Tunisia - Health and Population Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15857 PERFORMANCE AUDIT REPORT TUNISIA HEALTH AND POPULATION PROJECT (LOAN 2005-TUN) June 28, 1996 Operations Evaluation Department 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. Currency Equivalents (annual averages) Currency Unit = Dinars US$ SDR 1980 0.4050 0.5340 1981 0.4930 0.6002 1982 0.5907 0.6792 1983 0.6788 0.7612 1984 0.7768 0.8494 1985 0.8345 0.8314 1986 0.7940 1.0277 1987 0.8287 1.1035 1988 0.8578 1.2090 1989 0.9493 1.1888 Abbreviations and Acronyms BHC Basic Health Center BHS Basic Health Services CA Central Administration CBR Crude Birth Rate CNI National Computing Center CRFP Center for Pedagogical Research and Training DBHS Directorate of Basic Health Services EPI Expanded Program of Immunization EPS Public Health Establishment (Etablissement Public de Sant6) FP/MCH Family Planning/Maternal and Child Health GOT Government of Tunisia IDA International Development Agency JEC Information, Education and Communication IMR Infant Mortality Rate INE National Institute of Child Health INNTA National Institute of Nutrition and Food Technology MAS Ministry of Social Affairs MCH Maternal and Child Health MIS Management Information System MOP Ministry of Planning and Finance MOPH Ministry of Public Health OED Operations Evaluation Department ONPFP National Office of Family Planning and Population PAR Performance Audit Report PCR Project Completion Report PCU Project Coordination Unit PHN Population, Health, and Nutrition RDH Regional Directors of Health SAR Staff Appraisal Report SPD Studies and Planning Directorate (MOPH) TFR Total Fertility Rate UNFPA United Nations Fund for Population Activities UNICEF United Nations Childrens Fund USAID United States Agency for International Development WHO World Health Organization Fiscal Year Government: January I - December 31 FOR OFFICIAL USE ONLY The World Bank Washington, D.C. 20433 U.S.A. Office of the Director-General Operations Evaluation June 28, 1996 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Tunisia Health and Population Project (Loan 2005-TUN) Attached is the Performance Audit Report on the Tunisia Health and Population project prepared by the Operations Evaluation Department (OED). The loan for US$12.5 million equivalent was approved in May 1981. In March 1987, US$4.0 million was canceled at the request of the Tunisian Government, principally due to the dramatic appreciation of the US dollar. The project was closed in December 1988 after two extensions, and an undisbursed balance of US$0.4 million was canceled following final disbursement in October 1989. The project's objectives were to integrate preventive, curative, family planning, nutrition, health education and sanitation services to establish a basic health care delivery system in eight governorates; and to strengthen the planning and management capacities of the Ministry of Public Health (MOPH) in order to extend basic health services gradually to the rest of the country. The project had five components: (i) to strengthen the management capacity of the MOPH; (ii) to improve and expand the basic health care delivery system by extending physical infrastructure in eight governorates; (iii) to increase the scope and effectiveness of communications programs in health, population, and nutrition; (iv) to upgrade the training system and infrastructure, and train, upgrade and redeploy health personnel in the project area; and (v) to provide technical assistance in support of all components. Several factors hindered implementation, the most important being the institutional rivalry between the MOPH and the National Office of Family Planning and Population (ONPFP) that prevented the planned integration of family planning with basic health services. Despite the circumstances, the project successfully constructed and equipped 161 health facilities, representing more than a 50 percent increase in the number of basic health facilities, thus greatly improving access to basic health services in rural areas. The project also succeeded in strengthening the capabilities of the basic health system and the orientation of the MOPH through internal reorganization and decentralization. Because the planned management information system failed-the system never became operational although 30 computers were acquired-the original project indicators were never updated. Thus the project monitoring database is poor, making it difficult to measure attainment of specific project goals or assess health service outreach. Under the training component, the project constructed or remodeled and equipped six paramedical schools, developed curricula, and provided consultant services to prepare programs and seminars. The project also stimulated much training activity, though it was not directly funded by the project. Achievements under the health education component were modest, as health education was of uneven quality and most health posts were without basic educational materials. 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. 2 OED, in agreement with the PCR, rates the project as satisfactory, sustainability as likely, and Bank and borrower performance as satisfactory. The audit upgrades the institutional development rating to substantial, for two reasons: eight years after project closure the institutional framework it helped put in place for district health services still serves as the national model, and the service delivery methods developed under the project have been standardized across all governorates. In addition, the Directorate of Basic Health Services has been strengthened and the decentralization of authority for the Basic Health Services to the governorates is being continued under the 1991 Population and Family Health project. The project offers some important lessons. To ensure the feasibility of project goals, it is essential to examine the existing health system priorities, related political and organizational factors, the government's long-term policy goals, and the process envisioned for meeting them. To measure the success of population, health, and nutrition projects it is vital to build monitoring and evaluation systems into the project from the start. The Bank should not put undue reliance on technical assistance. Attachment FOR OFFICIAL USE ONLY Contents Preface..........................................................................................................................................3 Basic D ata Sheet.......................................................................................................................... 5 Evaluation Sum mary ............................................................................................................. 7 1. Sector Background ....... ..... ................................ .............................. 11 Health Status ...................................... ............. 11 Tunisia's Family Planning Program................................11 Integration of Family Planning and Health Services. ............... ....... 12 Decentralization.............. .............................. 12 Bank Activities in the Sector ............................ ......... 13 The Integration Debate ...................3...... ..............13 Subsequent Bank Activities in the Sector ....................... 14 2. Project Objectives and Content............................................................................. 16 Project Components ................................. ......... 16 Relevance................................................ 17 3. Project Implementation and Results...................................................................... 19 Overall Implementation .................... .............. 19 Disbursements.........................9...... ..............19 Procurement ................................................... 20 Management Capability.......................................20 Basic Health Care Services......................................22 Health, Population and Nutrition Communications ................. .....28 Training..................... ........................... 28 4. R atings .......................................................................................................................... 29 Outcome ................................................. 29 Sustainability ........................ .............. 29 Institutional Impact..............................................30 Borrower Performance......................................... 30 Bank Performance .................................. ......... 30 5. Issues and Lessons ................................................................................................... 32 Integration of Family Planning and Basic Health Services ......................32 Project Monitoring and Evaluation. ....................... .........33 Lessons .................................................. 34 This report was prepared by Ms. Laura Raney, Health Economist, guided by Mr. Robert van der Lugt, Task Manager who audited the project in December 1995. Ms. Diana Qualls and Ms. Sophie Lefebvre provided administrative support. The report was issued by the Agriculture and Human Development Division (Roger Slade, Chief) of the Operations Evaluation Department (Francisco Aguirre-Sacasa, Director). 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 wiLhout World Bank authorization. Figures Figure 1.1: Tunisia Health Sector Activities Timeline......................... 15 Tables Table 3.1: Evolution of Density of Basic Health Centers in the Eight Project Governorates 1982-1986............................................. 24 Table 3.2: Family Planning, Prenatal and Vaccination Medical Consultation Activities in Basic Health Centers, 1986............. ...... .......25 Table 3.3: Distribution of Responsibility for the Delivery of Family Planning Activities......26 Table 3.4: Basic Health Care Activities, 1993 .......................... 27 Annexes I. Tunisian Legislative Changes ......................................37 2. Major Phases of the MOPH Program and Decennial Characteristics .. ...........39 3. Borrower Comments..........................................44 Map IBRD 15368R 3 Preface This is a Performance Audit Report (PAR) for the Tunisia Health and Population Project, for which a loan in the amount of US$12.50 million equivalent was approved on May 28, 1981 (Loan 2005-TUN) 1. In March 1987, US$4.0 million was canceled at the request of the Tunisian government, principally due to dramatic appreciation of the US dollar. The government estimated and the Bank confirmed that only US$8.5 million of the loan would be needed. The loan closing date was extended twice, and the project was closed on December 31, 1988. The loan account was kept open until September 30, 1989 to complete disbursements. On October 1989, the remaining amount of $353,225.79 was canceled. The PAR is based on the Project Completion Report (PCR),2 the President's Report, the Staff Appraisal Report (SAR), a review of Bank files, sector and economic reports, and discussion with Bank staff. An OED mission visited Tunisia in December 1995, discussed project experience and outcomes with government officials, and visited several of the project sites in the governorates of Jendouba, Le Kef, and Sousse. The PCR was completed in October 1990 and provides an excellent account of the implementation of the project as well as lessons learned. Parts I and 3 were prepared by regional staff, and Part 2 includes the Borrower's comment. The PAR confirms the major findings of the PCR but examines in more detail several of these findings as well as some not fully covered in the PCR, in particular the integration of family planning and basic health services and monitoring and evaluation. It also examines the outcome of the project in the context of Bank lending for the sector as a whole. The findings of the PAR will be a building block for OED's forthcoming study of experience with health lending. In accordance with normal OED procedures, copies of the draft PAR were sent to the Government of Tunisia for comments. These comments have been included in Annex 3 with appropriate cross references and annotations. 1. All references to the Loan amount in the text and tables of this report are in US$ equivalent. 2. Project Completion Report, Tunisia Health and Population Project (Loan 2005-TUN), Report No. 9054, October 11, 1990. 3. President's Report, Tunisia Health and Population Project, Report No. P-3063, May 1, 1981. 4. Staff Appraisal Report, Tunisia Health and Population Project (Loan. 2005-TUN), Report No. 3204, May 6, 1981.  5 Basic Data Sheet HEALTH AND POPULATION PROJECT (LOAN 2005-TUN) Key Project Data (amounts in US$ million) Item Appraisal Estimate Actual Actual as % of Appraisal Estimate Total Project Cost 41.0 19.8 48.3 Loan Amount 12.5 8.5 68.0 Economic Rate of Return (%) n.a. n.a. n.a. Cumulative Estimated and Actual Disbursements (amounts in US$ million) FY82 FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 Appraisal Estimate 0.9 3.8 7.9 10.6 12.5 8.5 8.5 Actual - 0.08 1.44 1.97 2.77 3.79 5.64 6.69 8.3 Actual as% of Appraisal - 2.1 18.2 18.6 22.0 44.6 59.8 71.6 97.6 Project Dates Date Planned Date Actual Identification 1977 June 1977 Preparation September 1977 March 4 1978 Pre-appraisal June 1979, October 1979 First Stage Appraisal November 1979 Second Stage Appraisal May 1980 Post Appraisal September 1980 Negotiation n.a. April 27, 1981 Board Approval FY79 May 28, 1981 Loan Signing n.a. July 15, 1981 Loan Effectiveness October 15, 1981 March 23, 1982 Loan Closing' December 31, 1986 December 31, 1988 Loan Completionb December 31, 1986 September 30, 1989 a. Loan Closing date was extended twice from original of December 31, 1986, principally to complete procurement of computers for the MIS, and x-ray equipment. b. Loan Account held open to complete disbursements 6 Staff Inputs (staff weeks) FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY88 Identification/ Preparation 75.7 18.2 0.1 Appraisal 81.2 60.3 Negotiations 5.3 Processing Time 4.7 7.0 25.3 Administration 2.0 .6 .9 1.3 3.4 1.0 1.4 n.a. Supervision 2.5 14.4 14.4 15.2 14.6 20.4 19.6 19.6 Mission Data Month/ No. of Staff Specializationc Performance Types of Year Persons Days Rating db Problems' Supervision 1 10/81 4 17 EC, PH, AC 2 M Supervision 2 07/82 3 17 EC, PH, AC 1 M Supervision 3 03/83 3 9 EC, PH, AC 2 M Supervision 4 09/83 3 6 EC, PH, AC I M Supervision 5 06/84 I 3 EC 2 MP Supervision 6 06/85 2 10 EC, PH 2 MP Supervision 7 11/85 3 10 EC, PH 2 TP Supervision 8 07/86 2 14 EC, PH 2 T Supervision 9 12/86 2 1 EC, PH 2 TF Supervision 10 09/87 3 14 EC, PH 2 TF Supervision 11 02/88 1 2 PH 2 0 Supervision 12 07/88 2 5 PH 2 0 Supervision 13 10/88 1 5 PH 2 0 c. Specialization: AC= Architect, EC= Economist/Operations Officer, PH = Public Health Specialist d. Performance Rating: 1 = problem free or minor problems, 2 = moderate problems, 3 = major problems e. F = Financial Problem, T = Technical Problem, M = Management Problem, P = Political Problem, 0 = Other 7 Evaluation Summary Introduction 1. The Tunisia Health and Population Project was the second World Bank Group operation in the population, health, and nutrition (PHN) sector in Tunisia and the first in the health subsector. It assisted the government in implementing the new basic health services (BHS) strategy as set out in the Fifth National Development Plan (1977-81). The strategy focused on integrated health, nutrition, and family planning services delivered through a decentralized and more cost-effective system. Objectives 2. The project had five components aimed at (a) strengthening the management capability of the Ministry of Public Health (MOPH) through the provision of supporting staff and services; (b) improving and expanding the basic health care delivery system and infrastructure by extending the physical infrastructure and equipment in eight governorates; (c) improving and expanding the scope, content and effectiveness of national education and communication programs in health, population, and nutrition; (d) improving the health training system and its infrastructure and training, upgrading and redeploying health personnel in the project area; and (e) providing technical assistance in support of all components. Implementation 3. As stated in the PCR, despite rapid and promising initiation of major components, several factors hindered implementation. The most important of these factors was the institutional rivalry between the MOPH and the National Office Family Planning and Population (ONPFP) that prevented the planned integration of family planning with basic health services. Technical discussions regarding BHS were soured by conceptual and ideological differences regarding the future responsibilities functions of the MOPH and the ONPFP in addition to personality clashes among management of the two organizations. The change of ministers at the MOPH in 1983-84 compounded the ongoing difficulties and caused profound delays in project implementation, as it was accompanied by criticism and reassessment of project goals and BHS policy, turnover of staff (including the Project Coordinator) in the Directorate of Basic Health Services, and dissolution of the original MOPH team. It also resulted in a one-year cessation of project supervision (see Figure 1.1). 