India - Second Integrated Child Development Services Project

Document of The World Bank FOR OFFICIAL USE ONLY 4 ;, .1 ., , r, ;, ,; . ; i i ?1: ! I ) Report 'o.10572-IN :: : .;- r:;p it ',:1 Report... ,.F STAFF APPRAISAL REPORT INDIA SECOND INTEGRATED CHILD DEVELOPMENT SERVICES PROJECT JANUARY 22, 1993 South Asia Country Department II (India/ Population and Human Resources Operations Division is document has a restricted distribution and may be used by recipients only in the performance of !ir official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (as of November 30, 1992) Currency Unit = Rupee Rupee 30.0 = US$ 1.00 Rupee 1.00 = US$ 0.033 METRIC EQUIVALENTS 1 meter (m) = 3.28 Feet (ft) 1 Kilometer (km) = 0.62 Miles FISCAL YEAR April 1 - March 31 ABBREVIATIONS AND ACRONYMS ANTC - Anganwadi Training Center AW - Anganwadi Center AWW - Anganwadi Worker ARI - Acute Respiratory Infection Block - Unit of Administration in state governments CDPO - Child Development Project Officer CSSM - Child Survival and Safe Motherhood Project EC - Empowered Committee GOI - Government of India ICDS - Integrated Child Development Services IMR - Infant Mortality Rate Kcal - Thousand Calories LBW - Low Birth Weight MCH - Maternal and Child Health MDM - Mid-Day Meals MLTC - Middle Level Training Center MM - Mahila Mandal MPWF - Multi-Purpose Worker (Female) NIN - National Institute of Nutrition NIPCCD - National Institute for Public Cooperation and Child Development PDS - Public Distribution System PMC - Project Management Cell RTE - Ready-To-Eat SNP - Special Nutrition Program SSNSAP - Social Safety Net Sector Adjustment Program ST - Scheduled Tribe TINP - Tamal Nadu Integrated Nutrition Project WCD - Women and Child Development  FOR OMCIAL USE ONLY INDIA SECOND INTEGRATED CHILD DEVELOPMENT SERVICES PROJECT Table of Contents Page No. CREDIT AND PROJECT SUMHARY ....... . .... . . . . . i I. BACKGROUND A. Introduction . . . . . . . . . . . . . . . . . . . . . 1 B. National Nutrition Programs . . . . . . . . . . . . . . . 3 C. ICDS in Madhya Pradesh..... 7 D. ICDS in Bihar 8... ..8 E. Learning from Other Experiences . . . . . . . . . . . . . 10 F. Rationale for IDA Involvement . . . . . . . . . . . . . . 12 G. Other Donor Role.. . .. 13 II. THE PROJECT A. Project Goals and Objectives . . . . . . . . . . . . . . 13 B. Project Area . . . . . . . . . . . . . . . . . . . . . . 15 C. Project Approach . . . . . . . . . . . . . . . . . . . . 15 D. Project Description .................. . 16 III. PROJECT COSTS, FINANCING AND IMPLEMENTATION A. Cost Estimates.. 23 B. Financing Plan . . . . . . . . . . . . . . . . . . . . . 25 C. Recurrent Cost and Sustainability Implications . . . . . 25 D. Project Preparation and Implementation . . . . . . . . . 26 E. Disbursements . . ........... 27 F. Procurement . . ........ . 28 G. Accounting and Auditing ............. 31 IV. BENEFITS AND RISKS .... . . . . . . . . . . . . . . . . . . 31 V. AGREEMENTS REACHED AND RECOMMENDATIONS . . . . . . . . . . . . 32 * This report Ia based on the fIndIngs of an appraisal massion whIch vIsited IndIa In March 1992. The mission comprised James Green. (Principal Nutrition Spcialist and mission leader), Ellen Schaengold (Senior Operations Officer) and Aruna Chandran (Resoarch Assistant); consultants were ETd Dib (Procurement Specialist), John Keveny (Nutrition SpecTilist), Jay Satis (Institutional Specialist), Pradeep Kakar (Communication. Specialist), Asoko Bahl (Financial Analyst) and Hilary Standing (Anthropologist). The peer reviewers were: Alan Berg, Emmerich Schebeck, and Knlanidhi Subbarao. The report was endorsed by Mr. Richard Skolnik, Division Chief, India Population and Human Resoures Division, and Heinz Vergin, Director, South Asia, Country Department II. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table of Contents (cont'd) Page No. TABLES 3.1 Costs by Component .......... .......................... .. 24 3.2 Cost by Categories and Expenditures ...............to.. 24 3.3 Summary of Proposed Procurement Ar-angements .......... 30 ANNEXES 1. ICDS Performance in Project Areas ..................... 34 2. Process Objectives ........................ 40 3. Tribal Populations .................................... 42 4. Strategies for Reducing Malnutrition Among Young Children ...... ............... 50 5. Coordination between State Department of Women and Child Development & Health and Family Welfare ....... 54 6. Training ................................................ 61 7. Improving Work Routines in ICDS ...................... 67 8. Plan for Introduction of Services in a New Block ...... 72 9. Communications Component ........ .............................. . 76 10. Monitoring and Evaluation ............................. 81 11. Annual, Midterm and Final Evaluations ................. 84 12. Operations Research .... ...................... ............... 87 13. Detailed Project Cost Estimates ....................... 91 14. Implementation Schedule ............................... 104 15. Forecast of Annual Expenditures and Disbursements ..... 105 16. Supervision Plan . ............... . ..... . ........ .... 106 17. Strategies for Strengthening ICDS Among Tribal Populatm ons ...........ietFie............................. 107 18. Documents in Project File ....................................... 111  ^ iii INDIA SECOND INTEGRATED CHILD DEVELOPHENT SERVICES PROJECT CREDIT AND PROJECT SUNMMRY Borrower: India, acting by its President. Beneficlaries: The States of Bihar and Hadhya Pradesh. Amount: SDR 141.6 million (US$194.0 million equivalent). Terms: Standard, with 35 years maturity. On-lending terms: Government of India to Madhya Pradesh and Bihar in accordance with standard arrangements for development assistance to States and Union Territories. Description: The project would be carried out in Bihar and Hadhya Pradesh, two of India's poorest states. Project beneficiaries would be among India's poorest people, many of whom are tribal. The project would improve the nutrition and hea'.th status of children under 6 years of age, with special emphasis on those 0-3 years old, and pregnant and nursing women. It would also help to improve the capacity of the ICDS to deliver services in the two states, including among tribal people, over the longer term. The project would comprise the followi.ag components: (a) service delivers, to increase the range, coverage and quality of nutrition and health services to target groups through improvements in tne design and implementation of software systems, training for health and nutrition workers, provision of health referral services, and increasing the availability of medicines and equipment for maternal and child health; (b) communications, to provide innovative approaches and new messages to increase demand for the full range of project services and provide health and nutrition education; (c) community mobilization, to promote greater community "ownership" of ICDS services and involvement in meeting program objectives of better health and nutrition and women's and adolescent girls' development schemes; and uroiect management, monitoring and evaluation, to manage, monitor and evaluate the project and conduct operations research to test innovative activities and improve aspects of project design. The project would also include actions to strengthen ICDS among tribal populations.  - iv - Benefits: Tlbe msin benefit of the project would be to speed up tLe currently very gradual pace of improvement in pre-school nutrition status in project areas. The project also would contribute to a faster reduction in infant and child mortality. In the last year of the project alone, around 4.0 million pregnant and nursing women and about 12.0 million children under 6 years of age would benefit directly from the project. Risks: The project will be carried out in two of India's poorest states and among some of its poorest people. Thus, it will not be easy to implement. In addition, three particular risks face the project. The first risk is that implementation would be delayed by training and institutional shortfalls, especially in Bihar. The second risk is that additional responsibilities will be imposed on AWs without regard to existing workloads and undermine efforts to focus workers more on case management of malnourished children and improved household food behavior. The third risk is that nutrition-health coordination arrangements will fail to take adequate hold. To reduce these risks: (a) project management staff would be appointed as a condition of effectiveness and empowered committees have been set up in both states; (b) the findings of the mid-term review of the first ICDS project regarding work routines would be incorporated into the implementation of the proposed project; (c) improved health-nutrition coordination would be facilitated; and (d) the project would be subject to very careful and regular monitoring and evaluation, including special monitoring of the impact of the project on tribal people.  v Estimated ProJect Costs:\a Components Local Foreign Total -US$ millions------- A. Service Delivery 1. Nutrition 135.0 10.8 145.8 2. Health 15.8 3.7 19.5 3. Training 11.4 0.6 12.0 Subtotal 162.2 15.1 177.3 B. Communications 7.8 0.6 8.4 C. Community Mobilization 7.0 0.2 7.2 D. Project Management 1. Project Organization 11.5 1.2 12.7 2. Monitoring and Evaluation 0.3 0.1 0.4 3. Operations Research 2.5 0.5 3.0 Subtotal 14.3 1.8 16.1 Total Base Costs 191.3 17.7 209.0 Contingencies 35.7 4.1 39.8 Total Project Costs 227.0 21.8 248.8 \a Including taxes and duties equivalent to US$4.7 million. Project Financing Plan: Local Foreign Total -------US $ millions------- Government\a 54.8 0.0 54.8 IDA 172.2 21.8 194.0 Total Project Costs 227.0 21.8 248.8 \a Including taxes and duties. Estimated Disbursements: IDA - FY FY93 FY94 FY95 FY96 FY97 FY98 FY99 FYOO Annual 14.8 11.6 8.4 34.9 31.1 33.7 32.2 27.3 Cumulative 14.8 26.4 34.8 69.7 100.8 134.5 166.7 194.0  INDIA SECOND INTEGRATED CHILD DEVELOPMENT SERVICES PROJECT I. BACKGROUND A. Introduction 1.01 An important component of India's efforts to raise the living standards of its more than 300 million poor people has been lts human resource development programs which have focussed on maternal and child health, nutrition, and education. Several national and state programs operate specifically to improve child health and nutrition of the poor through service delivery. These programs have met with some success. However, India needs to expand and improve the coverage and effectiveness of its service delivery efforts to maximize their impact on child growth and development. 