Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00006349 IMPLEMENTATION COMPLETION AND RESULTS REPORT IDA D3330 ON A GRANT IN THE AMOUNT OF SDR 65,360,465 (US$ 86,500,000 EQUIVALENT) TO THE Democratic Republic of Congo FOR A DRC - Gender Based Violence Prevention and Response Project March 29, 2024 Social Sustainability And Inclusion Global Practice Eastern And Southern Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective March 29, 2024) Currency Unit = SDR 0.76 SDR = US$1 US$ 1.32 = SDR 1 FISCAL YEAR July 1 - June 30 Regional Vice President: Victoria Kwakwa Country Director: Albert G. Zeufack Regional Director: Iain G. Shuker Practice Manager: David Seth Warren Task Team Leader(s): Hiska Noemi Reyes ICR Main Contributor: Elizabeth Susan Graybill ABBREVIATIONS AND ACRONYMS AFRGIL Africa Gender Innovation Lab BCR Benefit cost ratio CAS Country Assistance Strategy CERC Contingency and Emergency Response Component CBO Community-Based Organizations CFEF Financing Unit in Favor of Fragile States ( Cellule d’Exécution des Financements en faveur des Etats Fragiles) CoE Center of Excellence CMU Country Management Unit CPF Country Partnership Framework CSPP Project and Program Monitoring Unit (Cellule de Suivi des Projets et Programmes) DRC Democratic Republic of Congo ESMF Environmental and Social Management Framework FCV Fragility, Conflict and Violence FM Financial Management FSRDC Social Fund for Democratic Republic of Congo (Fonds Social pour la République Démocratique du Congo) GBV Gender-Based Violence GL GBV Great Lakes Emergency Sexual and Gender-Based Violence and Women's Health Project GRM Grievance Redress Mechanism HCP Health Care Provider IGA Income Generating Activity IPV Intimate Partner Violence IRI Intermediate Results Indicator IRR Internal Rate of Return ISR Implementation Status and Results Report MGFC Ministry of Gender, Family and Children MoPH Ministry of Public Health MTR Mid-Term Review NET Narrative Exposure Therapy NGO Non-Governmental Organization NPV Net Present Value NSCGBV National Strategy to Combat Gender-Based Violence OSC One-Stop Center PAD Project Appraisal Document PBF Performance Based Financing PDO Project Development Objective PDSS Health System Strengthening for Better Maternal and Child Health Results Project PEP Post-Exposure Prophylaxis PHD Provincial Health Directorate PMNS Multisectoral Nutrition and Health Project PTSD Post-Traumatic Stress Disorder SEA/SH Sexual Exploitation and Abuse and Sexual Harassment STI/D Sexually Transmitted Infection/Disease VSLA Village Savings and Loan Association TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 5 A. CONTEXT AT APPRAISAL .........................................................................................................5 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) ..................................... 10 II. OUTCOME .................................................................................................................... 11 A. RELEVANCE OF PDOs ............................................................................................................ 11 B. ACHIEVEMENT OF PDOs (EFFICACY) ...................................................................................... 12 C. EFFICIENCY ........................................................................................................................... 17 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 20 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 20 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 21 A. KEY FACTORS DURING PREPARATION ................................................................................... 21 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 22 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 24 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 24 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 26 C. BANK PERFORMANCE ........................................................................................................... 27 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 27 V. LESSONS AND RECOMMENDATIONS ............................................................................. 28 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 31 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 43 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 46 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 47 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 60 ANNEX 6. SUPPORTING DOCUMENTS .................................................................................. 64 ANNEX 7. THEORY OF CHANGE AND RESULTS CHAIN DIAGRAMS ......................................... 66 ANNEX 8. PDO AND INTERMEDIATE RESULTS INDICATOR TABLES ........................................ 68 ANNEX 9. SUMMARY OF NET IMPACT EVALUATION ............................................................. 71 ANNEX 10. PROCESS EVALUATION ....................................................................................... 73 ANNEX 11. MAP OF PROJECT INTERVENTION AREAS ............................................................ 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name DRC - Gender Based Violence Prevention and Response P166763 Project Country Financing Instrument Congo, Democratic Republic of Investment Project Financing Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Organizations Borrower Implementing Agency Ministry of Finance, Democratic Republic of Congo Ministry of Finance Project Development Objective (PDO) Original PDO The objectives of the Project are to increase in targeted Health Zones: (i) the participation in Gender-Based Violence (GBV) prevention programs; (ii) the utilization of multi-sectoral response services for survivors of GBV; and (iii) in the event of an Eligible Crisis or Emergency, to provide immediate and effective response to said Eligible Crisis or Emergency. Page 1 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 100,000,000 86,500,000 77,204,116 IDA-D3330 Total 100,000,000 86,500,000 77,204,116 Non-World Bank Financing 0 0 0 Total 0 0 0 Total Project Cost 100,000,000 86,500,000 77,204,116 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 30-Aug-2018 19-Jul-2019 02-May-2022 30-Jun-2023 30-Sep-2023 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 30-Jun-2023 73.71 Change in Implementing Agency Change in Components and Cost Change in Loan Closing Date(s) Cancellation of Financing Reallocation between Disbursement Categories Change in Disbursements Arrangements Change in Institutional Arrangements 29-Jan-2024 76.68 Change in Components and Cost Cancellation of Financing Reallocation between Disbursement Categories KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Satisfactory Substantial Page 2 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 24-Dec-2018 Satisfactory Satisfactory .65 02 17-Jun-2019 Moderately Satisfactory Moderately Satisfactory .82 03 20-Dec-2019 Moderately Satisfactory Moderately Satisfactory 5.75 04 30-Jun-2020 Moderately Satisfactory Moderately Satisfactory 11.66 05 29-Jun-2021 Moderately Satisfactory Moderately Satisfactory 33.38 06 05-Aug-2022 Moderately Satisfactory Moderately Satisfactory 50.80 SECTORS AND THEMES Sectors Major Sector/Sector (%) Health 13 Health 10 Health Facilities and Construction 3 Social Protection 87 Social Protection 69 Public Administration - Social Protection 18 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Human Development and Gender 100 Gender 100 Health Systems and Policies 45 Health System Strengthening 45 Health Service Delivery 45 Page 3 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ADM STAFF Role At Approval At ICR Regional Vice President: Makhtar Diop Victoria Kwakwa Country Director: Jean-Christophe Carret Albert G. Zeufack Director: Ede Jorge Ijjasz-Vasquez Iain G. Shuker Practice Manager: Robin Mearns David Seth Warren Task Team Leader(s): Patricia Maria Fernandes Hiska Noemi Reyes ICR Contributing Author: Elizabeth Susan Graybill Page 4 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES 1. This implementation completion and results report (ICR) was prepared based on a desk review of World Bank supervision reports, project progress reports provided to the World Bank, World Bank country strategy documents and relevant project documents, relevant studies conducted during project implementation, project monitoring and evaluation data, as well as interviews with project implementation team, World Bank staff, and the Borrower.1 As noted below, in late April 2023, the project implementing agency, the Social Fund for the Democratic Republic of Congo (Fonds Social de la République Démocratique du Congo, FSRDC), was dissolved and recreated with a new legal status, ultimately postponing the original closure date of June 30, 2023. This ICR offers an analysis of project activities and relevant data reflecting the early cessation of project activities, as well as all project documents and progress reports issued with the formal closing of the project on September 30, 2023. A. CONTEXT AT APPRAISAL Context 2. At the time of project appraisal, the Democratic Republic of the Congo (DRC) was a country emerging from a long period of conflict which had had a devastating impact on institutions, the economy, and the social fabric . At appraisal, the country remained plagued by conflict, especially in its Eastern and Central regions, with a proliferation of armed groups and a military presence that generated tremendous humanitarian needs, especially in the North Kivu, South Kivu, Maniema, and Tanganyika provinces where project implementation took place. At this time, the security situation was aggravated by political instability generated by delays in elections due to take place in 2016; elections finally took place in December 2018, just after project approval, and ushered in a new political administration. 3. At appraisal, DRC ranked 176th out of 188 countries 2 in the 2016 Gender Inequality Index, a benchmark for national gender gaps using economic, political, education, and health criteria. Gender-based violence (GBV) represented a significant barrier to women’s full engagement in social and economic life in the DRC and was correlated with insecurity, especially in Eastern DRC. Prevalence rates of GBV in DRC were high.3, Overall, 52 percent of all women aged 15-49 reported experiencing physical violence (by any perpetrator)4 while 27 percent experienced sexual violence.5 For women aged 15-49 who experienced physical violence, the perpetrator was most often a current husband or partner (56.8 percent). Globally, average prevalence rates for violence against women6 were estimated by the World Health Organization (WHO) at 35.6 percent and the regional average in Africa at 37.7 percent.7 1 The following individuals provided additional key support and contributions to the ICR report: GBV and Gender Consultants Tamara Bah, Harald Hinkel, Helena Hwang, and Katie Robinette; Senior Economist Julia Vaillant for the Africa Gender Innovation Lab; and environment, social, and fiduciary specialists who were members of the task team (Bertille Ngameni Wepanjue, Cyrille Ngouana Kengne, Jean-Claude Azonfack, and Shamard Shamalirwa). 2 With a score of 0.663. 3 In 2017, over 26 000 new cases of sexual violence were identified in areas of the country affected by the humanitarian crisis (31% of which in Nord-Kivu). UNOCHA Humanitarian Response Plan, December 2017; see also DRC Sexual and Gender-Based Violence Sub-Cluster, August 2017; Demographic and Health Survey 2013-2014. 4 At least once since the age of 15 (Demographic and Health Survey, 2013-2014). 5 Of which 16 percent in the last 12 months (Demographic and Health Survey, 2013-2014). 6 Rates of physical or sexual violence for women aged 15-49. 7 World Health Organization (2013). Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. Geneva, World Health Organization. Page 5 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 4. Efforts to address gender inequality and gender-based violence were uneven. By project appraisal, important gains had been made in areas such as health and education, and in terms of legislation addressing gender bias.8 Persistent socio-cultural disparities as well as entrenched gender norms, however, continued to restrict women’s engagement in social and economic life as well as in public decision-making roles. 5. In addition, at appraisal, the Great Lakes Emergency Sexual and Gender-Based Violence and Women's Health project (GL GBV Project) – the first ever large-scale GBV and health project in the Africa region financed by the World Bank and implemented by the DRC Ministry of Public Health (MoPH) as well as FSRDC9 – was nearing closure, and this project was prepared as a follow-on operation. The GL GBV Project had partnered with the health sector and specialized Centers of Excellence (CoEs) – Panzi Foundation in South Kivu and Heal Africa in North Kivu – to offer integrated care to GBV survivors as well as with civil society to implement prevention programming and survivor response via NGOs and women’s community-based organizations (CBO). FSRDC had generally served as an efficient and reliable implementing agency under the GL GBV Project. As such, the FSRDC was well-placed to continue in the same role under this project, which benefited from FSRDC continued leadership at senior management levels between both projects. In addition, the CoEs had been strong technical partners for assuring survivor care under the GL GBV Project; with partnerships in place under the GL GBV Project, the CoEs were readily able to continue their collaboration under this project. 6. Lessons learned from the GL GBV Project’s mid-term review acknowledged challenges of service delivery through the health system and emphasized the importance of expanding GBV prevention and response interventions at the community level and decentralizing services to expand access.10 The GL GBV Project had experienced coordination challenges with MoPH under a complex regional structure and in assuring quality service delivery through government health centers. Under the GL GBV Project, CBOs were key to expanding project reach for survivors, including with delivery of Narrative Exposure Therapy (NET), a specialized mental health care intervention, which proved to be more reliably done through CBOs. Accordingly, this project shifted focus from working primarily with health structures and CoEs to working with local NGOs and women’s CBOs at a much greater scale, while partnering with CoEs and maintaining a health sector technical collaboration. The project also anticipated the need for assessing service quality, which had been weak under the GL GBV Project, by including quality audits at health centers and a process evaluation to complement or cover gaps in health system data. As under the GL GBV Project, at appraisal, insecurity was continuing in Eastern DRC, where gaps in GBV service provision were evident and rates of help-seeking behavior were low. 7. The Government of the DRC recognized the burden that gender inequality, including GBV, placed on social and economic development. The government had developed its Comprehensive Strategy on Combating Sexual Violence in the Democratic Republic of Congo in coordination with the United Nations Organization Stabilization Mission in the DRC (MONUSCO) in April 2009. Later that year, it also put into place its National Strategy on Combating Gender-Based Violence (NSCGBV) and related five-year National Action Plan under the leadership of the Ministry of Gender, Family and 8 For example, the revised Family Code adopted in 2016 eliminated discriminatory provisions regarding land and resources access for women and increased the minimum age of marriage for girls from 15 to 18; the DRC had also already adopted legislation in 2006 criminalizing sexual violence. DRC was also party to the following: (i) UN Convention on the Elimination of all Forms of Discrimination Against Women; (ii) UN Convention of the Rights of Child and African Charter on the Rights and Welfare of the Child; (iii) Declaration on Elimination of Violence Against Women; (iv) UN Security Council Resolution 1325 on Women, Peace and Security (2000) and UN Security Council Resolution 1820 on sexual violence in situations of armed conflict (2008); (v) African Charter on Human and Peoples’ Rights and Protocol to the African Charter on the Rights of Women in Africa (Maputo Protocol) (2003); (vi) Protocol on the Prevention and Suppression of Sexual Violence against Women and Children of the International Conference on the Great Lakes Region (2006); (vii) Goma Declaration on eradicating sexual violence & ending impunity in the Great Lakes Region (2008); and (viii) Kampala Declaration on ending impunity (2003). 9 The Financing Unit in Favor of Fragile States (Cellule d’Exécution des Financements en faveur des Etats Fragiles , CFEF), under the Ministry of Finance, was originally the main implementing agency with overall oversight for the GL GBV Project. Responsibility for project implementation was later shifted to the MoPH and FSRDC under subsequent project restructurings. 10 See generally Project PAD, GL GBV Project ICR. Page 6 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Children (MGFC), which set out a comprehensive framework for GBV prevention and response interventions.11 Nonetheless, gaps in framework implementation and MGFC’s ability to coordinate GBV programming remained. 8. The project was aligned with the World Bank’s Country Assistance Strategy (CAS) for the DRC for FY13-FY1612 and the Government’s own framework under the NSCGBV and related National Action Plan. Accordingly, the project aimed to contribute to the reduction of women’s and girls’ vulnerability in the DRC where high rates of GBV were exacerbated by ongoing conflict. The DRC CAS recognized the country’s profound gender inequalities and threat posed by GBV to sustainable development. The project’s underlying rationale aligned with two strategic CAS objectives: (i) Objective 3: improving social service delivery to raise human development indicators, and (ii) Objective 4: addressing the development deficits contributing to fragility and conflicts in DRC’s Eastern provinces. The project likewise reflected goals in the DRC Systematic Country Diagnostic13 to address gender inequality and GBV. The project also sought to contribute to the twin pillars of the World Bank’s global strategy and to its Gender Strategy, particularly regarding human endowments (Pillar 1), removing employment constraints (Pillar 2), and women’s voice and agency and GBV (Pillar 4). Theory of Change (Results Chain) 9. In accordance with Project Appraisal Document (PAD) preparation requirements at the time, the project presented a proposed approach in the PAD rather than a formal theory of change. Figure 1 (see Annex 7) is the proposed project approach that was outlined in the PAD. The analysis under this ICR reconstructed a project theory of change to illustrate the key components, activities, outputs, and intermediate outcomes as they link to the PDO and long-term impact of the project (see Figure 2, Annex 7). As described in the PAD, the premise underlying the project’s proposed approach was that by i) focusing on holistic community mobilization, livelihoods, and gender transformative interventions, and ii) combining prevention with survivor care, the following positive outcomes will ensue: i) change in harmful attitudes towards GBV, ii) increased household decision-making power for women, iii) decrease in experience of violence, and iv) higher rates of survivors seeking care. Project Development Objectives (PDOs) 10. The Project Development Objective as articulated in the financing agreement was to increase in Targeted Health Zones: (i) the participation in Gender-Based Violence (“GBV”) prevention programs; (ii) the utilization of multi-sectoral response services for survivors of GBV; and (iii) in the event of an Eligible Crisis or Emergency, to provide an immediate and effective response to said Eligible Crisis or Emergency. Key Expected Outcomes and Outcome Indicators 11. The PDO also stated that GBV prevention programs require a sustained investment over time to reduce incidence and change social norms. As such, the project’s aim was to contribute to the longer-term goal of reducing GBV prevalence by focusing on a set of intermediate level outcomes, as outlined in the project’s three objectives. 12. The project’s three objectives and five associated PDO-level outcome indicators were as follows: 11 The NSCGBV focused on five areas: (i) protection from and prevention of GBV; (ii) ending impunity; (iii) security sector reform; (iv) assistance for victims of violence; and (v) data collection and mapping. 12 As noted in the PAD, at appraisal, the FY13-FY16 CAS was extended to FY18, covering the project preparation period, while the new DRC Country Partnership Framework FY22-FY26 was being finalized. 13 Published March 2018. Page 7 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 1) Increase the participation in Gender-Based Violence (“GBV”) prevention programs by measuring: i) Numbers of direct project beneficiaries (percentage of women); and ii) Percentage reported decrease in accepting attitudes towards GBV in targeted Health Zones; 2) Increase the utilization of multi-sectoral response services for survivors of GBV by measuring: i) Percentage increase in reported cases who receive access to multidisciplinary services, defined as at least two of the following: (medical, psychosocial, security, legal support and livelihoods support); ii) Percentage of eligible reported GBV14 cases who receive Post-Exposure Prophylaxis (PEP) Treatment within 72 hours; and iii) Percentage of implementing partners providing services to GBV survivors in line with defined quality standards; and 3) In the event of an Eligible Crisis or Emergency, provide an immediate and effective response to said Eligible Crisis or Emergency. 13. The project’s third objective was provided as standard language to reflect the inclusion of a CERC component in the project, so this objective was not associated with specific PDO indicators. The overall PDO was supported by thirteen intermediate results indicators. Components 14. The project was prepared while the previous GL GBV Project was under implementation, with a focus on taking forward key lessons learned regarding community-level engagement and decentralization of service delivery to offer access to care through government, specialized, and community-based structures. As such, while the GL GBV Project had focused largely on health interventions via the government health system and CoEs, with some integration of prevention and survivor response at community level, this project significantly expanded community-based engagement to widen access to care and implement a broad range of prevention activities, while retaining the expertise of CoEs and integrating technical partnerships with MoPH and MGFC. The project was therefore structured around four key pillars: i) engagement at community level; ii) engagement with specialized structures and the health sector; iii) institutional support, research, and monitoring and evaluation (M&E); and iv) crisis response capacity. 15. These pillars also aligned across the three PDOs listed above. The first and second pillars aligned, respectively, with the first and second objectives; the third pillar provided the qualitative and quantitative research and M&E data to support the PDO indicators; and the fourth pillar aligned with the third objective. The project was implemented in four provinces in Eastern DRC, consolidating gains made under the GL GBV Project in North Kivu and South Kivu, and expanding later in implementation to Tanganyika and Maniema (see Annex 11). All four provinces were considered zones affected by conflict and instability with high rates of GBV. FSRDC was the implementing agency for both project management and fiduciary responsibilities, in close technical partnership with the MoPH, until its dissolution in April 2023, when CSPP assumed project management (see below). Component 1: Gender-Based Violence prevention and integrated support for survivors at community level (at appraisal US$54.5 million; restructuring US$46,457,390; actual US$37,391,144)15 16. This component supported the delivery of an integrated package of GBV prevention activities and dedicated survivor assistance at the community level. A detailed protocol developed by FSRDC offered a guide to these activities, 14 Eligible GBV cases for PEP (provision of antiretroviral medicine following potential exposure to HIV) are generally cases of rape that are reported at a service provider within 72 hours of the incident. 15 Project actual numbers are current as of June 30, 2023 (from the FSRDC Progress Report, August 2023), as project disbursements and the final financial report had not yet been completed at the time of the ICR filing (project was still in its grace period for payments). Page 8 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) which were led by one Umbrella NGO per province16 and supported through partnerships with local NGOs and women’s CBOs. Prevention activities were centered around three pillars of interventions: i) community mobilization and behavior change; ii) livelihood interventions; and iii) interventions focused on norms transformation.17 Integrated survivor support was anchored by safe spaces within CBOs, which were used for awareness-raising, life skills, and livelihoods activities. CBO focal points were trained to provide psychosocial support and service referrals, including specialized mental health care using the NET approach (see Annex 9). Component 2: Response to Gender-Based Violence (at appraisal US$27.5 million; no change at restructuring; actual US$24,637,821)18 17. This component supported delivery of multi-sectoral assistance through integrated CoEs and health sector investment. Activities were organized around two sub-components: i) support to two specialized CoEs and previous partners under the GL GBV Project (Panzi Hospital and Foundation in South Kivu and Heal Africa in North Kivu); and ii) strengthening GBV response in the health sector. The project supported CoEs to deliver multi-disciplinary services, including medical, legal, and psychosocial survivor support, and to provide specialized services in decentralized one-stop centers in Maniema and Tanganyika. 18. The project organized three principal activities for the health sector: training for health care providers (HCPs) and community health workers, funding for health services, and purchase of emergency medication for survivors, including PEP treatment. In addition to training, the project undertook minor rehabilitation of health facilities, overall at a lower level than planned due to very high construction costs owing to the poor state of health infrastructure.19 The project also provided for funding to health structures, based on the existing Performance Based Financing (PBF) approach in North and South Kivu. Building on lessons from the GL GBV Project, funds would be paid based on assessments of quality of survivor services. This component proved to be one of the most challenging to carry out as the health sector did not regularly conduct the quality assessments and therefore was unable to be paid with the same regularity as planned. Component 3: Support to Policy Development, Project Management and Monitoring and Evaluation (at appraisal: US$17.9 million; restructuring US$12,561,309; actual US$10,407,534)20 19. This component supported three sub-components: i) institutional strengthening activities, ii) project management, and iii) all monitoring, evaluation, and research activities.21 Under the first sub-component, the project funded coordination responsibilities for FSRDC and undertook a comprehensive assessment of the MGFC’s national GBV database in line with global best practices. FSRDC with other partners revised the National Medico-Legal certificate and guidelines on its utilization; dissemination efforts were later halted by the FSRDC’s dissolution. Under the second and third sub-components, the project received financial management support, including third-party assistance, and funded the upgrading and roll-out of a robust management information system (MIS) then in use by the GL GBV Project. The project also funded a third-party process evaluation (Annex 10) and the last phase of an impact evaluation of NET, which was begun under the GL GBV Project and conducted by the World Bank's Africa Gender Innovation Lab (AFRGIL) (Annex 9). 16 For North Kivu, South Kivu, and Maniema Provinces. 17 The project also initially anticipated prevention activities to target the armed forces and highly vulnerable population groups, but these resources were not developed due to competing project priorities and resource constraints with the devaluation of the Congolese currency. 18 See note 15. 19 Ultimately, smaller health facility rehabilitation was undertaken under World Bank-financed health projects (PDSS and PMNS). 20 See note 15. 21 A targeting and roll-out strategy for GBV programming in the Kasaï and Equateur Provinces was also planned but ultimately not undertaken as priority was given to Eastern DRC. Page 9 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Component 4: Contingency Emergency Response Component (CERC) (at appraisal: US$0 million; actual US$0 million) 20. This component was designed to provide immediate response in the event of an eligible crisis or emergency and was a “zero-dollar” Contingency and Emergency Response Component. While the project considered activating the CERC component to extend project activities to Kinshasa during the COVID-19 pandemic, the CERC component was never formally triggered during project implementation. The project determined at the time that it was not necessary to trigger the CERC because the financing agreement permitted the project to expand to other geographic areas via restructuring; however, restructuring and expansion were ultimately not pursued (see below). B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) Revised PDOs and Outcome Targets 21. The PDOs and outcome targets remained unchanged throughout the project lifespan. Revised PDO Indicators 22. No PDO indicators were added or dropped, or formal changes made to the definition or targets of the original indicators. 23. During the project mid-term review (MTR) and the latter half of 2022, the project discussed revising the results framework as part of a proposed restructuring by deleting three intermediate results indicators (IRI) for which no data could be obtained from the provincial health directorates (within the MoPH): (i) percentage of essential medication (PEP, STI Treatment and Emergency Contraception) for which there was no stock out22 during the implementation period, (ii) percentage of availability of basic equipment at the health facility level in accordance with the project ’s quality checklist, and (iii) percentage of minor works at the health facility level in compliance with ESMF requirements. The project also discussed during this time revising low indicator targets that did not originally take into account geographic expansion and mass communication outreach as well as indicators formulated as a percentage change for which data were difficult to collect and analyze. Ultimately, once the project decided not to undertake a restructuring in late 2022 (see below), the timeframe for revising the results framework before planned closure in June 2023 was too short and became impossible upon dissolution of the FSRDC. The results framework was therefore left unchanged. In its final report, FSRDC did not report data in its indicator matrix for the above three IRIs as well as a fourth IRI linked with grievance management under Component 323 (percentage of grievances received by the project that are addressed in line with GRM quality standards). Revised Components 24. There were no major changes to any project components. Other Changes 25. The project was formally restructured once during its lifespan. On April 27, 2023, presidential Ordonnance 23/04824 approved the dissolution and liquidation of the FSRDC, which required the project to revise its implementation 22 Stock-out denotes a depletion of inventory in essential medication listed above. 