4. In addition, the complexity and scale of the MIS component contributed to its ultimate failure, and the studies and the applied research program suffered from lack of supervision. The efforts of the borrower to obtain local expertise rather than utilizing more expensive foreign assistance limited training programs, made supervising civil works difficult, and reduced the use of consultants. Procurement was delayed due to cumbersome bidding and evaluation procedures and overestimation of borrower's capacity.' Slow disbursements were attributed to chronic tardiness in governorates' submission of documents supporting expenditures. 1. In the government's view it was not overestimation of borrower capacity, rather it came from difficulties related to the normal procedures followed by the public sector (Annex 3). 8 Results 5. Despite the circumstances, the project achieved its major objective of extending basic health care in the eight governorates. The project constructed and equipped 161 health posts and dispensaries simultaneously in the 8 governorates with only a year's delay. This construction represented an increase of more than 50 percent in the number of basic health facilities. In addition 57 staff houses and 7 vehicle maintenance workshops were constructed under the project, and an additional 103 existing health posts and dispensaries were renovated and/or remodeled. While this remarkable achievement in expanding the infrastructure most certainly had a positive impact on the health status of the population, it was difficult to measure the effects. The MIS which was supposed to monitor specific project indicators, never functioned as planned, and thus no results on project impact were available. The project also succeeded in strengthening the capabilities of the BHS and the orientation of the MOPH through internal reorganization and decentralization. 6. In spite of the fact that BHS was well-conceived and supported by beneficiaries, it encountered several difficulties: many centers were underutilized and overstaffed; bureaucratic inefficiencies caused long delays in the paying of staff salaries (with attendant loss of morale); and budgetary restrictions and procurement problems slowed the process of equipping health posts. Most importantly, the ONPFP, a well-staffed and well-funded vertical program to supply family planning services, successfully resisted integration into the BHS system.2 Thus, the goal of integrating family planning with BHS was not achieved due to failure to resolve the institutional dispute between the ONPFP and the MOPH.3 7. To bolster training infrastructure, the project constructed or remodeled and equipped six paramedical schools, developed curricula, and provided consultant services to prepare programs and seminars. Despite the fact that the project did not directly fund hands-on training, according to the PCR, it did stimulate much training activity, especially in the early years. Because the MIS component failed -as the system never became operational despite the fact that 30 computers were acquired-the original project indicators were never updated. The project monitoring database is poor, making it difficult to measure attainment of specific project goals or assess health service outreach. About half the new computers were used in the MOPH to process data collected from an inventory of health services. The other computers were used for patient registration at a major hospital, on a test basis in the DBHS in the Jendouba Governorate, within the planning and epidemiology units of the MOPH, and at the Center for Pedagogical Research and Training (CRFP). 8. Three studies were successfully carried out under the studies and applied research component on infant mortality, vaccinations and hepatitis. The study on infant mortality was published as a book that enjoyed wide circulation and discussion and greatly influenced policy in the health sector. Data was also collected for an inventory of health resources. Other planned research activities were not carried out. The PCR stated that achievements under the health 2. In the government's view, the measures were not taken to formulate the teams capable of providing the integrated services. 3. The PCR states that in addition to the unresolved institutional dispute, integration suffered further through the governments failure to specify arrangements for providing family planning services, after transferring the ONPFP to the new Ministry of the family in 1986, an agency with only a minimal budget and agenda. 9 education component were modest, as health education was of uneven quality and most health posts were without basic educational materials. Ratings 9. OED rates project outcome as satisfactory. The component to strengthen basic health care services was successfully implemented. Accessibility to basic health services improved owing to the construction or renovation and equipping of over 264 health facilities in underserved rural areas. The construction completed under this project represented an increase of more than 50 percent in the number of basic health facilities at the time of project appraisal. Maternal and child health (MCH) services were successfully integrated into basic health services, with the exception of family planning services. At the time of the audit, integration of family planning services with MCH services had taken place. The services have not yet been evaluated for efficiency and quality and can undoubtedly be improved. The training component was successful. The management improvement component had modest success, but one of the three research studies produced on infant mortality was very influential in later health policy, including the 1991 Population and Family Health project. 10. OED rates sustainability as likely, confirming the PCR rating. While the PCR rated institutional impact as negligible, the audit upgrades the rating to substantial, based largely on the fact that eight years after project closure, the institutional framework it helped put in place for district basic health services is still operational as the national model. The position of the DBHS has been strengthened, and the decentralization of authority for Basic Health Services to the regions is being continued under the 1991 Population and Family Health project. In addition, the service delivery methods developed under the project have become standardized across all governorates. 11. Overall, the audit agrees with the satisfactory ratings given by the PCR for borrower and Bank performance. Neither the borrower nor the Bank paid sufficient attention to the political and institutional impediments of the Government of Tunisia's (GOT) integration policy. The result was that family planning services were not integrated with BHS as planned under the project. With this exception, both borrower and Bank performance were very good with regard to project design and implementation until 1984, when institutional changes in the MOPH interrupted the momentum of the project. Despite the ensuing delays, the borrower met the project conditionalities and the project was successfully completed. Lessons 12. The lessons from this project are as follows: (a) In order to ensure the feasibility ofproject goals, it is essential to examine the existing health system priorities, the political and organizational factors, the government's long term policy goals, and the process envisaged for meeting them. The integration of health and family planning services envisaged in this project-a major policy change-did not occur. The parties concerned did not fully understand the policy of integration and therefore did not respond well to it. This project needed to give more attention to differing priorities (including conflicting views between external aid agencies); to place greater emphasis on rational, mutually agreed collaborative arrangements; to develop an 10 implementation time table; and to delegate the authority to take remedial measures. (b) Monitoring and evaluation systems are vital to measuring the success ofPHN projects. While the project did seek to implement a thorough MIS, the design ultimately proved to be too complex, owing to the chosen computer technology and the desire to do everything at once. Thus, the project had no mechanism to record achievements in component to improve basic health services. (c) The Bank should not put undue reliance on technical assistance. Tunisia, like many countries, has been reluctant to borrow for this purpose, preferring to obtain such assistance free of charge from other donors and agencies such as USAID, WHO, and UNICEF. Given the experience of the First Population Project, the Bank should have explored alternatives to technical assistance, which was seen to be crucial for the success of this project. (d) It is essential that procurement procedures are documented at the outset ofa project and/or the Bank work carefully to establish acceptable bidding and evaluation procedures. In this project, the Bank overestimated the procurement capacity of the MOPH and procurement progressively became the single greatest practical impediment to project implementation, as well an inordinate drain on supervision time.4 (e) Projects of massive scope and scale seeking to attain complex objectives simultaneously should be avoided in favor of a series o simpler, phased efforts which address logical steps in a longer-term program. This is especially true where institutional capacity is weak and vulnerable to political pressures. 4. The government has suggested that the problem was not capacity but overly cumbersome administrative procedures. 5. It is the opinion of the government that the scope of the project is not the main handicap if a good program of continuous evaluation is established, using all the components of the project during its execution, so as to be able to intervene on time in order to make the necessary corrections. The SAR identified as a major project risk the fact that the scale and design of the project were too difficult for effective MOPH coordination and implementation, an assessment which proved to be correct, despite actions taken or suggested to countervail their occurrence. 11 1. Sector Background 1.1 Since Tunisia became independent in 1956, the government has emphasized the development of the country's human resources and, in particular, the health of the population. Health expenditures averaged 2.5 percent of GDP from 1956 to 1980, contributing to impressive gains: life expectancy increased from 41 years in 1956 to 57 years in 1978, the crude birth rate fell from 49 to 32 per thousand, and infant mortality declined from 202 to 123 per thousand. Given its resources and potential, however, there were still improvements to be made at the time this project was identified. Health Status 1.2 In general, the status of the Tunisian health sector in 1980 reflected the emphasis of its public health policy on hospital-based curative medicine. This orientation, typical of industrialized countries, gave insufficient attention to preventive care and did not provide sufficient coverage of the 50 percent of the population living in rural areas. The system resulted in sharp contrasts in the health status of the urban and rural populations. For example, in 1978 infant mortality was 170 per thousand live births in rural areas compared with only 70 in urban areas. Access to prenatal and postnatal care was lacking, especially in rural areas, where most births went unattended. Malnutrition of both mothers and children increased the susceptibility of infants to disease. About half of all deaths in Tunisia in 1979 were of children under five, and about half of all child deaths occurred during the first year of life. In addition, morbidity associated with common diseases was high, due to insufficient coverage of the population by the health care delivery system. Poor health and nutrition resulted in high dropout and repeater rates in primary schools and contributed to the low labor productivity, especially in agriculture. The existing clinical infrastructure was underutilized, particularly in rural areas, due to the limited range of services and medication available and the limited qualifications of staff. Tunisia's Family Planning Program 1.3 The Tunisian family planning program, initiated in 1964, was the first in Africa. It offered free contraceptive advice and services and included provision for sterilization (tubal ligations) and social abortions for women with at least five living children. Considering the time and the social and cultural setting of the country, this policy was strongly progressive. In 1973, the National Office of Family Planning and Population (ONPFP) was created. While under the tutelage of the Ministry of Public Health (MOPH), the ONPFP acquired the legal status of a parastatal organization. The ONPFP formulated a family policy and introduced a wide range of programs, particularly in the areas of information, education, personnel training, research and evaluation. It provided services through its own facilities, but also those of the MOPH, especially the latter's Maternal and Child Health (MCH) Centers. The lack of clarity in the division of work between the ONPFP and the MOPH and the uncertain status of the family planning personnel hindered cooperation between the two entities. 