Index cards of children requiring special surveillance (such a s maltnourished I children) were kept i n a high priority f i l e to help keep track of them. A t one point, special weight cards were a l s o kept by the mother. But it was f e l t that this produced unnecessary duplication, because the chart on the c l i n i c record a l s o seemed t o provide an effective audio- visual device f o r educating the mother. Under conditions other than research, however, we would recommend that the family-retained weight card should be used. I t . WOMEN'S SEXVICES ~ a n ~ a s e r v i c eprovided t o women i n study villages were considered t:o a f f e c t s i womez and children a s a unit. This overlap became apparent a f t e r we designate2 * E prena+al services a s an integral component of children's services. There was * also much overlap i n the way curative services were provided t o women and children. But c e r t d n unique aspects of women's services need t o be emphasized. F i r s t , f o r maternity care the family health vorker functioned t!lrough the indigenous midwife (dai) a 5 the time of labor. She a l s o involved the d a i as much a s possible i n antenatal and postnatal care. This policy contrasted with the decision t h a t a l l child care sentices should be provided by the family health worker. Second, emergency o b s t e t r i c a l cases can of ten be more rapidly disastrous and can immobilize the patlent more completely than childhood eiaergencies. This gave rise t o the need f o r somevhat d i f f e r e n t means of providing support f o r the family health worker, including rapid acceas t o well-equipped hospital services. Third, many of the services t o women involved a cyclical pattern of pregnant and nonpregwnt s t a t e s , unlike child care, which w a s continuous and linear, paralleling the development of the child. F e r t i l i t y surveillance Each potentially f e r t i l e married woman was v i s i t e d every two months t o record the dates of her menses, detect eqrly pregnancy, and screen f o r symptoms of i l l n e s s , especially gynecologic disorders. This v i s i t was a central a c t i v i t y i n scheduling home v i s i t s and maintaining the l i n k between the family health worker and v i l l a g e women. L Antenatal care When a pregnancy was identified a t the t i m e of a routine bimonthly - - , v i s i t , c a r e was s t a r t e d with an a a e n a t a l check a t the visLt--plus - four more successive v i s i t s a t the home o r iB the c l i n i c ( a t roughly 22, 34, 36, and 38 !? weeks of pregnancy). The servicesYncluded measuring weight, height, and blood pressure, performing an abdominal examination, and testing urine and hemoglobin. The tests were done i n the subcenter by the fanily health worker. Routine treatment during the pregnancy included iron, f o l i c , acid, n~edication f o r parasit- - . two tetanus toxoid immunizations, and health and nutrition advice. Occas :zally, pregnant women received nutritional supplemerrtation on the recomrne~darloaof the family health worker o r physician. On the! l a s t v i s i t , the family health worker and the d a i vould go to the woman's home to make arrangements f o r the delivery. I n addition, care was taken to predict and prepare f o r possible complications a t the time of delivery. An important part of antenatal care w a s the classification of mmen i n four categories of risk. 0 Normal o Abnormality i n present pregnancy (such a s twins o r breech) o Previous obstetrical complications o Unfavorable age, parity, o r s t a t u r e (below a specified height) Appropriate management of the three high- risk categories was decided by the physician. I Labor and delivery - The families and ,the attending dai were instructed to notify the . family health worker a t the onset of labor. The family health worker was prepared zo attend the labor and delivery i f it did not appea-r, to be progress- 3 - i z g nomally. - She did not attempt t o observe the process contlnuouely but - - made reyest v i s i t s t o monitor progress. . The family health woqLers used a list of danger signals of labor f o r appropriate management and f o r determining whether to call for help and consultation. Physicians were provided with special emergency o b s t e t r i c a l k i t s f o r use i n such emergencies. Ale the time of labor and delivery, and during antenatal and postnatal v i s i t s , the family health worker consistently t r i e d to improve the knowledge and s W l l s of the dai. Emphasis was placed on teaching d a i s the danger signals of Labor and the indications f o r referral. A more intensive training program, along with special procedures to incorporate the d a i even more i n t o the team Btructure, was being developed when the project ended. Postpartum care Paralleling the neonatal v i s i t s a t twenty-fcur hours and a t