Document of The World Bank FOR OFFICIAL USE ONLY Report No: 66574-CF PROJECT APPRAISAL DOCUMENT ON A PROPOSED CREDIT IN THE AMOUNT OF SDR 6.05 MILLION (US$9.35 MILLION EQUIVALENT) AND A PROPOSED GRANT IN THE AMOUNT OF SDR 4.95 MILLION (US$7.65 MILLION EQUIVALENT) TO THE CENTRAL AFRICAN REPUBLIC FOR A HEALTH SYSTEM SUPPORT PROJECT April 23, 2012 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 (Exchange Rate Effective February 29, 2012) Currency Unit = CFAF CFAF 490 = US$1 US$ 0.642665260 = SDR 1 ABBREVIATIONS AND ACRONYMS AMI Aide Medicale Internationale ANC Antenatal careASSOMESCA Association des (Euvres Medicales des Eglises pour la Sante en RCA APRD Armee Populaire pour la Restoration de la Rpublique et la Democratie - People's Army for the Restoration of Democracy) AZT Zidovudine BINUCA United Nations Integrated Peace building Office in the Central African Republic CAR Central African Republic CAS Country Assistance Strategy CBA Cost-Benefit Analysis CBO Community-based organizations CEMAC Commission de la Communaute Economique et Montaire de l'Afrique Centrale - Economic Community of Central African States CPG Country Program Support Pilot Grant CPN Consultation Pre-natal CPS Country Partnership Strategy CY Calendar year DA Designated account DALY Disability-Adjusted Life Year DC Directors of Cabinet DDR Disarmament, Demobilization, and Reintegration DHS Demographic and Health Survey DPT3 Three doses of the combined diphtheria/ pertussis/ tetanus vaccine DRC Democratic Republic of Congo EEA External evaluation agency EONC Emergency Obstetric and Neonatal Care ERR Economic Rate of Return ESW Economic Sector Work EU European Union FBO Faith-based organization FM Financial management FP Family planning GDP Gross domestic product GDN Gross Domestic Net HCWMP Health Care Waste Management Plan HEAP Health, Education and AIDS Project - Projet Sante, Education et SIDA- PSES HIV-AIDS Human Immunodeficiency Virus - Acquired Immune Deficiency Syndrome HMIS Health Management Information System HNP Health Nutrition Population HR Health Region HRITF Health Results Innovation Trust Fund HSA Health Sector Assessment HSSP Health System Support Project IBRD International Bank for Reconstruction and Development ICB International Competitive Bidding IDA International Development Association IE Impact Evaluation IFAC International Federation of Accountants IFR Interim Financial Report IMR Infant mortality rate IP Implementation Partner IPPF Indigenous People Planning Framework KNK Kwa Na Kwa Party- National Convergence Party) LRA Lord Resistance Army M&E Monitoring and evaluation MCH Maternal and child health MDG Millennium Development Goal MICS Multi-indicator Cluster Survey MOH Ministry of Health NCB National Competitive Bidding NGO Nongovernmental organization NPV Net Present Value OHADA Organisation pour l'Harmonisation en Afrique du Droit des Affaires - Organisation for the Harmonization of Business Law in Africa PBF Performance-based financing PIU Project Implementation Unit PMTCT Prevention of mother-to-child HIV transmission PNDS Programme National de Developpement Sanitaire - National Health Sector Development Plan PPA Performance Purchasing Agency PRSP Poverty Reduction Strategy Paper RFP Request for Proposal RH Reproductive health SIL Specific Investment Lending SOE Statement of Expenses SPN Supervision SWAp Sector-wide approach TB Tuberculosis TF Trust Fund TFR Total fertility rate UFMR Under-five mortality rate VCT Voluntary counseling and testing Regional Vice President: Obiageli K. Ezekwesili Country Director: Gregor Binkert Sector Director: Ritva S. Reinikka Acting Sector Manager: Jean-Jacques de St. Antoine Task Team Leaders: Gaston Sorgho and Hadia Samaha (Co-TTLs)  Table of Contents PAD DATA SHEET i I. Strategic Context 1 A. Country Context 1 B. Sectoral and Institutional Context 2 C. Institutional and Political Context 7 D. Higher Level Objectives to Which the Project Contributes 8 II. Project Development Objectives (PDO) 10 A. PDO 10 B. Project Beneficiaries 10 C. PDO-Level Results Indicators 10 III. Project Description 11 A. Project Components 11 B. Project Financing 13 1. Lending instrument 13 2. Project cost and financing 13 C. Lessons Learned and Reflected in the Project Design 14 D. Alternatives Considered 18 IV. Implementation 18 A. Institutional and Implementation Arrangements 18 B. Results Monitoring and Evaluation 20 C. Sustainability 20 V. Key Risks and Mitigation Measures 21 A. Risk Ratings Summary Table 21 B. Overall Risk Rating Explanation 21 VI. Appraisal Summary 22 A. Economic and Financial Analysis 22 B. Technical 23 C. Financial Management & Disbursement 25 D. Procurement 25 E. Social (including safeguards) 26 F Environment (including safeguards) 27 Annex 1: Results Framework and Monitoring 30 Annex 2: Detailed Project Description 34 Annex 3: Implementation Arrangements 41 Annex 4: Operational Risk Assessment Framework (ORAF) 57 Annex 5: Implementation Support Plan 60 Annex 6: Economic and Financial Analysis 64 Annex 7: Map of Central African Republic 71 PAD DATA SHEET Central African Republic Health System Support Project PROJECT APPRAISAL DOCUMENT Region: AFR Sector Unit: AFTHE Basic Information Date: April 23, 2012 Sectors: Health 100% Country Director: Gregor Binkert Themes: Health System Performance (50%0; Population and Reproductive Health (25%); Child Health (25%) Acting Sector Manager/Director: Jean J. de St. Antoine EA Category: B-Partial Assessment Project ID: P119815 Lending Instrument: SIL Team Leader(s): Gaston Sorgho and Hadia Samaha Joint IFC: Borrower: Central African Republic (CAR) Responsible Agency: Ministry of Health, Population, and the Fight Against HIV/AIDS (MOH) Contact: Mr. Jean Michel Mandaba Title: Telephone No.: +236 75050840 Email: jm mandaba@yahoo.fr Project Implementation Start Date: May 17, 2012 End Date: May 17, 2018 Period: Expected Effectiveness Date: August 31, 2012 Expected Closing Date: March 31, 2018 Project Financing Data(US$M) Loan [X] Grant The credit terms are as follows: 10 years grace period, followed by 10 years (2022-2032) credit amount repayable at 1%, then last 20 years (2032-2052) credit amount repayable at 2%. For Loans/Credits/Others Total Project Cost: 28.2 Total Cofmancing: Total Bank Financing: 28.2 1 Financing Source Amount(US$M) BORROWER/RECIPIENT IBRD 0.0 IDA: New 17.0 (9.35 credit & 7.65 Grant) IDA: Recommitted Health Results Innovation Trust Fund (HRITF) 11.2 Financing Gap 0.0 Total 28.2 Expected Disbursements (in USD Million) Fiscal Year 2013 2014 2015 2016 2017 2018 Annual 4.0 6.5 6.5 5.7 4.0 2.0 Cumulative 4.0 10.5 16.5 22.2 26.2 28.2 Project Development Objective(s) Increase utilization and improve the quality of maternal and child health services in targeted rural areas of Central African Republic Components Component Name Cost (USD Millions) Component 1: Improving utilization and quality of maternal and child services through performance-based financing at different levels 25.7 Component 2: Strengthen monitoring and evaluation capacity and support project implementation unit 2.5 Compliance Policy Does the project depart from the CAS in content or in other significant respects? Yes [ ] No [X] Does the project require any waivers of Bank policies? Yes [ ] No [X] Have these been approved by Bank management? Yes [ ] No [] Is approval for any policy waiver sought from the Board? Yes [ ] No [X] Does the project meet the Regional criteria for readiness for implementation? Yes [ X] No [ I Safeguard Policies Triggered by the Project Yes No Environmental Assessment OP/BP 4.01 X Natural Habitats OP/BP 4.04 X Forests OP/BP 4.36 X Pest Management OP 4.09 X Physical Cultural Resources OP/BP 4.11 X Indigenous Peoples OP/BP 4.10 X Involuntary Resettlement OP/BP 4.12 X Safety of Dams OP/BP 4.37 X 11 Projects on International Waterways OP/BP 7.50 X Projects in Disputed Areas OP/BP 7.60 X Legal Covenants Name Recurrent Due Date Frequency Section 4.01 of the FA and the GA Effectiveness Description of Covenant: Concomitant effectiveness of the FA and the GA once evidence of validity and binding effect of both agreements have been received by the Bank Name Recurrent Due Date Frequency Section I.B.2(a) of Schedule 2 to the FA and GA 3 months after the signature of the PBF Agency Agreement with the PPA Description of Covenant The Recipient shall cause each PPA to finalize, in accordance with terms of reference acceptable to the Association and furnish to the Association for its approval, its PBF Agency Agreement, the proposed PBF Manual included in the PPA's technical bid for services, for the implementation of the Project in the Project areas in which said PPA will carry out its services under its PBF Agency Agreement Name Recurrent Due Date Frequency Section I.E. 1 of Schedule 2 to FA and GA 1 month after Effectiveness Description of Covenant The Recipient shall furnish a program of activities for the first year of Project implementation to the Association, for its review and approval by the Association Name Recurrent Due Date Frequency Section I.G of Schedule 2 to FA and GA 6 months after Effectiveness Description of Covenant The Recipient shall, engage and thereafter maintain, one or more evaluation agencies ("External Evaluation Agency" or "EEA") to monitor, evaluate and report on Part A (1) of the Project, and to conduct independent verifications of the delivery of Technical Support Packages and MCH Packages by each Health Service Provider under its respective MCH Sub-project, including through interviews with consumers of Technical Support Packages and MCH Packages and inspections of the Health Service Provider's documentation and facilities Name Recurrent Due Date Frequency Section 11.B.4 of Schedule 2 to the FA and GA 3 months after Effectiveness Description of Covenant The Recipient shall improve its computerized financial management system in a manner acceptable to the Association. Name Recurrent Due Date Frequency Section 11.B.5 of Schedule 2 to the FA and GA 6 months after Effectiveness Description of Covenant The Recipient shall engage an independent auditors Name Recurrent Due Date Frequency 111 Section IV.B. 1(b) of Schedule 2 to the FA and Disbursement GA of Category 1 (PBF financing) Description of Covenant No withdrawals under Category 1 until (i) the Recipient have entered into a PBF Agency Agreement with a PPA and (ii) the said PPA has a PBF Manual finalized and (iii) said PPA has opened a MCH Performance Account and a PPA Operation Account. Team Composition Bank Staff Name Title Specialization Unit Gaston Sorgho Lead Public Health Specialist Lead Public Health AFTHE Specialist Hadia Samaha Sr. Operations Officer Sr. Operations Officer AFTHE Benjamin Loevinsohn Lead Public Health Specialist Lead Public Health AFTHE Specialist & Result Based Financing Christophe Lemiere Sr. Public Health Specialist Public Health & RBF AFTHE Paul Jacob Robyn Health Specialist Health Specialist & Impact AFTHE Evaluation Helene Bertaud Senior Counsel Legal LEGAF Aissatou Diallo Sr. Financial Officer Financial Management and CTRLA Disbursement Ousmane Kolie Financial Management Specialist Financial Management AFTFM Haoussia Tchaoussala Procurement Specialist Procurement AFTPC Enagnon Ernest Eric Adda Financial Management Specialist Financial Management AFTFM Emeran Serge Menang Forestry Specialist Safeguards AFTEN Lucienne M. M'Baipor Sr. Social Development Specialist Social issues, safeguards AFTEN Daniele Jaekel Operations Analyst Operations AFTHE Natalie Tchoumba Team Assistant Team Assistant AFMCM Nicole Hamon Language Program Assistant Language Program AFTHE Assistant Evelyne Madozem Team Assistant Team Assistant AFMCF Omer Zang Health Economist Health Economics AFTHE Locations Country First Administrative Location Planned Actual Comments Division Central African Republic 1v I. Strategic Context A. Country Context 1. The Central African Republic (CAR) is a very poor country that has recently emerged from over a decade of armed conflict. CAR is land-locked, has an estimated population of 4.3 million (2009), and has an average per capita gross domestic product (GDP) of approximately US$350 (2009).1 Sixty-two percent of the population lives in poverty, and over three-fifths of the population subsists on less than US$1.25 per day.2 CAR has also suffered from more than a decade of conflict and political instability. Unsurprisingly, the conflict has had important adverse consequences for economic growth and welfare. CAR was ranked 179th out of 182 countries in the Human Development Index in 2011. 2. CAR's economy is dominated by the primary sector, which accounted for 52.2 percent of GDP in 2008, followed by the tertiary sector (the services sector) which accounts for 30.5 percent of GDP, and the secondary sector (the industrial sector) which accounts for 12.1 percent of GDP. Subsistence agriculture (28.5 percent of GDP) and livestock (12.5 percent of GDP) are the main activities in the primary sector. Forestry, extractive industries (mainly diamonds), and export crops (coffee and cotton) contribute a smaller percentage of GDP, although these sectors are the main sources of export revenue. 3. The global financial crisis has severely impacted economic trends in CAR. The main impact of transmission of the international financial crisis has been the weakening of world demand and prices for commodities, which led to a decline in economic activity, especially in the forestry and mining sector, which account for approximately 82 percent of exports. In 2008, timber exports fell by 24.1 percent compared to 2007 and by 36.8 percent compared to pre-crisis projections. Inflationary pressures fed by higher food prices in 2008 declined to an annual average rate of 0.7 percent in 2011 but inflation in 2012 could be higher than the convergence level of 3 percent of the CEMAC region owing to speculation associated to international oil prices movements. Economic recovery has been gaining momentum but growth rate is still under the 4 percent per annum of the period 2006-2007. CAR has registered a growth rate of 3.3 percent in 2011 from the 3.0 percent in 2010. The lower performance in 2011 as compared to the 4.1 percent of growth's projection is due to uncertainties that followed the legislatives and presidential elections of January 2011. 4. The fiscal slippages in 2010 raised governance concerns and prompted donors to suspend budget support in 2011, further reducing the fiscal space available for priority spending. The suspension of budget support disbursements by donors in 2011 together with the lower domestic revenue has generated a very tight fiscal situation and intensified the liquidity constraints3. The government responded with cuts in public expenditure, which are projected to decrease to 17.2 percent of GDP in 2012 from 19.4 percent in 2010, thus reducing the overall fiscal deficit excluding grants by 2.1 percentage points to 5.6 percent of GDP. However, the 1 The gross national income (GNI) per capita in 2009 is US$350. 2 World Bank, World Development Indicators (Washington, DC: 2010). Further to a positive assessment from the International Monetary Fund, the World Bank has recently resumed disbursing its budget support in March 2012. 1 government has made a determined effort to increase public expenditures on health despite the crisis. The share of government expenditures allocated to health jumped to 9.7 in 2012 from 8.9 percent in 2010 and an average of 7.6 percent in 2007-2009. However, though rising, government expenditure on health remains low, representing 1.9 percent of GDP in 2010 (up from 1.2 percent of GDP in 2009). Budgetary resources would need to rise further, but the composition and quality of spending would also have to improve significantly in order to improve the health status of the population. B. Sectoral and Institutional Context 5. CAR has made little progress in improving health status: maternal and child health remains extremely poor, particularly in rural areas. CAR is not on track to achieving the health Millennium Development Goals4 (MDGs). About 174 of every 1,000 babies die before the age of five, and about 105 of every 1,000 infants die before their first birthdays.5 As a result of political and social unrest in the 1990s and early 2000s, the maternal mortality ratio increased to 1,355 deaths per 100,000 live births in 2003. The available data-which may not be very reliable-suggest that it may have decreased sharply to 540 in 2006, but the target of 171 deaths per 100,000 in 2015 will not be achieved. The total fertility rate (TFR) also remains high at 4.8 6 births per woman as of 2008. Rural areas are considerably worse off in terms of health than urban areas. To illustrate, the under five mortality rate (U5MR) is 141 per 1,000 in urban areas compared to 193 per 1,000 in rural areas. Similarly, the infant mortality rate (IMR) is 115 per 1,000 in rural areas compared to 88 per 1,000 in urban areas. 6. HIV/AIDS remains a priority health concern in the Central African Republic. CAR has the tenth highest HIV/AIDS prevalence rate in the world. The Government has defined the HIV/AIDS epidemic as a national emergency, and infection rates are thought to be on the rise. The recent Multi-Indicator Cluster Survey-4 (MICS-4) study (2010) estimates HIV prevalence in CAR at about 5.9 percent among persons between 15 and 49 years of age (down from 6.3 percent in 2006), which is above the 3 percent to 5 percent rates in neighboring countries. HIV/AIDS and related opportunistic infections are a substantial threat to the country's economic and social development. 7. Health spending in CAR is below average, with households bearing the majority of the burden through out-of-pocket expenditures. The level of per capita expenditure estimated at US$20 is below the African average and the lowest in the sub-region. In addition, the state contribution to the total funding for health is very low; general government expenditures on health as a percentage of total expenditures on health was 39.3 percent in 2008, while general government expenditures on health as a percentage of total government expenditures was modest at 11.0 percent. The share of personal expenditures in total health spending continues to increase, ranging from 48.6 percent in 2008 to 51.5 percent in 2010. Since the mid-1990s, out-of-pocket 4 Trends in Maternal Mortality: 1990-2008. Estimates developed by WHO, UNICEF, UNFPA, and the World Bank. 5 UNICEF, Risultats de l'enqute nationale a indicateurs multiples en RCA 2011 (Results of the Multi Indicators Survey in CAR 2011) (Geneva: 2011). 6 World Bank, World Development Indicators (Washington, DC: 2010). 7 UNICEF, Risultats de l'enqute nationale a indicateurs multiples en RCA 2011 (Geneva: 2011). 2 expenditures accounted for 95 percent of private spending in health and about 57 percent of total spending on health. Compared with other countries, the proportion of health spending paid by households in CAR is much higher than the average for Sub-Saharan Africa (36.5 percent) and low-income countries (47.9 percent) in 2008. The majority of out-of-pocket expenditures are collected through cost recovery mechanisms such as service fees and the sale of drugs. In fact, drug sales and service fees constitute over 75 percent of health facility revenues in rural areas, while transfers by the Government amount to only 1 percent. Donors and nongovernmental organizations (NGOs) provide 25 percent and 19 percent of facility funding, respectively. 8. Allocative efficiency in the financing of the health sector is a major concern in CAR. There is a profound imbalance between allocations to Bangui and those for other areas of the country, and the total state funding for the health sector is not oriented toward the poorest populations. According to the Plan National de D6veloppement Sanitaire (PNDS) 11 (2006- 2015), more than half of the funds are allocated for the capital, where only 16.9 percent of the population lives. Such an urban prioritization leaves less than half of the health sector budget to the more than three-quarters of the population that lives in rural areas where poverty is higher than in the capital and the need for basic health services a priority. 9. Utilization of critical maternal, child, and reproductive health services remains very low and is especially low in rural areas. Only 54 percent of deliveries are attended by a skilled attendant in CAR. At 36 percent of deliveries, this figure is even lower for rural areas (see table 1).8 A mere 32 percent of all 12-to-23 month olds in CAR are fully immunized. The prevalence of malaria is very high and the disease is an important cause of morbidity and mortality: over 13 percent of children under five had had malaria in the four weeks preceding a household survey.9 Despite this, only a third of "under-fives" in the country sleeps under a bednet (treated or nontreated), and only 36 percent of under-fives in rural areas. Just 32 percent of under-fives with fever were treated with an appropriate antimalarial drug, and only 26 percent of rural under-fives with fever were being treated with an antimalarial drug. Modern contraceptive use is a mere 15.2 percent in the country as a whole.10 10. Services delivered per health worker indicate that service volumes are quite low relative to the human resources available. Averages of four outpatient consultations are delivered each day per health worker with curative care responsibilities. The corresponding figure for maternal health services is a little over four. Furthermore, in both cases there are considerable differences between regions. The analysis suggests that service utilization is quite limited relative to the clinical human resources available, and that considerable room for improvement exists both across the board and in regions such as Health Region 3 (HR 3-in French, R6gion Sanitaire 3, or RS3) where a health worker sees two outpatients each day on average. This analysis cannot shed any light on potential demand-side barriers that may be constraining service utilization. However, the available data on quality suggest that limited utilization may also reflect quality shortcomings and that creating a good quality supply of health 8 UNICEF, Rdsultats de l'enqute nationale a indicateurs multiples en RCA 2011 (Geneva: 2011). 9 Ministry of Planning, Economy and International Cooperation. Household Consumption Survey, ECASEB, 2008. 10 UNICEF, Rdsultats de l'enqute nationale a indicateurs multiples en RCA 2011 (Geneva: 2011). 3 services that can meet current levels of latent demand for health services may be an essential first step.11 Table 1: Evolution of key health indicators in CAR Indicators 1995 2000 2006 2010 HIV/AIDS prevalence 15-49 year olds (%) - 6.4 6.3 5.9 At least I antenatal visit (%) 67.0 62.1 69.0 68.3 Births attended by skilled personnel (%) 50.0 42 53.4 53.8 Contraception prevalence (%) 3.0 6.9 8.6 - (modem methods) Children 12-23 months old fully immunized 36.6 19.2 32.0 - (%) Measles vaccine coverage (%) 52.4 33.5 62.0 49.8 Source: Demographic and Health Survey (DHS) (1994/95), Multi-indicator Cluster Survey 2 (MICS-2) (2000); MICS-3 (2006); and MICS-4 (2011). 11. Available data suggests that limited access (physical and financial) and quality are important problems in the CAR health system. A national household survey found that 26 percent of respondents reported needing health services in the four preceding weeks.12 Of these, however, only 13 percent used any health services, as the majority seeks non-modem treatment or no treatment at all. Among those who used health services, over half (53.6 percent) reported being dissatisfied. High cost (57.2 percent), long wait times (27.0 percent), drug stockouts (27.2 percent), the long distance to health facilities (18.8 percent), and the perceived ineffectiveness of treatment (21.4 percent) were all cited as causes of dissatisfaction. Overall, 35 percent of respondents reported having to travel over an hour to reach a health facility. When disaggregated by residence, only 9 percent of urban residents reported travelling for over an hour while nearly half (48 percent) of rural residents reported travel times exceeding one hour, indicating that physical access to care may be especially limited in rural areas. 12. Service availability for non-HIV services fell in all the regions between 2006 and 2011, but Health Regions 3 and 6 show the biggest declines. With the exception of postnatal care, the proportion of health facilities offering non-HIV-related maternal and child health (MCH) services dropped for most regions between 2006 and 2011 (see figure 1). For example, in Health Region 6, only one-third of health facilities had functional delivery (34 percent) and cold chain equipment (35 percent). Less than 57 percent of health facilities offer any immunization services-this needs to be expanded considerably if the currently high rates of under-five mortality are to decline. Furthermore, only 44 percent of health facilities offer any family planning (FP) services and only 26 percent of facilities offer tuberculosis (TB) treatment. Half or less than half of all health facilities had vaccines, oral contraceptives, or vitamin A pills in stock on the day of the survey. Health Region 1, however, has been able to maintain its non-HIV services to a greater extent than the other regions. This is most likely due to the ongoing performance-based financing (PBF) pilot financed by the European Union (EU) health system 1 World Bank, CAR Service Delivery Report (Washington, DC: 2011). 12 Ministry of Planning, Economy and International Cooperation. Household Consumption Survey- ECASEB, 2008. 