1.4 In spite of some opposition to abortion and sterilization, family planning enjoyed broad political, religious, and legislative support, and was able to achieve a dramatic transformation in family planning knowledge, attitudes and practice. The number of family planning acceptors increased almost tenfold between 1966 and 1978. However, ONPFP services were largely 12 confined to the urban population; rural women accounted for less than a third of the clients. A slowing in the rate of increase of new acceptors serviced by the public sector program was noted during the period 1977-1980 and was attributed to logistical problems within the ONPFP's service delivery structure. After having largely satisfied urban demand, the ONPFP faced the need to expand services to rural areas, where demand had been stimulated by well-designed information and education programs. Integration of Family Planning and Health Services 1.5 Beginning in 1975, the government moved slowly toward a policy of integrating its health services with a series of pilot programs in various governorates (1975-1978). These programs combined preventive and curative health services, family planning and health education. The government permitted the ONPFP to retain its established delivery system as a practical way of maintaining momentum in the family planning program pending the outcome of those efforts. In 1977, a Presidential Decree was issued that called for integrated regional delivery of curative and preventive medicine, and family planning and nutrition activities and called for the ONPFP to divest itself of its field activities and concentrate on research, policy formulation, and the training of specialists in its own field. Routine delivery of family planning services would be managed by the MOPH. However, neither the Decree nor the Fifth National Development Plan (1977-81) which endorsed it provided any concrete proposals for implementation of the integration policy. A preparation mission noted that "there remains much ambiguity at the MOPH and ONPFP about the structural and administrative implications of the proposed concept of 'integration' of services", indicating the Bank's recognition that this structural problem was central to population policy.' 1.6 The government's integration policy was aimed to change the balance of the public health system as reflected in the Fifth National Development Plan, which gave priority in the health sector to basic health care and preventive medicine, improved cost effectiveness and better regional distribution of facilities and personnel. Integrating ONPFP and MOPH services would serve these ends. The proposed integration system also made financial sense, as the government was not prepared to finance a costly, parallel system of family planning service delivery in rural areas. Decentralization 1.7 The government also hoped decentralization of health, family planning and nutrition services would strengthen the public health system. The MOPH began decentralization by delegating increased authority to the six medical regions. In addition, the MOPH began the process of separating the management, administration, and the budget of hospital-based medicine from those of the peripheral basic health services. At the governorate level, Regional Directors of Health (RDHs) were created to coordinate and control all health activities, including budgeting matters. A Director of Basic Health Services was appointed, and in each region specially appointed physicians (Chefs des Services de Sant6 de Base) were responsible for integrating all basic health services under the Director's supervision. 1. Internal Bank document dated April 3, 1978. 13 Bank Activities in the Sector 1.8 In 1971, the World Bank extended a credit (Cr. 0238-TUN) to Tunisia for the First Population Project. This project aimed to expand and improve the family planning program through construction of maternity hospitals and maternal and child health centers and the provision of technical assistance. The project was first administered by the MOPH, and later, from 1974, by the ONPFP. Relations between the ONPFP and the Bank steadily deteriorated during project implementation. The Project Performance Audit Report concluded that while the Bank had a role in helping mold the status of the ONPFP, the project failed to achieve its institution-building goals. As explanation, the audit suggests that the Bank focused on construction and: ...abandoned its institutional interest in the ONPFP.....None of the supervision missions addressed the problem [of linkages] with concrete suggestions. In the atmosphere of tensioa that gradually developed between the MOPH and ONPFP, the Bank was caught in the middle and was unable to overcome the personality and political constraints which ensued.2 1.9 The outcome of the population project was rated as unsatisfactory. It was scheduled for completion by the end of 1975 but was not completed until 1981. As a result of the delays in implementation and price escalation, the total estimated cost increased to about US$35 million, up from the initial appraisal estimate of US$7 million. Additional financing was provided by the Norwegian Development Agency (NORAD) and the government. The failure of the First Population Project and the deterioration of relations between the Bank and the ONPFP resulted in a break in continuity between the First Population Project and the population component of the 1981 projects. Unresolved issues of institutional and programmatic integration between the ONPFP and the MOPH continued to constrain program effectiveness as the project was identified. The Integration Debate 1.10 At the start of the 1981 Health Development Project, the MOPH lacked the continuity, commitment, infrastructure and management skills to take up additional operational responsibilities for family planning delivery, and the ONPFP saw itself as filling the vacuum by expanding its own parallel delivery system despite the inefficiencies resulting from duplication of facilities, high unit costs, and the inability to cover rural areas adequately. These tensions over relative roles reflected, and perhaps fueled, a larger debate on the relative merits of service integration. The ONPFP, supported by USAID and UNFPA, was sensitive to the possibility that the integration process would slow the progress being made with the vertical approach. 1.11 The situation in Tunisia was a microcosm of the worldwide debate occurring among family planning practitioners, policy makers, and donors regarding integration. The 1971 Bank- financed Population Project reflected the view prevalent in the 1970s that vertical programs and hospital- and clinic-based postpartum programs were the most effective method of recruiting new family planning acceptors. But subsequent experience in Tunisia and other countries suggested that this approach was costly and could reach only a small proportion of the rural 2. Project Performance Audit Report, Tunisia First Population Project (Credit 238-TUN), Report No. 5135, June 21, 1984, paras. 38 and 48. 14 population. Integrationists further argued that health services had to be offered in conjunction with maternal and child care services and family planning to attract a larger proportion of the rural population. A Bank Sector Review in 1975, "The Population Program of the Government of Tunisia," recommended that family planning services be further increased and that they be provided by MOPH field staff under ONPFP supervision-that is, the services should be integcated. 1.12 The Bank's position throughout preparation of the 1981 Health and Population Project was that the government's strategy and program for developing the family health care system, based on the establishment of a network of basic health centers, should be clarified. It requested conceptual clarification of "integration" and of its operational implications for family planning and the ONPFP. However, at the beginning of the project, fundamental questions of organizational responsibility for health remained unresolved. The three key donors were also at variance on the issue of integration, a situation that did not bode well for future collaboration. Subsequent Bank Activities in the Sector 1.13 The Bank undertook a sector review, "The Demographic Challenge," in 1990 which was followed in 1991 by the simultaneous introduction of two projects, the Population and Family Health Project (Loan 3307) and the Hospital Restructuring Support Project (Loan 3308). The Population and Family Health Project pursued related goals to the 1981 Health Development Project. Its objectives were to assist the government to lower both fertility and mortality by targeting basic health care services to underprivileged groups with a strong focus on mothers and children. The project aims to reduce regional disparities in access to basic health care and in the availability of resources and improve the quality of basic health care, including family planning/maternal and child health services. One of the project's four components seeks to integrate FP/MCH services in BHC facilities. The 1991 Hospital Restructuring Support Project aims to support the government in improving the efficiency of major hospitals by containing costs while improving quality of services, and providing the information to permit adjustments in financial burden-sharing by better linking utilization of hospital services to financial contributions. Figure 1.1 details the evolution of Bank activity as well as key changes in the health sector in Tunisia. Tunisia Health Sector Activities Timeline Exterior Changes 1964 The Ministry of I ieali was replaced Tunisian Family twice in three months from October 1o Plannig ____________________ 1983 to January 1984, with attendant196E Planning 18 Program (TFPP) 1973 turnover in the Director of Basic lalth ONPFP transferreu E initiated as the National Office of Family Sevices, Project Cocidinator, and from MOPI to new first in Africa Planning and Population dissolution of he MO I team. of (ONPFP) created as a semi- autonomous organization 1984 1987 under the MOPHI. Project Coordinator left MOPHI ONFP transferred in January and replacement not back to MOPH named until 1985 1 64 1971 19/ 1975 1978 1980 1981 1983 1984 1985 86 1987 1988 1990 1991 1996 1981 1 1971 1980 1lelth & 1988 1990 1991 January 1996 Popultion Ban Repot: 1986 I Pouation \98 [Health & Bank Sector Population & [ roect1 E Proect"Tnisa: ocal lelt & op rom M Po e inPopulation Review The Family Health Identiication Project Cordna ojet OP ONFPpranserre PCrojc 23Tuniesia ocia Projec Prjet005ppaie End Date Demographic Project for I ealth (Crv23)lApet"n 2005) hut blocked by MOPII Challenge" (Ln. 3307) Sector Project 1975 Dvlpeidue to concerns about (Rpt. 8903) 19 Bank Sector the achievements 1987 Review (Rpt. 651 under the first project A Third Health & 1990 Hospital "The Population and the recurrent cost Population Project Health & Restructuring (mplications of the was identified but set Populatioc Project Program of the BIDpS policy aside by the Ministry Project PCR (L.n. 3308) Government of th a Tunisia" of Planning in favor of 1984 health sector work Population emphasizing services Project PPAR and finance Evolution of Baik Activity 16 2. Project Objectives and Content 2.1 The 1981 Health Development Project (Loan 2005-TUN) was designed to integrate basic health services delivery and to provide 90 percent coverage to the project population (eight governorates) in the following areas: family planning services; immunizations; preventive and curative services for children; general curative services, and hygiene and sanitation services among the rural population. These services were to be delivered using simple and economic health facilities, paying special attention to problems of management and training. 2.2 Based on the achievements of a pilot program in Medjez-el-Bab, the project design called for specific outcomes in the eight governorates: (a) reducing infant mortality to about 80/1000 from an estimated national average of 123/1000 considered representative of the project area; (b) lowering the birth rate from about 43/1000 to 36/1000; and (c) reducing morbidity from tuberculosis, other respiratory diseases, and parasitic and viral infections. These objectives, which were to be achieved over a five-year period once the project had become fully operational, were seen as ambitious but attainable if the project services operated as planned. Project Components 2.3 The project had five components: (a) strengthening the management capability of the Ministry of Public Health (MOPH) through the provision of supporting staff and services; (b) improving and expanding the basic health care delivery system and infrastructure by extending the physical infrastructure and equipment in eight governorates; (c) improving and expanding the scope, content and effectiveness of national education and communication programs in health, population, and nutrition; (d) improving the health training system and its infrastructure and training, upgrading and redeploying health personnel in the project area; and (e) providing technical assistance in support of all components. 2.4 Management Capability. The project proposed to develop a more efficient management information system (MIS) for the MOPH to permit better planning and control of resources. The existing system processed data at governorate and national levels, but little of the large amount of data collected was analyzed because the information was considered unreliable, and there was no effective feedback at any level. The basic features of the improved MIS were to be (a) standardized collection forms covering a limited number of basic health indicators; (b) systematic data collection by MOPH personnel; (c) the expansion of statistical and supporting staff; and (d) computerized data handling (including new programs) using the central computer of the National Computer Center and decentralized regional terminals. 2.5 In addition, the project included a program of studies and applied research to identify through experience improved ways of meeting health needs in rural areas and the cost and relative impact of essential services. The research program would be designed by the Directorate of Basic Health Services (DBHS) and the Center for Pedagogical Research and Training (CRFP). Several research topics were identified, including (a) the costs of preventive and curative services at the periphery; (b) family planning research into behavioral patterns; (c) management issues such as communication within the MOPH and interagency coordination and cooperation at central and local levels, as well as basic indicators for local health planning and drug distribution at the local level; and (d) the comparison between the impact of population, health, and nutrition 17 services in a project governorate and in one outside the project. To support these activities, the project would provide equipment and materials, vehicles, and external and local technical assistance. 