two, s i x and nine days postpartum, the mother's uterine involution, v i t a l signs, and l a c t a t i o n were checked. Follow-up v i s i t s were coordinated with child care on the second and s i x t h weeks and seventh and ninth months postpartum. Special a t t e n t i o n wasgiven i n these v i s i t s t o maintaining lactation. Curative services Women of a l l ages were provided care f o r i l l n e s s e s e i t h e r i n the subcenter c l i n i c o r a t the time of home v i s i t i n g , although the l a t t e r was discouraged unless the problem precluded travel by the woman. A s v i t h child care,, standing orders and guidelines were used by the* family heal tlh worker to e l i c i t h i s t o r i e s , examine the women, and give appropriate treatment o r r e f e r cases t o the physician. Hospitalization o r r e f e r r a l t o the specialty c l i n i c s - a t the health center o r hospital was provided when needed. The women's service component was identical i n W t h groups of v i l l a g e s receiving women's . services (FPWSCC and FPWS) . b The record system f o r women was similar t o t h a t f o r child care and included an index card, a high priority f i l e , a reminder card, a routinn v i s i t card ( f e r t i l i t y survey), and a patient record card. In addition, special records vere w e d t o record pertinent information from previous and current pregnancies, including antenatal v i s i t s , labor and delivery care, acd postnatal v i s i t s . These all vere kept k i t h the childrenOs cards i n the family folder. FAMILY PUNNING Family planning services i n a l l the experimental villages included education and motivation f o r the use of contraceptives t o limit family size and t o space children; contraceptive services; and follov-up of users. A t the beginning of the study, family planning edvsation and motivation were low key. It was thought that previous adverse reactions by v i l l a g e r s t o government campaigns of family planning and the backlash from the mass IUD campaign necessitated a cautious approach while developing rapport and understanding between the project s t a f f and the villagers. So we started by l e t t i n g people know that contraceptives vere avaiiable and by providing contraceptive wLthout aggres- sive family planning education to those who requested them. After other services were well established, family health workers L were knGouraged to discuss family pi nning with wcmen during contacts £0: other services. It was l e f t to the family health worker to sense appropriz-e - - times f o r such discussions. When a potential acceptor was identified, advan- '5 -L tages and disadvantages of various methods were discussed. But no pressure E Sr - t a c t i c s were used, and no monetary o r other incentives vere offered. Nor were targets s e t , but specific routines were established to follow up on acceptors. It soon was recognized that appropriate subjects and occasions to advise on family planning were not easy f o r a young (frequently unmarried) worker to recognize. Therttfore, a special e f f o r t was organized t o formulate guidelines f o r the family health worker on using other service con'scts more effectively f o r introducing family planning ad ice. The occasions when family planning vorkers had to discuss family plnnning bdth the women i n the village were termed "entry points." Ir addition to the family health workers, a male family planning worker was engaged i n educational e f f o r t s among the men. Systematic family planning Family health workers were trained to r e c ~ g n f z eand offer standard- ized advice on eight occasions (entry pofnts) during the cycle of women's care. Thfs l a t e r was expanded t o fourteen by incorporating cther occasions related t o the care of children. Because many women's and children's services overlapped, these entry points also frequently coincided. On such occasions, the family health worker was expected t- combine the edllcatioa appropriate t o those entry points. (See figutd 2.2 i n chapter 2 f 3 r the list of encry points.) The frequency and content of family planning advice depended on the wonen's reproductive history, family socioeconomic conditions, family health, and prior a t t i t u d e s toward family planning. Especially important was the use of a family planning profile developed for each e l i g i b l e .