Document of The World Bank Report No.: 39557 PROJECT PERFORMANCE ASSESSMENT REPORT GHANA AIDS RESPONSE PROJECT (GARFUND) (CREDIT NO. 3458) June 19, 2007 Sector, Thematic and Global Division Independent Evaluation Group Currency Equivalents Currency Unit = cedi (c) (as of 2000) (as of 2005) US$1.00 = 5,455 cedis US$1.00 = 9,073 cedis 1.00 cedi = US$0.00018 1.00 cedi = US$0.00011 Abbreviations and Acronyms AIDS Acquired immune deficiency MDA Ministries, departments and Syndrome agencies CBD Community-based Development MESW Ministry of Employment and CBO Community-based Organization Social Welfare CCM Common Coordination MICS Multiple Indicator Cluster Survey Mechanism MLGRD Ministry of Local Government and CDD Community-driven Development Rural Development CEDEP Center for the Development of M&E Monitoring and Evaluation People MoE Ministry of Education DA District Assembly MoH Ministry of Health DAC District AIDS Committee MSHAP Multi-sector HIV/AIDS Project DANIDA Danish International Development MTR Mid-Term Review Assistance NACP National AIDS/STD Control DCA Development Credit Agreement Program DFID UK Department of International NDPC National Development Planning Development Commission DRI District Response Initiative NGO Non-governmental Organization EU European Union NPC National Population Council FBO Faith-based Organization NSF National AIDS Strategic FHI Family Health International Framework FIDA International Federation of OVC Orphans and vulnerable children Women Lawyers PAD Project Appraisal Document GAC Ghana AIDS Commission PIU Project Implementation Unit GARFUND Ghana AIDS Response Fund PLWHA Person living with HIV/AIDS GDP Gross Domestic Product PMTCT Prevention of Mother to child GFATM Global Fund for AIDS, Transmission Tuberculosis and Malaria PPAR Project Performance Assessment GHS Ghana Health Service Review GoG Government of Ghana PRAC Project Review and Appraisal GSMF Ghana Social Marketing Committee Foundation SWAp Sector-Wide Approach GTZ German Development Assistance TA Technical Assistance HIV Human Immune-Deficiency Virus TTL Task Team Leader ICR Implementation Completion UNAIDS Joint United Nations Programme Report on HIV/AIDS IDA International Development UNICEF United Nations Children's Fund Association USAID United States Agency for IEC Information, Education and International Development Communication VCT Voluntary Counseling and Testing IEG Independent Evaluation Group WHO World Health Organization M & E Monitoring and Evaluation MAC Municipal AIDS Committee MAP Multi-country AIDS Program Fiscal Year Government: January 1-- December 31 Director-General, Evaluation : Mr.Vinod Thomas Director, Independent Evaluation Group, World Bank : Mr. Ajay Chhibber Manager, Sector, Thematic and Global Evaluation Division : Mr. Alain Barbu Task Manager : Ms. Denise Vaillancourt i IEGWB Mission: Enhancing development effectiveness through excellence and independence in evaluation. About this Report The Independent Evaluation Group assesses the programs and activities of the World Bank for two purposes: first, to ensure the integrity of the Bank's self-evaluation process and to verify that the Bank's work is producing the expected results, and second, to help develop improved directions, policies, and procedures through the dissemination of lessons drawn from experience. As part of this work, IEGWB annually assesses about 25 percent of the Bank's lending operations. In selecting operations for assessment, preference is given to those that are innovative, large, or complex; those that are relevant to upcoming studies or country evaluations; those for which Executive Directors or Bank management have requested assessments; and those that are likely to generate important lessons. The operations, topics, and analytical approaches selected for assessment support larger evaluation studies. A Project Performance Assessment Report (PPAR) is based on a review of the Implementation Completion Report (a self-evaluation by the responsible Bank department) and fieldwork conducted by IEGWB. To prepare PPARs, IEGWB staff examine project files and other documents, interview operational staff, and in most cases visit the borrowing country to discuss the operation with staff of the Bank and the government, other stakeholders, and beneficiaries. The PPAR thereby seeks to validate and augment the information provided in the ICR, as well as examine issues of special interest to broader IEGWB studies. Each PPAR is subject to peer review and IEGWB management approval. Once cleared internally, the PPAR is reviewed by the responsible Bank department and amended as necessary. The completed PPAR is then sent to the borrower for review; the borrowers' comments are attached to the document that is sent to the Bank's Board of Executive Directors. After an assessment report has been sent to the Board, it is disclosed to the public. About the IEGWB Rating System The time-tested evaluation methods used by IEGWB are suited to the broad range of the World Bank's work. The methods offer both rigor and a necessary level of flexibility to adapt to lending instrument, project design, or sectoral approach. IEGWB evaluators all apply the same basic method to arrive at their project ratings. Following is the definition and rating scale used for each evaluation criterion (additional information is available on the IEGWB website: http://worldbank.org/ieg). Outcome: The extent to which the operation's major relevant objectives were achieved, or are expected to be achieved, efficiently. The rating has three dimensions: relevance of objectives, efficacy, and efficiency. Relevance of objectives is the extent to which the project's objectives are consistent with the country's current development priorities and with current Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, and Operational Policies). Efficacy is the extent to which the project's objectives were achieved, or expected to be achieved, taking into account their relative importance. Efficiency is the extent to which the project achieved, or is expected to achieve, a return higher than the opportunity cost of capital and benefits at least cost compared to alternatives. The efficiency dimension generally is not applied to adjustment operations. Possible ratings: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Risk to Development Outcome: The risk, at the time of evaluation, that development outcomes (or expected outcomes) will not be maintained (or realized). Possible ratings: High Significant, Moderate, Negligible to Low, Not Evaluable. Bank Performance: The extent to which services provided by the Bank ensured quality at entry of the operation and supported effective implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of supported activities after loan/credit closing, toward the achievement of development outcomes. The rating has two dimensions: quality at entry and quality of supervision. Possible ratings: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Borrower Performance: The extent to which the borrower assumed ownership and responsibility to ensure quality of preparation and implementation, and complied with covenants and agreements, towards the achievement of development objectives and sustainability. The rating has two dimensions: government performance and implementing agency performance. Possible ratings: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. iii Contents Principal Ratings................................................................................................................v Key Staff Responsible........................................................................................................v Preface.............................................................................................................................. vii Summary........................................................................................................................... ix 1. Background and Context ............................................................................................1 HIV/AIDS Epidemic....................................................................................1 Ghana's Response........................................................................................2 World Bank Support for HIV/AIDS Control through 2000 ........................4 Support of Development Partners at the Time of Project Design................6 2. Objectives and Design..................................................................................................6 3. Implementation and Costs ........................................................................................10 Planned versus Actual Costs and Financing..............................................10 Planned versus Actual Inputs/Activities by Component ...........................10 4. Monitoring and Evaluation.......................................................................................15 Design ........................................................................................................15 Implementation ..........................................................................................16 Utilization of Data for Decision-Making...................................................18 5. Outputs and Outcomes by Objective .......................................................................18 Reduce the spread of HIV infection...........................................................19 Reduce the impact of AIDS on those who are infected and their families26 6. Ratings ........................................................................................................................27 7. Lessons and Challenges.............................................................................................33 Lessons.......................................................................................................33 Challenges..................................................................................................34 References.........................................................................................................................37 Annex A. Basic Data Sheet..............................................................................................41 Annex B. Persons and Organizations Consulted .........................................................43 Annex C. Supplemental Background and Contextual Information...........................47 Annex D. GARFUND: Planned vs. Actual Support by Component..........................51 iv Annex E. Subproject Data..............................................................................................57 Annex F. Costs and Financing........................................................................................61 Annex G. GARFUND: Project Outcomes by Objectives and Targets.......................65 Annex H. Borrower's Comments ..................................................................................71 Boxes Box 1. Ghana' HIV/AIDS National Strategic Framework (NSF) for 2001-2005.............. 