4 support project. HIV-related services have not been affected by the general downward trend in services; in fact their coverage has improved over the past five years.13 The World Bank Multisectoral HIV/AIDS project and Global Fund support have contributed to the improvement of the availability of HIV/AIDS services in rural areas in CAR. Figure 1: Change in facilities offering key services between 2006 and 2011 by Health Region (%) 100 P 50- e 0 r -50 C -100 e mHR1 *HR2 HR3 sHR4 mHR6 n 13. Faith-based organizations (FBOs) and NGOs are important providers of health services, especially in rural areas. In CAR, NGOs and FBOs are present in 75 percent of the health districts and represent 10 percent of all health facilities in the country. The member organizations are increasingly supporting the provision of the Ministry of Health's Population and the Fight against HIV/AIDS mandated standard package of health services in the health centers and district hospitals they manage. Health sector interventions that seek to extend the coverage to poor and rural populations must therefore expand their reach. 14. The health sector has a number of critical shortcomings, with the end result that rural populations are very underserved. First, public health spending is low in absolute terms: only 4.4 percent of GDP was devoted to the health sector in 2008.14 This is aggravated by the lack of allocative and technical efficiency. A key consequence of this is that the benefits from public spending are skewed toward the wealthier quintiles of the population living in Bangui. Second, the health service delivery system has very weak capacity to deliver and manage quality health services at scale. A key driver of this is the concentration of human resources in Bangui, with the result that service delivery capacity in rural areas is particularly low. CAR has one of the lowest health worker densities in sub-Saharan Africa, with only 1.4 health workers per 10,000 population. (See figure 2 for further information on the paucity of health workers.) Furthermore, the CAR health system is heavily centralized. Management capacity is very weak at the periphery, resulting in negative consequences such as high levels of health worker absenteeism in rural posts. Third, effective demand for health services is relatively low. This is driven at least partly by perceived low quality of care and financial barriers to care seeking. 13 World Bank, CAR Service Delivery Report (Washington, DC: 2011). 14 World Health Organization, World Health Statistics. 2008. 14 Ministry of Planning, Economy and International Cooperation. Household Consumption Survey- ECASEB, 2008. 5 Finally, ongoing political instability over the past 15 years has contributed to limiting the institutional capacity of the Government of Central African Republic and its Ministry of Health (MOH). Equally important, this instability has led to low levels of donor support for the health sector, leading to deterioration in the quality and coverage of health services in rural parts of the country. Figure 2: Average number of health workers (by type) per facility present on the day of the survey in public and nonprofit facilities 1 0.9 0.8 0.7 0.6 0.5 0.4 - 0.3 - 0.2 - 0.1 - 0 Doctors Health Nurses Assistant Lab Technicians Technicians midwives m Public m Private not-for-profit 15. Shortcomings in CAR's health information system do not allow for proper monitoring and evaluation (M&E) of health sector performance. Only 43 percent of health facilities overall had the necessary data collection tools (forms or registers) required for reporting to the routine reporting system (health management information system, or HMIS). Hospitals are far more likely to have tools (64 percent) than health centers (56 percent) or health posts (34 percent). Overall 66 percent of health facilities transmitted the trimester report for the most recently completed month. No mechanisms exist for data verification at the various levels of the health system, as the compiling and processing of data remains purely administrative. 16. Clinical health workers' knowledge on how to manage complicated maternal and child health cases are quite low. In a recent facility-based study, midwives and auxiliary midwives were questioned on maternal care while nurses and nursing assistants were questioned on malaria and diarrhea. All staff were questioned on immunization. The score on how to appropriately treat child malaria or diarrhea cases was less than 50/100. While the staff dedicated to maternal care were for the most part able to cite proper treatment protocol on antenatal care (ANC) and deliveries, they were less able (less than 40/100) to cite steps on how to behave in front of more complicated maternal cases: for example, women bleeding during or after delivery (see figure 3). 6 Figure 3: Average score for maternal healthcare worker knowledge on Emergency Obstetric and Neonatal Care (EONC) services (Maximum score: 100) 70 60 50 - 2T 40- 0 S30- 20 10 0 Pre-Natal Care Delivery Bleeding during Bleeding after surveillance delivery delivery Clinical knowledge topic C. Institutional and Political Context 17. The political situation in CAR remains fragile; however political parties continue talks with Government to maintain the peace consolidation dialogue. The Presidential and legislative elections' results of January 23, 2011, have strengthened the control of the country by President Boziz6's Party, KNK (Kwa Na Kwa), but the joint elections also created a division with opposition parties. President Francois Boziz6 has been re-elected in the first round with over 64 percent of the votes and his political party won 72 percent of seats in the Parliament. Post-election tensions have been exacerbated following allegations of fraud by political opposition, but political parties and the Government are now back to the negotiations table led by the United Nations Integrated Peacebuilding Office in the Central African Republic (BINUCA) and aiming to revise the electoral code as well as the reform of the existing independent supervisory committee of elections. 18. The security situation remains unstable and subject to rebel movements including from neighboring countries. The security situation is uncertain in the North and along the borders in the East with Chad and Sudan. The situation weakened after the withdrawal of the United Nations Mission in Central African Republic and Chad (MINURCAT). The peace in CAR is mainly threatened now by rebel groups from neighboring countries: (i) in the North of CAR by the presence of rebels led by Baba Ladd6, the leader of a Chadian rebel group and (ii) in the Eastern and North-Eastern parts of the country due to frequent incursions by the rebels of the Lord Resistance Army (LRA) from Uganda and now operating in CAR. 7 19. The internal reconciliation among political parties is promising but the Disarmament, Demobilization and Reintegration (DDR) program is lagging. While the country continues to struggle with rebel groups, the DDR program for demobilizing armed groups is lagging. President Boziz6 formally launched the first phase of the sensitization campaign on August 13, 2009 and the second phase in February 5, 2010. As of today, close to 5,000 ex-combatants of the People's Army for the Restoration of Democracy (Arm6e Populaire pour la Restoration de la R6publique et la D6mocratie; APRD) in the North-West have been disarmed and demobilized but are waiting for their reintegration. The disarmament in the North- West has had a positive impact on the overall security environment in this part of the country. The temporary reward obtained in securing the North-West can be maintained if the country completes the demobilization and begins re-integration. D. Higher Level Objectives to Which the Project Contributes 20. The International Development Association's (IDA's) engagement in the health sector is critical to unlocking the economic potential of CAR and reducing poverty. CAR is in dire need of more investment in health -- especially in maternal, child and reproductive health -- in order to make significant progress toward the MDGs. The health situation remains grim. Continued high levels of fertility means that the needs for health (and education) investment necessary to create a productive workforce are increasing. The availability of reproductive health services can potentially make an important contribution to increasing economic growth and welfare by empowering women and lowering the dependency ratio. The country's movement toward improved political stability implies that this is an opportune time for change in CAR. The gravity of the health situation combined with the small number of donors in the health sector underscore the importance of IDA's role in rebuilding human capital and thus contributing to economic growth and poverty reduction in CAR. 21. The proposed project is aligned well with CAR's Poverty Reduction Strategy Paper (PRSP) and with CAR's National Health Development Plan (NHDP/PNDS II) and the National Health Policy (NHP). The project is expected to contribute to strengthening and extending the delivery of maternal and child health services and thereby contributing to the achievement of MDGs four and five to reduce under-five mortality and improve maternal health. In so doing, the project will contribute to an increase in access to basic social services and will strengthen human capital, which is a specific long-term objective outlined in the PRSP. The proposed project also directly focuses on promoting partnerships with the private nonprofit sector and seeks to create a conducive environment to improve health by strengthening management capacity. Both of these are priorities outlined in the NHP and NHDP. 22. Client interest in the proposed project is high. The Ministry of Health requested the World Bank to help sustain the development and expand the implementation of performance- based financing (PBF) in the health sector. Some of the technical groundwork for PBF has already been laid at the leadership level. A PBF workshop,15 jointly organized by the MOH in 15 Results-based financing is the overarching term used by the World Bank to denote incentive approaches. PBF is a specific form of supply-side PBF. PBF is the in-house WB term, whilst PBF is the specific strategy envisaged for CAR. See R. Soeters, PBF in Action: Theory and Instruments, Course Guide in Performance-Based Ee. (The 8 collaboration with key development partners, defined the road map for PBF development in CAR. This included the creation of a national steering committee that would oversee the coordination of various initiatives related to PBF as well as the preparation of the World Bank's PBF operation. The Bank's proposed interventions in PBF would complement the existing efforts of EU and its technical partner Cordaid. 23. Performance-Based Financing is an approach, that is becoming more widely used in other parts of Africa (often financed by the Bank) in which individual health facilities are provided funds based on the quantity and quality of services they provide as independently verified. For example, (see table 2) if a health facility fully immunizes 100 children in a quarter, they could earn US$500 (100 x $5 per child fully vaccinated). The total amount would be adjusted for the remoteness or difficulty of the facility (equity bonus). In the example below, this particular facility would earn 50% more because of the difficulties it faces. The total would also be adjusted by a quality score based on a checklist administered at the facility every quarter. This facility would earn 60% of what it would be entitled to due to the quality correction. Facilities can use the funds ($1,080 in this example) for: (i) health facility operational costs (about 50%), including maintenance and repair, drugs and consumables, outreach and other quality- enhancement measures; and (ii) performance bonuses for health workers (up to 50%) according to defined criteria. Table 2: Example of Performance Based Financing (PBF) in a Health Facility Service Number Provided Unit Price Total Earned Child fully vaccinated 100 $5 $500 Skilled birth attendance 20 $10 $200 Curative care <5 years of age 1,000 $0.5 $500 Total $1,200 Remoteness (Equity) Bonus +50% $1,800 Quality correction 60% $1,080 24. The proposed project is consistent with the current Country Partnership (CPS), Health Nutrition and Population (HNP), and the Africa Regional Strategies. Axis 2 of the CPS is "to ease the infrastructure bottlenecks hindering the rehabilitation and development of key sectors, and to improve basic service delivery and the living conditions of the population." The project seeks to address both these aspects through PBF. Experience from other contexts shows that providers tend to invest PBF revenues in improving infrastructure and equipment at their facilities and the coverage and quality of health services also tend to increase. The World Bank has also carried out a Health Sector Assessment (HSA) to identify service delivery constraints, especially in rural areas. The findings from this assessment were used to further develop interventions for the proposed health project. In addition, the proposed project is fully aligned with the HNP strategy (2007) and the Africa Regional Strategy (February 2011). The regional strategy clearly stipulates that "the World Bank will harness its comparative advantage in building resilience to address cumulative effects of shocks (such as health, natural disasters, Hague: Cordaid, 2011); P. Musgrove, Financial and Other Rewards for Good Performance or Results: A Guided Tour of Concepts and Terms and a Short Glossary (Washington, DC: World Bank, 2010). 9 diseases, etc.) through financial support, knowledge, global experience and technical assistance in designing, monitoring and evaluating safety nets reforms, health system reforms as well as smoothing the effects of macroeconomic shocks." PBF is a significant health-financing and management reform that aims at (i) improving service delivery through contract-based, performance-linked payments and (ii) improving governance and reducing corruption through the independent verification of performance. 25. The proposed project complements the Bank's current activities in CAR and benefits from the results of the Economic Sector Work (ESW) assessing the health situation. The current multisectoral health/ education/ HIV/AIDS project focuses not only on HIV/AIDS but also on other urgent health and social needs, such as prevention of malaria. II. Project Development Objectives (PDO) A. PDO 26. The project development objective (PDO) of this proposed operation is to increase the utilization and improve the quality of maternal and child health services in targeted rural areas of Central African Republic. B. Project Beneficiaries 27. The total population to cover under the project is 2.5 million, which is close to two- thirds of CAR's population. The project is expected to be implemented in nine prefectures located in four out of seven Health Regions (HRs) in CAR: * HR 2 (prefectures of Mamb6r6 Kad6i and Sangha Mba6r6: 542,542 inhabitants, excluding Nana Memb6r6, where Cordaid has been already implementing PBF for three years); * HR 3 (prefectures of Ouham and Ouham-Pend6: 931,346 inhabitants); * HR 4 (prefectures of Ouaka, K6mo and Nana Gr6bizi: 597,367 inhabitants); and * HR 6 (prefectures of Mbomou and Basse-Kotto: 481,157 inhabitants). These four health regions included in the intervention were targeted by the project due to their poor health status and health service coverage indicators. In fact, for certain indicators, coverage has deteriorated in these regions between 2006 and 2011. Three regions were excluded from the project intervention zone: Health Region 5 was excluded due to its ongoing security concerns, while Health Region 1 (the closest to Bangui and currently benefitting from the ongoing European Union PBF intervention) and Health Region 7 (Bangui and surrounding localities) were not targeted as their health status and service coverage indicators are relatively better off than the four regions targeted by the project. The anticipated estimated level of funding for the project does allow covering a maximum of 2.5 million people, which is the size of the population in the targeted prefectures. C. PDO-Level Results Indicators 10 28. The proposed PDO key results indicators are as follows: (i) increasing the number of people with access to a basic package of health services (number of consultations/capita/year in targeted areas); (ii) increasing the number of children immunized (number of children under one year who received all vaccines in the project areas); (iii) increasing the number of births attended by trained personnel in health facility in targeted areas (percent); (iv) increasing the number of new cases of tuberculosis treated in targeted areas; and (v) increasing the number of new acceptors of modem contraceptive methods in the project areas. 29. All core sectoral indicators, except two (the number of facilities constructed, renovated and/or equipped; and the number of long-lasted insecticide-treated malaria nets purchased and/or distributed) are included in the project targeted indicators. (A detailed list of indicators targeted by the project is in annex 2, paragraph 8). The project will not address the issue of inputs related to infrastructure and malaria nets. III. Project Description A. Project Components 30. The proposed project will have the following components. A detailed project description is presented in annex 2. Component 1: Improvement of health facilities performance through performance-based financing (US$25.7 million) 31. This component would be supported by the IDA ($15.064 million) and the Multi- Donor Trust Fund for Health Results Innovation (HRITF) ($10.636 million). It will finance pilot PBF implementation in nine selected health prefectures through the contracting of performance-purchasing agencies (PPAs). This support would have the following two dimensions: 32. Subcomponent 1.1: Performance payments to health facilities ($22.56 million): Provision of performance-based financing to provide MCH packages and technical support as well as capacity building activities to the health centers in the targeted districts. 33. This subcomponent would directly pay for PBF grants. The Government of the Central African Republic will contract with PPAs to provide MCH, technical support as well as capacity building services in the targeted districts. Those services will be provided through contracting health facilities at regional, district, and local levels. In the nine selected prefectures, the PPA will contract out with rural public, FBO and NGO health care facilities. Result-based grants will be paid to these health facilities in proportion to, and in payment for, achieved results. Facility payments will be based on (i) the technical support delivered to health centers and MCH services delivered to the targeted population and (ii) the technical quality of these services. Facility payments will be made quarterly after service volumes have been verified and quality of technical support and care has been assessed and certified by the PPA. After the quantity and quality of services provided are certified, payment will be released to contracted health facilities 11 via the PPAs by the project management unit. The details on the PBF approach are provided in annex 2. 34. Subcomponent 1.2: Support to PBF implementation and supervision ($3.14 million): Given the technical challenges of PBF implementation and the shortage of technical capacity and human resources within the MOH, the project proposes to identify international NGOs with demonstrated experience with PBF to act as the PPAs. This subcomponent will finance activities related to PBF implementation and supervision. The PPA contract will include a component that will finance the following activities: (i) Provide training to strengthen the capacities of the health service providers in the efficient and effective provision of the MCH packages as well as train MOH staff in the efficient and effective management of the delivery of technical support packages and MCH packages; (ii) Develop, monitor, and evaluate the provision of MCH packages in targeted districts, including through verification activities with respect to the technical support packages and MCH packages delivered, and conduct surveys of the recipients of the technical support and the MCH packages; and (iii) Support coordination efforts among local, district, and regional stakeholders. Component 2: Strengthen monitoring and evaluation capacity and support project implementation unit ($2.5 million) 35. Component 2 would be supported by the IDA ($1.936 million) and the Multi-Donor Trust Fund for Health Results Innovation (HRITF) ($.564 million). It will have two subcomponents: 36. At present, the health sector's routine reporting system is very fragmented, and the data generated are too unreliable and inaccurate to use as a basis for measuring health facility performance. Since PBF payments are to be made based on service volumes and quality, having accurate data on these parameters is critical to effective implementation. Component 2 will therefore support the overall monitoring and evaluation and external verification of PBF- financed services through health facility and community-based surveys. These verification activities will be conducted by an independent third party. The External Evaluation Agency (EEA) will be contracted by the Ministry of Health. The EEA's roles will include ex-post fact verification of service volumes and quality delivered by health facilities contracted with the PPA and for which PBF payments have been made. Services will be verified through (i) home visits to a sub-sample of facility clients and (ii) interviews with identified clients to assess perceived quality and satisfaction. The EEA will also be tasked with building in-country monitoring and evaluation capacity. Component 2 will also cover the operating costs of the PIU, including financing goods, consultant services, some training, and other operating costs. 37. Sub-component 2.1(US1 million): Support the overall monitoring and evaluation of the project by the PBF technical team at the central level and the External Evaluation Agency (EEA) to carry out its activities which will include (a) monitoring and evaluation of Component 1 of the project through overseeing and guiding PPA activities and independent verifications of 12 the delivery of technical support packages and MCH packages by each health service provider under its respective MCH subproject; and (b) strengthening of the recipient's capacity to monitor and evaluate the delivery of the technical support packages and the MCH packages 38. Sub-component 2.2 (US$ 1.5 million): Project Implementation Unit (PIU) support for project coordination, implementation, and management. In addition, minor rehabilitation work will be undertaken under this sub-component to improve the PIU and PBF Technical Unit's offices. B. Project Financing 1. Lending instrument 39. Financing instrument: The lending instrument will be a Specific Investment Loan (SIL) financed by (i) IDA with a Credit on standard terms with a maturity of 40 years (including a grace period of 10 years) and a grant, and (ii) the Health Results Innovation Trust Fund (administered by both the International Bank for Reconstruction and Development and the International Development Association) with a grant. 2. Project cost and financing 40. Project cost and financing will be as illustrated in table 3: Table 3: Project cost and financing Project Cost IDA HRITF Project Components (US$ Credit/Grant Financin million) Financing g 1. Improving utilization and quality of matemal and child services through performance-based financing 25.7 15.064 10.636 at service delivery facility levels - Subcomponent 1.1: Performance payments to health 22.56 13.6 8.96 facilities - Subcomponent 1.2: Support to PBF implementation 3.14 1.464 1.676 and supervision 2. Strengthen monitoring and evaluation capacity to enable the health information system to generate up- 2.5 1.936 0.564 to-date, reliable data to measure both PBF and general health system performance as well as the operating costs of the PIU -Subcomponent 2.1: Support to monitoring and 1.0 .436 .564 evaluation and to verification by the EEA -Subcomponent 2.2: Support to the PIU as well as 1.5 1.5 0 minor rehabilitations of the PIU and PBF Technical Unit's offices. Total Baseline Costs 28.2 17.0 11.2 Project Preparation Facility 0 0 0 Price contingencies 0 0 0 13 Total Project Costs 28.2 17.0 11.2 Interest During Implementation Front-End Fees Total Financing Required C. Lessons Learned and Reflected in the Project Design 41. Traditionally, many donors have invested in health sector interventions through inputs-based financing, but the results achieved are far below expectations. As a provider payment method, input-based financing such as prospective line-item budgeting for input costs (personnel, utilities, medicines, and supplies) can lead to allocative and technical inefficiencies in health sector spending. Such methods can incentivize providers to under-provide services, refer patients to other providers, increase inputs, and spend all remaining funds by the end of a given fiscal year. Input-based financing provides few incentives to improve the quality of services provided or the efficiency in which resources are used. PBF is an approach to health financing that shifts attention from inputs to outputs, and eventually outcomes, in health services. While inputs are necessary to finance health services, a predominant focus on inputs has failed to deliver the results that are necessary if the country is to achieve its MDGs. PBF is one form of output-based aid, whose key premise is that it "seeks to address weaknesses by delegating service delivery to a third party under contracts that link payment to the outputs or results delivered. It thus has the potential to improve incentives and accountability, while also expanding opportunities for mobilizing private financing." 