Report No: ICR00318 IMPLEMENTATION COMPLETION AND RESULTS REPORT TF-10748; TF-15111; TF-A1598; TF-A5311; TF-A9037; TF-B3156 ON A SERIES OF GRANTS IN THE AMOUNT OF US$78.0 MILLION TO THE REPUBLIC OF ZIMBABWE FOR THE HEALTH SECTOR DEVELOPMENT SUPPORT PROJECT April 3, 2025 Health, Nutrition and Population Global Practice Eastern and Southern Africa Region CURRENCY EQUIVALENTS (Exchange Rate February 28, 2025) Currency Unit = Zimbabwean Dollar (ZWL) ZWL 322 = US$1 US$1.31 = SDR 1 FISCAL YEAR January 1 - December 31 Regional Vice President: Victoria Kwakwa Country Director: Nathan M. Belete Regional Director: Daniel Dulitzky Practice Manager: Ernest E. Massiah Task Team Leaders: Fernando Montenegro Torres, Carol Atieno Obure ICR Main Contributor: Miriam Schneidman ABBREVIATIONS AND ACRONYMS AF Additional Financing ANC Antenatal Care ARV Antiretroviral ARI Acute Respiratory Infection BCG Bacille Calmette-Guérin BP Bank Procedure CBO Community-based Organization CHW Community Health Worker COVID Coronavirus Disease 2019 CQI Continuous Quality Improvement CRI Corporate Results Indicator DH District Hospital DHE District Health Executive DHS Demographic and Health Survey DO Development Objective FP Family Planning FY Fiscal Year GDP Gross Domestic Product GFF Global Financing Facility GRM Grievance Redress Mechanism HCC Health Center Committee HDF Health Development Fund HIV Human Immunodeficiency Virus HNP Health, Nutrition and Population HRITF The Health Results Innovation Trust Fund ICER Incremental Cost Effectiveness Ratio ICR Implementation Completion and Results Report IE Impact Evaluation IPC Infection Prevention and Control IPF Investment Project Financing IRI Intermediate Results Indicator ISN Interim Strategy Note ISR Implementation Status and Results Report MICS Multiple Indicator Cluster Survey MNCH Maternal, Newborn and Child Health MMR Maternal Mortality Ratio MOHCC Ministry of Health and Child Care MTCT Mother-To-Child-Transmission NGO Non-Governmental Organization NPV Net Present Value OP Operational Policy ORS Oral Rehydration Salt ORT Oral Rehydration Therapy PAD Project Appraisal Document PCU Project Coordination Unit PDO Project Development Objective PHE Provincial Health Executive PIM Project Implementation Manual PIE Project Implementation Entity POI Project Outcome Indicator PPE Personal Protective Equipment QALY Quality Adjusted Life Year RBF Results-Based Financing RHC Rural Health Center RMNCAH Reproductive Maternal Neonatal Child and Adolescent Health SRHS Sexual and Reproductive Health Service TF Trust Fund TFR Total Fertility Rate UN United Nations UN DESA United Nations Development of Economic and Social Affairs UNFPA United Nations Population Fund UNICEF United Nations Children’s Fund WHO World Health Organization TABLE OF CONTENTS DATA SHEET ................................................................................................................................................. i I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ..................................................................................2 II. OUTCOME .................................................................................................................................................8 III. KEY FACTORS AFFECTED IMPLEMENTATION AND OUTCOME ..................................................................... 19 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .............................. 22 V. LESSONS AND RECOMMENDATIONS ......................................................................................................... 27 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................................................ 29 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ....................................................... 36 ANNEX 3. PROJECT COST BY COMPONENT ......................................................................................................... 38 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................................................ 39 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ................................. 41 ANNEX 6. SUMMARY OF KEY CHANGES DURING IMPLEMENTATION .................................................................. 52 ANNEX 7. ACHIEVEMENT OF PDO-LEVEL AND INTERMEDIATE RESULTS INDICATORS .......................................... 54 ANNEX 8. BROAD TRENDS IN KEY INDICATORS .................................................................................................. 58 DATA SHEET @#&OPS~Doctype~OPS^dynamics@icrbasicdata#doctemplate BASIC DATA Product Information Operation ID Operation Name P125229 Health Sector Development Support Project Product Operation Short Name Investment Project Financing (IPF) ZW-Health Sector Development Proj (FY12) Operation Status Approval Fiscal Year Closed 2012 Original EA Category Current EA Category Partial Assessment (B) (Restructuring Data Sheet - 26 Mar 2024) CLIENTS Borrower/Recipient Implementing Agency Republic of Zimbabwe Stichting Cordaid DEVELOPMENT OBJECTIVE Original Development Objective (Approved as part of Package on ) s s s s @#&OPS~Doctype~OPS^dynamics@icrfinancing#doctemplate FINANCING Financing Source Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Administered 77,691,907.68 77,691,907.58 78,000,000.00 Financing TF-B3156 25,000,000.00 24,691,907.68 24,691,907.68 TF-A9037 3,000,000.00 3,000,000.00 3,000,000.00 i TF-A1598 10,000,000.00 10,000,000.00 9,999,999.90 TF-A5311 5,000,000.00 5,000,000.00 5,000,000.00 TF-15111 20,000,000.00 20,000,000.00 20,000,000.00 TF-10748 15,000,000.00 15,000,000.00 15,000,000.00 Total 78,000,000.00 77,691,907.68 77,691,907.58 RESTRUCTURING AND/OR ADDITIONAL FINANCING Amount Disbursed Date(s) Type Key Revisions (US$M) • Loan Closing Date Extension 01-Sep-2015 Portal 33.51 • Reallocations 21-Feb-2017 Portal 43.89 • Loan Closing Date Extension 27-Jun-2018 Portal 50.00 • Loan Closing Date Extension • Loan Closing Date Extension 20-Dec-2018 Portal 50.00 • Reallocations • Loan Closing Date Extension 06-Dec-2019 Portal 51.38 • Reallocations 29-Jun-2020 Portal 53.00 • Loan Closing Date Extension • Components • Results • Disbursement Estimates 25-Apr-2023 Portal 67.58 • Loan Closing Date Extension • Reallocations • Implementation Schedule 26-Mar-2024 Portal 78.00 • Reallocations @#&OPS~Doctype~OPS^dynamics@icrkeydates#doctemplate KEY DATES Key Events Planned Date Actual Date Concept Review 13-Jan-2011 13-Jan-2011 Decision Review 07-Jun-2011 07-Jun-2011 Authorize Negotiations 26-Sep-2011 26-Sep-2011 Approval 29-Sep-2011 29-Sep-2011 Signing 29-Sep-2011 27-Sep-2011 ii Effectiveness 06-Dec-2011 08-Dec-2011 ICR/NCO 31-Mar-2025 03-Apr-2025 Restructuring Sequence.01 Not Applicable 01-Sep-2015 Restructuring Sequence.02 Not Applicable 21-Feb-2017 Restructuring Sequence.03 Not Applicable 27-Jun-2018 Restructuring Sequence.04 Not Applicable 20-Dec-2018 Restructuring Sequence.05 Not Applicable 06-Dec-2019 Restructuring Sequence.06 Not Applicable 29-Jun-2020 Restructuring Sequence.07 Not Applicable 25-Apr-2023 Restructuring Sequence.08 Not Applicable 26-Mar-2024 ICR Sequence.02 (Final) -- 02-Apr-2025 Operation Closing/Cancellation 31-Mar-2024 31-Mar-2024 @#&OPS~Doctype~OPS^dynamics@icrratings#doctemplate RATINGS SUMMARY Outcome Bank Performance M&E Quality Substantial ISR RATINGS Actual Disbursements No. Date ISR Archived DO Rating IP Rating (US$M) 01 28-Mar-2012 Satisfactory Satisfactory 3.80 02 27-Aug-2012 Satisfactory Satisfactory 6.60 03 30-Jan-2013 Satisfactory Satisfactory 10.26 04 20-Aug-2013 Satisfactory Satisfactory 15.00 05 07-Apr-2014 Satisfactory Moderately Satisfactory 25.54 06 09-Nov-2014 Satisfactory Satisfactory 29.92 07 14-May-2015 Satisfactory Satisfactory 32.62 08 09-Dec-2015 Satisfactory Satisfactory 34.90 iii 09 23-Jun-2016 Satisfactory Moderately Satisfactory 39.66 10 29-Dec-2016 Satisfactory Satisfactory 43.89 11 30-Jun-2017 Satisfactory Satisfactory 44.94 12 17-Jan-2018 Satisfactory Satisfactory 47.94 13 19-Jul-2018 Satisfactory Satisfactory 50.00 14 22-Dec-2018 Satisfactory Satisfactory 50.00 15 23-Jun-2019 Satisfactory Moderately Satisfactory 50.75 16 19-Feb-2020 Satisfactory Moderately Satisfactory 52.27 17 18-Sep-2020 Moderately Satisfactory Moderately Satisfactory 53.00 18 28-Mar-2021 Moderately Satisfactory Moderately Satisfactory 56.67 19 18-Oct-2021 Moderately Satisfactory Moderately Satisfactory 57.46 20 21-Apr-2022 Moderately Satisfactory Moderately Satisfactory 63.65 21 28-Nov-2022 Moderately Unsatisfactory Moderately Satisfactory 65.91 22 06-Jun-2023 Moderately Satisfactory Moderately Satisfactory 67.58 23 01-Aug-2024 Moderately Satisfactory Moderately Satisfactory 77.69 @#&OPS~Doctype~OPS^dynamics@icrsectortheme#doctemplate SECTORS AND THEMES Sectors Adaptation Mitigation Major Sector Sector % Co-benefits Co-benefits (%) (%) FY17 - Health 70 0 0 FY17 - Health FY17 - Public Administration - Health 30 0 0 Themes Major Theme Theme (Level 2) Theme (Level 3) % FY17 - Health System FY17 - Human 40 Strengthening Development and FY17 - Health Systems and Policies FY17 - Reproductive and Gender 40 Maternal Health iv FY17 - Social FY17 - Participation and Development and FY17 - Social Inclusion 20 Civic Engagement Protection v ADM STAFF Role At Approval At ICR Practice Manager Jean J. De St. Antoine Ernest E. Massiah Regional Director Ritva Reinikka Daniel Dulitzky Global Director Christian Baeza Juan Pablo Eusebio Uribe Restrepo Practice Group Vice President Tamar Manuelyan Atinc Mamta Murthi Country Director Kundhavi Kadiresan Nathan M. Belete Regional Vice President Obiageli Katryn Ezekwesili Victoria Kwakwa ADM Responsible Team Leader Monique Vledder Fernando Montenegro Torres Co-Team Leader(s) Carol Atieno Obure ICR Main Contributor Miriam Schneidman Page 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context1 1. During 2000-2008 Zimbabwe experienced a socio-economic crisis, which led to a sharp economic decline and collapse of basic social services. Gross Domestic Product dropped by 35 percent during this period, and hyperinflation peaked in 2008. At appraisal in 2010, unemployment remained high (65 percent), and over 50 percent of the population was living below the poverty line. The Government which assumed control in 2009 introduced economic reforms to stabilize the economy and created an enabling environment for restoring basic social services. 2. Key gains in the health sector began to unravel during the economic and political crisis . Historically, Zimbabwe had one of the best health systems in Africa with a strong network of health facilities which provided a comprehensive package of preventive and curative services. During the socio-economic crisis there was a steady decline in per capita public health spending (2001-2007), which translated into a drop in operational funding for health facilities; a deficit in human resources with an external brain drain; and shortages of drugs and supplies which hindered the quality of care. With the decline in public funding for health the financial burden fell increasingly on households with an estimated 36 percent of the population reporting difficulties with out-of- pocket spending. 2 3. The human cost of the crisis was large as reflected in the deterioration of health indicators. As noted in the Project Appraisal Document (PAD), life expectancy at birth declined from 58 (1990) to 43 (2008) years, partly due to the impact of HIV/AIDS. During 1990-2009 key indicators deteriorated: (i) maternal mortality ratio (MMR) rose from 390 to 790 deaths per 100,000 births; (ii) infant mortality increased from 51 to 60 deaths per 1,000 live births; and (iii) under-five mortality rose from 79 to 86 deaths per 1,000 live births. Zimbabwe ranked very low (#169) on the 2010 Human Development Index with a value of only .140. Coverage of key maternal and child health interventions declined steeply; skilled attendance at birth dropped from 73 (1990) to 60 percent (2009), and children fully immunized dropped from 53 percent (2006) to 49 percent (2009) with large geographic and socio-economic disparities. 4. The Government prioritized recovery of the health sector with a focus on reviving the primary healthcare system. This was reflected in the Short-Term Emergency Recovery Plan and in the Medium-Term Plan (2010- 2015). Key areas for the recovery effort included increasing health sector financing, improving productivity and retention of health workers, and strengthening the supply chain for drugs. The Government also jointed the Campaign on Accelerated Reduction of Maternal Mortality in Africa to address the persistently high MMR. 5. The PAD provided a strong rationale for the initial US$15 million grant from the Health Results Innovation Multi- donor Trust Fund (HRITF) to Zimbabwe.3 The country was in arrears, Government commitment to reforms was high, and the grant was expected to play a critical role in improving maternal and child health services. The project was aligned with the Government’s National Health Strategic Plan (2009-2013) and prioritized high impact interventions identified in the Health Sector Investment Case (2009-2013). The project was also aligned with the 1 This Implementation Completion and Results Report (ICR) draws on the Interim ICR filed on September 29, 2021, and updates the analysis to include Additional Financing (AF) V. The results are broadly similar. 2 2007 National Health Accounts. 3 The HRITF transitioned into the Global Financing Facility (GFF) in 2015. The initial US$15 million and three AFs were provided by the HRITF, while the GFF provided the fourth and the fifth AFs. Page 2 Results Based Management policy of the Government and the Work Program for the Social Cluster Ministries. The Results Based Financing (RBF) approach built and leveraged on the ongoing policy dialogue and on recent analytic work conducted by the World Bank. 6. The PAD did not include a Theory of Change as this was not a World Bank requirement at that time. For this Implementation Completion and Results Report (ICR), a detailed Theory of Change was constructed to include the socio-economic and health systems determinants. The Theory of Change has been retrofitted to include the Results Framework for both the original project and the five rounds of AFs, as well as the long-term goals to which the project was contributing (Figure 1). Figure 1: Zimbabwe Health Sector Development Support Project Theory of Change Page 3 Project Development Objectives (PDOs) 7. The Project Development Objective was to increase coverage of key maternal and child health interventions in the targeted rural districts ’ . Key Expected Outcomes and Outcome Indicators 8. There were five Project Outcome Indicators (POIs) to assess the achievement of the project, as noted below. • POI 1: Number of pregnant women receiving antenatal care (ANC) during a visit to a health provider in participating rural districts • POI 2: Increase in the percentage of births attended by skilled health personnel in a health institution in participating rural districts • POI 3: Number of pregnant women living with HIV in participating rural districts who received antiretrovirals (ARVs) to reduce the risk of Mother to Child Transmission (MTCT) • POI 4: Increase in the percentage of women 15-49 years receiving a modern family planning method in participating rural districts during their first or repeat visits • POI 5: Number of children immunized in participating rural districts Components 9. Component 1: Results-based Contracts with Provincial Health Executives, District Health Executives, District Hospitals, and Rural Health Centers for the Delivery of a Package of Key Maternal and Child Health Services (US$10.9 million). This component supported performance-based service delivery contracts with providers, which stipulated the services to be provided and the level of payments for their delivery, using a fee-for-service approach. 4 Performance-based payments took into consideration both the quantity and quality of services delivered as well as the remoteness of the health facility. A quality assessment tool was to be used to assess the services provided by Rural Health Centers, and another tool was developed for District Hospitals (DHs). The District Health Executives (DHEs) and Provincial Health Executives (PHEs) were to receive performance-based payments based on pre-agreed supervisory services. An international NGO (Cordaid) was competitively selected to contract service providers and serve as the Project Implementing Entity (PIE), managing the fund holding and purchasing functions. 10. Component 2: Management and Capacity Building in Results-Based Financing (US$3.9 million). This component supported management and capacity building on Results-Based Financing (RBF) for all service providers as well as District Steering Committees.5 Key activities included: (i) start-up training workshops for key staff from health facilities, Health Center Committees (HCCs), and DHEs; and (ii) on-the-job capacity development. The component also funded the verification and counter verification activities as well as the management cost of the PIE and the Local Purchasing Units (LPUs). Supply of medical equipment was to be provided to select facilities after an initial assessment during the start-up phase by the PIE. 11. Component 3: Monitoring and Documentation (US$0.2 million). This component supported monitoring, evaluation, and documentation of the implementation of the RBF program. This included tracking trends in coverage rates in project-supported districts; and conducting household surveys, health facility assessments, and exit interviews to assess the impact of interventions and to monitor patient satisfaction. A rigorous impact 4 The service providers included: (i) Rural Health Centers represented by Health Center Committees; (ii) District Hospitals (DHs); and (iii) District Health Executives and Provincial Health Executives for supervision of Rural Health Centers and DHs. 5 District Steering Committees were to be established in each project district to oversee and support implementation. They included representatives of the Government (national and local) and of faith-based health facilities. Page 4 evaluation and several process evaluations were funded through parallel trust funds to draw lessons from the introduction of RBF. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION 12. The project benefitted from five rounds of AF with each one building on previous achievements and expanding the impact. The AF had mutually complementary objectives: (i) scale up the RBF scheme (adding the urban RBF/voucher program) and set more ambitious targets (AF I, AF II); (ii) raise the importance of service quality and institutional capacity by introducing relevant PDO indicators (AF II, AF III, AF IV); (iii) broaden the scope of the interventions to be supported under RBF to include TB, HIV, and diabetes (AF III) and the package of Reproductive, Maternal, Neonatal, Child and Adolescent Health and Nutrition (RMNCAH-N) interventions (AF V); (iv) add activities to support the COVID-19 response (AF V); and (v) introduce innovations to address the persistently high rates of institutional maternal and newborn mortality by strengthening the continuum of care and quality at provincial and central hospitals through quality Focused RBF and by improving health seeking behavior through extended supply side community RBF (AF V). The closing dates were extended to ensure a smooth transition over time from one grant to another. In total, US$63 million of AF was provided in addition to the original US$15 million grant, bringing the overall total to US$78 million. The sequencing and timing of these modifications is depicted in Figure 2. While the project had five rounds of AF and closing date extensions this was because it served the sector well in supporting the roll out of the RBF program. Annex 6 provides more details on the objectives, rationale, and level of financing for each TF grant. Figure 2: Sequencing and Timing of Technical and Design Modifications Page 5 Revised PDOs and Outcome Targets 13. The PDO was revised three times during the life of the project to reflect evolving priorities. First, during the processing of the first AF approved in 2013, the PDO was amended to add targeted urban districts. Second, during the processing of the fourth AF approved in 2019, the PDO was amended to add quality and strengthen institutional capacity for RBF. Third, during the processing of the fifth AF in 2020, the PDO was changed “to increase coverage and quality of an integrated package of RMNCAH-N services and strengthen the COVID-19 response”. Revised POIs 14. The POIs were modified over time to use more robust measures, better align with the RBF objectives, address emerging issues, and drop indicators that were achieved. Below is the list of POIs which were added for each PDO part. In addition, two of the original POIs were converted to intermediate results indicators (IRIs).6 PDO Part A: Increase coverage and quality of an integrated package of RMNCAH-N services • POI 6: Percentage of women who had their first antenatal care visit during the first 16 weeks of pregnancy in participating rural districts • POI 7: Percentage of children under five with diarrhea receiving oral rehydration therapy (ORT) and Zinc in participating rural districts (*) • POI 8: Percentage of children 6-59 months receiving Vitamin A supplementation in participating rural districts • POI 9: Average quality scores by health facilities in participating rural and urban districts (*) • POI 10: Percentage of maternal deaths given audits as per protocol in participating districts (*) • POI 11: Percentage of partographs correctly filled in participating districts (*) • POI 12: Percentage of children under 5 years with pneumonia correctly managed in the participating districts (*) • POI 13: Percentage of participating district, provincial, and central hospitals that have registered an increase in quality scores since last quarter PDO Part B: Strengthen institutional capacity to manage performance-based contracts • POI 14: Percentage of health facilities managed under RBF contracts by the Ministry of Health and Child Care (MOHCC) in participating rural districts PDO Part C: Strengthen the COVID-19 response • POI 15: Percentage of close contacts of confirmed Coronavirus Disease (COVID-19) cases followed up based on national guidelines Note: (*) These POIs were achieved and subsequently dropped Revised Components 15. During the life of the project, the original components were broadly maintained with refinements introduced to improve the effectiveness of the RBF program, expand its impact, contain costs, and address emerging issues (COVID-19). Key refinements included: 6These were: (i) Number of pregnant women living with HIV in participating rural districts who received antiretrovirals to reduce the risk of Mother to Child Transmission; and (ii) Number of children fully immunized. Page 6 16. Component 1 (US$40.3 Million): (i) adding vouchers to improve access to health services in targeted urban districts; (ii) including performance-based grants to community-based organizations (CBOs) for monitoring and verification activities; and (iii) strengthening the quality of care by providing targeted capacity building to the National Quality Assurance Directorate and the Family and Child Health Directorate of the MOHCC to implement the National Quality Improvement Strategy and roll out a Continuous Quality Improvement (CQI) initiative in a subset of the RBF-supported facilities. A no user-fee requirement for services in the benefit package in participating facilities was introduced to minimize risk that providers introduce fees, ensuring compliance with the Government policy of no user fees for exempted groups (pregnant women and children under five). 17. Component 2 (US$28.7 Million): (i) supporting RBF institutionalization in collaboration with other development partners; (ii) introducing risk-based verification to address concerns with costs; and (iii) developing knowledge exchange platforms to disseminate good practices. 18. Component 3 (US$4.1 Million): supporting the Government to be mainly responsible for the rural RBF while the PIE would be mainly responsible for the urban RBF and voucher program. 