5 Box 2. Multi-Country AIDS Program (MAP) Eligibility Criteria...................................... 7 Box 3. Queen Mother's Community Foster Program for Children Affected by HIV/AIDS, Manya Krobo District, Eastern Region: Findings of a 2004 Review ............................... 27 Box 4. Role of MoH in the Context of the GAC .............................................................. 34 Figures Figure 1. Ghana 2002-2003 HIV/AIDS Expenditures by Category (in billions of Cedis)20 Figure 2. Percent of MDA Subprojects Supporting HIV/AIDS Interventions (n=70 subprojects)....................................................................................................................... 21 Figure 3. Percent of NGO Subprojects Supporting HIV/AIDS Interventions (n=501 Subprojects) ...................................................................................................................... 21 Figure 4. Expenditures by Target Group as a Percent of Total HIV/AIDS Program Expenditures, 2003 ........................................................................................................... 23 Tables Table 1. Ghana Health Indicators, 1988-2003.................................................................... 1 Table 2. Planned Versus Actual Costs by Component (US$million equivalent) ............. 10 Table 3. Subprojects Funded/Completed under GARFUND by Call for Proposals......... 11 Table 4. Timing of Data for Measuring Outcomes........................................................... 17 Table 5. Summary IEG Ratings* by Objective ................................................................ 28 This report was prepared by Michael Azefor, Consultant, and Denise Vaillancourt, IEGSG, who assessed the project in June 2006. Pilar Barquero and Marie-Jeanne Ndiaye provided administrative support. v Principal Ratings ICR* ICR Review* PPAR AIDS Response Project (GARFUND) (Credit 3458) Outcome Satisfactory Unsatisfactory Unsatisfactory Sustainabilitya Likely Unlikely - Risks to Development - - Substantial Outcomeb Institutional Development Substantial Modest - Impacta Bank Performance Satisfactory Unsatisfactory Unsatisfactory Borrower Performance Satisfactory Satisfactory Moderately Unsatisfactory * The Implementation Completion Report (ICR) is a self-evaluation by the responsible operational division of the Bank. The ICR Review is an intermediate Independent Evaluation Group (OED) product that seeks to independently verify the findings of the ICR. a. This rating was discontinued as of July 1, 2006 b. This rating was introduced as of July 1, 2006. Key Staff Responsible Project Task Manager/Leader Division Chief/ Country Director Sector Manager Preparation Sandra Rosenhouse Rosemary Bellew Peter Harrold Appraisal Eileen Murray Rosemary Bellew Peter Harrold Supervision Eileen Murray Alexandre V. Abrantes Peter Harrold & Mats Karlsson Completion Aissatou Diack Alexandre V. Abrantes Mats Karlsson vii Preface This is the Project Performance Assessment Report (PPAR) for the Ghana AIDS Response Project (GARFUND), financed through IDA Credit No. 3458 in the amount of US$25.0 million (19.6 million SDR), with planned government and community contributions of US$1.7 million and US$1.0 million, respectively. The credit was approved on December 28, 2000, became effective on May 8, 2002, and was 92 percent disbursed when it closed on December 31, 2005, six months after the original closing date. The findings of this assessment are based on an Independent Evaluation Group (IEG) mission to Ghana carried out in June 2006. The mission met in Accra with authorities and staff of: the Ministry of Health, Ghana Health Service and Ghana AIDS Commission; other public sector agencies implementing health and HIV/AIDS activities; selected NGOs and other representatives of civil society; and bilateral and international partners. The mission also visited the Regions (Districts/Municipalities) of: Greater Accra (Tema), Upper West (Wa, Nadowli), Upper East (Bolgatanga, Kassena Nankana and Navrongo), Northern (Tamale, Tolon) and Ashanti (Kumasi), where it had the opportunity to meet with HIV/AIDS focal persons, health authorities and other members of regional, district and municipal AIDS Committees and to witness activities supported by GARFUND. Key sources of evidence consulted include: (a) World Bank project files; (b) project-related reporting and evaluation; and (c) epidemiological data, studies, surveys and research on health and HIV/AIDS, much of it generated in Ghana. This PPAR is the first conducted on a World Bank-financed HIV/AIDS project since IEG's 2005 evaluation of the World Bank's support for HIV/AIDS control, Committing to Results: Improving the Effectiveness of HIV/AIDS Assistance, and the first of several to be conducted on projects prepared under the Africa Multi- Country AIDS Program (MAP). In addition, evidence from this assessment will contribute to the evaluation of multi-sectoral approaches in achieving health outcomes as part of IEG's forthcoming evaluation of the World Bank's support to health, nutrition, and population. As such, relatively more material has been presented in this "enhanced" PPAR than is the IEG standard. This report draws on the technical inputs of Dr. Moses Aikins, Health Economist, JSA Consultants, of Ghana. The IEG team gratefully acknowledges all those who made time for interviews and provided documents and information. Following standard IEG procedures, copies of the draft PPAR were sent to the relevant government officials and agencies for their review and feedback. Their comments are presented in Annex H. ix Summary The objectives of the Ghana AIDS Response Project (GARFUND) were to (1) intensify multi-sectoral activities designed to combat the spread of HIV/AIDS, and (2) reduce its impact on those already affected by HIV/AIDS. GARFUND's design specifically excluded the support of activities under the health sector's responsibility, assuming that these would be adequately covered under the pooled financing of partners (including the Bank) to the health sector, which had been ongoing since 1997. The underlying approach of the GARFUND was to provide emergency funding to accelerate and facilitate the disbursement of funds to a broad range of public sector and civil society actors, who would be mobilized to prepare and implement subprojects across a range of sectors and throughout the country. The project design envisaged the support of capacity building, monitoring and evaluation (M&E), supervision and operational research, both to compensate for an accelerated preparation and to support a learning-by-doing approach. Expected outcomes of prevention efforts articulated in the project logframe included reductions in the spread of HIV infection among the general population and among commercial sex workers (CSW), as well as reductions in risky behavior. The project's support was organized around four components: (a) prevention and care services: the financing of subprojects designed and implemented by public agencies and civil society organizations; (b) strengthening public/private institutions for HIV/AIDS control and care giving: the provision of training, technical assistance and other guidance to strengthen the capacity of implementing agencies; (c) knowledge management: the provision to implementers of best practice knowledge and a forum for learning and exchange; and (d) project management: the creation and support of the Ghana AIDS Commission (GAC), a multi-sectoral AIDS coordinating entity, responsible for project management; and the support of M&E. Until the creation of the GAC, the Ministry of Health (MoH) had been the lead agency for HIV/AIDS. In 1986 the first 42 cases of AIDS in Ghana were diagnosed primarily among CSW, who had traveled and lived abroad. By 2000 a cumulative total of 37,298 AIDS cases had been reported and the HIV prevalence rate was estimated at 3.0 percent, with highest rates in coastal Ghana. The epidemic was and remains largely concentrated among high-risk groups. While awareness of the existence of HIV/AIDS was almost universal by 1998, knowledge of HIV transmission and prevention methods was lower and behaviors did not fully reflect such knowledge. In 2001 Ghana's HIV/AIDS National Strategic Framework (NSF) for 2001-2005 was formally issued. Based on an extensive analysis of Ghana's response to the epidemic to date, the NSF emphasized prevention (among high-risk and vulnerable groups and among the general population), mitigation, the creation of an enabling environment, and the establishment of a multi-sectoral institutional framework. In support of NSF implementation, GARFUND was designed as one of the first operations under the Africa Region's Multi-Country AIDS Program (MAP) Adaptable Program Loan. Reduction in the spread of HIV infections. The GARFUND was successful in its efforts to increase rapidly and significantly the availability of financial resources to x support Ghana's fight against HIV/AIDS. However, the efficiency of total HIV/AIDS expenditures has not improved, with critical activities (behavior change interventions, interventions for high-risk groups, and essential health inputs) relatively neglected. The institutional framework set up to facilitate a multi-sectoral approach has thus far limited the role and potential contribution of the MoH. The under-utilization of the expertise and experience of the health sector has compromised the technical quality of interventions and overall program effectiveness. GARFUND support was instrumental in further decentralizing the fight against HIV/AIDS, most notably the establishment and functioning of 10 Regional and 110 District AIDS Committees (RACs and DACs) covering all regions/districts. However, shortfalls in the institutional framework, management processes and capacity persist. GARFUND mobilized a broad range of actors and sectors for a more expanded and expansive response. During its three-and-one-half-year implementation, 3,030 subprojects were financed covering all districts, implemented by a broad spectrum of public sector and civil society actors and spanning a full range of prevention and care activities. However, there are concerns about the quality and effectiveness of these subprojects. Prevention efforts were geared for the most part around the transmission of basic knowledge about the disease, instead of a more concerted effort to achieve targets to reduce risky behaviors, specified in both the NSF and in the project design document. Most subprojects were targeted to the general population and youth (in and out of school), with relatively few interventions for high-risk groups that were known to be the drivers of the epidemic. Despite nationally-set and project-specific objectives to