2.6 Basic Health Care Services. The project selected the medical regions of Sousse and Jendouba (comprising seven governorates) and the governorate of Zaghouan for the reinforcement of primary family health services. The selected governorates, with the exception of the areas surrounding Sousse and Monastir, were mainly rural and underequipped. In both regions, the findings of earlier primary health care pilot projects were taken into account in the design of the project. Health, family planning, and nutrition services would be delivered by teams of public health nurses, general physicians, and midwives, who would be assisted by nutritionists, nurse-supervisors, health educators and sanitary technicians. In rural areas, each team would serve an average of ten dispensaries on a regular schedule. The system would also provide an adequate supply of drugs and contraceptives at the peripheral level. To support this component the project provided for construction and remodeling/expansion of health posts, rural and urban dispensaries, staff houses, and vehicle maintenance workshops. The project also provided equipment and vehicles. 2.7 Health, Population, and Nutrition Communications. The communications component was designed to strengthen and develop the population, health, and nutrition activities of the ONPFP, the MOPH, the National Institute of Nutrition and Food Technology (INNTA) and the Ministry of Education. A working party was to be established by the MOPH to draw up and coordinate a communications development program to help field workers do a better job of motivating families to adopt improved health, family planning, and nutrition practices. The members of the Communications Working Party would organize in-service training courses for the field workers of all four agencies. The project provided for the construction of and equipment for two regional health education centers to be used by the working party as training and retraining centers and as centers for the production of simple visual aids, as well as for storage and distribution. The Central Health Education Center in Tunis would be expanded to produce more effective exhibition matter and additional types of educational visual aids. Equipment, vehicles and material were also to be provided. 2.8 Training. This component financed in-service training for middle-level health managers and statisticians, paramedical personnel and their teachers, health educators, and physicians in the project area. It also financed curriculum development for training of physicians and paramedical personnel, provided for the construction/remodeling and equipment for six paramedical schools, vehicles and local and external technical assistance, short-term scholarships for training abroad, and internal travel expenditure. Relevance 2.9 The project addressed the government's goals for the sector as laid out in the Fifth National Development Plan (1977-81), which itself appropriately focused on improving family planning and MCH services and strengthening the public health system. In the context of Tunisia's epidemiological status and demographic objectives, the project is therefore judged to have been relevant. However, on review of documents, it appears that although the integration of public health services, including family planning, was a clearly stated broad policy objective, the two concerned implementing agencies did not share a common vision of how integration was to 18 be achieved. The SAR states that: The Government has decided that all public health services, including family planning, will be provided by MOPH staff through MOPH channels, and that the ONPFP will therefore confine itself to its original functions of policy formulation, research, education and training. The integration of health and family planning activities, already underway in the medical regions of Sousse and Jendouba, is essential if the government is to extend its family planning program effectively throughout rural Tunisia.3 2.10 Thus, while the project's objectives were designed based on the government's integration policy, it is doubtful that the implementing agencies were sufficiently committed to this objective. The Bank recognized that the implications of the integration policy for project implementation had not been frankly faced within the government and noted that implementation could eventually have a profound effect on the structure, role, and function of the ONPFP. While the Bank saw integration as an internal issue to be decided by the government, it hoped that the project would add momentum to the process, as integration was essential for the project. In retrospect, the objectives of the project were clearly relevant in the demographic sense, but their fit with current organizational capacities was seriously flawed.4 3. Staff Appraisal Report, Tunisia Health and Population Project (Loan 2005-TUN), Report No. 3204, May 6, 1981, para. 1.15. 4. The region believes that while the above may be true, the project was indeed successful at creating momentum for reform, since now, fifteen years later, integration is close to being generalized. 19 3. Project Implementation and Results Overall Implementation 3.1 Project implementation can be divided into two time periods: 1981-1984 and 1985- 1988, before and after changes in the Ministry of Public Health. MOPH staff at all levels were highly motivated and committed to project objectives, and project implementation thus started in July 1981 in an optimistic climate. Project execution was also facilitated by the fact that the Project Coordinator simultaneously held the positions of Director of Basic Health Services as well as the political position of Chief of Staff for the Minister of Public Health, a situation that prevailed until 1984. After 1984, project implementation was hampered by political changes within the MOPH (the appointment of two Ministers within a three month period, October 1983- January 1984), the general economic crisis (bread riots which broke out in January 1984), and management problems (the Project Coordinator left and was not replaced for over a year).5 The results were paralysis in the MOPH due to political and financial problems; a halt in the momentum in project implementation; and a hiatus in project supervision of one year. The dialogue between the borrower and the Bank suffered. The project had two one-year extensions, mainly to complete procurement of computers for MIS and x-ray equipment. The project was completed in December 1988, and the loan account was kept open to September 30, 1989 to complete disbursement. Disbursements 3.2 The second supervision mission in July 1982 reported that no disbursements had taken place, though they should have reached $0.9 million by June 1982. By February 1983 disbursements were still slower than expected-$56,000 by March 31, 1983 vs. $1.86 million in the revised disbursement schedule. This chronic slow disbursal was attributed to administrative procedures involving regional authorities, including low incentives for submission of applications for withdrawal among regional authorities. In September 1983, arrangements were agreed with a Bank supervision mission for the Project Coordination Unit (PCU) to accelerate disbursements, improve and simplify the bids/evaluation procedure, and speed up the delegation of credits to government authorities. In June 1984, however, disbursements were still only US$1.4 million versus the revised disbursement target of $3.4 million and the appraisal estimate of $7.9 million for this period. Other factors contributing to slow disbursements included the MOPH's reluctance to use project funds earmarked for technical assistance;7 chronic delays in awarding contracts for goods; and dramatic appreciation of the US dollar (about 80 percent). At closing in December 1988, disbursements were US$5.6 million. In the end, US$4.35 million was canceled. 5. The borrower is correct in observing that while the changes within the MOPH were an important factor, they were not the only factor. 6. Preparation of a follow-up project during the same period suffered as well. 7. The borrower states that it is rather due to difficulties related to the procedures for the recruitment of the consultants that are followed in the country. 20 Procurement 3.3 In October 1983 a supervision mission reported lengthy (nine month) delays in the evaluation of bids for the procurement of equipment and furniture. These delays were due to the shortage of procurement specialists in the PCU as well as the MOPH, and cumbersome administrative procedures. To further complicate matters, the MOPH halted procurement in 1984 in order to re-examine its procurement and standardization policies, further slowing disbursements. The PCR states that the Bank overestimated the procurement capacity of the MOPH, and that the PCU was unable to expedite or effectively coordinate procurement, which became the single greatest practical obstacle to project implementation, as well as an inordinate drain on Bank supervision missions. It was also the main reason for the two one-year extensions of the project. Management Capability 3.4 Management Information System (MIS). The MIS was to remedy deficiencies in the collection and analysis of health statistics. Until March 1983, development proceeded on schedule. A technical report prepared in 1982 provided a critical evaluation of the existing system of data collection and information flows, described the objectives of the new MIS, and laid out performance and implementation indicators including activity levels, outreach and performance, impact and recurrent costs. However, technology improvements caused the National Computer Center (CNI) to make proposals for a new configuration-eight independent microcomputers instead of eight regional terminals linked to CNI. The new proposal entailed 50 percent higher investment costs but significant savings in terms of operating expenditure. The new proposal, however, was not supported by the MOPH or the Bank due to the increased costs. After the new Minister of Public Health was appointed in 1984, the MIS was put on hold as the MOPH was uncertain of the direction it should take regarding the system. A new proposal was approved by the MOPH and the Bank in 1986, and bids were launched for the computer equipment in early 1987. Due to procurement delays, 30 computers were acquired only during the last months of the project. 3.5 The MIS was never implemented. The PCR noted that half of the computers were being used by the MOPH to process data collected from an inventory of health services (Carte Sanitaire), not service utilization data. Other computers were used for (a) patient registration at a major hospital; (b) on a test basis in the DBHS in Jendouba Governorate; (c) within the planning and epidemiology units of MOPH; and (d) at the Center for Pedagogical Research and Training (CRFP). The PCR states that the indicators developed for the MIS were never collected, thus there was no data to evaluate the project outcomes in terms of reaching its basic health goals. The fact that the last three supervision reports from 1988 are missing from the files (and are presumed lost due to the Bank reorganization) rendered it impossible to determine what became of the indicators developed for the MIS and to ascertain why data was not collected against these indicators. 3.6 Studies and applied research. The program of studies and applied research, designed to identify better ways of meeting health needs in rural areas and the cost and relative impact of 8. The missing reports and related files also rendered it difficult to determine what happened with the studies and applied research and health education under the health, population and nutrition communications component (paras. 3.7 and 3.25). 21 essential services, was slow to start and suffered from lack of supervision. In early 1983 an ad hoc working group was established to identify and prepare an inventory of research topics in BHS, orient future research, identify consultants, and award and supervise contracts. In March of the same year, at a national seminar, research priorities were discussed and five subject areas were chosen, and additional research areas which were supported by WHO were included. These research priorities included tuberculosis; malnutrition; communicable diseases; improvements of records and statistics; accessibility and utilization of services; drinking water; school health; maternal mortality; child morbidity and mortality; and acute respiratory infections. As of June 1983, nine research proposals had been reviewed but research grants had not been awarded as the ad hoc working group was absorbed by other activities and did not function as planned. 3.7 A supervision report in September 1983 reported that applied research was being carried out in the field in three governorates, however, funds from the project were not being utilized, reportedly because of the formalities involved. The June 1984 supervision report recommended that the next mission focus on this component, but the planned supervision did not take place, and there was a year's hiatus. The report of the next supervision mission in June 1985 stated that little progress had been made on the software side (MIS, applied research studies, and health education) due to (a) changes in senior MOPH staff; (b) departure of the project coordinator; (c) failure of new staff to take critical decisions; and (d) a breakdown in relations with the Bank. With the appointment of a new coordinator later that year, communications with the Bank were back to normal; however, little progress was made with the applied research component. It was already several years behind schedule. It did not produce applied research as envisioned. However, three studies were completed under this component, on infant mortality, vaccination and hepatitis. The national study on infant morbidity and mortality was published as a book, "Un Enfant et Deux Tunisies"9 that has had considerable impact on policies in the sector and was used extensively in the preparation of the 1991 Bank-assisted Population and Family Health Project. 3.8 Health Planning. After some delay, due in part to the Tunisians' reluctance to use technical assistance funds, a health sector planning advisor was recruited in September 1983 for two months and assisted the Directorate of Basic Health Services (DBHS) in designing and implementing a methodology to identify specific interventions at the regional level. This work was aimed at the governorates to be included under the Second Health Development project that was under preparation but was to be expanded to all governorates under the proposed project. This project, the Second Health and Population Project, was appraised in December 1984 but was subsequently blocked by the new Minister of Health due to concerns about the achievements and the recurrent cost implications of the BHS policy. A Third Health and Population Project was identified in 1987 but was set aside by the Minister of Planning in favor of sector work emphasizing services and finance. 