16 42. The following lessons learned on PBF were derived from the design and implementation of both Bank and non-Bank operations in Africa and around the world and were found useful in the design of the project. 43. Appropriately designed PBF can potentially help to address many of the challenges identified above by the following actions: i. Improving the alignment between resources and maternal and child health priorities by purchasing priority service delivery indicators at higher rates ii. Allowing facilities to retain PBF funds for use at the operational level, and giving facility management autonomy on how to use these funds iii. Creating incentives for health facility managers and health workers to expand the coverage of important public health services and improve their quality by linking facility payments to service delivery and quality indicators, and offering health workers bonuses that are linked to facility performance iv. Improving governance through better verification and oversight of performance and incentives for good performance 16 Contracting for Public Services: Output-based Aid and Its Applications (Washington, DC: World Bank and International Monetary Fund, 2001), 91. 14 44. Several PBF-oriented programs have recently been introduced successfully in CAR. An EU-financed PBF pilot "Am6lioration des Soins de Sant6 de Base dans les R6gions Sanitaires 1 et 6" (ending in 2012) is currently being implemented in 4 of 16 districts (prefectures) in CAR and covers a total population of 1.01 million inhabitants. Cordaid (an international NGO with extensive PBF experience) has been contracted as the PPA since 2009. Within the context of this project, performance-based contracting in the education sector was also piloted in several primary schools. The mid-term evaluation of the project highlighted the critical role of a steering committee in both overseeing the project implementation from the central level as well as ensuring that lessons learned from the pilot were shared with other health sector partners interested in promoting PBF in CAR. Experience from the pilot project also identified that in CAR, paying for performance did not lower the overall cost borne by clients, despite the increase in utilization and quality of services. As a result, the mid-term evaluation also recommended that the reduction of the cost of services to clients should be included as a specific indicator of PBF. 45. Outside of the EU-financed project, since 2008 Cordaid has implemented PBF in an additional health prefecture in Region 2, while several other NGOs (Aide M6dicale Internationale, or AMI; the International Rescue Committee; etc.) have also introduced forms of performance-based bonuses to increase accountability and the quality of services of health centers. AMI has put a particular focus on reducing financial barriers to care. The level of bonus payments a contracted health facility is awarded is linked to how much service fees and the cost of drugs are reduced. An external evaluation of the Cordaid project in Region 2 provided several key lessons on PBF in CAR: (i) increased utilization was achieved through both improving the quality of care and reducing service fees and the price of medicine; (ii) gains in community mobilization and ownership were attained through contracting local organizations to conduct counter-verification activities; (iii) the health information system was strengthened through improvements in the routine reporting system and submission of monthly facility reports; and (iv) increased involvement of the local health management committee (Comit6 de Gestion, CoGes) was an important factor in contributing to good governance and accountability at the facility level. It was also noted that the poor health information system did not allow for the development of a strategic vision for PBF by the prefecture level health authorities. As a result, ownership of the PBF initiative at the regional and prefecture levels remained limited. Ensuring that local health authorities are able to define local health priorities is an essential element to successfully implementing PBF. 46. The success of PBF has also been well documented outside of CAR. Experience indicates that PBF approaches can be successful in rapidly increasing the use of cost-effective health interventions. Studies of PBF in Cambodia, Burundi, Haiti, and Afghanistan and a randomized controlled study in Rwanda have demonstrated its effectiveness. There are promising indications from a number of countries in Sub-Saharan Africa that suggest that PBF may be a useful approach to address the types of challenges evident in CAR. The Rwandan experience with PBF has attracted considerable interest and has had promising results in terms of increasing the proportion of staff in public sector facilities, increasing financing to the district level, and improving the coverage of key maternal and child health services. 17 Neighboring Burundi has also-albeit more recently-implemented a PBF program that is similar to the one 17 Agnes Soucat, "Performance-based financing in Rwanda," www.rbfhealth.org/rbfhealth/library, accessed January 18, 2011. 15 planned in CAR. Some promising preliminary results are available from Burundi. Since PBF has been implemented, facilities in Burundi are more likely to have the full complement of skilled staff, increasing from 37 percent in 2006 to 71 percent in 2010. Coverage of important health services such as skilled birth attendance has increased from 57 percent in 2006 to 82 percent in 2010, while contraceptive prevalence-often slower to change-has increased from 9 percent to 16 percent.18 47. Apart from the ongoing European Union and CORDAID projects in CAR, performance-based financing has already been successfully implemented in several other fragile states, including the Democratic Republic of the Congo, Burundi and Chad. Given this precedent, the fragile political situation in CAR should not be considered as a significant risk to the successful implementation of the proposed PBF project. 48. Besides providing an obvious performance-based motivation for health workers, PBF has a number of other advantages: (i) it is a clear signal to health workers about the priorities of the government and ensures that facilities maintain a focus on preventive and pro- poor interventions; (ii) it ensures that projects focus on producing tangible results and on strengthening M&E systems; and (iii) it decentralizes decision making to managers who are much closer to the community than those in the center. 49. Scaling-up of performance-based financing schemes across sub-Saharan Africa has developed rapidly over the past few years. Currently, PBF schemes have attained national coverage in Rwanda, Burundi, and Sierra Leone and are being piloted in several more countries: Sudan, Kenya, Zimbabwe, Zambia, the DRC, Benin, Chad, Nigeria, Burkina Faso, and Cameroon. Advanced planning for PBF piloting has been achieved in additional countries (Senegal, Mali, Ghana, Togo, and Congo), with preliminary discussions in yet another additional six countries (Liberia, Niger, Ethiopia, Botswana, Madagascar, and Malawi). 50. Initial evidence from recent PBF pilots in low-income countries shows that linking payment mechanisms to defined outcomes can lead to increased service coverage and improved quality of maternal and child health care. In Rwanda, results from two independent evaluations both showed a positive impact of PBF on utilization for institutional deliveries, growth in monitoring consultations, and increased levels of perceived and evaluated quality of care.1920 In the Democratic Republic of Congo (DRC), provision of performance-based subsidies resulted in lower direct payments to health facilities for patients, who received comparable or higher quality services than patients receiving care in control facilities. The disparity occurred 18 Longin Gashubije, Robert Soeters, Celestin Kamanuka, Godelieve van Heteren, and Michael Bossuyt, Presentation on PBF results in Burundi 2006-2010, Bujumbura, October 25, 2010. 19 L. Rusa, J. d. D. Ngirabega, W. Janssen, S. Van Bastelaere, D. Porignon, and W. Vandenbulcke, Performance- based financing for better quality of services in Rwandan health centres: 3-year experience, Tropical Medicine & International Health 14, no. 7 (2009): 830-837. 20 P. Basinga, P. J. Gertler, A. Binagwaho, A. L. Soucat J. Sturdy, and C. M. Vermeersch, Effect on maternal and child health services in Rwanda of payment to primary health-care providers for performance: an impact evaluation, Lancet, no. 377(9775) (2011): 1,421-8. 16 despite the fact that the districts receiving performance-based subsidies received substantially less external foreign assistance than the control districts.21 51. Gradual scale-up and simplicity in measurement are essential. No country has successfully introduced PBF without starting gradually. Lessons learned during the initial phase are crucial to a successful expansion. PBF is easiest to manage when facilities are paid for services that can be easily measured. Vaccinating a child can be easily measured and paid for, whereas behavior change is much more difficult to measure. 52. Ensuring transparency and independent verification are also critical. Robust, independent, and technically sound mechanisms are needed to verify the accuracy of reported results. There are typically two types of verification. First, ex-ante checking before payment is made to a facility that involves the review of registers and the checking of quality. This is done by the PPA and the district administrative structure and involves invoices submitted by the health facility. World Bank procurement reviews will be conducted to ensure that procurement rules are applied in accordance with Bank procurement policies. Second, ex-post verification is often done by independent groups or organizations and uses different techniques to detect "gaming," including the sampling of patients listed in registers to see whether they exist and whether they received the services listed. Specific sanctions will be imposed on facilities found to be misrepresenting their performance. 53. In most PBF schemes, it has been individual health facilities that receive payment, which are then managed at the facility level. This is the best way of ensuring that providers see the benefits of PBF. It often brings up the issue of how much financial and managerial autonomy health facilities have, and can become an important aspect of the policy dialogue. 54. It needs to be clear who approves the invoice submitted by the health facility and what procedures they will use. Those procedures pertain to (i) checking the calculations, (ii) verifying that services were actually delivered, and (iii) determining that appropriate adjustments for quality and equity are applied. A good practice is to establish a decentralized governance mechanism for this (a district PBF steering committee) with powers to oversee payments to health facilities. Usually MOH officials, local government officials, and civil society representatives sit in formal meetings. 55. A clear and concise manual of procedures that includes all the forms and information needed to implement the PBF scheme is important. Being explicit about policies will facilitate a meaningful dialogue with stakeholders and is an essential part of training providers and managers. 56. PBF usually involves important changes in a health care system and new procedures that need to be explained to providers and supervisors. Conducting high-quality training is an important challenge that will require a critical mass of people who are thoroughly familiar with the manual of procedures and PBF implementation. 21 R. Soeters, P. Peerenboom, P. Mushagalusa, and C. Kimanuka, Performance-Based Financing Experiment Improved Health Care in the Democratic Republic of Congo, Health Affairs 30, no. 8 (2011), 1,518-27. 17 57. Sufficient resources are needed to facilitate the supervision of health facilities. One of the major contributions of PBF is to make supervision more systematic and frequent. Ensuring that supervisors have the adequate resources to carry out frequent supervision will be important for success. D. Alternatives Considered 58. Why not consider traditional inputs-based health sector financing? As mentioned, input-based financing provides few incentives for health workers to improve the quality of services provided or the efficiency in which resources are used. When donors finance health services through input-based financing, prospective payments are made based on a predefined set of inputs (for example, staff salaries, medical supplies, transportation, building maintenance, and training). Given their prospective nature, such financing methods create the risk that more funding may be required, or that predefined goals will not be reached. Health service providers bear almost no risk in this arrangement, as they are paid regardless of the outcome, bearing limited responsibility for any failures. Performance-based financing shifts away from funding inputs and increases the accountability of fund recipients by holding providers accountable through performance measurement and evaluation. In practice, most health sector performance contracts are mixed-linking some payments to performance measures while other payments are treated as fixed or linked to inputs. 59. Why not consider sector budget support? Sector-wide approaches (SWAp) that provide sector budget support have evolved from the traditional investment project as a means to enhance development impact by scaling up, building donor partnerships in support of a country's long-term vision for its development, and improving the performance of institutions involved in implementing the SWAp. However, when new instruments such as PBF are introduced or a strong focus must be placed on a particular aspect of the sector, in the present case maternal and child health, a regular investment project is a better instrument as it allows a detailed design of the new features, the provision of technical assistance, as well as the allocation of financing to particular areas to be developed and supported. 60. Why not consider having no new project? The "no project" alternative is not desirable because (i) maternal and child health indicators are very poor, and maternal and child mortality rates remain high in Central African Republic, and (ii) the traditional way of financing the health sector has not produced the desired results, and a new project allows an avenue through which to introduce PBF. Furthermore, this new operation provides the needed support to help improve maternal and child health and introduce an innovative financing instrument which will contribute to CAR's trying to reach its MDGs. IV. Implementation A. Institutional and Implementation Arrangements 61. The Ministry of Health, Population and the Fight against HIV/AIDS (MOH) will implement the project through the Health, Education, and AIDS Project's (HEAP) Project 18 Implementation Unit (PIU). The PIU will be placed under the Director of the Cabinet (DC), with technical support from the relevant MOH directorates. The HEAP is staffed by a multidisciplinary team including nine members: a project manager, a financial and administrative management specialist, an accountant, a procurement specialist, two NGO/CBO specialists (who monitor the performance of health NGOs who were awarded contracts to implement community- based microprojects), one administrative assistant, and two support personnel. Members of the PIU have developed the skills and experience needed for project implementation through the implementation of the Bank-financed Multisectoral HIV/AIDS Project (P073525). Originally formed in 2005, the PIU was reorganized in 2009, leading to improved performance of the overall project. This restructuring was an important factor in having the project's rating change from unsatisfactory to satisfactory in 2010. Under the new PIU management, the project benefited from a second extension of the closing date granted in September 2010 for eighteen more months. The mandate of the PIU will remain in place for the Health System Support Project. The PIU, on March 21, 2012 has extended the contract and revised the terms of reference of the following staff: a project manager, a financial and administrative management specialist, an accountant and a procurement specialist. The PIU (under DC supervision) will carry out all of the disbursements and any procurement related to the project, in accordance with the Bankapproved procedures. It will serve as the funds holder, managing the operational accounts and transferring money to the PPAs, and submitting withdrawals to the Bank. 62. For Component 1, two performance purchasing agencies wiH be contracted by the Ministry of Health (one for Regions 2 and 3 and one for Regions 4 and 6) to manage the purchasing of predetermined technical support from the regional and district public facilities and the MCH services from rural public, FBO, and NGO facilities. The PIU will provide the PPA with the resources to pay health facilities for predetermined health services delivered to the population according to the norms and standards of service as defined by the Ministry of Health (subcomponent 1.1). The PPA will also receive resources for project support and verification activities (subcomponent 1.2). The PPA will contract with health posts, health centers and districts (prefectures), and regional hospitals that meet predetermined criteria to deliver an agreed package of technical support and MCH services. Payments will be based on (i) the technical support provided to the health centers and MCH services delivered to the population and (ii) the technical quality of these services. 63. For Component 2, strengthening project management and coordination, including monitoring and evaluation capacity, an independent third party, or external evaluation agency (EEA), will be contracted by the MOH. 64. At the central level a steering committee, including a national technical PBF unit, will guide PBF implementation to draw lessons and policy implications for the health sector. It is chaired by the Director of Cabinet of the MOH and includes key directorates of the MOH, the Ministry of Planning and Economy, the Ministry of Finance, and representatives from the donor community (the Bank will not sit as a member of the committee but only as an observer, to avoid going beyond its mandate). The steering committee will oversee the implementation of PBF, document the lessons learned from various initiatives of PBF in the country, provide guidance to the PIU and PPA, and generate policy direction for the institutionalization of PBF in CAR. 19 65. The National Technical PBF Unit is created by the Ministry of Health, Population and Fight against HIV/AIDS as a technical body to support the project Steering Committee as well as to promote performance based financing in CAR. More specifically, the PBF technical unit is in charge of: (i) preparing the steering committee meetings and supporting the implementation of the decisions made by the Steering Committee, (ii) supporting the regulatory function that the Ministry has to assume in PBF implementation, (iii) monitoring the progress of PBF implementation in the field, and promoting ownership of PBF by the Ministry, and (iv) exploring ways and mechanisms to institutionalizing PBF as a national policy in CAR and to scaling up PBF beyond the pilot phase. 66. The project policies and procedures were incorporated in an implementation manual, which was adopted by the PIU on February 24, 2012. It will be completed by a series of PBF manuals, each one prepared by a PPA as part of its offer of services and improved as practical experience develops from in-the-field implementation. The PIU and the Bank will ensure that the PBF manuals prepared by the PPAs are consistent with each other and with the overall implementation manual. A more detailed description of the implementing arrangements is presented in annex 3. B. Results Monitoring and Evaluation 67. The Results Framework focuses on accountability for results-it moves beyond the usual tracking of inputs and outputs and places a strong emphasis on intermediate outcomes. To the extent possible, the proposed results framework uses existing indicators and data to measure the progress of both the project and its contribution to the overall national program, not only for efficiency but also to build on and strengthen existing data collection mechanisms. For example, routine monthly and quarterly data collected from the Central African Republic's Health Management Information System (HMIS) will be aggregated for the project's annual indicators so as to reinforce the national system and avoid creating a parallel one. The project monitoring system will include (i) identification and consolidation of M&E indicators; (ii) training and capacity building initiatives at the national, regional, and local level, as needed; (iii) standardized methods and tools to facilitate consistent collection, consolidation, and sharing of information; (iv) an independent review led by an external technical consultants; and (v) annual program evaluations and strategic planning exercises for each component. 68. The results of the PBF component of the project will be assessed through a full Impact Evaluation (IE). The policy objectives of the Impact Evaluation are to (i) identify the impact of PBF on maternal and child health service coverage and quality, (ii) identify key factors responsible for this impact, and (iii) assess cost-effectiveness of PBF as a strategy to improve coverage and quality. In doing so, we expect that the results from the Impact Evaluation will be useful to fine tune the designing of the national PBF policy in CAR and will also contribute to the larger body of knowledge on PBF. C. Sustainability 20 69. Technical sustainability will be ensured by capacity building and knowledge transfer activities throughout the project. For Component 1, capacity-building efforts at all levels of the health sector will be coordinated and implemented by the PPAs and will take place within a precise timeline. Support from the PPAs will be progressively scaled down during the lifetime of the project. This will help the national PBF teams develop adequate tools and will provide on-the-job training, so that by the beginning of year three of the project, the PBF teams at the regional and district levels should be technically autonomous and the technical assistants can reduce their support. 70. Financial sustainability of PBF can be reasonably achieved, given the limited cost of this mechanism and the growing interest of other donors for it. The project will help improve the efficiency of health spending by improving the outcomes obtained from the current spending of US$20 per capita. By spending less than $3 per capita per year per beneficiary, the current cost is likely to be affordable and sustainable in the long term. However, as in most sub-Saharan countries, the relatively low levels of GDP per capita will require donor support to CAR for many years to come. Because of its direct focus on results, the support for the PBF methodology has been popular with a wide array of donors and is likely to remain so in the medium to long term. In addition, it is expected that the scaling-up of PBF after completion of the Bank operation would also be financed through the creation of a current expenditure budget line for PBF expenses within the national health sector budget. V. Key Risks and Mitigation Measures A. Risk Ratings Summary Table Rating Stakeholder Risk Low Implementing Agency Risk - Capacity Moderate - Governance High Project Risk - Design Moderate - Social and Environmental Low - Program and Donor Low - Delivery Monitoring and Sustainability Moderate Overall Implementation Risk Substantial B. Overall Risk Rating Explanation 71. The overall risk associated with the project is rated "substantial." The main risk is related to the introduction of a new Performance Based Financing (PBF) instrument in a relatively low capacity setting. However, the project's design addresses these issues by involving experienced Performance Purchasing Agencies (PPA) and emphasizing training efforts at the central, regional, and local levels. Furthermore, the risk of having health workers 21 manipulating results indicators is significantly reduced by ex-ante verification by the PPA, and ex-post verification by the external evaluation agency. In addition, experience from other countries shows that such can be successfully mitigated through a gradual scale-up, technical assistance, and training. These have already started and will continue during project implementation. The implementing agency, design, and delivery quality risks are rated as Moderate. The overall implementation risk rating for this project is Substantial. 