23 This indicator was not included in the Borrower’s final indicator matrix, though the final progress report (January 2024) stated that 100% of all complaints had been resolved by the end of the project. 24 The Ordonnance 23/048 was published in the Official Gazette no. 10 on May 15th, 2023. Page 10 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) and institutional arrangements. Through presidential Ordonnance 23/049 dated on April 28, 2023, a new FSRDC was created with a new legal status, expanded objectives and larger mission. As a result of its dissolution, the Ministry of Finance, through the Project and Program Monitoring Unit (Cellule de Suivi des Projets et Programmes, CSPP), assumed responsibility for project implementation, and project closure was moved to September 30, 2023. The project considered an earlier restructuring in order to extend activities to Kinshasa (following the initial request to trigger the CERC component25), Ituri Province, and new Health Zones in North Kivu, South Kivu, and Tanganyika Provinces, including to extend activities under Component 1 to Tanganyika. The project determined in late 2022 that restructuring would not be possible due to operational constraints with continuing the project under previous World Bank safeguards policies and a currency devaluation for the DRC, which considerably reduced remaining project funds. 26. As a result of the restructuring following FSRDC dissolution, the project cancelled US$13.5 million in funds. The Government of the DRC requested the cancellation to account for funds that would not be used as planned following the FSRDC dissolution. The cancellation resulted in changes in the final budgets for Components 1 and 3 as well as in the associated disbursement categories. Rationale for Changes and Their Implication on the Original Theory of Change 27. The project design never diverged from the original theory of change (project approach described above, see Annex 7), as the components, PDO, and PDO Indicators remained constant throughout project implementation. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating 28. The relevance of the PDO is rated as High. The PDO remained at appraisal, during implementation, and at completion highly relevant to the country’s current development priorities, as outlined in the DRC CAS FY13-FY16, in effect at appraisal26, and the most recent DRC Country Partnership Framework (CPF) FY22-FY26. 29. The DRC CPF references the same key contextual considerations underlying project preparation, namely, high rates of GBV as well as the profound and harmful effects of pervasive violence on women and girls. The CPF observes that conflict-related sexual violence serves to compound other forms of GBV that women and girls experience and cites GBV as a specific driver of fragility27. The persistent gender gaps and entrenched social norms, which result in substantial gender disparities for women and girls, are also noted in the CPF. 30. The PDO is also reflected in and remains highly relevant to two of the three focus areas in the CPF: i) Focus Area 1: Strengthen stabilization efforts for reduced risk of conflict and violence, and ii) Focus Area 2: Strengthen systems for improved service delivery and human capital development.28 The PDO’s focus on participation in GBV prevention and 25 See para. 20. 26 See para. 8. 27 According to results of the 2021 DRC Risk and Resilience Assessment. 28 Focus Area 1 seeks to strengthen crisis resilience for vulnerable, displaced, and conflict-affected populations as well as to improve community interconnectedness and regional integration. Focus Area 2 seeks to increase access to basic services in health, education, and social protection as well as to address gender disparities and underlying norms and factors contributing to GBV. Page 11 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) utilization of multi-sectoral services in the context of providing effective emergency response to conflict-affected communities continues to be relevant to both focus areas. In addition, the economic support activities funded by the project are directly relevant to Focus Area 1, strengthening resilience for vulnerable and conflict-affected populations by providing women tools and skills for income generation. The CPF also specifically cites under Focus Area 2 the implementation of this project, ongoing at the time of its approval, as one of the relevant World Bank-financed operations. 31. In addition, the PDO is reflected in the CPF’s cross-cutting theme around gender, which seeks specifically to reduce gender inequities in the DRC while addressing GBV. The project’s geographic focus in Eastern DRC, in particular North and South Kivu, likewise continue to be CPF priority areas for interventions. While Maniema and Tanganyika Provinces, likewise selected under the project as conflict zones with high GBV prevalence, are not target areas under the current CPF, the project is listed as a relevant World Bank-financed operation. The project remains equally relevant at completion to the World Bank’s Gender Strategy 2016-2023: i) increasing access to health care for GBV survivors (Pillar 1), ii) improving women’s income generating opportunities (Pillar 2), and iii) investing in holistic GBV prevention and response (Pillar 4). In addition, addressing GBV is a core objective in both the new Regional Gender Action Plan for Eastern and Southern Africa (AFE) FY24-FY28 as well as the forthcoming Gender Strategy 2024-2030. 32. As at appraisal, the PDO likewise continues to be aligned at completion with the recently revised World Bank’s global strategy to end extreme poverty and boost shared prosperity on a livable planet. The strategy specifically cites the importance of enhancing the World Bank’s work on social inclusion, for which the empowerment of women and addressing GBV are key. The project similarly remains aligned with the World Bank’s Strategy for Fragility, Conflict, and Violence (FCV) 2020-2025, which specifically cites GBV and violence against children as forms of community violence that the strategy addresses. In addition, the FCV Strategy identifies GBV as a major threat to development and cites gender inequalities, of which GBV and SEA/SH are a significant manifestation, as an aggravating factor in FCV settings. Current IDA20 objectives to address drivers of conflict well as gender bias and GBV are also reflected in the project. 33. The PDO also remains highly relevant to the Government’s current NSCGBV and associated five -year National Action Plan, which were being revised during project preparation and were issued in June 2020. The revised strategy is focused on addressing all forms of GBV, in addition to conflict-related sexual violence, and is organized around seven principal axes, which align with the PDO and the project’s activities.29 B. ACHIEVEMENT OF PDOs (EFFICACY) Assessment of Achievement of Each Objective/Outcome 34. The project included two primary outcomes, which aligned with the activities under Components 1 and 2: 1) increased participation in GBV prevention programs, and 2) increased utilization of multi-sectoral GBV response services. The third objective concerning provision of immediate and effective crisis or emergency response was included for the CERC component; there were no indicators associated with the third objective and no basis for evaluating this outcome, since the CERC component was never triggered. 35. The table at Annex 8 (Figure 1) outlines achievements made under each of the PDO indicators for both outcomes. The project’s data were also enriched by i) a process evaluation (see Annex 10), with both qualitative and 29 Axes are as follows: i) GBV prevention, especially at community level, ii) women’s socio-economic empowerment, iii) socialization and education of youth about GBV, iv) security and protection with a gender focus, v) holistic multi-sectoral GBV survivor care, vi) justice and addressing impunity, and vii) monitoring and evaluation around GBV. Page 12 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) quantitative research components, to assess partner collaborations, community interactions, quality of survivor care, and attitudinal shifts, and ii) the completion of an impact evaluation for the NET intervention, which overall had very positive results in terms of improvement in survivor mental health outcomes and effective delivery by local CBO actors (see Annex 9). The process evaluation was intended to help the project examine the potentially positive effects of the wide-ranging prevention and community mobilization activities as a whole on community attitudes and behaviors. Objective 1: Increase participation in Gender-Based Violence (GBV) prevention programs 36. The efficacy rating of this objective is judged to be Substantial given that prevention participation levels far exceeded their targets and were associated with an important reduction in attitudes accepting of GBV. As of the end of the project, over 8.5 million community members across four provinces in the DRC, and 38 Health Zones30, benefited from wide-ranging project interventions – exceeding the project’s original target of 785,000 by over 1000%. The project also maintained an equilibrium in participation by men and women, as women remained at 52% of the overall beneficiary population (compared with a target of 50%). In addition, the project exceeded its original target for numbers of Twa beneficiaries reached from indigenous communities (47,485 endline vs. 30,000 target). 37. The project also benefited from a comprehensive process evaluation to assess the PDO indicator linked with attitudinal change. A Knowledge, Attitude, and Practices (KAP) survey was conducted with different groups within the community as part of the process evaluation to assess changes in attitudes and acceptance of GBV. Overall, in view of the 14-point reduction between survey baseline and endline, the study found that the project played a significant and positive role in the diminution in attitudes of acceptance due to the mobilization and prevention activities at community level. 38. Additional intermediate outcomes related to engagement with women’s CBOs also reflected high levels of participation in prevention activities. The partnerships with CBOs demonstrated very positive results for both prevention and response efforts at the community level. The project supported 391 women’s CBOs across four provinces. Over 9,000 CBO members were trained on business skills and income-generating activities (IGAs) to promote management and sustainability of their organizations, and 3,733 members were trained as GBV focal points to provide effective case management to GBV survivors, including a subset trained to provide NET. Over the course of the project, these focal points provided case management services to over 50,000 GBV survivors (out of nearly 79,000 total supported) – the clear majority of survivors assisted by the project – as a first point of entry. The process evaluation likewise found that CBOs were well-received within their communities and provided quality, locally accessible care to survivors. In addition, in terms of long-term prospects, the study found that CBO sustainability depended upon IGA viability as a future revenue stream, adequate time for appropriate IGA training and monitoring, close collaboration with local state structures, broad community support, and a solid exit strategy. Given the myriad of factors influencing CBO sustainability and short timeframe for project and study implementation, the process evaluation noted that it was difficult to provide a definitive response on the prospects for their continuity after project closure. 39. The economic support activities reflected an equally high level of participation and addressed a key skills and financial resources gap for women. The number of economic support beneficiaries totaled 87,711, far exceeding its original target of 3,800. These activities offered women a menu of training opportunities in literacy, numeracy, business skills, and IGAs, and opportunities to participate in Village Savings and Loan Associations (VSLAs). Nearly 21,000 women at the community level participated in literacy and numeracy activities. In addition, more than 1,500 women31 identified as vulnerable, including GBV survivors identified anonymously, received vocational skills training (in addition to literacy 30 FSRDC Final Progress Report, January 2024. Health zones are administrative sub-divisions of provincial health directorates in the DRC. 31 Project data (economic support specialist). Page 13 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) and business skills) to support them to undertake livelihoods identified to be the most profitable and sustainable, such as agriculture and animal husbandry. The process evaluation found that IGA beneficiaries and VSLA members reported a high degree of satisfaction with the activities, viewing them as a means to improve quality of life and financial security for members and their households, strengthen holistic survivor care, and foster social cohesion. 40. The project established 1,128 women-only VSLAs with 27,414 members in North Kivu, South Kivu, and Maniema provinces. Loans typically ranged from 10,000 Congolese francs (US$ 3.75 at the time of the project) to 100,000 Congolese francs (US$ 37.50)32 and were used for IGAs.33 The high degree of participation and satisfaction noted above meant that a large majority of VSLAs continued onto subsequent savings cycles with minimal project supervision and a number of VSLAs were likewise spontaneously created at community level without project support. At project closure, VSLA members indicated that they would continue the savings and loan cycle following the project VSLA model, though precise numbers were not available. Members of selected female VSLAs and their male partners – 57,439 total female and male participants – also took part in couple’s dialogue groups to improve communications and support the importance of shared decision- making and non-violence within the household.34 41. Community-level participation in specialized mental health care interventions offered through CBOs also exceeded planned targets. Intermediate level indicators tracking the total number of beneficiaries receiving specialized mental health care (10,503 endline vs, 3,200 target) and number of service providers trained on NET (265 endline vs. 60 target35), both significantly exceeded their targets. Most service providers trained on NET were CBO focal points (221 total), with fewer CoE and health care personnel trained; lessons learned under the GL GBV Project showed that NET implementation at community level had been very effective. In addition to the project outputs achieved under NET, the impact evaluation led by AFRGIL likewise demonstrated that NET was as effectively carried out by trained community- based non specialists as by trained health sector professionals (see Annex 9) – a key finding for specialized mental health care delivery in a complex FCV setting such as the DRC. 42. Finally, the project realized high participation levels in male engagement programming, implemented at community level via the Engaging Men in Accountable Practice (EMAP)36 intervention. In total, 1,158 male and female EMAP facilitators were trained during the project, and 58,545 male and female community members participated in the 32 Estimates from project VSLA data (September 2023 aide-memoire). 33 The project collected regular monitoring data on participation rates in economic support activities though did not consolidate data collected at provincial level on other facets of the intervention, such as participant savings, rates of repayment, etc. 34 For reference, VSLA and couple’s dialogue interventions with female VSLA members and their partners have been the subject of previous impact evaluations in the DRC and elsewhere: Bass J, Murray S, Cole G, et al., Economic, social and mental health impacts of an economic intervention for female sexual violence survivors in Eastern Democratic Republic of Congo, Global Mental Health (2016); doi:10.1017/gmh.2016.13; Gupta J, Falb K, Lehmann H, et al., Gender norms and economic empowerment intervention to reduce intimate partner violence against women in rural Côte d’Ivoire: a randomized controlled pilot study, BMC International Health and Human Rights (2013), http://www.biomedcentral.com/1472-698X/13/46. These interventions have been associated with positive results in terms of increased food consumption and reduction of stigma for GBV survivors (VSLA) as well as reductions in experience of violence and acceptance of violence (dialogue groups). 35 The results framework in the PAD indicated an original target of 60; this target was increased to 100 in FSRDC progress reports during implementation, though no change was formally made to the results framework. 36 EMAP is a male behavior change intervention, developed by the International Rescue Committee (IRC) and structured in the form of a 16-week curriculum for male discussion groups and an accompanying 8-week curriculum for female discussion groups; each set of groups is led by same- sex facilitators. The curriculum covers a wide range of topics concerning GBV, its root causes and consequences, gender norms, and household power dynamics. EMAP in DRC was the subject of an earlier impact evaluation led by AFRGIL with the IRC, which found that the intervention improved couples’ relationships, fostered positive changes in male behavior associated with IPV, and led to improvements to men’s gender equitable attitudes, though no reduction in GBV prevalence was found. Vaillant J, Koussoubé E, Roth D, et al. Engaging men to transform inequitable gender attitudes and prevent intimate partner violence: a cluster randomised controlled trial in North and South Kivu, Democratic Republic of Congo. BMJ Global Health 2020; doi:10.1136/bmjgh-2019-002223. Page 14 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) discussion groups. Qualitative feedback from EMAP participants37 suggested a high appreciation of the intervention by communities, with women reporting positive changes in men’s behavior in their communities and shifts in gender equitable attitudes. The project likewise trained 1,687 facilitators on SASA!, a community mobilization approach aimed at preventing violence against women and reducing HIV-risk behaviors38. The project was able to achieve a partial implementation of SASA! though it faced challenges to complete the full program cycle as it was difficult to locate SASA! trained facilitators who could speak French and adapt the approach to the DRC context.39 The SASA! facilitators nonetheless contributed to wider community mobilization activities, as part of the collective of prevention interventions, which were reflected in the process evaluation’s assessment of attitudinal change owing to these activities. 43. Notwithstanding these achievements in community-level participation under Objective 1, the supporting PDO indicator on direct project beneficiaries might have benefited from disaggregation. Numbers of direct project beneficiaries (the first PDO indicator) reflected a wide range of prevention and response activities, including large- and small-scale awareness-raising via radio and local activists, as well as services and livelihoods. As such, while the data showed a huge increase in participation, the total number was very high, owing to the inclusion of radio programming and geographic expansion, and did not benefit from disaggregation of the data for small-scale prevention and community mobilization activities. 44. While the second PDO indicator that was supported by the process evaluation’s KAP survey was an appropriate measure to capture attitudinal shifts, it might likewise have benefitted from further clarity in the calculation methodology. This PDO indicator achieved a 32.3% decrease40 in accepting attitudes towards GBV, surpassing the 20% target decrease (see Annex 8, Figure 1). The final project progress report, however, reported this indicator as the difference between baseline and endline numbers (14.1%) rather than the percentage decrease in acceptance of GBV as defined in the results framework. For the other two IRIs supported by the KAP survey, specifically, percent reported change in women’s participation in household decision-making and percent change in help-seeking behavior for men and women aware of IPV cases at community level, these indicators did not achieve the 20% target increase and were similarly reported. The results showed a 9.2% decrease in women’s decision-making and a small 4% increase in help-seeking behavior by community members (see Annex 8, Figure 2). 45. Regardless of the results for these two additional indicators under the KAP survey, the process evaluation observed that study results demonstrated that the project had had a positive effect on improving attitudes towards gender equality and recognizing appropriate help-seeking behaviors, even if behaviors themselves were slow to change. The process evaluation explained that changes in attitudes often precede changes in behavior; hence, the project had greater success with the indicator on attitudinal change rather than changes in decision-making and help-seeking behaviors. In addition, behavioral change evolves over a long time, and the project implementation period for prevention activities was shorter than originally planned (see Annex 10).41 Objective 2: Increase utilization of multi-sectoral response services for survivors of GBV 46. The efficacy rating of this objective is judged to be Substantial as all three PDO indicators regarding timely access to quality survivor care were achieved. The project met the first PDO indicator regarding percentage increase in access to multi-disciplinary care, showing an increase from 50% to 53% in survivors who accessed at least two services 37 The process evaluation collected feedback on EMAP as part of its qualitative study (see Annex 10). 38 The approach was developed by Raising Voices, a violence prevention NGO based in Uganda (see Annex 4, note 69). 39 See Mid-Term Review aide-memoire, May 2022. 40 Calculated as difference between baseline (43.7%) and endline (29.6%) divided by baseline value (14.1% / 43.7% = 32.3% change). 41 See paras. 63-65. Page 15 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) within a holistic care package of medical, psychosocial, security, legal, and livelihoods assistance. The project likewise exceeded the 80% targets for the other two PDO indicators under Objective 2 regarding access to PEP treatment and quality of care: 100% of eligible reported GBV survivors received PEP treatment within 72 hours, and 82% of implementing partners provided services to GBV survivors in line with quality standards. 47. While the PDO indicator regarding access to multi-disciplinary care was a successful and appropriate measure to show an increase in utilization, it could have been strengthened by providing further clarity in calculation methodology and disaggregation of data from underlying service delivery models. This indicator was defined in the results framework as a percentage increase in reported cases accessing at least two multi-sectoral services, though it was generally reported as the percentage of GBV cases accessing services as it was a difficult distinction for implementing partners conducting data collection at the local level. During the mid-term review, the project discussed reformulating the indicator to track percentage access rather than a percentage increase in order to address these challenges in reporting, but the project had difficulties revising the results framework as noted above.42 48. The project might likewise have benefitted from the disaggregation of data from two different service delivery models for CoEs and community-based partners, in order to see more contextualized results. Having been established service providers within their communities and under the GL GBV Project, the CoEs already enjoyed high rates of survivor access to at least two services because of the nature of their multi-sectoral service provision and longstanding presence. As such, it would have been difficult for the project to show a significant increase in access at CoEs since the rates were already high. By contrast, local structures, such as Umbrella NGOs and CBOs, did not enjoy the same high levels of access to care at project start-up, as they were more nascent service delivery channels, with many beginning implementation during the COVID-19 pandemic, and did not have the same one-stop model. By project end, however, CBOs had demonstrated clear success in supporting increased access to care at community level, with CBOs serving approximately two-thirds of all survivors assisted under the project as the first point of entry for services (see below). 49. The associated IRIs regarding access to and quality of care likewise supported increased utilization of services. Most notably, the number of reported GBV cases accessing at least one service exceeded the original target of 60,000 survivors by 131%, with the project serving 78,466 survivors in total. When disaggregated by point of service entry, as noted, CBOs were the most frequent point of entry, with 52,240 out of 78,466 (66.6%) total survivors assisted accessing care via CBO focal points. Other points of entry included CoEs, legal and mobile clinics, and health centers, with far fewer survivors entering through these service delivery channels. 50. In addition, three other IRIs associated with GBV response under Component 2 supported an increase in use of services. A total of 433 health personnel received training on GBV service provision (target of 400). In addition, 98% of survivors met regularly with their case manager, exceeding a target of 80%. Nearly half (46%) of rape cases accessed care within 72 hours of the incident (50% target). This result was not unusual given the significant social, economic, and logistical barriers that rape survivors face in accessing care within the 72-hour timeframe for PEP treatment.43 51. While three additional IRIs were reliant upon data from the health sector, which were extremely challenging to collect and monitor due to general data management weaknesses in the health system, the process evaluation helped to cover these gaps. As noted above44, these indicators for the health sector tracked i) inventory stocks of essential 42 FSRDC routinely measured this indicator against a target of 80% for access to at least two services, which facilitated data collection with implementing partners, but the results framework was never formally changed to reflect a target or to reformulate the definition. See paras. 23, 25. 43 FSRDC Final Progress Report, January 2024. 44 See para. 23. Page 16 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) medication; ii) health facility equipment that met quality benchmarks; and iii) small-scale works compliant with ESMF requirements. Notwithstanding these challenges in data collection from health facilities, the process evaluation provided detailed qualitative data on access to and quality of medical care, PEP kit availability, and set-up of health facilities for survivor assistance – all information measured under the indicators linked with the health sector. Objective 3: Provide immediate and effective response in the event of an Eligible Crisis or Emergency 52. As noted above, the language under Objective 3 was provided as standard text in relation to the CERC component under the project. As such, there were no indicators associated with this objective and no activities to measure as the CERC was never formally triggered. 53. One final IRI linked to the project management activities under Component 3 was developed to monitor the percentage of grievances received by the project that were addressed in line with quality standards under the GRM. The indicator was achieved by project closure as the final FSRDC progress report indicated that 100% of all complaints were resolved adequately by the end of the project. Justification of Overall Efficacy Rating 54. The overall efficacy rating of the project is judged to be Substantial as both PDOs to increase participation in GBV participation and utilization of GBV response were achieved and judged Substantial. The project either met or exceeded all five PDO indicators, as well as the majority of IRIs, and the project’s investments in community-level prevention and response activities, especially mobilization, livelihoods, NET, and services within CBOs, resulted in high rates of participation and utilization. C. EFFICIENCY Assessment of Efficiency and Rating 55. The efficiency of the PDO is rated as Substantial based on the results of the economic and implementation efficiency analyses provided below. The economic analysis overall yielded a positive result (see Annex 4 for the full analysis). Implementation efficiency is overall judged to be substantial on the grounds that, while the project faced considerable operational constraints and delays, it ultimately achieved its objectives in a very difficult context. Economic analysis 56. The economic analysis that was presented in the PAD reviewed the economic costs of GBV, from the point of view of direct costs related to health care and legal expenses and indirect costs related to the value of lost productivity. The project did not conduct a full-fledged economic analysis at appraisal given the absence of data on monetary costs related to service provision and effects of GBV. 57. It was not possible to conduct an economic analysis of the entire project, given that there were not sufficient data to monetize all project interventions, particularly for GBV response interventions. Therefore, the analysis estimated benefits that could be quantified, based on assumptions, and for which data were available, such as data regarding beneficiaries of community mobilization activities, VSLA participants, beneficiaries of mental health care interventions, and beneficiaries of PEP kits and other gynecological interventions. Benefits were estimated around four Page 17 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) outcomes under Components 1 and 2: i) the number of cases of sexual violence and physical violence averted resulting from community mobilization activities (past 12 months); ii) productivity gains for women engaged in IGAs; iii) number of cases of HIV transmissions averted from the administration of PEP treatment within 72 hours; and iv) improvements in mental as well as and sexual and reproductive health as a result of psychosocial and medical support interventions. 