19. Component 4 (US$5.0 Million): A new component (COVID-19 Response) was added in 2020 to provide support to the Government to respond to the COVID-19 pandemic. Rationale for Changes and Their Implication on the Original Theory of Change 20. The rationale for changes made was three-fold. First, changes were made to continually adapt and expand the project based on implementation experience and country needs. Changes were also made as part of efforts to institutionalize the RBF program. To this end, the project: (i) scaled up the RBF program starting with 2 rural districts (2011), expanding to 16 rural districts (2012), and subsequently piloted an urban RBF program which included an innovative maternal health voucher scheme (2014); (ii) expanded the package of interventions, adding HIV/AIDS and non-communicable diseases (NCDs, 2015-16); (iii) introduced innovations progressively to systematically monitor, and track the quality of services, identify root causes of issues and plan corrective actions, such as the CQI initiative (2015/16); (iv) introduced different strategies for verification of results, including an innovative risk-based verification method and use of CBOs (2015-16); (v) transferred responsibility for subsidy payments for the rural RBF program to the MOHCC (2018); and (vi) conducted evaluations to draw lessons and make course corrections (e.g., transferring the counter-verification function from an external party to a Government structure, mainstreaming first level verification to existing ministry cadres, and introducing risk based verification to reduce cost). Second, changes were made to the Results Framework indicators to progressively adopt internationally recognized indicators for tracking progress on RMNCAH-N interventions. Third, changes were made to respond to evolving needs. The most notable example is the introduction of a new component to support the Government to deal with the COVID-19 pandemic. 21. The Theory of Change developed for the ICR illustrates the continued relevance of the project-supported RBF program. The focus on improving maternal and child health through RBF was maintained throughout the entire 12.5-year period. Adjustments and refinements were introduced over time to improve the effectiveness of the RBF scheme, scale up interventions, and include the COVID-19 response. These modifications did not change the Theory of Change, as the overriding thrust of the project remained the same, with a focus on using performance- based service delivery contracts with providers to improve both the coverage and quality of a comprehensive package of maternal and child services in the targeted districts, and with complementary support to bolster RBF capacity and institutionalize the program. The other cross cutting theme which was embedded in all project activities during the entire implementation period was a ‘learning-by-doing approach’ (locally coined ‘learning from implementation’), with systematic tracking of results, drawing lessons, and making course corrections, which is a good practice in RBF programs. Page 7 II. OUTCOME A. RELEVANCE OF PDO 22. The relevance of the PDO is rated High. The project was aligned with one of the key objectives of the second Interim Strategy Note (ISN) for FY08-09 for Zimbabwe which aimed to enhance country knowledge and improve policy dialogue with the Government. As noted in the PAD, the RBF activities built on the knowledge that the World Bank developed through its ongoing policy dialogue with the Government and on recent health sector analytic work. The project remained consistent with the World Bank Group priorities for Zimbabwe as outlined in the FY13-FY15 ISN (Report No. 74226-ZW) which included human development within Priority 3, Fostering an Enabling Environment for Reducing Vulnerabilities, Improving Resilience and Strengthening Human Development. The project was aligned with the Government’s National Health Strategic Plan (2009-2013) with an emphasis on strengthening primary health care services at health centers and DHs and with the Government’s overall Integrated Results Based Management Policy. The project was also aligned with Zimbabwe’s Transitional Stabilization Program (2018-2020) which targeted RMNCAH as a priority area. The project remains aligned with the Country Engagement Note for the Republic of Zimbabwe (FY25-26), particularly Pillar 2, Building Resilience, which aims to enhance critical health services, strengthening access, quality, and efficiency of maternal and newborn services through reforms in health financing and service delivery, focusing on vulnerable groups. 23. The focus of the PDO on increasing the coverage of key maternal and child health interventions in targeted rural districts was highly relevant, particularly given the collapse of social services. As noted in the PAD, experience with RBF programs had shown that these approaches can rapidly increase the coverage of cost-effective interventions by motivating health workers to focus on results, empowering them to use the funds to address local needs, and bolstering accountability and community participation. Key lessons from RBF programs were incorporated into the design of the project, including the importance of: (i) concentrating on rapidly improving service delivery that were starting from a low base and were within the control of providers;7 (ii) adopting flexible implementation arrangements to adapt to evolving needs; (iii) empowering and mobilizing local actors; and (iv) monitoring and evaluating results while using a learning-by-doing approach. Other studies prepared at the time to assess RBF programs also underscored their potential to improve coverage and quality of services.8 B. ACHIEVEMENT OF PDOs (EFFICACY) Assessment of Achievement of Each Objective/Outcome 24. The project substantially achieved the three PDO parts. The ICR focuses on three key PDO parts: (i) coverage and quality of RMNCAH-N services; (ii) institutional capacity to manage performance-based contracts; and the (iii) COVID-19 response. Results on improving coverage and quality of RMNCAH-N services are presented jointly (para. 25) given the importance of ‘effective coverage’ (i.e., capturing not only coverage but also the quality of the 7 While most indicators were well selected, a few (i.e., antenatal care in first trimester in rural areas and long-lasting contraceptives) were not fully within the control of providers as discussed under the M&E section. 8 One major review conducted by the Center for Global Development found that RBF programs can work in a variety of health systems and contexts, including in settings where governments demonstrate strong leadership (e.g., Nicaragua, Romania) and even in post-conflict countries where public delivery of health services was failing (e.g., Afghanistan, Rwanda). The study included a series of case studies of promising RBF approaches, concluding that incentives can be the catalyst that inspires providers and consumers to find solutions, and noting that what is most important is to identify the conditions under which they work and to minimize the risk of unintended consequences. Performance Incentives for Global Health: Potential and Pitfalls, Rena Eichler, Ruth Levine, and the Performance-Based Incentives Working Group, Center for Global Development, 2009. Page 8 interventions to best assess whether services will result in expected health gains) with disaggregated data provided below (paras. 26-30). All POIs and IRIs were systematically compared against the latest baseline and target figures (total or average).9 Achievement of each PDO part was assessed against a standard rating system.10 A split rating was not applied given that the overall level of ambition was expanded over time.11 As can be seen in Figure 3, the geographic coverage was gradually expanded to additional rural and urban districts; the package of interventions was increased; and the level of ambition rose about the institutionalization of the RBF program. Table 1 provides a summary of the achievement of the POIs and Annex 7 provides details of all achievements (POIs and IRIs). PDO Part A: Increase coverage and quality of an integrated package of RMNCAH-N services—substantial 25. PDO Part A was substantially achieved, as indicated by both POIs and IRIs. When analyzing the level of achievement of both the coverage and quality of RMNCAH-N services, 90 percent of all POIs/IRIs was achieved/surpassed (73 percent) or substantially achieved (17 percent); and only 10 percent was not achieved. PDO Part A is rated Substantial. These results are directly attributable to the project which incentivized the delivery of best buy interventions through the RBF program and held providers accountable at participating facilities in the 18 rural districts with a catchment population of roughly 4.5 million. Figure 3: Key Milestones in the Rollout and Institutionalization of the RBF Program Source: Zimbabwe Results Based Financing Knowledge & Learning Case Study, World Bank, 2021. 9 For POIs and IRIs that were dropped with the AF V, the baseline and target figures at that time were used. 10 The rating system used is as follows: surpassed/achieved: 100 percent+; substantially achieved: 80-99 percent; partially achieved: 65-79 percent; not achieved: <65 percent. For surpassed indicators, a weighted scale was used to reflect the degree of over- achievement. See Annex 7 for a detailed account of achieved results. 11 A split rating was not used given that the RBF program was scaled up geographically, more ambitious targets were introduced, and the scope was broadened over time to include additional interventions and innovations. Page 9 Table 1: Summary of Achievement of POIs Project Outcome Indicator (POI) Baseline Achieved Target Achievement (%) (A) (B) (C) (B-A)/(C-A) Part A: Increase coverage and quality of an integrated package of RMNCAH-N services Percentage of pregnant women 70.0 80.0 72.0 2020: 500.0 receiving antenatal care during their (2012) (Dec (2020) Surpassed visit to a health provider in 2020) participating rural districts Percentage of births attended by 58.0 82.0 88.0 2020: 80.0 skilled health personnel in a health (2012) (Multiple (2020) Substantially Achieved institution in participating rural Indicator districts based on survey data Cluster Survey [MICS] 2019) Percentage of women 15-49 years 56.0 70.0 70.0 2020: 100.0 receiving a modern family planning (2012) (2014) (2020) Achieved method in participating rural districts Percentage of women who had their 10.0 22.0 32.0 2024: 54.5 first ANC visit during the first sixteen (2012) (2024) (2024) Not Achieved weeks of pregnancy in participating rural districts Percentage of children under 5 with 13.8 16.5 16.0 2020: 122.7 diarrhea receiving ORT and Zinc in (2015) (MICS (2020) Surpassed participating districts 2019) Percentage of children 6-59 months 43.5 30.2 50.0 -204.6 with Vitamin A supplementation in (2019) (2024) (2024) Not Achieved participating rural districts-added Average quality scores by health 68.1 83.0 81.0 2020: 115.5 facilities in participating rural and (2012) (Dec (2020) Surpassed urban districts 2020) Percentage of maternal deaths given 0 76.0 80.0 2020: 95.0 audits as per protocol in participating (2015) (Dec (2020) Substantially Achieved districts 2020) Percentage of partographs correctly 0 62.0 65.0 2020: 95.4 filled in participating districts (2015) (Dec (2020) Substantially Achieved 2020) Percentage of children under 5 years 0 89.5 87.0 2020: 102.9 with Pneumonia correctly managed in (2015) (Dec (2020) Surpassed the participating districts 2020) Percentage of participating district, 51.0 51.35 60.0 2024: 3.9 provincial, and central hospitals that (2019) (2024) (2024) Not achieved have registered an increase in quality scores since last quarter Part B: Strengthen institutional capacity to manage performance-based contracts 0 72.5 80.0 Percentage of health facilities 2024: 90.6 (2012) (2024) (2024) managed under RBF contracts by the Substantially Achieved MOHCC PCU in participating rural districts Part C: Strengthen COVID-19 response Percentage of close contacts 0 82.0 80.0 2024:102.5 confirmed cases followed up based on (2020) (2024) (2024) Surpassed national guidelines Page 10 26. Performance on POIs contributing to increased coverage of an integrated package of RMNCAH-N services was substantially achieved with 78 percent of all POIs and IRIs targets achieved/surpassed; 9 percent substantially achieved; and only 13 percent not achieved. Three POIs (70 percent) were fully achieved or surpassed (women who receive ANC during their visit to a health provider in participating rural districts; women 15-49 years who currently use any modern family planning methods in participating rural districts; and children under five with diarrhea receiving ORT and Zinc in participating districts), one (10 percent) was substantially achieved (birth attended by skilled personnel in a health institution in participating rural districts: 80 percent), and two (20 percent) were not achieved (women who had their first antenatal visit during the first sixteen weeks of pregnancy in participating rural districts; and children 6-59 months with Vitamin A supplementation in participating rural districts). The limited progress on the coverage of ANC during the first trimester was due to persistent cultural taboos in rural areas with ongoing community behavior change communications requiring more time and effort. The Vitamin A supplementation target was not attained due primarily to supply chain constraints with some stakeholders noting that reporting forms changed as well. The delayed start-up of the Extended Supply Side community RBF scheme due to the COVID-19 pandemic also contributed to shortcomings in attaining these two POIs which required an intensification of behavior change interventions. 27. Performance on the IRIs contributing to increased coverage was substantial. Eight indicators (84 percent) surpassed their targets (number of pregnant women receiving their first ANC during a visit to a health provider in participating urban districts; number of pregnant women receiving their first ANC visit before 16 weeks of gestation period during a visit to a health provider in participating urban districts; children immunized; women and children who received basic nutrition services; deliveries attended by skilled personnel in health institutions in participating rural and urban districts; cumulative number of health facilities enrolled in RBF program in participating districts; people who received essential health, nutrition, and population services; and number of children under five who had their Mid Upper Arm Circumference and Height measured by community health workers (CHWs) as part of growth monitoring); one (8 percent) was substantially achieved (women living with HIV who are initiated on ARVs to reduce risk of MTCT in participating rural districts: 99 percent); and one (8 percent) was not achieved (women 15-49 years who are new acceptors of long term, reversible family planning (FP) methods in participating districts) due to myths and misconceptions around these methods, as well as implementation challenges (lack of skilled personnel at lower level facilities, high levels of attrition of trained personnel, and need for referrals). 28. The PDO part related to increased quality of an integrated package of RMNCAH-N services was also substantially achieved, with 59 percent of all POIs and IRIs targets achieved/surpassed; and 41 percent substantially achieved. These results are directly attributable to the project which added quality as part of the PDO with the approval of the AF IV in 2019 and incentivized providers to focus not only on coverage but quality as well. The RBF approach played a substantial role in strengthening the structural quality (e.g., investments in maternity waiting homes, waste management and water systems; procurement of essential equipment, drugs and supplies) as well as the quality of clinical care through the innovative CQI initiative. During the ICR mission, numerous stakeholders highlighted that they were able to identify problems jointly with community representatives through the revitalized HCCs to incorporate patient concerns and complaints (such as lack of drugs, long distances to travel for pregnant women), and had the autonomy and flexibility to use the RBF subsidies to address gaps and improve the quality of care. 29. The POIs related to the quality of services demonstrated substantial achievements. Two POIs (42 percent) were surpassed (quality scores by health facilities in participating rural and urban districts; children under five with pneumonia correctly managed in participating districts); two (39 percent) were substantially achieved (maternal Page 11 deaths given audits as per protocol in participating districts: 95 percent; partographs correctly filled in participating districts: 95.4 percent); and one (19 percent) was not achieved (participating district, provincial, and central hospitals that have registered an increase in quality scores since last quarter), reflecting the inherent difficulties and complexities of improving quality at hospitals, as well as the implementation period which became shorter. While the Quality Focused RBF at Provincial and Central Hospitals and the Extended Supply Side Community RBF, generated positive results, as discussed in detail in Annex 5 (Government’s ICR) there was insufficient time to fully test these approaches for efficacy and efficiency, with these activities remaining part of the unfinished agenda. 30. The service quality IRIs was substantially met. Out of a total of 3 IRIs, two (68 percent) were achieved/surpassed (DHEs in participating districts using quality tool for supervision of health facilities; and health facilities implementing CQI model in the participating rural districts), and one (32 percent) was substantially achieved (RBF contracted facilities in CQI districts with CQI standard operating procedures: 95 percent). These indicators reflect the solid progress made in providing integrated support and supervision using the quality checklist which was one of the key system strengthening dimensions of the RBF program. The intensification of supervision missions at the decentralized levels, which was one of the main RBF strategies, represents good practice as it serves to identify shortcomings, benefit from mentoring, and propose solutions with potential cross cutting benefits for the overall quality of care. Following the completion of the project, some sites continued to conduct regular self-assessments and data reviews even in the absence of funding as these activities were viewed as key to improving performance. PDO Part B: Strengthen institutional capacity to manage performance-based contracts—substantial 31. PDO Part B was substantially achieved, as indicated by both POIs and IRIs. When analyzing the overall level of achievement, 82 percent of all POIs/IRIs was achieved/surpassed (64 percent) or substantially achieved (18 percent), while 18 percent was not achieved. PDO Part B is rated substantial. The only POI (health facilities managed RBF contracts by the MOHCC Project Coordination Unit (PCU) in participating rural districts) included as part of the AF IV was substantially achieved (90.6 percent) as the Government assumed responsibility for the RBF program. Three IRIs (78 percent) were achieved/surpassed (health personnel and CHWs receiving training on RBF in participating districts; health workers that received first time training or refresher training on EONC using skills lab; and process evaluation to examine the institutionalization of RBF in the government system as a viable approach to sustainability of RBF), and one (22 percent) was not achieved (process evaluation to examine the effectiveness of quality focused RBF at provincial hospitals in improving quality of maternal and neonatal care for improved outcomes could not be carried out due to the limited time during which this was implemented). PDO Part C: Strengthen the COVID-19 response—substantial 32. PDO Part C was substantially achieved, as indicated by both POIs and IRIs. When analyzing the overall level of achievement 85 percent of all POIs/IRIs was achieved/surpassed (70 percent) or substantially achieved (15 percent); and 15 percent was not achieved. PDO Part C is rated substantial. The only POI (percentage of close contacts confirmed COVID-19 cases followed up based on national guidelines) was surpassed. Two IRIs (65 percent) were achieved/surpassed (isolation centers complying with infection prevention and control (IPC) guidelines; and number of affected people including health workers reached with clinical and psycho-social support); one (17 percent) was substantially achieved (health facilities with personal protective equipment (PPE) commodities: 93.3 percent); and one (17 percent) was not achieved (COVID-19 treatment centers with oxygen therapy). These results reflect interventions supported under the project to enhance infection prevention and control, strengthen surveillance, and improve availability of critical equipment. The Risk Communication and Community Engagement Strategy supported under the project strengthened communication and community engagement efforts, addressing misinformation and promoting correct knowledge. Annex 5 (Government’s ICR) provides details of the full range of interventions supported. Page 12 33. The Impact Evaluation (IE) conducted in 2015 found positive results in the RBF districts compared to non- intervention groups both in terms of coverage and quality of services.12 For incentivized indicators, findings were consistent with household survey results with significant increases in several services. The IE found that RBF interventions significantly improved coverage of institutional deliveries with a 12 percent point increase in delivery by skilled provider in RBF districts. RBF also significantly improved coverage of post-partum care from a qualified provider. The IE found improvements in the RBF-supported facilities in some measures of structural quality (e.g., higher incidence of biomedical waste disposal, increased availability of iron and folic acid, and select equipment) and in the quality of some services. Improvements in other aspects of quality of care were mixed but generally positive, with health facilities showing relatively better performance in some clinical processes, increased responsiveness to community needs, and stronger team-based decision-making. Providers working in RBF- supported facilities reported higher levels of autonomy on selective aspects of facility management such as how budgets are allocated. An equity analysis of RBF found that benefits accrued more to households below median wealth, particularly for delivery by skilled provider, delivery in facility, postpartum care, and post-natal care. For the non-incentivized services, there was no evidence of negative effects, which was one of the risks identified at appraisal. The generally encouraging results from the IE strengthened the justification for the decision of several development partners to support a streamlined RBF program in the remaining 42 districts, as discussed under the donor engagement section below. 34. Process evaluations identified key factors contributing to the RBF-supported interventions. Health workers appreciated the improved working conditions, greater autonomy, and supportive supervision, though typical challenges of RBF programs persisted (e.g., perceived unfair incentive distribution in some facilities, higher workloads). Positive spillover effects of RBF included improved availability of medicines for both incentivized and non-incentivized conditions, as well as logistical support for medicine distribution. Though the IE demonstrated that the approach had limited or no unintended negative effects on non-incentivized services, the process evaluations found that some facilities faced challenges to maintain a focus on some non-incentivized conditions. 35. Process evaluations also underscored the importance of the RBF strategies. First, facilities with strong supportive supervision and mentoring from the DHEs demonstrated better performance, while weaker supervision led to inconsistent results. Second, community engagement played a significant role in improving utilization and coverage with facilities that effectively incorporated community feedback and worked closely with HCCs enhancing their performance. Challenges with unclear roles and inconsistent participation persisted at some facilities. Third, target setting and incentive pricing inherent in the RBF design were critical drivers of facility-level responsiveness, with attractive pricing stimulating demand for services, while inadequate levels of financing created financial pressures for health facilities with lower catchment populations. These findings highlighted the need for continued refinement of supervision structures, community engagement strategies, and pricing models, to maximize the impact of the RBF program. 