&man--a p r o f i l e the family health worker used t o determine she intensity of sncour:2- ing the limitation a_r spacing of children. Rlis p r o f i l s was based on the '2 number of living chfldren, sex of living children, elapsed time since the l a s t * pregnancy, spacing, qccurrence of abortion o r family planning f a i l u r e s , and L health of the mother. A t f i r s t there were "high," "medium," and " low" prgfiles. Women with a high p r o f i l e r-ceived the most frequent and intense advice emphasizing limitation. The profile was modified near the ead of the project t o include only high and low categories. The f i n a l c r i t e r i a a f a high- profile woman consisted of: o A chronic i l l n e s s o O r an induced abortion o r family planning f a i l u r e within the previous two yeers r O r three or more living children with a t l e a s t two sovs and l e s s than five years sincc termination of the l a s t pregnancy o O r l e s s than three living children with l e s s than two years since termination of the l a s t pregnancy The family health worker rapidly learned to classify women appropriately, but they were a l ~ oencouraged t o use t h e i r judgment to determine when advice could be given. A round, clocklike device, was developed to help family health workers remember when t o give the appropriate advice. This device helped u n t i l the family health workers recognized the entry points reflexively. In the f i n a l form, entry points were w e d consistently by expertenced family health workers but with some f l e x i b i l i t y perditting more frequent o r l e s s C frequ@ntadvice, depending on their judgment of need and on the situation of the home v i s i t . For example, i f male members of the family Kere present a t the time of an entry point, advice or discussion was often postponed u n t i l the next opportunity t o see the woman alone. .I -Provision of contraceptives The contraceptive methods made available t o a l l potential acceptors included condoms, foam tablets, diaphragms and spermicidal cream, o r a l p i l l s , injectable Depo-provera (provided by Upjohn Pharmaceutical Company), intra- uterine devices ( t h e Lippes 27.5mm loop and the 'Taviti" a shield-shaped polyethylene device developed a t Narangval) , s t e r i l i z a t i o n of males and females, and instruction i n the rhythm method. After the introduction of the l i b e r d l z e d abortion laws, aesistance w a s provided t o wmen seeking indaced abortions. Because of Punjab regulations, both abortions a h sterilizations were performed i n accredited private or government hospitals t o which the acceptors were referred. All the other methods were provided by resident f i e l d s t a f f o r a t the t h e of the project physician's weekly v i s i t t o the subcenter. Uthough .iL; nethods were discussed with potential acceptors, suggestions f o r the most appropriate methods were made by the worker a f t e r a screening history and examination. Recommendations f o r a method other tlhan that chosen by t h e accepter were made only i f there vere medical o r =cia1 contraindications. For example, p i l l s were not given during lactation, but Depo-provera was used as an i n t e r h measure because it does not interfere with lactation. Depo-provera vas not, however, suggested f o r prolonged use, unless the couple was r e l a t i v e l y certain that they had completed t h e i r family. * S t e r i l i z a t i o n s were not urged unless the couple had tuo living sons, because of strong evidence that families wanted a t l e a s t two sons. Conventional contraceptives, including condoms and foam tablets, were dispensed by family health workers, family planning wo'rkers, and fanlily I planning educators. Diaphragms, used very infrequently, wete f i t t e d by the physician. A t f i r s t p i l l s , injections, and IUDs vere given only by the physician a f t e r examining the acceptor. In the f i n a l year of the project, the family health workers were taught to dispense p i l l s and i n s e r t IEDs a f t e r completing a check list and examination. If the checklist and examination were e n t i r e l y negative, the family health worker could proceed on h e r own; any p o s i t i v e findings on the check list o r i n the examination required r e f e r r a l t o the physician f o r evaluation. Although a s i m i l a r p a t t e r n w a s judged f e a s i b ~ l e f o r i n j e c t i o u s , it was not carrfed out, except on a few occasions. The rea.son was that more caution needed to be exercised as long a s Depo-provera vas considered a drug i n the c l i n i c a l- t e s t i n g stage. Follow-up of contraceptive u s e r s Family h e a l t h workers followed up on contraceptive users i n the c l i n i c and the home. Encouragement and support were s t r e s s e d a s important components of follow-up. Schedules f o r follow-up v i a its varied as follows. Conventional contraceptives. Enquiries were made at each home v i s i t associated with t h e f e r t i l i t y survey (bimonthly). Supplies were dis- pensed at t h a t time by the family health worker. Alternatively, t h e male family health worker d i s t r i b u t e d supplies (condoms) on h i s routinely scheduled v i s i t s . Oral p i l l s . Users of o r a l contraceptives were v i s i t e d once a month f o r t h r e e months and t h e r e a f t e