72. The team is convinced of the Government's strong political commitment to the project. Yet given CAR's ongoing security concerns in certain areas of the country, political and security instability may also be a risk to the project. A permanent monitoring of the political and security situation is ensured by the international community in CAR. The project team will continuously benefit from updated information in order to adjust the implementation of the operation as needed. VI. Appraisal Summary A. Economic and Financial Analysis 73. Macroeconomic and financing analysis. This analysis assumes that the scaling-up of PBF after completion of the Bank operation would mainly be financed through the creation of a current expenditure budget line for PBF expenses within the national health sector budget. This analysis develops two primary scenarios (a low-growth baseline scenario and a high-growth scenario) that estimate the expected level of government funding available for scaling-up and sustaining PBF in the years following the Bank operation. It should be noted that public health expenditure includes external donor contributions, alongside government expenditures in the analysis. 74. Sustainability of the project. The per capita benefit of the project amounts to US$$ 2.0 per year. If PBF was scaled up nationally, the operation would require an overall financing of US$$ 9.7 million per year. That amount represents less than the total government current expenditure per year as estimated for the last year base-case scenario (US$$ 11.2 million). Under the frame of the high-growth scenario, that amount accounts for only 70 percent. Moreover, out- of-pocket payment used for current expenses in the health facilities in CAR could only reinforce the evidence of the sustainability of the project (details in Annex 8). 75. Cost benefits analysis. To ascertain economic soundness of the project, a cost benefit analysis (CBA) was conducted. The analysis took into account the estimated incremental project's costs and benefits. A CBA is a technique of identifying, measuring and calculating a project's cost and benefits in order to ascertain the net returns [net present value (NPV)] and economic rate of return (ERR) of the project's investments. 76. Methodology. The 'input-output' approach to CBA was employed to generate net benefits of project components/activities. Within the context of this operation, the 'input-output' approach addresses the purchasing of high quality health services from targeted health facilities. From the net benefits, the net present value (NPV) and the economic rate of return (ERR) were calculated to determine economic viability of the project. To account for uncertainties associated 22 with the analysis, a sensitivity analysis was conducted to determine the extent of responsiveness of the analysis. We assumed a 10-percent discount rate for computing present values of net benefits and costs. The period of the analysis covers a 5-year period. 77. Project benefits and costs. The implementation of the project is expected to generate two main types of benefits (direct and indirect benefits). Direct benefits are benefits that would accrue to the project target beneficiaries; they are referred to as medical benefits resulting from improved access to a basic package of health, nutrition, or population services in targeted areas, children fully immunized, births attended by trained personnel in health facility, new cases of tuberculosis treated as well as acceptors of modem contraceptives. The project targets a quality improvement of at least 50 percent over 5 years. Indirect benefits include improved quality of life, prevention of unwanted pregnancies, a reduction in average household size, and over the long run, a key contribution to the society's well-being as a whole. Costs are the resources used in carrying out project activities. The direct costs of the project are the cost of implementing and evaluating the project. Indirect costs include, beneficiaries transport costs, costs associated with patients' visits, etc. Due to lack of data, the analysis did not take into account indirect benefits and costs, or the epidemiological situation before and after project implementation, and focused mainly on direct benefits and costs. The project is targeting as intermediate impact, health facilities achieving an average 50% of quality of care index (range from 0 to 100) in the targeted areas (see Results Framework in Annex 1). We have therefore monetized every "unit of quality delivered" with a supplement corresponding to the mean cost obtained per direct beneficiary of the project. 78. Based on these assumptions, the Cost-Benefit Analysis estimates the Net Present Value of the project as US$$ 8.4 million (Table 4). The positive NPV means that the project would generate the expected positive results. The Economic Rate of Return is estimated at around 50%. Sensitivity analysis did not significantly change the CBA results. Such a high ERR comes from the fact that the baseline evaluation of the project is assuming a zero percent quality index (Annex 1). We have conducted sensitivity analyses by changing the discount rate within a range from 3 to 12 percent and did not found any significant impact on the CBA results. Table 4: Cost-benefit analysis Year Index 0 1 2 3 4 5 Discount Factor 1.0000 0.9091 0.8264 0.7513 0.6830 0.6209 Discounted Flows Costs -$4,668,032 -$4,675,976 -$4,666,287 -$4,614,850 -$4,507,768 $0 Minimum Benefits $0 $6,365,498 $6,376,330 $6,363,119 $6,292,977 $6,146,956 Net -$4,668,032 $1,689,523 $1,710,043 $1,748,269 $1,785,209 $6,146,956 Cumulative -$4,668,032 -$2,978,509 -$1,268,466 $479,803 $2,265,012 $8,411,968 Minimum Net Present Value $8,411,968 Minimum Economic Rate of Return 50% B. Technical 23 79. The project supports a package of maternal and child health interventions aimed principally at reducing child and maternal mortality in the four selected regions. The approach of investing in maternal and child health interventions is supported by both a recent World Bank study in CAR and a body of evidence, notably in a series of Lancet articles 22 published in 2003, 2006, and 2008, as well as Cochrane collaboration reviews on interventions to reduce maternal mortality.23 The 2011 World Bank study identified how MCH service availability and quality have changed between 2006 and 2011 with data collected from public and nonprofit health facilities in five rural regions of CAR. Analysis findings suggest that the availability of health services is quite limited and has worsened over the preceding five years. In 2011, there were only 1.5 health facilities and 0.18 doctors for every 10,000 people. Furthermore, a very small proportion of health facilities offer critical MCH and family planning (FP) services (only 57 percent offer immunizations, 44 percent offer FP, and 26 percent offer TB treatment), and the proportion of health facilities offering non-HIV-related services has declined between 2006 and 2011. On average, health facilities had less than 44 percent of basic delivery equipment and less than half of the cold chain equipment they are required to have per national protocols. Clinical health workers' knowledge on MCH is very low and is a serious constraint to delivering quality services. About 40 percent of staff engaged in maternity wards in nonurban areas could not correctly list the components of prenatal care. The same proportion could not list the elements for monitoring the delivery. Worse, less than one agent in two (40 percent) know the signs to look for/monitor in a parturient abnormal bleeding and less than one agent in three (30 percent) knows the signs to look for in a woman who, immediately after delivery, is bleeding profusely. 80. The design of PBF arrangements in CAR is based on the best practices observed in other successful PBF projects. For instance, external entities (such as community-based organizations) will be strongly involved in monitoring PBF results (Component 2). Similarly, the mechanism to determine PBF credits is a "fee-for-service conditional on quality" system, which has been applied with successful results in other PBF projects in Rwanda, Burundi, and Zambia (Component 1). Such a system ensures that (i) the PBF mechanism is clear and can easily be understood by health workers and communities and (ii) the increase in the quantity of care is not detrimental to quality. 81. The procurement method that will be used to recruit the PPAs is "quality-based selection." The NGOs will submit a performance-based financing manual (PBF manual) outlining the details of implementation, organization, administration, monitoring and evaluation, environmental and social monitoring and mitigation, financial management, disbursement and procurement. Unit costs of each service to be purchased should be specified in the PBF manual. Based on best practices of results-based financing and adapting them to local conditions, each health service will be priced using a model that ensures that health centers are motivated to produce the optimal quantity and quality of the target indicators. In any case the unit price shall 22 The Lancet, founded in 1823, is one of the oldest peer-reviewed medical journals in the world, published weekly in England. The Lancet is considered to be one of the core general medical journals. 23 The Cochrane Collaboration, founded in 1993, was developed in response to Archie Cochrane's call for up-to- date, systematic reviews of all relevant randomized controlled trials of health care. A group of over 6,000 specialists in health care review biomedical trials and results of other research. 24 not exceed 100 percent of the total estimated cost of the technical support services or the MCH services, as the case may be, minus the amount of other funds allocated to finance such cost. C. Financial Management & Disbursement 82. The financial management (FM) assessment for the CAR Health System Support Project was conducted to determine whether the project has adequate financial management systems and related capacity for project implementation. The FM risk is "high." The proposed FM arrangements to be put in place by the entity meets the Bank's minimum requirements for project FM as per OP/BP 10.02. The FM assessment was carried out in accordance with the Financial Management Manual for World Bank-Financed Investment Operations issued by the Financial Management Sector Board that became effective on March 1, 2010. 83. The dated covenants include inheriting or acquiring new accounting software within three months of effectiveness and recruiting an auditor recognized by the Economic and Monetary Community of Central Africa (CEMAC) professional accounting body within six months of effectiveness to conduct annual audits. The recruitment of the PPAs endowed with acceptable financial management manuals and the EEA will be critical in setting the PBF framework and will be agreed with the team. The details of the FM assessments are summarized in annex 3. 84. For Component 1 (PBF Financing and supporting activities), after verification of reported results, PPAs will address a request to the PIU with relevant documents and the PIU will transfer funds to the PPA for services rendered in the health sector. Each PPA will open and operate two operational accounts in commercial banks, acceptable to IDA, one for the PPA's operating costs and one for results-based payments. Funds will flow to those accounts from the central Designated Account A which will be opened at the PIU for the financing of the PPAs. A dedicated financial channel will be set up and is described in annex 2. 85. Expenditures under Component 1 will be audited by an external financial auditor as part of the audit of the project's accounts. Given the risks of misappropriation and embezzlement at the local level, the MOH will select an external and independent auditing company in order to control expenditures related to PBF financing. This auditing company would check that the procedures defined in the procedures manual have been complied with. 86. For Component 2, part of the funding would be allocated for the work relevant to the M&E activities carried out by the PBF technical unit and the payment to the contracted EEA for work relevant to the M&E activities. This may be done via direct payment or from relevant designated account. The remaining funds would cover the costs of the PIU and remaining M&E activities. D. Procurement 25 87. Procurement would be carried out in accordance with the World Bank's "Guidelines: Procurement of Goods, Works, and Non-Consulting Services under IBRD Loans and IDA Credits & Grant by World Bank Borrowers," dated January 2011, and "Guidelines: Selection and Employment of Consultants under IBRD Loans and Credits & Grants by World Bank Borrowers," dated January 2011, and the provisions stipulated in the Financing Agreement and "Guidelines on Preventing and Combating Fraud and Corruption in Projects Financed by IBRD Loans and IDA Credits and Grants", dated October 15, 2006 and revised in January 2011. 88. Procurement activities at the central level will be implemented by the Project Implementation Unit of the HEAP which has substantial experience in the implementation of projects financed by the Bank. There will be very small procurement activities at the decentralized level. Each health facility may use part of its subsidies to improve quality of service by procuring small items estimated at less than US$20,000 per contract. Any beneficiary of PBF resources should procure small goods, works and services in accordance with provision of Procurement Guidelines, and will be responsible for compliance with procurement procedures. The PIU will be directly in charge of all other activity of significant, complex and pooled procurement across beneficiaries. 89. During the preparation phase of the project, a procurement assessment was carried out by the Bank. The assessment reviewed the organizational structure for implementing the project, the institutional arrangements, and the capacity of the HEAP project staff responsible for procurement. It concluded that the HEAP procurement department headed by a procurement specialist was successfully adhering to World Bank procurement guidelines. As the current HEAP PIU will continue to act as the PIU for the new project, it is expected that sound procurement policies will continue to be applied. 90. The Bank's prior and post reviews will be carried out on the basis of review thresholds. The Bank will conduct a six-month supervision mission and an annual Post- Procurement Review. The Bank could also conduct an Independent Procurement Review at any time until two years after the closing date of the project. 91. Risk mitigation measures have been discussed and agreed to with the HEAP, and the risk is assessed as "moderate." A draft procurement plan for the project has been prepared and found acceptable. It was finalized and agreed upon during appraisal. The procurement plan will be updated in agreement with the project team annually or as required to reflect the actual project implementation needs and improvements in institutional capacity. More details on the procurement capacity assessment and project procurement arrangements are provided in annex 3. E. Social (including safeguards) 92. The project is expected to have a positive social impact by improving access to health care services for the poorest households. Component 1 (through the payment for performance) will provide incentives for health facilities to reduce staff absenteeism and to improve staff responsiveness with patients. As a result, health facilities with PBF contracts will in turn provide more and better care for marginalized populations. 26 93. The project will also enhance community ownership for monitoring the quality of basic health services. CBOs will be identified and strengthened so that they can be involved in monitoring health facilities. The project will build on the experience of the HEAP project, which had successfully implemented a strong community based microproject component involving a great number of CBOs and community Health associations. 94. The preparation process of the project has been highly participatory with extensive work and consultation among the key stakeholders: selected line ministry representatives and representatives of the donor community, including NGOs. To avoid opposition to PBF, a strong emphasis on using a participatory approach was introduced very early in project preparation to explain PBF to the project stakeholders through workshops and meetings. The monitoring and evaluation system has been designed to ensure the adequate targeting of project activities, including their social impact. 95. The project will be implemented in the area where indigenous peoples (IPs) are located. While the project will not as such negatively affect IPs, OP/BP 4.10 policy is triggered to ensure that IPs will benefit from the project. An Indigenous Peoples Planning Framework (IPPF) has been prepared and disclosed publicly in country as well as in the World Bank's Infoshop before appraisal on March 13, 2012. The IPPF has the objective to define clearly how IPs will be taken into consideration in the project and what the indicators for doing so will be. In fact, a specific indicator related to the number of contacts per capita for IPs was included to motivate health professionals to pay particular attention to serving IPs. 96. The project will have a positive impact on gender in CAR. Given that the project's objectives are to improve maternal and child health in target areas, improving women's health is an essential component of the intervention. Particular attention will also be given to ensuring active participation of women in project areas through the use of community CBOs (local NGOs, women's groups, agricultural groups, etc.). The project is expected to have a positive impact not only on pregnant women but on all women, as PBF credits will improve the quality of care for the identified package of health services essential for the general population. F Environment (including safeguards) 97. The project triggers the Environmental Assessment (OP/BP 4.01) policy and has been assigned an environmental screening category of "B," given the risks associated with the handling and disposal of medical and general health waste. Health facilities use medical equipment that might entail an increased production of medical waste. Despite this risk, this project is not expected to generate any major adverse environmental impact. Possible environmental risks include the inappropriate handling and disposal of hazardous medical waste, including sharp needles. 98. The Government prepared in September 2010 the Health Care Waste Management Plan (HCWMP) as a condition to extending the closing date of the Health-Education- HIV/AIDS project. This plan is being implemented by the MOH, and it was agreed to revise the same plan to take into account this project. The HCWMP has been revised and re-disclosed by appraisal on March 13, 2012. In addition, health centers will report on hazardous medical material waste management since that action will be purchased and included as an indicator to be 27 evaluated in PBF. The quantified quality checklist used by PBF to pay for performance on the quality measure, in fact, measures this element. The weighting for the indicators related to health waste management will be increased, and the adherence to the guidelines will be checked, and paid for, quarterly.. 99. The borrower's capacity for environmental and social safeguards implementation and reporting remains limited despite the implementation by the line ministries of the Health- Education-HIV/AIDS project financed by the Bank. A limited understanding of the Bank's policies exists. Close technical support will be provided by the Bank's safeguard specialists during implementation to ensure compliance with all Banks safeguard policies as well as national policies. The capacity-strengthening measures were outlined in the project safeguards documents and integrated in the project budget, implementation, and monitoring plan. 100. IDA supervision will focus on (i) providing regular implementation support, (ii) carrying out field reviews of safeguards implementation, and (iii) monitoring safeguards implementation based on periodic progress reports. IDA supervision will be carried out by field-based Bank technical staff and complemented by specialist consultants not only during regular biannual supervision missions but also during interim technical safeguards missions that will respond to emerging issues. Monitoring will be based on the Health Care Waste Management Plan and IPP recommendations. 101. Two safeguards policies were triggered, as follow: Safeguard Policies Triggered by the Project Yes No Environmental Assessment (OP/BP 4.01) [X] [] Natural Habitats (OP/BP 4.04) [] [X] Pest Management (OP 4.09 I] [ ] Physical Cultural Resources (OP/BP 4.11) [] [X] Involuntary Resettlement (OP/BP 4.12) [] [] Indigenous Peoples (OP/BP 4.10) [X] [] Forests (OP/BP 4.36) [] [X] Safety of Dams (OP/BP 4.37) [] [X] Projects in Disputed Areas (OP/BP 7.60)* [] [X] Projects on International Waterways (OP/BP 7.50) [ ] [X] * By supporting the proposed project, the Bank does not intend to prejudice the final determination of the parties' claims on the disputed areas. 102. OP/BP 4.01-Environmental Assessment: This policy is triggered because of direct potential environmental impacts associated with the handling and disposal of medical and general health waste. The borrower has prepared a Health Care Waste Management Plan (HCWMP) in September 2010, which has been revised to take into account this new project. The HCWMP was re-disclosed prior to appraisal. The project will strengthen the capacities of health centers in handling health wastes. 28 103. OP/BP 4.10-Indigenous Plan: This policy is triggered to ensure that IPs will benefit from the project even though this project does not expect to impact negatively IPs. An IPPF was disclosed in the country and at the InfoShop by appraisal. 29 CENTRAL AFRICAN REPUBLIC: Health System Support Project Annex 1: Results Framework and Monitoring Results Framework Project Development Objective (PDO): Increase utilization and improve the quality of maternal and child health services in targeted rural areas of Central African Republic Cumulative Target Values Responsi Description (indicator PDO Level Results Indicators* W Unit of Baseline24 Fre- Data Source/ bility for definition etc.) Measure YR 1 YR 2 YR3 YR 4 YR5 quency Methodology Data Collection 1. People with access to a basic MOH Health Number of new package of health, nutrition, or 440,00 ppaaeof health, n trtn X Number 400,000 0 480,000 520,000 560,000 600,000 Annually Information MOH consultations/year. population services in targeted 0 Sse HS areas (Core Sector Indicator) fclty records facility records MOH Health Number of children at 1-year-old 2. Children immunized in X Number 31,360 34,500 36,472 37,744 39,416 41,000 Information having received all required targeted areas (Core Sector Annually System (HIS) - MOH vaccines. Indicator) facility records 3. N births attended by MOH Health Number of births assisted by a trained personnel in health Number 26,500 27,828 29,216 30,676 32,209 33,819 Annually Information MOH health professional (midwife, t in personel inea t System (HIS) - qualified nurse, doctor, etc.) facility in targeted areas facility records MOH Health 4. New cases of tuberculosis Number 1,450 1,640 1,830 2,020 2,150 2,300 Information Number of new TB cases treated treated in targeted areas Annually System (HIS) - MOH treated __in_targeted_areas_facility records MOH Health Women of reproductive age MOHr Hath who were counseled and 5. Number of acceptors of E Number 7,050 9,575 12,862 15,793 17,930 20,250 Information provided with modern modern contraceptives in Annually System (HIS) - targeted areas. facility records MOH contraceptive method Beneficiaries 6.1. Direct project Number 0 511,903 558,550 604,213 649,555 695,069 Annually Ministry reports Project ADDING UP INDICATORS 1, beneficiaries, Unit 2,3 & 5 ABOVE. 24 Baseline data based on 2010 HMIS: once the health facility and household surveys are finalized in summer 2012, the baseline data will be updated to reflect results from those surveys. 