reduce infections among commercial sex workers, actual expenditures on this high-risk, high-transmission group were minimal (an estimated 1 percent of GARFUND expenditures and 0.2 percent of total HIV/AIDS expenditures). The failure to undertake essential operational research also undermined the effectiveness of prevention interventions as well as the learning-by- doing approach inherent in the project design. Insufficient rigor in the review of subproject proposals and limited capacities of implementers compromised the quality, results orientation and potential development impact of subprojects. While the prevalence of HIV in Ghana fell, trends in HIV prevalence are not a valid measure for changes in the number of new infections (incidence). A decline in HIV prevalence could conceivably be the result of high AIDS mortality or unsuccessful treatment efforts, while an increase could signal more widespread and successful treatment. On the other hand, positive trends in the knowledge, attitudes and behaviors among the general population and among high-risk groups are consistent with a reduction in HIV incidence. Unfortunately these trends are not available because M&E was not undertaken as planned. While some baselines were established for some groups, they were not updated at the end of the project. Field visits suggest an increase in the knowledge of modes of HIV transmission and prevention methods among the general population and among youth and this is likely attributable in part to GARFUND support. However, there was no evidence of change in the levels of fear, stigma and discrimination associated with HIV/AIDS. Some RACs and DACs did report an increase in condom availability. Indeed, annual condom sales increased from 24 to 30 million, between 2002 and 2005, a result largely attributable to social marketing efforts financed by USAID. GARFUND xi investment in condom distribution was very modest and the bulk of prevention efforts were focused more on information, education, and communication (IEC) than on behavior change. Reduction of the impact of AIDS. The availability of home- and community- based services for the care and support of persons living with HIV/AIDS (PLWHA) is reported by GAC and by most informants in most regions/districts to have increased over the life of GARFUND. However, the failure to establish links between care and support services and the formal health system, which has the mandate for the care of AIDS patients, has left these services without the technical support and oversight needed to ensure their quality, effectiveness and sustainability. Some 17,500 orphans are reported to have benefited from different kinds of support under GARFUND, but the coverage has not been systematically tracked and a UNICEF study raises a number of issues about the quality, effectiveness and sustainability of orphan care provided with GARFUND support. Ratings. While recognizing that this project was among the first to be funded under the new, and largely untested, MAP design, this PPAR evaluates performance against objectives and expected outcomes documented in Ghana's own national strategy and in the project design document. On this basis, the outcome of GARFUND is rated unsatisfactory overall, based on its modest relevance, modest efficacy, and negligible efficiency. Risk to development outcome is rated as substantial. The Bank's performance was unsatisfactory; and the Borrower's performance was moderately unsatisfactory. Lessons · The drive to accelerate disbursements and to spread the financing across a wide range of implementers, activities and geographic areas poses the risk that these process goals could take precedence over the strategic selection and prioritization of the highest-impact activities. · A more strategic choice and sustained use of implementers with the experience and capacity to carry out the highest-impact interventions likely would have resulted in stronger and more sustainable results. · Efforts to strengthen a multi-sectoral response would likely have been more successful had synergies been created with the MoH, capitalizing on its technical and operational expertise. · "Learning-by-doing" has the potential to improve effectiveness, but it is unlikely to take place in the absence of strong preparation and incentives for systematic monitoring and evaluation. Ajay Chhibber Acting Director-General Evaluation xii 1 1. Background and Context 1.1 Ghana's population of 22.5 million is growing at 2.0 percent per annum. More than one third of Ghanaians live below the poverty line and unemployment is rampant, especially in urban centers. In recent years, the country has recorded good economic performance, but past economic hardships have left Ghana with relatively poor human development indicators. Between 1988 and 2003 health indicators have shown overall improvement, but infant and child health indicators have deteriorated slightly in recent years (Table 1). Annual GDP growth since 1995 has risen to over 5 percent and is likely to be sustained as current reforms are maintained within the current government's efforts to promote growth in income and employment, improve the delivery of services for human development and strengthen governance and public sector management. Table 1. Ghana Health Indicators, 1988-2003 Indicator 1988 1993 1998 2003 Infant mortality rate (per 1000 live births) 77 66 57 64 Under 5 mortality rate (per 1000 live births) 155 119 108 111 Life expectancy at birth (years) 54 56 57 58 Total fertility rate 6.4 5.5 4.6 4.4 Source: Ghana Demographic and Health Surveys, 1988, 1993, 1998 and 2003 HIV/AIDS Epidemic 1.2 Epidemiology. In 1986, the first 42 cases of AIDS in Ghana were diagnosed. Most initial cases occurred among women who had traveled and lived abroad, especially commercial sex workers (CSW)1. By 2000, a cumulative total of 37,298 AIDS cases had been reported2. According to UNAIDS estimates for that same year, some 330,000 adults and 14,000 children were HIV positive. Government of Ghana (GoG) statistics also estimated that 119,410 children under 15 had lost their mother or both parents to AIDS by 2000.3 1.3 As of 2000, an estimated 3 percent of women attending antenatal clinics were HIV-positive, revealing a downward trend from an estimated 4.0 percent in 1998 and 3.2 percent in 1999.4 These updated estimates are lower than those available in 1999, when some 4.6 percent of the adult population was estimated to be infected (1999 NACP) and HIV prevalence was noted to be increasing.5 HIV prevalence was highest in southern Ghana (Eastern, Western, Central, Greater Accra and lower Volta regions, 3.6 percent), somewhat lower (2.6 percent) in central Ghana (Brong Ahafo, Ashanti and higher Volta regions), and lowest (1.4 percent) in northern Ghana (Northern, 1Most of these women traveled to Côte d'Ivoire and returned to their homes in the Eastern and Greater Accra Regions. In the earliest stages of the epidemic many Ghanaians believed that HIV/AIDS was a disease that was limited to CSW and their customers. (Source: "AIDS in Africa During the Nineties: Ghana," 2003). 2Ghana AIDS Commission (GAC), 2001. 3Estimates included in the World Bank design documents were slightly higher, reflecting best estimates at the time: 400,000 (vs. 330,000) Ghanaians infected with HIV and 126,000 (vs. 119,410) orphans. 4 Measures Project and GAC, 2003. 5 World Bank, 2000. Some 80 percent of AIDS cases were attributed to heterosexual transmission, 15 percent to vertical transmission, and the remaining 5 percent to transmission through blood and blood products. 2 Upper West, and Upper East regions). However, more than three-quarters (77 percent) of sex workers who attended STD clinics in Accra were HIV-positive in 1997. More than 90 percent of AIDS cases were among persons aged 15-49 years, with the most cases among women aged 25-29 years and men 30-34 years. In 1998 the male to female ratio was 2:1. 1.4 Knowledge and Behaviors. In 1998 97 percent of women and virtually all men had heard about AIDS, and slightly over two-thirds of women and men knew that a person with HIV could look healthy. At least 80 percent of men and women, when prompted, knew that condom use protects against AIDS. Knowledge was generally higher among urban residents and among men (GDHS, 1998). A behavioral survey conducted in 2000 revealed that 58 percent of sex workers in Accra, 41 percent of sex workers in Obuasi,6 57 percent of male police in Accra, and 55 percent of Obuasi miners knew the main methods of preventing HIV (FHI, 2001). Median age at first sex increased from about 17 years in 1993 to about 18 years in 1998 for women. 1.5 While virtually all men and more than four out of five women had heard of male condoms in 1998, knowledge of where to get condoms was far lower (66 percent of women and 77 percent of men) and only 16 percent of women and 40 percent of men had ever used one, with noted differentials between urban and rural populations. Use of condoms among female sex workers was considerably higher, with 95 percent of those in Accra and 75 percent of those in Obuasi reporting use of condoms in their last paid sex act and 88 percent and 44 percent, respectively, reporting use of condoms during every sex act with clients. Ghana's Response 1.6 As early as 1985 a technical committee on AIDS was formed to develop a short-term plan for AIDS prevention and control. In 1987 the National AIDS Control Program (NACP) was established in the Disease Control Unit of the Ministry of Health (MoH), and in the following year a medium-term plan was developed with the World Health Organization's Global Program on AIDS (WHO/GPA). By 1994 sentinel surveillance of pregnant women attending antenatal clinics was established in all regions of the country allowing the tracking of the epidemic. 1.7 A review of Ghana's HIV/AIDS response was undertaken in 1998-1999 by a multi-disciplinary, multi-sectoral team under the leadership of the MoH.7 Annex C, Box C.2, itemizes the main findings of this review. Ghana's efforts to fight HIV/AIDS at that time consisted essentially of: promotion of safer sex (through a strategy stressing abstinence, fidelity, and condom use), clinical responses to prevention and care,8 inter-sectoral collaboration,9 partnerships with the non- 6Gold mining town in Ashanti Region. 7Adjei et al., 2000. 8Preventing and managing STDs, preventing HIV transmission through transfusion, reducing iatrogenic transmission, reducing mother-to-child transmission, institutional care of people living with HIV/AIDS, voluntary counseling and testing. 9Two examples include: Ministry of Education's efforts to introduce family life education into school curricula, and collaboration between MoH and the Ghana Police in the care and education of CSW through STD clinics and outreach activities. 