r every two months a t t h e t i m e of the f e r t i l i t y survey. Enough supplies were provided t o last u n t i l the next v i s i t , . L and careful enquiries were made about t h e manner of taking p i l l s , any menstrual - abnormalities, o r other possible side- effects. - The family health worker checked supplies'fo v e r i f y t h a t the p i l l s had been taken before providing subsequent cyclez !e ~ e ~ o - ~ r z v e r aMonthly fallow-up v i s i t s were c a r r i e d out f o r three . months and then every three months a t the time repeat i n j e c t i o n s were due. Examinations and enquiries were s i m i l a r t o th9se f o r p i l l users. The weight and blood pressure of women was also checked. Because of frequent pro- longed amenorrhea associated with Depo-provera, o r a l estrogen therapy was provided f o r ten days each month t o induce withdrawal bleeding. - IUD. Acceptors wre cliecked one week, one month, and three months a f t e r insertion. Thereafter the follow-up was coordinated every four months with the f e r t i l i t y survey. A t each v i s i t an enquiry w a s made whether the patient could still f e e l the threads of the IUD and whether she had any symptoms, such a s cramping and bleeding. I f there was any question about the IUD's being i n place, the family health worker o r the physician checked it i n a pelvic examination. Sterilization. After vasectomies, t h e man was v i s i t e d by the farcily health worker to find out whether any complications had developed and t o make sure the man understood the need t o continue another family plannng method u n t i l his semen was checked and found to contain no sperm. When tubectomy cases returned t o the village, the family health worker made frequent v l s i t s to check the wound and change the dressing u n t i l the incision healed. Maledamily planning workers , k The cosbination of family planning with women-s and children's - services produced a family planning prohYam centered on women. - It was thought L that family planning contacts with men %ere essential to provide adequate * coverage of both members of the couple, 3 n l i k e health services, which men could seek elsewhere. This w a s especially true because the husbands were generally considered dacisiomnakers on family planning. To provide special coverage f o r men, a male fvnily planning worker was used to supplement the a c t i v i t i e s of three o r four female family health workers. The male worker routinely v i s i t e d all husbands i n study villages-- to motivate them t o use contraceptives, t o supply condoms t o acceptors, and t o follow up users. A . important aspect of the village v i s i t s by note family planning workers was to discuss problem cases with the family health worker and to coordinate the motivation of both husband and wife i n r e s i s t a n t cases. The work load c f the family health w r k e r wass organized so t h a t he could spend ai, h i s time on family planning. On occasion, however, he assisted the family k.ea1r.h worker i n providing smallpox vaccinations and other immunizations. Family planning education The provision of family planning services i n FPED villages was similar to that i n other study villages with respect t o delivery of contra- ceptives and the f o l l o u v p of users. These two a c t i v i t i e s , however, were carried out primarily by one family health supervisor, who served a t four villages, and by one physician, who was available for supervision, r e f e r r a l , and consultation. The major difference i n the FPED villages was that home v i s i t i n g was limited t o family planning education, that group education meetings w e r e organized, and that no health services were provided. The basic village- level workers were called family. planning educators: they were . I C junior basic teachers a category of v i l l a g e teachers with essentially the szie t o t a l amount of educational preparation a s auxiliary nurse-midwives. - This precluded any chance t h a t health care might be delivered udder pressure - of the vill'agers. To give structure to the home v i s i t s , a sequmce of discus- * - sion topics'was developed for family planning educators t o use i n routine v i s i t s t o each e l i g i b l e woman about once a month. When an acceptor was identified, she was referred to the s u p e r ~ i s o ror physician f o r contraceptives. To supplement home contacts, group meetings of women were organized t o discuss aspects of family planning. These meetings proved useful, and before the end of the project similar meetings were being organized i n a l l the study villages according t o the pattern developed i n FPED villages. Family planning records In villages v i t h women's services, a l l family planning v i s i t s were recorded on the women's patient records. I f advice was given a t the time