36. Findings from national surveys highlight improvements in key maternal and child indicators during 2010-2019 but a deterioration during the COVID-19 pandemic and macroeconomic shocks (2019-2024). During 2010-2019 there were substantial reductions in the large coverage gaps between urban and rural areas. By 2019 coverage levels in rural areas exceeded those in urban areas for four of the six interventions supported under the project (i.e., first ANC visit before 16 weeks, ANC 4+, full immunization coverage of basic vaccines, and treatment of childhood diarrhea with ORS). The maternal mortality ratio dropped from 960 deaths per 100,000 live births in 12 Rewarding Provider Performance to Improve Quality and Coverage of Maternal and Child Outcomes, Evidence from Zimbabwe RBF Impact Evaluation, June 2016. The IE used a difference-in-difference estimation method to evaluate the impact of RBF with outcomes after two years of program exposure (2012-2014) compared to pre-intervention period (2008-2010). Page 13 2010-2011 to 462 deaths in 2019. 13 Under-five mortality declined from 84 to 73 deaths per 1,000 live births and stunting prevalence dropped from 32 to 23.5 percent during the same period. Although the progress made during 2010-2019 cannot be attributed solely to the RBF program, it is notable that these improvements occurred as the program was rolled out in the 18 rural districts, along with the Health Transition Fund (2012-2015) and the Health Development Fund (2016-2021) in the remaining 42 rural districts. 37. During 2019-2024 progress on reducing mortality, fertility and stunting levels stagnated or reversed due to the COVID-19 pandemic and deteriorating macroeconomic situation. Preliminary results from the 2023-2024 Demographic and Health Survey found a stagnation in coverage of key maternal health services (i.e., ANC 4+, skilled birth attendant), and a deterioration in child health interventions (i.e., fully vaccinated, treatment with ORS). During this difficult period progress in reducing under-five mortality and fertility stagnated while stunting prevalence (increased by about 15 percent) and neonatal mortality (rose by 19.4 percent) deteriorated. These trends reflect in part factors beyond the health sector including rising food insecurity and household poverty over the past five years. Analysis examining whether the trends during 2019-2024 would have been worse without the infusion of additional funds to support Zimbabwe’s COVID19 response and the RBF program is beyond the scope of this ICR. Evidence of the quick rebounding of key indicators such as institutional delivery by as early as 2021 indicate the important role these funds may have played in the recovery of Zimbabwe’s health system. Annex 8 provides a fuller discussion of national trends. Justification of Overall Efficacy Rating 38. The overall efficacy of the Zimbabwe Health Sector Development Support Project is rated Substantial. Overall, 88 percent of all POIs/IRIs was achieved/surpassed (71 percent) or substantially achieved (17 percent) and only 12 percent was not achieved, as summarized in Table 2. These results are remarkable given the negative impacts of the COVID-19 pandemic and the continuing precarious economic situation. Table 2: Summary Ratings PDO Part Achieved/Surpassed Substantially Achieved Not Achieved 100%+ 80-99% <65% A Coverage and Quality 73.0 17.0 10.0 B Institutional Capacity 64.0 18.0 18.0 C COVID-19 Response 70.0 15.0 15.0 Overall Efficacy 71.0 17.0 12.0 C. EFFICIENCY Assessment of Efficiency and Rating Cost-Benefit Analysis 39. The economic analysis for the project shows a solid economic rationale for the investment. The estimated impact is that over 12.5 years, roughly 38,260 children’s lives would have been saved, and 2,384 women’s lives would be saved. The net present value (NPV) of total project benefits is estimated at US$134.7 million at a 10 percent 13 The MMR of 960 per 100,000 live births comes from the 2010-2011 DHS, whereas the estimate in the original PAD (790 per 100,000 live births) came from Trends in Maternal Mortality: 1990-2008. World Health Organization (WHO)/United Nations Children’s Fund (UNICEF)/United Nations Population Fund (UNFPA)/World Bank. Page 14 discount rate. Given the net present cost of the project of US$55.6 million, this yields a cost-benefit ratio of 2.42 for every dollar invested and an overall Net Present Value of the total project benefits of US$79.1 million. The internal rate of return was estimated to be 14 percent. An expanded description of the economic analysis and the assumptions used are included in Annex 4. Implementation Efficiency Analysis 40. Implementation efficiency is rated Substantial. The project supported the delivery of evidence-based cost effective maternal and child health interventions (e.g., ANC, institutional deliveries, immunization) through the RBF program which directly tied funding to achievement of results. The RBF scheme provided incentives to providers to use resources efficiently to attain the desired results, hence there was a clear link between how the money was spent and the outcomes to be achieved. A Cost-Effectiveness Analysis of the Zimbabwe RBF Program conducted in the initial phase before the national roll-out found the program to be very cost effective with an Incremental Cost Effectiveness Ratio (ICER) below the country’s per capita GDP.14 41. Implementation efficiency was also enhanced by the introduction of risk-based verification in subsequent years to address concerns with the rising cost of the initial verification process.15 The risk-based verification stratified facilities according to risk to reduce the need for frequent visits while penalizing for over-reporting. As a result, there was a steep drop in verification costs from US$1.7 million at inception (2012) to roughly US$0.4 million (2018). The adoption of this less costly verification method made it easier for the MOHCC to mainstream these activities in the public system. The innovative urban voucher program also represented an efficient use of resources given its highly targeted focus on poor urban women (Box 1). The timeliness of disbursement of grant funds during 2012-2024 was satisfactory, with disbursements exceeding projections, as absorptive capacities were established. Against these positive aspects, however, the delays in the disbursement of Government funds after the institutionalization of the program combined with the eroding purchasing power of the local currency hindered implementation effectiveness during the final five years (e.g., high inflation contributed to reduction in the value of the RBF funds with mixed results in strategies used to cushion inflationary pressures such as invoicing in USD to retain the value of RBF funds; and the erosion in salaries led to nurses strikes and external migration requiring recruitment of locum nurses). The restructuring at the MOHCC in 2020 combined with competing demands stemming from the COVID-19 pandemic also translated into some delays in decision making. There were also delays in fund flows between AF transitions which were beyond the control of the World Bank as they were dependent on donor grant funding cycles. 42. Based on the economic analysis and implementation efficiency analyses, overall efficiency is rated Substantial. The substantial rating reflects the cost-effectiveness of project interventions and positive estimated returns with moderate shortcomings in implementation efficiency. 14 Cost-Effectiveness Analysis of Results-Based Financing in Zimbabwe, Shepard, D. and Zeng W., World Bank and Brandeis University, June 9, 2016. The ICER for coverage and quality was estimated at US$439 per Quality Adjusted Life Year (QALY) which compared favorably to similar RBF programs. The Cost Effectiveness Analysis compared the difference in costs between RBF and control districts to the differences in health outcomes expressed as QALYs to estimate the ICER. With a population of roughly 3.5 million in the RBF districts, RBF was estimated to generate 528 QALYs/100,000 population/year. 15 Implementation of Risk-Based Verification in Zimbabwe’s Results Based Financing Program: A Review of Benefits, Opportunities and Lessons from Implementation, August 2019 (updated 2021). Page 15 D. JUSTIFICATION OF OVERALL OUTCOME RATING 43. The overall outcome is rated as Satisfactory. This is based on the project’s continued high relevance, and substantial ratings on efficacy, and efficiency. 16 E. OTHER OUTCOMES AND IMPACTS Gender 44. ’ and children from the poorest and most vulnerable households. The project initially targeted beneficiaries in rural districts with an implicit pro-poor focus. Subsequently the project added the urban voucher program which targeted pregnant women in low- income urban communities in Harare and Bulawayo. While the needs in urban areas were huge, a rigorous data-driven initiative was used to identify unserved areas and screen eligible beneficiaries using the social development officers in the targeted low-income urban settings. 45. The project aimed to address significant barriers to health care access and quality of care for women. The women targeted through the project faced multiple constraints, with limited utilization of health services due to poor access, inadequate quality, high user fees, and socio-cultural barriers. The project incentivized the delivery of a comprehensive package of evidence based maternal and child health interventions through RBF, progressively adding additional activities over the 12.5-year implementation period. The project’s focus on improving maternal and reproductive health was critical to reducing the high level of maternal mortality at project inception. Institutional Strengthening 46. The project introduced performance-based contracting in Zimbabwe and established institutional capacity at all levels of the health system to mainstream the approach. While the roll-out started with the recruitment of a specialized agency (Cordaid) with a long-standing global track record on RBF which initially served as the PIE, structures were established and/or strengthened to progressively transfer capacity and responsibility to the Government. This included: (i) National RBF Steering Committee that provided oversight and guidance; (ii) District Steering Committees (which were social services subcommittees) that provided local oversight; and (iii) HCCs that prepared plans and budgets, collaborated in the procurement process, and oversaw the use of subsidies. Training and mentoring of health personnel and CHWs have been key to enhancing understanding, ownership, and effectiveness of the RBF program. This remains important to sustain engagement given staff turnover and the dynamic nature of the RBF scheme which requires continuous adaptations. 47. The project also played a pivotal role in supporting the institutionalization of RBF in Zimbabwe. This is part of the legacy of the project and one of its most transformative aspects. The trajectory of events and milestones that led to the institutionalization of the RBF approach is well captured in a case study produced by the GFF/World Bank to share with other countries embarking on similar reforms.17 The main actions taken included: (i) piloting the RBF approach in two front runner districts (2011) and scaling it up to another 16 rural districts with funding from the project (2012); (ii) introducing an RBF scheme in the 16 The overall satisfactory rating is broadly consistent with the final implementation status and results report (ISR) rating (marginally satisfactory) as the ICR takes a long-term holistic view of performance and concludes that on balance the development objectives were substantially achieved with minor shortcomings. 17 Zimbabwe: Results-Based Financing, Knowledge & Learning Case Study, World Bank, 2021. Page 16 remaining 42 rural districts with funding from the Health Development Fund (2014); (iii) introducing an innovative urban RBF voucher program (Box 1); (iv) confirming Government commitment and support to institutionalizing the RBF scheme, in line with the Mid-Term Framework for RBF which shaped the vision on institutionalization and proposed that the Programme Coordinating Unit of the MOHCC become the National Purchasing Agency (2015); (v) mobilizing counterpart funding to finance the 18 districts with the World Bank reducing financial support and providing technical assistance (2018-19); and (vi) transferring responsibility for funding of the program to the Government, starting with the 18 districts (2018) and subsequently the remaining 42 districts (2022). In total the Government provided US$53.3 million while the grant funding totaled US$78 million during the 12.5 years. The institutionalization process brought on new challenges with the deteriorating macroeconomic and fiscal situation and the COVID-19 pandemic during the final five years. As a result, disbursements were often delayed, sometimes fell short of allocations, and the value of the subsidies eroded as funds were disbursed in the local currency. 48. The RBF Approach had other positive spillover effects on the health system. First, it promoted community participation, a key national priority dating back to independence. The revitalized HCCs played a key role in seeking community views and collaborating with health personnel in preparation of plans to address gaps, serving as a key link between communities and the health system. Village Health Workers were mobilized and incentivized through the supply side RBF scheme to conduct awareness raising activities and improve health seeking behavior. CBOs were an integral part of the RBF approach, monitoring and tracking client satisfaction and feeding information back to health facilities. Second, the RBF scheme strengthened core health system functions, such as decentralized planning, budgeting, and integrated support and supervision. Health facilities had greater autonomy over how funds were spent. The integrated support and supervision visits played an important role in improving understanding of the RBF approach, applying the quality checklist, and providing mentorship on improving performance. Another innovation which was introduced at some participating sites was self-assessments in between supervision visits, which helped to bolster ownership. Third, RBF played a catalytic role in improving availability and quality of information systems as maintaining good records was critical to the incentive scheme, as discussed in the M&E section. Fourth, the innovations introduced (e.g., risk-based verification, electronic verification, electronic partograph, incentive calculator, CQI) bolstered health system performance as discussed in Annex 5 (Government ICR). Poverty Reduction and Shared Prosperity 49. The project had a pro-poor focus which likely contributed to poverty reduction and shared prosperity. The project targeted beneficiaries in rural areas as well as urban poor women, improving their access to critical maternal and child health services, and lowering financial barriers through the abolishment of user fees. Given Zimbabwe’s continuing economic challenges, and the erosion of household purchasing power, the project contributed to improving the health status and economic well-being of some of the most disadvantaged groups. The IE also found that that the RBF scheme had a pro-poor focus. Page 17 Box 1: Urban Voucher Program Background Building on the initial promising roll-out of RBF in rural districts, an Urban Voucher Program was introduced in 2014 to address inequities in access to Maternal, Neonatal and Child Health (MNCH) services for the urban poor. The Urban Voucher Program aimed to provide free MNCH services for the poor in targeted low-income districts in Harare and Bulawayo metropolitan provinces. The pilot districts were rigorously selected, in consultation with City Health Departments, based on levels of poverty and population size. Beneficiaries were identified by CHWs, and a poverty assessment tool was subsequently administered by social service officers to determine eligibility. Beneficiaries paid a nominal fee to ensure ownership. The voucher provided access to all maternity services, including ambulance transport and counselling on sexual and gender-based violence. As under the rural RBF, facilities received incentive payments based on improvements in the quality of care; CHWs were incentivized to raise awareness and improve health seeking behavior; and CBOs monitored performance. A process evaluation conducted in 2018 demonstrated the feasibility of the urban RBF program and recommended scaling it up to reach additional beneficiaries. The Government subsequently requested World Bank support to develop options for scaling-up and institutionalizing the program. In line with the Urban RBF Institutionalization Framework, the Government started to provide counterpart financing in 2019. During 2021-2022 additional health facilities were enrolled in the program, bringing the total to 35, in addition to the four central hospitals which participated in the scheme. Achievements -The voucher program successfully targeted poor women, reducing inequalities in access to healthcare in low-income urban areas. In total, over 81,000 beneficiaries have been enrolled in the program since its inception. -The pay-for-quality component succeeded in improving the quality of services at participating health facilities . Like the rural RBF scheme, participating facilities used the subsidies to hire locum nurses, procure drugs and supplies, improve the physical environment, and provide staff incentives. -The community component encouraged poor women to deliver in a health facility, implemented client satisfaction surveys, and provided feedback to health staff. -A review of the Mbare clinic performance found high levels of service quality (84 percent) and client satisfaction (78 percent). With the clinic receiving about US$16,000 per quarter in 2023, substantial investments were made in strengthening the provision of critical maternity services. As a result, the facility reported steep drops in early neonatal deaths and macerated stillbirths. (Source: Field Visit Brief: Urban Voucher/RBF Component-Mbare Clinic, July 22, 2024.) Sustainability Prospects for sustaining these achievements are mixed. On one hand, there is broad ownership of the Urban Voucher Program and of the broader RBF urban scheme, and recognition that the demand for these services remains very high. On the other hand, the Government’s ability to sustain and scale up activities is constrained by the overall fiscal situation with continuing dependence on donor support. Page 18 III. KEY FACTORS AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION Assessment of Project Design 50. Project design was informed by the growing positive experience with performance-based contracting. Given the dire situation in the country with the steep decline in the utilization of health services, performance-based contracting was expected to accelerate the uptake of key maternal and child health services. A pilot project in two front runner rural districts found promising results, and experience from other countries in the region with performance-based contracting found similar results. Therefore, there was mounting evidence that the approach was promising. Overall, the project design was simple and straightforward in line with other RBF programs. Government ownership was critical in shaping the design and adapting it to the Zimbabwe country context. Consultations with other development partners during preparation proved beneficial in leveraging technical expertise and building a foundation for further dialogue in subsequent stages of the project. Assessment of Project Risks 51. The PAD included a candid discussion of both country and sector-specific risks. The risks were appropriately identified: (i) deterioration in the political environment which may impede implementation with an international NGO (Cordaid) recruited to partly mitigate this risk; and (ii) RBF-related concerns, including inadequate understanding and ownership, excess focus on incentivized indicators with unintended declines in other indicators, over-reporting or mis-reporting of results, and uncertainty over sustainability, with mitigation measures put in place (e.g., providing training at all levels, introducing verification and counter verification measures, conducting impact evaluations to track results, carrying out periodic spot audits, and providing parallel technical assistance and policy advice on health financing, including mainstreaming the RBF approach). While the risks and mitigation measures were well defined, with the benefit of hindsight, the overall risk rating of medium may have been overly optimistic with a substantial risk probably more realistic, given that the macroeconomic situation remained precarious. Assessment of Institutional Arrangements 52. The institutional arrangements initially relied on external expertise in RBF to maximize chances of success and to mitigate risks related to the difficult country context. Cordaid, which had extensive global experience in RBF, was competitively selected to serve as the PIE. The World Bank entered into a Project Agreement with the PIE outlining its roles and responsibilities; and the Government entered into a Subsidiary Agreement with the PIE stipulating the arrangements for disbursing the proceeds of the grant. The PIE established satellite offices (Local Purchasing Units) at the district and provincial levels and worked collaboratively with the MOHCC to provide training and mentoring on RBF to key stakeholders. In line with good practices, a Project Implementation Manual (PIM) was produced and updated regularly, through a rigorous consultative process, to stipulate modifications to the package of services and the fee schedule. The National Steering Committee, which provided guidance and strategic support to the MOHCC and the PIE, included a sub-committee of key representatives of the Government, CSOs, and faith-based organizations, recognizing the need for broad-based consultation and consensus building on RBF. Likewise, the institutional arrangements included community participation which was viewed as critical to enhancing accountability at the local level, as well as consultations with other development partners, recognizing that collaboration was important for success of the project. The institutional arrangements were reasonable in the initial country context, given the need to show rapid results and to build national RBF capacity. Page 19 B. KEY FACTORS DURING IMPLEMENTATION Factors Subject to Control of Government and Implementing Agencies 53. Commitment and Leadership: The Government demonstrated strong commitment to RBF throughout the entire implementation period. The Ministry of Finance and Economic Development provided high-level leadership to support the effective implementation of RBF. The Ministry of Health and Child Welfare (which was later renamed as the MOHCC) played a pivotal role in embracing RBF and working collaboratively with Cordaid to roll out the program. Despite initial resistance to RBF among some high-level policymakers, the program was adapted to the local context with ongoing consultations and consensus building, avoiding what some feared might become a one- size-fits-all approach. The Government showed remarkable commitment to institutionalization of RBF, working diligently to build capacity in the public system, bolster ownership at the national and decentralized levels, and address emerging issues. This commitment was reflected in the increased allocation of counterpart funds for RBF. 54. Institutional Capacity: The recruitment of Cordaid proved critical to the rapid introduction of RBF into all 18 districts given the institution’s track record and expertise. The Government had the foresight to contract out the initial design and implementation of the RBF program, while building technical capacity to take over its management. The nuts and bolts of how the RBF would work and who was responsible for what were clearly spelled out in the PIM which was updated regularly as the design evolved and lessons learned were reflected in operational modalities and new governance arrangements. A notable example is the counter-verification function which was initially contracted out to a local university and subsequently delegated to the Health Professions Authority (HPA). These were both important cost-saving and capacity building/institutionalization measures. With the progressive institutionalization of RBF, there was also a rise in delays in disbursements and difficulties as reported in various ISRs. Delays in disbursement of counterpart funds occurred periodically during the early years and worsened in the final five years due to the deteriorating fiscal situation. 