58. The economic analysis estimated three indicators outlined under ICR guidelines45: 1) the benefit-cost ratio (BCR); 2) the net present value (NPV); and 3) the internal rate of return (IRR). The BCR expresses the result as a ratio of benefits to cost (benefit divided by cost), where a BCR greater than 1.0 indicates that the benefits of an investment outweigh the costs. The NPV is the difference between benefits and costs and measures the net economic gain; a positive NPV indicates a net economic benefit. The IRR is the annual rate of growth that an investment is expected to generate.46 Project costs were allocated among three components47: GBV prevention (Component 1), GBV response (Component 2), and project management (Component 3). Actual costs for Component 3 were proportionally allocated to Components 1 and 2 (see Annex 4). 59. Overall, the project was successful in economic terms. The economic benefits for Components 1 and 2, both combined and individually, are presented below in Table 1. The overall BCR for both components was 7.0 for the low range estimate and 13.3 for the high range estimate. This suggests every dollar invested in the project yielded an economic return ranging from US$ 7.00 to US$ 13.30. The total investment in the project of US$ 72.4 million generated an overall economic benefit (NPV) that ranged from US$ 509 million to US$ 962 million. The overall IRR ranged from 21.5% to 29.5%, suggesting a positive rate of growth owing to the investments made under the project. 60. Component 1 benefits were estimated with a low and a high range to account for the lack of demographic data for women participating in economic empowerment activities. The benefits for women engaged in IGAs were estimated from age of enrollment throughout their remaining work-life (until presumed retirement at age 60). Without information about age for women participants, two ages at enrollment were used i) 25 years equating to 35 years (60-25) of economically productive life to estimate a “high” benefits range; and ii) 40 years equating to 20 years (60 -40) of economically productive life to estimate a “low” benefits range. 61. Prevention activities measured under Component 1 showed very positive results, especially those related to economic support. Response interventions related to medical and basic psychosocial support, however, showed less favorable results under the economic analysis; this could be partly explained by the high cost of service provision in the DRC, given poor infrastructure, lack of widely available and detailed data on the cost of providing survivor care, and the overall low rates of and barriers to survivor access. It should also be noted that, in accordance with best practices, prevention interventions cannot ethically be conducted without the availability of basic medical and psychosocial support; as such, the favorable results for the prevention activities under Component 1 would not have been possible without the associated interventions in response services under Component 2. In addition, as noted above, this analysis was unable to monetize all of the benefits from every response intervention as the project did not have available data. 62. It is important to note that the economic benefits presented above are likely to be underestimated, given the limited scope to measure the multifaceted impacts of the project. For example, the analysis did not incorporate the benefits of treating women’s physical injuries or improvements in well-being more broadly, nor did it include the wider benefits of engaging with men. Including these additional benefits would likely increase the BCR, NPV and IRR. 45 Bank Guidance for ICR for IPF Operations (effective December 9, 2021), Appendix G. 46IRR = (End amount / Starting amount)(1/n) – 1 https://learning.treasurers.org/resources/how-to-calculate-IRR 47 See paras. 14-20. Page 18 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Table 1: Economic benefits of project IRR (over 10 US $ BCR NPV US $ years) Total cost (components 1 and 2) 72,436,500 Total benefit (components 1 low & 2) 508,725,296 7.0 436,288,797 21.5% Total benefit (components 1 high & 2) 962,058,124 13.3 889,621,624 29.5% Cost component 1 43,664,820 Benefit component 1 Low 498,506,520 11.4 454,841,700 27.6% Benefit component 1 High 951,839,348 21.8 908,174,528 36.1% Cost component 2 28,771,680 Benefit component 2 10,218,776 0.4 -18,552,904 -9.8% Implementation efficiency analysis 63. The project faced considerable operational challenges during implementation, principally from procurement delays and external shocks. While FSRDC met the required technical conditions for effectiveness as planned by January 2019, the project was subject to a further six-month delay in effectiveness due to an extended political transition period after the country’s presidential elections at the end of 2018.48 While CoE and Umbrella NGO activities in North and South Kivu started fairly promptly after effectiveness, with CoEs by the end of 2019 and NGOs by early 2020, other contracts stalled for 12-18 months due to extended procurement processes.49 Activities for Maniema and Tanganyika provinces began in earnest about a year after effectiveness50 and were slow to implement in Maniema until the end of 2022 due to difficulties with the CoE and Umbrella NGO collaboration. FSRDC also experienced recurring staff vacancies, including for a few key technical specialist positions in health, GBV, safeguards, and M&E, as well as early coordination challenges. In terms of external shocks, the COVID-19 pandemic, Ebola response in North Kivu, insecurity in Eastern DRC, a volcanic eruption in North Kivu, and impassable roads likewise caused delays and rendered monitoring extremely difficult. 64. Notwithstanding these challenges, the project adapted its strategies and remarkably managed to continue activity implementation under significant pandemic and security constraints in a difficult FCV context. The World Bank and FSRDC established a close monitoring program to address activity and procurement delays, scheduling weekly or biweekly support calls to review issues and identify solutions; FSRDC carried out close field monitoring with implementing partners to address bottlenecks. Partners adapted by relying upon remote telephonic support for monitoring and observance of social distancing practices to allow for activities to advance. The NET implementing partners conducted remote support via telephone, increased local supervisor transport allowances for field visits and established small satellite offices to facilitate close field supervision. The research team for the impact evaluation likewise benefited from a light, flexible structure with locally based data collection teams with easier access to project zones. The process evaluation consultant also adapted to a shorter contract period by shifting the KAP survey endline to a later date to allow for more time between baseline and endline. Contracts for NET, the process evaluation, and the feasibility study were 48 It should be noted that presidential elections were continually delayed over a period of two years from end 2016 to end 2018; project preparation fell amidst a period of unfortunate political instability, which led to unexpected delays in effectiveness. 49 Implementation partners for NET and the process evaluation began their activities at the end of 2020 or in early 2021, and the CoE feasibility study consultant was ultimately hired in late 2021. 50 Maniema and Tanganyika CoE contracts were signed end of 2020 and the Umbrella NGO for Maniema recruited in June 2021. There was no Umbrella NGO recruited for Tanganyika. Page 19 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) successfully completed on time. FSRDC intervention and determined consortium leadership in Maniema also helped to overcome partnership disagreements and advance activities considerably during the last several months of the project. 65. These efforts resulted in a strong turnaround in implementation during the latter half of the project’s lifespan. At the time of the mid-term review in May 2022, disbursements were slow, at 53% in August 202251, approximately one year before planned project closure in June 2023. By project closure in September 2023, however, it had successfully attained a 91% disbursement rate.52 The compressed timeframe for implementation meant that considerable activity progress occurred in the last 12-18 months of the project. Nonetheless, the project managed to overcome these challenges and achieve its objectives without sacrificing implementation quality. The process evaluation demonstrated that the project’s collective prevention efforts successfully resulted in attitudinal changes and also that service quality and access were overall maintained; monitoring data for CBOs and NET impact evaluation results likewise showed success in assuring access to community-level care that is also effective at improving mental health outcomes. D. JUSTIFICATION OF OVERALL OUTCOME RATING 66. The overall outcome is rated as Satisfactory on the ground that PDO relevance is judged to be High, efficacy is judged to be Substantial, and efficiency is judged to be Substantial. E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 67. The project made significant contributions to building gender considerations into project design and closing gender gaps in implementation, notably through the project’s economic empowerment and community mobilization activities. As noted above, the project achieved particular gains by working directly with women’s CBOs as an anchor both for survivor support and safe spaces for livelihoods as well as community mobilization and GBV prevention activities. These partnerships showed positive results in promoting women’s voice, agency, and participation in decision-making structures. The EMAP intervention was another successful vehicle for addressing harmful gender norms and received overwhelmingly positive community feedback.53 One key EMAP element was a feedback mechanism allowing inputs from female community members to ensure that the needs and risks of women and girls remained central to men’s group discussions. 68. The project was also extremely proactive at addressing gender imbalance in implementation agency staff ranks. The core FSRDC project team began as one comprised of only male personnel and evolved by the end of the project to a team comprised of almost half female personnel. As part of these efforts, the project hired women in several key technical specialist roles and adopted flexible work conditions to attract and retain female staff. FSRDC was equally proactive at prioritizing female candidates and undertaking intentional headhunting efforts to identify female candidates. Institutional Strengthening 51 Project ISR, August 2022. As of the filing of the ICR, the difference between actual component costs and estimated costs at appraisal are reflected in the reallocation and cancellation of funds that happened at restructuring (see paras. 25-26); in addition, project actuals in the ICR are current as of June 20, 2023 (see note 15) and do not reflect all payments made with project funds, as the project was still in its grace period. 52 Project aide-memoire, September 2023; restructuring delayed final project payments, which are to be completed during the grace period. 53 See para. 42. Page 20 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 69. Institutional strengthening was a core element of the project and focused on three principal pillars: i) government entities, including FSRDC, ii) specialized technical providers, and iii) civil society and community-level structures. As a follow-on operation to the GL GBV Project, the project placed an important emphasis on continuing collaborations with the health sector as well as with the MGFC to strengthen its coordination and data management infrastructure. The project likewise worked with FSRDC as a government agency to strengthen staff and consultant skills regarding its fiduciary responsibilities, environmental and social safeguards, and technical GBV prevention and response interventions. 70. The project continued significant investments in two specialized technical CoEs. In addition to support for multi- sectoral holistic assistance and expansion of one-stop centers, the project funded feasibility studies for both CoEs to assess existing infrastructure, internal governance and fiduciary mechanisms, as well as quality of care. 71. The project also considerably expanded collaborations with community-based actors. Partnerships with CBOs were very successful in helping community-level structures reach more GBV survivors and decentralize service delivery. These partnerships also served as key anchors within the community for a broad range of prevention interventions. 72. While institutional strengthening was a key investment under the project, the project’s long-term footprint for working with government and specialized technical structures experienced important implementation challenges. The FSRDC was overall an effective fiduciary and disbursement partner until it was abruptly dissolved. The MoPH and MGFC faced significant governance and infrastructure challenges. Likewise, while the CoEs continued to demonstrate top-notch technical expertise, the feasibility studies documented internal capacity and fiduciary constraints. Mobilizing Private Sector Financing Not Applicable. Private sector funding was not utilized in this project. Poverty Reduction and Shared Prosperity 73. While the project did not include any measurements or indicators specifically related to poverty reduction, it was nonetheless aligned with the current World Bank DRC CPF’s goals of reducing poverty and boosting shared prosperity with a specific emphasis on addressing drivers of conflict. The project at closure continued to contribute to Focus Areas 1 and 2 under the World Bank DRC CPF FY22-26.54 The DRC CPF likewise focuses its support on provinces with displaced and conflict-affected populations, including North and South Kivu where this project was implemented. Other Unintended Outcomes and Impacts Not Applicable. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 54 See para. 30. Page 21 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 74. The project included at preparation clear project objectives that defined achievable goals around increasing participation in prevention programs and utilization of multi-sectoral services. Both objectives focused on elements of GBV prevention and response that a project could realistically hope to achieve within a traditional five-year project span, for instance, rather than setting unrealistic goals related to reduction in GBV prevalence or significant improvements in behavior or norms change, both of which require sustained, multi-faceted interventions. Acknowledging these constraints, the project was clear at preparation that it was seeking to contribute to the long-term goal of reducing GBV by focusing on intermediate-level objectives. Project design also included clearly structured components focused, respectively, on GBV prevention, survivor response, and project support, drawing directly from lessons learned under the GL GBV project to expand decentralization and continue to engage the health sector. 75. The project developed a detailed results framework and a comprehensive M&E plan. In general, the PDO indicators were aligned with the operational objectives under the project. The project’s M&E plan also anticipated significant investment to strengthen the project’s data management systems and improve the quality and reliability of project data. The M&E plan provided for regular progress reports from FSRDC and implementing partners and also included a process evaluation as well as an impact evaluation to assess project interventions (see Annexes 9-10). 76. The project also relied upon experiences and lessons learned under the GL GBV Project when designing its institutional and implementation arrangements. Notably, the project was able to rely upon the same implementing agency, FSRDC, and PIU, drawing upon their fiduciary and project management capabilities and institutional memory. The project also streamlined partnerships by consolidating community-based partners under the Umbrella NGO structure and directing complex cases to the CoEs. The project continued the existing collaboration with the health sector, working with health personnel on service delivery and reducing fiduciary risk by disbursing funds through the FSRDC. 77. Project risks and associated mitigation measures were identified during project preparation and assessed as high, with the possibility to reduce to substantial risk with effective implementation of proposed mitigation measures. The primary risks were outlined as follows: i) macro-economic risk, ii) governance and security challenges, iii) need for strong coordination, iv) capacity and fiduciary constraints, and v) limited safeguards risks. The project responded by disbursing funds in a stable currency (the US dollar), working with established providers and reducing the number of partner contracts, hiring appropriate personnel, and putting into place measures for fiduciary oversight. B. KEY FACTORS DURING IMPLEMENTATION 78. Project implementation progress was rated Satisfactory and Moderately Satisfactory throughout the lifespan of the project. A variety of factors informed the project’s ratings during implementation. Factors subject to the control of government and/or implementing entities 79. The project benefited from the presence of an overall robust staff organization within FSRDC. The continuation of FSRDC as implementing agency from the GL GBV Project, especially at senior project management level, offered helpful continuity during project implementation. FSRDC engaged a range of technical and fiduciary specialists, as well as a strong M&E team at national and provincial levels. FSRDC, however, did experience recurring staffing gaps, as noted above,55 especially for social and environmental specialists at provincial level as the project employed only one social and environmental specialist at national level. 55 See para. 63. Page 22 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 80. The project also invested significantly in updating the M&E data management system from the GL GBV Project. The project simplified tools, streamlined data collection points, and developed a national level compilation system that displayed current results and in-depth analysis in an online dashboard. The dashboard resolved previous errors in results calculations and rendered project results and data more accessible to all project actors, including the World Bank. 81. The project experienced delays, especially at the beginning, in completing procurement of several contracts, including for key implementing partners, the process evaluation and feasibility study consultant firms, and NET implementation consultants. Several contracts were delayed for 12-18 months after effectiveness, and disbursement was therefore slow, just above 50% in August 2022 after the mid-term review held in May.56 82. The project also experienced challenges regarding multi-sectoral collaborations with MoPH and MGFC. These elements of the project were among the most difficult to implement, due to inter-sectoral coordination challenges and delays with both ministries in the signing and implementation of partnership agreements. Payments to health structures for services were also very delayed due to difficulties in obtaining proper funds justification. Factors subject to the control of the World Bank 83. The project benefitted from the presence of a comprehensive World Bank team to support project implementation on a regular and consistent basis, including the timely support of the World Bank’s technical and fiduciary teams. The task team was well-staffed to support FSRDC to implement technically sound programming and fulfill its fiduciary responsibilities, with a range of experienced operations staff, safeguards and fiduciary experts, as well as technical specialists with significant experience in the DRC context and with GBV prevention and response programming. 84. The World Bank team offered frequent implementation support, organizing regular monitoring meetings and missions. The team conducted a total of twelve implementation support missions with an average of two missions per year, including three virtual missions during the COVID-19 pandemic when support was especially critical. During the pandemic, the team conducted weekly calls to assure close supervision when in-person visits were not possible and at least biweekly support calls throughout project implementation. At the end of every mission, the team regularly reviewed with FSRDC and CMU leadership mission findings and issues for management attention. The task team maintained a positive relationship with FSRDC, which allowed for effective and candid communication key for both implementation and problem-solving. The World Bank team produced comprehensive aide-memoires for each mission, reporting on matters pertaining to fiduciary and safeguards compliance, as well as progress updates on technical implementation and M&E. 85. Notwithstanding a strong base of expertise and regular implementation support in an extremely challenging context, the team faced gaps in documenting support. The team struggled somewhat to make aide-memoires available for this ICR57 and produced limited ISRs over the course of the project, with four in the first 24 months of the project and two between 2021-2022 (the last ISR issuing approximately one year before project closure). This gap in ISR submissions at the end of the project may be attributed to the dissolution of the FSRDC in April 2023. This ICR analysis also notes that, with FSRDC dissolution and delay in project closure, regular M&E updates were difficult to provide. This unfortunately meant that the M&E dashboard and results from the process evaluation were unable to be updated and discussed with FSRDC regularly, affecting reporting of certain indicator results (see section on Efficacy above). 56See paras. 63-65. 57Not all aide-memoires were timely filed on the operations portal or in team archives; the aide-memoire for the December 2021 mission was provided in draft form only. Page 23 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Factors outside the control of government and/or implementing entities 86. The project experienced several external challenges during implementation, owing especially to the COVID-19 pandemic and deteriorating security conditions in Eastern DRC. The conflict, including the “state of siege” and resurgence of armed group M-23 in North Kivu in 2021, and violent protests in Eastern DRC constrained both the FSRDC and World Bank teams from providing regular on-the-ground supervision. The FSRDC team also experienced difficulties with project implementation and supervision due to dilapidated roads in some project zones, the Ebola outbreak in certain areas of North Kivu in 2019-2020, and the eruption of the Nyiragongo volcano outside of Goma, North Kivu, in 2021. 87. In addition, the onset of the global COVID-19 pandemic in March 2020 created significant challenges for both FSRDC and World Bank teams. The pandemic resulted in significant travel restrictions and social distancing requirements, which made in-person implementation support missions extremely difficult or impossible for a period of time. In addition, pandemic constraints also resulted in a significant slowdown of project activities and supervision challenges for both the FSRDC and World Bank teams. Accordingly, as noted, the FSRDC and World Bank teams conducted weekly calls, especially during the pandemic, to assure close support when in-person visits were difficult to organize. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 88. The PDOs for prevention and response were well-developed and reflected logical links as intermediate-level outcomes in support of the long-term goal of reducing GBV incidence. The project’s proposed approach (see Annex 7) likewise drew logical links between planned activities and the barriers to reduction in GBV incidence that the project sought to address, though it did not define as explicitly the links between project inputs, outputs, and outcomes. 89. Project indicators were relevant and supported PDO achievement. Under the first PDO objective, as noted above,58 while the indicators could have been strengthened in different ways, both indicators around attitudinal change and numbers of project beneficiaries, and associated IRIs, were appropriate to demonstrate an increase in participation in GBV prevention. The indicators were also linked to the prevention and community mobilization activities outlined under Component 1 in the project’s approach. Similarly, while the indicator for the second objective regarding access to multi- disciplinary services could have been strengthened,59 it was appropriate to monitor progress in access to and quality of care, as were PDO indicators regarding PEP treatment and quality of care, as well as associated IRIs. The indicators were also linked with the response interventions for GBV service providers, including CoEs, and the health sector outlined under Component 2 of the project. With respect to difficulties in receiving health facility data, the process evaluation served as a helpful means of evaluating access to and quality of medical care provided, which helped to account for information from the health sector where data could not be collected in a difficult environment (see Annex 10). 90. The project consolidated limited monitoring data in relation to economic support or male engagement activities. As such, the results of those activities, which otherwise proved to be successful in terms of high participation 58 See paras. 43-44. 59 See paras. 47-48. Page 24 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) and for which monitoring data were collected at provincial level, were not as comprehensively reflected in overall project reporting. M&E Implementation 91. Project design incorporated a separate M&E component and allocated resources to upgrade the existing MIS in use under the GL GBV Project and to conduct a process evaluation to monitor quality of support services (see Annex 10). As noted, the MIS was rehauled to simplify and streamline the data collection and reporting process and establish a comprehensive dashboard.60 While the FSRDC M&E team was technically strong, they relied heavily upon World Bank M&E specialists to manipulate the dashboard effectively, affecting sustainability of the system. 92. The project also developed a detailed indicator matrix, which outlined all PDO, intermediate results, and sub- indicators needed to respond to the PAD indicators, as well as definitions, calculation methodology, and responsible parties. This process was accompanied by training for partners on a standard set of indicator definitions and data collection methodology, to ensure that data points were comparable and of good quality. A strong M&E team at national and provincial levels also accompanied partners closely in data collection and reporting. 93. Project data were regularly collected and reported during implementation, and the process evaluation provided a rich, additional source of qualitative and quantitative data on services, CBO engagement, and attitudinal change. The project regularly recorded indicator data in the ISRs as well as aide-memoires and maintained a satisfactory or moderately satisfactory rating for M&E during implementation, even with COVID-19 pandemic constraints. The project addressed challenges with overly complex reporting tools by working with partners to simplify the collection tools and eliminate superfluous data points. While the project had difficulties collecting data for health sector indicators due to system weaknesses in the health care system, the qualitative data collected during the process evaluation on medical care helped serve to cover gaps in health sector data around quality of and access to care and medications as well as set-up of health facilities for survivor assistance. As noted above, the project, however, had difficulties in revising the results framework to adjust for low targets or challenges in collection or reporting once restructuring did not move forward.61 M&E Utilization 94. The process evaluation examined in detail quality of GBV services, community attitudes towards GBV, and sustainability of women’s CBOs as providers, which allowed for course correction in implementation. Data from the process evaluation were used during the mid-term review to advocate for additional resources for CBOs, including ongoing training needs and materials for GBV focal points. It was also used to improve access to care at the CoE in Tanganyika. 95. M&E data were likewise used throughout the life of the project to inform implementation and introduce course corrections where needed. Data were used to continue support for CBOs and incorporate them in expansion of services, given their broader geographic coverage and the diverse types of GBV for which CBOs offered support. Justification of Overall Rating of Quality of M&E 60 See para. 80. 61 See paras. 23, 25. Page 25 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 96. The overall rating for M&E is Substantial. The overall M&E system reflected indicators that were aligned with the PDOs and appropriate to assess their achievement. Project M&E was also supported by a robust data management system with trained staff who provided close technical support to partners. The M&E system was also accompanied by a detailed, well-executed process evaluation, which allowed the project to make informed decisions regarding activities during implementation and also to strengthen information on quality of services, including from the health sector, where information was not always forthcoming. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 97. The project’s compliance with environmental and social safeguards was rated as Satisfactory and Moderately Satisfactory over the course of the project. At appraisal, the project was rated as Category B given limited environmental and social risks that could be mitigated and adequately managed by the project. The environmental risks under the project were due to planned minor rehabilitation works and small-scale livelihoods activities; the social risks under the project were largely owing to the presence of indigenous peoples in project zones. Three environmental and social safeguards policies were triggered: Environmental and Social Assessment (OP/BP 4.01), Pest Management (OP/BP 4.09), and Indigenous Peoples (OP/BP 4.10). An Environmental and Social Management Framework (ESMF) as well as an Indigenous People’s Policy Planning Framework (IPPF) were prepared, cleared, and disclosed by the Borrower prior to project appraisal. The project also developed a grievance redress mechanism (GRM) to manage potential complaints, including those that might arise from indigenous communities; the project also implemented mitigation measures to manage SEA/SH risks, including complaints. All grievances were resolved satisfactorily by the end of the project.62 Owing to the staffing shortage noted above, project site risk screenings were delayed, and GRM implementation experienced some challenges, including management of local grievance committees by partners rather than the project and gaps in community sensitization. Notwithstanding these gaps, the safeguards assessment carried out during the project closing mission in September 2023 reported no major outstanding issues of non-compliance. 