1.02 Abetted by economic development which has tripled foodgrain production and raised national incomes and productivity, life expectancy has doubled since the decade ending in 1951. India has also achieved considerable success in improving child survival, particularly over the last decade, diuring which the infant mortality rate (IMR) is estimated to have declined by about a third, to 86 per thousand live births in 1988. The death rate for children 1-4 years of age also declined from 19 to 11 per thousand during the same period. Largely as a result of these advances, life expectoacy is estimated to have increased from 50 to 57 years. Improvements are widespread over all regions and in both rural and urban areas. However, considerable variations persist. For example, the IMR in Madhya Pradesh (MP) and in the rural areas of Bihar is 111 and 100 per thousand live births respectively, while in Kerala, it is 31 per thousand. In states with high IMN, a predictable pattern of morbidity including diarrhea, respiratory infections, worm infestations and skin diseases afflicts a large proportion of survivors. 1.03 Despite higher rates of child survival, an estimated one-third of India's pre-school children have some form of serious growth deficit from malnutrition. Available data, while inadequate, suggest only a very gradual improvement in nutrition status over the last 20 years. Malnutrition is particularly intense in Andhra Pradesh, Bihar, Gujarat, Madhya Pradesh, Maharashtra, Karnataka, Orissa, and Rajasthan. According to the latest national data, the proportion of children 1-5 years of age with moderate and severe malnutrition (less than 75X weight for age) varied in 1982 from 18.9Z percent in Kerala to 59.61 in Gujarat. The extent of malnutrition among scheduled tribes and castes was higher than averages in most places. The incidence of severe malnutrition (less than 60Z of weight for age) is declining but exists among all classes. Predictably, the incidence of energy deficiency is higher among the children of landless laborers and small farmers. Evidence of gender discrimination is not conclusive; regional figures suggest some discrimination in north India, especially amongst very young girls. 1.04 Children under 3 years are the most vulnerable. The Indian Academy of Pediatrics uses the 50th percentile of the internationally accepted Harvard standard as its weight for age standard and classifies malnutrition status as  - 2 - follows: more than 80? as normal, between 71? and 80? as mild (grade I) malnutrition, between 61Z and 70? as moderate (grade II) malnutrition, and below 60Z as severe malnutrition (between 51? and 60? is grade III and below 50? is grade IV). Recent survey results from the project areas of MP show that about 30? of the children 0-6 years suffer from moderate or severe malnutrition. Data from Bihar suggests that about 40? of children age 0-6 years suffer from moderate or severe malnutrition. 1.05 The consequences of malnutrition are lowered potential for physical and mental capacity and greater susceptibility to disease. The risk of premature death among severely malnourished younger children is three times that of better nourished ones. Hospital statistics in Tamil Nadu indic that malnutrition is the underlying cause of death in 10 and an associL 5 cause of death in 75? of deaths in the age group 0-5 years. A large proportion of malnourished children also remain stunted with inadequate height for age. 1.06 A prir;sipal cause of malnutrition, of course, is inadequate food intake due to low household incomes. Latest available data (1975-80) show that average calorie intake in almost all states was lower than the recommended allowance of 2400 Kcal/consuming unit. Further, the calorie intake of scheduled tribes and castes was lower than state averages. Since subsequent poverty decline was the lowest among these social groups, their nutrition has probably not improved substantially over the last decade. Although calorie intake for all income groups shows a modest increase, no trend increase was discernible for the ultra-poor, who spend more than 80? of their income to achieve less than 80? of their calorie needs. Evidsnce also indicates that less developed districts have a preponderance of malnourished children among scheduled tribes and castes. 1.07 However, neither adequate food at the household level nor economic development of a state seems to protect children from malnutrition to the same extent as it does adults. In several states, about 252 of all households had inadequate calorie protein levels, while more than 40Z of the children were malnourished (less than 75? weight for age). 1.08 Beyond inadequate food intake, the main causes of malnutrition among young children are three. First is the high incidence of childhood diseases, particularly acute respiratory infections (ARI), measles and diarrhea. More than one fourth of all infant/childhood deaths are attributed to respiratory infections. An Indian rural child also suffers from several episodes of diarrhea per year. Behavioral factors, particularly faulty breast feeding and weaning practices and inadequate patterns of household food distribution, are a second cause of early childhood malnutrition. Colostrum is often discarded and breast feeding frequently begins only three days after birth. Weaning not only begins late but weaning foods do not contain enough nutrients. Third, low birth weight (LBW), estimated at around 30? of all births for India as a whole, often results either in early infant death or subsequent malnourishment. LBW is largely a result of the low weight of women  before, and inadequate weight gain during pregnancy. In addition, children suffer from micro-nutrient deficiencies, particularly Vitamin A. About 30,000 - 40,000 children become blind every year due to Vitamin A deficiency. B. National Nutrition Programs 1.09 The Government of India (GOI) has sought to address nutrition issues in a variety of ways, in addition to income generation. The problem of malnutrition was recognized as early as the First Five Year Plan in 1950 and the first three Plans treated nutrition principally as a component of the health sector. The Applied Nutrition Program, introduced in the Fourth Plan (1969), aimed at raising the nutritional status of the poor by nutrition education and local food production but achieved limited coverage and was abandoned. The Mid-day lieals Program (MDM) was introduced by some states in 1962 to provide supplementary food to primary school children and later (1974) became a part of the Minimum Needs Program. Although its impact on school. attendance, academic performance and nutritional status is not well evaluated, the program continues to operate and currently covers &n estimated 17 million children in India. The Special Nutrition Program (SNP) was introduced in 1970 to provide nutrition supplementation to pre-school children and pregnant and nursing women, primarily from tribal and backward areas and urban slums. An SNP support group was formed in each program village; a village organizer receives a token honorarium to cook and feed beneficiaries. By 1980, the scheme had more than 6 million beneficiaries but the enrolment of children below 3 years of age, the most vulnerable group, was low. In recent years, the program has been implemented in a few states in areas not covered by the Integrated Child Development Services scheme. 1.10 A public distribution system (PDS) to improve the household's access to food has also occupied a central place in public policy since the 1970s. The scheme provides for a network of fair price shops run by the Ministry of Food and Agriculture which provides low-cost staples to poor households against ration cards. However, a substantial share of PDS supplies goes to the cities. In some states, such as Kerala, Tamil Nadu and Gujarat, PDS has succeeded in reaching the poor. But states that account for a substantial proportion of India's poverty population like Bihar, Uttar Pradesh and Madhya Pradesh account for only a small share of PDS offtake. Evidence suggests that a well managed PDS has potential for improving the nutrition status of vulnerable households, although in practice the record is mixed. 1.11 Integrated Child Developmept, Services (ICDS). India's most important national child nutrition intervention is the ICDS scheme, probably the largest program of its kind in the world and certainly one of the most comprehensive and imaginative. ICDS aims to achieve four objectives: (1) to improve the health and nutrition status of children 0-6 years by providing supplementary food to beneficiaries 300 days per year; (2) to provide conditions necessary for child psychological and social development through early stimulation and education; (3) to enhance the mother's ability to provide proper child care through health and nutrition education; and (4) to achieve effective coordination of policy and implementation among the various departments to promote child development. ICDS delivers a package of services comprising supplementary nutrition, immunization, health check ups, referral  - 4 - services, and health and nutrition education to children under 6 years of age, pregnant and nursing women, and pre-school education to children between 3 and 6 years of age. Supplementary feeding generally consists of a hot meal of lentils, rice and vegetables or a mixture of grains. CARE provides food supplements of bulgur wheat and oil. ICDS adopts a holistic approach to improved child development to reduce mortality, morbidity, malnutrition and school drop outs. Initiated in 1975 on an experimental basis in 33 blocks, ICDS now covers 2,200 out of a total of 5,500 rural blocks in India. (Blocks have an average population of 110,000.) 1.12 ICDS is centrally sponsored through the Department of Women and Child Development (WCD) in the Ministry of Human Resources Development in the Government of India (GOI) but is administered by the state governmE,vs. The COI and the states also share ICDS costs. The GOI provides training and operating costs including salaries, equipment, supplies, play materials, petrol and oil expenses, and medical kits, estimated at Rs. 1.0 million per block-year. State governments meet the costs of supplementary food, currently estimated at around Rs. 1.7 million per block-year. Some states have funded ICDS blocks independently and pay all the expenses themselves. 