55. Donor Engagement: The early engagement of development partners was key to expanding RBF to the remaining 42 rural districts. Funding for RBF was introduced as part of a broader, long-term donor fund supported by various partners (United Kingdom’s Foreign, Commonwealth and Development Office, European Union, Swedish International Development Cooperation Agency, Irish Aid, Gavi, the Vaccine Alliance) in close collaboration with United Nations Children’s Fund (UNICEF) as the fundholder, and Crown Agents as the PIE. The Health Transition Fund (2012-2015), which transitioned to the Health Development Fund (HDF) (2016-2021) and later became the Health Resilience Fund (2022-2025), provided substantial funding for over a decade for the procurement of essential drugs and supplies; community health RBF program; and a streamlined health facility RBF program (which did not include external verification or formally implement CQI). An independent evaluation of the performance of the HDF concluded that the financing contributed to sustaining the availability, quality, accessibility, acceptability, and utilization of care for RMNCAH nationwide.18 The evaluation found that during 2016-2018 the HDF: (i) increased the availability of medicines but stockouts persisted; (ii) played a key role in strengthening Village Health Workers; (iii) improved access and availability of care and strengthened accountability through the health facility and community level RBF; and (iv) enhanced governance at all levels of the health system. Crown Agents (which coordinated the RBF scheme) introduced several innovations, such as using community sisters to conduct first line verification, rolling out mHealth technologies (e.g., mobile electronic invoicing which lowered costs and improved accountability; staff incentives calculator which improved transparency in distributing incentives; videos that served as training guides to reduce need for off-site training). 18 Independent Evaluation of the Health Development Fund, London School of Tropical Medicine Centre for Maternal and Newborn Health, December 20, 2018. Page 20 Factors Outside Government and Implementing Agency Control 56. The continuing macroeconomic and fiscal challenges faced by the Government intermittently impeded implementation of project activities. This was a recurrent theme particularly during the final five years of project implementation. The macroeconomic and fiscal challenges had several implications. First, at the national level, the Government faced challenges in disbursing funds in a timely fashion. Despite the commendable commitment to RBF, disbursements were delayed by several months, demotivating staff and impeding activities, such as the rural RBF, the urban voucher program, and the community and hospital RBF pilots. Moreover, in 2024 there was a decline in the allocation of funds for the health sector, which further exacerbated the situation. Second, at the provider level, the eroded value of the local currency resulted in the deterioration in salaries which in turn led to occasional strikes and staff attrition, with doctors and nurses seeking better opportunities abroad. These exogenous shocks made it increasingly difficult to sustain the level of effort and funding for the RBF program with facilities adjusting activities to resource availability. The support provided by Cordaid and the MOHCC-PCU partly mitigated these challenges by introducing corrective measures (e.g., recruiting locum nurses, providing technical assistance). 57. Like other countries, the COVID-19 pandemic resulted in lockdowns and staff diverted towards the emergency response. In Zimbabwe, the COVID-19 pandemic coincided with a deterioration in the macroeconomic context with triple-digit inflation, periodic health worker strikes, a decline in the value of the RBF funds, and staff attrition, creating a perfect storm. The 2021 Zimbabwe Economic Update found that the COVID-19 pandemic disrupted livelihoods, expanding the number of extremely poor citizens by 1.3 million, and increasing extreme poverty overall to 49 percent in 2020. 19 The pandemic also put pressure on strained public resources, exacerbating implementation challenges and severely affecting the delivery of public services. In addition, supply-side challenges facing the health system—following a prolonged period of doctor strikes, reduced working hours for nurses, and limited and slow access to personal protective equipment—initially contributed to a decline in the coverage and quality of essential health services. The number of institutional maternal deaths increased by 29 percent in 2020 compared to 2018, while home deliveries rose by 30 percent. Factors subject to the control of the World Bank 58. The World Bank team provided strong support for the RBF program. As discussed in greater detail in the Bank Performance section below, the World Bank team mobilized grant funding over a 12-year period, supported the Government to design and implement an innovative program, and provided complementary technical assistance to address health financing and human resources issues. The World Bank team strategically mobilized technical support to support the roll-out, such as the Population Council which was mobilized to review the urban voucher program and make recommendations to improve it. As is good practice, the World Bank team conducted regular supervision missions, supported the Government to adapt the design to evolving needs, and ensured regular evaluations were conducted to draw lessons and make course corrections. The Zimbabwe RBF experience was featured in a case study produced to share experiences and lessons with other countries in the region as it had multiple features considered good practice (strong government ownership, solid design, learning and knowledge sharing, scale up and institutionalization strategy). Based on the positive initial results of the RBF program, Zimbabwe was selected to host the 2016 annual meeting of the Health, Nutrition and Population (HNP) Global Practice conference on Health Results Innovation which brought together representatives from 26 countries. 19 The Zimbabwe Economic Update, Overcoming Economic Challenges, Natural Disasters, and the Pandemic: Social and Economic Impacts, June 2021. Page 21 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 59. The M&E Design was Satisfactory. The PAD included five well defined POIs which focused on priority interventions that were lagging (ANC, institutional deliveries, ARV therapy for pregnant women, family planning, and immunization). The POIs and IRIs were well selected and in line with international practice in terms of tracking and incentivizing the delivery of evidence-based interventions to improve coverage of life-saving maternal and child health services. The PAD included baseline and target figures, data sources, and institutional responsibility for data collection with most data coming from the Health Management Information System, using the District Health Information Software and complemented by national surveys. The design appropriately focused on introducing a strong verification and counter-verification function to be conducted by a third party which was important for the effectiveness of the RBF program. Finally, the attention given to conducting periodic evaluations (both quantitative and qualitative) to draw lessons and make course corrections from the roll-out of the RBF program was in line with similar well-designed programs. M&E Implementation 60. M&E implementation was generally well done. During the 12.5-year implementation period, the indicators were progressively adjusted to raise the bar in terms of achievements, dropping indicators that were attained, changing some POIs into IRIs, and adding new indicators to reflect the increased focus on quality, RBF institutionalization, and the COVID-19 response. The progressive amendment of indicators is in line with good practice in RBF as incentivized indicators need to be continually monitored and adjusted to have an impact on service delivery over time. The verification and counter-verification functions experienced some difficulties and delays and were found to be too costly. In response, Zimbabwe gradually introduced risk-adjusted verification and transferred responsibility for counter-verification from a local university to the public sector. The M&E responsibility was also transferred from Cordaid to the MOHCC as the RBF program was progressively institutionalized. There were also moderate difficulties and delays in reporting on some indicators resulting in occasional delays in fund disbursements, which led to a temporary downgrade of the Monitoring, Documentation, and Verification component rating from Satisfactory to Moderately Satisfactory (ISR #20, 2022). In retrospect, some targets may have been overly ambitious and beyond the control of providers (e.g., ANC during the first 16 weeks in rural areas is linked to deeply embedded socio-cultural norms; uptake of long-term family planning methods is influenced by women’s preferences, side effects, and other demand side issues). M&E Utilization 61. M&E utilization was strong. There was systematic tracking of results to facilitate fund disbursements and inform planning and budgeting. The results of the IE and various process evaluations were used to review progress, bolster commitment, and make course corrections. These evaluations also provided critical evidence for the value added of RBF approach which inspired the replication of RBF in the remaining 42 rural districts. 62. The RBF scheme contributed to improving data quality and availability which were critical for the approach. Prior to RBF implementation, many health facilities in participating districts lacked a systematic process for organizing patient records and reporting forms. Because of the reporting requirements of the RBF program, facility record keeping and reporting improved so that facilities could receive and benefit from the incentives. Studies conducted on the verification function showed that large gaps between reported and verified data for key RBF indicators narrowed over time, a marker of improved data quality. Field visits conducted during the ICR Page 22 mission also suggested that improvements occurred in CHW reporting to health facilities, a critical aspect of strengthening linkages between community and facility service delivery mechanisms. 63. The RBF approach also strengthened data use which was critical to the incentive scheme. During the ICR mission field visit interviews, health care staff across Mashonaland East, Matabeleland South, and urban sites in Bulawayo and Harare consistently indicated positive shifts in the institutional culture around regular joint self-assessment practices based on the RBF checklists and data on incentivized interventions. These self-assessments were reviewed during the supervisory and mentoring visits regularly conducted when the RBF program was being fully implemented, another best practice in the use of data to improve the quality of care. The RBF program also contributed to improving adherence to clinical guideline20 Several interviewees commented that the introduction of an accountability mechanism through the RBF program generated healthy competition across participating health facilities as all strove to be top performers.21 Another best practice introduced through the RBF program was a client satisfaction survey and the revitalization of HCCs which served as a platform for discussing survey results and improving responsiveness of health facilities to community health needs. Justification of Overall Rating of Quality of M&E 64. The overall rating for M&E is Substantial. The project was designed with a simple M&E framework which was subsequently modified and refined with minor shortcomings in implementation. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 65. Environmental Safeguards. As noted in the PAD, the project triggered Operational Policy (OP)/Bank Policy (BP) 4.01 and was classified as Category B-Partial Assessment. No civil works was to be undertaken under the program. The main environmental risk was the anticipated increase in health care waste generated by health facilities. An in-country consultative process took place, and a Health Care Waste Management Plan was prepared by the Ministry of Health and Child Welfare and publicly disclosed on May 19, 2011. In addition, the PAD noted that a quarterly quality verification tool used to supervise participating health facilities included indicators to track performance on medical waste practices, hence providers had a financial incentive to adopt safe practices. 66. During project implementation, quarterly assessments were conducted to monitor compliance on waste management with notable improvements in RBF-supported health facilities (i.e., RBF-supported facilities had a 16-percentage point higher incidence of biomedical waste disposal), which benefited from greater autonomy and flexibility to use RBF funds to procure critical equipment and supplies. As the project design evolved over the 12.5-year period, health facilities were allowed to use RBF funds for minor civil works, in which case national and local guidelines were followed. Likewise, with the introduction of the COVID-19 response component as part of AF V in 2020, the environmental requirements were further modified with the preparation of an Environmental and Social Management Framework and an Infection Control and Waste Management Plans (ICPWMPs), as risks increased, and institutional capacity was strengthened through the recruitment of an Environmental Specialist. 67. The project supported participating health facilities to prepare their own ICPWMPs which were reviewed and cleared by MOHCC, with technical support from the PIU. Even though instruments were prepared, implementation of some activities (i.e., waste management plans) proved challenging due to lack of budgets to 20 In more than one facility visited, health staff relayed that they observed improvements in key indicators around clinical management of obstetric complications (e.g., case management of pre-eclampsia/eclampsia and hypertensive disorder of pregnancy, management of prolonged rupture of membranes, and reduced time from decision to perform a c-section delivery and performance of the procedure) and commented that the RBF program resulted in greater staff adherence to clinical guidelines. 21 For example, transparency around health facility performance on the RBF indicators enabled facility staff to compare their results including financial incentives received with other facilities. Page 23 support all gaps, resulting in a temporary downgrade to moderately satisfactory. 22 PIU commitment was instrumental in supporting participating facilities to improve compliance with one environmental specialist making significant efforts to effectively support more than 1,700 facilities. The PIU set aside a reasonable budget to support compliance activities, transportation of waste, capacity building, selective aspects of medical waste management, occupational health and safety activities, and risk assessments before any works commenced. During the January 2025 ICR field visits, various stakeholders described how the RBF subsidies were used to improve waste management practices and availability of water. Overall performance at completion is rated Satisfactory. 68. Social Safeguards. As noted in the PAD, the project was anticipated to have a positive social impact by improving access to health care for the poorest households, which has been observed over the 12.5 years of project implementation. The elimination of user fees significantly improved access to maternal and child health services at participating facilities. The RBF scheme included a special bonus for the remoteness of the facility, which may have contributed to improve staff retention and incentivize provision of services for the most vulnerable. Community ownership was bolstered through the revitalized HCCs which played a key role in decision-making on the optimal use of the RBF funds and in addressing community concerns and complaints. The results from the RBF scheme were shared at the community level to enhance transparency and accountability. To enhance client engagement and seek feedback various tools and approaches were used as discussed in Annex 5 (Government ICR). As noted in the interim ICR produced in 2021, Zimbabwe had a well-established grievance redress mechanism (GRM) that was effectively utilized during implementation. 23 All grievances received during implementation were resolved and closed by the end of the project. During the last few years with AF V, the project also provided support to a COVID-19 emergency operations call center, enhanced access to critical public health information, and promoted citizen engagement through social media platforms. Overall, the performance on social safeguards is rated Satisfactory. 69. Financial Management. A financial management assessment was carried out at appraisal in accordance with the Financial Management Manual for World Bank-Financed Investment Operations. It concluded that the PIE’s capacity met the World Bank’s minimum requirements for project financial management as per OP/BP 10.02, with the financial management arrangements proposed by the PIE to be implemented together with a Financial Management Action Plan. The overall residual risk rating after mitigation measures was rated moderate. During implementation, independent external audits of project financial statements, in accordance with the International Standards on Auditing, were carried out regularly, and no qualified audits were produced. The project has been compliant with the World Bank requirements of submitting quarterly interim unaudited financial reports. Despite a few difficulties (e.g., problems with financial forecasting) which emerged in the early years and necessitated a temporary downgrade of the rating, the financial management performance at completion is rated Satisfactory. 70. Procurement. Procurement of goods, services and non-consulting services for the project was to be carried out in accordance with the World Bank Procurement Regulations for IPF Borrowers: Procurement in Investment 22 During field visits numerous gaps were identified such as non-functional incinerators, inadequate waste handling, storage, transportation and management capacity; lack of adequate environmental health personnel; and limited training and capacity building activities with patients and community members exposed to risks. 23 At the facility level the MOHCC utilizes suggestion boxes where patients can anonymously drop written grievances. The facility administration, including the HCCs and hospital boards, take these grievances seriously and try to solve them as they are recorded in minutes of meetings of the respective facility. Client satisfaction surveys are also used to identify grievances. A complementary part of the GRM works at the community level with patients presenting grievances to members of the HCCs. This community approach has been incorporated into national law in Zimbabwe. The community level also has an alternative GRM approach through community outreach by CBOs. Page 24 Project Financing Goods, Works, Non-Consulting and Consulting Services Guidelines (First published in July 2016, Second Edition in November 2017, Third Edition in July 2018, Fourth Edition in November 2020, Fifth Edition in September 2023). The Bank’s Anti-Corruption Guidelines were to be applied. Procurement of operational costs was in accordance with the PIE’s Field Office Manual which was reviewed by the World Bank and found to be adequate. A procurement capacity assessment of the PIE was carried out by the World Bank during appraisal. The procurement capacity of the PIE, including the organizational setup and the staff to be recruited was considered adequate and the risk was considered medium. During project implementation, staffing of the PIE was found to be insufficient leading to the hiring of an additional Procurement Officer to ensure proper documentation and accuracy of documents. Despite some occasional delays in procurement and a few activities carried outside the Systematic Tracking of Exchanges in Procurement system which resulted in a temporary downgrade to moderately satisfactory, overall procurement performance was rated satisfactory. The final rating for procurement is Satisfactory. C. BANK PERFORMANCE Quality at Entry 71. The project was highly relevant, and the design was innovative, serving as a flagship operation. The World Bank team mobilized US$15 million in grant funding and supported the Government to introduce an innovative strategy in a difficult country context. The results-based focus aimed to bolster accountability and rapidly increase coverage of critical maternal and child health services which had plummeted. Zimbabwe was in arrears and could not benefit from International Development Association financing, hence the World Bank’s effort to mobilize US$15 million to support the introduction of an innovative RBF program was greatly appreciated by multiple stakeholders. The project was well designed, incorporating lessons from Zimbabwe and other countries, and leveraging high-level commitment of key policymakers. The risks and mitigation measures were generally well articulated even though the overall ‘medium’ risk was probably excessively optimistic. The institutional arrangements were appropriate, leveraging the expertise of an international NGO with extensive experience in RBF. Zimbabwe also benefited from complementary technical assistance and policy advice from the World Bank and other development partners and from being part of a larger network of countries introducing RBF in the early 2010s, with multiple peer-to-peer learning opportunities to share lessons and experiences. Quality of Supervision 72. The World Bank team provided strong support throughout the 12.5-year implementation period. Implementation support missions were conducted regularly (typically twice a year) and were well-staffed with a good mix of technical and fiduciary and safeguards experts and held virtually during the COVID-19 pandemic. These missions systematically reported on overall progress, identified emerging challenges and opportunities, and proposed practical solutions. Progress was monitored regularly between missions by field-based staff who provided outstanding support, interacting regularly with development partners and providing hands-on support to resolve issues. The field-based staff served as the institutional memory behind the flagship project, with some engaged from inception to completion. The appointment of a GFF Liaison Officer further strengthened implementation support in the final years. One task team leader (TTL) managed the project for the first six years, another TTL oversaw implementation for the next five years, with the final one-plus years covered by a third TTL. 73. Project ratings were generally well-adjusted as implementation slipped or upgraded with improvements.24 The team was proactive in modifying the Results Framework indicators over time to reflect progress and emerging issues. The team successfully mobilized grant financing, providing the country five rounds of AF for over a decade. This was not always an easy task and required perseverance with the World Bank team mobilizing even modest 24 One small shortcoming was what appears as an inconsistency between the overall satisfactory rating on implementation which was maintained as satisfactory while two components were both rated as moderately satisfactory (ISR #12, 2017). Page 25 amounts of trust fund resources to keep the program on track (AF III and AF IV secured US$5 and US$3 million, respectively). 74. The World Bank provided parallel technical assistance. This included several evaluations and reviews to improve the cost-effectiveness of the RBF program (e.g., risk-based verification) and to explore strategies for supporting the Government to make better decisions about which services to provide at different levels of the health system, including: • The Modelling Combination of HIV Prevention Study (2018) applied the Optima Optimization Model to demonstrate how new HIV infections could be reduced by 89 percent by 2030 with a 25 percent increase in funding. • The Allocative Efficiency Study (2021) applying the Health Interventions Prioritization (HIP) tool provided evidence of the potential efficiency and health benefits of integrated care at the primary health level with increased allocations for maternal, child health, and NCD services. • The 2018 study on the Efficiency Gains of Integration of HIV and Sexual and Reproductive Health Services (SRHS) found that there was more integration during the study period (2013-16), particularly in terms of HIV interventions integrated into SRHS (e.g., ANC, family planning); service delivery became more efficient by reducing costs and reaping economies of scale, especially at DHs; and integration did not lead to a reduction in quality or the costs faced by clients. • A Fiscal Space Study was carried out in 2017 to assess the scope for improving efficiency in health spending. • An innovative machine-learning-based optimization tool was introduced in 2017 and used in subsequent price reviews to derive prices for interventions in the RBF incentivized package. 