of an entry point, speclal notation was made i n the space provided on t.he form. In the FPCC and FPED villages, a special family planning record was used because selnrices were separated. I n addition, the family planning workers and educators kept separate records f o r each man o r womanvisited. L i s t s of a l l acceptors as well a s special index f i l e of all users vas maintained t o ensure adequate follow-up. STSFING OF THE SERVICE PROGRAMS The basic staffing pattern of the Narangwal services consisted of one female health worker assigned t o each village subcenter; one male family planning worker f o r every three of four female workers; and one family health super- visor f o r every three o r four female workers. I n addition, public health I I nurses provided overall supervision of the female f i e l d workers, wit11 one -_ nurse generally responsible f o r eight or more village subcenters when services were originally developed. Physicians participated i n the coordination and - supervision of services and provided service back-up to family health workers.E by weekly v i s i t s to each subcenter and by being available for emergencies. A male social s c i e n t i s t supervised the family planning workers. Each sub- center served one v i l l a g e wfth an average population of 1,300 t p 1,500. Modifications t o this pattern vere made l a the experimental groups and included the following. First, i n the FPdS r f l l a g e s four subcenters were served by t h e family health vorkets, with one family health worker covering t ~ smaller o villages. Second, the FPPC villages, also part of t h e nutrition and infection study, had five family health vorkers i n the three J i l l a g e s , because a major part of t h e i r responsibility included a detailed morbidity survey. There was, however, no family health s u p e ~ i s o tbecause the additional research e f f o r t required that the workers have training equivalent t o the supervisors i n other villages. A public health nurse provided d i r e c t supervision t o the workers i n these villages. In May 1973 t h e fieldwork of the nutrition project was completed, and thereafter one vorker served each village. In addition, i n the f i r s t tvc years of family planning seivices i n the FPCC villages, one full- time and one half- time family plaming worker provided services t o the men. This vas reduced to one worker i n 1973 t o bring services i n l i n e with those i n the other experimentd groups. Third, for approxhately one year i n 1972-73, only one ,£amily health I I supervisor w a s available to cover both FPWSCC villages. During t h i s t i m e - she supervised a l l sevsn family health vorkers. Although t h i s restricted her involvement i n resear&, it provided an opportunity to observe the feasibility of greater scrvlce covGrage by the supervisor, and the cotrerage seemed adequate. !? I Fourth, i n 1973 one of the more experienced family planning workers was given responsibility f o r additional group education i n all study villages. his consisted of helping the other family planning workers organize group meetings and film shove. - Fifth, the FPED villages had staffing r a t i o s similar to other . villages, except family planning educators replaced family health workers. I n addition, there wa8 one family health supervisor to provide contraceptive services f o r the four villages, but she was nor the educator's supervisor. Sixth, the staffing pattern f o r nurses and physiciacs varied depending on the number available and the intensity of t h e i r research activi- ties. It was greatest a: the ineginning when the services were f i r s t being developed and was decreased a s the e f f o r t s were standardized. The r a t i o of physicians and nurses t o subcenters varied from 1:4 t o 1: 8. Most important was a frequent lack of lady physicians t o supervise both child care and family planning a c t i v i t i e s . The FPWSSCC, FWS, and FPED villages had the lady physicians most consistently. Only during the l a s t year was one lady physician available, and she had to divide her t i m e amoy a l l groups of villa.ges. As a result, male physician8 on several occasions provided a l l services except gynecology and family planning. In the FPCC v i l l a g e s especially, t h i s dual coverage-by aimale physician f o r child care and a female f o r family planning--proved necessary f o r most of the study and made integration of - services d i f f i c u l t . Another variation was that a'nurse with special training ? - i i n family planning served a s a physician-substituLe during the l a s t year of - the project. A t one time or another, she backed family planning services i n all villages of the project. - 313 - Seventh, an important requirement f o r all three types of peripheral workers (family health workers, family planning workers, and family