3 governmental sector,10 and modest expansion of district- and community-level activities (under the District Response Initiative, DRI, pilot). Ghana's response involved key development sectors and engaged a range of actors, including communities, NGOs, and district-level actors. Shortfalls in that response, included: poor targeting and coverage of prevention efforts;11 inadequate attention to behavior change (vs. increasing awareness and knowledge); poor coverage and quality of prevention of mother-to-child transmission and interventions to prevent transmission by non-sterilized instruments in clinical and non-clinical settings; very limited efforts on legal and human rights; and (with the notable exception of sentinel surveillance in antenatal clinics), very weak monitoring and evaluation and inadequate research, which resulted in a very limited understanding of program effectiveness. A "way forward" was defined that sought to address these weaknesses as well as to build on ongoing activities and experiences, especially the acceleration and expansion of the DRI.12 1.8 The institutional framework for the HIV/AIDS response existed at both the national and regional levels and was being piloted at the district level. At the national level the NACP was responsible for coordinating the national response, consensus building, policy development, technical support to other stakeholders, resource mobilization, implementation of HIV/AIDS activities under MoH's mandate, epidemiological surveillance, research, and monitoring and evaluation. In addition, a technical working group, chaired by the Deputy Minister of Health provided a platform for exchange among its members, comprised of representatives from Ministries of Education, Social Welfare, Youth and Sports, Communications and selected donor agencies supporting HIV/AIDS efforts. Other networks/fora for coordination of HIV/AIDS efforts included: a UNAIDS HIV/AIDS Thematic Group; Christian Health Association of Ghana (CHAG), coordinating all mission facilities; a Ministry of Education (MoE) Task Force for AIDS prevention; and the Ghana HIV/AIDS Network (GHANET), created in 1996 to coordinate the work of over 100 NGOs in an attempt to minimize duplication of efforts and waste of human and material resources on AIDS efforts. At the regional level NACP coordinators, responsible for program planning, oversight and coordination, were only part-time, juggling numerous other responsibilities/disease programs, and few regional AIDS Advisory Committees were functional. At the district level, the DRI pilot was said to be promising, but had not yet been evaluated. 1.9 The analysis concluded that the NACP had carried out and overseen good work, including policy formulation and planning, but its potential to fulfill its mandate was seriously undermined by: (a) inadequate staffing at the national and 10Examples include: social marketing of condoms and NGO work with various in- and out-of-school youth. 11STD prevention and care of CSW in three clinics; limited interventions with miners, truck drivers, migrant and other mobile groups, uniformed services (police and military). 12 The DRI was piloted in 10 administrative districts in the Ashanti Region to support a move towards a more integrated, multi- sectoral and decentralized approach to the fight against HIV/AIDS led by the Ministry of Employment and Social Welfare (MoESW), with full involvement of the MoH, and the support of UNAIDS and WHO. This approach involved management capacity building and activities by communities, local NGOs and institutions. 4 regional levels;13 (b) inadequate financial support; and (c) its low position within MoH's hierarchy, all of which were incompatible with its role to coordinate a national response across sectors. It recommended a rigorous institutional assessment of the institutional and organizational framework for Ghana's HIV/AIDS response. However, in 1999, without the benefit of such an analysis, a five-day mission of the Institutional Partnership Against AIDS in Africa14 culminated in a "joint decision taken by Government and the International Partnership" to: establish a supra- ministerial and multi-sectoral body to direct and coordinate the involvement of all ministries, the private sector and non-governmental organizations (NGOs) in the fight against AIDS and its impact; and strengthen the NACP and elevate it to a higher position in MoH. 1.10 In 2001 Ghana's HIV/AIDS National Strategic Framework (NSF) for 2001- 2005 was issued, the product of a multi-disciplinary technical team of 16 experts,15 funded by UNAIDS and DFID, supported by WHO and endorsed by the Head of State (Box 1). The evidence cited in this document points to the concentration of the epidemic among certain groups and in certain towns and cities.16 World Bank Support for HIV/AIDS Control through 2000 1.11 Non-Lending Support. A 1989 review of Ghana's population, health and nutrition sectors17 reported low incidence of HIV and AIDS, with 276 AIDS cases reported as of December 1987, and documented the concentration of the epidemic in high-risk groups. No specific strategy or action was recommended for addressing HIV/AIDS, nor was any additional analysis of AIDS in Ghana undertaken in the 1990s. While not specific to Ghana, two widely disseminated analyses on the HIV/AIDS epidemic were issued, in 1997 and 2000, respectively, calling for more Government involvement in the fight against HIV/AIDS and for the prioritization of high-impact interventions.18 13 The NACP had five full-time and four part time staff. Full time staff were the Program Manager, a Public Health Specialist, Counseling and Laboratory Coordinators and a Senior Typist. Part-time staff included a Clinical Management Coordinator, IEC Coordinator, a Surveillance Officer with no formal training in epidemiology and an Accountant Regional level staff were given HIV/AIDS coordination responsibilities along with responsibilities for other disease programs and were thus not available full- time. 14 This mission was carried out under UNAIDS leadership, with participation by the World Bank Country Director, and representatives of other UNAIDS cosponsors (UNICEF, UNDP, UNFPA, UNDCP, ILO, UNESCO, WHO) (International Partnership against AIDS in Africa, 1999). 15 The 16 experts included representatives from: the National Population Council (NPC), MoH, the National Development Planning Commission (NDPC), MoESW, Ghana Social Marketing Foundation (GSMF), International Federation of Women Lawyers (FIDA), JSA Consultants, Ltd., and the Center for Development of People. 16 Seventy-six percent of commercial sex workers in Accra and 82 percent of those in Kumasi were reported to be HIV positive and 80 percent of all HIV/AIDS infection was due to heterosexual transmission. It was also reported that a higher number of cases occurred in the southern cities (like Kumasi, Koforidua and Accra), in mining towns (like Obuasi and Takwa) as well as in border towns. 17 World Bank, 1989. 18 In 1997 the Bank's research department released Confronting AIDS: Public Priorities in a Global Epidemic, which made the economic case for government involvement in fighting AIDS and provided epidemiological and economic evidence justifying the importance of investing in public goods provision and targeting of high-risk groups (World Bank, 1997). In 1999 a new AIDS strategy issued by the Bank's Africa Region, Intensifying Action against HIV/AIDS in Africa, pointed to the need to strengthen government commitment and highlighted the importance of targeting high-risk groups in both low- and high- prevalence settings (World Bank, 2000). 5 Box 1. Ghana' HIV/AIDS National Strategic Framework (NSF) for 2001-2005 Objectives: · Reduce new HIV infections among the 15-49 age-group and other vulnerable groups, especially youth · Improve service delivery and mitigate the impact of HIV/AIDS on individuals, families and communities · Reduce individual and societal vulnerability and susceptibility to HIV/AIDS by creating an enabling environment · Establish a well-managed multi-sectoral and multi-disciplinary institutional framework for coordination and implementation of HIV/AIDS programmes in the country Components (or Thematic Areas): · Prevention of new transmission of HIV, with specific focus on: youth, women, CSW, mobile and migrant populations, uniformed service personnel, workers, and the general public · Care and support for Persons Living with HIV/AIDS (PLWHA) · Creating an enabling (legal, ethical and policy) environment for national response · Decentralised implementation and institutional arrangements · Research, monitoring and evaluation Expected Outcomes: · Reduce incidence of new HIV infections among the youth and other vulnerable groups by 30 percent · Raise condom use during casual sex to 60 percent · Equip 30 percent of communities and health facilities to care for PLWHA · Enact and enforce necessary laws on HIV/AIDS · Establish institutions at national, regional, district and community levels Source: Ghana AIDS Commission (GAC), 2001 1.12 Country Assistance Strategy (CAS). The primary goal of the Bank's strategy for Ghana, in the 1990s through 2000 was to support Government in its efforts to reduce poverty and improve the quality of life of its citizens through macroeconomic stability and economic growth, higher private sector investment, capacity building and utilization, improved quality and access to basic social services and other direct poverty-alleviation measures (World Bank, 1995, 1997, and 2000).19 Neither the 1995 nor the 1997 have CAS addressed the HIV/AIDS epidemic.20 By contrast, the 2000 CAS emphasizes the importance of containing the epidemic through a multi-sectoral strategy, and advocated restructuring ongoing projects in key sectors and supplemental support to health. 1.13 Lending. At the time of project design the Bank had financed three health sector operations, of which the first two were completed (Health and Education Rehabilitation Project, approved in 1986; and the Second Health and Population Project, approved in 1990) and the third was ongoing (Health Sector Support Project, a Sector-Wide Approach [SWAp] operation, approved in 1997). This series of investments supported improvements to service quality, coverage, efficiency and financial sustainability. Only the third project supported the HIV/AIDS program (by 19All three CAS support Ghana's poverty alleviation and economic development goals as set out in Ghana-Vision 2020. 