98. The project’s financial management (FM) performance rating was consistently Moderately Satisfactory, and the risk rating at appraisal was Substantial. With FSRDC dissolution in April 2023, however, the risk on the final FM supervision was rated High with respect to the institutional arrangements modification, selection of the CSPP, an agency with project oversight experience but no implementation experience, and uncertainty created by the abrupt dissolution. External audits were conducted annually for 2019-2022 and issued as unmodified; as of the filing of this report, the external audit for 2023 is due in June 2024. In addition, an external financial management agent was contracted to provide additional FM support on internal audits to the FSRDC in the early stages of the project. During implementation, the project was overall in compliance with FM rules and policies. The principal concerns related to delays in lifting disbursement conditions to finalize timely documentation for health structure payments,63 lack of adequate supporting documentation from partner NGOs, and delays in payments at project closure. The FM team worked to address NGO gaps by helping to ensure consistent processes to verify completion of activities against NGO payment requests. With the FSRDC dissolution, the project encountered delays in payments to implementing partners, which have been addressed by extending the grace period after project closure. 99. The project’s procurement performance rating was consistently Moderately Satisfactory, and the risk rating was Substantial. The World Bank’s procurement team conducted regular support missions. Procurement arrangements set forth in the project’s legal agreement were functional throughout the project, and the project was in compliance with 62 This includes SEA/SH grievances, which were resolved by FSRDC and were pending final administrative closure by the World Bank team at the time of the ICR filing. 63 Final payments were still pending as of the filing of the ICR. Page 26 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) procurement rules. The FSRDC project team included a Procurement Specialist at national level and Procurement Officers in each of the four provinces. CoE and Umbrella NGO performance was satisfactory due to mitigation strategies established following the GL GBV Project to ensure that payments to CoEs, Umbrella NGOs, and other partners were linked to execution of activities. The FSRDC dissolution as well as difficulties in obtaining exemptions from taxes and customs duties hindered the completion of some contracts. In the future, task teams should consider contracts to be paid inclusive of all taxes to avoid delays. C. BANK PERFORMANCE Quality at Entry 100. At the time of project preparation, the project team actively drew from the experiences under the GL GBV project to inform project design and streamline implementation arrangements. As noted above,64 the World Bank team took advantage of existing FSRDC expertise to lead this project. The team adhered to a timely project preparation schedule, identifying relevant risks and incorporating appropriate design elements and mitigation measures, including in its M&E plan. The team was sufficiently staffed with a range of operations, fiduciary, and technical experts. The rating of World Bank performance for quality at entry is judged to be Satisfactory. Quality of Supervision 101. During implementation, as noted above, the World Bank team provided frequent and comprehensive implementation support and adapted to challenges during the COVID-19 pandemic. Notwithstanding the team’s dedicated support, project documentation demonstrated certain gaps as noted.65 The rating of World Bank performance for quality of supervision is judged to be Moderately Satisfactory. Justification of Overall Rating of Bank Performance 102. Based on the above ratings for individual dimensions of World Bank performance (Satisfactory for quality at entry and Moderately Satisfactory for quality of supervision), the overall Bank Performance rating is Satisfactory. This rating is based on the overall level of the World Bank team’s support to FSRDC on technical design, M&E, and fiduciary responsibilities during preparation as well as the task team’s consistent dedication to regular and thorough implementation support despite observed documentation gaps. D. RISK TO DEVELOPMENT OUTCOME 103. Notwithstanding the satisfactory performance of the FSRDC and World Bank teams in the achievement of project outcomes, the risk to development outcomes is Substantial. The principal risk to sustainability of the project’s development outcomes is the institutional environment under which the project was prepared and implemented. Drawing upon the GL GBV project experience, this project relied upon the project management expertise of the FSRDC, given the coordination and fiduciary challenges with the MoPH. Local NGOs and CBOs played a greater role in improving community- level accessibility of services and participation in prevention activities, while survivor support was maintained through collaborations with CoEs and the health sector. In this regard, the project performed very well. 64 See para. 79. 65 See para. 85. Page 27 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 104. Reliance, however, upon the FSRDC as implementing agency and private entities as service providers meant that project resources were directed to parallel structures operating outside of the traditional government sphere. Collaborations with the MoPH and MGFC faced challenges given poor fiduciary and coordination mechanisms as well as weak personnel and physical infrastructure. Nonetheless, while FSRDC was a more reliable fiduciary and disbursement partner, the footprint for building government capacity as a long-term partner remained limited in scope. The FSRDC was itself unfortunately subject to DRC’s political landscape with its abrupt dissolution, prompting its restructuring and a delay in project close-out. Further, despite CoE technical expertise in complex service delivery, CoEs displayed gaps in infrastructure and governance as project feasibility studies demonstrated.66 105. In addition, coordination between government and civil society structures, including at the local level, was a challenge. While the project supported regular coordination meetings for all partners, the collaborations with the MoPH and MGFC proved to be difficult, including at the provincial and local levels. MoPH and MGFC were slow to develop operational action plans for their designated activities, and the project was slow to procure equipment under the project for the MGFC; it was also difficult to work with the MoPH to complete facility rehabilitation and provide health center payments. At the provincial and community levels, local MGFC offices and community health focal points did not always have consistent links with the local NGOs and CBOs, which made coordination difficult. The project was nonetheless successful in collaborating with the MGFC on the national GBV database and information dissemination and likewise with training health care personnel. The sustainability of holistic GBV prevention and response interventions necessarily depends upon a cohesive and stable partnership between government and civil society to ensure the broadest coverage for service delivery and community mobilization. Strengthening these links is critical to the long-term outlook of delivering quality and timely survivor care and changing harmful norms and behaviors. V. LESSONS AND RECOMMENDATIONS 106. Lessons learned. Based on the preceding analysis, several lessons learned from the operation are presented below. 107. Lesson 1. Heavy reliance upon a sole independent government agency for implementation, such as the FSRDC, as well as specialized technical structures, limited the project’s scope for strengthening fiduciary and technical capacity of traditional ministry structures for future sustainability of GBV prevention and response programming. The project relied upon the FSRDC as fiduciary partner since earlier implementation with the MoPH had been problematic. These same difficulties were likewise reflected in challenges in carrying out the health sector partnership under this project. Nonetheless, as a result, the project suffered from a lack of appropriation by MoPH and MGFC. While FSRDC and CoEs were, respectively, strong fiduciary and technical partners, FSRDC proved itself vulnerable to the DRC political sphere, and CoEs themselves showed gaps in capacity and management with regard to their possible expansion. As such, future similar operations could employ mixed institutional arrangements whereby a solid fiduciary partner in government manages the overall project with MoPH and MGFC undertaking broader technical roles, with limited fiduciary responsibilities. Government actors would likewise benefit from significant investments in training and skills development, both technical and fiduciary; partnerships with the CoEs and civil society would be key to expanding access to care and training government actors on key technical interventions. 108. Lesson 2. Engagement with women’s community-based structures provided the project with a local base for launching a wide range of community mobilization and economic support activities and also served as an effective bridge between prevention activities and survivor support. The process evaluation demonstrated that women’s CBOs can 66 See paras. 70, 72. Page 28 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) continue to operate as independent structures in the long term provided that they receive appropriate skills strengthening support, develop sustainable IGAs, and benefit from broad community support. Participants in economic support activities overall expressed a high level of satisfaction, seeing VSLAs and IGAs as important elements in women’s empowerment and sustainability of CBO activities. 109. Lesson 3. Decentralization of service delivery and referral pathways to local NGOs and women’s CBOs rendered survivor care much more accessible. Project data demonstrated that survivors accessed care through CBO focal points at much higher numbers than any other entry points. Positioning caregivers at multiple levels, through government, multi- sectoral response centers, and local CBOs, allowed survivors to access care quickly and be referred as needed. 110. Lesson 4. Specialized mental health care interventions, such as NET, can be delivered as effectively by trained community-based non specialists as technical health professionals. The impact evaluation for NET demonstrated that trained CBO focal points in conflict-affected settings were as effective at delivering the therapy as trained health care professionals. This approach allowed the project to assist more survivors through non-specialists who would remain in the community. 111. Lesson 5. Lack of consolidated monitoring data for economic support as well as male engagement and community mobilization activities hampered the project’s ability to report on these activities as a whole. The project benefitted from its design flexibility to be able to respond to the high demand for VSLAs from the communities and invested resources to implement EMAP and other community mobilization activities; it accordingly collected associated monitoring data from partners at provincial level. Given the significant expansion of services and increase in participation in these activities, however, the project would have benefitted from consolidating locally reported monitoring data to track and provide an overall picture of activity performance for these various prevention and mobilization interventions. 112. Recommendations. Based on the lessons learned outlined above, this report also provides the following recommendations that may be applied to future GBV prevention and response operations in the DRC or in other settings. 113. Recommendation 1. GBV prevention and response operational engagements should consider the complementarity of institutional partners and how they will best interlink, especially ministry actors, formal structures and specialized service providers, and local and community-based actors. GBV prevention and response operations are naturally multi-sectoral endeavors. Especially in FCV contexts, there is unlikely to be a government actor wholly able to serve as both an agile fiduciary and technical partner. Projects will need to work with both formal and informal civil society actors to complement the government’s role. Projects should likewise invest in critical collaborations with community- based actors, especially to foster their sustainability at local level. In addition, while acknowledging the challenges, operations should continue to invest in strengthening the project management, fiduciary, and technical skill sets of government actors, even on an iterative basis. Operational engagements in education and health offer such potential entry points to work with government and civil society to improve access to and quality of holistic survivor care and to continue work on GBV prevention, especially in community and school settings. In addition, engagements to address women’s economic empowerment and leadership initiatives also offer opportunities to strengthen women’s access to and control over resources and participation in decision-making roles. 114. Recommendation 2. Women’s community-based structures should be engaged as a local platform to offer both GBV prevention and response, especially where services are sparse. Services must be made locally available when implementing any prevention activities. As such, members may be trained as focal points to offer quality, locally available psychosocial and specialized mental health care, which serve as critical anchors for prevention and livelihoods Page 29 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) interventions. These organizations also serve more generally as safe spaces, reducing stigma for survivors and promoting broad community engagement in their work. In addition, to ensure that community-based structures can operate independently after project closure, investment in sustainability during implementation must be a priority. This includes strengthening management capacities through a robust and tailored capacity-building plan; ensuring that organizations own their own physical operating space; fostering collaboration with existing state actors, especially health; and developing a well-structured exit plan. 115. Recommendation 3. GBV prevention and response programming should feature economic support activities for women that is a standalone activity and also incorporated into project design and M&E activities. These activities should include livelihoods, IGAs, VSLA programming, and a range of training opportunities. They will benefit from adequate resource allocation, and the project will be able to monitor activity outcomes when they are integrated into the M&E system. These activities are key to sustainability as they offer both women’s community-based structures an ongoing revenue stream as well as vulnerable women and GBV survivors socio-economic support. 116. Recommendation 4. Future GBV prevention and response operations should develop well-defined and relevant indicators for all interventions and avoid indicators that may be at risk of different interpretations or reporting difficulties. Incorporating monitoring indicators, and consolidating the relevant data, that track the progress of a particular intervention, especially those related to violence prevention and economic empowerment, allow a project to measure more comprehensively how well an activity is functioning overall for the project. Accordingly, projects should ensure that sufficient resources are allocated for M&E specialists and other livelihoods and prevention specialists to ensure that these data are consolidated and integrated into regular project-level reporting. . Page 30 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: Increase participation in Gender-Based Violence prevention programs Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage reported Percentage 0.00 20.00 32.30 decrease in accepting attitudes towards GBV in 29-Jun-2018 30-Jun-2023 21-Jul-2023 targeted Health Zones Comments (achievements against targets): Note that end result reports percentage decrease. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Numbers of direct project Number 0.00 785,000.00 8,614,146.00 beneficiaries (percentage of women) 29-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Page 31 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Female beneficiaries Percentage 0.00 50.00 52.00 28-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Direct project beneficiaries is Number 0.00 30,000.00 47,485.00 Twa communities 28-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Objective/Outcome: Increase utilization of multi-sectoral response services for survivors of Gender-Based Violence Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage increase in Percentage 50.00 0.00 53.00 reported cases who receive access to multidisciplinary 29-Jun-2018 30-Jun-2023 30-Jun-2023 services, defined as at least Page 32 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) two of the following (medical, psychosocial, security, legal support and livelihoods suppo Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of eligible Percentage 13.00 80.00 100.00 reported cases of eligible GBV who receive Post 29-Dec-2017 30-Jun-2023 30-Jun-2023 Exposure Prophylaxis (PEP) Treatment within 72 hours. Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of implementing Percentage 0.00 80.00 82.00 partners providing services to GBV survivors in line with 28-Jun-2018 30-Jun-2023 30-Jun-2023 quality standards Comments (achievements against targets): Page 33 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) A.2 Intermediate Results Indicators Component: Gender-Based Violence prevention and integrated support for survivors at community level Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Numbers of beneficiaries Number 0.00 3,800.00 87,711.00 participating in community level economic support 29-Jun-2018 30-Jun-2023 30-Jun-2023 services Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of beneficiaries Number 0.00 3,200.00 10,503.00 receiving specialized mental health care 28-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Page 34 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Number of service providers Number 0.00 60.00 265.00 trained in NET 29-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent reported change in Percentage 0.00 20.00 9.20 women´s participation in household decision-making 28-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Note that end result reports percentage decrease. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent change in help Percentage 0.00 20.00 4.00 seeking behaviour for women and men aware of 31-Mar-2021 30-Jun-2023 30-Jun-2023 IPV cases at community level Comments (achievements against targets): Page 35 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Note that end result reports percentage increase. Component: Response to Gender Based Violence Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of health personnel Number 0.00 400.00 433.00 receiving training on GBV service provision 29-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of reported cases of Number 0.00 60,000.00 78,466.00 GBV that access at least one service supported by the 29-Jun-2018 30-Jun-2023 30-Jun-2023 project (disaggregated by entry point) Comments (achievements against targets): Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Formally Revised Completion Page 36 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Target Percentage of rape cases Percentage 0.00 50.00 46.00 that access services within 72 hours of the incident 29-Jun-2018 30-Jun-2023 30-Jun-2023 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of beneficiaries Percentage 0.00 80.00 98.00 who meet regularly with their case manager, as 29-Jun-2018 30-Jun-2023 30-Jun-2023 defined in the project manual Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of essential Percentage 0.00 20.00 0.00 medication (PEP, STI Treatment and Emergency 29-Jun-2018 30-Jun-2023 30-Jun-2023 Contraception) for which there was no stock out during the implementation Page 37 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) period Comments (achievements against targets): No data were able to be obtained from the Ministry of Public Health. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage availability of Percentage 0.00 80.00 0.00 basic equipment at health facility level in line with the 28-Jun-2018 30-Jun-2023 30-Jun-2023 project´s quality check-list Comments (achievements against targets): No data were able to be obtained from the Ministry of Public Health. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of small-scale Percentage 0.00 100.00 0.00 works at health facility level complying with ESMF 29-Jun-2018 30-Jun-2023 30-Jun-2023 requirements Comments (achievements against targets): Page 38 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) No data were able to be obtained from the Ministry of Public Health. Component: Support to Policy Development, Project Management and Monitoring and Evaluation Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of grievances Percentage 0.00 100.00 100.00 received by the project that are addressed in line with 28-Jun-2018 30-Jun-2023 30-Jun-2023 quality standards defined in the GRM manual Comments (achievements against targets): Page 39 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1 1. Percentage reported decrease in accepting attitudes towards GBV in targeted Health Zones Outcome Indicators 2. Numbers of direct project beneficiaries (disaggregated for percentage of women and number of Twa populations) 1. Numbers of beneficiaries participating in community level economic support services 2. Number of beneficiaries receiving specialized mental health care 3. Number of service providers trained in NET Intermediate Results Indicators 4. Percent reported change in women’s participation in household decision-making 5. Percent change in help-seeking behavior for women and men aware of IPV cases at community level 1. Numbers of beneficiaries participating in community level economic support services 2. Number of beneficiaries receiving specialized mental health care Key Outputs by Component 1 3. Number of service providers trained in NET (linked to the achievement of the Objective/Outcome 1) 4. Percent reported change in women’s participation in household decision-making 5. Percent change in help-seeking behavior for women and men aware of IPV cases at community level Key Outputs by Component 3 1. Percentage of grievances received by the project that are addressed (linked to the achievement of the Objective/Outcome 1) in line with quality standards defined in the GRM manual Page 40 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Objective/Outcome 2 1. Percentage increase in reported cases who receive access to multidisciplinary services, defined as at least two of the following (medical, psychosocial, security, legal support and livelihoods support) Outcome Indicators 2. Percentage of eligible reported GBV cases who receive Post Exposure Prophylaxis (PEP) Treatment within 72 hours 3. Percentage of implementing partners providing services to GBV survivors in line with quality standards 1. Number of health personnel receiving training on GBV service provision 2. Number of reported cases of GBV that access at least one service supported by the project (disaggregated by entry point) 3. Percentage of rape cases that access services within 72 hours of the incident 4. Percentage of beneficiaries who meet regularly with their case Intermediate Results Indicators manager, as defined in the project manual 5. Percentage of essential medicines (PEP, STI Treatment and Emergency Contraception) for which there was no stock out during the implementation period 6. Percentage availability of basic equipment at health facility level in line with the project’s quality check-list 7. Percentage of small-scale works at health facility level complying with ESMF requirements 1. Number of health personnel receiving training on GBV service provision Key Outputs by Component 2 2. Number of reported cases of GBV that access at least one service (linked to the achievement of the Objective/Outcome 2) supported by the project (disaggregated by entry point) 3. Percentage of rape cases that access services within 72 hours of the incident Page 41 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 4. Percentage of beneficiaries who meet regularly with their case manager, as defined in the project manual 5. Percentage of essential medication (PEP, STI Treatment and Emergency Contraception) for which there was no stock out during the implementation period 6. Percentage availability of basic equipment at health facility level in line with the project’s quality check-list 7. Percentage of small-scale works at health facility level complying with ESMF requirements Key Outputs by Component 3 1. Percentage of grievances received by the project that are addressed (linked to the achievement of the Objective/Outcome 2) in line with quality standards defined in the GRM manual Objective/Outcome 3 Outcome Indicators None Intermediate Results Indicators None Page 42 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Patricia Maria Fernandes Task Team Leader(s) Lanssina Traore Procurement Specialist(s) Bella Diallo Financial Management Specialist Maurizia Tovo Peer Reviewer Hadia Nazem Samaha Team Member Caroline L. Guazzo Team Member Lucienne M. M'Baipor Social Specialist Hugues Agossou Peer Reviewer Joseph-Antoine Ellong Team Member Diana Jimena Arango Peer Reviewer Harald Hugo Hinkel Team Member Julia Vaillant Team Member Grace Muhimpundu Social Specialist Sofia De Abreu Ferreira Team Member Koho Francine Takoy Team Member Lisette Meno Khonde Team Member Issa Thiam Team Member Michel Muvudi Lushimba Team Member Guy Kiaku Kindoki Team Member Page 43 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Saba Nabeel M Gheshan Counsel Julie Luvisa Bazolana Team Member Joelle Nkombela Mukungu Environmental Specialist Tresor Angela Ikobo Team Member Daniela Greco Team Member Jonathan Volger Greenland Team Member Katie Lauren Robinette Team Member Supervision/ICR Hiska Noemi Reyes Task Team Leader(s) Jean-Claude Azonfack Procurement Specialist(s) Bertille Gerardine Ngameni Wepanjue Financial Management Specialist Cyrille V. Ngouana Kengne Environmental Specialist Shamard Ya Jua Mungu Shamalirwa Social Specialist Yasmine Binti Sangwa Team Member Katie Lauren Robinette Team Member Lydia Filunga Ndaya Kanyembo Environmental Specialist Daniela Greco Team Member Pascaline Okako Okongo Team Member Guy Kiaku Kindoki Team Member Lisette Meno Khonde Team Member Koho Francine Takoy Team Member Richard Everett Social Specialist Luc Sukadi Mbayo Procurement Team Julia Vaillant Team Member Lucie Lufiauluisu Bobola Procurement Team Verena Phipps-Ebeler Team Member Nikolai Alexei Sviedrys Wittich Procurement Team Page 44 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Helena Hwang Team Member Elena Segura Labadia Team Member Harald Hugo Hinkel Team Member Diana Jimena Arango Peer Reviewer Joseph-Antoine Ellong Team Member Mohammad Ilyas Butt Procurement Team B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY18 8.085 225,739.46 FY19 4.000 29,618.93 FY20 1.950 7,450.14 Total 14.04 262,808.53 Supervision/ICR FY19 7.847 184,005.28 FY20 28.259 298,836.68 FY21 50.895 366,634.66 FY22 28.112 287,605.80 FY23 45.209 399,011.99 FY24 28.704 182,170.47 Total 189.03 1,718,264.88 Page 45 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 3. PROJECT COST BY COMPONENT Amount at 2023 Revised Percentage of 2023 Actual at Project Components Approval Allocation Revised Allocation Closing (US$M)67 (US$M) (US$M) (US$M) Gender-Based Violence prevention and integrated 53696972.14 46457390.00 37391144.38 80.48 support for survivors at community level Response to Gender-Based 27481300.39 27481300.00 24637821.31 89.65 Violence Support to Policy Development, Project 18821727.47 12561309.00 10407533.84 82.85 Management and Monitoring and Evaluation Contingency Emergency 0 0 0 0 Response Component Total 100000000.00 86499999.00 72436499.53 83.74 67Figures are current as of June 30, 2023 (FSRDC Progress Report, August 2023); final actuals were still pending at the time of this report as project was still in the grace period. Page 46 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 4. EFFICIENCY ANALYSIS This analysis assesses whether the dollar benefit of the Project outweighs its dollar costs. To do so, it monetizes the major benefits and costs associated with the project, and reports on three measures: a. Benefit to cost ratio (BCR): is the ratio between the benefits and costs of the project, expressed in monetary units at discounted present values. A ratio greater than one indicates that project benefits outweigh its costs. b. Net present value (NPV): is the absolute difference between the benefits and the costs. An NPV above zero indicates that project was profitable. c. Internal rate of return (IRR): is the discount rate that equates the present value of the project’s cash inflow to the present value of its outflow. The IRR measures the projects growth rate. LIMITATIONS A notable limitation of this economic assessment is that the full economic benefits, particularly for GBV response interventions, could not be estimated due to a lack of data. As a result, the benefits of the project are underestimated. Despite this limitation, the analysis aims to offer valuable insights based on the available data while recognizing the need for further research to capture the entire spectrum of economic benefits. DATA SOURCES The data required to estimate the monetary costs and benefits of the outcomes assessed for this project are vast and were obtained from many sources. Examples of data requirements are project cost data; macroeconomic data; violence prevalence estimates, administrative data on the number of women seeking services; and disability weights associated with certain adverse health conditions. Details of the information gathered are shown in Table 1. Table 1: Data Sources used for economic assessment Indicator Source Project cost data Project report • No. women beneficiaries of Project dashboard community sensitization • No. women enrolled in income- generating activities • Number of women/men beneficiaries who received PEP Project treatment beneficiaries • Number of women/men NET beneficiaries • Number of women/men beneficiaries of non-NET mental health care • No. women treated for health conditions Page 47 