1.13 The GOI has followed a gradual ICDS expansion policy, based on targeting to the most disadvantaged areas. The initial geographic focuE has been on tribal, drought-prone areas and blocks with a significant proportion of scheduled caste population. The program has also been targeted towards malnourished children, but in practice most beneficiaries of supplementary feeding are not selected through nutritional screening, but are fed on the basis of attendance at the Anganwadi center. 1.14 The Anganwadi center (AW) is the focal point for delivering ICDS services at the rate of one AW per 1000 population in rural and urban areas, and one per 700 population in tribal areas. AWs are often located in small rented rooms or in open courtyards. The AW is staffed by a locally-recruited woman worker (AWW) and a female helper. The AWW is a part-time honorary worker paid between Rs. 225 and As. 325 per month depending on educational qualifications. 1.15 The AWW is responsible for growth monitoring of children under 6 years of age, conducting quarterly and annual surveys of the village, organizing supplementary feeding, providing non-formal pre-school education for the older children, imparting nutrition and health education to mothers, making periodic house-to-house visits in the village, organizing primary health care for children and mote 's, referring the needy to health personnel, eliciting community support and participation .ncluding the formation of women's organizations known as Mahila Mandals (MMs), and maintaining records and furnishing reports. 1.16 While ICDS has been recognized internationally for its ambitious goals and innovative features, there is considerable variability in its impact on the target population and also in the quality of services provided. There are several aspects of the program which generally need strengthening such as: (a) training; (b) supervision; (c) targeting of supplementary feeding; (d) case management of malnourished children; (e) linkages with the health  system; (f) nutrition and health education; and (g) community participation. Available evidence suggests that the impact of ICDS depends mostly on these points; it does not reflect significant differences in tribal compared to non- tribal areas, except in respect of coverage, which is consistently lower in tribal than other areas. A recent national ICDS evaluation by the National Institute for Public Cooperation and Child Development (NIPCCD) reported no significant weight or nutrition status differences between ICDS and non-ICDS children under three years of age in comparable areas, mainly because "under- threes are not that easily being captured by the program." 1.17 Training and Supervision. Both training and supervision can play important roles in improving quality service delivery. In particular, four training areas need strengthening. First, there is considerable unevenness in the quality of the more than 200 training centers spread all over the country. Second, despite good training manuals, the training syllabus and materials need to be adapted to allow for variations in differing rural, urban and tribal conditions. Third, the training of supervisors is very weak. Consequently, although many supervisors are aware of ICDS deficiencies in their areas of responsibility, vigorous actions to remedy the situation through more frequent and more extended visits to AWs are not undertaken. The need to supervise a large number of AWs and inadequate mobility also limit their attention to quality issues. Fourth, existing funding places inaedquate emphasis on in-service training of all categories of field staff. The NIPCCD evaluation called supervision perhaps the "weakest link in the ICDS programme." Cited problems included thin supervisor ratios, infrequent visits to AWs and inappropriate supervisor styles and work patterns. 1.18 Supplementary Feeding. The regularity and targeting of ICDS supplementary feeding remains an area of concern. The NIPCCD evaluation showed wide variations in the number of feeding days; over 60X of all AWs had yearly feeding disruptions of more than 60 days, including 25X which were affected for more than 90 days. The timely delivery of supplies in adequate quantities is one of the main problems encountered. An important problem in the supplementary feeding program is the low coverage of children under three years of age, and even lower coverage of pregnant and nursing women. Proportionately, a much higher percentage of 3-6 year olds receive food than in the more vulnerable under-three age group, partly because younger children cannot come to the AW by themselves. The low coverage of pregnant and nursing women has been a problem because of the low priority placed on their feeding by individual AWWs, the inconvenient timings of the AWs for working women, and the reluctance of some women to accept the food supplementatioll under the program. 1.19 ICDS set feeding quotas per block which usually are fully subscribed in non-tribal but underfllled in tribal areas. Most AW nutritional screening is to identify severely malnourished children, who are entitled to food supplementation but because of listlessness and lack of appetite rarely can consume a double ration. Once enrolled, most child beneficiaries continue to receive food regardless of nutritional status until they reach school age. ICDS maintains with some justification that in locales where poverty levels approach 752, such as many tribal blocks, virtually all pre-school children are at malnutrition risk and warrant continued supplementation. However, the  - 6 - cost-effectiveness argument for area targeting rather than individual nutrition screening is weaker in ICDS blocks where poverty levels are substantially lower. 1.20 Case Management. The present supplementation system has other drawbacks. Once a feeding quota is filled, the AWW has little incentive to seek out additional malnourished children for other AW services. Thus, area targeting tends to reduce the emphasis on monitoring individual child growth and can detract from worker focus on case management of malnourished children through health check-ups and referral. NIPCCD found that no more than two- thirds of workers regularly monitor child growth in their AWs. Evaluations have also indicated that long-term supplementation may substitute for food which the child otherwise would receive at home and, thus, run counter to the development of maternal understanding of the special needs of malnourished children and improved family food behavior. On the other hand, the direct impact of the present feeding program on child nutrition may be less than desirable but may induce mothers and children to come to AWs. 1.21 Linkage with Health System. Preventing energy losses from illness is another area of prospective ICDS improvement. Although immunization coverage has increased, both oral rehydration therapy for diarrhea and management of ARI are not systematically practiced at the village level. This situation can partly be attributed to under-utilization or inadequacy of health services. Evidence suggests that the link between health and nutrition services needs strengthening at all levels. The general problem of uneven referral to higher health care levels also affects ICDS; neither the AWW, nor, in most cases, the local health workers keep satisfactory referral records or follow up to ensure that children reach and receive treatment at health facilities. ICDS has taken several steps to improve its linkages with the health system. Coordinating committees with health sector participation have been established at block, district and state levels. Key block and district health professionals have been appointed as technical advisers to ICDS. Joint monthly meetings also take place at and below the block level to review performance; supervisors are expected to plan joint visits. Despite these measures, much remains to be done to ensure coordinated delivery of health and ICDS services at the village level. 1.22 Information, Education and Communications (IEC) and Community Participation. Both nutrition education and community participation need to be strengthened further. According to NIPCCD, only a very low percentage of families provided any form of material support to ICDS; AWW initiative in approaching community representatives appeared lacking. The task of eliciting community particip&-ion is admittedly difficult in villages stratified by social barriers and economic differentials; therefore, health and nutrition activities are rarely conducted in women's working groups, which were envisaged as a major vehicle for community participation. House-to-house visits by the AWW also need to be more regular. Moreover, special efforts are needed for communities to become effective partners with ICDS in overall child development. 1.23 Pre-school for the older ICDS beneficiaries is an integral part of the program, which evaluations have indicated could be strengthened through  improved worker training, supervision and materials. A 1987 NIPCCD study of ICDS's social components observed that around one-third of AWs had enough play materials and teaching aids for pre-school; however, workers themselves have reported deficiencies in being taught how to use them properly. In addition, as with other aspects of ICDS, pre-school functions well in some AWs and poorly or not at all in others. 1.24 Other Program Issues. In addition to the technical issues cited above, there are program issues related to the ICDS such as: (a) the extent to which supplementary feeding should be targeted to the nutritionally most needy (e.g., moderate and severely malnourished children under 3 years and pregnant and nursing women) and the costs and benefits of such targeting; (b) the rapid expansion to new blocks which can leave existing ICDS blocks with suboptimal service levels; (c) the lack of a well-developed communication program aimed at improving community participation, improved health and nutrition education, and creating greater demand for the full range of health and nutrition services offered by ICDS; and (d) the lack of clearly defined work routines for AWWs and supervisors to meet the specific health and nutrition objectives of the program. 