3.9 In terms of strengthening the MOPH's capacity for policy formulation, planning, management, and evaluation, the project was not very successful. The reliance on the MIS for policy formulation, management, and evaluation was a poor design choice, as it was too narrow an instrument to accomplish all the goals. In addition, its sheer complexity did not bode well for its smooth operation. The reliance on technical assistance for the services of a health planner was 9. Gueddana, N. Ben AYcha, N. et Jarraya, S., February 1989, "Un Enfant et Deux Tunisies, Rdsultats d'une Enquete Nationale sur la Mortalit6 et la Morbidit6 Infantile, Indices de Mortalitd, D6terminants, Causes Mddicales des D6cks, Stratdgie pour une Meilleure Survie," R6publique Tunisienne, Ministare de la Sant6 Publique, Tunis. 22 also unwise as Tunisia was known to be reluctant to pay for such services, preferring to obtain them free of charge from other donors. The studies and applied research component might have been rescued with better supervision. Basic Health Care Services 3.10 Decentralizativn. Even before loan effectiveness, important changes were made in training and organization and budgetary allocation (method and level) within the MOPH. In June 1981 the MOPH required management staff to provide a breakdown of 1980 expenditures between hospitals and basic health services as a step in the preparation of the 1982 budget. In addition, the MOPH's decentralization policy delegated budgetary, technical, and administrative responsibilities to regional authorities, including training. In 1983 a supervision mission reported that the MOPH's decentralization policy had led to significant and rapid progress in strengthening basic health services. A Deputy Director of BHS was appointed in February 1987, further strengthening the DBHS team. 3.11 Organizational Changes. At the outset of the project, BHS consisted essentially of health facilities scattered throughout eight governorates, with little cohesion. They provided minimum, low-standard medical care. Beginning in 1981 with the project, the MOPH appointed well- trained and experienced staff in key technical and administrative positions for the development of its integrated health development system and implementation of the project. Within the first eighteen months of the project, BHS structures were strengthened with the addition of staff, staff were trained, and norms and instructions for medical activities were issued. The number of public health physicians in the project area practically doubled, the health budget was substantially increased, and separate budgets were prepared for BHS in all medical regions. These budgets included all expenditures except salaries which continued to be administered at the centrally. Additional vehicles, office accommodations, and equipment improved working conditions. 3.12 Civil Works. Price increases for cement and labor resulted in higher unit construction costs than estimated at appraisal that was partially compensated by regrouping contracts. In addition, to stay within the limits of the financial envelope agreed upon at appraisal, the MOPH decided in 1981 to accelerate implementation and to reduce the number of new facilities to be 10 built from the 265 planned at appraisal to 225. The first phase of the construction program was launched in March 1982, all sites were acquired by July 1982, good bidding documents were prepared, and almost all contracts were awarded by February 1983. New construction was given priority over remodeling and/or extensions which began in May 1983. The number of facilities to be remodeled and/or extended was reduced to 103 from the 288 identified at appraisal. Despite the fact that construction was begun in eight governorates simultaneously and there were recurrent problems with reimbursement of travel allowances which hindered supervision, construction proceeded smoothly with only a year's delay. The positive results achieved in the implementation of the construction program can be credited to the efficiency of the PCU, the project architect and the two construction supervisors. 3.13 Integration. Regarding the new policy of integration, the SAR noted that sustained government efforts would be required to ensure full cooperation of the ONPFP, which was 10. Plans for the construction of II facilities were canceled while the remaining 29 were postponed-to be financed in subsequent years by government. 23 accustomed to a high degree of autonomy, in the integrated delivery of services that it formerly 11 provided separately. Progress in defining services, training staff and mobilizing additional resources in family planning was limited, in contrast to the progress that was being made in other basic health services under the project. While the MOPH relied heavily on ONPFP cooperation for the effectiveness of the new DBHS and its regional representatives in establishing the integrated basic health services, it was unable to promote a working dialogue free of personal and political rancor. The ONPFP felt no incentive to take the initiative given that it believed, genuinely, that its work and existence were at stake. The sustained technical and logistical support needed by the DBHS from the ONPFP to reinforce the former's ability to provide family planning services within the basic health services framework thus did not materialize. Relations between the MOPH and the ONPFP deteriorated progressively in the first 18 months of implementation with adverse operational effects. Efforts by the Bank to establish dialogue between the two organizations proved futile. Donor agencies also became embroiled in the conflict, on the side of ONPFP and advised the Bank to leave ONPFP alone, as the Bank was seen as siding with the MOPH. In 1984, Bank staff noted that: It is evident that there have been major differences of opinion, to the point of animosity, on these issues within the Tunisian Government, between the Government and the Bank, and also among the donors. Everybody apparently was squabbling with each other!! The question. in Tunisia's case, is what is the Government's policy'? Although Tuni,ia is said to be pursuing an integrated approach, it seems that powerful elements within the Government are fiercely resisting its adoption. 1 3.14 Family planning services continued to be delivered by the ONPFP. The MOPH was unable or unwilling to work with the ONPFP on this basis. In 1986, the ONPFP was moved to a newly created Ministry of the Family and Protection of Women but was moved back under the MOPH the following year, and a new director was appointed. Despite these changes. neither the ONPFP's legal status, the conflict with the DBHS, nor the integration schedule was settled. The political and institutional environment which prevailed was not conducive to integration. It was reported that some integration of family planning services with basic health services ocrr d. in the field, yet the institutional squabbles prevented any progress in integration at the central level, i.e. agreement over an implementation plan and a timetable. Thus, the population focus of tht project was lost, and the project did not succeed in fully integrating family planning and NICH services. In 1991 it was reported that such integrated services were being provided in only !6 percent of health facilities, com ared to 55 percent of clinics receiving family planning services from the ONPFP though visits. 3.15 Service Delivery. The project constructed 67 health posts, 79 rural dispensaries, and 15 urban dispensaries-a total of 161 new health facilities, representing an increase of more than 50 percent in the number of basic health facilities at appraisal (287). An additional 103 existing health facilities were renovated and/or remodeled, and 57 staff houses, and 7 vehicle maintenance workshops were constructed. While no figures are available for the total lumber of I1. SAR 198 1, para. 6.04. 12. Internal Bank memo dated February 27, 1984. 13. Staff Appraisal Report, Tunisia Population and Family Health Project (Loan 3307-TUN). February 19, 1991, Report No. 9128, para. 1.23. 24 basic health facilities in the eight governorates in 1988 at the time of project completion, the audit mission was able to collect such data for 1993 (627).14 These data show that the construction completed under the project represented 26 percent of existing basic health facilities in 1993. This remarkable achievement in expanding the infrastructure most certainly had a positive impact on health services delivery. However, the impact of the project is difficult to measure as the MIS that was supposed to monitor specific project indicators never functioned as planned, and no results are available about the project's goals of increased coverage, reduction in IMR, CBR and morbidity. 3.16 In spite of the fact that BHS was well-conceived and supported by beneficiaries, several difficulties were encountered: many centers were underutilized and overstaffed; bureaucratic inefficiencies caused long delays in the payment of staff salaries with attendant loss of morale; and equipping of health posts was very slow due to budgetary restrictions and procurement problems. 3.17 The PCR states that the borrower reported a doubling of the rate of medical consultations (from 15 to 30 percent) across the project areas, though the PCR provides no figures indicating if the coverage goal was met-from 50 percent of the population to 90 percent. The audit mission was able to obtain some data. Table 3.1 shows the evolution of the density of basic health centers (BHCs) in the eight governorates from 1982-1986, where density is defined as the ratio of population per BHC. The figures illustrate that coverage did improve though it is unclear exactly how much can be attributed to this project.15 The ratio of population per basic health center decreased significantly, with improvements ranging from 42 percent in Zaghouan, 39 percent in Jendouba, but only six percent in Monastir. Table 3.1: Population per Basic Health Center in the Eight Project Governorates, 1982-86 1982 1986 % Improvement B6ja 5,853 4,622 21 Jendouba 9,063 5,517 39 Kairouan 9,113 6,458 29 Kef 5,196 3,947 24 Mahdia 4,533 4,043 11 Monastir 4,940 4,662 6 Sousse 6,285 4,886 22 Zaghouan 6,290 3,620 42 Source: Carte Sanitaire-Plan des CSB, 1986, MOPH. 3.18 As stated above, the project did not achieve integrated family planning and basic health service delivery. However, family planning services, prenatal care, and vaccinations were provided in most basic health facilities in the project governorates. Table 3.2 provides a snapshot in 1986 of the family planning, prenatal, and vaccination consultation activities in BHCs. In Mahdia in 1986, only seven percent of BHCs offered family planning, versus the high of 82 percent in Beja. In terms of prenatal consultations, again a very low figure is found in Kairouan, 14. Source: MSP/DSSB/Evaluation Database (funded by the 1991 Population and Family Health Project), November 1995. 15. Other facilities were constructed during the time of the project, financed by the government, the Bank-financed Rural Development Program, and USAID. 25 seven percent, as opposed to rates between 42.5 and 82 percent in the other governorates. Most centers offered vaccinations, however. Table 3.2: Percent of Basic Health Centers Offering Family Planning, Prenatal and Vaccination Medical Consultation Activities, 1986 Family Prenatal Vaccination Planning B6ja 82.0 82.0 93.5 Jendouba 78.0 53.0 92.5 Kairouan 67.0 7.0 81.5 Kef 67.5 67.5 84.5 Mahdia 7.0 72.0 85.0 Monastir 60.5 60.5 81.0 Sousse 42.5 42.5 83.0 Zaghouan 60.0 65.5 88.5 Source: Ibid. 3.19 In 1993 a DBHS study examined the volume of preventive and curative services, the availability of resources and supervisions, and the integration of MCH/FP services with BHS for a random sample of BHCs across the country.16 The results showed that, on average, 58 percent of BHCs offered prenatal and family planning services-100 percent in those centers with the program for mothers and children (PMI), 87 percent in the larger BHCs (type IV), and between 17 40-60 percent in the majority of facilities, the smaller BHCs (types II and III). It is these latter facilities that include those constructed under the project. 3.20 In Table 3.2 above, family planning availability does not indicate which agency, ONPFP or MOPH/DBHS, delivered the services. Table 3.3 shows the distribution of responsibility for the delivery of family planning services in 1993.18 By then, unlike at the time of the PCR, there was cooperation between the two agencies, as midwives from both DBHS and ONPFP delivered family planning services. At the national level in 1993, ONPFP midwives supplied 57 percent of family planning services. Data from the eight project governorates show that this percentage varied from 16 percent in Sousse to 98 percent in Monastir.19 16. "National Evaluation Study of the Degree of Achievement and the Quality of Services offered in the Basic Health Centers." 1993. DSSB, MOPH. Tunis. 17. At the time of the audit there were 102 PMI, 89 Type IV BHCs (the largest facilities), 291 Type III BHCs, 875 Type 11 BHCs, and 263 Type I BHCs (rural dispensaries). 18. The borrower notes that these data concern only the sample of basic health centers chosen for this 1993 survey. Data concerning all the structures are available at the Directorate of Basic Health Services. Likewise, regarding Table 3.2, in 1993 78 percent of basic health centers offered these activities. It is also possible to calculate the rates for the structures built within the framework of the project. 19. It is unclear if the breakdown in data reflects the distinction between ONPFP mobile teams and the DBHS MCH records, or if the records merely indicate which staff provided the service, regardless of how it was provided (mobile teams or fixed staff. 26 Table 3.3: Distribution of Responsibility for the Delivery of Family Planning Activities (DBHS Midwives or ONPFP Midwives) 1993 Governorate DBHS Midwives ONPFP Midwives (%) (%) Zaghouan 33 67 Bdja 16 84 Jendouba 32 68 Le Kef 34 66 Kairouan 43 57 Sousse 84 16 Monastir 2 98 Mahdia 27 73 National Total 43 57 Source: 1993 Annual Report, DSSB, MOPH, Tunisia, volume 1. 3.21 Thus, while there is no doubt that access to facilities and services improved under the project, measuring the projects' outcome against its objectives is almost impossible due to the lack of data, as the MIS did not function as envisioned. As of 1988, the borrower reported a 50 percent decline in infant mortality (which was 123/1000 in 1981). In addition, project staff reported that total fertility rates had fallen to 4.3-though there was no comparison figure for the pre-project era, as the indicator in the SAR was the CBR. The national TFR in 1981 was 5.3.20 The contraceptive prevalence rates were reported to be 51 percent in the project area in 1988 (compared with 17 percent in 1981). However, with the exception of medical consultations, it is unlikely that these changes can be attributed solely to project intervention. 