planning educators) was the a v a i l a b i l i t y of "relievers." These were reserve workers of the ! fie category who f i l l e d in when regular workers were on leave. This provided a continuity of fieldvork, continuity considered essential f o r collecting research data and for making services more consistent and more credible t o villagers. Such a system should be used mainly f o r large blocks of leave (such a s maternity leave) i f government manpower is limited. Eighth, i n each v i l h g e v i t h children's services (FPWSCC and FPCC), two local women were hired t o assist the family health worker i n feeding centers and i n home follow-up. These full- time village attendants were directly supervised by the family health worker. BACKGROUND, SELECTION, AND ROLE OF DIFFERENT STAFF CATEGORIES A l l service s t a f f were essentially similar t o workers i n government health services, with a few modifications to permit flexibility i n the research a t Narangwal . Family health workers Two types of health worker were recruited for t h i s position. I n L the FPWSCC and FPWS villages, they had received standard government-prescribed training a s a auxiliary nurse midvife. This training consisted of two years - - '5 of hospital ar-d f i e l d training a f t e r eight' to ten years of basic education. - The emphasis wzs mainly hospital midwifery. I n the FPCC villages, the e- E family health worker was H lady health v i s i t o r , which meant that she had received two and a half years of training beyond the tenth grade i n a special program of iiospital and f i e l d training. The f i e l d and public health aspects of t h e i r training were more intensive than those of the midwife. The s l i g h t l y g r e a t e r academic q u a l i f i c a t i o n s of the lady h e a l t h v i s i t o r were considered necessary because the rsquirements of gathering d a t a f o r the morbidity survey were particularly demanding. A l l family h e a l t h workers were r e c r u i t e d through newspaper advertise - ments and selected by a n interview and t h e results of a nonverbal i n t e l l i g e n c e test (Raven's Matrix Text). Final s e l e c t i o n was based on t h e i r performance during training. Reading and writing a b i l i t y i n English was required because of the research component of the p r o j e c t , b u t t h e a b i l i t y t o converse i n English w a s not e s s e n t i a l . From project experience w e learned t h a t workers with little experience were nore receptive t o r e t r a i n i n g i n new s e r v i c e p a t t e r n s than those with experience i n government h e a l t h centers. Tlhe family h e a l t h worker was t h e primary h e a l t h worker i n each village- - providing both home and c l i n i c s e r v i c e s a s required by the p r o j e c t design. She a l s o r e f e r r e d cases t o the family health supervisor, public h e a l t h nurse, and physician. Family planning educators The background of family planning educators was matched a s c l o s e l y as possible to that of the auxiliary nurse midwife, except the training was L 1 not i n health. Females of similar age and general education with two years' Lcaining a s junior b a s i c teachers were s e l e c t e d i n a manner similar t o family health workers and r e t r a i n e d f o r t h e i r new roles. The educator's conunand of w r i t t e n English was o f t e n not a s adequate a s t h a t of family health workers I (because of differences i n t h e i r previous medium of i n s t r u c t i o n ) , so they kept a l l records i n h n j a h i o r Hindi. The educator w a s the primary family planning worker t o o f f e r some s p e c i f i c contraceptive and follow-up senrices and to r e f e r c a s e s t o the family h e a l t h supervisor and physician. Family planning workers Hen with a high-school degree o r undergraduate college degree were selected a s family planning workers. No formal health training o r experience was required. Selection vas based on c r i t e r i a similar to those f o r family health workers, except emphasis vaa placed on apparent maturity and probably a b i l i t y t o r e l a t e and communicate with v i l l a g e men. The worker was t o develop good relationships with the men i n study villages to educate them about family planning and to motivat them to use contraceptives when appropriate. They also supplied condoms and followed up on husbands of a l l contracepting couples. - Family health supervisors As i s the pattern i n the Punjab government services, the supervisors of family health workers (the Glidwife type) had been trained a s lady health visitors. They were chosen e i t h e r from among the family health workers i n the villages that overlapped with the nutrition study o r ( l e s s often) directly. Their main r o l e was to support and supervise the family health workt~rs. I n d i f