1.25 PDS-ICDS Complementarity. The Public Distribution System (PDS) and ICDS are complementary elements of India's social safety net. PDS seeks to promote household food security for poor families while ICDS is concerned with the individual growth and development of pre-school children and the nutritional health status of pregnant and nursing women in socially and economically disadvantaged areas. The Government of India is now carrying out an adjustment program that may have at least a transitional and adverse impact on the nutritional status of the poor. In addition, food prices in India have risen by almost 30Z over the last twelve months. In this context, the proposed project would help to strengthen the nutrition and health safety net among some of India's poorest people, in two of India's poorest states. C. ICDS in Madhya Pradesh (HP) 1.26 ICDS faces a special challenge in tribal areas of MP, because of the poor economic conditions, and problems with service delivery due to the widely dispersed population. Tribal populations tend to live in small clusters or hamlets. Thus, about one-third of the villages in Madhya Pradesh have a population of less than 500 and one-eighth of these have less than 200. Its population density is 149 persons per sq. km., about 56? of the India average. A 1982-84 situation analysis of the tribals in Madhya Pradesh by the National Institute of Nutrition (NIN) showed that around a fifth of pre-school children suffered from ARI and nearly half were stunted. The extent of severe anaemia in the population ranged from 4Z to 23?. 1.27 According to the 1991 census, Madhya Pradesh has an estimated total population of 66.2 million, 7.8? of India's total population. It has a tribal population of about 12.0 million which is 23? of the country's total tribal population. Of its 46 tribes, 12 account for over an estimated 80? of the total HP tribal population. The birth and death rates in 1990 were estimated to be 36.9 and 12.5 per thousand respectively, more than 20? higher than the average for India. The state also has a very high infant mortality rate,  - 8 - 111 per thousand live births. Only about 30Z of females are literate. Madhya Pradesh is also one of the poorest states, with per capita state domestic product in 1987 of about 72? of the Indian average. 1.28 ICDS is targeted to the most disadvantaged areas. Priority has been given to blocks with a high proportion of tribal or scheduled caste populations, and drought prone areas. There are 45 districts and 459 blocks in Madhya Pradesh. ICDS currently covers 115 rural blocks, of which, about 98 are tribal. CARE and the World Food Program (WFP) provide food commodity assistance to 95 and 34 blocks respectively. The remaining blocks are supplied ready-to-eat energy food procured by the state from Karnataka State Agro-Corn Industries. 1.29 Even in the difficult environment of Madhya Pradesh with its scattered population, ICDS coverage of some services is good. The NIN base line survey (Annex 1), carried out in 1990, showed that around 60? of children and 69? of pregnant and nursing women (as compared to a target of 75Z) received supplementary nutrition regularly. However, about a half of the women shared it with others at home. Maternal and Child Health (MCH) service coverage in ICDS areas was much higher than in non-ICDS areas. About half the children under 1 year of age received a full schedule of immunizations, nearly three times the coverage in non-ICDS areas. About 60? of pregnant women received tetanus immunization, and iron and folate tablets, nearly twice that in non-ICDS areas. Similarly, 60Z of women received some health and nutrition education, nearly 6 times that in non-ICDS areas. The AWs were found to be accessible but the accommodations were generally not considered good by the respondents. Growth monitoring was being carried out in most AWs and the AWWs were generally competent to carry it out. There is evidence to suggest that ICDS service delivery in Madhya Pradesh is perhaps somewhat better than but at least comparable to that of the program as a whole in India. 1.30 The reduction of severe malnutrition would remain an important priority for reducing the high infant mortality rates in Madhya Pradesh. After several years of participation in ICDS, around 8? of children under 2 years of age were still severely malnourished (grade III and IV). The base line survey of ICDS areas also showed that the proportion of moderately malnourished children under 6 years of age (grade II) was 24.6Z. Only about a third of the children had normal nutrition status. D. ICDS in Bihar 1.31 Bihar is one of the poorest states in India. In 1987-88, 40.7? of population was estimated to be below the poverty line compared to 29.2? for India as a whole. Base line data from the proposed project areas suggested that 1 in 10 children suffers from severe malnutrition (grade III and IV) in ICDS areas. A third of the children under 6 years of age are moderately malnourished (grade II). Only about a fifth of the children have normal nutrition status. 1.32 According to the 1991 census, the population of Bihar is 86.4 million, 10.32 of India's total. Of this, over 10.0 million are scheduled castes and about 6.0 million are Scheduled Tribes (ST). Twelve of Bihar's 30  - 9 - ST account for more than 952 of the state's tribal population. The birth and death rates in 1990 were estimated to be 32.9 and 10.6 per thousand respectively, about 102 higher than the average for India as a whole. The IMR for rural areas is 104 per thousand live births. Less than a fourth of females are literate. 1.33 Currently, ICDS operates in 168 of the state's 591 blocks; about half of these ICDS blocks are tribal. CARE supplies food commodities to about 100 blocks and the remaining are supplied by the State Food Corporation or the RTE Food supplement. Rice is made available from the public distribution supplies, and lentils are procured from the local market. 1.34 ICDS service delivery and coverage in Bihar is considerably lower than in Madhya Pradesh (Annex 1) and much below the program as a whole. Although about two-thirds of targeted beneficiaries received at least one service, less than a third of children received regular supplementation. Only 112 ever received Vitamin A, and about half the women received some health and nutrition education. Of pregnant and nursing women, 712 received supplementation but less than half consumed it at the AWs and nearly three- fourths of those taking it home shared it with others. MCH service coverage, although higher in ICDS areas, is also very low. Although coverage is increasing rapidly, only 302 of children under 1 year old had received a third dose of diphtheria-polio-tetanus immunization and only 332 of pregnant women received at least one dose of tetanus immunization. Food storage facilities were inadequate and water sources were located far from the AW. Very few AWWs understood the basis for growth monitoring and consequently did not maintain adequate age-weight charts. Coordination between ICDS and the health service at the field level was poor. Supervisors rarely participated in house-to- house visits. 1.35 The base line survey showed that nearly a third of AWs did not reside in the AW center village. A September 1991 analysis of sample monthly monitoring data suggested that (a) about a third of the blocks reported having no food supplies; (b) about 40X of the sanctioned supervisory positions were vacant; (c) although pre-school attendance was reported to be high, population enumeration was inaccurate and growth monitoring of the registered children was irregular; and (d) the house-to-house visits of AWWs in the service area were rarely undertaken. 1.36 The lower ICDS performance in Bihar reflects its overall weak implementation capacity which affects almost all development programs in the state. However, recently the State has taken several steps to strengthen ICDS operations, such as (a) improving government transport arrangements for CARE- supplied commodities so that the proportion of food delivered to AWs has increased from 412 to 592 of the food supplied; (b) arranging for decentralized procurement of lentils in most of the districts; (c) making arrangements with a public sector corporation to supply ready-to-eat food for 35 blocks; and (d) increasing health-ICDS coordination through the scheduling of visits by Multi-purpose Workers (Female) (MPWF) to AWs, distribution of iron and folate through AWs and holding immunization camps at AWs.  - in _ E. Learning from Other Experiences 1.37 The lessons of experience incorporated into the proposed Second ICDS project are drawn from the numerous evaluations of the ICDS program which have been conducted over the years, the lessons learned from the ongoing Bank- supported Tamil Nadu Integrated Nutrition'Project (TINP) and those from other Bank-financed projects in the population and health sectors. 1.38 Case Management. The present supplementation system has other drawbacks. Once a feeding quota is filled, the AWW has little incentive to seek out additional malnourished children for other AW services. Thus, area targeting tends to reduce the emphasis on monitoring individual child growth and can detract from worker focus on case management of malnourished children through health check-ups and referral. Evaluations have indicated that long- term supplementation may substitute for food which the child otherwise would receive at home and, thus, run counter to the development of maternal understanding of the special needs of malnourished children and improved family food behavior. On the other hand, the direct impact of the present feeding program on child nutrition may be less than desirable but may induce mothers and children to come to AWs. 1.39 It is estimated that TINP contributed to a reduction of a third to a half in severe malnutrition among 6-24 month olds, and a reduction of about 5O0 in severe malnutrition among 6-60 month olds. While not conclusive, the available data suggest a strong TINP impact in improving children's nutrition status. TINP also reduced inequalities in the incidence of malnutrition among different districts. In project areas, the overall proportion of children in normal and grade I increased substantially; the proportion in grade II remained nearly the same, implying a favorably upward shift in the overall nutrition curve. There is also some evidence that TINP effects persist beyond the age of 36 months. At 5 years of age, children participating in the project's initial block weighed almost 2 Kgs. more than those in a control group. However, the project was less successful in reaching its health goals. This was largely due to service delivery problems with the health system. 