75. The World Bank team leveraged support from key stakeholders. The team worked closely with the Country Management Unit (CMU) to flag issues at the highest levels of Government. The CMU was instrumental in sending strong messages on the importance of continued commitment and co-financing at the level of Ministry of Finance and Economic Development, which resulted in the progressive increase in funding that also strengthened the case for including Zimbabwe as a GFF beneficiary. One TTL noted that the CMU was fully engaged from preparation to implementation, providing helpful advice on navigating implementation challenges. The World Bank team developed close working relationships with Government counterparts, Cordaid, and development partners, which proved key to resolving issues, institutionalizing RBF, and scaling it up to other districts. Justification of Overall Rating of Bank Performance 76. ’ is rated Satisfactory. The World Bank team prepared an innovative project in collaboration with the Government, mobilized grant financing over a 12-year period, and provided complementary technical assistance. D. RISK TO DEVELOPMENT OUTCOME 77. The main risk to sustaining gains made under the project is the continuous precarious macroeconomic and fiscal situation and unstable financing for the health sector in Zimbabwe. While the Government’s commitment to RBF remains strong, the ability to provide adequate, predictable resources in a timely manner remains a concern. With the transfer of the RBF program to the Government, the overall envelope has been reduced with some facilities struggling to make ends meet and finding it difficult to maintain the free user fee policy for exempted groups with risks of serving fewer clients and/or reintroducing fees. The devaluation of the local currency continues to erode purchasing power with working conditions not proving viable for health workers who continue to seek better economic opportunities abroad. The combination of the precarious country context and difficult sector conditions present important risks that need to be carefully managed going forward, Page 26 sustaining and consolidating the gains from the RBF program through reforms in health financing and service delivery. 78. Prospects for sustaining gains are enhanced by the institutionalization of RBF during the implementation phase. With the transfer of the RBF management to the MOHCC for the 18 front runner rural districts, there is greater capacity to manage the program at different levels of the health system. The governance structures are in place, and the tools are available to sustain the RBF program. Most importantly, there is strong ownership of the RBF approach which has enhanced accountability and autonomy. The National RBF Steering Committee played an important role in institutionalizing RBF as well as harmonizing the program supported under the project, supported by other development partners. V. LESSONS AND RECOMMENDATIONS 79. The introduction of RBF in Zimbabwe has generated important lessons that have been continually adapted for over a decade. Key lessons regarding the design, impact, and sustainability are summarized below, along with stakeholder views from the front lines (Box 2). • The Zimbabwe RBF program has shown that RBF can be effectively implemented and sustained at scale to generate rapid results, particularly for services which are starting from a low base and are within the control of providers. • The sequencing of reforms and innovations represents good practice with continual modifications to indicators, Box 2: Views on RBF from the front lines: targets, and incentives to maintain relevance and ensure -Ownership of the approach has made the biggest alignment to national strategies. difference • The RBF approach has promoted accountability and enhanced autonomy of front-line providers, incentivizing -Incentives were sometimes small but really helped and empowering them to find solutions and deliver improve the working environment for staff and results. patients • The strong results focus has contributed to enhanced -Urban voucher program was a game changer, planning, monitoring, and supervision as RBF requires resulting in direct benefits both to the target facilities to be proactive and managers to track progress beneficiaries and to health facilities and provide oversight. • The financial incentives served as a catalyst that -Teamwork is critical to success, need to manage motivated providers to find solutions. The RBF subsidies expectations about how incentives are distributed enabled providers to improve the functionality of health clinics and enhance staff motivation with cross cutting -Need to see RBF not only from a financing benefits for participating facilities. perspective but also from a health system angle • The introduction of risk-based verification contained costs and use of a bonus system provided additional incentives to facilities in remote areas. • The engagement of communities helped to improve service delivery and ensure the program met local needs while strengthening accountability. • The tracking of the impact of RBF on non-incentivized services is good practice and finding no major negative impacts was encouraging, as this could be a potential pitfall of these schemes. 80. The institutionalization of RBF in Zimbabwe is part of the legacy of the project with important lessons both for the country and the region. The process used by Zimbabwe to build capacity and gradually transfer the program from an international NGO to the Government is promising. The Government of Zimbabwe has shown sustained Page 27 commitment and ownership of the RBF approach which has been a critical factor contributing to the institutionalization. The sustained commitment translated into a substantial financial contribution to the project (US$53.3 million) despite the difficult macroeconomic situation. From inception the Government worked closely with all stakeholders to adapt the RBF to the local context and ensure broad-based ownership. The Government had the foresight to recruit an external agency with a long-standing track record of implementing similar programs. In subsequent years, a systematic process was adopted to train and mentor staff at all levels of the health system (e.g., health facility staff responsible for service delivery, management cadres responsible for institutionalization) and to establish governance structures (e.g., National Steering Committee, District Steering Committees). These structures ensured accountability and alignment to national policies. The institutionalization of RBF was carried out in phases, which allowed for gradual integration and adaptation of the program within the health system. The phased approach helped manage challenges and facilitated smoother transitions. While institutionalization encountered some difficulties (e.g., staff turnover, resistance, budget cuts), the Government demonstrated resilience and perseverance in assuming leadership of the program and providing co-financing despite the macroeconomic challenges. 81. The learning by doing approach was an integral aspect of the Zimbabwe RBF program with strong monitoring and evaluation underpinning the design and roll out. The Zimbabwe RBF program was accompanied by a strong emphasis on evaluating results and making course corrections, with both an IE and several process evaluations conducted over the 12.5-year period. Using a science of implementation approach was key to generating evidence to inform the scale up. Evaluations helped to dispel some concerns and created greater confidence in the RBF approach. The evaluations documented the relative performance of RBF districts, described shortcomings, and provided recommendations on areas requiring continual improvement. The learning by doing approach also resulted in the introduction of various innovations (e.g., Urban Voucher Program, CQI Initiative, Risk-based Verification, Electronic Partograph) which improved the effectiveness of the RBF program and strengthened health system capacity. 82. The sustained dialogue with development partners was key to expanding RBF to the remaining 42 districts . Both the Government and World Bank team established strong partnerships and maintained a sustained dialogue with partners through the National Steering Committee. Given the shared goal to improve maternal and child health and the need to use resources efficiently, a streamlined version of RBF was rolled out in the 42 remaining rural districts along with various RBF innovations and broader investments in health systems. 83. Looking forward, it is important to keep the RBF program on track as it has served the sector well in a ‘high risk, high reward’ context. To this end, there are several strategic priorities: (i) maintain strong commitment at the highest ministerial levels; (ii) harmonize the RBF approaches supported under the project with the one funded by other development partners to have ‘one’ harmonized national scheme while resuscitating the National RBF Steering Committee; (iii) mobilize additional external funding to support the continuation of the program, focusing on lagging areas (e.g., neonatal mortality, stunting) and prioritizing support for M&E to continue generating lessons; and (iv) address systemic challenges facing the health sector, including human resources and supply chain issues, by exploring innovative complementary strategies. Page 28 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS @#&OPS~Doctype~OPS^dynamics@icrresultframework#doctemplate A. RESULTS FRAMEWORK PDO Indicators by Outcomes Increase coverage of key RMNCAH-N services in participating districts Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Percentage of women who had their 10.00 Dec/2012 32.00 Mar/2024 22.00 Mar/2024 first ANC visit during the first 16 Comments on achieving targets Not Achieved weeks of pregnancy in participating rural districts (Percentage) Percentage of children 6 to 59 43.50 Dec/2019 50.00 Apr/2023 30.20 Mar/2024 months with vitamin A Comments on achieving targets Not Achieved supplementation in participating rural districts. (Percentage) People who have received 0.00 2,515,859.00 2,888,688.00 essential health, nutrition, and population (HNP) services - Female (RMS requirement) (Number) Number of children 0.00 Dec/2012 929,004.00 Mar/2024 1,218,431.00 Mar/2024 immunized (Number) Number of women and 0.00 Dec/2012 2,768,670.00 Mar/2024 3,329,251.00 Mar/2024 children who have received basic nutrition services (Number) Page 29 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Number of deliveries attended 0.00 Dec/2012 1,149,839.00 Mar/2024 1,358,721.00 Mar/2024 by skilled health personnel (Number) Increase quality of key MCH services in participating districts Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Percentage of participating district, 51.00 Dec/2019 60.00 Mar/2024 51.35 Mar/2024 provincial and central hospitals that Comments on achieving targets Not Achieved have registered an increase in quality scores since last quarter (Percentage) Strengthen institutional capacity to manage performance-based contracts Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Percentage of health facilities 0.00 Dec/2012 80.00 Mar/2024 72.50 Mar/2024 managed under RBF contracts by the Comments on achieving targets Substantially achieved MOHCC Program Coordination Unit in participating rural districts (Percentage) Strengthen COVID-19 response Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Percentage of close contacts of 0.00 Feb/2020 80.00 Mar/2024 82.00 Mar/2024 confirmed COVID-19 cases followed Comments on achieving targets Surpassed up based on national guidelines (Percentage) Intermediate Indicators by Components Increase quality of an integrated package of RMNCAH-N Services Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Page 30 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Percentage of 0.00 Sep/2015 50.00 Mar/2024 46.00 Mar/2024 health facilities implementing Continuous Quality Improvement (CQI) model in the participating rural districts. (Percentage) Number of District 0.00 Apr/2020 25.00 Mar/2024 25.00 Mar/2024 Health Executives Comments on Target was revised upwards at the last restructuring to 25 and this has been achieved. (DHEs) in achieving targets participating districts using updated quality checklist with the COVID-19 component for supervision of health facilities. (Number) Strengthen institution capacity to manage performance based contracts Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Number of health 0 Dec/2012 280 Mar/2024 302 Mar/2024 workers that Comments on Surpassed received first time achieving targets training or refresher training on EONC using skills labs (Number) 0 Dec/2012 100 Mar/2024 0 Mar/2024 Page 31 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Percentage of Comments on Delayed MOHCC cost achieving targets centers utilizing resource allocation formula in line with the whole Government Management Performance System (Percentage) Process evaluation No Dec/2012 Yes Mar/2024 Yes Mar/2024 to examine the Comments on Achieved institutionalization achieving targets of RBF in the government system as a viable approach to sustainability of RBF (Yes/No) COVID-19 Response Indicator Name Baseline Closing Period (Original) Closing Period (Current) Actual Achieved at Completion Result Month/Year Result Month/Year Result Month/Year Result Month/Year Percentage of 5.00 Jan/2020 95.00 Mar/2024 89.00 Mar/2024 health facilities Comments on The reference source, the Vital Medicines Availability and Health Services Survey, which is undertaken by the Ministry with support from partners, with PPE achieving targets has not been conducted since Q2 2022. commodities (Percentage) Number of 0.00 Feb/2020 1,680.00 Mar/2024 4,573.00 Mar/2024 affected people Comments on Target exceeded including health achieving targets workers reached with clinical and psycho-social Page 32 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT support (Number) Number of 0.00 Jan/2020 6.00 Mar/2024 6.00 Mar/2024 isolation centers Comments on Target of 6 isolation centers reached. complying with achieving targets infection prevention control (IPC) guidelines (Number) Number of COVID 0.00 Jan/2020 30.00 Mar/2024 2.00 Mar/2024 treatment centers Comments on Target not met. Initial plan was to support 20 hospitals with oxygen concentrators. MOHCC decided, instead, to support with oxygen achieving targets therapy with oxygen reticulation (a higher value investment per facility). Due to budget constraints only 2 facilitiesere supported (Number) with oxygen reticulation system. Page 33 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT B. KEY OUTPUTS Increase coverage of key RMNCAH-N services in participating districts PDO Indicators -Percentage of pregnant women receiving antenatal care during their visit to a health provider in participating rural districts -Percentage of births attended by skilled health personnel in a health institution in participating rural districts based on survey data -Percentage of women 15-49 receiving a modern family planning method in participating rural districts -Number of children immunized in participating rural districts -Percentage of women who had their first ANC visit during the first 16 weeks of pregnancy in participating rural districts -Percentage of children under 5 with diarrhea receiving ORT and Zinc in participating rural districts -Percentage of children 6 to 59 months with vitamin A supplementation in participating rural districts. Key Outputs -Cumulative number of pregnant women living with HIV who are initiated on antiretrovirals to (linked to the achievement of the PDO Outcome) reduce the risk of MTCT in participating rural districts -Cumulative number of pregnant women receiving first antenatal care during a visit to a health provider in participating urban districts -Cumulative number of pregnant women receiving first antenatal care before 16 weeks of gestation period during a visit to a health provider in participating urban districts -Number of children immunized -Number of women and children who have received basic nutrition services -Number of deliveries attended by skilled health personnel (in health institutions in participating rural and urban districts -Cumulative number of health facilities enrolled in RBF program in participating districts -People who have received essential health, nutrition, and population services -Cumulative number of children <5 who had their Mid Upper Arm Circumference and Height measured by CHWs as part of growth monitoring -Percentage of 15-49 year old women who are new acceptors of long term, reversible FP methods in participating districts Page 34 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Increase quality of key MCH services in participating districts PDO Indicators -Average quality scores by health facilities in participating rural and urban districts -Percentage of maternal deaths given audits as per protocol in participating districts -Percentage of partographs correctly filled in participating districts -Percentage of children under 5 years with Pneumonia correctly managed in the participating districts -Percentage of participating district, provincial and central hospitals that have registered an increase in quality scores since last quarter Key Outputs -Percentage of health facilities implementing Continous Quality Improvement model in the (linked to the achievement of the PDO Outcome) participating rural districts -Percentage of RBF contracted facilities in Continous Quality Improvement (CQI) districts with CQI Standard Operating Procedires -Number of District Health Executives in participating districts using quality tool for supervision of health facilities Strengthen institutional capacity in RBF Contract Management PDO Indicators -Percentage of health facilities managed under RBF contracts by the MOHCC Program Coordinating Unit in participating rural districts Key Outputs -Number of health personnel and CHWs receiving training on RBF in participating districts (linked to the achievement of the PDO Outcome) -Number of health workers that received first time training or refresher training on EONC using skills labs -Percentage of MOHCC cost centers utilizing resource allocation formula in line with the whole Government Management Performance System -Process evaluation to examine the institutionalization of RBF in the government system as a viable approach to sustainability of RBF -Process evaluation to examine the effectiveness of quality focused RBF at provincial hospitals in improving the quality of maternal and neonatal care for improved outcomes Strengthen COVID-19 response PDO Indicators -Percentage of close contacts of confirmed COVID-19 cases followed up based on national guidelines Key Outputs -Percentage of health facilities with PPE commodities (linked to the achievement of the PDO Outcome) -Number of affected people including health workers reached with clinical and pscyho-social support -Number of isolation centers complying with IPC guidelines -Number of COVID-10 treatment centers with oxygen therapy Page 35 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Fernando Montenegro Torres Team Leader Carol Atieno Obure Team Leader Baison Banda Financial Management Specialist Eliot Kalinda Procurement Specialist Mercy Chimpokosera-Mseu Environmental Specialist Kudakwashe Dube Social Specialist Blessing Karadzandima Procurement Team Kutemba Chilila Kambole Procurement Team Patron Titsha Mafaune Team Member Yvette M. Atkins Team Member Priscilla Netsai Mutikani Team Member Sandra M Kuwaza Team Member Chenjerai N. Sisimayi Team Member @#&OPS~Doctype~OPS^dynamics@icrannexstafftime#doctemplate B. STAFF TIME & COST Staff Time & Cost Stage of Project Cycle No. of Staff Weeks US$ (including travel and consultant costs) Preparation FY11 6.352 71,019.00 FY12 6.600 17,758.11 FY20 0.000 1,045.00 Page 36 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Total 12.95 89,822.11 Supervision/ICR FY12 48.925 278,418.32 FY13 57.637 264,652.34 FY14 32.073 280,791.57 FY15 51.288 386,875.54 FY16 42.846 313,081.87 FY17 37.753 273,946.48 FY18 23.660 223,723.50 FY19 20.684 168,211.35 FY20 22.390 277,731.55 Total 337.26 2,467,432.52 Page 37 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 3. PROJECT COST BY COMPONENT Component Amount at Approval (US$M) Actual at Project Closing (US$M) Results Based Financing in Delivery of Packages of Key Maternal, Child, 10.9 40.3 and Other Related Health Services (RMNCAH-N) Management and Capacity Building 3.9 28.7 Monitoring, Documentation, and Verification of Results under 0.2 4.1 Performance-based Contracts COVID-19 Response 5.0 Page 38 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 4. EFFICIENCY ANALYSIS 1. Project cost. Total grant disbursements for the project up to the closing of AF V was US$78 million. The original grant and five subsequent AFs were all fully disbursed. The NPV of the total cost is US$55.6 million, assuming a 10 percent discount rate. 2. Project impact. Following the theory of change, the estimated project impact is such that over 12 years, 38,259 children’s lives and 2,384 women’s lives would be saved (see Table 4.1 below). The economic analysis used the total population of the 18 rural districts plus that of the cities of Bulawayo and Harare as a proxy for project beneficiaries. Assumptions taken in estimating the project impact are detailed in Box 1 below. Table 4.1. Estimated Project Impact Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Year 11 Year 12 Year 13 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 Total Disbursed: $US (M) 15 20 10 5 3 25 $ 78 Benefits # of maternal deaths averted 186 187 199 207 218 221 225 226 227 242 247 2,384 # of newborn deaths averted 779 769 823 847 886 957 958 971 999 1,097 1,136 10,222 # of children <5 deaths averted - - 1,086 1,412 1,727 2,042 2,963 2,956 3,405 3,784 4,147 4,516 28,037 - Total number of maternal deaths averted 186 187 199 207 218 221 225 226 227 242 247 2,384 Total number of children's deaths averted (newborn to <5) 779 1,855 2,234 2,574 2,928 3,920 3,914 4,376 4,783 5,243 5,652 38,259 Box 4.1: Assumptions for estimating project impact The impact of the project was modeled using the data that were available on coverage of key interventions and peer-reviewed literature on the impact of those interventions, combined with data on population over the lifecycle of the project. The analysis only examined the impact of select interventions known to impact on maternal, child and neonatal mortality, but did not include the estimated impact on overall mortality or disability due to COVID-19 interventions. Thus, the benefit-cost ratio is expected to be underestimated. Estimating the number of maternal deaths averted: This number was estimated because of the following indicators: (i) pregnant women receiving ANC, (ii) births attended by skilled health personnel, (iii) use of modern contraceptive methods among women 15-49 years old, and the (iv) average quality scores by health facilities. Jowett (2000) estimates that in low-income countries, 26 percent of maternal deaths are avoidable through antenatal/community-based interventions and access to quality essential obstetric care can prevent a further 48 percent of maternal deaths. Chou et al. (2018) also found that high quality services in health systems would result in a 28 percent decrease in maternal deaths. Ahmed et al. (2012) found that the total impact of increased contraceptive use (i.e., through spacing births or reducing the number of pregnancies and thus deliveries and unsafe abortions) reduces the maternal death by 44 percent. Estimating the number of neonatal deaths averted: We estimated this number because of pregnant women receiving ANC, births attended by skilled health personnel, the average quality scores by health facilities, the percentage of health facilities managed under RBF (all of which impact the number of deaths), but it also modeled the impact of contraceptive use (which reduces the number of births). Based on Tekelab et al. (2019), “Utilization 39 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT of at least one ANC visit by a skilled provider during pregnancy reduces the risk of neonatal mortality by 39 percent in sub-Saharan African countries." In relation to births attended by skilled health personnel, according to Amouzou et al. (2017) also focused on sub-Saharan Africa, “After the first day of life, newborns delivered with a skilled attendant at birth were 16 percent less likely to die within 2–27 days than those without a skilled attendant at birth.” Furthermore, according to Chou et al. (2018), high-quality systems in low-and-middle-income countries delivering services to mothers and newborns would result in an estimated decrease of 28 percent in neonatal deaths compared to a situation without any improvement in the delivery of services. Using anti-corruption audit interventions as a proxy measure for institutional reforms, Nichter et al. (2020) also found reduced neonatal mortality by 8.1 percent in the non-white population, comparable to the populations that this Project served. This value was used as a proxy for the impact of RBF. Estimating the number of deaths averted, among children under 5: The number of deaths averted in this age group was estimated based on the effects of the following indicators: (i) proper management of pneumonia and diarrhea (with the use of ORT and Zinc) among children under 5, and (ii) health facilities managed under RBF. The study by Nichter et al. (2020), where we used anti-corruption audit interventions as a proxy measure for institutional reforms, found a reduction of child mortality of 9.2 percent among the non-white population. In relation to proper diarrhea management particularly using ORT and Zinc, Munos et al. (2010) found that oral rehydration salts prevent up to 93 percent of childhood diarrhea deaths. In a study of developing countries, Theodoratou et al. (2010) estimated that community case management of pneumonia could result in a 70 percent reduction in mortality from pneumonia among children between 0-5. Unlike the Interim ICR, this final ICR did not estimate the number of cases of stunting prevented, as ANC is just one of many factors contributing to stunting. Thus, the benefit-cost ratio is likely to underestimate the impact of the project. 