30 6.2. Of which female Number 0 278,628 307,814 340,319 372,147 400,429 Annually Ministry Project Estimate based on the fact that (beneficiaries) Unit there is no gender bias. INTERMEDIATE RESULTS Intermediate Result (Component One): Improvement of health facilities performance through performance-based financing (PBF) 1: Proportion of mothers in the Number of women who received project areas who received two Report from PBF two doses of preventive doses of intermittent preventive E % 4% 6% 8% 10% 11% 15% Annually statistics MOH treatment during last pregnancy/ treatment for malaria during last (Quantity) Total number of births pregnancy Number of outpatient 2. Outpatient consultations Report from PBF consultations provided to provided to indigenous people E Number 0 20 40 60 80 100 Annually statistics MOH indigenous people as measured (Quantity) by the origin of the patient in the project areas. Number of health facilities 3. Health facilities achieving an Report from PBF achieving an average 50% of average 50% of quality of care E % 0% 25% 50% 70% 75% 80% Annually statistics (Quality MOH quality of care index (range from index (range from 0 to 100) in evaluation grill) 0 to 100) / Total number of the targeted areas health facilities implementing PBF Number of facilities that 4. Percentage of facilities that Report from PPA received PBF payments within received PBF payments within E % 0% 25% 50% 70% 75% 80% Annually statistics PPA 60 days / total number of 60 days of invoicing facilities implementing PBF 5. Pregnant Women receiving "..receiving antenatal care" antenatal care during a visit to a E Number 87.000 95,700 104,400 113,100 121,800 130,500 Annually Report from PPA PPA means any antenatal consultation health provider I I I I I statistics (new or repeated consultation) Intermediate Result (Component Two): Strengthen monitoring and evaluation capacity to enable the health information system to generate up-to-date, reliable data to measure both PBF and general health system performance Number of facilities 1. Percentage of health facilities whose PBF results have whose PBF results have been H % 0% 40% 50% 60% 70% 80% Annually Report from PPA PPA been fully verified / Total verified statistics number of facilities implementing PBF Number of health facilities who 2. Percentage of health facilities Report from PBF displayed the quantitative that displayed the two previous H % 0% 30% 40% 50% 60% 70% Quarter- statistics (Quality MOH statistics of the two previous quarter health HMIS statistics to ly evaluation grill) quarters / Total number of the public health facilities Percentage of health facilities who 3. Percentage of health facilities MOH Health have transmitted each trimester that have transmitted their recent % 0% 40% 50% 60% 70% 80% Information MOH HMIS report within 60 days of the trimester HMIS report Annually System (HIS) - end of the quarter / Total number of facility records health facilities implementing PBE 31 Figure A1.1: How PBF affects MCH service coverage and quality MCH-4 service coverage MCH service quality Demand-s ide factors Supply-sidefa ctors * Ffnancal accessfb[lity of services - Userchargesforservices * Physicalaccessibility of services - Ava!fab!My and organizatfonof * Socio-cufturaFaccessibity of servrces servies * Perceived quality of care o Hour,whenser,4ce, are avallaWIe o Servge outreach * aQuaBty ofMcare o Presence of knowCedgeabe and cnent-orented skied staff * AvailabifFty of drugs o AvaiLabRty ofecuipment o Availability of essentia suppies o Clean and adequate physical faclifty Ex pe ted changesdueto RBF Expected changesduetc RBF More and betterBehaviorChange Communicat[on * Reduction in nformalusercharges (BCC) focused on mrnprovfng demand for heath service* sGreaterstaffmotivation * More and betterBehaviorChange Communication * Strongercientfocus focused on changingcrfcaf heafth behaviors. such as o Servfces available at more client frendly hours infant and child feedfngpract[ces o Services available through outreach o Staff have more c[fent-friend[y attitude Better management and greater autonomy in drugs procurement o Reduced staff absenteeism o tmproved drug inventory and supply management o improved equipment and facility maintenance 1. The Results Framework for this project is strongly informed by the advances in M&E thinking in the Bank.' The Results Framework focuses on accountability for results-it moves beyond the usual tracking of inputs and outputs and places a strong emphasis on intermediate outcomes. In addition to the accountability function of evaluation, this Results Framework also emphasizes the learning function of evaluation. To the extent possible, the proposed results framework uses existing indicators and data to measure the progress of both the project and its contribution to the overall national program, not only for efficiency but also to build on and strengthen existing data collection mechanisms. 2. The indicators and methods of measurement designated in the Results Framework have been designed to rely on preexisting information systems and data sets whenever possible. For example, routine monthly and quarterly data collected from the Central African Republic's Health Management Information System (HMIS) will be aggregated for the project's annual indicators so as to reinforce the national system and avoid creating a parallel one. The Multi- 1 In support of the 2009 IEG HNP Evaluation, a background paper on M&E quality in HNP investment operations raised concerns about (i) the poor quality of results frameworks, (ii) the absence of baseline data;,(iii) poor efforts or failure to collect baseline data or within first year of the project approval, (iii) unrealistic indicator targets, and (iv) poor data quality. 32 Indicator Cluster Survey (MICS) will be used for population-level indicators because the survey rounds coincide with the five-year project cycle. The baseline and follow-up survey rounds of the PBF Impact Evaluation will also provide data for monitoring the progress of indicators identified in the Results Framework. 3. Ex-post external verification will be conducted by the external evaluation agency (EEA). The PBF component will take place in nine health prefectures. The EEA will be contracted by the Ministry of Health as an independent third party to conduct quality checks and independently assess implementation performance for the PBF verification in Component 1. 4. The project monitoring system will include (i) identification and consolidation of M&E indicators; (ii) training and capacity building initiatives at the national, regional, and local level, as needed; (iii) standardized methods and tools to facilitate consistent collection, consolidation, and sharing of information; (iv) an independent review led by an external technical consultants; and (v) annual program evaluations and strategic planning exercises for each component. 5. The results of the PBF component of the project will be assessed through a full Impact Evaluation (IE). More details on the IE in annex 2. 6. The policy objectives of the Impact Evaluation are to (i) identify the impact of PBF on maternal and child health service coverage and quality, (ii) identify key factors responsible for this impact, and (iii) assess cost-effectiveness of PBF as a strategy to improve coverage and quality. In doing so, we expect that the results from the Impact Evaluation will be useful to fine tune the designing of the national PBF policy in CAR and will also contribute to the larger body of knowledge on PBF. 33 CENTRAL AFRICAN REPUBLIC: Health System Support Project Annex 2: Detailed Project Description 1. The development objective of this proposed operation is to increase utilization and improve the quality of maternal and child health services in targeted rural areas of the Central African Republic. The project will have the following components: Component 1: Improvement of health facilities performance through performance-based financing (PBF) US$ 25.7 million. 2. This component would be financed by IDA (US$15.064 million) and the Multi-Donor Trust Fund for Health Results Innovation (US$10.636 million). It will support pilot PBF implementation in nine selected health prefectures through the contracting of performance- purchasing agencies (PPAs). This support would have the following two dimensions: 3. Subcomponent 1.1: Performance payments to health facilities (US$22.56 million): Provision of performance-based financing for carrying out of specific development projects, to provide MCH packages and technical support as well as capacity building activities to the health centers in the targeted districts. 4. This subcomponent would directly pay for PBF grants. The Government of the Central African Republic will contract with PPAs to provide MCH, technical support as well as capacity building services in the targeted districts. Those services will be provided through contracting health facilities at regional, district, and local levels. In the nine selected prefectures, the PPA will contract out with rural public, FBO and NGO health care facilities. Result-based grant will be paid to these health facilities in proportion to, and in payment for, achieved results. Facility payments will be based on (i) the volume of predefined technical support services delivered to health care centers and MCH services delivered to the targeted population and (ii) the technical quality of these services. Facility payments will be made quarterly after service volumes have been verified and quality of technical support and care has been assessed and certified by the PPA. After the quantity and quality of services provided are certified, payment will be released to contracted health facilities via the PPAs by the project management unit. 5. Subcomponent 1.2: Support to PBF implementation and supervision (US$3.14 million): To ensure the smooth implementation of PBF, this subcomponent finances activities related to PBF implementation and supervision through the PPAs. Activities will include (i) provision of training to strengthen the capacities of the health service providers in the efficient and effective provision of the technical support and MCH packages and strengthen the capacities of the staff of the recipient working in the health sector in the efficient and effective management of the delivery of technical support packages and MCH packages; (ii) development, monitoring, and evaluation of the provision of technical support packages and MCH packages in targeted districts, including through verification activities and surveys of the recipients of the technical support packages and the MCH packages; and (c) support coordination efforts among local, district, and regional stakeholders. 34 6. Given the shortage of technical capacity and human resources within the MOH, the project will identify international NGOs with demonstrated experience with PBF to act as the PPAs. International institutions / NGOs with relevant experience to develop PBF will be hired to be the performance-purchasing agencies (PPAs) for the four regions where the project will be implemented. There will be one international institution/NGO for two close regions (Health Region 2 and 3, and Health Region 4 and 6). 7. The project will provide each PPA with the resources to implement results based financing within its area of responsibility. This will be done by (i) providing training and ongoing capacity building, (ii) contracting with health facilities, (iii) strategically purchasing services, (iv) supporting coordination efforts, (v) conducting verification activities, and (vi) implementing community client satisfaction surveys. The PPA will contract with health posts, health centers, and prefecture- and regional-level hospitals to deliver an agreed package of MCH services. Facility payments will be based on (a) the volume of predefined MCH services delivered and (b) the technical quality of these services. Payment rates will also be adjusted for factors such as remoteness of health facilities. Facility payments will be made quarterly after service volumes have been verified and quality of care has been assessed and certified by the PPA. Regional- and prefecture-level health offices will provide regulation services, including quality assessments of contracted facilities. After the quantity and quality of services provided are certified and request for payment received, the Project Implementation Unit (PIU) will transfer the funds for performance to the PPA. The PPA will release the payment to contracted health facilities in accordance with the terms of the performance-based contracts. An external evaluation agency (EEA) will perform ex post-facto verification of the quantity and quality of the services provided to the population (described in Component 2). 8. The EU and Cordaid experiences were used by the MOH for developing the list of health services to be purchased and the level of (partial) reimbursement of unit costs via PBF payments. The list approved by the Ministry of Health, as of March 2012, is presented in table A2.1 and is subject to revision on an annual basis Table A2.1 Project services and indicators No. Indicator Price (US$$) Primary Care Services 1 External consultation (new cases) 0.28 2 Treatment of HIV/AIDS opportunistic infections 0.40 3 Hospitalization days 0.58 4 Child under 1 year old completely vaccinated 3.60 5 Pregnant woman completely vaccinated (tetanus) 1.40 6 Vitamin A distribution 0.28 7 Referral of positive case of TB 2.00 8 Referral to secondary care 2.20 9 Assisted delivery 4.00 10 Family planning (pill + injection) 2.00 11 Family planning (implant) 0.00 12 Pre-natal consultations (1 new or 3 returning patients) 1.40 35 13 HIV VCT: test 2.00 14 HIV VCT: HIV+ cases referred to district hospital 2.00 15 PMTCT: pregnant woman tested 2.00 16 PMTCT: HIV+ cases referred to district hospital 2.00 17 PMTCT: HIV+ women treated (AZT + Niverapine) 2.15 18 PMTCT: HIV+ newborn treated (AZT + Niverapine) 2.15 Secondary Care Services 19 Consultations provided by doctor 1.40 20 Referral from primary care 1.40 21 Major surgery performed 22.00 22 Minor surgery performed 6.00 23 Blood transfusion (donor-recipient) 2.20 24 Treatment of positive case of TB 20.00 25 Assisted delivery (normal) 2.80 26 Assisted delivery (Cesarean) 22.00 27 Assisted delivery (complicated, non-Cesarean) 16.00 28 Family planning (ligatures) 16.00 9. The PPAs will submit a performance-based financing manual (PBF manual) outlining the details of implementation, organization, administration, monitoring and evaluation, environmental and social monitoring and mitigation, financial management, disbursement and procurement. Based on best practices of results-based financing and adapting them to local conditions, each health service will be priced using a model that ensures that health centers are motivated to produce the optimal quantity and quality of the target indicators. Unit costs of each service to be purchased will be specified in the PBF manual. 10. The total population to cover under the project is 2.5 million, which is close to two- thirds of CAR's population. The project is expected to be implemented in nine prefectures located in four out of seven Health Regions (HR) in CAR: HR 2, prefectures of Mamb6r6 Kad6i and Sangha Mba6r6: 542,542 inhabitants, excluding Nana Memb6r&, where Cordaid is already implementing PBF for three years; HR 3, prefectures of Ouham and Ouham-Pend6: 931,346 inhabitants); HR 4, prefectures of Ouaka, K6mo and Nana Gr6bizi: 597,367 inhabitants; and HR 6 (Prefectures of Mbomou and Basse-Kotto: 481,157 inhabitants). In HR 6 (very far from Bangui), the plan is that the project will take over PFB intervention from the EU, whose support ends in 2012. HR 5 is too insecure for PBF intervention, while HR 1 (the closest to Bangui, and also benefiting from EU intervention) and HR 7 (Bangui and surrounding localities) are not targeted by the current project. (See table A2.2.) 36 Table A2.2: Targeted districts, population, and number of health facilities, 2011 Region Prefecture Population # Health facilities Total (2011) Hospital Health Health Center Post 2 Mamb6r6 424,833 Kad6i 2 20 12 34 2 Sangha 117,709 Mba6r6 1 11 14 26 3 Ouham 429,987 1 18 33 52 3 Ouham-Pend6 501,359 2 11 13 26 4 Ouaka 322,251 1 22 32 55 4 K6mo 137,910 1 10 19 30 4 Nana Gr6bizi 137,206 1 5 22 28 6 Mbomou 191,002 0 21 26 47 6 Basse-Kotto 290,155 1 15 28 44 Total 2,552,412 10 133 199 342 Component 2: Strengthen monitoring and evaluation capacity and support project implementation unit ($2.5 million) 11. Component 2 would be supported by the IDA ($1.936 million) and the Multi-Donor Trust Fund for Health Results Innovation (HRITF) ($.564 million). It will have two subcomponents: 12. Sub-component 2.1: Support the overall monitoring and evaluation of the project by PBF technical team at the central level and the External Evaluation Agency to carry out its activities which will include (a) monitoring and evaluation of Component 1 of the project through independent verifications of the delivery of technical support packages and MCH packages by each health service provider under its respective MCH subproject; and (b) strengthening of the recipient's capacity to monitor and evaluate the delivery of the technical support packages and the MCH packages. 13. Sub-component 2.2: Project Implementation Unit support for project coordination, implementation, and management. Subcomponent 2 will finance goods, consultant services, some training, and operating costs. In addition, minor rehabilitation work will be undertaken under this sub-component to improve the PIU and PBF Technical Unit's offices. 14. Strengthening monitoring and evaluation capacity to enable the health information system to generate up-to-date, reliable data to measure both PBF and general health system performance: At present, the health sector routine reporting system is very fragmented and the data generated are too unreliable and inaccurate to use as a basis for measuring health facility performance. Since PBF payments are to be made based on service volumes and quality, having accurate data on these parameters is critical to effective implementation. To avoid over- reporting, the external reviewers will collect primary data to measure achieved results. As there is a high probability that PBF facilities would manipulate their routine data to artificially inflate 37 their results (and therefore their PBF payments), ex post verification activities will be conducted by an independent third party, an external evaluation agency (EEA) such as the Department of Public Health at the University of Bangui. The EEA will be contracted by the MOH to check the veracity of the information provided by health services. The EEA's roles will include (i) random ex-post verification of service volumes and quality delivered by contracted health facilities, verified through a combination of facility record checks, technical quality assessments (using a checklist), home visits to a subsample of facility clients, and interviews with community group members to assess perceived quality; and (ii) building in-country M&E capacity. The EEA will partner with community-based organizations to conduct targeted household surveys. Data collected by the PPA and EEA will not only serve to verify results and validate PBF payments, but will also contribute to strengthening the health management information system through more routine and higher quality reporting by health facilities that sign PBF contracts. 15. Furthermore, the Government of CAR wants to use evidence-based information to decide if PBF is worthy to be extended countrywide. Component 2 will therefore include an Impact Evaluation using health facility and population-based household surveys. This Impact Evaluation is funded by a Bank-executed Trust Fund (TF) that is separate from this IDA funding and in addition to the Health Results Innovation Trust Fund (HRITF) funding of US$11.2 million. It is a separate HRITF grant of US$1.5 million. Below is a description of the objective of the Impact Evaluation and how it will be used by the Government to determine the merits of PBF. 16. The results of the PBF component of the project will be assessed through a full Impact Evaluation (IE). As PBF is being implemented in three prefectures in CAR but has never been systematically evaluated, the IE will be implemented in seven of the nine prefectures the project will cover (the prefectures of Mbomou and Basse-Kotto in Region 6 are excluded because they are under an EU PBF). The IE prefectures are Mamb6r6 Kad6i and Sangha Mba6r6 (Health Region 2); Ouham and Ouham-Pend6 (Health Region 3); and Ouaka, K6mo and Nana Gr6bizi in Health Region 4. 17. The policy objectives of the Impact Evaluation are to (i) identify the impact of PBF on maternal and child health service coverage and quality, (ii) identify key factors responsible for this impact, and (iii) assess cost-effectiveness of PBF as a strategy to improve coverage and quality. The results from the Impact Evaluation will be useful to fine tune the designing of the national PBF policy in CAR and will also contribute to the larger body of knowledge on PBF. 18. The Impact Evaluation will collect data on service coverage and health behaviors using household surveys, while facility surveys will be implemented for the quality indicators. Both household and facility surveys will be conducted by a third-party research firm that is not involved in any aspect of PBF implementation. 19. The study will have a pre-post with comparison design. It will rely primarily on experimental control to answer the research questions. Individual health facilities in each prefecture will be randomized to one of three study groups. Individual public and private health centers and health posts who meet pre-established criteria' in seven prefectures2 from the three I Not all health facilities (particularly Postes de Sante) provide the minimum package of services, nor are staffed with qualified health personnel, necessary for contracting with the PPAs. Only health facilities that meet criteria 38 pilot regions will be randomly assigned to each study group. This process of random allocation seeks to ensure that the two study groups are comparable in terms of observed and unobserved characteristics. Since prefecture hospitals play a critical role in supervising and acting as source of referral services for all facilities in the prefecture, all prefecture hospitals in these seven districts will be included in the full PBF (i.e., treatment) arm. 20. Tentatively, three study groups have been defined in the concept note of the impact evaluation as follows: * Group T1: Full PBF Package: Facilities in group TI will implement this full PBF package with health worker performance bonuses; * Group Cl: Full PBF Package but with PBF payment levels at 50% of Group T1 payment levels * Facilities in group TI will implement this full PBF package with health worker performance bonuses, but payment levels for targeted health services will be 50% lower than those found in Group Ti. For example, if the unit price for a child fully vaccinated is $5.00 in Group Ti, the unit price will be $2.50 in Group C1. * Group C2: Status quo: C2 facilities will be the 'business as usual' facilities and will not receive any additional resources, inputs or supervision. C2 facility managers will not have the autonomy to hire staff or financial autonomy. Table A2.3: Study groups and sample size T1: PBF with health worker performance Cl: PBF with health worker performance bonuses at 100% PBF payment levels bonuses at 50% PBF payment levels 80 facilities* 80 facilities* 1280 households (16 households per facility 1280 households (16 households per facility catchment area) catchment area) C2: Status quo 80 facilities* 1280 households (16 households per facility catchment area) 21. Separate HRITF funding will cover all primary activities related to the PBF impact evaluation. These include data collection, data analysis and report writing, and information dissemination activities. A baseline survey will be conducted prior to the piloting of PBF in the seven targeted prefectures. An endline survey will be conducted two years after the baseline survey. A mid-term documentation activity will also be conducted. Dissemination activities will include restitution of baseline and endline surveys results to key stakeholders in CAR, and peer- reviewed journal publication. Project Cost 22. The detailed project cost is as presented in table A2.3 and table A2.4. regarding service provision and health worker staffing will be included in the Impact Evaluation. These criteria will be formally established prior to the piloting of PBF and the Impact Evaluation. 2 As previously indicated, Mbomou and Basse-Kotto Prefectures in Region 6 are not included in the IE, although implementation of PBF in public and private facilities will be financed in these prefectures through the project. 39 Table A2.3: Estimated project costs Component Modality Estimated Cost (US$ million) Total IDA HRITF CPG Cost Credit/Grant Funds3 Performance payments to health facilities 22.56 13.6 8.96 __________________ (80%) (80%) (80%) Performance Based Support to PBF implementation and Payment supervision 3.14 1.464 1.676 (11.1%) (8.7%) (15%) Strengthening monitoring and evaluation 1.0 .436 0.564 capacity (3.5%) (2.5%) (5%) Technical Assistance PIU as well as minor rehabilitations of the 1.5 1.5 0 PIU and PBF Technical Unit's offices. (5.4%) (8.8) (0%) TOTAL 28.2 17.0 11.2 100% 100% 100% Table A2.4: Breakdown of HRITF financing HRITF Grant CY CY 2012 CY 2013 CY 2014 CY 2015 CY 2016 Total 2011 Preparation (Bank- 100,000 150,000 0 0 0 0 250,000 executed) Preparation 0 300,000 100,000 0 0 0 400,000 (Recipient- executed) Country Program 0 1,000,000 2,500,000 3,000,000 3,000,000 1,700,000 11,200,000 Support Pilot Grant (CPG) (Recipient- executed) Impact Evaluation 0 600,000 150,000 600,000 150,000 0 1,500,000 (Bank-executed) SPN (Bank- 0 60,000 60,000 60,000 60,000 60,000 300,000 executed) Total 100,000 2,110,000 2,810,000 3,660,000 3,210,000 1,760,000 13,650,000 3 The IDA Credit will finance approximately 60% of Project Costs and the HRITF funding will provide for 40%, not including the IE costs. 