20The 1995 CAS mentions emerging challenges of new diseases "...such as AIDS and cardiovascular disease..."; and the 1997 CAS refers to UNAIDS as a good example of donor collaboration. But neither provides any information on the dynamics or implications of the epidemic or programs lending or non-lending support to address it. 6 nature of its sector-wide approach), but support was modest21 (see Annex C, Table C.1). Support of Development Partners at the Time of Project Design 1.14 Before 2000 partners' support to HIV/AIDS was channeled largely through the health sector's Programme of Work (PoW) for 1997-2001. In addition to the World Bank credit and GoG's counterpart, pooled funding for the health sector was also provided by DFID, DANIDA, Nordic Fund, the Netherlands, and the European Union. HIV/AIDS activities carried out in the context of the PoW included voluntary counseling and testing (VCT) services, prevention of mother to child transmission (PMTCT), information, education and communication (IEC), home-based care and social support (in collaboration with civil society organizations), improved blood screening, training for syndromic management of STDs, tuberculosis control and surveillance. Health sector financing also enabled the NACP to engage other ministries and key NGOs, faith-based organizations (FBOs) and private organizations to pilot research and other HIV/AIDS activities in selected high-prevalence areas. Additional earmarked funds for HIV/AIDS were provided by USAID, CIDA, GTZ, UNAIDS, UNICEF, UNDP, UNFPA and WHO. (See Annex C, Table C.2.)22 2. Objectives and Design 2.1 The Ghana AIDS Response Project (GARFUND) was financed through an IDA credit of US$25.0 million equivalent,23 approved on December 28, 2000 and declared effective on May 8, 2002, with planned government and community contributions of US$1.7 million and US$1.0 million, respectively. 2.2 The objective of the GARFUND was "(1) to intensify multi-sectoral activities designed to combat the spread of HIV/AIDS, and (2) reduce its impact on those already affected by HIV/AIDS."24 Expected outcomes of prevention efforts articulated in the project logframe included reductions in the spread of HIV infection among the general population and among commercial sex workers, as well as reductions in risky behavior. Because the ongoing Health SWAp was assumed to be providing adequate financial support to all of MoH's programs, including HIV/AIDS, GARFUND's design specifically excluded the support of HIV/AIDS activities falling within MoH's mandate (World Bank 2000b). 2.3 Designed and financed under the umbrella of the Africa Region's Multi- Country AIDS Program (MAP),25 the underlying approach of the GARFUND (one of 21The Implementation completion report (ICR) for this project, not issued until 2003, found that performance of the HIV/AIDS program was unsatisfactory (World Bank, 2003). 22Overall costs and financing of the national strategic framework for HIV/AIDS were not available. 23The US$ equivalent of Special Drawing Rights (SDRs) or other currencies. 24Development Credit Agreement between the Republic of Ghana and IDA, 2000. 25On September 12, 2000, the Bank's Board of Executive Directors approved a US$500 million emergency Multi-Country AIDS Program (MAP), conceived as an Adaptable Program Loan (APL) that would rapidly provide financial support to governments in Africa seeking to mount and/or implement multi-sectoral HIV/AIDS programs. To access MAP funding each country was required to satisfy four eligibility criteria (Box 2). Processing of individual countries' requests for financial support under the MAP was expedited through an accelerated review process and approval by the Regional Vice-President, validated within two weeks on the basis of the Board's non-objection (World Bank, 2000a). 7 the earlier MAPs) was to provide emergency funding to support and expand a multi- sectoral approach. Through the development of fiduciary infrastructure, it sought to increase the amount of money available to Ghana's HIV/AIDS efforts and to accelerate and facilitate the disbursement of these funds to a broad range of public sector and civil society actors, who would be mobilized to prepare and implement subprojects across a range of sectors and at the community level. The project placed strong emphasis on capacity building, considered essential to the rapid expansion of activities. Another important feature of the GARFUND design was its strong emphasis on monitoring and evaluation (M&E), supervision and operational research, both to (a) compensate for an accelerated preparation phase in the interest of making funds quickly available; and (b) to support a learning-by-doing approach. Approval of this project was contingent on the satisfaction of MAP eligibility criteria (Box 2). Box 2. Multi-Country AIDS Program (MAP) Eligibility Criteria · Evidence of a strategic approach to HIV/AIDS, developed in a participatory manner, or a participatory strategic planning process underway, with a clear roadmap and timetable · Existence of a high-level HIV/AIDS coordinating body, with broad representation of key stakeholders from all sectors, including people living with HIV/AIDS · Government commitment to quick implementation arrangements, including channeling grant funds directly to communities, civil society and the private sector · Agreement by the government to use multiple implementation agencies, especially NGOs and CBOs Source: World Bank, 2000b 2.4 The project's support was organized around the following four components summarized below (with estimated costs at appraisal, including contingencies). A more detailed inventory of planned support, including a list of eligible activities for financing, by type of implementing agency, is presented in Annex D. 2.5 Prevention and Care Services (US$21.0 million, 75% of total): This component was designed to finance the implementation of subproject proposals prepared by non-health public agencies and civil society entities with a view to mobilizing a broad spectrum of preventive and care activities at national, regional and local levels. Seventy percent of the cost of this component (US$15 million) was to be earmarked for civil society entities, the remaining amount available for public sector activities. Prevention activities were intended to reduce high risk behavior and exposure to risk, and reduce vulnerabilities by raising awareness and "de- stigmatizing" the disease with a view to improving access to prevention and care. Care activities would reduce the vulnerability of families affected by HIV/AIDS through financing: services to protect their rights; income generation activities; care for orphans; and home-based care for the ill. Financing was to be provided through three windows, each with slightly different access rules: Window A for proposals from public sector ministries, departments and agencies (MDAs) for internal and external clients (contracts up to US$100,000); Window B for proposals from NGOs, District Assemblies, the private sector (including private hospitals/clinics, as well as profit-making enterprises in all sectors), trade and professional associations, associations of PLWHAs, or groups of these entities, including affiliation networks (umbrella organizations) (contracts up to US$25,000); and Window C for seed money (not to exceed US$2,500) for very small community-based organizations (CBOs) and 8 associations. Eligible activities, summarized in Annex D by Window, included, inter alia: advocacy; development and dissemination of IEC; peer education; condom distribution; counseling; training and technical assistance; care giving; social support; research; legal advice; and support for PLWHAs. Some eligibility restrictions applied to civil works, vehicles, overseas training, salaries of permanent staff, clinical research and taxes. 2.6 Strengthening Public/Private Institutions for HIV/AIDS Control and Care Giving (US$2.4 million, 9% of total): Provision of training and technical assistance and the preparation of manuals and guidelines with a view to improving capacities in: project proposal writing, management, monitoring and evaluation; procurement and financial management; and building technical competencies in HIV/AIDS prevention and care. Some capacity building activities were to be conceived by the GAC Secretariat, while others were expected to be included in subproject proposals. The training was to be subcontracted to NGOs and line ministries with requisite experience and expertise. 2.7 Knowledge Management (US$1.4 million, 5% of total): This component was designed to set up and maintain an information clearing house and database on HIV/AIDS that would help implementing agencies improve their effectiveness. Implementers were to be provided with the latest information on HIV/AIDS prevention and care, best practice examples from other countries, research results, and any other reports that would be useful to them. Information sharing and dissemination would be supported through: fora for exchange among implementing entities, newsletters, published papers and research, and the internet. It was expected that umbrella organizations would have a key role in information dissemination. Regional AIDS Committees (RACs) were also expected to serve as focal points for dissemination, once established. 2.8 Project Management (US$3.0 million, 11% of total): This component was designed to support a new institution for National HIV/AIDS Program coordination, the Ghana AIDS Commission (GAC), which would be chaired by the Head of State, comprised of high-level representatives of key ministries, the private sector and civil society, and supported by a GAC Secretariat. Establishment of the GAC Secretariat and decentralized branches of the GAC (regional AIDS Committees - RACs, and district AIDS Committees ­ DAC) was also envisaged. 2.9 A strong monitoring and evaluation subcomponent was intended to assess the progress attained in the fight against AIDS nationwide on the basis of targets defined in the Strategic Framework and to monitor GARFUND implementation. It envisaged the collection and analysis of trend data through ongoing surveillance of prevalence among pregnant women attending ANC clinics, population-based studies on knowledge, attitudes, practices and behavior (on both the general population and on high-risk groups), and a DHS survey. It also envisaged the involvement of civil society in monitoring and ensuring accountability at the grassroots level. However, no specific budget allocation was assigned to this subcomponent. 2.10 Implementation Arrangements. GARFUND was established essentially as a fund to finance subprojects executed by a wide range of organizations at the 9 national, regional and local levels. GARFUND was to be placed under the overall authority of the GAC, which would: define broad priorities for action, oversee the selection of subproject proposals and monitor project performance. The GAC Secretariat was given direct responsibility for GARFUND oversight and management. It was supposed to solicit and vet subproject proposals, ensuring that proposed activities correspond to GARFUND's menu of activities eligible for funding. A Project Review and Appraisal Committee (PRAC), appointed by GAC, was to meet twice a year to select eligible subproject proposals, vetted by the Secretariat.26 Once approved subproposals would be endorsed by the GAC, the Secretariat would process and finance the contracts. To access funds, line ministries would be asked to submit proposals based on previously prepared HIV/AIDS action plans and to define a line item for HIV/AIDS in their respective budgets in order to contribute their own matching funds (10 percent of the cost of activities). To ensure that local needs were addressed and duplication was minimized, all implementing entities (NGOs, FBOs, CBOs) would be asked to obtain the endorsement of the districts where they proposed to work. While public sector, NGO, FBO, district and other proposals were to be reviewed and approved centrally, the review and approval of CBO proposals was to be carried out at the district level,27 based on district allocations fixed by the GAC secretariat. Annex E, Table E.1 presents institutional and technical eligibility criteria for the selection of subprojects for funding, by type of implementing agency. 