3. Return on investment. The NPV of total project benefits is estimated at US$134.7 at a 10 percent discount rate, and the net present value of costs is estimated at US$55.6 million. This yields a benefit-cost ratio of 2.42 for every dollar invested and an overall NPV return of US$79.1 M, at a 10 percent discount rate. The internal rate of return was estimated to be 14 percent. 40 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ’ I. BACKGROUND 1. In the 1990s, Zimbabwe boasted of the best primary health care services in Sub-Saharan Africa. However, protracted economic challenges from 2000 to 2008 led to a significant decline in funding for the health sector. This period saw a severe brain drain, lack of resources, and a near-collapse of the health system, with primary level facilities charging service fees. Consequently, access to and utilization of services plummeted25. The Maternal Mortality Ratio surged from 555 deaths per 100,000 live births in 2005/06 to 960 deaths per 100,000 live births in 2010/1126. Similarly, the Under-5 Mortality Rate increased from 82 deaths per 1,000 live births to 84 deaths per 1,000 live births in the same period. Health worker motivation was low, with salaries far below the poverty line, exacerbating the challenges. In response, the Government of Zimbabwe (GoZ) mobilized resources to revitalize the health sector, focusing on Maternal, Newborn, and Child Health (MNCH) services. Recognizing the critical need for external support, the GoZ successfully secured the World Bank (WB)'s interest in a health investment case. This collaboration led to the introduction of the Health Sector Development Support Project (HSDSP) in 2011 to improve maternal and child health indicators by addressing the poor health indicators that had stagnated or declined during the years of economic crisis. The intervention tackled various demographic and public health challenges, particularly maternal and child morbidity and mortality. HSDSP Objectives 2. The Project Development Objective was to increase coverage and quality of an integrated package of Reproductive, Maternal, Neonatal, Child, Adolescent Health and Nutrition (RMNCAH&N) services and to strengthen institutional capacity to manage performance-based contracts consistent with the recipient’s ongoing health initiatives. During the COVID-19 outbreak, an additional objective was included (i.e., to strengthen the national COVID-19 response). The following were the specific objectives/activities: • To improve the availability, accessibility and quality of health services with a focus on integrated key RMNCAH&N, NCDs, TB, HIV and Malaria services and their optimal utilization • To increase utilization and strengthen supply of quality MNCH services in low-income urban health districts • To increase the coverage of RMNCAH&N services at community and primary care levels in the 4 districts implementing the Extended Supply Side Community Results Based Financing (ESScRBF) initiative) • To improve outcomes of Obstetric and Newborn complications by enhancing management of such as per national guidelines at tertiary and quaternary levels of care through the Quality Focused RBF (QFRBF) initiative • To strengthen the institutional capacity of the Ministry of Health and Child Care (MoHCC) to respond to the COVID- 19 pandemic 25 Ministry of Health and Child Care, 2011 26 ZDHS 2005/6 &2010/11 41 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT II. INTERVENTIONS DONE UNDER HSDSP Rural Results Based Financing 3. The RBF initiative in Zimbabwe began in 2011 under the HSDSP in two rural districts: Marondera and Zvishavane, to reduce urban-rural disparities in RMNCAH&N outcomes. In 2012, the RBF project expanded to 16 more rural districts with support from the World Bank and Cordaid as the implementing entity. Guided by the Programme Implementation Manual (PIM), revised in 2014, 2017, and 2019, the initiative demonstrated significant success, as shown by findings from observational studies and some published secondary quantitative and qualitative studies, leading the GoZ to extend it to all 42 remaining rural districts with financial support from the Health Transition Fund which transitioned to the Health Development Fund (HDF) and later became the Health Transition Fund and the World Bank's phased support enabled the introduction of innovative approaches, while the GoZ gradually increased its own funding as World Bank contributions declined. 4. The institutionalization of RBF in Zimbabwe was recommended in 2015 through the Medium-Term Strategic Framework, which emphasized governance, funding adaptations, and verification methods to enhance local ownership. This led to three generations of RBF implementation: the first focused on increasing health service utilization, particularly for maternal and child health, by providing financial incentives to health facilities. The second generation emphasized service quality, introducing performance-based contracts and engaging local governance structures. The third generation aimed at health system sustainability by strengthening domestic financing mechanisms, reducing dependency on external funds, and integrating RBF into broader health financing strategies. This phase also prioritized capacity building at all health system levels and improved health information systems for monitoring and evaluation. 5. The success of Zimbabwe's RBF model attracted interest from other health partners, leading to partnerships for targeted health interventions. World Education supported pediatric HIV interventions in Matabeleland South (2016-2017), the Clinton Health Access Initiative (CHAI) funded TB-related indicators in Manicaland (2019), and Population Solutions for Health (PSH) currently uses RBF to enhance Voluntary Medical Male Circumcision (VMMC) programs. These collaborations have effectively addressed critical health challenges and improved service delivery in targeted areas by leveraging the RBF approach. Institutionalization of RBF 6. Institutionalization efforts began in 2017, with the MoHCC assuming the purchasing function within the Programme Coordination Unit (PCU), reducing administrative costs from 18 to 4.5 percent and shifting Cordaid's role to technical support. By 2018, the GoZ began rolling out RBF institutionalization in 18 rural districts previously managed by Cordaid and later in 2022 in the 42 districts supported by HDF as outlined in Figure 5.1 below. 7. The Government of Zimbabwe has increased domestic funding for RBF institutionalization to reduce reliance on external donors. However, frequent currency changes and economic instability have severely impacted timely disbursement of RBF subsidies and procurement processes. Over the past decade, Zimbabwe transitioned from a multi-currency system (2015-2018) to the reintroduction of the Zimbabwean dollar (2019), which rapidly devalued amid inflation. Despite interventions like the Foreign Exchange Auction System (2020) and the temporary return of US dollar transactions (2023), the local currency continued to lose value, leading to the introduction of the Zimbabwe Gold (ZiG) currency in 2024. These financial challenges have hindered the predictability and sustainability of RBF funding. 42 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT FIGURE 5.1: ROADMAP TO INSTITUTIONALIZATION Innovations Complementing Rural RBF 8. During implementation, the RBF program brought about new innovations that aimed to strengthen health systems, as outlined below. • Risk Based Verification (RBV) i : This approach was initiated during the 2nd phase of RBF in 2014 after recommendations had been made to the project to reduce verification costs, (i.e., staff and operational costs), and in the process create sustainability for the project. The adoption of RBV was noted to indeed reduce costs without increasing the risk of over-reporting and subsequent related overpayments. The quality of data was not compromised by the change in verification model 27 . The overall costs for verification declined over time and resources saved were channeled towards new innovations in RBF. • Electronic verification: Data collection for both Quality Support and Supervision, invoicing, and the Quantity Verification 28 previously used paper-based tools until the end of 2016 when an electronic system was developed using an open-source system called ODK (Open Data Kit) and linked to MoHCC DHIS2. This enabled the generation of electronic invoices and visualizing performance data. • Electronic Partograph (ePartograph): Cordaid, through its own funding, supported the MoHCC in transitioning from a paper-based partograph to an electronic version (ePartograph). This digital tool was integrated into the Labor & Delivery module of the EHR system. The partograph is essential for clinicians to monitor the progress of labor in pregnant women, enabling timely interventions to improve maternal and neonatal outcomes. Due to the unreliability of power in facilities the paper-based is also still in use. • Incentive Calculator: The Incentive Calculator was modified from an Excel-based manual to an electronic solution that provides a convenient and equitable way to calculate staff incentives of a particular health facility29. Managers at a health facility feed the required data, which includes the post, number of years worked, responsibility rate, 27 Cordaid, 2014 28 Ministry of Health and Child Care, 2019 29 MoHCC, 2017 43 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT hours worked etc. for all the facility staff quarterly. The Incentive Calculator automatically generates amounts each cadre will receive using a predefined algorithm. The process ensured transparency and helped curtail conflicts at different care levels. • Continuous Quality Improvement (CQI): In response to the 2014 RBF program impact evaluation, the MoHCC, with World Bank support, piloted a CQI initiative to enhance clinical process measures that had initially shown limited responsiveness to RBF. Implemented in eight randomly selected districts, the CQI pilot adopted a quasi- experimental design, leading to notable improvements in tracked indicators. World Bank-led process monitoring and evaluation (PME) in 2018 and 2019 highlighted CQI as a key initiative for improving maternal, newborn, and child health (MNCH) services in Zimbabwe. Stakeholders widely acknowledged its role in fostering a culture of quality improvement, local problem-solving, and corrective action planning. While CQI has contributed to better working environments and enhanced routine care practices, it is viewed as a component of a broader health systems quality improvement strategy. • Data driven pricing of indicators: Zimbabwe has implemented data-driven pricing of indicators in RBF programs, systematically using data to determine the monetary value assigned to performance indicators to ensure fair, efficient, and objective-aligned payments to healthcare providers. Key data sources included historical health data, utilization rates, and demographic trends, all of which inform pricing decisions. This approach has also been adopted by Liberia and Mauritania in their Performance-Based Financing (PBF) pricing models. Performance of Rural RBF Indicators Quantity performance 9. The analysis of trends in this completion report focused on those indicators that have maintained presence on the indicator package throughout the implementation period. Table 5.1 below shows that there was an overall increase in performance on antenatal care (ANC), institutional deliveries and uptake of long-term family planning methods between 2012 and 2019. However, these indicators declined during and post the COVID-19 period. TABLE 5.1: PERFORMANCE OF RURAL RBF INDICATORS IN THE 60 DISTRICTS Indicator Baseline (2012) 2019 2020 2021 2022 Progress by 2023 Antenatal care 11% 38% 36% 35% 31% 29% before 16 weeks gestation Institutional 69% 81% 79% 81% 76% 74% deliveries Long term 1% 21% 19% 17% 15% 14% family planning Quality performance 10. Quality of Care in the MoHCC is assessed routinely every quarter using the quality supportive and supervision (QSS) checklist (for rural and urban health centres and district hospitals) and client satisfaction survey (CSS) tools. Overall, the performance showed an increase in overall quality of care from 77 percent in 2014 to 83 percent in 2020. The rate of increase then slowed down between 2021 (83 percent) and 2023 (84 percent). a) Client satisfaction (perceived quality of care): From 2014 to 2019, CSS scores remained consistently above 80 percent, ranging from 81 to 86 percent. This high level of satisfaction was largely attributed to improvements in staff attitudes, better retention due to RBF, and reduced waiting times. 44 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT 'RBF improved staff attitude, they no longer shouted at patients, and they no longer went to other clinics as service delivery had improved at our own clinic [Mutema Clinic CBO]. However, after 2019, CSS scores declined to an average 77 percent, reflecting growing dissatisfaction. This drop was primarily driven by delays in subsidy payments, which led to shortages of essential medicines and negatively impacted health services. Additionally, the dormancy of many CBO groups that conducted the surveys, along with data quality and network challenges, contributed to inconsistencies in reporting. While the inactivity of some CBOs may have affected data collection, the primary factor behind declining client satisfaction was the inability to access necessary medications due to funding delays. b) Service level quality of care: RBF has significantly improved both hospital and clinic infrastructure, enhancing the overall appearance of health facilities through equipment procurement, signage installation, and ward refurbishments. Additionally, there have been notable advancements in management and planning; however, supervision reports highlight the ongoing need for continuous support in handling financial documents. Despite these improvements, challenges persist, particularly in documentation, reliance on improvised registers, and poor filing systems, which continue to affect clinical domain scores. Nonetheless, scores in the service level domain (structural, management and planning and the clinical components) have shown progress, increasing from 75 percent in 2014 to 85 percent in 2023. Key achievements for Rural RBF a) Improved Infrastructure and Equipment: RBF implementation has led to significant improvements in healthcare infrastructure and equipment, including the installation of incinerators, construction of fences, and provision of Maternity Waiting Homes at selected high volume primary level facilities. b) Teamwork: RBF has fostered improved communication and teamwork among healthcare teams. Decision-making processes have been decentralized, ensuring that all team members and staff at the facility are involved. There was improved adherence to standards of care and use of data for improving service quality. c) Improved Client Satisfaction: There was an improved patient and nurse relationship, and the Community was more supportive of the health program. This further boosted the need for facility staff to improve their attitude and treat patients fairly as it affected their overall quality score. The client satisfaction further decongested the district, provincial and tertiary hospitals as clients no longer had self-referrals for conditions that were not emergencies. d) Increased Community Participation and Awareness: RBF has successfully promoted community participation and increased their awareness of healthcare services. Communities now perceive healthcare facilities as their own, resulting in a higher level of ownership and engagement. This heightened community involvement has positively influenced health-seeking behaviour and contributed to improved health outcomes. e) Enhanced Service Availability and Utilization: Healthcare facilities implementing RBF have improved on service availability and the communities they serve have subsequently improved on utilization of the services. Reductions in home deliveries, particularly within religious sects, have been observed, indicating increased trust in facility- based deliveries. f) Increased autonomy for procurement: The financial benefits of RBF implementation have been notable. Facilities have been able to purchase necessary items based on their specific needs, as opposed to centralized procurement. The availability of subsidies and incentives has facilitated the procurement of critical equipment and essential supplies including medicines and sundries. Additionally, the use of the USD currency during the initial stages of RBF implementation simplified financial processes, enabling facilities to manage funds more effectively. 45 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT III. ADDITIONAL INTERVENTIONS UNDER HSDSP The Urban Voucher (UV) Program 11. In 2014, the UV program was designed as complementary to the rural RBF Program to ensure the urban poor where not left behind. The program aimed to increase access to essential healthcare services and to ensure health equity for poor pregnant women in the cities of Harare and Bulawayo, as a way of reducing preventable maternal and perinatal mortality and morbidity. The program provided foundational steps for the targeting of beneficiaries as vouchers were provided to eligible pregnant women, enabling them to access free antenatal, delivery, and postnatal care services at designated clinics and hospitals. The following were the key achievements: a) Increased Access to Services: Vouchers increased access to health services. ANC bookings increased by 37 percent whilst institutional deliveries had a 23 percent increase between 2021 and 2022 after the full scale up across the cities of Bulawayo and Harare. During implementation, perinatal death outcomes decreased compared to the period before 2014. b) Improved Governance and Leadership: Through streamlining management structures and the separation of functions, District Health Executive (DHE)s became more involved in ensuring efficient and effective planning and execution of programs. The clinic as a management unit became more autonomous in operational planning and managing clinic resources. c) Improved Community Participation: With the formation and operationalization of Health Centre Committees, there was an increase in the community voice in the management of clinics' affairs and in managing the clinic- community interface. The appointment of community groups to verify client satisfaction further improved the relationship between the clinics and the patients. d) Enhanced Service Quality: Through supervision strengthening by Provincial Health Executives (PHEs) and DHEs as well as systematic execution of the Quality Checklist, there have been noticeable improvements in quality of services provided at health institutions. Health workers were also mentored on identifying and managing obstetric and neonatal complications. e) Improved Motivation: The performance-based incentive structure for service providers has improved motivation among health care staff, thereby also increasing some form of accountability in the system. Results Based Financing QFRBF and CRBF Initiatives 12. Under the HSDSP AFV, running from 2020 to 2024, two innovative RBF strategies, the Extended Supply side Community Results-Based financing (ESScRBF) and the Quality Focused Results-Based Financing (QFRBF), were introduced in 2022 in selected districts and referral facilities, to improve the coverage of selected RMNCAH&N indicators and strengthen the provision of quality of care at Provincial and selected Central Hospitals in the country. Extended Supply Side Community RBF Initiative (ESScRBF) 13. The ESScRBF initiative aimed to enhance the performance of RMNCAH&N indicators, particularly in Family Planning, ANC, PNC, Child Health, and Nutrition. These indicators had either stagnated or were underperforming. 46 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT The initiative also sought to increase demand and utilization of RMNCAH&N services at the community and primary care levels by introducing performance-based incentives for Village Health Workers (VHWs) in Mutare Rural, Centenary, Mangwe, and Gokwe South districts. Implemented between February and December 2023, the initiative engaged 1,527 VHWs working with 100 health facilities across the four districts. Key activities included training and sensitization of 156 health workers, 1,673 VHWs, and 1,028 community leaders, procurement and distribution of essential medical supplies and equipment (e.g., pregnancy test kits, satchels, and referral slip), development of standardized data collection tools, training of 84 nurses on long-acting reversible contraceptive methods (intra-uterine contraceptive devices and implants), and verification and incentivization of VHW and facility performance. 14. The initiative resulted in significant improvements in quantity and quality of health services. It enhanced collaboration and teamwork among VHWs, boosted morale and job satisfaction due to incentives received, and strengthened data collection and reporting on health services. A VHW at one of the health facilities in Mangwe had this to say: ‘We are 13 here and each one of us has received US$152 for the work that we did last quarter. I have no words to describe how much joy we have and how thankful we are to you. We have been energised and are raring to go as a team.’ VHW after receiving subsidies for Q2 2023. 15. Additionally, there was better supervision and monitoring of VHWs by nurses at health facilities. Most importantly, the initiative contributed to increased demand and utilization of RMNCAH&N services, with early ANC bookings rising from 26 percent in 2022 to 29 percent in 2023 and Postnatal Care (PNC) coverage on days 3 and 7 improving from 40.9 percent and 39.4 percent in 2022 to 47.4 percent and 47.3 percent in 2023, respectively. These achievements demonstrate the positive impact of performance-based incentives in strengthening community health systems and improving maternal and child health services Quality Focused RBF Initiative (QRBF) 16. The implementation of this initiative began as a pilot in November 2022 at two central hospitals (Sally Mugabe and Mpilo), as well as at 4 provincial hospitals (Gweru, Gwanda, Marondera, and Victoria Chitepo). In August 2023, it was expanded to the remaining four Provincial Hospitals (Masvingo, Bindura, St. Luke’s Mission, and Chinhoyi) and Chitungwiza Central Hospital. The initiative aimed to enhance the quality of care for pregnant and lactating mothers and their newborns at tertiary and quaternary levels. Key interventions included the revitalization of Emergency Obstetric and Neonatal Care (EmONC) skills laboratories, training healthcare workers in EmONC and Results-Based Financing (RBF), procuring and distributing essential Maternal and Newborn Health (MNH) equipment, and incentivizing health workers in maternity and other related departments through performance- based subsidies. 