40 CENTRAL AFRICAN REPUBLIC: Health System Support Project Annex 3: Implementation Arrangements A. Project Administration Mechanisms 1. The Ministry of Health Population and Fight against HIV/AIDS (MOH) will implement the project through the Health, Education and AIDS Project (HEAP) Project Implementation Unit (PIU). The PIU will be placed under the Director of the Cabinet (DC), with technical support from the relevant MOH directorates. The HEAP is staffed by a multidisciplinary team including nine members: a project manager, a financial and administrative management specialist, an accountant, a procurement specialist, two NGO/CBO specialists (who monitor the performance of health NGOs that were awarded contracts to implement community- based microprojects), one administrative assistant, and two support personnel. Members of the PIU have developed the skills and experience needed for project implementation through the implementation of the Bank-financed Multisectoral HIV/AIDS Project (P073525). Originally formed in 2005, the PIU was reorganized in 2009, leading to improved performance of the overall project. This restructuring was an important factor in having the project's rating change from "unsatisfactory" to "satisfactory" in 2010. Under the new PIU management, the project benefited from a second extension of the closing date granted in September 2010 for eighteen more months. For the Health System Support Project, the Government has extended the contract and revised the terms of reference of the following staff of the PIU: the project manager, the financial and administrative management specialist, the accountant and the procurement specialist. The PIU (under DC supervision) will carry out all of the disbursements and any procurement related to the project, in accordance with IDA-approved procedures. It will serve as the funds holder, managing the operational accounts and transferring money to the PPAs and submitting withdrawals to IDA. See figure A3. 1. 2. For Component 1, two performance-purchasing agencies (PPAs) will be contracted by the Ministry of Health (one for Health Regions 2 and 3 and one for Health Regions 4 and 6) to manage the purchasing of predetermined technical Support from regional and district public facilities and MCH from rural public, FBO and NGO facilities. The PIU will provide the PPA with the resources to pay health facilities for predetermined health services delivered to the population and technical services delivered to health centers, according to the norms and standards of service as defined by the Ministry of Health (subcomponent 1.1). The PPA will also receive resources for project support and verification activities (subcomponent 1.2). The PPA will contract with health posts, health centers, and district-level (prefecture-level) and regional-level hospitals that meet predetermined criteria to deliver an agreed package of technical support and MCH. Payments will be based on (i) the volume of predefined technical support and MCH services delivered; (ii) the technical quality of these services; and iii) the remoteness of the health facility. 3. For Component 2, strengthening project management and coordination including monitoring and evaluation capacity (and support PIU costs), an independent third party, or external evaluation agency (EEA), will be contracted by the MOH. 41 4. At the periphery level, quality of care will be evaluated by prefecture-level health teams. In order to ensure that facilities are focusing on the quality of the care that they provide, the payments given to the facility will be adjusted according to a quality score for the quarter. The quality score will be based on an assessment carried out at the facility conjointly by prefecture health teams in the case of health centers and by a peer-review process involving prefecture hospitals in the case of hospitals. These assessments will be carried out quarterly and will rate the facility on the basis of a broad variety of indicators such as health service provider behavior, the application of norms and protocols, the availability of staff and medicines, cleanliness, and the quality of recordkeeping. Based on the results of this assessment, the PPA will calculate a quality score for the facility and will adjust its payment to the health facility accordingly. 5. At the central level a steering committee, including a PBF technical team, will guide PBF implementation to draw lessons and policy implications for the health sector. It will be chaired by the Director of Cabinet of the Ministry of Health and include key directorates of the MOH, the Ministry of Planning and Economy, the Ministry of Finance, and representatives from the donor community. The Bank will not sit as a member of the committee but only as an observer, to avoid going beyond its mandate. The steering committee will oversee the implementation of PBF, document the lessons learned from various initiatives of PBF in the country, provide guidance to the PIU and PPA, and generate policy direction for the institutionalization of PBF in CAR. In particular, it will approve the annual work plans of the PIU, the national PBF framework document, and other key documents. 6. The project policies and procedures are incorporated in an implementation manual that was adopted by the PIU before negotiations. It will be completed by a series of PBF manuals, each one prepared by a PPA as part of its offer of services and improved as practical experience develops from in-the-field implementation,. Figure A3.1: PBF implementation structure Fund transfer Develome t ov. of CAR - Verification DeelpmntiInistryof - - -- Contract/Convention iartners Healt a Health service provision orsupervision Steering Commitee Performance Project External Evaluation Purchasing Agency Implementation Unit Agency (EEA) (PPA) EXTERNAL INTERNAL VERIFFICATION VERIFICATION District Hospital H-ealth Service Health Center/Post clients PROVIDER POPULATION 42 B. Financial Management and Disbursement Arrangements Conclusion of the assessment 7. Overall, the residual financial management risk for the project is assessed as "high." The Financial Management arrangement below indicates that they satisfy the Bank's minimum requirements under OP/BP 10.02. CAR Health System Support Project will implement the FM action plan below during the course of preparation and implementation as part of strengthening its financial arrangements to provide with reasonable assurance that the Credit and Trust funds will be used for the intended purpose. Budgeting 8. Budget arrangements are developed with the procedures manuals and are adequate to support the project. The procedures will ensure that budget variance analysis will be carried out to ensure budget variances are adequately addressed in timely manner by the coordination unit. Accounting arrangements 9. Financial management (FM) manual: A manual was developed as part of the project implementation manual, with a separate one for partners involved in the implementation of the PBF mechanism. It includes the development of social accountability and reporting mechanism (display of credit allocations, payments made, and third party monitoring). 10. Staffing arrangements: The current HIV/AIDS project which closed in March 2012, has experience in implementing decentralized projects and an experienced finance manager and assistant accountant having implemented a decentralized project. Based on their success with the previous project, staff from the current PIU staff was retained; their contracts extended and their terms of reference were revised to cover the responsibilities under the new project. The FM will be strengthened with the recruitment of another accountant during implementation in view of the potential number of operations in this project. 11. Books of accounts: The finance manager and the assistant accountant will be responsible for setting up and maintaining the CAR Health System Support Project accounts and preparation and submission of the action plans with inputs from partners for accessing the project proceeds and reports. Setting up the accounting system will be described in the financial management manual and will be used to track, record, analyze, and summarize the Health System Support Project's (HSSP's) financial transactions. HSSP's accounts will be prepared in conformity with the OHADA accounting framework and the provisions of the financing agreement. Appropriate internal controls over the preparation and approval of transactions and recording and consolidated reporting will be documented in the manual to ensure that all transactions are correctly made. 12. Information systems: The resources of the current HIV/AIDS project which closed in March 2012 will be redeployed into the new PIU. It also uses Tompro accounting software to 43 maintain its books of accounts. The new PIU will inherit the software existing in the HIV/AIDS project closing in early 2012 if the licensing permits or a new software will be acquired before implementation altogether within 3 months after effectiveness. Financial reporting 13. CAR HSSP will issue consolidated quarterly interim financial reports (IFRs) and financial statements at the end of each fiscal year. A first level of consolidation will be performed by the PPAs. The formats and contents for the IFRs will be prepared by the PIU and agreed upon with the Bank. These will include and not be limited to (i) a sources and uses statement; (ii) financial assets and liabilities statement if prescribed by the OHADA accounting framework; (iii) uses of funds by activities for the project-inter alia, the type of service, the price, the total cost, and the center/district-(iv) a segregated account reconciliation, and (v) accounting policies and explanatory notes adopted by the project. These IFRs will be submitted to the Bank 45 days after the end of quarterly period. These terms have been agreed upon with the Bank during project negotiations. Internal control and internal audit 14. Multiple internal control mechanisms used by the PPAs involved may not be adequate to support the financial operations of the project. A financial management manual will be developed for project and a separate one for PPA involved with the decentralized agencies. The aspects covered in the financial management manual will include (i) flow of funds; (ii) financial and accounting policies; (iii) accounting system (including centers for maintenance of accounting records, chart of accounts, formats of books and records, accounting and financial procedures); (iv) procedures for authorization of transactions including ex ante control, budgeting, and financial forecasting; (v) financial reporting; and (vi) auditing arrangements. 15. A weak audit environment is prevailing in CAR and hence an insufficient internal audit support. The financial management manual for the PPA will include social audit mechanisms that will involve verifications of the performance from beneficiaries receiving health services. After the quantity and quality of services provided are certified by the PPA, payment will be released to contracted health facilities via the PPAs by the project management unit. To this end, the PPA should meet some specifics criteria (strong internal audit function, appropriate FM capacity at both central and decentralized level, and appropriate manual of procedures) to be detailed in the ToRs. In addition, EEA will be recruited to perform ex-post physical verification of service volumes and quality delivered by health facilities. Audit arrangements 16. The annual financial statements prepared by the PIU as well as internal control system will be subject to an annual audit by auditor listed at CEMAC based on terms of reference that will include the review of PPAs' procedures and an acceptable sample of health centers. The auditor will provide one single opinion on the annual financial statements in compliance with the International Federation of Automatic Control (IFAC) Standards on Auditing. In addition to the audit reports, the external auditors will be expected to prepare a management letter giving observations and comments, and providing recommendations for improvements in accounting 44 records, systems, controls, and compliance with financial covenants in the financing agreement. The project will be required to produce, no later than June 30 of the following fiscal year, audited annual financial statements. In line with the new access to information policy, the project will comply with the Bank disclosure policy of audit reports-make them publicly available promptly after receipt of all final financial audit reports (including qualified audit reports) and place the information on the project's official Web site within one month of the report being accepted as final by the team. Disbursement arrangements 17. A joint financing approach will be used for CAR HSSP in which two segregated Designated Accounts will be opened to receive the proceeds of both IDA and Health Results Innovations Trust Fund grants as follows: the first account (DA-A) will be used for PBF activities which are financed jointly by IDA and the HRITF while the second account (DA-B) will be used for all other activities. The combined ceiling of the pooled DA-A will be CFAF 890 million split as follows: CFAF 540 for IDA and CFAF 350 million for the HRITF. Upon effectiveness of the credit and the Trust Fund, an initial advance up to the ceiling of each DA will be deposited into the Designated Accounts. The Designated Accounts will be denominated in CFAF and held in Banque Populaire Maroco-Centrafricaine which is acceptable to IDA. The PPA will open and operate two transaction accounts in commercial banks, acceptable to IDA, one for operating costs and one for results-based payments. Funds will flow to those accounts from the central Designated Accounts, which will be opened and managed by the PIU. (See figure A3.2.) For the first 12 months, the project will use transaction-based disbursement procedures for the centrally located designated accounts. After a fiduciary assessment, the project might switch to the report based disbursement procedures whereby the ceiling of the DA is set on the basis of six-month cash forecasts derived from the approved work plan for the initial advance and from unaudited interim financial reports (IFRs) thereafter. . The advance will be documented according to the frequency of submission of the IFR. The IFR will include a certified statement of expenditures on PBF proceeds. Disbursement will be made pari passu from IDA and the HRITF for PBF related expenditures. Disbursements between the PPAs and the health centers will be made against certified statements of expenditures and for services already rendered based on PBF determined unit price that takes into account, inter alia, quality and cost. The certified Statement of Expenses (SOE) verified by the PPA will be submitted to IDA in support of applications for withdrawal of proceeds under the PBF components. 18. Disbursements to the decentralized levels will be supported by PPA using several mechanisms, which will include but not be limited to using bank networks and established NGOs to be agreed upon with IDA; these will be recruited, where there is low banking penetration. Transactions will be substantiated by source documents to be maintained both at the central PIU level and the PPA level. A standardized detailed sheet (Statement of Expenditures) will be maintained by the PPA for all centers/districts substantiating financial information transmitted for disbursement purposes as the basis of payment of their performance. 19. The project's authorized accounts signatories will include the project coordinator and the financial management specialist at the PIU level. At PPAs and health facilities level, the signatories will be described in the manual of procedures. 45 Figure A3.2: Funds Flow Chart IDA IDA HRITF Financial Reporting IF 's Designated Designated Account Account B (CFA) A (pooled account) PIu (CFA) PIU PPAs (regional level) Two Transaction Accounts per PPA Health Centers PIu's Sirvices providers and Services providers and suppliers sunDliers Disbursements by category 20. Table A3.1 sets out the expenditure categories to be financed out of the credit proceeds. This table takes into recognition the prevailing country financing parameter for CAR in setting out the financing levels. Component I will be fully co-financed by IDA and HRITF based on the percentage described in the table below. Component 2, the PIU costs will be fully paid by the IDA funds whereas the M&E activities will be co-financed by the IDA and HRITF funding. 46 Table A3.1: Project expenditure categories IDA IDA Credit/ IDA Designated IA Amount of Grant! con Credit/Grant Amount HRITF / Percentage of Account Amount of Amount of the Expenditures Financing ofthe Categories the Finacn Fthe Financing to be Financed Financing (US$ Allocated Allocated (US$ (inclusive of Allocated mlin (US$ million) taxes) (US$ million)(US million) (1) Goods, non- consulting services, consultants' services, 13.6 7.480 6.120 8.96 60.3% [IDA] A Training and 39.7% Operating Costs under [HRITF] component 1.1 (Performance payments to health facilities) (2) Goods, non- consulting services, consultants services, 1.464 .8052 .6588 1.676 46.6% [IDA] A Training and 53.4% Operating Costs under [HRITF] component 1.2 ( Support to PBF implementation and supervision) (3) Goods, consulting services, .436 .2398 .1962 .564 43.6.% [IDA] A Training under 56.4% component 2.1 [HRITF] (4) PIU costs as 1.5 .825 .675 0 100% [IDA] B well as minor rehabilitations of the PIU and PBF Technical Unit's offices. Total Amount 17 9.35 7.65 11.2 Financial management action plan 21. Table A3.2 below indicates the actions agreed upon with the CAR HSSP that will be taken to strengthen its financial management system. Relevant dates for completion and responsibilities are also indicated. 47 Table A3.2: Project financial management plan Action Date due by Responsible 1 Develop a financial management manual for Negotiations/ HSSP project Done Recruit a PPA based on ToR acceptable to the A condition of HSSP Bank disbursement, 2 expected to be completed three months into effectiveness 3 Inherit or acquire new accounting software within Three months HSSP three months of effectiveness. into effectiveness 4 Agree on terms of reference for the audit of HSSP Negotiations/ HSSP/ IDA Done 5 Auditor recognized in the CEMAC region will be Six months after HSSP recruited effectiveness 6 Agree on report formats and contents that will be Negotiations/ HSSP/ IDA prepared by the PIU Done C. Procurement Arrangements Guidelines 22. Procurement for the proposed project would be carried out in accordance with the World Bank's Guidelines: Procurement of Goods, Works, and Non-Consulting Services under IBRD Loans and IDA Credits & Grant by World Bank Borrowers, dated January 2011, and Guidelines: Selection and Employment of Consultants under IBRD Loans and Credits & Grants by World Bank Borrowers, dated January 2011, and the provisions stipulated in the Financing Agreement. The general description of various items under different expenditure categories are described below. For each contract to be financed by the proposed financing, the different procurement methods or consultant selection methods and the need for prequalification, estimated costs, prior review requirements, and time-frame are agreed between the Borrower and the Bank in respective procurement plans. The procurement plans will be updated at least annually as required to reflect the actual project implementation needs and improvements in institutional capacity. 23. National competitive bidding (NCB) procedures may be used provided that (i) bids are advertised in national newspapers with wide circulation ; (ii) bid evaluation, bidder qualification, and award criteria are specified clearly in the biddings documents; (iii) bidders are given adequate response time (minimum four weeks) to prepare and submit bids; (iv) bids are awarded to the lowest evaluated bidder proven this bidder is qualified; (v) eligible bidders, including foreign bidders, are not precluded from participating; and (vi) no preference margin is granted to domestic suppliers. 48 Procurement Documents 24. The procurement will be carried out using the Bank's standard bidding documents or standard request for proposal (RFP) respectively for all ICB for goods and recruitment of consultant. For NCB, the borrower shall submit a sample form of bidding documents to the Bank for prior review and use this type of document throughout the project once agreed upon. The Sample Form of Evaluation Reports developed by the Bank, will used. Advertising Procedures 25. In order to get the broadest possible interest from eligible bidders, a general procurement notice (GPN) will be prepared by the borrower and published in the United Nations Development Business online (UNDB online), in the Development Gateway Market (dg Market), and in national newspapers of wide circulation to advertise for major consulting assignments above US$200,000 equivalent through ICB. The borrowers will keep a roster of the responses received from potential bidders interested in the contracts. 26. Specific procurement notices (SPN) for goods to be procured under ICB and NCB and for consultant services will be published in at least one national newspaper of wide national circulation. Moreover, for all ICB procurement and all requests for expressions of interest relating to large consulting services, above US$200,000 equivalent, the SPN will be published in accordance with advertising provisions in the following guidelines: Procurement of Goods, Works, and Non-Consulting Services under IBRD Loans and IDA Credits & Grant by World Bank Borrowers, dated January 2011, and Guidelines: Selection and Employment of Consultants under IBRD Loans and Credits & Grants by World Bank Borrowers, dated January 2011 Procurement Methods 27. Procurement of Civil Works. No civil works are planed under this operation. 28. Procurement of Goods. Goods procured under this project would include software IT equipment, communications, vehicles, furniture and offices supplies, etc. Goods procurement will be done using the Bank's SBD for all ICB and national SBD agreed with the Bank for NCB). 29. International Competitive Bidding. Each goods contract package estimated to cost US$500,000 equivalent and more per bid package would be procured through ICB. 30. National Competitive Bidding (NCB). Goods estimated to cost less than US$500,000 equivalent and locally available at commercial price would be procured through National Competitive Bidding (NCB) procedures acceptable to IDA. 31. Shopping. Procurement for readily available off-the-shelf goods that cannot be grouped, or standard specification commodities for individual contracts of less than US$50,000 equivalent, may be procured under shopping procedures as detailed in paragraph 3.5 of the 49 Guidelines: Procurement under IBRD Loans and IDA Credits, dated January 2011, and the Guidance on Shopping Memorandum, issued by IDA, June 9, 2000. The requests of solicitation will specify that offers should be submitted sealed and opened at a public opening session. Direct contracting for goods may be used only on an exceptional basis and with the prior approval of the Bank in accordance with the provisions of paragraph 3.9 of the Procurement Guidelines. 32. The selection of consultants would be as follows. Firms: Consultancy services will be selected using request for expressions of interest, short-lists, and the Bank's standard RFP, where required by the Bank's guidelines. The selection method would include Quality and Cost-Based Selection (QCBS) whenever possible, Quality-Based Selection (QBS), Fixed Budget (FBS), Least Cost Selection (LCS), and Single Source Selection (SSS) as appropriate; all consultancy services contracts estimated to cost less than US$100,000 equivalent for firms could be awarded through Consultant's Qualifications (CQ). 33. Individual consultants: Specialized advisory services would be provided by individual consultants selected by comparison of qualifications of at least three candidates and hired in accordance with the provisions of Section V of the Consultant Guidelines. 34. Short-lists of firms for services estimated to cost less than US$100,000 equivalent per contract may be composed entirely of the national consultants in accordance with the provisions of paragraphs 2.7 through 2.8 of the Consultant Guidelines. SSS may be used exceptionally in accordance with paragraph 3.9 to 3.12 of the Consultant Guidelines. 35. Training, Workshops, Seminars, and Conferences: Selection of consulting firms for training services estimated to cost US$100,000 equivalent or more shall be procured on basis of QCBS or QBS as appropriate. Training services estimated to cost less than US$100,000 equivalent per contract may be procured through CQS method. When appropriate, training may also be procured on the basis of direct contracting subject to review and approval by the Bank. The training (including training material and support), workshops, conference attendance, and study tours will be carried out on the basis of approved annual programs. 36. Operating costs: The operating costs shall include staff, travel expenditures, and other travel- related allowances with prior clearance from IDA. Operating costs financed by the project will be procured using the implementing agency's administrative procedures described in the project implementation manual reviewed and found acceptable to the IDA. Other Methods of Procurement 37. Procurement from PBF Proceeds. Each beneficiary of PBF will procure goods, works, and services in accordance with provisions of administrative procedures described in the project implementation manual reviewed and found acceptable to the IDA. 38. Country Issues. On June 6, 2008, the Government enacted a new Procurement Code and institutional framework in compliance with OECD/DAC recommendations on public 50 procurement reform. Procurement regulatory and prior review functions are carried out by adequate entities, and decentralized procurement is assigned to line ministries and agencies. Assessment of the Agency's Capacity to Implement Procurement. 