2.11 All subprojects were required to be completed within one year and the implementing agency was required to submit: (a) interim reports, on which basis subsequent tranches of financing under the contracts would be released; and (b) a project completion report, containing technical and financial details of the subproject with the supporting documents including final bank statements. Project completion reports would need to be endorsed as satisfactorily completed before future funding for new proposals submitted by those implementing agencies would be considered.28 Random ex-post technical audits/assessments were to be subcontracted out by the Secretariat to verify activities completed. 2.12 The GAC Secretariat was also given responsibility for the financial and administrative aspects of GARFUND, capacity building of implementing agencies, assurance of compliance with Bank procedures, preparation and administration of contracts with executing entities, monitoring their technical and financial performance, preparation of financial and technical reports, and commissioning of annual audits, all in compliance with the GARFUND operational manual. Project financing of the GAC Secretariat covered the costs of doing business and salaries of all staff. 26In its review of subproject proposals, the PRAC, was to apply the following criteria ensuring that: the project would benefit the target group; proposed activities are feasible; costs would be reasonable; current priorities, as determined by the GAC, would be addressed; in-kind contributions would be incremental to the subproject; and participatory approaches would be used for subproject design (GAC, 2004b). 27The DAC, under the leadership of the M&E Focal Person, was responsible for evaluating, approving and selecting CBO subproject proposals. Contracts would be signed by the District Chief Executive and a copy would be sent to GAC for funding. 28While there was no explicit policy in the operations manual to prohibit the award of subsequent contracts to NGOs and CBOs, upon satisfactory completion of a subproject, two factors might have made discouraged follow-on contracts. First was the unwritten policy of the project to mobilize as many NGOs and CBOs as possible (interviews w/ Bank team members and w/ GAC staff). Second was the reference in the design of the document to CBO support as "seed money" (World Bank, 2000b) implying a one-time investment to stimulate action. 10 3. Implementation and Costs 3.1 The 16-month delay between project approval and effectiveness was due in part to the Presidential elections of December 2000 and in part to the need to establish the GAC before launching activities. The project was implemented over a period of just over three and one half years, closing on December 31, 2005, six months after the original closing date. Planned versus Actual Costs and Financing 3.2 The total project cost was US$26.6 million or 96 percent of the cost estimated at appraisal (Table 2). While the allocation for the Prevention and Care Services component was fully utilized, about two-thirds of the allocations of the remaining three components were utilized. Of the original IDA credit amount of 19.6 SDR, 92 percent was disbursed and the remaining amount was cancelled.29 Counterpart funding provided by government amounted to US$1.6 million versus the planned amount of US$1.8 million. Because of fluctuations in the exchange rate, actual counterpart funding in terms of cedis, is estimated to have been 118 percent of initial commitments. However, delays in counterpart financing and inadequate budgeting of contributions by the line ministries were both issues during implementation.30 Data on actual community counterpart financing was not available. IDA financing through GARFUND is estimated to be about 30 percent of total financing for HIV/AIDS in Ghana, based on HIV/AIDS National Accounts data available for 2003, the first full year of GARFUND implementation.31 Table 2. Planned Versus Actual Costs by Component (US$million equivalent) Component Planned b/ Actual c/ Actual as % of Planned Prevention and Care Services 21.0 21.9 104% Strengthening Public/Private Institutions for HIV/AIDS Control and Care 2.4 1.7 71% Giving Knowledge Management 1.4 0.9 64% Project Management (including Monitoring & Evaluation) a/ 3.0 2.1 70% Total 27.8 26.6 96% a/ No data were available on planned versus actual costs of the monitoring and evaluation subcomponent b/ World Bank, 2000 c/ World Bank, 2006 Planned versus Actual Inputs/Activities by Component32 3.3 Prevention and Care. A total of 3,03033 subprojects were approved and funded under GARFUND through a series of four calls for subproject proposals (Table 3). Of the total subprojects funded, 2,777 (92 percent) were completed.34 All CBO subprojects were completed, while completion rates for the remaining implementing agencies averaged 70 percent. The 253 projects reported as not having 29 Annex F, Table F.1. Because of the fluctuations in the US$ during the life of the project, actual disbursements are estimated at US$25.71 million, and cancelled amount at US$2.24 million, against the original equivalent value of US$25.0 million when the credit was approved (Source: World Bank Loan Department database). 30 Public agency counterpart contributions were reduced from 10 to 5 percent of work plan costs, but difficulties in meeting obligations persisted throughout the project. 31 Annex F, Table F.5 32 For more information, see Annex D which details project implementation by component. 33 This figure, provided by the GAC Secretariat to IEG in June 2006, represents an update of the figure of 3,026 reported in GoG's contribution to the Implementation Completion Report (March 2006). 34 Completed subprojects are defined as those having received all financing tranches under their respective contracts. 11 been completed did manage to implement an estimated 70 percent their planned activities (GAC estimate), but were unable to access their last tranche of financing.35 Table 3. Subprojects Funded/Completed under GARFUND by Call for Proposals Implementing 1st Call 2nd call 3rd call 4th call Total Total Completed Agency June 2002 November 2002 June 2003 April 2004 # Projects s de s de s de s de Funded s ts ect ect ect ect ect proj# complet# proj# complet# proj# complet# proj# complet# proj# projec % completed MDAs 17 13 33 22 9 8 16 11 75 54 72 NGOs, FBOs, DAs 112 94 258 154 207 134 183 153 760 535 70 CBOs 342 342 872 872 960 960 0 0 2,174 2174 100 Private Sector 0 0 0 0 0 0 21 14 21 14 67 Total 471 449 1,163 1,048 1,176 1,102 220 178 3,030 2,777 92 Source: GAC Secretariat, December 2006 3.4 Distributions of subprojects and of total subproject disbursements across the different types of implementing agencies show that CBOs implemented the majority (72 percent) of approved subprojects, followed by NGOs, FBOs and DAs (25 percent), MDAs (2 percent) and the private sector (1 percent). On the other hand, NGOs, FBOs and DAs, together, received the bulk (65 percent) of subproject financing, followed by CBOs (21 percent), MDAs (12 percent) and the private sector (2 percent). Subprojects financed under the GARFUND covered all ten regions of Ghana, with the highest HIV prevalence regions (Ashanti, Brong Ahafo, Eastern, Greater Accra and Volta) having the largest number of subprojects and the greatest shares of subproject financing (Annex E, Figure E.2). This is in part an outcome of the decision to prioritize these regions in the first call, but is also likely to be a function of their closer location to Accra, their previous experience in HIV/AIDS activities, stronger civil society capacity and greater access to support and information. 3.5 The average size of subproject contracts varied according to the type of implementer: US$34,000 for MDAs, US$20,000 for NGOs/FBOs, US$27,000 for associations of PLWHA, US$2,000 for CBOs, and US$19,000 for the private sector (Annex E, Table E2).36 3.6 Throughout implementation GARFUND sought to finance new, different NGOs and CBOs for every new round of proposals, as a means to incite activities in as many parts of the country and to spread the financing to as many beneficiaries as possible. There are no policies or guidelines in the GARFUND operational manual requiring or recommending this practice. Nevertheless it was documented by GAC and UNAIDS (GAC, 2004) and raised in numerous in-country interviews. Bank staff supervising GARFUND confirmed that, while not necessarily explicit, GAC's policy was "to mobilize as many NGOs and CBOs as possible." Reviews of subproject 35GAC did not receive completion reports on all subprojects that were fully disbursed. For example, a tally of 501 NGO/FBO subprojects shows that all of these subprojects were fully disbursed, but that only 287 (or 57 percent) had filed completion reports to GAC. 36The private sector was slow to submit subproject proposals to GARFUND, and it was only under the fourth call for proposals (April 2004) that private sector subproject proposals were approved and implemented with GARFUND support. 12 contract tallies corroborate this information, revealing very little incidence of "repeater contracts."37 3.7 Strengthening Public/Private Institutions for HIV/AIDS Control and Care Giving. Considerable amounts of training and workshops were financed under this component aimed at building the capacity of public sector and civil society implementers. At the national level five different training courses were carried out, each of which benefited trainees in all 10 regions. This training focused on skills development in proposal writing, financial management, procurement, project implementation, M&E, and "HIV/AIDS competency" (not defined). This training was delivered to a total of 1,337 trainees38 representing a range of stakeholders (DAs, RCCs, NGOs, CBOs, and MDAs). Regional and district level training benefited some 19,510 trainees,39 81 percent of, which received training in "HIV/AIDS competency," the remaining 19 percent distributed more or less equally across the remaining topics.40 Training was not evaluated for its impact. Capacity building activities were included in a significant proportion of subproject contracts: 60 percent of MDAs, 41 percent of NGOs/FBOs, and 41 percent of private enterprises. GHANET did receive support to carry out subproject activities, but only a small proportion of funds received were allocated for capacity building. GARFUND's practice of financing new, different NGOs/CBOs for every round of proposals undermined the effectiveness and sustainability of capacity building efforts. 