1.40 TINP demonstrates that it is possible to reach a high proportion of younger children who are nutritionally the most vulnerable, and significantly reduce the incidence of severe malnutrition through well targeted health and nutrition services aided by communication and community mobilization activities. While it did not achieve all of its goals, the project has an unusual number of lessons for the design and implementation of nutrition programs, particularly in the areas of training, supervision and monitoring. Key features include carefully defined recruitment criteria for local workers; limiting field worker tasks to those which are manageable and high priority; specification of daily and monthly work routines; decentralized training systems; supervisory practices which facilitate on the job training; the u of local women's groups to support project activities; the display of performance information to clients and workers at the village nutrition center; and a management information system which could rapidly detect performers falling below established norms. On the health side, the main lesson is that large-scale investment in health infrastructure and supplies is not sufficient to improve performance. Complementary software measures are  - 11 - needed to optimize health workers' performance. With a few design changes in the supplementation criteria, and more focus on maternal nutrition and improved health-nutrition coordination, it may be possible to reduce the incidence of moderate (grade II) malnutrition resulting in a greater proportion of children in normal and grade I category. 1.41 The Bank's assistance to ICDS began in 1991 with the approval of the first ICDS project in Andhra Pradesh and Orissa (Ln. 3253/Cr. 2173), now under implementation. The design of that project takes account of lessons from TINP and other experiences in and outside of India. Disbursements are largely on schedule. It is still too early to determine any lessons of experience from the project, but the results of a mid-term review in 1993 will be incorporated into the implementation of both it and the proposed second project. Implementation of the first project has underscored the importance of early establishment of state-level project management units. It also has shown the need for speedy development of communications activities and for ensuring a good balance between expanding ICDS and strengthening the existing program. The proposed project has been designed largely along the lines of the first project and incorporates preventive measures against the kinds of snags that initially hindered implementation of that project. 1.42 Since 1973, IDA has supported seven population projects and one child survival and safe motherhood project. There is also a major health component in the Second Calcutta Urban Development Project. The First, Second and Third Population Projects have been completed and Project Completion Reports (PCRs) and Project Performance Audit Reports (PPARs) have been issued for all except the Third Population Project for which the PCR is under preparation. The First and Second Population Projects suffered some delays in implementation, but each has been evaluated as having achieved its principal objectives. Nonetheless, the gains were only marginally higher when compared to non-project districts. The third project has done very well in Kerala, but less well in Karnataka. Implementation of the on-going Fourth, Fifth, Sixth and Seventh Population projects has been largely satisfactory, with the Fourth project already having a significant impact on morbidity and contraceptive prevalence. The Sixth and Seventh projects are still relatively new, but initial reviews of implementation have reported progress in most areas. The Sixth Project covers the state of Bihar and the Seventh Project, Madhya Pradesh. Provisions have been made under the proposed Second ICDS Project to ensure coordination between the ongoing Bank-financed projects in each state. The health component of the Calcutta Urban Project has had a significant health impact. It was recently judged by a WHO evaluation to be among the most successful programs of its type in the world. 1.43 The Child Survival and Safe Motherhood (CSSM) Project, which became effective on March 5, 1992, supports the enhancement and expansion of the Government of India's Maternal and Child Health Program for 1991-1995. The project is national in scope but with an emphasis on specific districts where maternal and infant mortality rates are higher than the national average. Its specific objectives are to enhance child survival, prevent maternal mortality and morbidity, and increase the effectiveness of service delivery through strengthening diarrhea control, the control of ARI, prophylaxis against blindness and eye lesions due to Vitamin A deficiency, enhanced newborn care  - 12 - and the active promotion of breast feeding. The CSSM Project contains specific guidelines for improved coordination between the health care system and ICDS at the field level (Annex 5). Efforts are being made to ensure that the implementation of the proposed Second ICDS and the CSSM Projects are phased in such a way that services are introduced simultaneously in overlapping project districts. 1.44 The design of the proposed project takes account of both the technical and project implementation lessons learned from the above projects. On the technical side, the project pays particular attention to: developing a strong communications component to promote health and nutrition education and greater community involvement; enhanced training and supervision of workers; refining job descriptions so that they focus on a select number of well focused tasks that the workers can effectively carry out; and sharpening the targeting of food supplementation. In terms of project implementation, the project would support the creation of empowered comuittees to speed implementation; make use of standard bidding documents; introduce annual reviews to assess implementation progress; and build upon an existing program with advanced preparation of many software activities. F. Rationale for IDA Involvement 1.45 IDA's country assistance strategy involves accelerating the pace of India's human resource development and the strengthening of anti-poverty programs, particularly those which comprise a safety net for the poor during adjustment. An important aspect of IDA's human resource strategy is to help reduce excess fertility, mortality and morbidity, increase school enrollment levels and the quality of educational output, and improve nutritional status. IDA's main nutrition objective is to assist the Central and state governments in adopting policies, strategies and cost-effective programs to deal with the nutrition problems of pre-school children and pregnant and nursing women. The proposed project would provide targeted assistance to poor women and children, many of whom are tribal. These are the groups most at risk of falling through the social safety net during the present period of economic adjustment and severe budgetary constraints. As the principal nutrition donor in India today, as well as a financier of structural adjustment, IDA has a comparative advantage in promoting improved effectiveness, efficiency and coverage of Indian nutrition programs. Without the project, the pace of expanding ICDS to beneficiaries would be slower, many of those needing ICDS services would not have access to them, and the program's nutritional impact would be less in both existing and new ICDS areas covered by the project. 1.46 The Government recognizes that a critical aspect of the ongoing transformation of India's development strategy is an effort to speed the national pace of human resource development. The recently approved Credit for the Social Safety Net Sector Adjustment Program (SSNSAP) responds to that opportunity. As part of the Program, the GOI has committed ICDS to focus on particularly disad-vantaged districts with a large proportion of tribal, scheduled caste or slum dwellers and to improve service quality.  - 13 - 1.47 The proposed project would assist the achievement of the SSNSAP objectives by promoting the increased effectiveness and efficiency of ICDS services and program expansion to provide targeted assistance to poor women and children, many of whom are tribal. These groups are most at risk of falling through the social safety net during the present period of economic adjustment and seveze budgetary constraints. As the principal nutrition donor in India today, as well as a financier of structural adjustment, IDA has a comparative advantage in promoting improved effectiveness, efficiency and coverage of Indian nutrition programs. Without the project, the pace of expanding ICDS to beneficiaries would be slower, many of those needing ICDS services would not have access to them, and the program's nutritional impact would be less in both existing and new JY'JS areas to be covered by the project. G. Other Donor Role 1.48 External donor involvement in ICDS is relatively small. However, CARE-donated food provides supplementary nutrition for around 5 million ICDS beneficiaries in 7 states. The WFP provides supplementary food for around 2.1 million beneficiaries in 5 states. UNICEF funds, on a one time basis, equipment and training to establish new ICDS blocks not covered under Bans- financed proi3cts. Its assistance in recent years has averaged about US$5.0 million yearly. An innovative USAID-assisted project has supported ICDS expansion and strengthening in one district each of Gujarat and Maharashtra. It seeks to improve ICDS performance in project areas through in-service training, increased supervision, strengthened communication and development of a management information system. Although the results of its mid-term evaluation were vitiated by severe drought conditions, it showed a considerable increase in coverage by various services. It is still early to assess the impact of the project since a final evaluation has not been carried out. More recently, NORAD is assisting ICDS in Uttar Pradesh and SIDA is assisting ICDS in one district in Tamil Nadu. II. THE PROJECT A. Project Goals and Objectives 