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Indicator Source • GDP per capita World Bank indicators Macroeconomic • Annual % GDP growth https://data.worldbank.org/indicator data • Inflation • Population prevalence of sexual DRC 2014 Demographic and Health Survey Violence violence https://dhsprogram.com/publications/publication- prevalence • Population prevalence of fr300-dhs-final-reports.cfm estimates physical violence • Severe depression Global Burden of Disease study • Mild depression https://ghdx.healthdata.org/record/ihme- Disability weights • Fistula & other gynecological data/gbd-2019-disability-weights conditions • Prolapse PROJECT BENEFITS The project included a wide range of interventions that could lead to economic benefits. However, many of the project’s activities were difficult to quantify in economic terms due to a lack of detailed data. This analysis therefore focuses on key areas where project benefits could be captured and monetized: cases of sexual and physical violence as well as of HIV transmission averted; productivity gain from women’s engagement in income generating activities as a result of economic support; and improvements in mental and sexual and reproductive health through medical service provision. The benefits assessed capture the project development objectives to increase participation in GBV prevention interventions and to increase use of multi-sectoral response services through the provision of health and other types of survivor assistance. Monetary benefits for the following project outcomes were modelled: • Number of cases of sexual violence and physical violence averted resulting from community mobilization / sensitization activities (Component 1) • Productivity gain from women’s engagement in income generating activities through economic support (Component 1) • Number of cases of HIV transmission averted from the administration of post-exposure prophylaxis (PEP) within 72 hours (Component 2) • Gains from improvements in mental and sexual and reproductive health as a result of psychosocial and medical support interventions. Beneficiaries include women and men who received psychosocial support – either Narrative Exposure Therapy (NET) (Component 1), or other specialized mental health care, or basic psychosocial support (Component 2), and medical procedures for gynecological conditions (Component 2). a. Estimating the number of cases of sexual violence and physical violence averted from community mobilization and behavior change The project includes a community mobilization and promotion of behavior change component aimed at decreasing experience of violence. The number of cases of sexual violence and physical violence averted (NSEXUAL_AVERTED and NPHYSICAL_AVERTED) were estimated using a methodology applied in the economic Page 48 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) evaluation of the SASA! intervention68. The approach estimates the difference between the expected number of cases in the absence of the intervention and the estimated number of cases among those exposed to the intervention, and was based on the following information: • population prevalence of sexual violence (SV) and physical violence (PV), • sexual violence and physical violence risk reduction factors (SVRR & PVRR respectively) estimated from the SASA economic evaluation (see note 68), and • the number of beneficiaries from the community mobilization sub-component (NBENEFICIARIES). The 2013-2014 DRC Demographic and Health Survey reported that at the population level, 16.3% of women (ages 15 to 49 years) experienced sexual violence and 27.2% experienced physical violence in the preceding 12 months to interview (see Table 1 for source). An economic evaluation of SASA! (see note 68), estimated a risk reduction in past year sexual violence of 6.0% and past year physical violence of 11.6%. Finally, project monitoring indicators estimated 1,195,345 women were beneficiaries of community mobilization and awareness-raising as well as gender transformative training activities.69 Multiplying these together yielded an estimated 11,690 cases of sexual violence averted and an estimated 37,716 cases of physical violence averted. NSEXUAL_AVERTED = SV x SVRR x NBENEFICIARIES = 16.3% x 6.0% x 1,195,345 = 11,690* NPHYSICAL_ AVERTED = PV x PVRR x NBENEFICIARIES = 27.2% x 11.6% x 1,195,345 = 37,716* The monetary benefit of the community mobilization and promotion of behavior change component was modelled as the number of sexual violence and physical violence cases averted multiplied by the savings in medical, legal, psychosocial support, and social services provision to survivors of violence (sexual and physical). The total unit cost of these services ($SERVICE) was calculated to be US$ 1,29170. TOTAL MONETARY BENEFIT FROM COMMUNITY (NSEXUAL AVERTED+NPHYSICAL AVERTED) x $SERVICE = = US$ 63,783,145 MOBILISATION AND (11,690* + 37,716*) X US$1,291 BEHAVIOUR CHANGE * Figures represented to the nearest whole number. The estimated number of sexual violence cases averted was 11,690.47 (2dp) and the estimated number of physical violence cases averted was 37,715.53 (2dp); a discrepancy in estimated total monetary benefit is due to rounding. 68 Michaels-Igbokwe et al. BMC Public Health DOI 10.1186/s12889-016-2883. SASA! (Kiswahili word for “now” and acronym for the program’s four phases – Start, Awareness, Support, and Action) is a community mobilization intervention aimed at preventing violence against women and reducing HIV-risk behaviors, which was subject to a successful impact evaluation in Uganda (program designed by Raising Voices and implemented by CEDOVIP). The study showed a reduction in the acceptance of physical violence and increase in the social acceptance of the belief that women can refuse sex from her partner; the study also showed that levels of physical partner violence were lower in SASA! Intervention communities and that women in intervention communities who experienced violence were more than twice as likely to report that they had experienced a supportive community response. Watts, C. et al. 2015, The SASA! study: a cluster randomised trial to assess the impact of a violence and HIV prevention programme in Kampala, Uganda, 3ie Impact Evaluation Report 24. 69 These activities included a wide range of community mobilization and behavior change interventions, including some SASA! activities at community level. Project data on community mobilization activities were not disaggregated for every individual activity. The number of beneficiaries included in this calculation represented beneficiaries of small-scale interventions, such as SASA!. 70 Avenant 001-2020 CDE Fondation Panzi SK (signed version), page 24. The unit costs for medical services, legal services, psychosocial support, and social services, respectively. were US$ 700, US$ 379, US$ 93 and US$ 119. Page 49 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) b. Productivity gain from women’s engagement in income-generating activities resulting from economic support interventions Another prevention intervention component provided economic support to women through participation in women-only Village Savings and Loan Associations to stimulate engagement in income-generating activities. In total, the project reached 27,414 women71 (NINCOME) through this activity, and all women were reported to have become engaged in income-generating activities72. The monetary benefit of this activity was conceptualized as the productivity gain over the beneficiary’s remaining productive life course, which was assumed until age of retirement at 60 years for women.73 Because demographic information from women participating in this program, including their age, was not collected, a low and high productive life span was assumed, i.e., 20 years (based on the assumption the average age of women participants was 40 years) and 35 years (based on the assumption the average age of women participants was 25 years). The productivity gain was then modelled as the number of beneficiaries (NINCOME) multiplied by the discounted accumulation of GDP per capita over the productive life. The DRC GDP per capita for 2022 (US$ 653.66)74 was taken as the base year. An average annual GDP growth rate of 5% (which is the average of the annual growth rates between 2018 and 202275) was applied to estimate annual GDP per capita for each successive year over 35 years. To adjust for future values of GDP per capita, a 3% discount rate was applied using the formula described in footnote below76. The discounted GDP per capita rates were summed over a period of: a) 20 years (low productive life span $GAINLOW) to yield US$ 15,790 per capita, and b) 35 years (high productive span $GAINHIGH) to yield US$ 32,327 per capita. To estimate the total monetary benefit the number of beneficiaries (NINCOME) was multiplied by the discounted per person productivity gain: NINCOME x $GAINLOW TOTAL MONETARY LOW (20 YEARS) = = US$ 432,876,047 27,414 x US$ 15,790 BENEFIT FROM NINCOME x $ GAINHIGH ECONOMIC SUPPORT HIGH (35 YEARS) = = US$ 886,208,875 27,414 x US$ 32,327 * Figures represented to the nearest whole number. Estimated $GAINLOW was US$ 15,790.33 (2dp) and estimated $GAINHIGH was US$ 32,326.87 (2dp); discrepancies in estimated total monetary benefit are due to rounding. 71 Note that this metric is only for women involved in new VSLAs (which were women-only) under the project rather than other associations called Mutualité de Solidarité (MUSO), some of which were previously in existence with the Centers of Excellence and also with mixed male and female membership. 72 Communication with project lead 73 https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiEuff747WEAxX0g_0HHeYnBrAQFnoECBkQAQ&url=h ttps%3A%2F%2Fwbl.worldbank.org%2Fcontent%2Fdam%2Fdocuments%2Fwbl%2F2020%2Fsep%2FCongo-dem- rep.pdf&usg=AOvVaw2xc7DeuH5Q1fgpVxS-vrJF&opi=89978449 74 https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?locations=CD (2022 being the most recent year for which GDP data are available) 75 https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=CD 76 Based on ICR methodology from economic analysis for Great Lakes Emergency Sexual and Gender-Based Violence and Women's Health Project (P147489). Discounting is the process of weighting costs and benefits that occur at different time points. A discount rate is normally applied using $0 $1 $2 $ the formula $ = + + …….. where $ is the present value of the monetary benefit, $ is the future benefit at year (1+)0 (1+)1 (1+)2 (1+) and is the rate of discount. Gray A.M., Clarke P.M., Wolstenholme J. L., Wordsworth S. 2011. Applied Methods of Cost-Effectiveness Analysis in Health Care. Oxford University Press, Oxford, page 131. Typically, within health economic evaluations, the most frequently applied discount rates are 3% or 5% https://www.york.ac.uk/che/pdf/tp19.pdf. Page 50 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) c. Number of cases of HIV transmission averted due to administration of post-exposure prophylaxis (PEP) within 72 hours The number of HIV transmissions averted (among men and women) due to the administration of PEP within 72 hours was estimated using the methodology applied in the ICR for the Great Lakes Emergency Sexual and Gender-Based Violence and Women’s Health Project (GL GBV Project).77 To estimate the number of HIV transmissions averted among women (NHIV_AVERTED_FEMALE), the number of women exposed who received PEP (NPEP_FEMALE) was multiplied by the HIV prevalence (HIVPREV) and by the estimated effectiveness of PEP in preventing HIV transmission (EffectPEP). The same approach was applied to estimate the number of HIV transmissions averted among men (NHIV_AVERTED_MALE): the number of men exposed who received PEP (NPEP_MALE) was multiplied by HIV prevalence (HIVPREV) and the effectiveness of PEP (EffectPEP). The HIV prevalence in DRC was estimated to be 0.7%78 in 2022; the total number of women and men who received PEP treatment within 72 hours under the project was 5,334 and 120 respectively; and the effectiveness of PEP in reducing the risk of getting HIV is documented at over 80%79. With this information, the numbers of HIV transmissions averted were as follows: NHIV_AVERTED_FEMALE = NPEP_FEMALE x HIVPREV x EffectPEP = 5334 x 0.7% x 80%= 30 cases averted* NHIV_AVERTED_MALE = NPEP_MALE x HIVPREV x EffectPEP = 120 x 0.7% x 80% = 1 case averted* The monetary benefit resulting from averting cases of HIV transmission was estimated by multiplying the number of HIV cases averted and the cost of HIV treatment extrapolated over the average number of years living with HIV. Using the same assumptions as applied in the GL GBV Project ICR efficiency analysis (see note 81), the average number of years living with HIV was estimated as half of life expectancy in DRC. The life expectancy for women and men in DRC was 62 years and 57 years respectively in 2021.80 Therefore, the number of years living with HIV for women (YRSHIV_FEMALE) and men (YRSHIV_MALE) is 31 years and 29 years respectively. The annual cost of HIV treatment ($HIV) was US$ 208, which was extrapolated over the expected number of years living with HIV by adjusting for inflation (π) of 0.6% each year81. The lifetime costs of HIV treatment for women and men respectively are: $HIV_LIFE_FEMALE = $HIV x [(1-(1+ π /1)^YRSHIV_FEMALE) / (1-(1+ π /1)] = 208 x [(1-(1+0.6/1)^31) / (1-(1+0.6/1)] = US$ 7,063* 77 GL GBV Project ICR, page 85 78https://data.worldbank.org/indicator/SH.DYN.AIDS.ZS?locations=CG-CD (2022 being the most recent year for which HIV data are available) 79https://hivinfo.nih.gov/understanding-hiv/fact-sheets/post-exposure-prophylaxis-pep 80 Life expectancy for women and for men in the DRC (2021 data are the most recent available): https://data.worldbank.org/indicator/SP.DYN.LE00.FE.IN?locations=CD (women); https://data.worldbank.org/indicator/SP.DYN.LE00.MA.IN?locations=CD (men) 81 GL GBV Project ICR, page 88 Page 51 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) $HIV_LIFE_MALE = $HIV X [(1-(1+ π /1)^ YRSHIV_MALE) / (1-(1+ π /1)] = 208 X [(1-(1+0.6/1)^29) / (1-(1+0.6/1)] = US$ 6,567* TOTAL MONETARY NHIV_AVERTED_FEMALE x $HIV_LIFE_FEMALE WOMEN = = US$ 210,988 BENEFIT FROM HIV = 30* x US$ 7,063* TRANSMISSIONS NHIV_AVERTED_MALE x $HIV_LIFE_MALE MEN = = US$ 4,413 AVERTED DUE TO PEP = 1* x US$ 6,567* * Figures represented to nearest whole number. The estimated numbers of cases of HIV transmissions averted was 29.87 (2dp) among females and was 0.67 (2dp) among males. The estimated lifetime cost of HIV treatment for females was US$ 7,063.44 (2dp) and the estimated lifetime cost of HIV treatment for males was US$ 6567.15 (2dp). Discrepancies in estimated total monetary benefit are due to rounding. d. Improvements in mental and sexual and reproductive health outcomes To estimate the monetary benefits of improved mental and sexual and reproductive health outcomes, gains in health-related quality of life were estimated using disability weights. Disability weights reflect the severity of different health states and range from 0 (equivalent to a state of full health) to 1 (equivalent to death). The weights are used to calculate disease burden typically, by multiplying the disability weight for a disease or health state and the number of years lived in that health state. Disability weights used for the analysis of health outcomes are presented in Table 282. Table 2: Disability weights associated with health conditions (see note 82) Health state Disability weight Depression – severe 0.658 Depression – moderate 0.396 Depression – mild 0.145 Fistula 0.324 Other gynecological conditions 0.324 Prolapse 0.031 Improvements in mental health This analysis measured the monetary benefits resulting from improvements in mental health following three mental health interventions i) Narrative Exposure Therapy (NET); ii) other non-NET specialized mental health care; and iii) provision of basic psychosocial support. i) NET The number of people who were treated with the NET approach and whose mental health improved (NIMPROVE_NET) was first estimated by multiplying the number of people who received NET ( NNET) by the estimated percentage of people whose mental health was no longer classified as severe depression (PERCENTDIFF) post-intervention. In total, 7,198 people with mental health conditions were supported using the NET approach (6,489 women and 709 men). A pre-post evaluation of the NET intervention 82 Global Burden of Disease https://ghdx.healthdata.org/record/ihme-data/gbd-2019-disability-weights Page 52 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) documented that at baseline 79.2% of those assigned to receive NET had symptoms of severe depression and this proportion was 41.3% 12 months post-intervention, yielding a difference of 37.9 percentage points.83 NIMPROVE_NET = NNET x PERCENTDIFF = 7198 x 37.9% = 2728* cases of improved mental health To estimate the monetary benefit because of NET, the number of people whose mental health improved (NIMPROVE_NET) was multiplied by the disability weighted gain in GDP per capita ($GAINNET). The GDP per capita gain was calculated as GDP per capita multiplied by the difference in disability weight associated with severe and with mild depression over the expected number of years the treatment is effective. The long-term impacts of mental health interventions are unknown, and therefore, a conservative estimate of two years improvement post-intervention was assumed. The DRC GDP per capita for 2022 is US$ 653.66 (see note 74) and an average annual GDP growth rate of 5% (the average of the annual growth rates between 2018 and 2022 – see note 75) were applied to estimate the GDP per capita for two years [i.e., US$ 653.66 year 1 and US$ 686.34 year 2]. The difference in disability weights for severe depression and mild depression is 0.51384. The disability weight adjusted GDP per capita for year 1 was US$ 335 (US$ 653.66 * 0.513) and the disability weight adjusted GDP per capita for year 2 was US$ 352 (US$ 686.34 * 0.513). The GDP per capita in year 2 was then discounted (GDP per capita in year 1 is the base year and, therefore, already represents present value), applying a 3% rate to yield a gain of US$ 342 in year 2 (see note 76). The total gain in per capita GDP ($GAINNET) was therefore US$ 677 (US$ 335 + US$ 342). TOTAL MONETARY BENEFIT NIMPROVE_NET x $GAINNET FROM IMPROVED MENTAL = = US$ 1,847,328 2728* X US$ 677* HEALTH FROM NET * Figures represented to nearest whole number. The estimated number of cases with improved mental health was 2728.04 (2dp). The estimated gain in GDP from improved mental health was US$ 677.16 (2dp). Discrepancies in estimated total monetary benefit are due to rounding. ii) Other non-NET specialized mental health care The methodological approach used to estimate the benefit of NET was used to estimate the benefit of receiving other forms of specialized mental health care (other than NET). The number of people who were treated with other forms of specialized mental health care (NIMPROVE_OTHER) was first estimated by multiplying the total number of people who received non-NET specialized mental health care (NOTHER) by the estimated percentage of people whose mental health was no longer classified as severe depression (PERCENTDIFF) post-intervention. In total, 3,305 people were provided with other forms of specialized mental health care (3,151 women and 154 men were supported with specialized mental health care other than NET and including NETFait). The pre-post evaluation of the NET intervention documented that, at baseline, 79.4% of those assigned to the control group who received other mental health support, had symptoms of severe depression and this proportion was 46.1% at 12 months follow-up, a difference of 33.3 percentage points. 83 Project data courtesyof Africa Gender Innovation Lab 84 From Table 2: the disability weight associated with severe depression is 0.658 and the disability weight associated with mild depression is 0.145, therefore the difference is 0.658 – 0.145 = 0.513. Page 53 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) NIMPROVE_OTHER = NOTHER x PERCENTDIFF = 3,305 x 33.3% = 1101* cases of improved mental health To estimate the monetary benefit because of other forms of mental health care, the number of people whose mental health improved (NIMPROVE_OTHER) was multiplied by the disability weighted gain in GDP per capita ($GAINOTHER). The GDP per capita gain was calculated as GDP per capita multiplied by the difference in disability weight associated with severe and mild depression over the expected number of years that the treatment is effective. As above, a conservative estimate of two years improvement post-intervention was assumed. The DRC GDP per capita for 2022 is US$ 653.66 (see note 74) and an average annual GDP growth rate of 5% (the average of the annual growth rates between 2018 and 2022 – see note 75) was applied to estimate the GDP per capita for two years [i.e., US$ 653.66 year 1 and US$ 686.34 year 2). The difference in disability weights for severe depression and mild depression is 0.51385. The disability weight adjusted GDP per capita for year 1 was US$ 335 (US$653.66 * 0.513) and the disability weight adjusted GDP per capita for year 2 was US$ 352 (US$ 686.34 * 0.513). The GDP per capita in year 2 was then discounted (GDP per capita in year 1 is the base year and, therefore, already represents present value), applying a 3% rate to yield a gain of US$ 342 in year 2 (see note 76). The total gain in per capita GDP ($GAINOTHER) was therefore US$ 677 (US$ 335 + US$ 342). The monetary benefit from women’s and men’s improved mental health because of other forms of psychosocial care was estimated as follows: TOTAL MONETARY BENEFIT FROM IMPROVED MENTAL NIMPROVE_OTHER x $GAINOTHER HEALTH FROM OTHER = = US$ 745,261 1101* X US$ 677* SPECIALIZED MENTAL HEALTH CARE * Figures represented to nearest whole number. The estimated number of cases with improved mental health was 1100.57 (2dp). The estimated gain in GDP from improved mental health was US$ 677.16 (2dp). Discrepancies in estimated total monetary benefit are due to rounding. iii) Basic psychosocial support A similar approach used to estimate the monetary benefit from NET and from other specialized mental health care was used to estimate the monetary benefit of the provision of basic psychosocial support (PSS). The number of people who were treated with basic PSS from women’s community-based organizations (CBOs) and the Centers of Excellence and whose mental health improved (NIMPROVE_PSS) was first estimated by multiplying the number of people who received basic psychosocial support (NPSS) by the estimated percentage of people whose mental health was no longer classified as moderate depression (PERCENTDIFF) post-treatment. In total, 63,751 people were supported with basic psychosocial care (11,511 from Centers of Excellence, and 52,240 from CBOs). As the extent of recipients’ mental health status and the proportion whose mental health improved because they received basic psychosocial support are not known, it was assumed that beneficiaries of psychosocial support presented with 85 See calculation at note 85. Page 54 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) moderate levels of depression, and the same percent improvement in mental health as documented for other forms of specialized mental health care was applied (33.3%) to obtain the number of cases of improved mental health. NIMPROVE_PSS = NPSS x PERCENTDIFF = 63,751 x 33.3% = 21,229* cases of improved mental health To estimate the monetary benefit because of basic psychosocial support, the number of people whose mental health improved (NIMPROVE_PSS) was multiplied by the disability weighted gain in GDP per capita ($GAINPSS). The GDP per capita gain was calculated as GDP per capita multiplied by the difference in disability weight associated with moderate and with mild depression over the expected number of years the treatment is effective. The long-term impacts of mental health interventions, and in particular basic psychosocial support, are unknown, and therefore, a conservative estimate of two years improvement post treatment was assumed. The DRC GDP per capita for 2022 is US$ 653.66 (see note 74) and an average annual GDP growth rate of 5% (the average of the annual growth rates between 2018 and 2022 – see note 75) were applied to estimate the GDP per capita for two years [i.e., US$ 653.66 year 1 and US$ 686.34 year 2]. The difference in disability weights for moderate depression and mild depression is 0.25186. The disability weight adjusted GDP per capita for year 1 was US$ 164 (US$ 653.66 * 0.251) and the disability weight adjusted GDP per capita for year 2 was US$ 172 (US$ 686.34 * 0.251). The GDP per capita in year 2 was then discounted (GDP per capita in year 1 is the base year and, therefore, already represents present value), applying a 3% rate to yield a gain of US$ 167 in year 2 (see note 76). The total gain in per capita GDP ($GAINPSS) was therefore US$ 331 (US$ 164 + US$ 167). TOTAL MONETARY BENEFIT NIMPROVE_PSS x $GAINPSS FROM IMPROVED MENTAL = = US$ 7,033,645 21,229* X US$ 331* HEALTH FROM BASIC PSS * Figures represented to nearest whole number. The estimated number of cases with improved mental health was 21229.08 (2dp). The estimated gain in GDP from improved mental health was US$ 331.32 (2dp). Discrepancies in estimated total monetary benefit are due to rounding. Improvement in women’s reproductive health through treatment of fistula, prolapse and other gynecological conditions This analysis measured the monetary benefits resulting from treatment of the gynecological conditions: i) fistula and other gynecological conditions; and ii) prolapse. i) Fistula and other gynecological conditions The number of women successfully treated for fistula (NREPAIRED) was estimated by multiplying the number of women who were treated (NFISTULA) by the percentage of women who were considered “repaired” after 12 months (PERCENTFISTULA). In total, 1,620 women with fistula or other gynecological conditions received treatment. Studies have estimated that 83.9% are considered repaired after 12 months87. 86 From Table 2: the disability weight associated with moderate depression is 0.396 and the disability weight associated with mild depression is 0.145, therefore the difference is 0.396 – 0.145 = 0.251. 87 Benski et al. 2019. Prognostic factors and long-term outcomes of obstetric fistula care using the Tanguiéta model. DOI: 10.1002/ijgo.13071 Page 55 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) NREPAIRED = NFISTULA x PERCENTFISTULA = 1620 x 84% = 1359* successfully treated after 12 months To estimate the monetary benefit of women’s improved health, the number of women successfully treated for fistula (NREPAIRED) was multiplied by the gain in GDP per capita ($GAINFISTULA). The gain is estimated as the disability weighted GDP per capita over the expected number of years the treatment is effective. The long-term impacts of fistula surgery are unknown, and, therefore, a conservative estimate of five years improvement post-surgery was assumed. The DRC GDP per capita for 2022 is US$ 653.66 (see note 74) and an average annual GDP growth rate of 5% (the average of the annual growth rates between 2018 and 2022 – see note 75) was applied to estimate the GDP per capita for the five years post-treatment. The disability weighted GDP per capita is calculated by multiplying the GDP per capita over the five years by the disability weight for fistula and other gynecological conditions which is 0.324 (from Table 2). To convert future monetary benefits into present value, a 3% discount rate was applied to yield an overall GDP per capita gain ($GAINFISTULA) of US$ 1,101 (see note 76 for discounting formula applied). TOTAL MONETARY BENEFIT FROM IMPROVED HEALTH FOLLOWING NREPAIRED x $GAINFISTULA = = US$ 1,496,257 FISTULA/OTHER GYNAECOLOGICAL 1359* x $US 1,101* SURGERY * Figures represented to nearest whole number. The estimated number of cases with improved health was 1359.18 (2dp). The estimated gain in GDP from improved health as a result of fistula surgery was US$ 1,100.85 (2dp). Discrepancies in estimated total monetary benefit are due to rounding. ii) Prolapse The number of women successfully treated for prolapse was estimated by multiplying the number of women who were treated by the percentage of women who were considered “repaired”. In total, 3,764 women with prolapse received treatment (NPROLAPSE). Studies have estimated that, following treatment for prolapse, 89% were considered repaired after 5 years (PERCENTPROLAPSE_5) and 86% after 10 years (PERCENTPROLAPSE_10).88 The numbers of women considered repaired after 5 years (NREPAIRED_5) and 10 years (NREPAIRED_10) were estimated as follows: NREPAIRED_5 = NPROLAPSE x PERCENTPROLAPSE_5 = 3,764 x 89%= 3350* successfully treated after 5 years NREPAIRED_10 = NPROLAPSE x PERCENTPROLAPSE_10 = 3,764 x 86%= 3237* successfully treated after 10 years To estimate the monetary benefit of women’s improved health, the number of women successfully treated for prolapse was multiplied by the gain in GDP per capita from treatment. The estimate of the per capita GDP gain resulting from treatment for prolapse was undertaken in two stages. First, GDP gain for all 3237 women considered repaired at 10 years was estimated for the full ten years ($GAINPROLAPSE_10). Second, the GDP gain for the 113 women (3350 – 3237) who were repaired at 5 years (but not at 10 years) 88 Nüssleret al. 2022. Long-term outcome after routine surgery for pelvic organ prolapse—A national register-based cohort study. https://doi.org/10.1007/s00192-022-05156-y Page 56 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) was estimated over the first five years ($GAINPROLAPSE_5). GDP gain is estimated as the disability weighted GDP per capita over the expected number of years the treatment is effective (i.e., 10 years and 5 years) where monetary values of future benefits were discounted into present value. The DRC GDP per capita for 2022 is US$ 653.66 (see note 74) and an average annual GDP growth rate of 5% (the average of the annual growth rates between 2018 and 2022 – see note 75) was applied to estimate the following annual GDP per capita over ten years. The disability weighted GDP per capita multiplies the accumulated GDP per capita (for 10 years and for 5 years) by the disability weight for prolapse which is 0.031 (from Table 2). To convert future monetary benefits into present value a 3% discount rate was applied to yield an overall GDP per capita gain ($GAINPROLAPSE_10) of US$ 221 over 10 years and a GDP per capita gain ($GAINPROLAPSE_5) of US$ 105 over 5 years (see note 76 for discounting formula applied). The total monetary benefit is the sum of the estimated number of women repaired at 10 years multiplied by the productivity gain over 10 years and the estimated number of women who were repaired at 5 years but not at 10 years multiplied by the productivity gain over 5 years. TOTAL MONETARY NREPAIRED_10 x $GAINPROLAPSE_10 BENEFIT FROM AFTER 10 YEARS = = US$ 716,318 = 3237* x $US 221* IMPROVED HEALTH FOLLOWING PROLAPSE NREPAIRED_5 x $GAINPROLAPSE_5 TREATMENT AFTER 5 YEARS = = US$ 11,894 = (3350-3237)* x $US 105* * Figures represented to nearest whole number. The estimated number of cases with improved health was 3349.96 (2dp) after 5 years and 3,237.04 after 10 years. The estimated gain in GDP from improved health as