3.8 Knowledge Management activities supported and disseminated several research studies.41 The Bank's team and the GAC Secretariat report that materials and best practices were disseminated to decentralized implementers in the regions and districts. However, (a) IEG could not find evidence to document to what extent and on what basis best practices were determined and how effectively and often they were disseminated42; (b) field interviews reveal a strong, unmet demand for technical advice, guidance and support, information on good practices, and financial and technical assistance for the support of relevant operational research43; and (c) GAC was unable to respond to IEG's request for a tally of all research financed under GARFUND itemizing the topic, research results and whether or how results were used to improve effectiveness. 3.9 Project Management. The GAC was formally established in January 200244 to direct and coordinate the involvement of all ministries, the private sector and NGOs in the fight against HIV/AIDS in Ghana. As the highest policy-making body on HIV/AIDS, its mission is to provide leadership in the coordination of all programs 37GAC Secretariat staff have since confirmed that only four of the 3,030 subprojects were repeater projects (communication to IEG of January 2007). However, in its official comments (Annex H-1) the Borrower states that this information is incorrect, but provides no additional data. 38It is possible that this total includes some double-counting due to the likelihood of the same individual attending more than one training session. 39 Ibid. 40 Proposal development/project management: 5 percent; financial management: 6 percent; care and support: 5 percent; and district strategic plans/HIV workplace policy development: 3 percent. 41Research/studies covered the following topics: schistosomiasis and HIV/AIDS; HIV among prisoners in Accra; HBC and community care of PLWHAs; nutrition and PLWHA quality of life; physically challenged and HIV/AIDS; seroprevalence among health workers; home-based care; PLWHA associations; safety of plant medicines, variations in prevalence/trends across sentinel sites; traditional practitioners in Ashanti and HIV/AIDS (GAC 2006b). 42In its official comments (Annex H-1) the Borrower notes that there is such evidence, but it is not provided. 43 All regions and districts visited gave unprompted comments to this effect. 44 By Act 613 of the Parliament of the Republic of Ghana in 2002. 13 and activities of all stakeholders (public, private and civil society) in the fight against HIV/AIDS through advocacy, joint planning, monitoring and evaluation. The President and Vice-President of Ghana serve, respectively, as Chairman and Vice- Chairman of the GAC, whose membership includes high-level representation of some 15 ministries, including MoH, and representatives of a wide array of civil society organizations. 3.10 By the same act of Parliament the GAC Secretariat was also established as the executive arm of the GAC to provide it with technical and administrative support. Headed by a Director-General, the Secretariat is made up of four Divisions: Finance; Administration; Policy, Planning, Research, Monitoring and Evaluation; and Technical Services. Although recruitment of these key positions was launched in 2001, most of these positions, including that of the Director-General, were not effectively filled until 2002, a key reason for delayed effectiveness. One crucial position, Director of Technical Services, was never filled. 3.11 The GAC met regularly three times a year to review progress towards the implementation of Ghana's Strategic Framework. The GAC PRAC (whose membership was made public and rotated once every two years) met twice annually to review and recommend to the GAC financing of subproject proposals. However, the planned mapping of high priority target groups was not implemented, making it impossible to track coverage of these groups with specific interventions. 3.12 Legal Covenants. The Borrower complied with all seven covenants related to financial management, audits and project reporting. 3.13 Procurement. Goods and services purchased by the GAC Secretariat were procured in accordance with procedures stipulated in the legal agreement. Procurement by implementing agencies, in the form of local purchases, was subject to post reviews, but review of documentation and field visits revealed little evidence of sufficiently frequent or systematic post reviews, or of corrective action taken. An audit highlighted the issue of lack of adherence by CBOs to procurement guidelines requiring three quotations for local purchases.45 Another audit report noted that because of the practice (for each new round of proposals) of favoring new NGOs and CBOs over those who had previously benefited from GARFUND support, procurement procedures may not have been appropriate.46 The mid-term review did not undertake a primary analysis of procurement. A 2003 post review was cited and simplification of procedures was underway. It did not cite cases of non-compliance of procurement procedures or recommend ways to improve capacity and practice. IEG review of documents and field visits revealed many incidences of items purchased under NGO and CBO contracts that were not always in sync with subproject activities or duration. 3.14 Financial Management. External audits and periodic supervision by GAC Secretariat and Bank supervision missions confirmed that financial management and accounting standards required by IDA were respected overall. Nevertheless, an audit of a sample of 75 CBOs financed under Window C revealed that many failed to prepare bank reconciliation statements and register or maintain a register for fixed 45Ernst and Young, 2003. 46KPMG, 2006. 14 assets funded under the project.47 This audit also revealed that CBOs made changes in signatories to bank accounts without GAC prior approval, many lacked supporting documents for items purchased or expenditures incurred, some NGO proposals were funded without a budget, many ineligible items were purchased, and the required beneficiary's contribution (10 percent) was not always documented. There were cases of duplication of payments and of relatively large sums transferred from subproject account to CBO accounts, as large as 20 percent of the contract amount. Some entities modified funding for specific activity without GAC approval. The use of third party receipts for payments was detected by auditors, but never followed up. Audit reports also cited instances of use of project money for meals, accommodation and purchase of computers, printers and office equipment that seemed inappropriate or excessive considering the nature and duration of project activities. Expenditures that were inappropriate or unaccounted for were corroborated by IEG's field visits, its own examination of project documentation and random inquiries with beneficiaries. Project funds to parliamentarians and traditional chiefs for HIV/AIDS campaigns (under Window B) were not fully accounted for at the time of the IEG mission.48 Random checks of subproject proposals by IEG corroborated incidences of unnecessary hardware purchases for short-term outreach activities. Inadequate screening and lack of accountability of NGOs/CBOs were also corroborated by interviews, field visits and documentation reviewed.49 Lack of information at central, regional, district and local levels on what other partners were funding led to duplication of financing and efforts. Implementation arrangements gave the RACs and DACs no authority to examine accounts of NGOs/FBOs and audit reports were never shared with these decentralized authorities. 3.15 The GAC Secretariat's efforts to monitor the use of funds for intended purposes and for the accountability of said funds did prompt it to carry out activities to obtain reimbursement from civil society organizations that were not to be using funds properly. Annex E, Table E.3 itemizes the results of these efforts, but notes that they were limited in scope (focused primarily on the 20 percent of projects audited/reviewed) and that for funds now accounted (especially funds used by Parliamentarians) supporting documents were not available for validation.50 3.16 Disbursements. The release of follow-on tranches to NGOs and CBOs under ongoing contracts was widely noted to be very slow, threatening timely project execution, continuity and completion, and, as a consequence, the credibility of implementing agencies. Delays were due to: the learning curve of GAC Secretariat and recipient NGOs/CBOs; absence of officials designated to sign payment vouchers; lack of familiarity of CBOs with banking procedures; and inadequate skills to prepare 47 Ernst and Young, 2003. 48 At the time of IEG's mission, six months after project closing (a) only 43 of the 158 members of Parliament who had received funds for HIV/AIDS campaigns had submitted a report to GAC. After IEG's mission an additional 65 submitted reports (see also Annex E, Table E.4); and (b) a traditional council contacted by IEG received 20 million cedis (US$2,200), a first tranche of a 100 million cedis contract, but only spent 3.5 million on a workshop to launch its activities. It reported that it was assessing how it would use the remaining 16.5 million. 49For one example, a meeting with a RAC, including representatives from NGOs and CBOs, noted that NGOs/CBOs have mushroomed since GARFUND has made so much financing available. IEG was told that it is very easy for two to three persons to write a proposal and access GARFUND money, and then disappear. GAC meeting minutes also document this concern. The Operations Manual specifies that in order to be eligible for financing NGOs must be registered, whereas eligibility criteria for CBOs are much more lax. IEG could not determine to what extent (a) the NGO criteria under the selection process were adhered to; (b) it is relatively quick and easy for NGOs to officially register; and/or (c) this issue was limited to CBO subprojects. 50 While in the field IEG was unsuccessful in obtaining sample reports from Parliamentarians to understand how funds were used, what activities were carried out and what results were achieved. 15 and submit statements of expenditures, on which basis a subsequent tranche would be released. 