17. Key achievements of the initiative included improved clinical management of obstetric and neonatal emergencies, as seen in hospitals like Gwanda, where clinical management scores improved from 38 percent in Q4 2022 to 77 percent in Q3 2023, and Mpilo, where scores rose from 40 percent in Q4 2022 to 68 percent in Q4 2023. This means there were increases in the proportion of women and newborn babies with complications who were managed properly (per guideline) and those for whom emergency caesarean sections were performed within a decision to cutting time interval of ≤ 30 minutes. 47 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT 18. Facilities such as Sally Mugabe and Marondera also recorded notable progress in newborn care and postnatal health checks. Additionally, the initiative boosted staff morale and motivation by using subsidies to incentivize maternity department personnel. It also enhanced the work environment, with hospitals utilizing subsidies to procure essential MNH equipment and improve facility infrastructure—for example, Gwanda procured new delivery beds and installed visible signage, while Mpilo acquired surgical sundries and improved maternity hospital signage. The documentation and record-keeping system was revitalized, ensuring timely retrieval of patient notes through incentives for records personnel. Furthermore, hospitals procured critical medical equipment, including operating theatre tables, digital patient monitors, anesthetic machines, incubators, and defibrillators for eight provincial and three central hospitals. A health manager at one of the provincial hospitals had this to say: ‘These subsidies have helped us a lot. I am now working with motivated staff with a changed attitude towards work. We have also used part of the money to procure essential supplies for our pregnant and lactating women and newborn babies’ Matron-Provincial Hospital 19. Despite these achievements, the short implementation period of the initiatives—10 months for ESScRBF and 12 months for QFRBF—may have limited their full impact on RMNCAH coverage indicators. To sustain and build upon these gains, there is a need to secure additional domestic and external funding to extend implementation time, prepare for institutionalization to ensure long-term sustainability, and scale up the initiatives to other districts and secondary-level facilities. However, there are some long-term benefits realized through the initiatives such as essential equipment procured, skills and knowledge gained through the training and monitoring and evaluation tools developed and used during implementation will continue beyond the implementation period. IV. ENVIRONMENTAL AND SOCIAL SAFEGUARDS Environmental Safeguards 20. The civil works for COVID-19 isolation centres, including minor renovations, water tank installations, and oxygen reticulation systems, were categorized as Category B due to their minimal environmental impact. To mitigate these impacts, Environmental and Social Management Plans (ESMPs) were implemented under the Environmental and Social Management Framework (ESMF). Additionally, the increase in medical waste from Reproductive, Maternal, Newborn, Child, and Adolescent Health and Nutrition (RMNCAH&N) services led to the development of an Infection Control and Waste Management Plan (ICWMP) in November 2022. Key activities implemented under these frameworks included stakeholder consultations, monitoring, capacity building, and training of healthcare workers. 21. As part of the stakeholder engagement process, consultations were conducted at provincial, district, and public health facility levels to gather inputs for the development of the ESMF and ICWMP. A validation workshop was held to ensure alignment with MoHCC and stakeholder expectations. Additionally, HSDSP project staff, MoHCC engineers, and Ministry of Local Government and Public Works (MoLGPW) representatives conducted field visits to COVID-19 isolation centres before and during construction. These visits focused on ESMP implementation, occupational health and safety, personal hygiene, COVID-19 risk management, and environmental monitoring. Monitoring activities were carried out at six health facilities undergoing minor renovations and water tank installations, as well as two facilities with oxygen reticulation systems. Teams from MoHCC, MoLGPW, and Cordaid assessed potential safeguard issues and implemented corrective measures as necessary. To strengthen contractor compliance, capacity-building sessions were conducted for contractors, site supervisors, and health workers on environmental monitoring, social safeguards, occupational health and safety, and COVID-19 prevention. 48 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Healthcare workers also received refresher orientations to enhance their understanding of ESMP implementation. In parallel, 450 health workers from all eight rural provinces were trained on safe medical waste management, focusing on risk assessment, infection control, and proper waste disposal. 22. For sustainability, provincial and district teams from MoHCC, EMA, and the Public Works Department were fully capacitated to oversee ESMP implementation. Moving forward, MoHCC plans to integrate environmental and social reviews into overall program planning, implementation, and monitoring to maximize environmental and social benefits. Social Safeguards 23. The introduction of social safeguards under the HSDSP AFV aimed to minimize social and environmental risks, ensuring that project benefits reached all societal segments, particularly marginalized groups. These safeguards promoted stakeholder engagement, including the active participation of women and indigenous communities in decision-making. Additionally, they helped identify and mitigate potential negative impacts, ensuring that health outcomes remained equitable and sustainable. To enhance client engagement and feedback, various tools and approaches were utilized, including Client Satisfaction Survey Reports, Community-Based Organization Summary Reports, Suggestion Books, Complaints Books, Management Books, Suggestion Boxes, Focus Group Discussions (FGDs), Cordaid Landline, WhatsApp platforms, Twitter, and Facility Telephone Numbers. These mechanisms provided patients with multiple avenues to express concerns. As a result, the resolution rate of complaints within four weeks improved from 46.7 percent to 100 percent during the project period. The most common complaints related to medicine and sundry shortages, unpaid RBF funds, long waiting times, and staff attitudes and behaviour. Other concerns included short staffing, hygiene and cleanliness, maternity waiting home infrastructure, security, water shortages, delays at facilities, privacy and confidentiality, and long waiting times for vulnerable beneficiaries. By the end of the project, complaints about staff attitudes and behaviour had significantly decreased, and healthcare managers reported a 10 percent reduction in staff absenteeism. 24. To ensure sustainability, the project strengthened the Grievance Redress Mechanism (GRM) across all MoHCC levels by establishing GRM Focal Points for ongoing capacity building and awareness-raising. Additionally, GRM Focal Champions were identified and trained, and a GRM Working Group was formed to oversee continuous improvements. However, inconsistent complaint documentation and follow-ip remained a challenge. This issue is being addressed through Standard Operating Procedures (SoPs), capacity building, digitalization and continuous support and supervision. To enhance efficiency, there is exploration of a centralized and digitalized system to track GRM issues and facilitate MoHCC-wide complaint management. V. RISK COMMUNICATION AND COMMUNITY ENGAGEMENT 25. The project introduced Risk Communication and Community Engagement (RCCE) in 2020 to strengthen communication and community engagement efforts in line with its objective of increasing access to and utilization of RMNCAH&N services. Working primarily with the Health Promotion Unit and Public Relations departments across all levels of the Ministry of Health and Child Care (MoHCC) structures, as well as within the City Health Directorate for the Urban Voucher (UV) program, several key achievements were realized. 26. One of the major accomplishments was the revision of the RCCE strategy, which involved reviewing risk communication, community engagement, and COVID-19 demand creation strategies in collaboration with Health Promotion Officers (HPOs) at the provincial and district levels, alongside national-level staff and development partners. The revised strategy was designed to be adaptable for any future disease outbreaks. Additionally, the 49 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT project focused on capacity building of health communicators in sign language, equipping HPOs, public relations officers, environmental health practitioners, and community health nurses in Harare, Chitungwiza, and Bulawayo with essential skills to enhance health education and information dissemination. 27. To strengthen documentation and knowledge sharing, the project introduced training on the Most Significant Change (MSC) storytelling methodology for HPOs and other MoHCC staff. As a result, program successes were documented and shared through the project newsletter. Media engagement initiatives were also implemented to enhance project visibility and improve health reporting. This included media sensitization workshops with 49 health journalists from print, broadcast, and online platforms, as well as media health tours in the Northern and Southern regions, allowing journalists to engage directly with health service providers and beneficiaries. Furthermore, the project established a website (www.healthprojectzim.org.zw) and social media platforms to facilitate regular updates on project implementation. 28. The project also supported the development and distribution of IEC materials on topics such as MNH, EPI, and COVID-19. These materials played a crucial role in addressing misinformation, promoting correct health knowledge, and encouraging early health-seeking behaviours in communities. To ensure the sustainability of RCCE efforts, the Health Promotion Unit received capacity-building training in documentation and qualitative research methods. Additionally, the unit benefited from procurement of essential equipment, including laptops, cameras, hailers, and public address systems, which are vital for effective communication and documentation. 29. Despite these achievements, the implementation of RCCE activities faced challenges. The COVID-19 lockdowns significantly disrupted community engagement efforts, as public gatherings were restricted, limiting platforms for information dissemination. Additionally, underfunding hindered the implementation of critical community engagement interventions. Given RCCE's importance in disease prevention and health promotion, it requires adequate financial and logistical support to maximize its impact. Ensuring sustained investment in RCCE will enhance community resilience, improve health-seeking behaviors, and strengthen public health responses to future outbreaks. VI. COVID-19 RESPONSE 30. The AFV COVID-19 component, with a budget of USD 5 million, aimed to support the Zimbabwean government’s response to COVID-19 in preventing spread by focusing on infection, prevention, control and strengthening the national response. It enabled the mobilization of surge capacity through trained, well-equipped frontline health workers and better-equipped facilities, also addressing the needs of migrant and displaced populations in fragile settings. By June 2023, Zimbabwe had reported 265,261 confirmed cases, 5,701 deaths, and 259,272 recoveries, with a Case Fatality Rate of 2.1 percent. Progress was made under various pillars, with retroactive financing following the grant's effectiveness in December 2020. 31. To enhance coordination, the project installed video conferencing facilities in eight rural provinces, the Permanent Secretary’s boardroom, and the Minister of Health and Child Care’s boardroom. These facilities enabled nationwide COVID-19 coordination meetings and are now widely used by the MoHCC for stakeholder engagements. In strengthening surveillance, the project facilitated training and implementation of the WHO’s Go. Data system (https://www.who.int/tools/godata) to track and manage case investigations, contact follow-ups, and transmission chains. Initially piloted in three districts of Mashonaland East, the system was later expanded to Harare and Bulawayo Metropolitan provinces for cholera outbreak management, with additional training in Matabeleland North and South. A total of 900 cadres were trained, and real-time data collection, especially during outbreaks, has since improved. 50 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT 32. IPC efforts were reinforced through the installation of 10,000-liter water storage tanks at five hospitals and renovations of COVID-19 isolation wards at Mutawatawa and Binga hospitals. The refurbishments included repairs to lighting, roofing, electrical systems, and ablution facilities, creating safer environments for both patients and healthcare workers. Additionally, IPC training and provision of PPE were extended to healthcare workers, with routine PPE support covering 104 hospitals. The monitoring of PPE usage has since improved through the Impilo Electronic Health Records system. 33. Laboratory capacity was strengthened by developing an electronic reporting tool for case detection and data management. The project procured and distributed 21,000 GeneXpert Xpress SARS-CoV-2 test cartridges to 26 laboratories nationwide, with a tracking system ensuring efficient usage. The last-mile tracking system successfully traced products to the end user and identified the 3 percent that had been lost due to theft. 34. In case management and health systems strengthening, the project provided critical ICU equipment, including eight Linet beds, eight ventilators, and other essential medical supplies to major hospitals such as Sally Mugabe, Mpilo, and UBH. Eight ICU ambulances and suction machines were procured through UNDP and distributed to central and provincial hospitals. Additionally, oxygen reticulation systems were installed at Murehwa and Silobela hospitals to enhance oxygen availability in outpatient departments, emergency rooms, and wards. The project had initially been planned to support 20 hospitals with oxygen concentrators. MOHCC then decided to support oxygen reticulation and due to budgetary constraints only 2 facilities were supported with oxygen reticulation system. Furthermore, the RBF approach was utilized for contact tracing, improving the efficiency and responsiveness of disease surveillance and outbreak management. 35. Under the Risk Communication pillar, the project significantly bolstered the Public Health Emergency Operations Centre (PHEOC) by training and funding shift payments for call operators, ensuring seamless call center operations. To enhance efficiency, an IVR system and additional equipment were provided which are still in use to date supporting the Annex staff at the EOC center. For sustainability, the MoHCC is working to integrate staff into the PHEOC to maintain the call center’s functionality beyond COVID-19, given its critical role in responding to public health emergencies as reliance on interns is unsustainable and raises accountability concerns. Additionally, re- engaging partners for continued support is crucial to maintaining the call center’s operations in the long term. VII. CONCLUSIONS 36. The HSDSP has played a transformative role in strengthening Zimbabwe's maternal, newborn, and child health (MNCH) services, enhancing pandemic preparedness, and improving overall health system resilience over its 13- year implementation period. Through innovative approaches such as Results-Based Financing (RBF), risk communication, infection prevention and control, and institutional capacity building, the project successfully addressed health service delivery gaps, increased healthcare access, and improved quality of care. The Zimbabwean government has demonstrated strong commitment by allocating financial, human, and material resources to sustain the initiative. Given its proven cost-effectiveness and impact, it is crucial to institutionalize key components of the program, secure long-term funding, and further scale up interventions to ensure sustainable health system strengthening and continued progress in MNCH and pandemic response efforts. The HSDSP’s targeted approach has been instrumental in addressing both supply and demand side barriers, leading to significant improvements in maternal and child health indicators in Zimbabwe. 51 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 6. SUMMARY OF KEY CHANGES DURING IMPLEMENTATION Table 6.1. Key Changes Introduced during Project Implementation / World Bank Approval (i) Supported RBF contracts in 2 rural districts and expanded to 16 additional 3.6 million U $ rural districts through July 31, 2014 U $20 (i) Continued RBF contracts in 18 rural districts 4.1 million J 20 3 (ii) Introduced urban RBF innovations • Maternal health voucher scheme (demand-side) • Results based contracts (supply side) (iii) Extended project closing date until October 30, 2015 U $ 0 (i) Continued RBF contracts in 18 rural districts and expanded the package 4.1 million of primary and secondary services 20 (ii) Continued urban RBF innovations • Maternal health voucher scheme (demand-side) • Results based contracts (supply-side) (iii) Introduced Continuous Quality Improvement (CQI) innovations (iv) Extended project closing date until February 28, 2017, and subsequently to January 30, 2018, through level 2 restructuring (i) Continued RBF contracts in 18 rural districts and revised package to 4.1 million U $ include additional interventions (TB, HIV, diabetes) J 20 7 (ii) Continued urban RBF innovations and expanded the package of services, supported improved strategies for household enrollment, and strengthened use of electronic health records • Maternal health voucher scheme (demand-side) • Results based contracts (supply-side) (iii) Rolled out CQI innovations in 5 districts (iv) Rolled out a process evaluation on cost-effective approaches to verification of results under RBF projects (v) Extended project closing date until June 30, 2018 (i) Modified PDO to better reflect the types of interventions supported 4.1 million U $3 (increase coverage and quality of key health interventions with an J emphasis on MCH services and strengthen institutional capacity) 20 (ii) Continued results-based contracts in 18 rural districts with government assuming full responsibility for financing these activities (iii) Continued urban RBF innovations while finetuning selection criteria to better target poor households - Maternal health voucher scheme (demand-side) - Results based contracts (supply-side) (iv) Continued with CQI innovations in 5 districts (v) Continued with implementing process evaluation on cost- effective approaches to verification of results under RBF projects (v) Supported Ministry of Health and Child Care in institutionalizing and Harmonizing the RBF program in Zimbabwe 52 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT (vi) Extended project closing date until December 31, 2020 (i) Modified PDO to support COVID-19 response and to provide an RMNCAH- U $2 integrated package of RMNCAH-N services N: 4.8 (ii) Continued with RBF in 18 rural districts in health centers and DHs. million 2020 (iii)Piloted Community-based RBF and Quality based RBF in provincial and central hospitals COVID (iii)Continued with the urban RBF innovations Response: - Maternal health voucher scheme (demand-side) 14.4 million - Results based contracts (supply-side) (iv)Continued with CQI innovations in 5 districts (v) Included a new component --COVID-19 Emergency Response—including introduction of last mile commodity tracking system for COVID-19 supplies using MOHCC Electronic Health Records System (vi)Conducted blockchain feasibility assessment for RBF verification and commodity tracking (vii)Continued to support MOHCC to institutionalize and harmonize the RBF program in Zimbabwe (viii) Extended project closing date until April 2023 Sources: Interim Implementation Completion and Results Report, World Bank, September 29, 2021; January 2025 ICR mission. 53 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 7. ACHIEVEMENT OF PDO-LEVEL AND INTERMEDIATE RESULTS INDICATORS % / Increase coverage of integrated package of RMNCAH-N services POI 1: Percentage of pregnant women 70.0 80.0 72.0 2020: 500.0 (Surpassed) receiving antenatal care during their (2012) (Dec 2020) (2020) visit to a health provider in participating rural districts* POI 2: Percentage of births attended 58.0 82.0 (MICS 88.0 2020: 80.0 (Substantially Achieved) by skilled health personnel in a health (2012) 2019) (2020) institution in participating rural districts based on survey data* POI 3: Number of pregnant women This indicator was changed into an IRI (#1) living with HIV in participating rural districts who received antiretrovirals to reduce the risk of Mother to Child Transmission POI 4: Percentage of women 15-49 56.0 70.0 70.0 2020: 100.0 (Achieved) years receiving a modern family (2012) (2014) (2020) planning method in participating rural districts* POI 5: Number of children immunized Indicator tracked under IRI 4 in participating rural districts POI 6: Percentage of women who had 10.0 22.0 32.0 2024: 54.5 (Not Achieved) their first ANC visit during the first (2012) (2024) (2024) sixteen weeks of pregnancy in participating rural districts POI 7: Percentage of children under 5 13.8 16.5 16.0 2020: 122.7 (Surpassed) with diarrhea receiving ORT and Zinc (2015) (MICS 2019) (2020) in participating districts* POI 8: Percentage of children 6-59 43.5 30.2 50.0 -204.6 (Not Achieved) months with Vitamin A (2019) (2024) (2024) supplementation in participating rural districts-added IRI 1: Cumulative number of pregnant 9,399 59,572 60,000 2020: 99.2 (Substantially Achieved) women living with HIV who are (2012) (Dec 2020) (2020) initiated on antiretrovirals to reduce the risk of MTCT in participating rural districts* IRI 2: Cumulative number of pregnant 12,737 100,941 89,031 2020: 115.6 (Surpassed) women receiving first antenatal care (2014) (Dec 2020) (2020) during a visit to a health provider in participating urban districts* 54 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT IRI 3: Cumulative number of pregnant 572 4,989 4,035 2020: 127.5 (Surpassed) women receiving first antenatal care (2012) (Dec 2020) (2020) before 16 weeks of gestation period during a visit to a health provider in participating urban districts* IRI 4: Number of children immunized 0 1,218,432 929,004 2024: 131.1 (Surpassed) (completing primary course on (2012) (2024) (2024) immunization in participating districts: BCG, oral polio vaccine 1 to 3, Penta 1 to 3, and measles and rubella 1) (HNP CRI) IRI 5: Number of women and children 0 3,329,251 2,768,670 2024: 120.2 (Surpassed) who have received basic nutrition (2012) (2024) (2024) services (HNP CRI) IRI 6: Number of deliveries attended 0 1,358,721 1,149,839 2024: 118.2 (Surpassed) by skilled health personnel (in health (2012) (2024) (2024) institutions in participating rural and urban districts) (HNP CRI) IRI 7: Cumulative number of health 0 1,408 1,190 2024: 118.3 (Surpassed) facilities enrolled in RBF program in (2024) (2024) participating districts IRI 8: People who have received 0 5,906,403 4,847,512 2024: 121.8 (Surpassed) essential health, nutrition, and (2024) (2024) population services (HNP CRI) IRI 9: Cumulative number of children 0 224,810 121,500 2024: 185.0 (Surpassed) <5 who had their Mid Upper Arm (2024) (2024) Circumference and Height measured by CHWs as part of growth monitoring IRI 10: Percentage of 15–49-year-old 0 17.0 25.0 -122.2 (Not achieved) women who are new acceptors of (2024) (2024) long term, reversible FP methods in participating districts Increase quality of an integrated package of RMNCAH-N services POI 9: Average quality scores by 68.1 83.0 81.0 2020: 115.5 (Surpassed) health facilities in participating rural (2012) (Dec 2020) (2020) and urban districts* POI 10: Percentage of maternal 0 (2015) 76.0 80.0 2020: 95.0 (Substantially Achieved) deaths given audits as per protocol in (Dec 2020) (2020) participating districts* POI 11: Percentage of partographs 0 62.0 65.0 2020: 95.4 (Substantially Achieved) correctly filled in participating (2015) (Dec 2020) (2020) districts* POI 12: Percentage of children under 0 89.5 87.0 2020: 102.9 (Surpassed) 5 years with Pneumonia correctly (2015) (Dec 2020) (2020) managed in the participating districts* 55 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT POI 13: Percentage of participating 51.0 51.35 60.0 2024: 3.9 (Not achieved) district, provincial, and central (2019) (2024) (2024) hospitals that have registered an increase in quality scores since last quarter IRI 11: Percentage of health facilities 0 (2015) 50.0 46.0 2024: 108.7 (Surpassed) implementing Continuous Quality (2024) (2024) Improvement model in the participating rural districts IRI 12: Percentage of RBF contracted 0 (2015) 76.0 80.0 2020: 95.0 (Substantially Achieved) facilities in Continuous Quality (Dec 2020) (2020) Improvement (CQI) Districts with CQI Standard Operating Procedures* IRI 13: Number of District Health 0 25.0 25.0 2024: 100.0 (Achieved) Executives (DHEs) in participating (Jan 2012) (2024) (2024) districts using quality tool for supervision of health facilities 72.5 80.0 0 (2012) 2024: 90.6 (Substantially Achieved) POI 14: Percentage of health facilities (2024) (2024) managed under RBF contracts by the MOHCC Program Coordination Unit in participating rural districts 8,223 5,792 IRI 14: Number of health personnel 0 2024: 141.9 (Surpassed) (2024) (2024) and CHWs receiving training on RBF in participating districts 302 280 IRI 15: Number of health workers that 0 2024: 107.9 (Surpassed) (2024) (2024) received first time training or refresher training on EONC using skills labs IRI 16: Percentage of MOHCC cost 0 100.0 Delayed centers utilizing resource allocation formula in line with the whole Government Management Performance System 56 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT IRI 17: Process evaluation to examine No Yes Yes 2024: Achieved the institutionalization of RBF in the government system as a viable approach to sustainability of RBF IRI 18: Process evaluation to examine No No Yes 2024: Not achieved the effectiveness of quality focused RBF at provincial hospitals in improving the quality of maternal and neonatal care for improved outcomes POI 15: Percentage of close contacts 0 82.0 80.0 2024: 102.5 (Surpassed) confirmed COVID-19 cases followed (2020) (2024) (2024) up based on national guidelines IRI 19: Percentage of health facilities 5.0 89.0 95.0 with PPE commodities 2024: 93.3 (Substantially achieved) (2020) (2024) (2024) IRI 20: Number of affected people including health workers reached 0 4,573 1,680 2024: 272.2 (Surpassed) with clinical and psycho-social (2020) (2024) (2024) support IRI 21: Number of isolation centers 0 6 6 2024: 100.0 (Achieved) complying with IPC guidelines (2020) (2024) (2024) IRI 22: Number of COVID-19 0 2 30.0 treatment centers with oxygen 2024: 6.7 (Not achieved) (2020) (2024) (2024) therapy Note: The rating system used is as follows: surpassed/achieved: 100 percent+; substantially achieved: 80-99 percent; partially achieved: 65-79 percent; not achieved: <65 percent. For surpassed indicators, a weighted scale was used to reflect the degree of over-achievement. *These indicators were dropped in AF V because they were substantially or fully achieved. 