a result of prolapse treatment was US$ 105.33 (2dp) after 5 years and US$ 221.29 after 10 years. Discrepancies in estimated total monetary benefit are due to rounding. Total project benefits across intervention components The estimated monetary benefits from the intervention components are shown in Table 3. The total monetary benefit across all intervention components ranged from US$ 504.3 million (low estimate) to US$ 957.6 million (high estimate). Table 3: Total project benefits across intervention components Monetary Intervention Benefit estimated Benefit (US $) Community mobilization and Sexual violence and physical cases violence 63,783,145 promotion of behavior change averted Economic support through Productivity gain from women's Low 432,876,047 participation in women-only Village engagement in income generating Savings and Loan Associations activities High 886,208,875 Administration of PEP within 72 Cases of HIV transmission averted: women 210,988 hours of exposure Cases of HIV transmission averted: men 4,413 Narrative Exposure Therapy Improvement in mental health 1,847,328 Page 57 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Monetary Intervention Benefit estimated Benefit (US $) Other non-NET specialized mental Improvement in mental health 745,261 health care Basic psychosocial support89 Improvement in mental health 7,033,645 Treatment of fistula / other Improvement in women's health 1,496,257 gynecological conditions Treatment of prolapse (10 yrs + 5 Improvement in women's health 728,212 yrs) TOTAL MONETARY BENEFIT: LOW ESTIMATE 508,725,296 TOTAL MONETARY BENEFIT: HIGH ESTIMATE 962,058,124 PROJECT COSTS Table 4a summarizes the cost of the project-by-project component and by year. The total project expenditures amounted to US$72.4 million (as of June 30, 2023). Costs for Component 3 were proportionately allocated to Components 1 and 2 (Table 4b). Table 4a: Program costs by project component from 2019 to 2023 Comp Total Total 2019 2020 2021 2022 2023* onent Actual# Budgeted+ 1 324,192 5,386,976 11,482,213 14,362,241 5,835,523 37,391,144 46,457,390 1.1 324,192 5,136,073 9,574,259 11,989,210 4,913,396 31,937,129 39,957,390 1.2 250,903 1,907,954 2,373,032 922,127 5,454,016 6,500,000 2 820,629 4,827,501 6,828,862 8,847,091 3,313,738 24,637,821 27,481,300 2.1 820,629 4,699,028 6,693,618 8,176,514 3,285,710 23,675,499 20,714,841 2.2 128,473 135,244 670,577 28,028 962,322 6,766,459 3 1,450,014 2,125,914 2,424,292 3,217,368 1,189,947 10,407,534 12,561,309 Total 72,436,500 86,499,999 * First semester 2023 only (August 2023 Progress Report, FSRDC) # As of June 30, 2023 (most recent financial figures available as of the filing of the ICR) + Note that total budgeted for the project was reduced from $100M to $86.5M after project restructuring reduced budget to $89.5M and client requested to return an additional $3M. Table 4b: Program costs by project component with costs for Component 3 proportionately allocated to Components 1 and 2 Estimate project Actual projects US $ (with US $ for component 3 costs (US $) costs (US $)* apportioned to components 1 and 2) Component 1 46,457,390 37,391,144 43,664,820 Component 2 27,481,300 24,637,821 28,771,680 Component 3 12,561,309 10,407,534 89 Note that basic psychosocial support benefits are calculated together for both Centers of Excellence and CBOs under Component 2. Page 58 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Total 86,499,999 72,436,500 72,436,500 * As of June 30, 2023 (most recent financial figures available as of the filing of the ICR) RESULTS The economic analysis estimated three indicators: 1) the benefit-cost ratio (BCR); 2) the Net Present Value (NPV); and 3) the internal rate of return (IRR). The BCR expresses the result as a ratio of benefits to cost (benefit divided by cost), where a BCR greater than 1.0 indicates that the benefits of an investment outweigh the costs. The NPV is the difference between benefits and costs and measures the net economic gain; a positive NPV indicates a net economic benefit. The IRR measures the annual rate of return of the investment in the project.90 The project was overall successful in economic terms, looking at interventions under both Components 1 and 2 together. The overall BCR was 7.0 for low range estimate and 13.3 for the high range estimate. This suggests every dollar invested in project yielded an economic return ranging from US$ 7.00 to US$ 13.30. The total investment in the project of US$ 72.4 million generated an overall economic benefit (NPV) that ranged from US$ 509 million to US$ 962 million. The overall IRR ranged from 21.5% to 29.5%, suggesting a positive rate of growth owing to the investments made under the project. Prevention activities measured under Component 1 showed very positive results, especially those related to economic support. Response interventions related to medical and other non-NET forms of psychosocial support, however, showed less favorable results under the economic analysis; this could be partly explained by the high cost of service provision in the DRC, given poor infrastructure, lack of widely available and detailed data on the cost of providing survivor care, and the overall low rates of and barriers to survivor access. It should also be noted that, in accordance with best practices, prevention interventions cannot ethically be conducted without the availability of basic medical and psychosocial support; as such, the favorable results for the prevention activities under Component 1 would not have been possible without the associated interventions in response services under Component 2. In addition, as noted above, this analysis was unable to monetize all of the benefits from every response intervention as the project did not have available data. Table 5: Economic benefits of project IRR (over 10 US $ BCR NPV US $ years) Total cost (components 1 and 2) 72,436,500 Total benefit (components 1 low & 2) 508,725,296 7.0 436,288,797 21.5% Total benefit (components 1 high & 2) 962,058,124 13.3 889,621,624 29.5% Cost component 1 43,664,820 Benefit component 1 Low 498,506,520 11.4 454,841,700 27.6% Benefit component 1 High 951,839,348 21.8 908,174,528 36.1% Cost component 2 28,771,680 Benefit component 2 10,218,776 0.4 -18,552,904 -9.8% 90 IRR = (End amount / Starting amount)(1/n) – 1 https://learning.treasurers.org/resources/how-to-calculate-IRR Page 59 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS Borrower comments received by the World Bank task team were embedded in the full ICR draft that the Borrower reviewed in French (the Borrower received copies of the ICR in English and French). As such, some comments are specific to translation questions for the document that was translated from English into French. The Borrower’s comments, with World Bank responses, have been reproduced below in the form of a response matrix for ease of presentation (both in French as shared with the Borrower and in English for purposes of reporting in the ICR). Commentaires de l'Emprunteur – FSRDC Rapport d’achèvement et de résultats de la Banque mondiale RDC : Projet de prévention et de lutte contre les VBG (P166763) Section et texte/observation de Commentaire FSRDC Réponse de la Banque mondiale l'ICR Section : Page de couverture Le montant posté dans client connexion Selon les lignes directrices de la BM pour la ce jour est exactement de 86 520 000 préparation du rapport d'achèvement, on Observation : Question concernant USD. Cependant d’autres transactions doit soumettre le rapport les 6 mois après la le montant total du projet à vont se poursuivre, notamment une DRF clôture du projet, ou la même date du 30 rapporter (86,5 M$ à la clôture) d’environ 1,3 Millions en cours de mars 2024. soumission via client connexion et d’autres DPD et factures à payer en cours. Donc la situation exhaustive ne sera disponible qu’à la clôture effective desdits dossiers. L’objectif est focalisé sur le 30 mars 2024, comme clôture de la période de grâce. Section : Page de couverture Compte tenu de ce qui ressort ci-haut et Comme noté, le rapport d’achèvement doit la nécessité qu’il y a nécessité de tout être soumis les 6 mois après la clôture du Observation : Question concernant clôturer et ensuite compléter ce rapport, projet – ici avant le 30 mars 2024. la date à inclure sur la page de je suggère de mettre au 02 Avril 2024. couverture de l'ICR Section : Composantes du projet En attente du montant exact exécuté Comme noté ci-haut, compte tenu de la date (pp. 8-10, paragraphes 15-19) après clôture en cours des contrats des butoir pour le rapport d'achèvement de la activités de la composante 1. BM, on a dû s'appuyer sur les chiffres du 30 Observation : Questions juin 2023 n'ayant pas pour l'instant les concernant les effectifs du projet à chiffres finaux du projet. On a ajouté une inclure par description de note de bas de page dans la version finale du composante rapport pour expliquer. Section : Composante 2 (p. 9, Seule la Fondation Panzi est cité[e] dans Le commentaire est noté – pour contexte, la paragraphe 16) les clauses contractuelles. référence aux partenaires dans cette phrase fait généralement référence à la Fondation et Observation : Commentaire à l'hôpital de Panzi et à Heal Africa. concernant le texte suivant - ‘Les activités étaient organisées autour de deux sous-composantes : i) soutien à deux centres d'excellence spécialisés et à d'anciens partenaires dans le cadre du projet GL VBG (l'hôpital et la fondation Page 60 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Commentaires de l'Emprunteur – FSRDC Rapport d’achèvement et de résultats de la Banque mondiale RDC : Projet de prévention et de lutte contre les VBG (P166763) Section et texte/observation de Commentaire FSRDC Réponse de la Banque mondiale l'ICR de Panzi dans le Sud-Kivu, et Heal Africa dans le Nord-Kivu) ; et ii) le renforcement de la réponse à la VBG dans le secteur de la santé.’ Section : Autres modifications (p. Cela a été rendu effectif dans les 3 Le commentaire sur les activités sous la 11, paragraphe 24) provinces quelques mois avant la RMP composante 1 à intégrer au Tanganyika est [revue à mi-parcours]. La RMP avait pris en compte. Pour référence, ces détails Observation : Commentaire plutôt envisagé la mise en place des sur la restructuration ont été tirés de l’AM de concernant le texte suivant - ‘Le activités de la composante 1 au la RMP (paragraphes 8-11, texte copié dans la projet a envisagé une Tanganyika mais en vain mais les défis version du rapport ICR renvoyé au client).91 restructuration antérieure afin déjà énuméré ci-dessous ne l’ont pas d'étendre les activités à Kinshasa permis. (suite à la demande initiale de déclenchement de la composante CERC), à la province de l'Ituri et à de nouvelles zones de santé dans les provinces du Nord-Kivu, du Sud-Kivu et du Tanganyika.’ Section : Multiples références Conseil de l'Europe au lieu de Centres Bien noté et expliqué dans le texte de l’ICR d'Excellence a été noté par l’Emprunteur (pour une raison ou une autre le logiciel a Observation : Commentaires capté Conseil de l'Europe, ce qui aurait dû généraux sur la traduction faible du être traduit comme les Centres d'Excellence terme ‘Conseil de l'Europe’ au lieu ou avec le sigle CdE). de Centres d'Excellence Section : Analyse de l’efficacité de La dissolution du FSRDC a eu lieu plutôt Une traduction mal faite : On confirme que la mise en œuvre (p. 19, ‘avant’ qu'après la clôture du projet. cette version antérieure du rapport en anglais paragraphe 62) a utilisé la phrase ‘close to project closure,' en voulant dire la période avant la clôture du Observation : Commentaire projet. concernant le texte suivant - ‘La dissolution brutale du FSRDC peu de temps après la clôture du projet a été un autre facteur de déstabilisation.’ Section : Annexe 5, Commentaires Confirmer si l’équipe projet doit Selon les lignes directrices sur la préparation de l'Emprunteur s’exprimer à ce niveau. Si oui, sur quels de l’ICR, la BM souhaite avoir le point de vue points clés essentiellement ? du FSRDC ou formellement CSPP sur la Observations : Question général description des résultats et de la performance sur les commentaires de du projet dans le rapport de la BM (voir le l'emprunteur point de vue du Coordonnateur du projet qui l’a géré dans sa globalité et/ou d'autres personnes clés dans la hiérarchie au sein du FSRDC si besoin). 91 Aide-mémoire, Revue à mi-parcours, mai 2022 (paragraphes. 8-11). Page 61 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Borrower Comments – FSRDC Implementation Completion and Results Report DRC GBV Prevention and Control Project (P166763) ICR Section and Text/Comment FSRDC Comment World Bank Response Section: Cover Page The amount posted in Client Connection According to the WB guidelines for the today is exactly 86,520,000 USD. preparation of the completion report, the Note: Question regarding total However, other transactions will report must be submitted 6 months after the amount of project to report continue, including a DRF of closing of the project, or the same date of ($86.5M at closing) approximately 1.3 million currently being March 30, 2024. submitted via Client Connection and other DPDs and bills to be paid in progress. Therefore, the complete situation will not be available until the actual closure of the said files. The target is focused on March 30, 2024, as the end of the grace period. Section: Cover Page In view of the above and the need to As noted, the completion report should be close everything and then complete this submitted within 6 months of project closing Comment: Question regarding date report, I suggest that it be set to April 2, – in this case before March 30, 2024. to be included on ICR cover page 2024. Section: Project Components (pp. Pending exact amount executed after As noted above, given the cutoff date for the 8-10, paragraphs 15-19) ongoing closure of contracts for WB completion report, we had to rely on component 1 activities. June 30, 2023, figures as we do not yet have Note: Questions on project actuals the final figures for the project. A footnote to be included per component has been added in the final version of the description report to explain. Section: Component 2 (p. 9, Only the Panzi Foundation is mentioned The comment is noted – for context, the paragraph 16) in the contractual clauses. reference to partners in this sentence refers generally to the Panzi Foundation and Observation: Comment on the Hospital and Heal Africa. following text - 'The activities were organized around two sub- components: i) support to two specialized centers of excellence and former partners in the GL GBV project (the Panzi Hospital and Foundation in South Kivu, and Heal Africa in North Kivu); and ii) strengthening the response to GBV in the health sector.' Section: Other Changes (p. 11, This was made effective in the 3 The comment on the activities under para. 24) provinces a few months before the MTR component 1 to be integrated in Tanganyika (mid-term review). The MTR had instead is taken into account. For reference, these Comment: Comment on the considered the implementation of details of the restructuring were taken from following text - 'The project component 1 activities in Tanganyika but the MTR aide-memoire (paragraphs 8-11, text considered a previous restructuring to no avail, but the challenges already copied in the version of the ICR returned to to extend activities to Kinshasa the client).92 92 Aide-memoire, Mid-Term Review, May 2022 (paragraphs 8-11). Page 62 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Borrower Comments – FSRDC Implementation Completion and Results Report DRC GBV Prevention and Control Project (P166763) ICR Section and Text/Comment FSRDC Comment World Bank Response (following the initial request to listed below prevented this from trigger the CERC component), Ituri occurring. province and new health zones in North Kivu, South Kivu and Tanganyika provinces.' Section: Multiple references Council of Europe instead of Centers of Well noted and explained in the ICR text (for Excellence was noted by the Borrower one reason or another the software captured Comment: General comments on Council of Europe, which should have been the weak translation of the term translated as the Centers of Excellence or with ‘Council of Europe’ instead of the acronym CoE). Centers of Excellence Section: Implementation efficiency The dissolution of the FSRDC took place Mistranslation: It was confirmed that this analysis (p. 19, para. 62) rather 'before' than after the project earlier version of the report had used the closed. phrase 'close to project closure,' which meant Comment: Comment on the the period before the closure of the project. following text - 'The abrupt dissolution of the FSRDC shortly after the closing of the project was another destabilizing factor.' Section: Schedule 5, Borrower Confirm if the project team needs to According to the ICR Preparation Guidelines, Comments speak at this level. If so, on what key the WB would like to hear from the FSRDC or points? formally CSPP on the description of project Comments: General question on results and performance in the WB report borrower comments (meaning the views of the Project Coordinator who managed it as a whole and/or other key individuals in the FSRDC hierarchy if necessary). Page 63 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 6. SUPPORTING DOCUMENTS Project Appraisal Document, DRC – Gender-Based Violence Prevention and Response Project (P166763), World Bank, June 6, 2018. Report No: PAD2782. Project Restructuring Paper, DRC – Gender Based Violence Prevention and Response Project (P166763), World Bank, Report No.: RES53131. Implementation Status Reports: December 24, 2018; June 17, 2019; December 20 ,2019; June 30, 2020; June 29, 2021; August 5, 2022. Aide-memoire Reports: November 2018, February 2019, September-October 2019, June 2020, August- September 2020, March 202193, December 202194, May 2022 (Mid-Term Review), October 2022, February 2023, September 2023 Fonds Social Progress Reports: February 2020, August 2020, January 2021, July 2021, January 2022, July 2022, February 2023, August 2023, January 2024 (Final Report and Borrower ICR) World Bank, Country Assistance Strategy for the Democratic Republic of Congo, FY2013-FY2016. World Bank, Democratic Republic of Congo: Systematic Country Diagnostic, March 2018. World Bank, Country Partnership Framework for the Democratic Republic of Congo, FY2022-FY2026. World Bank Group Strategy for Fragility, Conflict, and Violence, 2020-2025. World Bank Group, Gender Equality, Poverty Reduction, and Inclusive Growth: Gender Strategy, 2016- 2023. DRC Ministry of Gender, Family & Children, National Strategy to Combat Gender-Based Violence, November 2009 & Revised Strategy, June 2020. Implementation Completion and Results Report, Great Lakes Emergency Sexual and Gender-Based Violence and Women’s Health Project (P147489), September 30, 2020. Independent Evaluation Group, Implementation Completion Report Review, Great Lakes Emergency Sexual and Gender-Based Violence and Women’s Health Project (P147489). Financing Agreement, Gender-Based Violence Prevention and Response Project, September 21, 2018. DRC GBV Prevention and Response Project, Project Operations Manual, December 2018. 93 December 2020 mission findings were rolled over into the AM for March 2021. 94 The aide-memoire for December 2021 was provided in draft form only. Page 64 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Process Evaluation Final Report for the DRC GBV Prevention and Response Project, Health Focus, July 21, 2023. Africa Gender Innovation Lab, The Impact of Narrative Exposure Therapy (NET) in the Democratic Republic of Congo (DRC): Preliminary Midline Results, June 2020; Improving Mental Health and Socioeconomic Outcomes of Women Survivors of Violence: Impact Evaluation of Narrative Exposure Therapy in Eastern DRC, March 2023. Final Report on NET Implementation, DRC GBV Prevention and Response Project, University of Konstanz, September 29, 2023. Final Report on NET Implementation, DRC GBV Prevention and Response Project, VIVO International, September 29, 2023. Project Monitoring and Evaluation Dashboard and Indicator Matrix (November 2019). Implementation Protocol for Component 1 of the DRC GBV Prevention and Response Project, FSRDC. FRSDC, Feasibility Study for Extension of Centers of Excellence Panzi Foundation (Bukavu) and Heal Africa (Goma): Heal Africa ASBL, SOFRECO, February 2023. FRSDC, Feasibility Study for Extension of Centers of Excellence Panzi Foundation (Bukavu) and Heal Africa (Goma): Panzi Foundation, SOFRECO, March 2023. Page 65 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 7. THEORY OF CHANGE AND RESULTS CHAIN DIAGRAMS Figure 1. Proposed Project Approach Page 66 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Figure 2. Reconstructed Theory of Change95 Components Ac vi es Outputs Intermediate Outcomes PDO Level Outcomes Long Term Impact Component : Community mobilization and Number of economic support % reported change in women s GBV Preven on promotion of behavior change bene ciaries participation in household Economic support: VSLAs, Number of bene ciaries of decision-making business skills training, income- specialized mental health care % change in help -seeking generating activities Number of service providers behavior for women and men Gender transformative training trained in NET aware of IPV Objec ve Increase Safe spaces in CBOs: par cipa on in GBV sensitization and livelihoods, preven on: psychosocial support (with NET) % reported decrease in and referrals accepting a tudes towards GBV Component 2: Sub-component 2A: Support to Number of health personnel % of rape cases accessing GBV Response multi-sectoral response services receiving training on GBV Numbers of direct project services within 72H at Centers of Excellence, services bene ciaries % of bene ciaries meeting including OSC Number of GBV cases accessing regularly with their case Sub-component 2B: at least one service manager Objec ve 2 Increase Reduce incidence of Strengthening health sector % of essential medications for u liza on of mul sectoral GBV response to GBV (trainings, PBF which no stock-out response services: approach, and PEP kits) % availability basic equipment % increase in cases receiving at health facility in line with access to at least two multi - quality disciplinary services % of small-scale works % of reported eligible GBV complying with ESMF cases receiving PEP within 72H Component : Sub-component 3A: Policy Policy, Project development and capacity - % of partners providing anagement, building (coordination, MGFC, services to survivors in line and E dissemination) with quality standards Sub-component 3B: Project % of grievances received addressed in line with quality standards management under GRM Sub-component 3C: M&E (client MIS, third -party process evaluation, and NET impact evaluation) 95Objective 3 related to crisis response capacity, which is linked to Component 4 (CERC component under the project), is not included in this results chain as language under Objective 3 was included as a standard provision to reflect the CERC component and was not associated with any indicators. Page 67 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 8. PDO AND INTERMEDIATE RESULTS INDICATOR TABLES Figure 1: PDO Indicator Results No. PDO Indicator Target Baseline End of project Achievement (as of 30 Sept 2023) Objective 1: Increase participation in GBV prevention 1. Percentage reported decrease in 20% decrease 43.7% 29.6% Achieved accepting attitudes towards GBV in targeted Health Zones 14.1% difference OR 32.3% decrease in accepting attitudes96 (162% rate of completion) 2. Numbers of direct project 785,000 0 8,614,146 Achieved beneficiaries (disaggregated for percentage of women and number of (1097% rate of Twa populations) completion) Percentage of female beneficiaries 50% 0 52% Achieved (104% rate of completion) Number of Twa community members 30,000 0 47,485 Achieved (158% rate of completion) Objective 2: Increase utilization of multi-sectoral response 3. Percentage increase in reported 0%97 50%98 53% Achieved cases who receive access to multidisciplinary services, defined as 6% increase in at least two of the following (medical, access to psychosocial, security, legal support services and livelihoods support) (106% rate of completion) 4. Percentage of eligible reported GBV 80% 13%99 100% Achieved cases who receive Post Exposure Prophylaxis (PEP) Treatment within (125% rate of 72 hours completion) 96 Percentage change calculated by dividing difference in baseline and endline values by baseline value. 97 FSRDC routinely measured this indicator against a target of 80% for access to at least two services, which facilitated data collection with implementing partners, but the results framework was never formally changed to reflect a target or to reformulate the definition. 98 Baseline estimated from data under the GL GBV Project. 99 Baseline estimated from data under the GL GBV Project (in relation to indicator for percentage of all – rather than only eligible - GBV cases receiving PEP treatment), which resulted in a lower overall percentage of cases accessing PEP treatment. Page 68 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) 5. Percentage of implementing partners 80% 0 82% Achieved providing services to GBV survivors in line with quality standards (103% rate of completion) Figure 2: Intermediate Indicator Results No. Intermediate Results Indicator Target End of project Baseline Achievement (as of 30 Sept 2023) Component 1: GBV prevention and integrated support for survivors at community level 1. Numbers of beneficiaries 3,800 0 87,711 Achieved participating in community level economic support services (2308% rate of completion) 2. Number of beneficiaries receiving 3,200 0 10,503 Achieved specialized mental health care (328% rate of completion) 3. Number of service providers trained 60100 0 265 Achieved in NET (442% rate of completion) 4. Percent reported change in women’s 20% increase 45.8% 41.6% Not achieved participation in household decision- making 4.2% difference OR 9.2% decrease in women’s participation in decision- making101 5. Percent change in help-seeking 20% increase 85.0% 88.4% Not achieved behavior for women and men aware of IPV cases at community level 3.4% difference OR 4% increase in help-seeking behavior by bystanders102 Component 2: Response to GBV 6. Number of health personnel 400 0 433 Achieved receiving training on GBV service provision (108% rate of completion) 7. Number of reported cases of GBV 60,000 0 78,466 Achieved that access at least one service 100 See note 35 (para. 41). 101 Percentage change calculated by dividing difference in baseline and endline values by baseline value. 102 Percentage change calculated by dividing difference in baseline and endline values by baseline value. For this last indicator, with a baseline of 85% and target 20% positive change, the project was effectively seeking an endline result to exceed 100%. Page 69 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) No. Intermediate Results Indicator Target Baseline End of project Achievement (as of 30 Sept 2023) supported by the project (131% rate of (disaggregated by entry point) completion) CBO entry point 0 0 52,240 CoE entry point 0 0 10,551 Legal clinic entry point 0 0 13,961 Mobile clinic entry point 0 0 1,467 Health facility entry point103 0 0 247 8. Percentage of rape cases that access 50% 0 46% Not achieved104 services within 72 hours of the incident 9. Percentage of beneficiaries who 80% 0 98% Achieved meet regularly with their case manager, as defined in the project (123% rate of manual completion) 10. Percentage of essential medication 20% 0 0 No data (PEP, STI Treatment and Emergency collected105 Contraception) for which there was no stock out during the implementation period 11. Percentage availability of basic 80% 0 0 No data equipment at health facility level in collected line with the project’s quality check- list 12. Percentage of small-scale works at 100% 0 0 No data health facility level complying with collected106 ESMF requirements Component 3: Support to Policy Development, Project Management and Monitoring and Evaluation 13. Percentage of grievances received by 100% 0 100%107 Achieved the project that are addressed in line with quality standards defined in the GRM manual 103 Data from health facilities were very limited; results reported here are reflected only from Tanganyika. 104 As noted above (see para. 50), a result of under 50% was not unusual given the significant social, economic, and logistical barriers that rape survivors face in accessing services within the 72-hour time frame for receiving PEP treatment. 105 As noted above (see paras. 51, 90), data from health facilities were very difficult to collect; ultimately, the project was unable to report on these indicators. 106 The FSRDC stopped including this indicator in its reports during implementation (as of the July 2021 report). 107 The FSRDC stopped including this indicator in its reports during implementation (as of the July 2021 report); nonetheless, the Borrower’s final progress report indicates that 100% of all complaints were resolved adequately by the end of the project. Page 70 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 9. SUMMARY OF NET IMPACT EVALUATION Summary of Narrative Exposure Therapy Impact Evaluation for the PRVBG ICR Alev Gurbuz Cuneo, Léa Rouanet, Julia Vaillant Narrative Exposure Therapy (NET) is a short-term mental health treatment that aims to reduce symptoms of PTSD (Post-Traumatic Stress Disorder) among survivors of mass violence and torture. NET was piloted as part of the Great Lakes Emergency Sexual and Gender Based Violence and Women’s Health Project (P147489), in North and South Kivu, then scaled up through the DRC – Gender Based Violence Prevention and Response Project (P166763). A rigorous impact evaluation of NET was embedded in the project to measure its short- and medium-term impacts on the mental health and socioeconomic outcomes of survivors of sexual violence in North and South Kivu. NET treatment recognizes that trauma survivors undergo multiple distressing events within their lifetime. Focusing on the traumatic events, the patient constructs a chronological narrative of her life story and a coherent narrative is then structured with the assistance of the therapist. NET’s approach of systematically unpacking multiple traumatic experiences in an autobiographical structure is expected to achieve positive outcomes for survivors who have been diagnosed with PTSD and other trauma-related disorders. The therapy consists of 8 to 12 individual sessions of approximately 90 minutes each. In the Great Lakes project, NET was delivered by counselors in the Center of Excellence Heal Africa, in health centers, and in Community-Based Organizations (CBO). To evaluate the impact of NET, we used a randomized controlled trial method. Women were screened in cohorts of 12 by counselors in health centers or CBOs and deemed eligible for the study if they were diagnosed with PTSD.108 Although being a survivor of conflict-related violence was not an eligibility condition, the intervention was designed to address the mental health needs of women survivors of sexual violence. Eligible individuals in each cohort were randomized into receiving NET now or receiving NET 15 months later. A baseline survey was conducted under the Great Lakes Project between June 2017 and October 2019, including 525 women in the treatment group and 528 in the control group. The midline and endline surveys were completed by the GBV Prevention and Response Project in May 2020 and February 2021 respectively. 