2.01 The proposed project would be carried out in Bihar and Madhya Pradesh. The project would seek to accelerate the pace of improvement in the nutrition and health status of pre-school children, particularly children under 3 years of age, and pregnant and nursing women, focussing on households with incomes below the poverty line. Specific impact objectives would be:  - 14 - Madhya Pradesh Bihar (Percent) Current Current Estimate Target Reduction Estimate Target Reduction Severe malnutrWon among children 6.36 months 8.0 4.0 50.0 11.4 6.8 40.0 Moderate malnutrition among children 6-36 months 24.0 19.0 20.0 32.7 24.5 26.0 Low birth weight incidence 40.0 28.0 30.0 40.0 28.0 30.0 It is projected that the above improvements in nutrition status, along with other MCH interventions, would contribute towards a reduction of 30Z in DMR. 2.02 Process and activity targets (Annex 2) to achieve the above impact have been established for: (a) Children under 3 Years. Regular growth monitoring, immunizations, Vitamin A administration, care of diarrheal diseases and ARI, supplementation of those malnourished, and health care and referral of the severely malnourished; (b) Children 3-6 Years. Regular growth monitoring, pre-school attendance, and health care and referral of those severely malnourished; (c) Pregnant Women. Early registration of pregnant women, antenatal care, and iron and folate supplementation; (d) Pregnant and Nursing Women. Counselling on appropriate and adequate breast feeding, weaning, and birth spacing practices; and supplementation of those with inadequate nutrition status; End of ProJect Targets Process Ob ectives (percentage) Madhya Pradesh Bihar Total registratfon of pregnant women 80 75 Food supplementation of pregnant women with inadequate nutrition status (at least 20 weeks) 80 80 Food supplementation of registered nursing women with malnutrition In pregnancy (at least 16 weeks) 90 75 Regular growth monitoring of children under 3 years of age (quartely) 100 80 Supplementation of moderately and severely malnourished children 6-36 months 90 80  - 15 - B. Project Area 2.03 The project would cover 229 blocks in Madhya Pradesh and 207 blocks in Bihar. In Madhya Pradesh, the project would strengthen services in 98 existing ICDS blocks and extend the program to an additional 131 blocks covering an estimated 7.6 million population. This would improve services in 85Z of the existing blocks and expand the total area under ICDS by 114X. Similarly in Bihar, the project would strengthen services in 72 existing blocks and extend the program to an additional 135 blocks covering a population of 12.4 million. This would improve services in 43Z of existing blocks and expand the total area under ICDS by 802. The blocks were selected on the basis of the percentage of population which was either tribal or scheduled caste. In a few cases, blocks were included which are in drought- prone areas.1 2.04 Bihar and Madhya Pradesh contain substantial proportions of India's Scheduled Tribe (ST) population (Annex 3). These populations are distinguished socio-economically by their preponderance in drought prone, marginal farming areas and their dependence on a range of relatively insecure sources of livelihood. These tribal populations are also characterized by low social and health indicators in relation to literacy levels, child survival and child and maternal nutrition. They are distinguished, to varying degrees. by language and cultural practices. In both states, the ST population is concentrated in the more hilly, forested areas which constitute a more or less continuous ecological zone across Madhya Pradesh and southern Bihar. The largest ST groups in Bihar are the Santals, Mundas, Hos, Oraons and Kharias. In Madhya Pradesh, some of the main ST groups are the Gonds, Baigas, Oraons, Muria and Maria Gonds, and Bhils. The project would cover 112 tribal blocks in Bihar with a total population of about 10.1 million and 156 tribal blocks in Madhya Pradesh, with a total population of about 14.0 million. This represents 100Z of the tribal blocks in Bihar and 892 of the tribal blocks in Madhya Pradesh. C. Project Approach 2.05 The proposed project would take the following approach: - It would build on the existing model of ICDS. That model has shown that it can produce positive results, including in Madhya Pradesh and Bihar, and including in tribal areas, despite the shortcomings of the manner in which the program is sometimes implemented. - The project would seek to enhance the quality and effectiveness of ICDS services, by promoting better worker training, improved supervision of workers, strengthened delivery of health and nutrition services, development of a health and nutrition education program, enhanced monitoring and evaluation, and increased community 1 A tribal block is defined as any block whose tribal population constitutes over 502 of its population, as well as any block with an area covered under the tribal sub-plan of the two states which is in the project area.  - 16 - participation. These are the areas most closely associated with a successful outcome for ICDS investments. - The project would aim at expanding ICDS, in its improved form, in areas in which the program is already active in Hadhya Pradesh and Bihar, including tribal areas. - The project would also aim at expanding the enhanced ICDS into some new areas in both states, including tribal areas. - The project would finance efforts to refine further several key areas of ICDS involvement, by operations research, independent review, and impact evaluations. This will be especially the case for supplementary nutrition, community participation, services for remote regions, and services for tribal people. These are all complex areas where no single approach can be used and where it is hoped that the project, over time, will point to a variety of approaches that might be used in the future. - The project should have a positive impact on tribal people and no negative impact. The main ICDS interventions would, in fact, be quite beneficial to the tribal beneficiaries. The project, however, would seek to improve the ability of ICDS to serve tribal people even further by taking a number of steps related to community participation, the start-up of ICDS services in particular blocks, monitoring and evaluation, operations research, and by linking the project with actions to strengthen ICDS among tribal populations. 2.06 Measures were undertaken during project preparation to strengthen participation in the project by tribal people. Project design derives partly from NIPCCD recommendations in respect of ICDS operations in tribal areas as well as from a Bank-initiated review to: (a) make an indicative assessment of ICDS service quality to tribal people in the proposed project areas, and (b) explore methodologies for community participation during implementation. The three-part review comprised: (a) a survey and analysis of literature on the social, demographic and ethnographic characteristics of proposed tribal beneficiaries; (b) a Bihar and Madhya Pradesh field study on the reach and quality of ICDS services and beneficiary attitudes toward them in tribal areas, and (c) a workshop in each project state to verify the findings of the field study and evolve ways of strengthening participation in the project, particularly by indigenous people. The workshops involved tribal people and their leaders, non-government organizations working with tribal people, ICDS personnel and tribal specialists from academic and research institutions and government departments. D. Project Description 2.07 The project would support the strengthening of ICDS service delivery which includes growth monitoring, case management, food supplementation, pre- school education, health and nutrition education, immunization, and health check-ups and referrals. The project would also assist in improving the capacity of Bihar and Madhya Pradesh to deliver ICDS services more effectively  - 17 - in the future. The project would comprise four components: service delivery, to strengthen training, work organization, supervision, and the supply of materials and equipment; communications through innovative approaches and new messages to increase demand for the full range of project services and provide health and nutrition education; community mobilization, to promote greater community "ownership' of ICDS services, and develop pilot schemes for women's income generation, adolescent girls' schemes and creche programs; and project management, monitoring and evaluation, including operations research to test innovative activities and improve aspects of project design such ast (a) improved work routines for supervisors and AWs, and (b) testing of options for therapeutic nutrition supplementation for children of different nutritional grade status. The project would also include an action program for strengthening ICDS among tribal populations. Service Delivery (US$211.7 million) 2.08 The project would concentrate on improving the effectiveness and efficiency of ICDS services ir. the project area by: (a) assisting in upgrading the quality of services in existing AWs; (b) providing an enhanced service package to new AWs in ICDS blocks which are currently underserved according to program guidelines; and (c) establishing new AWs also with an improved ICDS package for blocks in which ICDS is not yet active. A minimum target of 75X of population coverage within any block would be established for both states. This component would finance equipment, medicines, furniture, vehicles, civil works, consultant services and incremental operating costs. It is estimated that the cost of service delivery per beneficiary under the age of six is US$10 per year. 2.09 Improved Planning for the Introduction of ICDS. A Plan for the Introduction of Services (PIS) (Annex 8) would be followed for the introduction of each new block in both states. The purpose of the PIS is to ensure that the following are carried out in a well-planned, systematic manner: (a) all field positions from the district down to the AW are filled; (b) mapping of blocks is carried out to determine the number of new AWs; (c) training is carried out for all new staff; (d) community mobilization activities are satisfactorily implemented; and (e) adequate equipment and supplies are provided to the AW. At negotiations, Madhya Pradesh and Bihar provided assurances that they would prepare and, thereafter, implement Plans for the Introduction of Services for each new block covered with content satisfactory to IDA. 2.10 Expansion and Strengthening of AW. The provision of new AWs for existing and new ICDS blocks would be carried out on the basis of distance and population criteria. The project would finance the construction of about 11,000 AWs averaging about 450 sq.ft. The project would also finance civil works construction of block offices and food storage facilities in selected blocks in both states. At negotiations, Madhya Pradesh and Bihar provided assurances that they would select the location of AWs to be constructed under the project in accordance with criteria satisfactory to IDA, including adequate consultation with the concerned local communities.  - 18 - 2.11 Deliver* of Supplementary Foods. Improvements in the delivery of supplementary foods would be achieved through the provision c additional storage facilities at the block and district levels, simplifications in the distribution system, and improved monitoring of food distribution by independent consultants. At negotiations, Madhya Pradesh and Bihar provided assurances that they would take necessary steps to make available to Anganwadip adequate supplemental food required for the carrying out of the project. 2.12 Support for Pre-School Education. Although considerable time is devoted to pre-school education, base-line studies and other studies suggest that its quality is uneven. The project would support initial provision and replenishment of educational toys and play materials at AWs. It would also promote use of a more interactive and participatory approach to education through AWW training; and through communications enhance the capacity of parents and other care givers to provide a stimulating environment at home. 2.13 Improving ICDS Supervision. The project would improve the quality of program supervision by strengthening in-service training (see para. 2.17) and through improved mobility by providing mopeds to field supervisors. The project would increase the amount of supervision by adding a block level supervisor with special responsibilities for organizing in-service trairing, communications and community mobilization activities. There are eristing shortages of ICDS personnel at the Supervisory and the Child Development Project Officer (CDPO) levels, and among health personnel at the field level in the existing blocks in both states. At negotiations, Madhya Pradesh and Bihar provided assurances that they would fill, in accordance with a staffing plan, including a time schedule satisfactory to IDA, all vacancies under ICDS and the health programs in the project blocks. 2.14 Improved Health-Nutrition Coordination. Cooperation between ICDS and family welfare workers in the field has improved in recent years. However, much more is needed to strengthen health-nutrition coordination. The project would support additional actions to reinforce and complement coordination activities in such areas as joint supervision, informal joint training, and harmonization of formal training curricula (Annex 5). 2.15 Pharmaceuticals. The project would finance only a modest addition of medicines to health centers and subcenters. These supplies would complement those provided under other Bank-financed projects in the states such as the Child Survival and Safe Motherhood Project. 2.16 Strengthening Operational Research. The project would support operational research for such areas as: (a) improved work routines for supervisors and AWWs; (b) testing of various options to therapeutic nutrition supplements of children in 14 blocks with different nutrition status; and (c) service delivery for scattered populations. Annex 12 contains a detailed discussion of proposed operations research, the important aim of which is to assist in refining key program areas. At negotiations, the states agreed on a program of operational research and provided assurances that they would carry out that research according to a timetable agreed with IDA and discuss the  - 19 - findings of the research in a timely way after the completion of individual studies, including how the results of the study may be incorporated into the delivery of ICDS services. 2.17 Training. To strengthen the expansion of ICDS, the project would provide (a) orientation training to staff of 98 existing ICDS blocks (10,800 helpers, 10,800 AWWS, 750 Supervisors, 98 CDPOs) in Madhya Pradesh and 72 existing ICDS blocks (8,000 helpers, 8,000 AWWs, 550 supervisors, 72 CDPOs) in Bihar; (b) pre-service training to staff of 131 new ICDS blocks (16,375 helpers, 16,375 AWWS, 1,100 Supervisors, 131 CDPOs) in Madhya Pradesh and 135 new ICDS blocks (17,000 helpers, 17,000 AWWs, 1,150 supervisors, 135 CDPOs) in Bihar; and (c) annual in-service training to all the staff. In Bihar, special field-based training would be provided to those whose performance is found deficient after orientation training. Key areas of training for different categories of staff and how they would be organized are given in Annex 6. 2.18 To improve the quality of training, the project would strengthen training institutions and support curriculum review and development. The current ICDS pre-service training pattern is as follows: AWWs at AW Training Centers (ANTCs), supervisors at Middle Level Training Centers (MLTCs) and CDPOs at the National Institute of Public Cooperation and Child development (NIPCCD), which also provides training to trainers. NIPCCD recently carried out a national ANTC evaluation which indicated major differences among ANTCs in quality of training being imparted. Therefore, the project would support strengthening 35 and 26 ANTCs in Madhya Pradesh and Bihar respectively though training of trainers, provision of necessary equipment and supplies, review of their curriculum and implementation arrangements for field placement. In Bihar, an additional MLTC would be set up to handle the increased supervisory training load. 2.19 In-service training to improve program performance would be field- based and linked with program development by focusing on a few key themes. It would emphasize upgrading of technical and problem solving skills and would be conducted jointly for ICDS and health staff. Current refresher training is imparted to staff once in two years by the pre-service training institutions. However, this training is treated as residual after pre-service training obligations are met and remains ad hoc. The project would establish 26 mobile training teams each in Bihar and Madhya Pradesh with specially appointed staff. In collaboration with other district and block level staff and teams being established under the Bank-financed Sixth Population Project in Madhya Pradesh and the Seventh Population Project in Bihar, these teams will train supervisors, who in turn will train village level staff under a team's guidance. The senior block level supervisor would be responsible for organizing these activities at the block level. 2.20 In both states, curriculum preparation, material development, and planning and monitoring of training activities would be the responsibility of project management cells under the project coordinators. To ensure that training is both relevant and effective, the project would also support  - 20 - frequent training of trainers and technical assistance for curriculum review and development. Provisions would be made for the introduction of participatory techniques for mobile training teams by qualified NGOs. Communications (US$10.2 million) 2.21 The communications component would consist of the development of a sound data base, the development of a communications strategy, the use of creative skills to implement the strategy, production of materials such as slides, tapes, posters, and the selection of appropriate media and identification of target groups (Annex 9). Knowing the audience is a prerequisite to successful communications. The coverage of tribal areas in Bihar and Madhya Pradesh makes it all the more necessary that a comprehensive and current data base be used to develop this component. There is a need for both quantitative and qualitative data collection including information on signs, symbols, festivals, rituals and taboos. The communications strategy would be particularly important for reaching pregnant and nursing women. 2.22 Communication strategy development which will take place during the first year of the project would address such issues as the selection of the target audience, the setting of objectives for each target audience, the selection of the broad approach with which each would be pursued. There are six main groups to be reached under this component: mothers of children between 6 and 36 months, mothers of children botween 3 and 6 years, pregnant women, potential mothers (adolescent girls) and older women. To be effective, communication must attract attention and generate involvement. Creative inputs can make a significant difference in the quality of the impacts of the communications component. 2.23 The communications strategy, message development and assistance in implementation would be undertaken by a professional consulting firm with experience in communications, mass media and social marketing. A condition of credit effectiveness would be that Madhya Pradesh and Bihar would have appointed consultants with qualifications and terms of reference satisfactory to IDA, to assist in carrying out the communication components. At negotiations, the states provided assurances that a communications strategy would be completed and sent to the IDA for review and approval by October 1, 1993, and that they would thereafter implement such strategy, as agreed with IDA. The project would finance a contract for publicity services which would cover the development of a strategy and specific messages and, pretesting of these messages. It would also finance IEC materials and equipment. Community Mobilization (US$7.9 million) 2.24 To ensure effective delivery and sustainability of the program at village level, the project's community mobilization activities would aim at promoting community "ownership" of ICDS services and encouraging individual and community self-reliance. Currently, community participation is low in both states. ICDS largely relies on formation of women's groups such as Mahila Mandals (MMs). However, these are generally dormant. In Bihar, only half of the existing AWs have established MMs. TINP and other experiences  - 21 - show that women's groups can be very effective in gaining community support for project activities. The project would support pilot schemes to develop a system for establishing and sustaining village level groups. 