57 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT ANNEX 8. BROAD TRENDS IN KEY INDICATORS Summary points: • Zimbabwe achieved progress on some key MNCH impact and intervention coverage indicators during 2010-2019 when the RBF program was introduced and then scaled up. Although this progress cannot be attributed solely to the RBF program, it is notable that these improvements occurred as the program was rolled out and when subsidies were being paid in USD in the 18 districts originally supported by the World Bank. • Substantial reductions in the large intervention coverage gaps between urban and rural areas in 2010 were achieved by 2019. Coverage levels in rural areas matched or exceeded those in urban areas by 2019 for 4 of the 6 interventions examined in this report and supported under the project (i.e., first ANC visit before 16 weeks, ANC 4+, full immunization coverage of basic vaccines, and treatment of childhood diarrhea with ORS). These patterns suggest that the RBF program may have contributed to rapid progress in improving service coverage in rural areas and provided the justification for the eventual introduction of the urban voucher program to ensure the urban poor were not left behind. • Coverage levels of maternal and child health interventions deteriorated between 2019 and 2024 and progress towards reducing mortality, fertility and stunting levels stagnated or reversed . These deteriorating trends have occurred amidst the backdrop of the COVID19 pandemic and several crippling macroeconomic shocks. Analysis examining whether these patterns would have been worse without the infusion of additional funds to support Zimbabwe’s COVID19 response and the RBF program is beyond the scope of this ICR. Evidence of the quick rebounding of key indicators such as institutional delivery by as early as 2021 indicate the important role these funds played in the recovery of Zimbabwe’s health system. However, measures such as maternal and child mortality and stunting are impacted by many factors beyond the health sector including food insecurity and household poverty that have worsened over the past five years because of the macroeconomic situation. Stunting prevalence, for example, declined approximately 27 percent (from 32 percent to about 23.5 percent) between 2010 and 2019, but then increased about 15 percent (reaching 27 percent prevalence) between 2019 and 2024. • Z ’ U nited Nations (UN) Z ’ . Zimbabwe has historically performed better on stunting (e.g., 22 percent prevalence in 2022 compared to 31 percent in the region), although latest survey data suggests an increase in stunting prevalence in the past five years. Based on the UNICEF global database (https://data.unicef.org/resources/data_explorer/unicef_f/), intervention coverage levels for four or more ANC visits and skilled birth attendant has been consistently higher in Zimbabwe in the past 15 years than in Eastern and Southern Africa (72 percent vs. 54 percent in 2019 for ANC4+, and 85 percent vs. 70 percent for births attended by skilled health personnel. Similarly, Zimbabwe has historically performed better on childhood immunization rates (for DPT3, MCV1, PCV3) in comparison to the Eastern and Southern Africa average. However, Zimbabwe’s coverage levels of care seeking for children with symptoms of acute respiratory infection (ARI) and diarrhea treatment have trended lower than the regional average since 2007. For graphic illustrations of Zimbabwe’s progress on a range of reproductive, maternal, newborn child health and nutrition indicators in comparison to the Eastern and Southern Africa region, see the Countdown to 2030 Zimbabwe country profile: https://data.unicef.org/countdown-2030/country/Zimbabwe/1/ 1. This brief analysis is organized into the following sections: 1) Review of select MNCH impact indicators based on Demographic and Health Survey (DHS) and MICS surveys from 2010 to 2024, 2) Review of essential MNCH intervention coverage indicators based on DHS and MICS surveys from 2010 to 2024, and routine health facility data from 2019 to 2023 based on analysis conducted by the Zimbabwe Countdown country collaboration, and 3) Comparison of Zimbabwe’s progress to the Eastern and Southern Africa average using United Nations (UN) inter-agency estimates and global databases. Analyses are provided where possible at the national level and by urban and rural. Because the RBF program was eventually scaled up to all districts and comparable trend data are not readily available from 2010 to 2024 58 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT at the district level for all key indicators, a district level analysis was not conducted. This analysis provides a picture of general trends in Zimbabwe. The data sources and level of analysis differ from those used for monitoring the RBF program and are not directly comparable. Section 1. MNCH Impact indicators – household survey analysis 2. At the national level, Zimbabwe experienced mixed progress in key maternal and child health impact indicators over the timeframe 2010-2024 (Table 8.1). There are important distinctions in the survey trends between the 10-year period prior to the COVID19 pandemic (2010-2019) when the RBF program was first introduced and later scaled up, and the following five years (2019-2024) when Zimbabwe experienced multiple exogenous shocks. Although data are not yet available from the 2023-2024 DHS for maternal mortality, the survey data show consistent decreases in the maternal mortality ratio over the time frame 2010-2019 (approximately 29 percent decline). The 2022 National Housing and Population Census reported a Maternal Mortality Ratio of 363 Maternal Death per 100,000 live births, indicating continued reductions after 2019. The total fertility rate showed modest declines from 2010-2019 and no change between 2019 and 2024. The under-five mortality rate declined from 84 deaths per 1,000 live births in 2010 to around 69 deaths per 1,000 live births in 2024 (an approximate 18 percent decrease). However, the pace of decline for under- five mortality was more substantial in the 2010-2019 period (around 13.1 percent) compared to the 2019-2024 period (about 5.5 percent. Most concerning are the patterns for stunting prevalence and neonatal mortality, both of which showed large percentage increases during 2019-2024 (14.9 percent and 19.4 percent, respectively). In sum, Zimbabwe achieved reductions in key MNCH impact indicators except neonatal mortality in the time frame 2010-2019. After 2019, progress slowed or reversed for all the impact indicators for which preliminary 2023-2024 DHS data are available. Table 8.1. Trends in select maternal and child health impact indicators, national level, 2010-2024 2010-2024 2010-2024 2010-2019 2019-2024 Indicator 2010-2011 2015 2019 2023-2024 (absolute difference) (% change) (% change) (% change) Maternal Maternal mortality ratio (deaths per 100,000 live births) 960 651 462 N/A N/A N/A -29.0 N/A Total fertility rate (women 15-49 years of age) 4.1 4 3.9 3.9 0.2 -4.9 -4.9 0.0 Child Neonatal mortality rate (deaths per 1,000 live births) 31 29 31 37 6 19.4 0.0 19.4 Under five mortality rate (deaths per 1,000 live births) 84 69 73 69 -15 -17.9 -13.1 -5.5 Stunting prevalence (moderate and severe) (%) 32 27 23.5 27 -8.5 -15.6 -26.6 14.9 N/A: data not available Data sources: 2010-2011 DHS, 2015 DHS, 2010 MICS, 2023-2024 DHS preliminary report 3. Comparisons of progress in urban and rural areas show similar trends for stunting prevalence and total fertility rate (TFR), the two indicators for which urban/rural data are available across the surveys (Table 8.2). The TFR remained consistently higher in rural areas compared to urban areas over 2010-2024. Small reductions were achieved in both urban and rural areas during 2010-2019 followed by small increases in the TFR in 2019-2024. Although stunting levels were consistently lower in urban areas over 2010-2024 compared to rural areas, the percentage increase in stunting was larger (more than double) in urban areas in 2019-2024. 59 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT Table 8.2: Trends in total fertility rate and stunting prevalence, urban and rural, 2010-2024 Year of survey Urban Rural Indicator 2010-2011 2015 2019 2023-2024 2010-2019 2019-2024 2010-2019 2019-2024 Urban Rural Urban Rural Urban Rural Urban Rural (% change) (% change) (% change)(% change) Total fertility rate (women 15-49 years of age) 3.1 4.8 3 4.7 3 4.5 3.1 4.6 -3.2 3.3 -6.3 2.2 Stunting prevalence (moderate and severe) (%) 27.5 36.1 22.1 28.5 18.5 25.6 23.5 28.8 -32.7 27.0 -29.1 12.5 Data sources: 2010-2011 DHS, 2015 DHS, 2019 MICS, 2023-2024 DHS preliminary report Section 2. MNCH outcome indicators – selected indicators consistent with prioritized interventions in the RBF program 4. Z ’ mothers and children over 2010-2024 is mixed (tables 8.3, 8.4, 8.5), with reversals in the time frame 2019 to 2024 for indicators for which data are available from the preliminary DHS results. At the national level, substantial improvements in maternal health interventions occurred between 2010 and 2019, ranging from percentage point increases of around 7 for ANC4+ visits to 55 for postnatal care for mothers (within 2 days). Coverage levels for full vaccination (basic vaccines) also increased a substantial 21 percentage points during this time frame, but coverage of ORS declined about 1 percentage point. Coverage of all selected maternal and child health interventions declined between 2019 and 2024. 5. Examination of urban and rural patterns show that coverage levels of almost all the maternal and child health interventions examined were higher in urban areas in 2010 (the exception was first ANC visit before 16 weeks, which had equal low coverage of about 19 percent in both urban and rural areas), ranging from 2 percentage points (ANC 4+ visits) to 28 percentage points (skilled birth attendant) higher. These differentials narrowed over time with coverage levels in rural areas approximating or exceeding those in urban areas for 4 of the 6 interventions by 2019 (e.g., coverage levels were nearly the same for full vaccination with basic antigens, 19 percentage points higher for first ANC visits by 16 weeks in rural areas in comparison to urban areas, 5.5 percentage points higher in rural areas for 4+ ANC visits, and 1.5 percentage points higher for treatment with ORS). The coverage gap between urban and rural areas for skilled birth attendance also dropped from 28 percentage points in 2010 to 12 percentage points in 2019. These trends suggest that the rural RBF program may have been successful in rapidly increasing coverage levels in rural areas and provided further justification for the introduction of the urban voucher program to ensure urban poor women and children are not left behind. Coverage levels dropped more precipitously in rural compared to urban areas in the 2019-2024 timeframe, with coverage levels falling below urban levels for all but ANC 4+ visits. This trend is a stark reminder that reversals can occur during pandemics and difficult macroeconomic periods. Table 8.3. Maternal and child health, select intervention coverage measures, 2010-2024, national, urban, rural 2010-2011 2015 2019 2023-2024 Indicator National (%) Urban (%) Rural (%) National (%) Urban (%) Rural (%) National (%) Urban (%) Rural (%) National (%) Urban (%) Rural (%) Maternal health Antenatal care (first visit before 16 weeks) 19.4 19 19.6 38.5 33.9 40.7 40.3 27.4 46.3 N/A N/A N/A Antenatal care (4+ visits) 64.8 66 64.3 75.7 77.4 74.9 71.5 67.8 73.3 71.2 69.9 71.7 Skilled birth attendant 66 86 58 78.1 92.9 71.3 86 94.3 82.2 84.5 93.6 79.4 Postnatal care for mothers (within 2 days) 27.1 40.6 21.5 56.6 66.7 52.6 82.1 84.3 79.8 67.8 77.5 62.6 Child health Fully vaccinated (12-23 months)* 64.5 69.9 62.3 76 80.5 74.2 85.9 88.9 84.7 71.5 72.4 70.9 Treatment with ORS** 63.3 68 60.9 69.5 72 68.3 62.1 61 62.5 34.3 36.8 32.8 Careseeking for symptoms of ARI 48 43.7 48.9 50.9 69.1 45.2 N/A N/A N/A 44.5 65 32.4 60 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT *All basic vaccinations ** This includes ORS or government recommended home fluid Data sources: 2010-2011 DHS, 2014 MICS, 2015 DHS, 2019 MICS, 2023-2024 DHS preliminary report N/A: Not available Table 8.4. Maternal and child health, select intervention coverage measures, national level trends, 2010-2024 2010-2011 2015 2019 2023-2024 2010-2024 2010-2024 2010-2019 2019-2024 Indicator (%) (%) (%) (%) Prct. point change Change (%) Change (%) Change (%) Maternal health Antenatal care (first visit before 16 weeks) 19.4 38.5 40.3 N/A N/A N/A 107.7 N/A Antenatal care (4+ visits) 64.8 75.7 71.5 71.2 6.4 9.9 10.3 -0.4 Skilled birth attendant 66 78.1 86 84.5 18.5 28.0 30.3 -1.7 Postnatal care for mothers (within 2 days) 27.1 56.6 82.1 67.8 40.7 150.2 203.0 -17.4 Child health Fully vaccinated (12-23 months)* 64.5 76 85.9 71.5 7 10.9 33.2 -16.8 Treatment with ORS** 63.3 78 62.1 34.3 -29 -45.8 -1.9 -44.8 Careseeking for symptoms of ARI 48 50.9 N/A 44.5 -3.5 -7.3 N/A N/A *All basic vaccinations Data sources: 2010-2011 DHS, 2014 MICS, 2015 DHS, 2019 MICS, 2023-2024 DHS preliminary report ** This is ORS or government recommended homemade fluid N/A: Not available Table 8.5. Maternal and child health, select intervention coverage measures, urban and rural trends, 2010-2024 Intervention coverage levels Urban Rural Indicator 2010-2011 2015 2019 2023-2024 2010-2024 2010-2019 2019-2024 2010-2024 2010-2019 2019-2024 Urban (%) Rural (%) Urban (%) Rural (%) Urban (%) Rural (%) Urban (%) Rural (%) Prct. Point Prct. Point Prct. Point Prct. Point Prct. Point Prct. Point Maternal health Antenatal care (first visit before 16 weeks) 19 19.6 33.9 40.7 27.4 46.3 N/A N/A N/A 8.4 N/A N/A 26.7 N/A Antenatal care (4+ visits) 66 64.3 77.4 74.9 67.8 73.3 69.9 71.7 3.9 1.8 2.1 7.4 9 -1.6 Skilled birth attendant 86 58 92.9 71.3 94.3 82.2 93.6 79.4 7.6 8.3 -0.7 21.4 24.2 -2.8 Postnatal care for mothers (within 2 days) 40.6 21.5 66.7 52.6 84.3 79.8 77.5 62.6 36.9 43.7 -6.8 41.1 58.3 -17.2 Child health Fully vaccinated (12-23 months)* 69.9 62.3 80.5 74.2 88.9 84.7 72.4 70.9 2.5 19 -16.5 8.6 22.4 -13.8 Treatment with ORS** 68 60.9 72 68.3 61 62.5 36.8 32.8 -31.2 -7 -24.2 -28.1 1.6 -29.7 Careseeking for symptoms of ARI 43.7 48.9 69.1 45.2 N/A N/A 65 32.4 21.3 N/A N/A -16.5 N/A N/A *All basic vaccines Data sources: 2010-2011 DHS, 2014 MICS, 2015 DHS, 2019 MICS, 2023-2024 DHS preliminary report ** This is ORS or government recommended homemade fluid (essentially ORT, which is consistent with the RBF indicator) N/A: Not available Countdown country chartbooks – trends from 2019 to 2024 6. In April 2024, the Countdown Zimbabwe country collaboration completed a 5-year time series analysis of essential MNCH interventions based on routine Health Management Information System/facility data (2019-2023). These data were compiled into a chartbook available on the GFF data portal (Zimbabwe country webpage) and Countdown to 2030 website. 7. The Countdown facility-based analyses show a more positive picture for immunization coverage than the survey analysis, but this is likely an artifact of the indicator used. The above tables based on the household survey data use full immunization coverage with basic vaccines. The chartbook presents trends for individual antigens such as pentavalent, MCV1, MCV2, and BCG vaccination, which are consistent with the UN interagency estimates and Household surveys. The trend data show that coverage levels of all four individual vaccines improved between 2019 and 2023, and that national targets for pentavalent and BCG were achieved. This finding is consistent with statements from interviewees about the strength of Zimbabwe’s nurse-led EPI program and ongoing efforts to improve community vaccination campaigns (including to address the current vaccine-derived polio outbreak). Further investigation may be 61 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT warranted into challenges in achieving universal coverage of full vaccination coverage with basic vaccines and according to the national vaccine schedule in Zimbabwe even if coverage of individual vaccines is high. 8. Consistent with the survey analysis above, the Countdown team found that coverage of ANC services began declining after 2020 because of the COVID19 pandemic and subsequent high levels of staff attrition. The Countdown analysis found that a dip in institutional delivery coverage occurred in 2020, likely due to COVID19, but rebounded to pre-covid levels by 2021. The main findings of the Countdown health system performance assessment corroborate concerns raised by interviewees about the high levels of health worker out-migration and highlight areas in the country where health facility density falls far short of the international benchmark of 2 per 10,000 population. 9. In summary, the Countdown and household survey analyses show that Zimbabwe had been making some progress during 2010-2019 with increasing access to lifesaving care, mostly with maternal health services and improvements in maternal and child mortality and stunting. Multiple shocks – macroeconomic and COVID19 – experienced in the past five years have negatively impacted MNCH service coverage and health system readiness to provide high quality care. These downturns have been notably reflected in increases in neonatal mortality – a finding remarked upon in both the DHS preliminary results summary and in the Countdown analysis – as well as in stunting. Section 3: Placing Zimbabwe in context: comparison with Eastern and Southern Africa based on UN estimates and UNICEF global database 10. Zimbabwe’s progress on mortality, fertility and stunting levels is consistent with secular trends across Eastern and Southern Africa. Although Zimbabwe has achieved relatively steady declines in maternal mortality, its maternal mortality levels have remained slightly higher than the regional average for over the past ten years. Based on the latest UN interagency estimates, Zimbabwe’s MMR was 357 deaths per 100,000 live births in 2020 compared to 324 in East and Southern Africa (MMEIG, 2023). Zimbabwe’s neonatal and under-five mortality rates have tracked more closely with regional patterns. According to UN interagency estimates, Zimbabwe’s neonatal mortality rate was 24 deaths per 1,000 live births in 2022 compared to 23 in the East and Southern Africa region. Similarly, Zimbabwe’s under-five mortality rate in 2022 was 48 deaths per 1,000 live births compared to 51 in the region.30 In terms of the total fertility rate, Zimbabwe’s rate was around 3.77 in 2022, compared to 4.39 in sub-Saharan Africa (4.19 in Eastern Africa, 2.31 in Southern Africa) (World Population Prospects: The 2024 revision). The joint UN malnutrition estimates show that up to 2022, Zimbabwe was performing consistently better than the Eastern and Southern Africa regional average (22 percent prevalence in 2022 compared to the regional average of 31 percent). 11. Available data from UNICEF global databases on intervention coverage trends show that Zimbabwe has consistently outperformed the Eastern and Southern Africa region on key maternal health interventions such as ANC 4+ visits and births attended by skilled health personnel, as well as on childhood vaccination services. Zimbabwe has underperformed in comparison to the region on key measures of case management of childhood illnesses such as care seeking for symptoms of ARI and treatment of diarrhea with ORS (See pages 1 and 2 from the latest Zimbabwe Countdown to 2030 country profile below, which compares many RMNCAH&N indicators to the Eastern and Southern Africa regional average). References: 1) UN IGME, Levels & Trends in Child Mortality: Report 2023, Estimates developed by the United Nations Inter- agency Group for Child Mortality Estimation, United Nations Children’s Fund, New York, 2024. 30 2024. United Nations Inter-agency Working Group on Child Mortality Estimation (UN IGME). 62 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT 2) Trends in maternal mortality 2000 to 2020: estimates by WHO, UNICEF, UNFPA, World Bank Group and UN Department of Economic and Social Affairs (DESA)/Population Division. Geneva: World Health Organization; 2023. Licence: CC BY-NC-SA 3.0 IGO. 3) United Nations, Department of Economic and Social Affairs, Population Division (2024). World Population Prospects 2024: Summary of Results (UN DESA/POP/2024/TR/NO.9). 4) Zimbabwe Countdown to 2030 profile, pages 1 and 2: https://data.unicef.org/countdown- 2030/country/Zimbabwe/1/; https://data.unicef.org/countdown-2030/country/Zimbabwe/2/ (snapshot below) 63 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT 64 The World Bank Health Sector Development Support Project (P125229) ICR DOCUMENT i Risk-Based Verification refers to a targeted and proportionate approach to verifying the accuracy and reliability of reported health service data. Facilities are categorized into different risk levels (Green - Low Risk, Amber - Medium Risk, Red - High Risk) based on their historical reporting accuracy, specifically the margin of error in reported data over the past two quarters. The frequency and intensity of verification efforts are adjusted according to the assigned risk category. 65