979 women out of the initial 1053 were found and re-interviewed at endline (7% attrition rate). The midline and endline surveys were designed to measure impacts of NET respectively three and 12 months after the treatment group had received NET. Data were collected by enumerators hired by the Project Implementation Unit (PIU), and data collection was supervised jointly by the PIU and GIL researchers, with the presence of a GIL consultant on the ground. The study assessed the impact of NET on PTSD, depression, anxiety, and a range of secondary psychosocial and economic empowerment variables. The findings show that receiving NET resulted in an immediate improvement in the mental health of beneficiaries with significant decreases in their PTSD and depression and anxiety scores and in their likelihood of having probable PTSD and probable depression or anxiety (see Figure 1). In the short-term, women eligible to receiving NET were 13.1% less likely to be suffering from PTSD and 8.4% less likely to be suffering from depression and/or anxiety than the control group. This immediate impact on mental 108 Women with PTSD but in urgent need of treatment were excluded from the study and prioritized to receive NET. Page 71 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) health disorders in the short term was sustained in the medium term, around 12 months after completion of their NET treatment. At endline, the impact of NET on mental health outcomes of beneficiaries are the same as at midline: PTSD and depression and anxiety are significantly lower in the NET group than in the control group. The theory of change of NET posited that mental health improvements could be observed immediately, whereas impact on economic and psychosocial outcomes would take longer. Therefore, we examine impact on psychosocial and economic outcomes at endline only. NET beneficiaries have significantly higher self-esteem at endline and a lower score on their local functioning impairment. There are no significant impacts on management of problems, network, mobility, or social group participation at endline. A rather unexpected finding is that NET beneficiaries work significantly fewer hours (2 hours) per week compared to the control group, but this does not translate into lower earnings. NET beneficiaries are also significantly more likely to have saved money in the past 6 months. Lastly, NET beneficiaries have more decision-making power in the household than women randomized to the control group. NET was initially delivered through the DRC government health system in the Center of Excellence Heal Africa and in health centers. After it became clear that implementation through the health system was stalled, the project shifted to a CBO-based delivery of NET, with non-professional, psychosocial assistants, delivering the therapy to survivors. A heterogeneity analysis of the impact of the NET by delivery mode revealed that delivery through CBOs was at least as efficient as through the health system. This has important implications for the scalability of the intervention: the study shows that NET can be delivered by community-based non specialists as effectively as when it is administered by healthcare professionals. This is particularly relevant in the DRC and similar FCV contexts, which suffer from a dire lack of mental health professionals. Taken together, results of the study suggest that receiving NET had significant and sustained impacts on the mental health outcomes of beneficiaries. While these were associated with some improvements in psychosocial outcomes, they did not translate into meaningful impacts on the economic empowerment of beneficiaries. Figure 1: Effect of NET on PTSD and Depression/anxiety scores at midline and endline Page 72 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 10. PROCESS EVALUATION Excerpts Adapted from Introduction, Discussion, Lessons Learned, and Conclusion Sections to the Final Report Process Evaluation from Health Focus / Médecins du Monde Translated from French into English (with assistance from DeepL and WBG Translate) TRANSLATED INTRODUCTION SECTION: The Health Focus final report presented the triangulated results based on qualitative and quantitative data, as well as key recommendations from the process evaluation of the Gender-Based Violence Prevention and Response project (PRVBG). The evaluation drew on the findings of the KAP baseline and endline studies, as well as on the results of three qualitative visits to the provinces of Sud-Kivu, Nord-Kivu and Maniema, and two qualitative visits to the province of Tanganyika. The results of this evaluation demonstrated a positive impact of the PRVBG project on attitudes towards gender-based violence (GBV). It would therefore be essential to maintain collaboration with stakeholders and mobilize the resources needed to sustain the work started by the PRVBG project. This report had several objectives. Firstly, it concisely but comprehensively presented the results of the qualitative and KAP studies, which covered six themes assessed, in four provinces (North Kivu, South Kivu, Maniema and Tanganyika), over a period extending from March 2021 to May 2023. Full details of each study are available in the corresponding specific reports. This first section also included the main recommendations arising from the results, which are more of an operational nature. The second part of the report addressed strategic issues and their implications, based on the results, discussions, team expertise and field experience. The aim of this part was to serve as a basis for improving the implementation of future similar programs and ensuring sustainable results. Finally, the last part of the report offered reflections on the implementation of process evaluation, as well as on the successes and challenges encountered. This section aimed to identify lessons learned and provide recommendations for similar process evaluations in the future. 1.1 Project background Gender-based violence (GBV) is a dominant and debilitating problem in the Democratic Republic of Congo (DRC), ranked 151st out of 170 countries according to the 2021 Gender Inequality Index. Overall, 52% of women aged 15 to 49 said they had been victims of physical violence (by an aggressor of any kind) at least once since the age of 15, while 27% had experienced sexual violence. For women aged 15 to 49 who had suffered physical violence, the aggressor was most often a husband or current partner (56.8%). As such, the GBVRP project is designed to address four components, namely 1) GBV prevention and support for survivors integrated at community level, 2) response to GBV, 3) support for policy development, project management and monitoring and evaluation, and 4) emergency response. This process evaluation focused on Component 1 of the PRVBG. Component 1 activities focused on GBV prevention and integrated support for survivors at community level. These activities therefore aimed to tackle the root causes of GBV and, consequently, the social norms and values on which they are based at Page 73 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) individual, family and community level. This included awareness-raising and behavior-change communication activities, which focused not only on women and girls, but also on men and boys. A second important activity was support for income-generating activities (IGAs), which benefited CBOs, pools of community activists and SASA activists, and Village Savings and Loan Associations (VSLAs) to strengthen the social and economic empowerment of women and girls. The project also sought to address the problem of insufficient or inadequate medical, psychosocial and mental health services in the community response to GBV, as well as legal support services, by providing decentralized responses and ensuring adequate referral to specialized service providers. In this way, the PRVBG project aimed to produce sustainable change for beneficiaries by supporting, through training and essential resources, notably Umbrella NGOs, health facilities, community-based women's organizations (CBOs) and community-based project actors, including focal points, paralegals and community mobilizers. 1.2 General objectives of the process evaluation The aim of the process evaluation of the PRVBG was therefore to a) collect and analyze data gathered by the various implementing entities between June 2021 and March 2023, b) assess the quality and accessibility of services in terms of use by beneficiaries, c) obtain an independent overview of the level of satisfaction of service users and d) carry out a detailed analysis of attitudes of acceptance of GBV among beneficiaries and service providers. During the implementation of the evaluation, the Health Focus/Médecins du Monde (HF/MdM) consortium produced periodic reports providing suggestions and recommendations to the FSRDC for the improvement, ownership and sustainability of the quality of services provided by the project. In accordance with the terms of reference, the evaluation focused on six predefined research themes specific to the PRVBG project. These thematic areas were: 1. Capacity-building and support models 2. Community interaction 3. Quality of service delivery 4. Livelihood activities 5. Attitude and behavior change 6. Sustainability of interventions Each of these themes was developed using specific and cross-cutting research questions. Box 1 : Cross-cutting research questions 1. How is the program implemented? 2. Is the program being implemented as planned? 3. What practical problems have been encountered and how are they being resolved? 4. What types and quantities of services are provided? 5. How are resources used to provide services? 6. Is the program accessible and acceptable to the target population? 7. Does the program integrate and involve particularly marginalized groups, including indigenous populations (Twa)? Page 74 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Indeed, this research process, and the conclusions and recommendations arising from it, focused on three fundamental elements that underpin process evaluations, namely 1) fidelity (whether the project is carried out as planned), 2) dose (how much of the intervention is implemented) and 3) reach (whether and how the intended target groups are reached and benefit from the project). Another key element emerging from process evaluations was context, in which the evaluator seeks to better understand anything that is considered “external” to a project or intervention, and which may hinder or facilitate its implementation or effects. The following general research questions (Box 1), which include “fidelity”, “dose”, “scope” and “context”, are addressed. 1.3 Objectives of the qualitative study The qualitative study provided crucial information for project operations, as it enabled partners to assess how implementation was proceeding on the ground, identifying potential gaps in training, understanding of procedures and engagement of communities, particularly excluded and marginalized groups. It also enabled teams to understand contextual factors that can influence the pace or quality of implementation, such as local community dynamics, which are more difficult to capture in a quantitative study like the two KAP surveys conducted. Of the six thematic areas, topics 1 to 4 and 6, as well as cross-cutting issues, were covered by the qualitative study. 1. Capacity-building and support models: • Are there enough existing CBOs in the new project areas with which to partner to launch project activities, or how is the process of creating CBOs in the new project areas being implemented? • What requirements need to be in place for the CBO to implement project activities in the communities? • What technical assistance and financial/operational support is required from to NGOs? • How do NGOs, centers of excellence and health services work together to coordinate service delivery? • What feedback is received from the Complaints Management Mechanism (CMM) or other sources of feedback related to project implementation? 2. Interactions with the community: • How do the CBOs interact with community members? • What is the perception of CBO members and women involved in CBO activities about their own status or decision-making power within the community? • Are the prevention activities of the community mobilizers being implemented as designed in the project operations manual? • What mechanisms exist for the resolution of complaints by the community? • Do community members know how and where to submit comments or complaints about project activities? 3. Quality of service provision at community level: Page 75 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) • What services do survivors request, and what is the availability of these services at community level? • Are psychosocial services provided by CBO focal points implemented as designed in the operations manual? • Are medical services being implemented as envisaged in the operations manual? • Are legal services being implemented as envisaged in the project operations manual? • What is the level of satisfaction of survivors accessing these services? • How do they improve the overall functionality and well-being of survivors? 4. Livelihood activities: • How profitable are income-generating activities and business plans at organizational level for CBOs? • Are individual-level livelihood activities for CBO members (income-generating activities) being implemented as planned in the operations manual? • What is the financial and social reintegration impact for participants in these activities? 6. Sustainability of interventions: Comparison between newly-created CBOs and those that have been operating as service providers for several years • What is the capacity of existing CBOs at the start of this project, having already participated in previous projects, compared with newly created CBOs? • How does the previous experience and training of existing CBOs affect organizational capacity and the quality of service delivery? • How does the organizational capacity of CBOs change over the project period? • Can CBOs operate independently at the end of the project, or what do they need to do so? The research questions relating to each topic gathered information on how operational procedures were followed, and on systemic or procedural problems, in order to suggest suitable solutions or process modifications. The research questions also addressed the appropriateness of the GBVRP project design and procedures for ensuring that GBVRP project activities are accessible to survivors. In addition, the research included a comparative analysis of the quality of implementation at new and existing project sites. The information gathered will help refine project procedures for use in other locations and/or similar future projects. Methodology of the qualitative study The qualitative study was carried out with the aim of assessing the project's implementation in the field, identifying potential gaps in training, understanding of procedures and community involvement, particularly among excluded and marginalized groups. It also enabled us to understand the contextual factors that can influence implementation in terms of pace and quality, such as local community dynamics. A variety of data collection tools were used, including key informant interviews (KIIs), focus group discussions (FGDs) and observations. The selection of participants included key informants such as FSRDC members, implementing partner coordination staff, health professionals, representatives of community stakeholder groups, NGO Umbrella supervisors and service beneficiaries. Focus groups were held with different groups, such as community members, CBO and AVEC members, displaced populations and indigenous populations. Page 76 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Pilot studies were carried out to verify the composition of tools and research methods. Data quality assurance mechanisms were put in place, including regular debriefing sessions with field teams. In-depth analysis of the transcripts was subsequently carried out using qualitative analysis software. The complete methodology for collecting and analyzing qualitative data from the field was described in the annual reports and in the qualitative analysis reports. 1.4 Objectives of the KAP quantitative study Main objective The main objective of the KAP study was to assess the impact of the PRVBG project in terms of behavioral and attitudinal changes in the context of gender-based violence in the project’s selected health areas in the provinces of North Kivu, South Kivu and Maniema. Tanganyika province was not included in the KAP survey due to the specific characteristics of the PRVBG project in this province, notably the absence of Component 1. The KAP survey covered thematic area “5 – Attitude and behaviour change” and thematic area “2 – Community interaction.” The latter was also covered by the qualitative study. Guided by the impact indicators of the PRVBG project, the specific objectives of the quantitative research were as follows: 1) To assess changes in attitudes and acceptance of gender-based violence in the target health areas between the baseline and final evaluations. • Percentage of service providers expressing attitudes of acceptance of GBV. • Percentage of community activists expressing attitudes of acceptance of GBV. • Percentage of participants in prevention activities who express attitudes of acceptance of GBV. • Percentage of community members expressing attitudes of acceptance of GBV. 2) Assess the percentage of change in women's participation in household decision-making between the baseline and final assessments. 3) To assess the percentage change in help-seeking among men and women aware of domestic violence at community level between baseline and final assessments. Secondary objective The secondary objective of the KAP study was to answer selected research questions relevant to research theme “2 – Interactions with the community”: • What is the community’s perception of the services and activities of the community-based organization (CBO)? • How do CBO members and women involved in CBO activities perceive their own status or decision-making power? Page 77 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Tertiary objective The third objective of the KAP study was to answer the following four cross-cutting research questions: • How is the program implemented? • Types and quantities of services provided? • Is the program accessible and acceptable to the target population? • Does the program integrate and involve particularly marginalized groups, including indigenous populations (Twa)? KAP study methodology The research team comprised experts from Health Focus and the NGO Marakuja. HF was responsible for study design and planning, coordination, quality assurance of training, supervision of data collection and analysis of results. Marakuja was responsible for recruiting data collection staff, organizing training and implementing data collection. KAP study sites were selected on a stratified basis, taking into account criteria such as security situation, proximity to the chief town of the health zone, CBO status, presence of vulnerable/marginalized groups and accessibility. Participants were selected at random from each site, using the Lot Quality Assurance Sampling (LQAS) method. Different categories of participants were targeted, including community members, service users, community activists and service providers. The main data collection tool was a structured quantitative questionnaire, developed in line with the objectives of the PRVBG project. The questionnaire covered several areas, such as women's participation in household decision-making, attitudes towards gender-based violence, helping behaviors, and interactions with the community. Pilot studies were carried out to test the data collection tools and finalize the guides. Data quality assurance mechanisms were put in place, including plausibility checks and regular debriefings with field teams. With regard to the analysis of results, the approach comprised two main approaches: univariate analysis and difference-in-difference (DiD) analysis. The univariate analysis examined the changes observed in the PRVBG project indicators between the baseline and endline studies, using Chi-square tests to determine significant differences. It also compared percentage changes between control and intervention samples. Next, the DiD analysis assessed the causal impact of the PRVBG project by comparing changes in outcomes between the intervention group (receiving the intervention) and the control group (not receiving the intervention) over time. Adjusted and unadjusted DiD models took into account covariates such as geographic location, gender, age, household size, level of education, type of participant and proximity of the CBO to the chief town of the health zone. Interaction tests were also carried out to better understand the impact of the intervention on the indicators and the possible interactions between the covariates and the intervention. Page 78 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) The complete methodology for collecting and analyzing quantitative data from the field was described in the related reports for the two phases of the KAP study (baseline and endline), as well as in the annual reports. 1.5 Data collection The project ran from March 2021 to May 2023, with specific data collection for the KAP study (baseline then endline) and qualitative data collection comprising three visits for North Kivu, South Kivu and Maniema and two visits for Tanganyika province. For the KAP study, data collection was carried out in the provinces of South Kivu, North Kivu and Maniema. Tanganyika province was excluded from KAP data collection due to the absence of community activities in the volatile security context. For the qualitative study, the first and last visits covered the four provinces included in the PRVBG project, while the second visit took place in the provinces of South Kivu, North Kivu and Maniema. Data collection for the KAP study Data collection for the KAP baseline study began in September 2021, with the simultaneous deployment of investigators in the provinces of North and South Kivu, and continued in the province of Maniema during October 2021. For the KAP Endline study, after reflection and discussion between the different teams, it was decided by mutual agreement to voluntarily delay data collection for the KAP Endline study in order to maximize the time between the two data collections and thus allow for any changes to be observed. Thus, preparations began in December 2022 in collaboration with Marakuja. Following fighting between state armed forces and the M23 armed group in two PRVBG project intervention zones and one control zone in North Kivu, a reassessment of the security situation was necessary, and it was decided to withdraw three health areas from the collection plan, thus reducing the number of health areas to 17 for the three provinces. Collection was therefore able to take place in North Kivu in February 2023, in South Kivu from late January to February 2023 and in Maniema from late January to early February 2023. Data collection for the qualitative study With regard to qualitative data collection, the first pilot collection took place in November 2021 in the health area of Kaniola, South Kivu. Data collection for the first visit then began at the end of November 2021 and ended at the end of February 2022 in North and South Kivu, and continued until March 13, 2022 in Tanganyika and March 19, 2022 in Maniema. The second visit took place in North Kivu from late May to early July 2022. In South Kivu, the team of evaluators carried out the second visit between May 23 and the end of June 2022. In Maniema, the second visit began in mid-July 2022 and ended in mid-August 2022. At the time of this second visit, livelihood activities had not yet begun in Maniema, including the establishment of VSLAs. Consequently, in consultation with the FSRDC, the themes of this visit were adapted accordingly. It was decided not to Page 79 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) include theme 4 (livelihoods) and to focus instead on themes 1 (capacity building) and 2 (community interaction), with as much attention as possible paid to theme 3 (quality of services). The third wave of data collection began in mid-November 2022 and ended in early December 2022 for the provinces of Sud-Kivu and Nord-Kivu. In Maniema, the third visit took place in the second half of January 2023 and lasted until February 19. In Tanganyika, the second round of qualitative data collection took place in January and February 2023. These successive visits enabled qualitative data to be collected throughout the duration of the PRVBG project, focusing on different themes and adapting certain aspects according to the specific circumstances of each province. 1.6 Description of samples surveyed Description of the KAP study sample The KAP study involved a total of 1611 participants in the baseline phase and 1380 participants in the endline phase, comprising both intervention and control groups. Of the individuals included in the baseline phase, 75% were in the intervention group, while 25% were in the control group. In the final phase, 70% of participants were from the intervention group and 30% from the control group. The gender breakdown of participants showed a higher proportion of women (around 55%) than men (around 45%), partly due to the majority of women among service providers (63% at baseline and 66% at endline). There were also more women among participants in prevention activities, with 70% being women at baseline and 55% at endline. The male-to-female ratio among community members, however, was equal to 50:50 at baseline and endline. Table 1 : Description of the global baseline and endline study sample Baseline Endline All Health Areas All Health Areas N % N % p-value Total 1611 100% 1380 100% Type of participant Members of the community 946 58.7 604 43.8 Participants in prevention activities 316 19.6 364 26.4 Community activists 121 7.5 141 10.2 Service providers 228 14.2 271 19.6 0.000 Gender Women 885 54.9 747 54.1 Men 726 45.1 366 45.9 0.660 Province South Kivu 645 40.0 646 46.8 North Kivu 566 35.1 326 23.6 Maniema 400 24.8 408 29.6 0.000 * A value p≤0.05 indicates evidence of a significant statistical difference between the compared groups. Page 80 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) Description of the qualitative study sample It is essential to emphasize the breadth of the data collected during the qualitative part of this process evaluation carried out over a two-year period. It is remarkable to note that no fewer than 317 focus groups were organized, testifying to a significant amount of qualitative data collected. A total of 1,930 people took part in the qualitative data collection activities, either in individual interviews or group discussions. This large number of participants reflects the diversity and richness of this study. This in-depth and varied qualitative data is an invaluable resource for assessing the impact of the PRVBG project, understanding the mechanisms of change and identifying successes, challenges and prospects for improvement. The considerable amount of qualitative data provides a solid basis for analysis and recommendations. Table 2 : Number of FGD organized by province and by visit South Kivu North-Kivu Maniema Tanganyika Visit 1 27 26 21 12 86 Visit 2 41 39 21 101 Visit 3 43 41 30 16 130 TOTAL FGD 317 Table 3 : Number of persons questioned by province and by visit South Kivu North Kivu Maniema Tanganyika Visit 1 182 184 150 97 613 Visit 2 206 212 156 574 Visit 3 235 220 180 108 743 TOTAL PERSONS QUESTIONED 1930 TRANSLATED DISCUSSION SECTION: This section highlights elements that were consistently emphasized across the themes and that underlie broader considerations. It therefore has a strategic dimension and presents the strengths and limitations of the PRVBG project. It can serve as a basis for future reflections and improvements aimed at increasing the effectiveness, efficiency and sustainability of similar projects. The lessons learned from the PRVBG project can thus be used as valuable guidance to guide future efforts in the prevention and response to GBV. Strengths The implementation of the PRVBG project stands out for its global and holistic approach, being among the most comprehensive strategies encountered at this level in the DRC. It provides a relatively comprehensive overview of the complexity of gender-based violence in DRC, with a focus on survivors' needs. This ambitious program aims to expand and improve prevention measures and responses to GBV, with a key priority given to the well-being and empowerment of survivors. Page 81 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) One of the strengths of this project is its thoughtful design, with a focus on inclusion and protection of vulnerable populations, including GBV survivors and other vulnerable women. By adopting a community- based approach, the project seeks to empower women in their communities and households. It also recognizes the importance of the involvement of men in efforts to combat GBV, promoting a collective response to effectively address this issue. The project also stands out for its commitment to scientific research and processes. The allocation of resources to a comprehensive and in-depth evaluation of the multi-pronged PRVBG process provides a means to collect good practices and lessons learned on a variety of key themes. This commitment to learning and improvement is invaluable for future GBV and related interventions in complex environments. It fosters collaboration between organizations from different sectors and geographic regions, fostering knowledge exchange and enhancing the overall effectiveness of GBV efforts. FSRDC's willingness to learn and adapt the PRVBG project based on the results of the process evaluation demonstrates a commitment to continuous improvement and a proactive approach to addressing the challenges encountered in preventing and responding to GBV. It is critical to recognize that the PRVBG project was carried out under difficult circumstances, both domestically and globally, including due to the ongoing conflicts in eastern DRC and the global COVID-19 crisis. These external factors inevitably had an impact on the implementation of some aspects of the project. However, the commitment and perseverance shown by the various teams involved in the PRVBG project to overcome these challenges deserve to be commended. The selection of intervention zones highlights a strategic approach that combines areas with prior experience in GBV programming with newer areas located both inside and outside the Kivu provinces. This approach makes it possible to introduce GBV programming and CoE actions in new areas and provinces, thus contributing to the expansion of GBV efforts and the promotion of survivor-centered responses. In its oversight role as a government entity, the FSRDC demonstrates