3.22 Indicators are not available for other health and population targets under the project, namely reduction in the CBR (43/1000 in 1981); morbidity; maternal mortality rates; neonatal deaths; and BCG vaccination rates. National figures are available to trace the movement of several health and population indicators, however the audit mission was unable to disaggregate these to the governorate level, despite working with the Evaluation Unit in the DBHS. At the national level, TFR stood at 5.3 in 1981 and had fallen to 3.3 in 1992; the CBR in 1981 was 34.6/1000 but curiously it was reported to be higher in 1993 at 44.1/1000 (different sources). For other indicators, the only ones available are comparisons (obviously, unsatisfactory) of the project area in 1981 with the national level in 1992/4: IMR in the project area in 1981 was 123/1000; at the national level in 1992 it was 32/1000. Vaccination rates for BCG were 45 percent in 1981 in the project area; nationally they were 98.2 percent in 1994. 3.23 Despite the lack of comparable, pre-project indicators, the audit mission was able to collect statistics on utilization for 1993 for the eight project governorates and compare them with the national averages.21 Table 3.4 shows that the average number of medical consultations per person per year is less than one (0.93) nationally but across the eight governorates the average is higher (1.21), ranging from a low of 0.77 in Kairouan to a high of 1.82 in Mahdia. The average 20. Source: Office National de la Famille et de la Population, !995, "Population, Family. Health: Key Figures 1994," Republic of Tunisia. September. 21. Caveat: these data refer to medical consultations provided by the public sector only (MOPH data). As the private sector provides a large portion of medical services in urban areas, data reported here for governorates with large urban populations (Sousse, Monastir) as well as data for the national average underestimate the actual number of medical consultations. 27 number of family planning consultations per married woman of reproductive age is also higher on average in the project governorates (0.59) than the national average (0.45). Again, there is high variability across governorates, from 0.12 in Monastir to 1.28 in Zaghouan. Similarly, the average number of prenatal consultations per pregnant woman across the eight governorates is higher (2.44) than the national average (1.50). The reduction of neonatal mortality and maternal mortality were project goals. However no data are available on the average number of prenatal consultants per pregnant woman in 1981 in order to ascertain if this objective was met. Table 3.4: Basic Health Care Activities, 1993 Ave. Frequency of Visits to CSB CL.u Mnsi1.7 012 14 .31 1.9 192 09 B . Governorate Makdia 1.82 1.13 2.26 3.15 1.78 1.78 1.51 Zaghouan 1.53 1.28 2.89 2.43 1.45 1.44 0.54 Sousse 1.08 0.38 3.49 3.76 2.79 2.79 1.24 Monastir 1.37 0.12 1.49 4.31 1.89 L.92 0.90 U~ja 1.10 0.47 2.67 2.00 1.34 1.34 1.19 Le Kef 1.13 0.79 2.83 2.58 1.26 1.23 1.03 Jendouba 0.87 0.22 1.93 1.92 1.38 1.33 1.32 Kairouan 0.77 0.29 1.94 2.09 2.05 2.05 1.55 Governorate Average 1.21 0.59 2.44 2.78 1.74 1.74 1.74 National Average 0.93 0.45 1.50 2.82 1.94 1.93 1.19 Source: MSP/DSSB/Evaluation Database (Funded by the 1991 Population and Family Health Project), November 1995 3.24 Table 3.4 also indicates the average frequency of visits to basic health centers (BHCs), disaggregated by type of visit. The number of consultations per week in 1993 averaged across the project governorates (2.78) was slightly less than the national average (2.82) . The governorate average for family planning consultations (1.74) was slightly less than the national average (1.94). A similar pattern is to be found for the number of prenatal consultations per week. However, the number of vaccinations per week in 1993 across the eight governorates is substantially higher (1.74) than the national average (1.19). Vaccinations also showed the greatest variance across the governorates, with the number of vaccinations per week in Mahdia almost three times that in Zaghouan. 3.25 Thus, despite the lack of data with which to evaluate the project, there is some evidence to support the borrower's claims that coverage has improved and utilization has increased. Information, such as that in Tabe 3.4, would have been extremely helpful in measuring the impact of the project had it been gathered in the project area before and after the project. Due to the absence from the files of the last three supervision reports, there is no information as to why nothing was done to measure the impact of the project, such as a cross-sectional survey, when the MIS was obviously failing. 28 Health, Population and Nutrition Communication 3.26 The health, population, and nutrition communication component was designed to help field workers do a better job of motivating families to adopt improved health, family planning and nutrition practices. There was very little information either in the files or the PCR regarding the implementation or results of this component. A supervision report in June 1985 reported that little progress had been made, however there was no elaboration. The audit mission visited the Regional Health Education Center in Sousse and found that the building was being utilized for other purposes. The audio visual equipment and materials were financed by the government not by the project. In Jendouba, the center was constructed, but due to a shifting foundation, the building was never occupied. The National Education and Communication center in Tunis was supposed to be renovated but the audit mission did not obtain any evidence to confirm this. The PCR states that achievements under the health education component were modest, as health education was of uneven quality and most health posts were without basic educational materials. The audit finds no reason to disagree with this assessment. Training 3.27 The Center for Pedagogical Research and Training (CRFP) was responsible for training and was strongly supported by the DBHS. Many training programs took place in 1981-1983 spurred by the project, though they were not all funded under this component. The project financed training for middle-level health managers and statisticians (40), full-time teachers (45), physicians (80), and paramedical personnel in the project areas. The retraining program for physicians and paramedicals focused on preventive care, outpatient treatment of common ailments, health education, and communication techniques. The project also financed curriculum development for training of physicians and paramedical personnel. Through the revision of medical and paramedical curricula, an increasing number of graduates were made aware of the specific needs of rural populations. In addition, two new paramedical were constructed, four paramedical schools were remodeled, and equipment and vehicles were provided. 3.28 Five of the eight regional directors of BHS attended short-term tra-ning courses in Europe under WHO fellowships. In addition, four national sem;nars were held under the project, and an international training course was given in Tunis on immunization. In June 1983, a supervision report stated that training activities were continuing at a rapid pace and were reaching an increasing audience through decentralized responsibilities given to regional authorities. The continuing education program was actively implemented in the regions and acceptance and participation were excellent. The first national seminar on the implementation of the BHS strategy was very well attended. The MDs participating organized similar seminars for paramedics in their representative regions. Thus training completed during the time of the project was very successful. Though only a few training activities were actively funded under the project, they built up a momentum for other training activities. 29 4. Ratings 4.1 The PCR rated the project outcome as satisfactory, sustainability as likely, and institutional development as negligible. The audit confirms the outcome and sustainability ratings and upgrades the institutional development rating to substantial. Borrower and Bank performance are rated as satisfactory, although there were substantial periods during project implementation when this was not the case. Outcome 4.2 The project was a success. It succeeded in the establishment, staffing, and equipping of major health infrastructure in eight governorates. At the time of the project, this new construction represented more than a 50 percent increase in the number of basic health facilities. The basic health care services component was satisfactorily implemented despite slight delays. Maternal and child health services (MCH) were successfully integrated into basic health care services (vaccinations, prenatal, and postnatal visits). Family planning services were integrated into basic health care services to a much lesser extent during project implementation, but the audit found that significant changes have taken place since, and at present integration is now no longer an issue. Accessibility to basic health services improved in the eight governorates due to the increased infrastructure, and general health almost certainly improved despite the difficulties in measuring project outcomes. The main issues that remain concern the efficiency of service delivery and quality of care. These issues are being addressed in the 1991 Population and Family Health Project. The project also strengthened BHS capacity and the orientation of the MOPH through internal reorganization including the establishment of the DBHS whose functions were in turn decentralized in an effective manner to the governorates. The project also created momentum with its small but successful training component which resulted in an improved health training system. 4.3 Achievements in other components were not as strong. The studies and applied research program had a modest outcome, though one of the studies (on infant mortality) did prove to be extremely influential in future health policy and the follow-on 1991 Population and Family Health Project. The health, population, and nutrition communications component also had limited success. The project failed to implement the MIS component, which suffered from complex design, changing technology, and lacked support in the MOPH. 4.4 Given the overriding importance of the project's health objectives and the remarkable achievements in expanding health infrastructure and the availability of services, the project is rated successful.23 Sustainability 4.5 The service delivery methods developed under the project have now been replicated across all governorates in Tunisia. The project provided the catalyst for decentralization of basic 22. Or, rather, these were the OED confirmed ratings during PCR review. 23. The PCR called the project a qualified success, probably a reflection of some remaining institutional problems at the time. 30 health services to the regions. Beginning in 1982 the MOPH budgets separated BHS and hospital expenditures in the Sixth Development Plan (1982-87). In addition, the 1991 Population and Family Health project reaffirmed the government's commitment to the BHS program. Sustainability is therefore likely. However, most of Tunisia's health system is in the public sector, and therefore there is concern about the financial sustainability of the sector as a whole. Institutional Impact 4.6 Institutional impact is rated as substantial; eight years after project closure, the institutional framework it helped put in place is not only still operational but has also became the national model. The position of the DBHS has been strengthened and the decentralization of responsibilities to the regional directors has been effective. The institutional rivalry between the ONPFP and the MOPH which plagued the project has largely disappeared. The scope and level of each organization's interventions have also been clarified, and their staffs share responsibility for the delivery of family planning services (para. 3.20). Borrower Performance 4.7 From the beginning, insufficient attention was paid to the political and institutional impediments of the integration policy. Combined with the personality conflict and high level MOPH staff changes, these unresolved issues of implementation of the integration policy caused complications and hampered the dialogue with the Bank. With this exception, borrower performance was very good with regard to the design of the project and implementation until 1984 when institutional changes interrupted the momentum of the project and seemed to call the basic BHS policy into question. As the PCR stated, the principal factors affecting borrower performance throughout the project were the MOPH's fluctuating level of influence in the system and managerial shortcomings, unsteady commitment to the central project concept, political pressures, and deteriorating economic conditions which adversely affected social spending. As evidenced in the 1971 First Population Project and again in this project, the borrower was reluctant to use loan funds for technical assistance, preferring grants from other donors. The new Minister of Public Health who was appointed in 1984 was critical of the project and repeatedly postponed and delayed supervision missions. Ultimately, however, the project was successfully implemented despite the delays and the borrower imposed break in supervision. The borrower met the project conditionalities, and a major effort was made to maintain the momentum of a project covering eight governorates simultaneously despite the lack of supervision staff and funding. Borrower performance is thus rated as satisfactory. Bank Performance 4.8 The Bank spent substantial resources and time (1977-1980) on the identification and preparation of this project. The PCR praised the great effort and vision that went into the design of a new program reflecting the government's policy shift to national basic health care but states that the Bank failed to adequately address the implications of the government's new policy-the interests at stake and its feasibility. Potential implementation difficulties which should have been foreseen were not addressed during preparation or appraisal. A review of the files shows that the Bank did, however, try on numerous occasions to get the MOPH to clarify the implementation plan and provide a timeline. It regarded the issue as an internal matter and felt that pressure should not be applied, especially given the very real possibility of further damaging the already 31 strained relations with the ONPFP. Despite the Bank's failure to realistically consider the organizational capacities to implement the integration policy, on the whole, project design was relevant to Tunisia's demographic goals, and supervision was regular, with the exception of the 1984-85 hiatus. The audit agrees with the PCR that the MIS component and the applied research studies could perhaps have been rescued following the 1985 resumption of supervision if enough attention had been directed towards them. 4.9 At times, however, there were serious problems. Some were caused by over optimism. Insufficient attention to bottlenecks during implementation also caused major problems. Following the change in the Minister of Public Health in 1984, dialogue with the borrower was constrained at times, as is illustrated by the disruption of supervision and the failure to follow through on project preparation on two occasions (1984 and 1987). At times it must have been tempting to stop the Bank's involvement in health. However, with the benefit of hindsight it is clear that the Bank's perseverance paid off. In a sector where policy issues are long term and funding issues mostly recurrent, policy dialogue over an extended period is the most important instrument the Bank has at its disposal to promote achievement of development objectives. Overall, Bank performance is rated as satisfactory. 