39. During the preparation phase of the Health System Support Project (HSSP), a procurement assessment was carried out by the procurement specialist based in the CAR Country Office. The assessment reviewed the organizational structure, the procurement process and management, and the interaction between the PIU and other entities involved in procurement process. This assessment revealed that the PIU of the current health sector project (HEAP) was well staffed by a procurement specialist. Procurement implementation arrangements 40. Procurement activities and overall fiduciary responsibility at the central level will be implemented by the PIU of the HEAP, which is a structure of the Ministry of Health. The ongoing PIU, a redefined version of the existing unit, is likewise being established under the Ministry of Health and will be responsible for the project implementation. 41. Procurement activities at the central level will be implemented by the PIU of the HEAP, which has substantial experience in the implementation of project financed by the Bank. The PIU will be responsible for the coordination of all procurement activities. All project procurement prior review documents shall be submitted to IDA through PIU. The procurement specialist will oversee and manage the project's activities and will ensure that these activities are proceeding in a timely manner and according to project objectives. The specialist will be responsible for the preparation of project's procurement reports every other month. 42. Procurement from PBF Proceeds. There will be very small procurement activity at the decentralized level. Each health district may use part of its subsidies to improve quality of service by procuring small items estimated at less than US$20,000 per contract. Any beneficiary of PBF resources should procure small goods, works and services in accordance with provision of Procurement Guidelines, and will be responsible of compliance of procurement procedures. Procurement Plan 43. The PIU, on behalf of the Government, has developed a draft detailed and consolidated Procurement Plan for project implementation, which provides the basis for the procurement methods. The plan covers the first 18 months of project implementation. This plan has been agreed to by the beneficiary and the IDA Project Team during the negotiations and is available at the implementation agency. The Procurement Plan will be updated in agreement with the Bank at least annually in conjunction with an updated Project Annual Work Program or as required to reflect the actual project implementation needs and improvements in institutional capacity improvements. 51 Overall Procurement Risk Assessment High Moderate X Low Risks and Mitigation Measures 44. In view of the experience of the PIU in carrying out procurement activities for the HEAP, the overall project procurement risk has been rated as moderate. The following measures were agreed upon to mitigate the remaining risks: Task Completion Responsibility i. Submit to the Bank the first Done Project 18 month procurement plan Preparation Team ii. Organize a procurement workshop focusing on First Project shopping for Health districts semester Implementatio benefit into n Unit effectivenes s Fraud, Coercion, and Corruption 45. All procuring entities, as well as bidders, suppliers, and contractors, shall observe the highest standard of ethics during the procurement and execution of contracts financed under the project in accordance with paragraphs 1.15 & 1.16 of the Procurement Guidelines and paragraphs 1.25 & 1.26 of the Consultants Guidelines. Frequency of Procurement Supervision Missions and Audits 46. In addition to the prior review to be carried out from World Bank offices, the capacity assessment of procurement activities has recommended supervision every six months and at least one Bank annual Post Procurement Review (PPR); the ratio for post-review is at least one out five contracts. The independent post review could be conducted through a consultant selected by the Bank at any time until two years of the closing date of project. The PIU shall send to the Bank, every six months, the procurement management report with details on implementation progress, problems, and solutions proposed. 52 Table A7.1: Procurement and selection review threshold Expenditure Contract value (US$) Procurement Contract Subject to Category (Threshold) Method Prior Review 1. Works >3,000,000 ICB All contracts < 3000,000 NCB Review of the first two contracts from all entities in charge of procurement < 50,000 At least three quotations No Threshold Direct Contracting All contracts 2. Goods > 500,000 All contracts < 500,000 NCB Review of the first two contracts from all entities in charge of procurement < 50,000 Shopping No Threshold Direct Contracting All contracts 3. Consultants > 200,000 QCBS All contracts Firms < 200,000 LCS; FBS Review of the first two <100,000 CQ contracts from all entities in charge of procurement Individual IC (at least 3 CVs) All contracts of US$ 50,000 and more No Threshold Single Source All contracts 4. Trainings and No Threshold On basis of detailed and approved annual Workshops plan (with indication of venue, number of participants, duration, exhaustive budget, etc.) D. Social and Environmental (including safeguards) Social 47. The project is expected to have a positive social impact by improving access to health care services for the poorest households. Component 1 (through the PBF mechanism) will provide incentives for health facilities to reduce staff absenteeism, to improve staff responsiveness with patients, and to provide more and better care for the poorest patients. 48. The project will also enhance community ownership for monitoring the quality of basic health services. Community-based organizations (CBOs) will be identified and strengthened so that they can be involved in monitoring health facilities. The project will build on the experience of the HEAP project, which had successfully implemented a strong community- 53 based microproject component involving a great number of CBOs and community Health associations. 49. The preparation process of the project has been highly participatory with extensive work and consultation among the key stakeholders: selected line ministry representatives and representatives of donor community, including NGOs. To avoid opposition to PBF, a strong effort was made very early in project preparation to explain the project to stakeholders through workshops and meetings. The monitoring and evaluation system has been designed to ensure the adequate targeting of project activities, including their social impact. 50. The project will be implemented in the area where indigenous peoples (IPs) are located. While the project will not as such negatively affect IPs, OP/BP 4.10 policy is triggered to ensure that IPs will benefit from the project. An Indigenous Peoples Plan was prepared and publicly disclosed in country as well as in the World Bank's Infoshop before appraisal on March 13, 2012. The IPP has the objective to define clearly how IPs will be taken into consideration in the project and what the measurement indicators will be. A specific indicator related to the number of contacts per capita for indigenous persons was included to motivate health professionals to pay particular attention to serving IPs. 51. A particular attention will be given to ensure active participation of women in project areas. The project is expected to have a positive impact on women and all females, not only those who are pregnant but all vulnerable women in general. Environment 52. The project triggers the Environmental Assessment (OP/BP 4.01) policy and has been assigned an Environmental Screening Category of "B," given the risks associated with the handling and disposal of medical and general health waste. Health facilities use medical equipment that might entail an increased production of medical waste. Despite this risk, the project is not expected to generate any major adverse environmental impact. Possible environmental risks include the inappropriate handling and disposal of hazardous medical waste, including sharp needles, and especially the inadequate management of disposal sites in urban or peri-urban areas, where domestic and medical waste may be mixed and where scavenging is common. 53. The Government prepared in September 2010 a Health Care Waste Management Plan as a condition to extending the closing date of the Health-Education-HIV/AIDS project. This plan is being implemented by the MOH and a revised plan has been prepared to take into account this project and has been disclosed before appraisal. In addition, health centers will report on hazardous medical material waste management since that action will be included as an indicator to be evaluated and purchased in PBF. The quantified quality checklist used by PBF to pay for performance on the quality measure, in fact, measures this element. The weighting for this aspect will be increased, and the adherence to the guidelines will be checked, and paid for, quarterly. These issues are not seen as an increasing threat in the context of this project, because of the comparative advantage of PBF to solving such traditionally intangible problems. 54 54. To deal with all environmental and social potential negative impacts related mainly to the health waste management, the PPA will be responsible for following up the implementation of all environmental and social mitigation measures described in two relevant safeguards documents. Two safeguards policies were triggered, as follows: Safeguard Policies Triggered by the Project Yes No Environmental Assessment (OP/BP 4.01) [X] [] Natural Habitats (OP/BP 4.04) [] [ X] Pest Management (OP 4.09) [ ] [] Physical Cultural Resources (OP/BP 4.11) [] [X ] Involuntary Resettlement (OP/BP 4.12) [] [] Indigenous Peoples (OP/BP 4.10) [X] [X ] Forests (OP/BP 4.36) [ [ X] Safety of Dams (OP/BP 4.37) [] [X] Projects in Disputed Areas (OP/BP 7.60)* [] [X] Projects on International Waterways (OP/BP 7.50) 11 [X] OP/BP 4.01 Environmental Assessment: This policy is triggered because of direct potential environmental impacts associated with the handling and disposal of medical and general health waste. The borrower prepared a Health Care Waste Management Plan in September 2010, which will continue to be applied to this project. The HCWMP has been re-disclosed prior to appraisal. The project will strengthen the capacities of health centers in handling health wastes. OP/BP 4.10 Indigenous Plan: This policy is triggered to ensure that IPs will benefit from the project even though this project is not expected to impact negatively IPs. An IP Planning Framework has been prepared and disclosed in Central African Republic and at the InfoShop prior to appraisal. Consultation and disclosure: Extensive consultations at the national and local level were undertaken during the safeguards documents preparation process. This consultation process will continue during the implementation. Borrower Capacity to Implement Safeguard Policies 55. The line borrower capacity for environmental and social safeguards implementation and reporting remains limited despite the implementation by the line ministries of the Health- Education-HIV/AIDS project financed by the World Bank. A limited understanding of the Bank's policies exists. Close technical support will be provided by the Bank's safeguards specialists during implementation to ensure compliance with all Bank safeguards policies as well as national policies. The capacity-strengthening measures were outlined in the project safeguards documents and integrated in the project budget, implementation, and monitoring plan. 55 Safeguards Supervision Plan 56. Given the borrower's limited but growing experience with implementation of environmental and social safeguards instruments, close safeguards supervision and implementation support will be carried out during the early stage of project implementation until adequate safeguards experience is developed. The PPA, the relevant health centers of the Ministry of Environment, and other relevant local government staff will supervise the implementation of the safeguards instruments discussed above. IDA supervision will focus on (i) providing regular implementation support, (ii) carrying out field reviews of safeguards implementation, and (iii) monitoring safeguards implementation based on periodic progress reports. IDA supervision will be carried out by field-based Bank technical staff and complemented by specialist consultants not only during regular biannual supervision missions but also during interim technical safeguards missions that will respond to emerging issues. Monitoring will be based on Health Care Waste Management Plan and IP Plan recommendations. 56 Central African: Health System Support Project (P119815) Annex 4: Operational Risk Assessment Framework (ORAF) Stakeholder Risk Rating Low Description: Risk Management: 1- Demand for maternal and child health services from targeted population remains low 1- even in facilities that implement PBF - Project Implementation Unit (PIU) will track trends on service utilization across all project facilities to identify potential problems with sustained low demand and identify possible solutions. 2- International NGOs with strong experience on PBF may not be interested in working in - Effective behavior change communication strategies targeting key women leaders and other decision makers in each CAR given the political instability and unattractive living conditions in the country. community will be implemented. 2- The request for proposals and other bidding documents to be sent to NGOs will incorporate and clearly communicate attractive financial conditions. Resp: Client Stage: Implementation Due Date: 16-Dec-2013 Status: Not Yet Due Capacity Rating Moderate Description: Risk Management: The current shortage of skilled human resources in rural areas could block smooth Mitigation measures include: (i) offering health workers performance-linked bonuses that are attractive enough for health professionals implementation of project activities. to relocate to and stay in rural areas, and (ii) contracting out with FBO and NGO facilities to strengthen and extend coverage of health services in rural areas Resp: Client Stage: Implementation Due Date: 30-Sep-2013 Status: Not Yet Due PBF strategy is not, as yet, widely understood by some senior staff at central level or by regional, district and health center personnel. Thus the absence of a critical mass of PBF Risk Management: supporters within the MOH increases the risk of opposition to its implementation, and the Health Results Innovation Trust Fund (HRITF) knowledge and learning grant was secured to create a policy dialogue which aims to risk of failure from poor implementation promote PBF among key stakeholders in the government and the donor community. Workshops will also be organized for health While timely in cash payment to health service providers is key to PBF success, the weak professionals at different levels of the system banking system in CAR may hamper the transfer of funds to health facilities in rural areas Resp: Bank Stage: Implementation Due Date: 16-Jun-2012 Status: Not Yet Due Risk Management: The project will explore alternatives to transfer money to such rural areas. One option would be to use intermediaries to transfer funds to rural areas. The on-going Health-Education-HIV/AIDS project, for instance, successfully used Catholic Churches as intermediaries to pay teachers' salaries in remote areas. Other options being explored include using PPA using several mechanisms which will include mobile phones to transfer funds to health centers Resp: Client Stage: Implementation Due Date: 30-Mar-2013 Status: Not Yet Due Governance Rating High Description: Risk Management: 57 The MOH will ensure that a PIU is in place before project effectiveness. The members of PRU will mainly be drawn from the PRU which Weak governance including in PBF contract negotiations and management (contract successfully turned around the procurement and financial management of the ongoing HIV/AIDS- Health-and Education project. award, monitoring & evaluation, fraud and corruption prevention, etc...) could result in NGO contracts that do not deliver value-for-money Resp: Client Stage: Implementation Due Date: 16-Apr-2012 Status: Completed Risk Management: Autonomy to health centers to spend PBF revenues may lead to mismanagement of funds Project health facilities will have to adhere to guidelines on how PBF revenues can be spent, and also to measures designed to increase transparency such as disclosure of information on how PBF revenues have been spent. PBF credit allocated, funds transferred and payments made will be displayed in health center to ensure transparency. Resp: Client Stage: Implementation Due Date: 16-Sep-2012 Status: Not Yet Due Risk Management: While the PPA will certify the performance of the health centers before any payment is made to the health service providers, the EEA will provide independent assessment and verification of the quantity and quality of services provided to the population based on ex-post evaluation including at the beneficiary level. Resp: Client Stage: Implementation Due Date: 30-Sep-2013 Status: Not Yet Due Design Rating Moderate Description: Risk Management: Challenges related to scaling up PBF interventions at decentralized levels. Sufficient time will be taken during preparation to define target areas and population and processes to build local expertise the project will relay on to scale up PBF. (200 char) Health workers could manipulate results to get PBF bonuses. Resp: Client Stage: Implementation Due Date: 30-Sep-2014 Status: Not Yet Due Risk Management: Adequate control processes will be defined in the PBF procedure manual to avoid the "number game". The EEA will audit facility records and conduct home visits to a random sub-sample of facility clients to ensure that service volumes are not mis-reported. The quality of care at health facilities will also be checked independently to ensure accurate reporting. Resp: Client Stage: Implementation Due Date: 17-May-2013 Status: Not Yet Due Social and Environmental Rating Low Description: Risk Management: Risks related to handling and disposal of medical and health waste: the proposed project is A biological waste management plan was re-designed and strengthened in September 2010 as a condition to extend the closing date of classified as Category B for environmental screening purposes given the risks associated the ongoing Health-Education-HIV/AIDS project. The plan is being implemented by the MOH. A revised planned has been prepared with the handling and disposal of medical and health waste such as syringes and material and disclosed prior to appraisal and will be adopted by the project. In addition, health centers will report on hazardous medical material used during deliveries, waste management since that action will be included as an indicator to be evaluated and purchased in PBF Resp:Cliet Stge: PepartionDue Date: 30-Mar-2012Stus Coped In case the project is developed nationwide, it will be implemented in the some prefectures of regions 1 and 2 hosting Indigenous people, for that purpose, OP. 4.10- Risk Management: Indigenous People was triggered - An Indigenous People Planning Framework has beene developed and disclosed prior to appraisal to ensure that Indigenous people who in general are marginalized will also benefit from this project in a manner suitable to their culture and needs. Moreover, health service quantity and quality checklists used by PBF, to pay for performance, will precisely measure the extension to which indigenous people are served by health facilities. The weighting for this indicator will be increased, and the adherence to the Indigenous People Plan will be checked, and paid for, quarterly Resp: Client Stage: Preparation Due Date: 30-Mar-2012 Status: Completed 58 Program and Donor Rating Low Description: Risk Management: The project may be perceived as potentially widening regional disparities and failing to Selectivity being the rule of engagement for most development partners, there is little likelihood that this will pose a serious problem. promote poverty reduction for all. The initial focus on districts that are accessible and safe The road map for PBF institutionalization developed by the MOH during the high level workshop promotes a progressive scale-up of prmle soememersction o the. publnitoal believe ndistrict that theeproce PBF nationwide. Forthcoming workshops on PBF involving district levels stakeholders will ensure that this information is conveyed to may lead some members of the public to believe that the project may exacerbate differences between safe and accessible districts and those that are not covered. the public and to stakeholder in order to s. (200 char) Resp: Stage: Implementation Due Date: 30-Sep-2012 Status: Not Yet Due Limited donor involvement in PBF approach in CAR: only a limited number of donors is Risk Management: involved in PBF funding (EU and couple of NGOs) and technical support (CORDAID), while tremendous efforts are geared towards vertical disease control programs. The MOH is preparing a road map for PBF implementation which aims to raise awareness of donors and MOH personnel that PBF is a promising strategy to accelerate progress towards health related MDGs. This is likely to mobilize additional support and funds for PBF. Experience in other countries suggests that PBF can be a platform to unit funding from different disease control programs as disease- Possibility that some NGOs may not share project approach: As in most post-conflict focused stakeholders can 'purchase' indicators that relate to this disease within the PBF program. countries, several NGOs are present on the ground; some of them, well established in Resp: Client Stage: Implementation Due Date: 30-Sep-2012 Status: Not Yet Due geographic and thematic areas, may have conflicting approaches, Risk Management: NGOs will be involved in PBF workshops prior to project development. The project launch seminar will be broadened to enable the participation of NGOs and other stakeholders. Resp: Stage: Implementation Due Date: 30-Dec-2012 Status: Not Yet Due Delivery Monitoring and Sustainability Rating Moderate Description: Risk Management: Government of CAR commitment to continuing PBF approach and financial sustainability A key milestone the road map for PBF implementation is to integrate the financing of PBF strategy in national budget; In addition, are growing but need to be strengthened over time. experience in many countries shows that successful implementation of PBF does mobilize additional donors and encourages government to invest its own budget into PBF. Close collaboration with other donors led to the fact that UNFPA and UNICEF PIUs have the unintended consequence of separating management of the project from the expressed their willingness to co-finance the PBF project in CAR. mainstream of government administration. This increases the risk of insufficient Resp: Stage: Implementation Due Date: 30-Sep-2014 Status: Not Yet Due ownership of the PBF approach in the government when the PIU is phased out, as it usually is on project completion. This risk is heightened in post-conflict settings where Risk Management: governments cannot absorb the PIU staff after project completion due to the high cost involved. Qualified staff will be recruited for the PIU. Ensuring the transfer of skills to the relevant government institutions through formal and informal mechanisms will be included in the terms of reference of PIU staff and will be strongly encouraged by the project. Resp: Client Stage: Implementation Due Date: 30-Jun-2012 Status: Not Yet Due Resp: Stage: Due Date: Status: Overall Implementation Risk Rating Substantial Description: Close supervision of the project will take place as the team intend to hire a full time person to be Risk Management: Close supervision of the project will take place as the team intend to hire a full time person to based in Bangui to follow up the project implementation and respond to challenges in an efficient way. be based in Bangui to follow up the project implementation and respond to challenges in an efficient way. Resp: Client Stage: Implementation Due Date: 30-Jun-2017 Status: In Progress 59 CENTRAL AFRICAN REPUBLIC: Health System Support Project Annex 5: Implementation Support Plan 1. The approach for implementation support has been developed based on the nature of the project, including its risk profile. It will aim to render implementation support to the client more flexible and efficient, and will focus on implementation of the risk mitigation measures defined in the ORAF. 2. The project will need intensive supervision given the geographic spread of the proposed operation (9 prefectures in 4 regions), and given implementation capacity weaknesses at the country and project level. The project will be implemented at three levels: the central MoH, regions, and prefectures. A budget of US$150,000 would be required for the Bank team to thoroughly supervise the project during the first 12 months of implementation. 3. The supervision by the Bank will be leveraged by the supervision carried out by the MoH and the HSSP on a regular basis. The MoH will have teams visiting each prefecture several times per year, and will prepare action-oriented supervision reports that will be reviewed by the Bank and donors during their bi-annual supervision missions, and through desk reviews. This system will allow the MoH to distinguish between the better and lesser-performing regions and prefectures, and provide more assistance to the latter. Sufficient funds to that effect have been included in the project design with a total of about US$400,000 over the five-year lifespan of the project. 