outstanding expertise and experience in the sector. The successful implementation of the project demonstrates full national ownership and commitment to this crucial area of work, providing a model for potential similar initiatives in the future. Limitations Operational challenges and constraints The implementation of this vast project has faced some major challenges due to the complexity and sensitivity of the context in which it is deployed. On the one hand, there are challenging operating environments, while on the other, sensitive areas of work must be managed. In addition, a set of organizations and stakeholders must be brought together, each with their own processes and expectations for the PRVBG project, as well as varying levels of experience. These factors add complexity and require a strategic and adaptive approach to overcome the obstacles encountered throughout implementation. Delays in the implementation of various elements and in different geographical areas hindered the full Page 82 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) potential of the PRVBG project from its inception. This is essential to take into consideration when evaluating the project's results and impact. Overall, the different components of the PRVBG project did not have the same opportunities to set up and implement their specific parts of the intervention, which had obvious repercussions on the overall impact of the PRVBG project activities within and between the different zones. As previously mentioned, a robust supervision system has been put in place, across sectors and activities, to support capacity building of community service providers as well as project implementers. However, it is important to note that Component 1 of the PRVBG project did not receive sufficient funding to fully conduct regular monitoring and visits, as well as supervision. In addition, it should be highlighted that communication channels were not sufficiently transparent and consistent, which limited the ability to provide effective remote support. For example, supervisors may encounter difficulties with telephone connectivity and some numbers may be inactive without notification, which hinders the fluidity of exchanges and the availability of the necessary support. Mixed effects on empowerment of women members of CBOs One of the underlying tenets of community women’s empowerment was the idea that CBO members develop skills and take on new roles, in order to boost their self-confidence, improve their self-esteem, and increase their decision-making power in their homes and communities. However, the process evaluation revealed mixed results. Although CBO members have reported greater self-confidence and there is an improvement in women's perception of CBOs in the community, where they are more respected, the expected benefits in terms of women’s increased participation in household decision- making are not clearly demonstrated. One possible hypothesis is that women’s work for CBO does not generate direct monetary income, and time spent on CBO activities is time not spent generating family income, which can lead to intra-household conflict and indirectly reduce women's decision-making power. It is therefore essential that targeted research be conducted by the World Bank and others to promote women’s empowerment through similar community-based capacity building mechanisms. This would allow for a better understanding of the pathways, opportunities and limits to empowering women in communities and households through their participation in such interventions. Sufficient means of implementation The use of unpaid volunteer work is an important consideration and needs further study. This programmatic approach is based on several underlying assumptions: first, that the work will be done by the community itself; second, that the work and activities will be sustainable even after the end of the external funding; Third, that communities will be able to generate income to finance their activities, thus ensuring their sustainability independently of an external funder. However, the results of the study challenge these assumptions. The expected high and specialized workload of community actors under Component 1 of the PRVBG project leads to a limitation of their availability for IGAs within their own households, which can have negative repercussions on the people involved in the project as well as their families. In this context, it is also legitimate to question the amount of time that can be required of them. In order to optimize their contribution, it would be advisable to include in the PRVBG protocols a clarification Page 83 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) regarding the time required to fulfill their role, possibly based on existing data in order to better quantify the time needed to allocate to these activities. At the same time, it is essential to reassess the complexity of the tasks expected of community actors, including documentation and reporting, to determine their relevance and usefulness. Challenges were identified regarding the completion of the tracking sheets by focal points, as well as the high number of them. This calls for consideration of whether the additional administrative work is warranted and for simplifying documentation processes to facilitate monitoring and learning. In addition, the specialized nature of some tasks requires more in-depth and ongoing training, particularly for community mobilizers, focal points, paralegals, and other community project actors involved in community and beneficiary-related activities. It has been observed that community mobilizers in charge of outreach and facilitation of EMAP groups may lack in-depth knowledge in the thematic areas being addressed, which limits their ability to adequately respond to questions or resistance expressed by community members. This gap hinders the potential of advocacy activities. Iterative risks Various consequences, sometimes negative, of the PRVBG project have been identified, having an impact on the actors responsible for implementing the activities. The studies have highlighted several significant examples in this regard. Of particular concern are issues related to insecurity, including fears of retaliation expressed by paralegals and focal points for their role in supporting the legal and judicial process, as well as security concerns reported by CBO members, such as removal and destruction of CBO and PRVBG signs in some areas. Another important aspect is the lack of tools for focal points to manage their own emotional burden related to supporting survivors. They need support to cope with these challenges and be able to provide quality assistance to victims. In addition, the lack of sufficient resources for the care of victims leads to compensation on the part of CBO members. For example, some CBO members have held sessions in their homes or even hosted survivors of SGBV. Others advanced transportation costs, provided food or clothing to meet their immediate needs. The hosting of survivors by focal points exposes CBO members to significant personal risks, both in terms of safety and marital relations. The lack of remuneration for the work of CBO members is also a concern, as it creates tension within their households. To address these issues, it is recommended to support CBO members so that they have a clearer understanding of the risks they may face within their community or household, thereby enabling them to take appropriate safety measures to protect themselves. This can be achieved by offering training, workshops, or awareness sessions focused on risk assessment and safety measures. By equipping them with this knowledge, they will be able to make informed decisions and take the necessary precautions to mitigate potential risks. Accessibility of services The impact of distance on the services offered and their accessibility is another point to consider, especially in remote areas. Challenges related to transport and logistics are identified as a major obstacle to the implementation of the PRVBG project activities in these regions. It is therefore essential to find appropriate solutions to reach these areas and ensure the provision of necessary services. In this context, Page 84 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) a suggestion was made concerning the use of the community relays already present as community mobilizers. These local actors can play a critical role as intermediaries, passing on relevant information and mobilizing the community. By relying on these community relays, it becomes possible to overcome distance constraints and ensure that services reach remote areas efficiently. The PRVBG project was designed in a complex and multi-sectoral way to support the medical care of survivors, taking into account various aspects, including transportation to medical facilities for surgeries. It is crucial to consider post-surgical/medical care and provide appropriate support, including transportation. Accepting beneficiaries to walk or ride long distances on motorbikes can lead to new medical risks. Therefore, it is essential to put in place measures to ensure safe and adequate transportation for survivors, in order to minimize additional risks to their health. In the same vein, the lack of availability of post-exposure prophylaxis (PEP) kits has been repeatedly identified as one of the problematic findings. This challenge has a direct impact on the medical care provided to survivors and represents a weakness of the PRVBG project. This requires special attention and targeted actions to ensure adequate availability of PEP kits to ensure optimal medical care for survivors. Inclusion of specific populations With regard to the involvement of different target populations, several observations can be made. Despite the measures taken and the importance given to the inclusion of vulnerable populations, such as Indigenous Peoples (IPs) and IDPs, during the design of the PRVBG project, it is crucial to recognize that the project has not been able to fully achieve this objective in practice. Issues related to the inclusion of indigenous peoples have been identified and are discussed in detail in the first part of the report. It is concerning that some IP people do not feel included and even reject the PRVBG project. One of the major problems is the limited accessibility to the services of the PRVBG project, in particular the VSLAs. Revisiting approaches and mechanisms to engage with these populations is critical to overcome barriers and respond more appropriately to their specific needs. It was pointed out that younger women were less likely to attend the services of the PRVBG project, which raises the issue of awareness and engagement of this population layer. Additional efforts need to be made to sensitize young women to the services offered by the PRVBG project, for example through targeted awareness campaigns, using communication channels suitable for young women such as social media, schools or community spaces frequented by this population. In addition, it is crucial to create a safe and inclusive environment for young women so that they feel comfortable seeking help and participating in the activities of the PRVBG project. Similarly, although the PRVBG project judiciously implements initiatives to involve men in efforts to prevent and respond to SGBV in the intervention areas, including through EMAP and other awareness- raising activities, it appeared that some men felt neglected by the PRVBG project and expressed feelings of jealousy towards women benefiting from IGAs. In order to solve this problem, it is necessary to increase awareness in the community and to involve men more in the project. Several suggestions were made, such as strengthening outreach activities towards men so that they understand why the PRVBG project gives priority to activities towards women. A couple-based approach, which integrates the needs and interests of both sexes, can also be an effective strategy to foster more equitable participation. In addition, Page 85 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) it may be useful to reflect and consider that men also benefit from operations for specific health conditions, such as hernias, to increase acceptance of the project and its GBV messages, using a transformative health approach. Men's involvement is critical to promoting healthy relationships, challenging harmful social norms, and encouraging equal participation in violence prevention. Collaboration between different actors Still with regard to ownership of the PRVBG project, the active involvement of key community actors in the project is often highlighted as crucial to ensure its proper functioning. This can be achieved in particular by holding regular meetings under the collaboration mechanism, which bring together different actors and representatives of the community. These meetings help coordinate efforts and facilitate information sharing. In addition, it is essential to promote interaction among community service providers, including between focal points and paralegals. This interaction can take place within the same health zone or between different health zones and allows different service providers to foster coordination, collaboration and exchange of experiences. This improves the quality of interventions and better responds to the needs of beneficiaries. Missed opportunity In addition, it is crucial to highlight that the GRM of the PRVBG project has many shortcomings, which limit its proper functioning. The GRM is an essential prevention tool and promotes a "do no harm" approach. The dysfunction of the GRM is worrisome because it prevents the collection and effective management of complaints from persons affected by the PRVBG project. This means that people's problems and concerns are not properly addressed and resolved, which can have negative consequences for their well-being and engagement in the project. In addition, the establishment of a robust GRM would have allowed valuable lessons to be learned from the experience of the PRVBG project and adjustments and improvements made in real time. Unfortunately, due to its weaknesses, the GRM has not been able to play this crucial role fully. This is therefore a missed opportunity to learn from the project throughout its implementation. Finally, the project was based on the assumption that the success of collective IGAs would be crucial to ensure the financial self-sufficiency of CBOs and mobilizer pools and for the continuity of their activities. Unfortunately, this assumption did not hold due to implementation, supervision, and training issues related to the management of collective IGAs. However, it is important to distinguish between the management of IGAs by CBOs and by community mobilizer pools. It was observed that IGAs of CBOs were, in general, better managed than those of community mobilizers, where there was a lower level of ownership and accountability towards the IGAs awarded. This disparity may be due to the demotivation of most community mobilizers and paralegals due to insufficient compensation for their work. These actors tend to. This disparity may be a result of the demotivation of most community mobilizers and paralegals from not receiving adequate compensation for their work. This group of actors has shown a tendency to focus on quick wins rather than long-term investments in their structures and services. This is an important aspect to consider when planning similar interventions in the future. Page 86 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) In conclusion, it is crucial to ensure the regular generation and especially the proper management of revenues to ensure the sustainability of the PRVBG project activities. This is essential to maintain the involvement and participation of service providers, and thus ensure a certain sustainability of the activities beyond the project closure. Discussion of the results of the KAP study The KAP study was conducted to assess the impact of the PRVBG project on attitudes towards GBV and women's participation in decision-making. The results of the study show that the community-based approach adopted under the PRVBG project has the potential to change harmful attitudes towards women at the population level. In addition, the results indicate encouraging prospects. However, several methodological aspects must be considered for the interpretation of the results. Limitations and strengths of the KAP study There are several limitations to the KAP study that need to be considered. First, it was conducted in a relatively small number of health areas (HAs) compared to all HAs served by the PRVBG project. The indicators were assessed in only 15 intervention HAs while the PRVBG project was implemented in 267 HAs distributed among them. Therefore, the findings of the study should be interpreted with caution and be limited to areas where access and security are reasonable, thus excluding areas with high security issues or geographical inaccessibility that are not represented. Three HAs were lost in Nord-Kivu as part of the endline study due to the armed conflict, resulting in the unavailability of end-of-follow-up data for these HAs. Sensitivity analyses were conducted to assess the impact of this loss and the results showed no significant change. It should be highlighted that the sampling and recruitment strategy of the KAP study followed a strictly defined stratified random sampling approach and protocol, ranging from the selection of health zones, health areas within the selected zones, villages to the household and individual participant levels. Moreover, although the sample size is large enough to detect changes, the conclusions are based on a limited number of clusters. This can influence the generalization of results to the general population. It should also be noted that the baseline study was carried out at a time when project activities had not yet begun in Maniema. On the other hand, activities were already underway in North and South Kivu at the time of baseline data collection. This raises questions about the representativeness of the findings of the baseline study for “unexposed” attitudes and behaviors in the context of gender-based violence. In addition, the 14-month time period between baseline and endline can be considered relatively short to achieve significant behavior change at the population level, given the nature of behavior change programs such as the PRVBG project. It is also important to note that the study did not collect direct data on gender-based violence, which was a deliberate decision. Despite these limitations, the study also has strengths. The inclusion of a control sample allows for meaningful comparisons between baseline and final study data, which is a major asset for the impact evaluation of the PRVBG project. For PRVBG actors such as community mobilizers and service providers, Page 87 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) all identifiable individuals working in the HAs covered by the KAP study were interviewed. Therefore, the results in these subsets have a high level of validity. Interpretation of indicator results Secondly, it is important to take into account several considerations when evaluating the results presented in this report regarding the PRVBG project indicators. The pre-established targets for each indicator were set before the start of the project and we believe that the success or failure of the PRVBG project should not be determined solely on the basis of these raw values. First, it should be noted that changes in attitudes are often observed earlier and more significantly than changes in behavior. It is therefore necessary to plan for a longer project duration to produce significant behavioral changes. Second, the indicator of change in attitudes towards GBV was assessed for four distinct groups of participants. Significant change is expected among community members not directly exposed to project activities, but not necessarily among prevention participants and project actors. The results confirm this notion, demonstrating differences in the magnitude of the change in attitude between groups. Third, impact analysis is most useful for measuring changes at the population level, in this case the community members. For the other three groups of participants, statistical changes between the baseline and endline studies are not expected to the same extent. However, information on project actors and participants in prevention activities is valuable in identifying potential weaknesses in knowledge, attitudes, and behaviors. For example, a significant deterioration in project actors' participation in intra- household decision-making was observed at the end of the project, raising questions about the negative impact of the time-intensive volunteer model of women in CBOs on their power within the household. The sustainability and viability of this “volunteer model” needs to be examined. However, to obtain a reliable explanation for this hypothesis, additional data collection methods and deeper investigations are needed. Future studies may opt for qualitative methods to explore KAP indicators (attitudes of acceptance towards GBV, participation in decision-making within the household, and support for domestic violence in the community) with project actors and participants in prevention activities. In contrast, for the quantitative KAP study, by focusing on members of the general population, statistical power would be increased to detect population-level changes of particular subgroups of interest within the population. At the same time, it would be beneficial to systematically sample and recruit different types of participants as subgroups of the general community, such as indigenous groups, rural and urban communities, or different age groups, e.g. young men and young women. This dual approach would provide a clearer picture of the impact of the project at the community level and provide more detailed information on the processes and dynamics among project actors and participants in prevention activities. Finally, the PRVBG project was initially planned to run for five years. However, due to various circumstances, project activities were carried out over a maximum period of 24 months, and in some health zones, notably in Maniema, for even shorter periods. Therefore, it is important to highlight that the pre-defined target should have been adjusted accordingly to account for this shorter implementation period. Matrix 4: Percent change in all three primary indicators between baseline and endline Page 88 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) COMMUNITY QUANTITATIVE SURVEY Percentages of primary indicators at baseline at endline Percentage Estimated IR Intermediate Results Indicators Benchmark Baseline Endline Change Target % decrease in attitudes of acceptance of GBV in targeted health zones (service providers, 1 ND 43,7% 29,6% -14,1% ↓20% community activists, participants in prevention activities, community members) % Change in Women's Participation in Household 2 ND 45,8% 41,6% -4,2% ↑20% Decision Making % change in aid provision among men and women 3 aware of intimate partner violence at community ND 85,0% 88,4% 3,4% ↑20% level TRANSLATED LESSONS LEARNED SECTION: The Consortium would like to acknowledge the fruitful collaboration with the FSRDC and the WB which has been a key success factor. The weekly meetings were particularly beneficial in the initial phase, contributing to the development of the implementation plan and strategic decisions. In addition, the cooperation with our consortium partner, Médecins du Monde, has made it possible to benefit from additional expertise, thus making the project to evaluate the process possible. Internally, regular debriefing sessions and close collaboration within the team fostered communication, mutual learning, and continuous improvement. These exchanges have been invaluable for learning lessons, comparing results between provinces and adjusting our actions accordingly. Effective communication between stakeholders and within teams is therefore essential for the success of such a process evaluation. In addition, it is crucial to be able to count on competent and professional staff to carry out such work. In this regard, we would like to highlight the fruitful collaboration with the NGO Marakuja, which has been a real success. The quality of the work provided, the effective communication and the flexibility demonstrated by Marakuja have greatly contributed to the success of the KAP study. Unfortunately, we have also faced breaches of trust within our own team. Despite this, we managed to mitigate the consequences and maintain the quality of the work. We see this experience as an important lesson to be learned. It would be beneficial to engage in joint reflections to put in place measures to further prevent such situations in the future. We also faced challenges in terms of balancing technical and administrative work. The coordination team was tasked with both technical and managerial tasks, which led to delays, especially in reporting. Going forward, it would be a good idea to assess the administrative and coordination workload more accurately, and to provide dedicated reinforcement for these tasks. The substantial amount of data collected during the evaluation, with a total of 1930 people interviewed as part of the qualitative data collection during three visits, and 1611 participants in the baseline KAP study and 1380 participants in the end-of-study study, provided an overview of the implementation of the Page 89 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) PRVBG project in the prevention and response to GBV in four provinces of the country. This provided a solid basis for analysis and recommendations. The inclusion of a large number of participants in the focus groups reinforced the validity and representativeness of the results obtained. However, it is important to point out that managing this large amount of data has been a considerable and partly underestimated task. The process of collecting, transcribing, coding, and analyzing required significant effort. There were delays in these stages, in part due to the large volume of data to be processed. In addition, the combination of analysis and coordination responsibilities at the level of the evaluation coordination team also had an impact on the progress of the process. This highlights the need for adequate attention to the planning and execution of data management activities for future evaluation and research projects. The operational, multisectoral and policy analyses and recommendations generated by the process evaluation have significant value for future interventions in the area of prevention and responses to SGBV. They cover key themes and provide practical guidance. In addition, the data and analyses revealed important trends that require further research for a better understanding of the dynamics of SGBV and optimal orientation of future actions. For future similar initiatives, it is essential to build in a continuous evaluation process from the outset to track progress, make adjustments and continuously improve interventions while the project is still ongoing. It is therefore recommended that a process evaluation be planned at the project design stage to maximize the benefits of such an evaluation. TRANSLATED CONCLUSION SECTION: In conclusion, this process evaluation of the PRVBG project highlighted several key elements that underpin more general considerations. The PRVBG project stood out for its comprehensive, holistic approach and its commitment to empowering survivors. The PRVBG project has succeeded in comprehensively addressing the complexity of GBV with a focus on the needs of survivors, and its implementation has been widely praised despite the challenges of conflict and the COVID-19 pandemic. The commitment to scientific research and continuous learning was a positive aspect that fostered collaboration between organizations and reinforced the overall effectiveness of interventions against GBV. This commitment also promoted the accountability of the entire PRVBG project. FSRDC’s role as a governmental entity offered a model for future similar initiatives. Capacity-building for CBO providers has been an important investment, enabling the training of professional, competent community players capable of delivering quality services and effectively managing interventions. This has helped to improve the results and impact of the PRVBG project in the fight against gender-based violence. However, the implementation of the PRVBG project has been fraught with challenges. Setting up the project in a complex and sensitive context presented operational challenges and constraints linked to the various stakeholders and their expectations. Implementation deadlines restricted the full potential of the PRVBG project, and some interventions did not have the same opportunity to develop in all zones. More specifically, the financial autonomy of the CBOs proved to be a major challenge. Income-generating Page 90 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) activities have not produced the expected results, compromising the CBOs’ ability to operate independently and autonomously. It is therefore essential to pay particular attention to the planning and management of IGAs, while ensuring the timely availability of the necessary resources. The lack of adequate resources and monitoring, as well as communication difficulties, have also hampered the overall effectiveness of the PRVBG project. In particular, collaboration with the health system was identified as a crucial challenge to ensuring quality medical care for the beneficiaries of the PRVBG project. Specific concerns were raised, such as the negative consequences for the personnel involved. Security issues were identified, including the fear of reprisals for actors, both from the community and professional staff from umbrella NGOs and the health sector. Measures need to be taken to guarantee their safety and emotional support, as well as to provide adequate compensation for caring for victims. The mixed results regarding the empowerment of women within communities, particularly with regard to their participation in household decision-making, and the high workload for the program's community players also require careful consideration. Recommendations have been made to improve future projects, including strengthening the security of actors, guaranteeing adequate transportation for survivors, ensuring the availability of post-exposure prophylaxis kits, further integrating vulnerable populations and actively involving men in GBV prevention efforts. Measures are also suggested to support CBO members, assess the administrative workload for community players and strengthen data management. In conclusion, the lessons learned from the PRVBG project can serve as a valuable basis for future reflections and improvements aimed at increasing the effectiveness and sustainability of similar projects. These lessons can guide future efforts in the fight against GBV, focusing on the empowerment of survivors, the inclusion of vulnerable populations and effective coordination between actors. Page 91 of 92 The World Bank DRC - Gender Based Violence Prevention and Response Project (P166763) ANNEX 11. MAP OF PROJECT INTERVENTION AREAS Page 92 of 92