f i c u l t or emergency situations they occasionally provided d i r e c t services t o women and children. After workers were trained to screen potential contra- ceptive users, dispense o r a l p i l l s , and i n s e r t IUDs, the supervisors generally . were the f i r s t t o i n s t i t u t e these delegated a c t i v i t i e s i n each group of villages. In the FPED villages, the supervisors worked more a s workers i n providing contracept5ve services directly. But because of her greater experience, she '.i - - L - also w a s frequently used t o a s s i s t the family planning supervisor and social s c i e n t i s t S supervising the family planning educators and i n group education. I I Social s c i e n t i s t s Although selected entirely on the basis of potential cont~eibutions t o the research aspects of the Harangval studies, s o c i a l s c i e n t i s t s vere a l s o - designated t o provide f i e l d supervision of family planning vorkers and t o assume responsibility f o r the FPED villages. They had backgrounds Iln sociology o r s o c i a l work a t the master's o r doctoral level. The j u s t i f i c a t i o n for using people with a high level of training vas the experimental and d e v e l o p mental nature of the work. After the s o c i a l s c i e n t i s t s organized a c t i v i t i e s of workers and most of the FPED program, they vere responsible f o r the continu- ing modification of services. Supervision of family planning workers was limited to a small fraction of a social scientist's time; overall direction of the FPED villages required about half of a social scientist's t i m e . Physicians The physicians i n the project had diverse backgrounds and experience. Some had no specialty training but many years of f i e l d expezience i n primary health centers and goverment family planning programs. Others had specialty training i n pediatrics, obstetrics- gynecology, and public health. Generally those u i t h extensive special training served a s project o f f i c e r s and infrequent- l y dglivered services. They directed and developed t h e service and research components of the project. The other physicians vere also deeply involved i n research and development, but t h e i r priiary role was to provide services unzer '9 r e f e r r a l and to coordinate f i e l d s;zpervfsion with supervisors and nurses. - A l l * physicians took part i n the training prQrams and i n the development of C manuals and standing orders. Village attendants The v i l l a g e attendants generally Lare i l l i t e r a t e women who lived i n study v i l l a g e s and needed employment. They varied from teenagers t o women past the reproductive age. Some happended a l s o to be daxs. Their main work was to maintain the feeding center, prepare 'food supplements, supemrise feeding i n the center, feed nonattenders at home, and a s s i s t the family health worker i n weighing and maintainiog records of malnourished children. In addition, they assisted in other a c t i v i t i e s , such a s c a l l i n g r e f e r r a l cases t o the c l i n i c , rounding up children f o r routine weighing or immunizations, and acting a s a d i r e c t channel of communication t ethe v i l l a g e womec. - Others Laboratory technicians provided simple c l i n i c a l laboratory cests, such a s s t o o l exams, s p u t m smears, white blood counts, urine sediment exams, and blood smears f o r RBC morphology and malaria parasites. They a l s o p e r formed special laboratory t e s t s f o r research substudies and prepared medicines f o r use i n the villages. A n i ~ t r i t i o n i s temployed f o r the research aspects of the n u t r i t i o n project, also assisted i n developing and supervising the feeding centers. . I Especially important was her r o l e k n developing, from the basic s t a p l e s available, the varied food supplements used a s weaning foods. - - 'i OTHER PROGRAM COMPONENTS i F a c i l i t i e s W In each study village, one of the f i r s t a c t i v i t i e s a f t e r its selection was to renovate a village building or house a s a subcenter c l i n i c and headquarters for the village services. Generally the village or an individuai donated an existing or new building f o r renovation. Tie r e ~ o v a t i o n of older structures included adding cement floors, screens, a l a t r i n e , and a- hand-driven pump for wa5er. When posaible, the space was divided i n t o ts~o rooms, one f o r consultation, and the other for examinations, IUD insertioce, and so on. Attempts were made to ensure that each c l f s i c was reasbnably accessible to a l l v i l l a g e families a w did not antagonize any one faction by it3 location. The subcenters were sparsely equipped with the minimum essential furnitare and equipment, much of it ma?