4. The overall supervision of PBF implementation will be the responsibility of the national PBF steering committee and technical team within the MoH. Health prefecture, regional directorates and the Performance Purchasing Agencies (international NGOs with extensive experience in PBF) contracted by the MoH for the verification of health services delivered will also contribute to regulatory and supervision activities. The External Evaluation Agency will verify the reported quality performance through regular data quality audits at health facilities and carry out regular community client satisfaction surveys. Bank supervision will consist in ensuring that the PPA and EEA are performing their functions properly according to the terms of its contract through direct interaction with agencies, sample verification of its records, and interviews and feedback from the MoH, selected prefectures, and health facilities interacting directly with the entity. 5. Some of the skills required by the Bank team for supervision will be needed on a regular basis while others will be required on an ad hoc basis. It is therefore proposed to establish a core supervision group, that will emphasize financial, procurement, PBF, and operational basic needs, complemented by technical specialists, in particular those covering monitoring and evaluation. 6. While regular Bank supervision will take place at least twice a year, this will be leveraged by regular visits by the country-based Bank health sector, procurement and financial management specialists who take advantage of their field presence to verify progress and provide ongoing assistance to the client. 7. A much more intensive than normal supervision program should be carried out during the first year of the project to put in place a sound institutional base and properly begin interventions to be undertaken by the operation. 60 8. The supervision team includes the following members: (i) the Task Team Leader; (ii) the co-Task Team Leader; (iii) a monitoring and evaluation specialist whose experience includes health sector impact evaluations; (iv) a PBF specialist with technical expertise in the field; (v) a financial management specialist who will review adherence to Bank procedures with regard to fiduciary responsibilities; (vi) procurement specialist; and (vii) an environmental and social specialists. 9. Financial management. FM implementation support missions will use a risk-based approach, and collaborate with the Task Team, including the procurement specialist. A first implementation support mission will be performed six months or earlier after effectiveness, especially for the review of PBF mechanisms. Afterwards, the missions will be scheduled by using a risk-based approach model and will include the following: (i) monitoring of the financial management arrangements at intervals determined by the risk rating assigned to the overall FM Assessment at appraisal, and subsequently during implementation; (ii) review the IFRs; (iii) review the audit reports and management letters from the external auditors and follow-up on material accountability issues by engaging with the Task Team, client, and auditors. The quality of the audits (internal and external) will be monitored closely to ensure that they are comprehensive and provide enough confidence on the appropriate use of funds by the client; and (iv) physical supervision especially for the PBF mechanism; and (v) assistance to build or maintain appropriate financial management capacity. 10. Procurement. Implementation support will include: (i) provision of training to the PIU staff as needed; (ii) review of procurement documents and provision of timely feedback to PIU; (iii) provision of guidance on the Bank's Procurement Guidelines to the PIU; (iv) monitoring of procurement progress against the detailed Procurement Plan; (v) monitoring that implementation of contracts is compliant with the World Bank's fiduciary guidelines as well as with contract obligations. 11. Environmental and Social Safeguards. Implementation support will include: (i) guidance on the preparation and disclosure of an Environmental and Social Assessment; (ii) supervision of the implementation of the prepared Process Framework and provision of training and guidance to the PIU team; (iii) third party monitoring assessing compliance with safeguards as a specific, separate component will be included in the M&E system. 12. Coordination with other Development Partners. Implementation support will include: (i) planning for joint local and national meetings and missions with UNICEF, WHO, and UNFPA; (ii) close coordination with other development partners, research institutions and international, national and local NGOs engaged in the health sector in Central African Republic. Implementation Support Plan 13. The project will require substantive technical support due the rather complex and technical nature of the activities to be financed. Most of the World Bank team members are based in the region, to ensure timely, efficient, and effective implementation support to the client. Formal implementation support missions and field visits will be carried out every six months. Detailed inputs from the Bank team and partners are outlined below: * Technical inputs: Technical inputs will be provided by members of the abovementioned supervision team, and additional Bank staff who have expertise in Monitoring & Evaluation. As needed, the task team will seek additional highly-specialized technical inputs from 61 technical partners with whom close coordination and collaboration has been established during project preparation. * Fiduciary requirements and inputs: Training will be provided by the Bank's financial management specialist and procurement specialist before the commencement of project implementation. The task team will further provide support to the PIU to improve fiduciary efficiency. Formal supervision of financial management will be carried out semi-annually, while procurement supervision will be carried out on a timely basis as required by the client. * Safeguards: Inputs from an environment specialist and a social specialist will be provided, despite the project's limited expected social and environmental impacts. An Indigenous Peoples Planning Framework (IPPF) and Health Care Waste Management Plan (HCWMP) have been prepared and disclosed publicly in country as well as in the World Bank's Infoshop. Capacity building will be required for environment monitoring and reporting. On the social side, supervision will focus on implementation of the Process Framework and indigenous peoples' issues. Field visits will be conducted on a semi-annual basis. The social and environmental specialists are based in the sub-region. * Operation: The TTL and co-TTL will provide timely supervision of all operational aspects, as well as ensure coordination with the client and among World Bank team members. The TTL will lead two formal field supervisions a year and, as needed, conduct punctual missions to resolve operational issues. Timeline of main focus of support to implementation: Time Focus Skills Needed Resource Partner Role Estimate First twelve Preparing for Team Leadership, $150,000 per year Technical months implementation of PBF, M & E, support and PBF FM, policy dialogue Procurement, Environment, Social 12-48 months Implementation of Same as above Same as above Same as above PBF and institutional strengthening Other 62 Skills Mix Required Skills Needed Number of Staff Weeks Number of Trips Comments Team Leadership 8 annually Three the first year Based in the Region/ and two annually Washington based from the second year M & E Specialist 4 annually 2 annually Based in the Region PBF Specialist 4 annually Field trips as Washington based needed Financial 4 annually Field trips as Based in the Region Management needed Procurement 3 annually Field trips as Country office based needed Environment 2 annually Field trips as Based in the Region needed Partners Name Institution/Country Role UNICEF, WHO, and Contributing to UNFPA PBF implementation and technical knowledge of country health sector 63 CENTRAL AFRICAN REPUBLIC: Health System Support Project Annex 6: Economic and Financial Analysis 1. Financial Analysis. Sustainability of the benefits to be obtained from the project depends, to a large extent, on the government's commitment to position the strengthening of MCH services as a priority objective and allocate sufficient funds after completion of the Bank operation. 2. Key macroeconomic outcomes. CAR's macroeconomic performance has witnessed a certain recovery between years 2003 and 2007. Within the period, the country's GDP growth rate evolved from -7.1 percent to 4.2 percent in a context progressively characterized by sociopolitical stability and the implementation of reforms in many key domains. Despite the low level of public investment in key sectors, the overall level of investment has increased, particularly due to private investment in the mining, forestry, transportation and telecommunication sectors. 3. However, since 2008 CAR's economic performance has been affected by the global economic downturn, alongside acute shocks such as the fuel and food price crisis, the depreciation of the dollar relative to the euro, and the international financial crisis. In 2008, CAR's GDP growth rate declined to 2.2 percent, instead of reaching the 6.9 percent growth rate that was forecasted. Moreover, the real GDP growth rate was estimated at a meager 1.2 percent in 200929. The financial crisis led to an important drop in net external demand and exports (almost 23 percent) in 2008. The collapse of exports led to considerable deficit into the public current account from -6.1 percent of GDP in 2007 to -8.8 percent in 2008 and to about -9.8 percent in 2009. 4. The country implemented a series of reforms within the framework of the Poverty Reduction and Growth Strategy 2007-2009, which allowed the attainment of the decision point of the Heavily Indebted Poor Countries (HIPC) initiative in September 2007. The budgetary policy was aimed at maintaining the stability of the macroeconomic framework as well as a better mobilization of internal and external resources accompanied with the reinforcement of social spending. In 2008, the total revenue collection was about 10.5 percent of GDP, while total expenditures accounted for 16.7 percent. 5. As a whole, during the last four years, the annual growth of GDP per capita was only 0.93 percent, much less than population growth rate estimated at 2.5 percent for the same period. Poverty levels remain high, estimated at 62 percent in 2008. 29 CAR government, IMF and BEAC 64 6. Health sector financing. In CAR, the health sector is financed from three sources: government budget, external donor contributions, and private, out-of- pocket expenditures on health. Public sector funding on health is very low: over the past 15 years, public funds (external donors excluded) have accounted for less than 10 percent of total health expenditures in the country. This low public contribution to health expenses also reflects the low level of health budget consumption, averaging only 71 percent between 2008 and 2010. Disbursement linked to the payment of salaries appears to be better (97 percent); disbursement related to goods and services is lower (between 68 and 50 percent); and disbursement related investment expenses is very low (32 percent). Graph A8.1: Distribution of health funding in CAR as a percentage of total expenditure 120 100 - N Private health 80 -expenditures 60 - External health 40 - expenditures 20 - Government health expenditures Ln 0 r 00 (n 0 - 0 M L 0 0, 00 MMMMM000000000 M~ M~ M~ M M 0 0 0 0 0 0 0 0 0 H rH rH rH rH N1 N1 N1 N1 N1 N1 N1 N* N* 7. Moreover, according to the Poverty Reduction and Growth Strategy 2008- 2010 document, based on the funds effectively disbursed, the health sector financing of the country accounted for less than 6 percent of the national budget for more than a decade (5.3 percent in 2010). The proportion of total health expenditure linked to external investments continues to increase in CAR - mainly to the detriment of government expenses. 8. The structure of health sector expenditures in CAR has changed significantly. Between 2009 and 2010 total health sector investments doubled, while goods and services purchases were declining in the sector. In the period 2008-2009, investment expenditures were quite low (5.1 percent of public expenses in health) compared to goods and services as well as salaries and transfer expenses. Between 2010 and 2011, investment expenditures and goods and services purchases increased significantly; while salaries and premiums expenditures declined. 65 Graph A8.2: Evolution of the health expenses as a percentage of GDP from 1995 to 2008 5 4 .... 3 2 1 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 9. Macroeconomic and financing analysis. We are assuming for this analysis that the PBF project, if scaled up and adopted by the national health policy, would mainly be financed through the creation of a current expenditure budget line for PBF expenses within the national health sector budget. This analysis develops two primary scenarios (a low-growth baseline scenario and a high-growth scenario) that enable to estimate the expected level of government funding available for scaling-up and sustaining PBF in the years following the Bank operation. It should be noted that public health expenditure includes external donor contributions, alongside government expenditures in the analysis. Graph A8.3: Health expenditures according to their economic classification 2008-2011 100 90 80 70 60 50 40 30 20 10 0 2008 2009 2010 2011 - Salaries and premium - Good and services Transfers - Investment expenses Source: Public Expenditure Review (The World Bank, 2011) 10. For the baseline scenario (see Table A8.1), the following assumptions have also been made. The analysis assumed an annual real GDP growth rate of 1.2 percent. Between 2006 and 2009, total public expenditure as a percentage of 66 GDP is increasing by a growth rate steadily accounting from 9 to 10 percent. There is an assumed baseline ratio of 10.25 percent of GDP, which will increase 0.25 percentage points every year over the 5 years of the project. From 2000 to 2004, the mean ratio of government health expenditures relative to the national budget was 8.7 percent. For this analysis, the ratio was considered constant throughout the duration of the project. The proportion of current expenditures in the health sector budget varied slightly over the years (between 50 and 55 % from 2000 to 2004), with a mean ratio of 53 percent. Providing the PBF project yields an increase in health facility utilization, one can expect an increase in current expenditures, set at an average of 1 percentage point per year over the lifetime of the project. 11. The results of the baseline scenario analysis estimate an increase in the total government health expenditure from US$ 17.7 million in 2011 to US$$ 19 million in 2017. This represents an increase of 7.4 percent over the period. The increased government health expenditure is expected to translate into an increase in current health expenditures benefiting from PBF investments, from US$ 9.4 million in 2011 to US$ 11.2 million in 2017. Table A8.1: Results of low-growth baseline scenario financing analysis 2011 2012 2013 2014 2015 2016 2017 Baseline real GDP growth 1,2 1,2 1,2 1,2 1,2 1,2 1,2 GDP in US$$ million -2012 values 1983,8 2007,6 2031,7 2056,1 2080,7 2105,7 2131,0 Total Government expenditures (% GDP) 10,25 10,5 10,75 11 11,25 11,5 11,75 Total Government expenditure (US$$ million) 203,3 205,8 208,2 210,7 213,3 215,8 218,4 Total Government health expenditure (% Total Government expenditures) 8,7 8,7 8,7 8,7 8,7 8,7 8,7 Total Government health expenditure (US$$ million) 17,7 17,9 18,1 18,3 18,6 18,8 19,0 Total Government current health expenditure (% of Total Government health expenditure) 53 54 55 56 57 58 59 Total Government current health expenditure (US$$ million) 9,4 9,7 10,0 10,3 10,6 10,9 11,2 12. For the high-growth scenario (see Table A8.2), the analysis assumed a more optimistic GDP growth rate under which the country witnesses the growth rate of 6.9 percent previously forecasted in 2008, from 2012 to 2017. With a GDP growth rate increase of 1 percent per year, CAR would be able to attain a 7.2 percent growth rate by 2017. All other assumptions are identical as in the low- growth baseline scenario. 13. The results of the high-growth scenario analysis shows a rise in total government health expenditure from US$ 17.7 million in 2011 to US$ 23.3 million in 2017 - an increase of 31.6 percent over the period. This last estimate is consistent with the PNDS II funding estimates of CAR, done in 2008 for a 10 year period. The share of current expenditure is therefore estimated from US$ 9.4 million in 2011 to US$ 13.7 million in 2017 - an increase of 46.5 percent over the period. 67 Table A8.2: Results of high-growth case scenario funding analysis 2011 2012 2013 2014 2015 2016 2017 Baseline real GDP growth 1,2 2,2 3,2 4,2 5,2 6,2 7,2 GDP in US$ million -2012 values 1983,8 2027,4 2092,3 2180,2 2293,6 2435,8 2611,1 Total Government expenditures (% GDP) 10,25 10,5 10,75 11 11,25 11,5 11,75 Total Government expenditure (US$ million) 203,3 207,8 214,5 223,5 235,1 249,7 267,6 Total Government health expenditure (% Total Government expenditures) 8,7 8,7 8,7 8,7 8,7 8,7 8,7 Total Government health expenditure (US$ million) 17,7 18,1 18,7 19,4 20,5 21,7 23,3 Total Government current health expenditure (% of Total Government health expenditure) 53 54 55 56 57 58 59 Total Government current health expenditure (US$ million) 9,4 9,8 10,3 10,9 11,7 12,6 13,7 14. Sustainability of the project. The per capita benefit of the project amounts to US$ 2.0 per year. If PBF were scaled up nationally, the operation would require an overall financing of US$ 9.7 million per year. That amount represents less than the total government current expenditure per year as estimated for the last year base-case scenario (US$ 11.2 million). Under the frame of the high-growth scenario, that amount accounts for only 70 percent. Moreover, out-of-pocket payment used for current expenses in the health facilities in CAR could only reinforce the evidence of the sustainability of the project. 15. Economic analysis. The development objective of this proposed operation is to increase utilization and improve the quality of maternal and child health services in targeted rural areas of Central African Republic. The project will be implemented through two components: (1) Improvement of health facilities performance through Performance-Based Financing and (2) Strengthen monitoring and evaluation capacity and support project implementation unit. The first component includes performance payments to health facilities as well as support to PBF implementation and supervision. The anticipated level of funding for the project does allow covering a maximum of 2.5 million populations, which is size of the population in the targeted prefectures. 16. Cost benefits analysis. To ascertain economic soundness of the project, a cost benefit analysis (CBA) was conducted. The analysis took into account the estimated incremental project's costs and benefits. A CBA is a technique of identifying, measuring and calculating a project's cost and benefits in order to ascertain the net returns [net present value (NPV)] and economic rate of return (ERR) of the project's investments. 17. Methodology. The 'input-output' approach to CBA was employed to generate net benefits of project components/activities. Within the context of this operation, the 'input-output' approach addresses the purchasing of high quality health services from targeted health facilities. From the net benefits, the net present value (NPV) and the economic rate of return (ERR) were calculated to determine economic viability of the project. To account for uncertainties 68 associated with the analysis, a sensitivity analysis was conducted to determine the extent of responsiveness of the analysis. 18. The chronology of steps to the conduction of the analysis is outlined below: * Identify project inputs and outputs; * Quantify project inputs and outputs based on project objectives and expected results; * Construct value flow tables to value inputs and outputs that would generate the project's net benefits; * Use techniques of discounting to discount project benefits and costs, and compute the net present value (NPV) and the economic rate of return (ERR) of the project to determine economic soundness of the project; 19. We assumed a 10-percent discount rate for computing present values of net benefits and costs. The period of the analysis covers a 5-year period. 20. Project benefits and costs. The implementation of the project is expected to generate two main types of benefits (direct and indirect benefits). Direct benefits are benefits that would accrue to the project target beneficiaries; they are referred to as medical benefits resulting from improved access to a basic package of health, nutrition, or population services in targeted areas, children fully immunized, births attended by trained personnel in health facility, new cases of tuberculosis treated as well as acceptors of modern contraceptives. The project also targets a quality improvement of at least 50 percent over 5 years. Indirect benefits include increase of quality of life in years of the population, prevention of unwanted pregnancies, reduction of household size and in the long run a key contribution to the society's well-being as a whole. Costs are the resources used in carrying out project activities. The direct costs of the project are the cost of implementing and evaluating the project (impact evaluation excluded). Indirect costs include, beneficiaries transport costs, costs associated with patients' visits, etc. Due to lack of data, the analysis did not take into account indirect benefits and costs, nor the epidemiological frame before/after the project implemented, and focused mainly on direct benefits and costs. As stated above in Annex 1, the project is targeting as intermediate impact, health facilities achieving an average 50% of quality of care index (range from 0 to 100) in the targeted areas. We have therefore monetized every "unit of quality delivered" with a supplement corresponding to the mean cost obtained per direct beneficiary of the project. 21. The project cost - Impact evaluation costs excluded - amounts to US$ 28.2 million. As no material input is provided by the project, costs of the project are directly benefiting to the population through the purchase of healthcare provided to them. Therefore, direct benefits estimates were done only by summing the amount of quality healthcare purchased (expected coverage) with the project funds over the 5 years. We have assumed a 50-percent increase of 69 quality of health care starting from the second year of implementing the project that we have monetized at the same amount than the per direct beneficiary healthcare delivered. 22. Based on these assumptions, the Cost-Benefit Analysis estimates the Net Present Value of the project as US$ 8.4 million (Table A8.3). The positive NPV means that the project would generate the expected positive results. The Economic Rate of Return is estimated at around 50%. Sensitivity analysis did not significantly change the CBA results. Such a high ERR comes from the fact that the baseline evaluation of the project is assuming a zero percent quality index (Annex 1). We have conducted sensitivity analyses by changing the discount rate within a range from 3 to 12 percent and did not found any significant impact on the CBA results. Table A8.3: Cost-benefit analysis Year Index 0 1 2 3 4 5 Discount Factor 1.0000 0.9091 0.8264 0.7513 0.6830 0.6209 Discounted Flows Costs -$4,668,032 -$4,675,976 -$4,666,287 -$4,614,850 -$4,507,768 $0 Minimum Benefits $0 $6,365,498 $6,376,330 $6,363,119 $6,292,977 $6,146,956 Net -$4,668,032 $1,689,523 $1,710,043 $1,748,269 $1,785,209 $6,146,956 Cumulative -$4,668,032 -$2,978,509 -$1,268,466 $479,803 $2,265,012 $8,411,968 Minimum Net Present Value $8,411,968 Minimum Economic Rate of Return 50% 70 * i i 16°E 18°E 20°E 22°E 24°E This maop ~'o prodosd by the Mop Design Unoit of The World Bonk. The boundones, cobos, d~iona,tio,s and ony othe inf,omtion j sho~n on fhis map do not irnÅ oo th port of The world Bank SUDA Group, ony judgnlent on the 499a1 stotos o( any åerritoiy, or any endorsemnto o...laonce oJ soh boordoä- T NOl C H A D To . *Birao \ Am ri ~m CENTRAL10 10°N- . AFRICAN _..---1° r: ° VA/KA GA REPUBL|C TO N -. Ouorda . Kafia Käogi To NDicillé Sorf BAMINGUI-BANGG- N 0 g Mok 2p, Ndélé -c n 8° - ---T,·· Ouadda Kaouadja 8°N 8'N To J Pata Doba SOUT H - J Markounda HAUTE-KOTTO Botangofo NANA- SUDAN ou GREBIZI Bocorango OUHAM .aga Yangalia PENI9DE Molgag Bozoum Bossangoaa MM OU 6°N Bouar. KEMO 6N M B OM OU ouarrö Béfi,éC Oy NA NA - °Bakouma MAMBERE o13o Bossermbélé ° OM.B ELLA- dTMingala zém o -- - . Noooi Comot M'POKO ~Gma °Koang° KOTO BaBgAssou MAMBERE-KADEI Gmoa.BebotBN I 'oye . MonI"CENTRAL AFRICAN Boaoo Gosobulo L B \E 4°NMbaïli Ubenge REPUBLIC To •'SANG Mongounb Yokdoumo\ Nola Eri MBAERE DEM. REP. o SELECTED CITIES AND TOWNS Salo OF CONGO 0 40 80 120 160 200 Kilometers PREFECTURE CAPITALS CAMEROON ® NATIONAL CAPITAL 0 40 80 120 Miles RIVERS _. CONGO MAIN ROADS 2°N Na .RAILROADS ..... PRFECTURE BOUNDARIES 16°E 1E 20°E 22°E - - - INTERNATIONAL BOUNDARIES 16EWE2'E2*