32 5. Issues and Lessons 5.1 This audit raised a number of issues but this section focuses two deemed the most important-the integration of family planning and basic health services, and project monitoring and evaluation. Integration of Family Planning and Basic Health Services 5.2 The goal of integration was the extension of family planning services to underserved rural areas where logistical problems and the absence of links with the health services limited the effectiveness of the ONPFP. This shift in emphasis from a vertical family planning program to its integration with health signaled profound changes in the structure, role and function of the ONPFP and its relationship to the MOPH. It called for collaboration and coordination between two powerful national agencies whose willingness to integrate their activities was declared but not guaranteed. These institutional changes were insufficiently addressed by the government and the Bank and contributed to a prolonged preparation and implementation delays. The government's early proposals focused on civil works construction and skirted the definition of project organization, administration and financing. 5.3 Integration was an over-riding but unstated goal in the 1981 Health and Population Project. There were two schools of thought regarding integration, and the Bank took a position of being in favor, while the major players in the population field, USAID and UNFPA, took the opposite position and continued to support the vertical family planning program of the ONPFP. With hindsight, the Bank's stance appears to have been divorced from the political reality in Tunisia. It also brings into question the government's commitment to its integration policy. Neither the Decree in 1977 nor the Fifth National Development Plan (1977-82) provided any concrete plans for implementation of the integration policy, leaving it to the MOPH and the ONPFP to settle. This pronouncement, at least in the eyes of the ONPFP, appeared to come at the behest of the Bank and the MOPH. Thus from the beginning, an atmosphere of distrust existed, and it was very difficult to get the concerned parties to come to an understanding or to prepare an implementation plan. These institutional rivalries existed throughout the project and hindered implementation of its family planning goal (paras. 3.13-3.14). 5.4 The Bank has not acknowledged that it may have played a role in creating the conflict. Beginning with the 1971 First Population Project, the Bank's relations with the ONPFP deteriorated (para. 1.8). Tensions were further exacerbated when the responsibility for preparation of the 1981 Health and Population Project that had been delegated to the ONPFP was taken over by the MOPH (after the Bank deemed the ONPFP's efforts lacking and its proposal unrealistic). This planning exercise highlighted the lack of cooperation and coordination between the ONPFP and the MOPH, was a harbinger of what was to come, and served to further exacerbate the tensions between the Bank and the ONPFP as well as the ONPFP and the MOPH. The Bank did not attempt an analysis of the political and organizational factors in Tunisia, preferring to be swayed by the enthusiasm of the MOPH for the integration policy. The Bank and the other donors became enmeshed in the ensuing conflict, and integration did not occur in the time frame of this project. At the time of the audit, the conflict over integration appeared to have been settled (paras. 3.20 and 4.2). However, it is evident that the enormous loss of time and energy, not to mention good will, could have been avoided or at least the impact lessened in the 33 first place had the Bank paid adequate attention to the realities of the situation in Tunisia and to the practicalities of attempting institutional change. 5.5 Integration was pursued again as an objective in the follow-on 1991 Population and Family Health Project. The SAR states that the government, in full agreement with the MOPH and the ONPFP, decided to accelerate the integration of family planning and MCH into the routine BHS. A 10 year timetable was proposed and was said to be a realistic as it took into account the need to transfer skills and experience from one institution to the other.24 Bank staff say that this time integration was fully supported by the government, whereas previously it had been viewed as imposed from the outside. Project Monitoring and Evaluation 5.6 At the time of appraisal the government, with Bank concurrence, decided against either a pre-project baseline study or the monitoring of a control population outside the project areas due to prohibitive costs, technical difficulties and most serious, human resource implications. Monitoring and evaluation was to be based on comparisons with the first (baseline) year of operations and was to be a part of the MIS to be designed under the project. Quantitative objectives on several demographic indicators were mentioned in the SAR: reduction of IMR and the CBR, and increase in attended births, contraceptive prevalence, and vaccination rates. 5.7 The proposed MIS for the health system was to provide systematic and timely data about project activity in the eight governorates and thus enable speedy identification and correction of problems (para. 2.3). It was to rely on activity, performance, and impact indicators compiled monthly by health supervisors at dispensaries and consolidated by the governorate statistician, who would then forward the information to Tunis for analysis. The original design called for computer terminals for data processing at the regional hospital, and regional level data processing was to take place at the regional directorate. National level data processing was to take place at the CNI, where the central computer would be located. All the data results would go to the Central Administration (CA) for studies and planning as well as basic health services. The CA would have the responsibility for diffusion of results and instruction. 5.8 Specific indicators were to encompass (a) activity levels, (b) outreach and performance, and (c) impact. The main indicators are described in the SAR (paras. 5.15-5.16). The collection, processing and interpretation of indicators which constituted the MIS were reviewed with the chief epidemiologist of the DBHS in early 1983. As a result, some indicators were refined while others were expanded or abandoned as impractical. Data was to be collected both in dispensaries and by visiting nurses whose registers would record births, deaths and pregnancies in the estimated 200 households in each assigned community. Supervisory staff would check these registers periodically for completeness and at random for accuracy. In addition, the DBHS was to collect indicators of consumer's satisfaction by sample studies. 5.9 The MIS was also to collect and analyze data on recurrent costs of the dispensary and outreach parts of the health system in project areas in order to disaggregate the MOPH budget to permit specific provision for financing the services placed under the authority of the DBHS. These data would also make it possible to asses the alternative unit costs of various services provided at the peripheral level. 24. SAR, 1991, paras. 1.37-1.38. 34 5.10 The proposed MIS was too complex, as the CNI attempted to keep up with changes in computer technology occurring at the time. Thus, even though all the right elements were planned, its comprehensiveness and its complexity were its downfall. With hindsight, placing reliance on one complex system was unwise. It would have been more advantageous to begin with a less intricate system and to build it up. The MIS never became operational, and as a result there are few data with which to judge the project's outreach, integration, or effectiveness. 5.11 The audit team was unsuccessful in its attempt to collect data relating to the goals of the project. Extensive data was collected by the regional directors in the governorates, pooled from records of the basic health centers, and include detailed information about consultations (sex, age, social group/payment category, reason for visit, and treatment), vaccinations, respiratory infections, chronic illnesses, and special campaigns against tuberculosis, scorpions, diarrhea, and rabies. In addition, data is available on the number and type of facilities, staffing, and health education activities. However, there are no statistics enabling a comparison of project goals with actual outcomes. The existing data does not conform with those indicators originally designed to measure the outcome of the project-they appear to be more programmatic in nature, allowing comparisons of process more than an evaluation of change in health status. Thus the health and population indicators envisioned for this project are not being collected and no definitive answer can be put forth regarding the outcomes of this project in terms of achievement of its health goals. Lessons 5.12 The lessons from this project are as follows: (a) In order to ensure the feasibility ofproject goals, it is essential to examine the existing health system priorities, the political and organizational factors, as well as the government's long-term policy goals, and the process envisaged for meeting them. The integration of health and family planning services envisaged in this project-a major policy change-did not occur. The parties concerned did not fully understand the policy of integration and therefore did not respond well to it. Attention to differing priorities, a greater emphasis on rational, mutually agreed collaborative arrangements, an implementation time table, and the authority to take remedial measures were needed in this project. (b) Monitoring and evaluation systems are vital to measuring the success of PHN projects. While the project did seek to implement a thorough MIS, the design ultimately proved to be too complex, owing to the chosen computer technology and the desire to do everything at once. Thus, the project had no mechanism to record achievements in component to improve basic health services. (c) The Bank should not put undue reliance on technical assistance. Tunisia, like many countries, has been reluctant to borrow for this purpose, preferring to obtain such assistance free of charge from other donors and agencies such as USAID, WHO, and UNICEF. Given the experience of the First Population Project, the Bank should have explored alternatives to technical assistance, which was seen to be crucial for the success of this project. 35 (d) It is essential that procurement procedures are documented at the outset of a project and/or the Bank work carefully to establish acceptable bidding and evaluation procedures. In this project, the Bank overestimated the procurement capacity of the MOPH and procurement progressively became the single greatest practical impediment to project implementation as well an inordinate drain on supervision time.25 (e) Projects of massive scope and scale seeking to attain complex objectives simultaneously should be avoided in favor of a series of simpler, phased efforts which address logical steps in a longer-term program. This is especially true where institutional capacity is weak and vulnerable to political pressures. 25. The government has suggested that the problem was not capacity but overly cumbersome administrative procedures. 26. It is the opinion of the government that the scope of the project is not the main handicap if a good program of continuous evaluation is established, using all the components of the project during its execution, so as to be able to intervene on time in order to make the necessary corrections. The SAR identified as a major project risk the fact that the scale and design of the project were too difficult for effective MOPH coordination and implementation, an assessment which proved to be correct, despite actions taken or suggested to countervail their occurrence.  37 Annex I Tunisian Legislative Changes 1. Personal Status Code of August 13, 1956 This code granted Tunisian women the civil status of majority. The code abrogated polygamy and abandonment, and established divorce laws. The code also forbade the marriage of pre-adolescent girls (minimum legal age of 15) and guaranteed freedom of choice of the spouse. 2. Law of November 1958 Adopted the plan for school attendance, however, no mandatory attendance requirements were placed on education. 3. Law of January 9, 1961 Re-established the legal importation, sale and distribution of contraceptive products (repealed the 1920 French law which prohibited such products). 4. Law of December 31, 1962 A. Limited the payment of welfare benefits for families with dependent children to the first four children only. B. Limited tax allowances and salary allowances for the head of the household to the first four children only. 5. Law of February 20, 1964 Established a new minimum legal age for marriage, 17 for women and 20 for men, thus changing the Personal Status Code of 1956. 6. Law of July 1, 1965 This law legalized abortion under sanitary conditions by a doctor during the first three months of the pregnancy and only after the birth of the fourth child. 7. National Family Planning Program, 1966 Established the following fundamental objectives: A. Improvement of the quality of life of the citizen; B. Realization of demographic balance through controlling the process of procreation; C. Safeguarding the health of mothers and children; and D. Promotion of the flourishing of the basic cell upon which all society is build, the family. Annex 1 38 8. National Institute of Family Planning and Maternal and Infant Health Care Assists the Family Planning Administration of the Ministry of Public Health with the National Family Planning Program. The institute was responsible for all activities relevant to family planning (i.e., medical, educational, and training). 9. Labor Law of April 30, 1968 A. Guaranteed women the equal right to employment. B. Forbade the employment of children under 15 years of age in industry. 10. Law of March 23, 1973 Shifted the Family Planning Program to the Office of Family Planning and Population under the auspices of the Ministry of Public Health. 11. Law of September 26, 1973 Further liberalized the practice of abortion. Legalized abortion within the first three months of pregnancy provided that it be carried out in a "suitable facility". 12. Decree of January 31, 1974 Provided for joint operation of the National Office of the Family and Population and the Superior Council of Population. Established regional population councils. 13. Decree of December 27, 1985 A. Regarding regulation of marriage certificates: Established that each prospective spouse would undergo a complete medical examination, including blood count, prior to marriage. The examining physician would then provide suggestions on birth spacing. 