= by project c a q e n t e r s : a desk, chairs, storage cabinets, wooden examining tables, and f i l i n g cabinets for records. Purchased equipoent included trays, scales, instruments, and kerosene stoves. In villager providing nutrition services, an additional room--or a room in one or more separate buildings-ras used a s a feeding center. The location was planned t o give eauy access t o poor, high- risk families. Apart from rugs or nats foz the children to sit on, the only e q u i p e n t re- quired .wag stove, utensils, and food storage bins. The workers' village residence was a major concern. Space was usually rented from a respected family, and one o r two rooms were provided with such improvements a s screening and cercent floors. A bath and a kitchen - always had to be b u i l t for the worker. E The Narangwal headqwrters of research and administrativ; ofzr'l c e s were convzrted village buildings. A large training room was used for s t a f f e d ~ c a t i o n ,and a small but comprehensive library was available. A room for consultations and examinations was maintained for cases referred from the villages to the project physicians. Supplies -x Each fortnight, when the field- workers were brought by project vehicles t o the headqwrters f o r traf-ning sessions, they used the opportun.:ty t o replenish t h e i r subcenter supplies. Alternatively, requests were forwarded through supervisors, and supplies were sent i n the vehicles a t the t i m e of f i e l d v i s i t s . A set formula was established f o r drugs dispensed by family health workers and physicians. Supplies f o r the use af the physicians were kept separately in each village, and each of them carried a complete medical box of special supplies on f i e l d v i s i t s . The fonuulnries were frequently reviewed to ensrue t h a t they were l i n i t e d to essential drugs. Almosr a l l drugs were purchased from supply houses in India, except such special items a s Depo- I provera and uoasleo vaccine. Food supplies were obtained from r e l i e f agencies, but crude sugar and ~ iwere purchased. l -Transport - Family health workers and supervisors--and family planning educators and ~orkers--all used bicycles f o r transportation i n and between villages. . I Tbe a g i l i t y t o r i d e a bicycle was a positive although not essential considera- cion i n the selection of new workers. But i f they did not already r i d e * - bicycles, they had t o learn a f t e r joining the project. The bicycle was i especially important to the family plannfng workers and family health super- * visors, who bad to cover more than one vHlage. Nurses and physicians used project vehtcles, i n c l u d i ~motorcycles and scooters, to travel to villages. Vehicles, mostly with four-wheel drive, were also used to bring f i e l d workers to training sessions a t the Narangwal headquarters. - 320 - REFERENCES Antia, N. H. Anrmal Report of Foundation f o r Research i n Community Health. Bombay: Ohokuade/Uran, 1979. Xrole, H., and K. Arole. "The Comprehensive Rural Health Project i n Jamkheti, India," i n K. W. N e w e l l (ed. ), Health by the People. Geneva: World Health Organization, 1975. Coale, A. J. "Tke Demographic Trans1tion, " IUUSP, International Population Conference a t Liege, Belgium, uol. 1, no. 2 (1973), pp. 53-72. Coyaji, B. J. Annual Report on Nadu Rural Health Project. Pune, India: KEM Research Center, 1979. DeSweemer, C., N. K. Sengupta, and S. B. Takulia. Manual f o r Child Nutrition i n Rural India. New Delhi: Voluntary Health Association and Concept Pub- lishing, 1978. Esterlin, R. "The Economic and Sociology of Fertility: A Synthesis," i n C. T i l l y (ed.), Early Industrialization: S h i f t s i n F e r t i l i t y and Changes i- n Family Structure. Princeton: Princeton gniversity Press, 1974. Grant, J. Disparity Reduction Rates i n Social Indicators: A Proposal f o r Measuring and Targeting Progress i n Meeting Basic Needs. Monograph no. 11. Washington, D.C.: Overseas Development Council, 1978. Henry, Louis. "Same Data on Natural Fertility," Eugenics Quarterly, vol. 8, no. 2 (1971). Indian Council on Medical Research. National Conference on Evaluation of Primary Health Care Programmes. Proceedings of a conference held i n New Delhi, April 21-23, 1980. New Delhi, 1981. International Cowci: f o r Educational Develo~ment. The LamDane Health " Development Pro;ett: Thailand's Fresh Approach t o ~ u r a iPrimary Health ,, Care. Case study no. 7. Essex, Conn., 1979. Johns Kopkins University, Department of International Health. Functional Analysis of Health Needs and Services. New Delhi: Asia Publishing House, 1976. '? & :(endall, H.' "The World Fertility Survey: Current Status and Frndings." Populaaon Reports, s e r i e s H, no. 3 (July 1979). . !e - Kielinann, A., C. E. Taylor, and others. "The Narangwal experiment on interaction cf nutrition and infection: 11. Horbidity and mortality effects," Indian Journal of Hedical Research, vol. 68 supplement