2.25 To continue the search for effective means for community mobilization, the project would encourage pil,t schemes to (a) provide a clearer definition of what can be realistically achieved through increased community mobilization; (b) involve NGOs in the design, planning and implementation of community mobilization activities to take advantage of their successful experiences in this area; (c) provide models which would be replicable on a large scale in such areas as women's income generation, adolescent girls' schemes and creche programs; and (d) develop innovative approaches to integrating other women's activities at the village level. At negotiations, Madhya Pradesh and Bihar provided assurances that they would furnish proposals for the pilot schemes according to timing and criteria agreed with IDA. They also provided assurances that they would select NGOs to assist in carrying out the project, in accordance with criteria and procedures satisfactory to IDA. These criteria and procedures would ensure the transparency of the role of NGOs in the project and the selection procedures. The project would finance consultant and NGO services, equipment, and incremental operating costs. Project Management, Monitoring and Evaluation (US$19.0 million) 2.26 Project Management. A project management cell (PMC) in each state will be headed by an Additional Director drawn from the Indian Administrative Service as Project Coordinator (PC). He or she would report to the state Director, ICDS, and would be in charge of day-to-day project activities. Current fiscal stringency dictates that only an absolute minimum of new staff positions be added. However, to reflect the increasing technical demands of expansion in scope and range of ICDS services, professionalization of state levei management teams is necessary. Therefore, the project, in each state, would support a core PMC group consisting of a nutritionist, a communications specialist, a sociologist/anthropologist with experience in tribal areas, and a training specialist. The PMC would also include an ICDS Plus Unit which would consist of persons trained in .ommunity organization and technical specialists. The Unit will consist of at least two persons located at the PMC headquarters and at least one person located in a minimum of ten district headquarters. The Unit would be responsible for implementing the community mobilization component and the strategies for strengthening ICDS among tribal populations. At negotiations, Madhya Pradesh provided assurances that it would establish a regional directorate, with staffing, facilities, and terms of reference satisfactory to IDA to assist in the implementation of the project in the eastern parts of that state, in which there are several difficult tribal districts. At negotiations, the states provided assurances that they would establish and thereafter maintain an Empowered Committee (EC), chaired by the Chief Secretary and comprising the Secretaries responsible for ICDS, Health and Finance. The committees would approve plans, issue necessary government sanctions for implementation, review the annual implementation plans, and monitor their progress. The EC would be expected to meet on a quarterly basis. The PC would be a Member-Secretary of the EC. The project would finance vehicles, equipment, office furniture, incremental operating  - 22 - costs, and consultant services. A Condition of Credit Effectiveness would be that the Governments of Madhya Pradesh and Bihar would establish, with adequate staff, Project Management Cells. 2.27 District Management. As ICDS expands, need for coordination with health, education and other departments, managerial requirements of planning, monitoring, personnel and logistics, and implementation of enhanced communication and community mobilization, all require that district level management be strengthened. The project would support establishment of 17 offices in Madhya Pradesh and 13 offices in Bihar to cover the districts without such offices. All district officers would be trained in planning and management, and guidelines would be developed for their functioning. In collaboration with the project-supported district training teams, these district offices would: prepare district plans in coordination with the health system; monitor performance, enable, empower and motivate service providers through continuing in-service training and ensuring the provision of supplies; and implement the project's communication and community mobilization activities. 2.28 Monitoring and Evaluation. Service statistics would be used to monitor coverage by various services and provision of project inputs. A revised ICDS reporting system is currently being implemented in both the states. It would be adapted to reflect a wider range of services such as the communication and community mobilization activities of the project. However, the main difficulty is the lack of use of this data by program managers for monitoring and taking corrective actions. A USAID-assisted ICDS project has developed a computerized progress reporting system which can provide feedback to program managers on a variety of performance indicators as well as time- series and cross-sectional comparisons. The project would support implementation of this system in both states. Supervisors would be responsible for improving the reliability of data. They would carry out sample checks of field records, train those AWWs who are deficient in record keeping and annually share AW performance data with the community. The project would finance consultant services. 2.29 The first ICDS project includes operations research on supplementation, field-based in-service training, nutritional rehabilitation centers, work routines and supervisory practices, communication and community mobilization, all of which would be evaluated by mid-term. A process of linked annual reviews is planned for both the first and second project states where lessons learned in a state can be shared with others. 2.30 Base line surveys have been carried out in the project areas (Annex 1). Annual and mid-term evaluations to assess progress are planned with the latter taking place in the third year of the project (Annex 11). These evaluations would provide an opportunity to make any necessary mid- course corrections. The annual and mid-term reviews would evaluate ICDS in terms of nutritional status of children under 3, availability of services, community perceptions of ICDS, procurement, logistics, budget releases and the implementdtion of ICDS in tribal areas. A final evaluation would be carried out at the end of the project (Annex 11). At negotiations, Madhya Pradesh and Bihar provided assurances that they would carry out annual and mid-term  - 23 - reviews and final evaluations of project operations and provide the final reports to IDA for review; the Government of India provided assurances that it would participate in the carrying out of these reviews and evaluations. 2.31 Indigenous Peoples. The project is expected to have a positive impact on the tribal populations in the project area. Where ICDS already exists in tribal areas, there is demand for its services (Annex 3). This project does not include an indigenous people's development plan. However, measures, including actions set out in "Strategies For Strengthening ICDS Among Tribal Populations' (Annex 17 of the Staff Appraisal Report), were agreed on to ensure that the project is implemented in a manner compatible with the social and cultural values of tribal communities, and includes the informed participation of tribal beneficiaries. Assurances were received at negotiations that the project would be implemented in accordance with the Strategies, which include (a) strengthening community participation of tribal beneficiaries and the implementation of ICDS in remote areas, (b) the introduction of training programs developed for tribal personnel employed by ICDS and (c) the development of a data base on tribal groups covered under the project. The design of monitoring and evaluation instruments would capture differences in process and impact indicators among tribal and non-tribal groups. In order to provide for the informed participation of tribal communities during implementation, at negotiations, Bihar and Madhya Pradesh provided assurances that district and block level ICDS coordination committees would be established or reconstituted to provide adequate representation for tribal communities and that these committees would provide advice on and monitor the implementation of ICDS among tribal populations in the two sates. To assist the state governments in ensuring that the project takes account of tribal social and cultural values during implementation, the Government of India at negotiations provided assurances that it would establish a panel of advisors of national repute in nutrition, maternal and child health and social sciences related to tribal affairs by December 31, 1993, to review annually project implementation in tribal areas in both states and discuss the results with IDA, including proposed steps resulting from the review. 2.32 In Madhya Pradesh, a special regional directorate for implementation is being established under the project to deal with problems of sparsely populated tribal areas in the eastern part of the state. Lessons learned from this center will be applied to the program in other parts of the project areas. The ICDS Plus Unit in the PMC would have primary responsibility ,or implementing the actions for strengthening ICDS among tribal populations under the direction of the PC. III. PROJECT COSTS, FINANCING AND IMPLEMENTATION A. Cost Estimates 3.01 Cost Summaries. The total cost of the project, net of duties and taxes is estimated at about Rs. 7449.1 million or US$244.1 million equivalent. Duties and taxes are around US$4.7 million. A breakdown of costs of the proposed project by component and categories of expenditure appears in Tables 3.1 and 3.2, respectively. Detailed project costs by component, categories of expenditure and year appear in Annex 13.  - 24 - Table 3.1: Costs by Component Components Local Foreign Total Local Fore gn Total -----Rupees (millions)----- -----US B (millions)