14. Law of January 13, 1987 Changed the National Office of Family Planning and Population to the National Office of the Family and Population. 15. Law of May 6, 1988 Limited benefit payments to families for the first three dependent children. Source: Cochrane, Susan and David K. Guilkey, 1992. "How Access to Contraception Affects Fertility and Contraceptive Use in Tunisia." World Bank Policy Research Working Paper Series No.841, Washington, DC: The World Bank. 39 Annex 2 Republic of Tunisia-Ministry of Public Health Major Phases of the Program and Decennial Characteristics The Fifties Setting up of Institutional Foundations for a Modern State 1956: Independence year (March 20th) 1956: General Census of the Population (February 13th) 1956: Promulgation of the "Code of Individual Rights" Abolition of Polygamy and Repudiation Limitation of the age of marriage (15 years for woman and 18 years for man) 1958: Reform of Education (November 4th) Access to school for all the children of both sexes from the age of six years 1959: The Tunisian Constitution (June Ist) Equality of rights and duties between men and women: constitutional rights, as well as administrative, economic and social rights Main Characteristics (1956) Population: 3 Million 782 thousand inhabitants Total Fertility Rate: 7.2 children per woman Birth Rate: about 50 per thousand Death Rate: 25 per thousand Infant Mortality Rate: 200 per thousand Life Expectancy at Birth: 47 years The 2/3 of the population live in rural areas Illiteracy Rate: 85% (99% for women) 1 Physician per 7000 inhabitants Half of the population is under 15 years (49%) Annex 2 40 The Sixties Starting of the First Economic and Social Development Plans 1960: Family Allowance System (December 14th) 1961: Selling contraceptive products became permitted 1962: First Economic and Social Development prospects Decennial Development Perspectives (1962-1971) First Triennial Development Plan (1962-1964) Second Triennial Development Plan (1965-1967) Third Quadrennial Development Plan (1968-1971) 1964: Starting of an Experimental Family Planning Program 1964: Increase in the age of marriage (Woman: 17 years; Man: 20 years) 1965: Abortion permitted from the fifth child (July Ist) 1966: Starting of a National Family Planning Program 1966: Institution of the Labor code: Equality between both sexes (April 30th). 1966: General Census of Population and Housing (May 3rd) Main Characteristics Population: 4 Million 533 thousand inhabitants Total Fertility rate: 7.1 children per woman Birth Rate: 45 per thousand Death Rate: 15 per thousand Infant Mortality rate: 120 per thousand 60% of the population live in rural areas Illiteracy rate: 67% (82% for women) 46% of the population is under 15 years 41 Annex 2 The Seventies Establishment of Institutional Framework of the Demographic Policy 1972: Fourth quinquennial Development Plan (1972-1976) 1973: Creation of the Office National du Planning Familial et de la Population 1973: Abortion became permitted without restriction regarding the number of children 1974: Creation of the High Population Council (January 31st) 1975: General Census of Population and Housing 1977: Fifth quinquennial development Plan (1977-198 1) 1978: National Survey on Fertility Main Characteristics (1975) Population: 5 Million 577 thousand inhabitants Total Fertility Rate: 5.77 children per woman Birth Rate: 36 per thousand Death Rate: 9.9 per thousand Infant Mortality Rate: 84 per thousand 50% of the population live in rural areas Illiteracy rate: 55% (68% for women) 42% of the total population is under 15 years 31.4% of married women use a contraceptive method Annex 2 42 The Eighties Success ofthe Demographic Policy 1982: Sixth quinquennial Development Plan (1982-1986) 1983: National Survey on Contraceptive Prevalence 1984: The "ONPFP" became Office National De La Famille et de la Population 1984: General Census of Population and Housing 1987: Political Change of November 7th-H.E. Zine el Abidine Ben ali, New President of the Republic of Tunisia 1987: United Nations Award for Population Activities 1988: Meeti g of the Higher Population Council 1989: Family allowances granted only to the first three children Main characteristics (1984) Population: 6 Million 975 thousand inhabitants Total Fertility rate: 4.7 children per women Birth Rate: 31.1 per thousand Death Rate: 6.6 per thousand Infant Mortality Rate: 65 per thousand Illiteracy Rate: 46% (58% for women) 40% of the population is under 15 years 41.1% of married women use a contraceptive method in 1983 49.8% of married women use a contraceptive method in 1988 43 Annex 2 The Nineties New Qualitative Orientation of the Demographic Policy: Family Health and Sustainable Development 1991: Compulsory education up to the age of 15 years 1991: New orientations of the Program---"Family Health" 1992: Eighth Development Plan (1992-1996) 1992: New presidential measures in favor of Women and Family (August 13th) 1992: President of the Republic Award for the promotion of family 1992: United Nations Award for the exemplary contribution made by ONPFP for the preparation of the International Year of the Family, 1994 1992: First Regional Award for the promotion of family awarded to the Regional Family Planning Center of Sidi Bouzid 1993: Priority given to the intervention in rural regions and unprivileged areas 1993: First Regional Award for the promotion of family awarded to the Regional Family Planning Center of Kasserine 1994: General Census of Population and Housing (April 20th) 1994: National Survey on Mother and Child Health (PAP.CHILD) Main Characteristics Population: 8 Million 785 thousand inhabitants (1994) Total Fertility Rate: 3.32 children per woman (1992) Birth Rate: 24.1 per thousand (1993) Death Rate: 5.8 per thousand (1993) Natural Growth Rate: 1.84 percent (1993) Infant Mortality Rate: 36 per thousand (1992) Life Expectancy at Birth: 68.8 years (1991) Female: 70.2 years and Male: 67.4 years 39% of the population live in rural areas (1994) 33% of the total population is under 15 years (1994) 1 Physician per 1,660 inhabitants (1992) 59.5% of married women use a contraceptive method (1994) Source: Office National de la Famille et de la Population, undated, "Major Phases of the Programme and Decennial Characteristics," Republic of Tunisia, Ministry of Health. UNOFFICIAL TRANSLATION 44 Annex 3 Tunisia Republic Ministry of Public Health No MSP/153/UCT June 11, 1996 Mr. Roger Slade, Chief Agriculture and Human Development Division Operations Evaluation Department World Bank Subject: Performance Audit Report Tunisia - Health and Population Project (Loan 2005-TUN) Reference: Your letter dated of May 2, 1996 Our fax dated of June 1, 1996. In answer to your transmission mentioned above and relative to the report cited in subject, I have the honor to present you the comments of my department: We agree that the situation was very 1. Some explanations were given to justify the difficulties or some complex. However observed failures. In those explanations, it is quoted that some changes of changes in the responsible at the head of the Ministry. However, it is not certain that those eadership of the Ministy were events were at the origin of those difficulties. We think that the explanations among the factors given are very superficial and that the situation was more complex. It is for this that influenced the execution and reason that we are proposing the suppression of those passage in the report. outcome ofthe project. The same observation had earlier been made in the PCR report no. 9054 dated 10/11/90, para. 5.02. View cross- referenced in footnote 5 (text). 45 Annex 3 2. We totally agree on the diagnostic of the actual situation. 3. A few passages of this document bear overall pejorative judgments It is OED's resonsiibility' in and we ask for their suppression in the final report. Those passages are as a tofairy follows: assess the situation and report what we believe to be the facts. The conclusions we have drawn are based on reading the files and interviews with both Bank and government staff involved with the project. Page7w para. 4: Lines 12-13: "Borrower reluctance to use technical assistance funds limited training programs, civil works supervision, Text amended and use of consultants". This affirmation does not seem to be justified. Lines 14-15-16: "overestimated Borrower This comment :It originates from difficulties related to the normal procedures to be aded infootnote I capacity" : Itoiiae rmdfiute eae otenra rcdrst e (evaluative followed by the public sector. summary). PageJ8- para. 6. Lines 10-11-12-13: "many centers were The same underutilized and overstaffed; bureaucratic inefficiencies caused long delays in observations e made in the PCP, the payment of staff salaries..." "equipping of health posts was very slow...": para. 5.18. these are very disparaging affirmations. Lines 14-15: "Most importantly the ONPFP, a This view cross- well-staffed and well-funded vertical program ...": It is in fact due to an absence referenced in footnote 2 of implementation so as to build up teams able to provide the integrated services. (evaluative summary). Page-9 para. 9: lines 4-5: "...though issue remain regarding efficiency and quality" : how can we bring this non justified judgment in the Text amended. presence of the possible results recorded? Pagg9a lessons and conclusion: a) line 13-14: "Instead, antagonism quickly arose from the perception that the integration was Text amended being imposed". This is a pejorative judgment. Page 10: d) lines 11-12: "the Bank overestimated the procurement capacity of the MOPH...." is it the capacity of administration of the This view cross- MOPH or the necessity to respect the administrative procedures in effect. referenced in footnote 4 (evaluative summary) and footnote 25 (text). Annex 3 46 This view cross- e) lines 15-16: "Projects of massive scope and scale referenced in seeking to attain complex objectives simultaneously should be avoided in favor footnoteie of a series of simpler, phased efforts": The scope of the project is not the main summary) and handicap if a good program of continuous evaluation is established, using all the footnote 26 (text). components of the project during its execution, to intervene on time in order to make the necessary correctives. Text amended. Page 11: para. 1.3: lines 17-18-19-20: "Relations with the MOPH, however, were never smooth because of an unclear institutional division of labor in the population sector and the equally unclear status of family planning personnel borrowed from the MOPH.": This is a very disparaging judgment. Page Ie para. 1.8: lines 11-12-13-14-15 "...abandoned its institutional interest in the ONPFP...In the atmosphere of tension that gradually developed between the MOPH and ONPFP..." The report insisted on this aspect without trying to go in depth in the analysis. Page 13 para. 1.10: lines 12-13 "The ONPFP opposed Text amended integration and was supported by USAID and UNFPA." : disparaging judgment. Text amended Page 14a para. 1.12: lines 9-10-11-12 "... three key donors disagreed on the issue, and relations between the operating agencies in Tunisia were contentious at best" : disparaging judgment. Page 17: para. 2.6 Monastir instead of "Mogadir". Translation error corrected. Pare 19: para. 3.1: lines 9 to 15: "After 1984, project This view cross- Eag~19.par. 31 chnge wihinthe OPH ..the referenced in implementation was hampered by political changes within the MOPH....the eonc(text). general economic crisis...": refer to point I of the observations. Page 19 para. 3.2: lines 11-12: "...the MOPH's reluctance to use This comment project funds earmarked for technical assistance...": it is rather due to difficulties added in footnote 7 related to the procedures for the recruitment of the consultants that are applied in (text). the country. Page 21: para. 3.8: lines 7-8 "This project, the Second Health Text adjusted and Population project, was appraised in December 1994..." In fact it is accordingly. December 1984. 47 Annex 3 Page 23: para. 3.13: last paragraph: "It is evident that there have Ibis sentence was a been major differences of opinion, to the point of animosity..." disparaging quoteo ma judgment. memo dated 2/27/84. Page 24: para. 3.16: many centers were underutilized and The same overstaffed; bureaucratic inefficiencies caused long delays in the payment of observation was staff salaries with attendant loss of morale..."very disparaging affirmations. para. 5.18. Page 25: para. 3.20: table 3.2: the data concerning the proportion Text adjusted in of basic health centers that offer prenatal and family planning consultations footnote 18 (text). concern the sample of basic health centers chosen for this survey. It is to be noted that the data concerning all the structures are available at the Directorate of Basic Health Services (DBHS). In 1993, 78% of the structures offered these activities. Likewise, it is possible to calculate these rates for the structures built within the framework of the project. Page 26: para. 3.21: concerning the table 3.3, the proportion of Table 3.3 amended basic health centers where family planning consultations were offered by a midwife in 1993 were 16% in B6jA and not 1%. Note: As noted Hoping that you will take into consideration, during the revision of the above, to the extent that it was feasible, preliminary version of the report, the above comments, OED has taken into account the Sincerely, borrowers' comments. Some adjustments to the text have been The Minister of Public Health made; however, in other cases, changes could not Signed Dr. H6di MHENNI be made as OED feels that we are reporting facts and our best judgment based on our review of the project.  MAP SECTION  -I, TET RAS DJEBEL E T, E- Di ANE MENZEt S -37» ~MATEu BOURGulB TIQUE 5EL AOUAIRIA 3. -37 - NEF i HABET. NI1 OBE J BOUR lA 0 DRAHA AMDOUN Tu MMAJAF NA BEILU MENZIL EJA • 50IMAN TEMIME JE DOUB.3 ME NZE BOUYELfA BOLUSALE ED Et B ATGRIOMBANA TEST B. KRBA PARJECT B E ZAGHO AN NASEUJ S g GH MAO , A AD HAMAE SA1T NEBEUL A u Z G FASO U A N NAMAE YO SSEF 6,/11 5 i A N A LEKEF 51UANAdADOURoEN CDAVill E FdSERS1al 36- TAD.EOUIN E DAHMAIN SBIKH KIA KSOURI S- SOUSSE M94C.AR EL .MONAST IR KALAAT OUJSSELTIA KAI A' KSAR HELLAL ESENA -K A R O U A N M AKEN BOULGA R- HAFFOUZ2 - DE (HAEA - -. M A TIR , ---- MAHDIA u t K A S S E R l N E BOU i A KSOUR-E55AF HEEtNASR AUN ALLAH DJM -HEBBA ., CHOUR 5BEIIL E L KA55ERiNE ENC.A "SID' O, tIJliD, DjENMANA 5, FEIN S1 D 1 81 U1Z D /PmZEt 1AKER ..I5DI Ali BEN AOUN R S F A X (!EN KHAL1IFA E Z iO MAFHARES • , oe GAF-- METLAou, EtLUETAR PEDEYEFES-SEKHIRA G A F S A T.11.R EIG.IH MEIHOUJIAOUTS K NE ZEU DEAC GABES HUTSU . NFAEl HAMMA 4 ID o ADJIM K EBI L1 DJORF MAIMATA ZARIS5 DOUZ TUNISIA HEALTH AND POPULATION MEDENINE PROJECT KEDCEE 33'Project Area GHOUIMRA55 R NE33 J- - - -- Medical Region Boundarnesl~N Health Education Centers 11NUN Medical Schools Paramedical Schools to be RemodeIied M E D E N I N E Under he ProjectIUs 0 Medical Region& Capitals S Goverinoratte Capitals A F R l C A Governorare Boundaries P~vd Roads Q Earth and Grovel Roads Ra.lways 0 REMADA Poirts International Airportsr , ,,, , , Salt Lakes m cn.,a.i,.d,o -37. R~vrs reeo,aos ee m --International Boundaries 0 , 0 , 40 80 , 0 -o-p2""".'f'2J 'f'" KI1LOM E TE RSpy e eige ltso