3.17 Mid-Term Review (MTR). The MTR took place in March 2004, at which time over 2,700 NGOs, FBOs, CBOs and MDAs had received GARFUND support. It concluded that progress in project implementation and in the achievement of development objectives was satisfactory and that project management was "best practice." Recommendations for improved performance and effectiveness included: shifting emphasis away from general IEC towards more targeted interventions aimed at specific groups (CSWs, homosexuals, youth, PLWHA associations, miners and transport) and toward high-prevalence areas. Efforts were also recommended to step up weak and late-start areas of project performance, particularly MDAs, private sector activities, and activities under the Knowledge Management and Strengthening of Public and Private Institutions components. The MTR also recommended working closely with the NACP/MoH for the sake of project efficiency and the analysis and use of available data on HIV prevalence and behaviors. 4. Monitoring and Evaluation Design 4.1 Program Monitoring and Evaluation (M&E).51 Strengthening research, monitoring and evaluation was accorded high priority by GoG, designated as one of the five key intervention areas of the 2001-05 NSF (Box 1). A National M&E Plan for the NSF was developed by GAC, with the technical assistance and support of USAID, DFID and other development partners. At the central level, the GAC would coordinate and monitor the activities of national level MDAs, NGOs, and RACs. A Directorate of Policy, Planning, Research, Monitoring and Evaluation was to be staffed by a Director, a Research Coordinator, an Information Officer and an M&E Coordinator. In addition, the Research, Monitoring and Evaluation Technical Committee of the GAC was given responsibility for creating a national set of core indicators for M&E of the national response. At the regional level, RACs were charged with monitoring community HIV/AIDS activities within their respective regions. A Regional M&E Focal Person would be responsible for preparing and implementing a regional M&E plan which would coordinate and track the activities of the DAC's. At the district level, DACs through the District M&E Focal Person were responsible for: (a) monitoring AIDS Programs implementation (tracking inputs, outputs, activities); and (b) providing data relevant for national level M&E, such as the tracking of trends in behavior change. At the implementation level, agencies would report on indicators to monitor the various activities of their respective programs. 4.2 Prevention indicators included HIV prevalence among women attending ANC (sentinel surveillance); HIV prevalence among female CSW in Kumasi; HIV prevalence of infants born to HIV infected mothers; behaviors and knowledge of young people (15-24 years); the delivery of HIV/AIDS education in schools; quality 51Source: GAC, 2002. 16 of STI services; percent of blood supply adequately screened; availability and utilization of services to mother-to-child transmission; and availability and affordability of VCT services. However, program indicators failed to capture and monitor progress in the knowledge and behaviors of high-transmission, vulnerable and other groups specified in the NSF. Indicators were also chosen to measure process and outputs on the other four components of the NSF: care and support; creation of an enabling environment; decentralized implementation; and program research, monitoring and evaluation. 4.3 GARFUND M&E. The project logframe included indicators and targets commensurate, for the most part, with those established for the national program, grouped around the two main objectives of reducing new infections (HIV prevalence, knowledge and behaviors) and mitigating the impact of the affected and infected (provision of specialized services).52 However, the logframe did not clearly articulate a results chain, which linked inputs, outputs, outcomes, and impacts. Some indicators were not appropriate (e.g., trends in HIV prevalence as a measure of reducing new infections), while others that would have measured the program's intent to target high-transmission and high-risk groups, in line with comparative advantages of the various actors, were missing. Notwithstanding these omissions, GARFUND did plan to map high-risk groups and monitor coverage of these groups and to use the data collected to set funding priorities for the subsequent funding cycle. There was no plan to assess the outcomes of public sector and civil society activities, however, even on a random basis.53 4.4 Trends in knowledge, attitudes, practices and behaviors were to be documented through: (a) the 1998 GDHS, with a follow-on GDHS planned for 2003 or 2004; and (b) a 2000 survey on knowledge, attitudes and behaviors for high-risk groups (youth, CSW, miners and police), with a similar end-of-project survey. Roles and responsibilities for M&E were not clearly articulated in the project design document, but the Borrower has confirmed that the new framework for M&E described in the national M&E plan was relied upon (Annex H-1). Interviews, field observations, and reviews of Bank and Borrower reports on project progress all provide evidence that incentives to disburse funds were stronger than incentives to measure results. Implementation 4.5 In May 2005 a national progress report on the implementation of the M&E plan noted that data was not being routinely and systematically reported to GAC; it called for the preparation of guidelines for the timely reporting of essential data for GAC.54 It was also recognized that capacity building and the provision of feedback for those collecting and analyzing data at the lower levels were critical to improving program M&E as was the need to review and revise reporting formats for improved consistency and accuracy. This review did not specifically note any gaps in indicators on targeted interventions for high-risk groups, only noting that indicators 52Annex G itemizes indicators, and baselines (where available) for GARFUND, by objective. The GARFUND logframe also included numerous process indicators (numbers of subprojects, implementers, districts supported) which are also reported on in Annex G. 53Report prototypes focused, rather, on inputs and activities. 54GAC, 2005a. 17 could be reviewed and revised only at the time of preparation of the M&E plan for 2006-2011 to ensure consistency in data collection and reporting. Development partners noted that throughout the life of GARFUND, GAC's M&E plan for the NSF was not operational and their dialogue with GAC to strengthen and operationalize it became a source of tension. 4.6 The GARFUND logframe and impact/outcome indicators were not revised during implementation.55 Baselines were never established for some indicators; other indicators were not adequately tracked. The timing of the two GDHSs was not ideal: the first GDHS was too early to serve as a baseline (1998) and the second one took place only one year after project effectiveness (2003, Table 4). The 2000 FHI- sponsored survey on behaviors of key target groups was a useful baseline, but was never followed up by an end-of-project (or end-of program) survey. There was no end-of-project data for the majority of indicators (Annex G). A Ghana Multiple Indicator Cluster Survey (MICS) was carried out in 2006, which collected some data on knowledge and behavior. However, because of the very recent availability of these data and their preliminary nature, IEG was unable to use them to assess trends. It was thus not possible to assess trends over time. GARFUND's reporting requirements strongly favored the tracking of financial flows and execution rates over results reporting, which did not facilitate the MAP's policy of "learning by doing." A mapping exercise of subprojects was carried out, which documented the geographic coverage of subprojects with a view to extending coverage to all geographic areas. IEC could find no evidence of a strategic assessment of the coverage of specific groups with specific interventions in response to epidemiological and behavioral evidence. Table 4. Timing of Data for Measuring Outcomes Year 1998 1999 2000 2001 2002 2003 2004 2005 Outcome GDHS BSS BSS GDHS Measures 1998 2000 2002 2003 Project start Effectiveness: Completion: and end May 8 December 30 4.7 While the institutional and organizational framework and staffing for M&E envisaged under Ghana's national M&E plan were clear, implementing agencies did not have the proper training or tools for setting targets and measuring their outcomes.56 The job descriptions for regional and district M&E focal persons demanded the skills of highly qualified, full-time professionals. In contrast, district and regional M&E focal persons were civil servants with full-time responsibilities to their sectors, fulfilling HIV/AIDS M&E responsibilities on their own time. Training, technical supervision and support, and tools were inadequate. RAC and DAC attempts to monitor and evaluate activities were undermined by the fact that NGOs were approved by, financed by, and reported directly to, GAC and did not consider themselves accountable to regional/district AIDS Committees or focal persons. Time, 55At the time of the MTR it was decided to keep the original impact and outcome indicators. Nevertheless, a few process indicators were added to the mix to facilitate assessment of implementation progress: share of line ministries, districts and communities that have prepared HIV/AIDS work programs; number of associations of PLWHA that are participating in the project; lag between submission and approval of plans by level; lag between approval of plans and disbursement by level; percentage of districts/communities that have mapped high transmission areas. 56 The Borrower in its comments (Annex H-1) cites a number of M&E capacity building activities that were financed by sources other than GARFUND, but the fact remains that M&E capacity remained weak. 18 skills and resource constraints limited the work of M&E focal persons largely to the consolidation of subproject reports (with or without field visits) and dispatching them to GAC, without analysis, validation, or (for the most part) local use. Utilization of Data for Decision-Making 4.8 Information on key prevention and care indicators57 was not collected and thus could not be used for decision-making. More fundamentally, the incentives to collect and analyze data to improve program/project impact were weak. Rather, incentives to expand the fight against HIV/AIDS (in terms of number and range of implementers and level of disbursements) were very strong. IEG's review of reports in the regions and districts visited revealed that M&E activities were considered a requirement by GAC/GARFUND, rather than a management or accountability tool. While M&E reporting requirements were not considered interesting or relevant to M&E focal persons, they did express a strong demand for operational research. There was an overwhelming opinion among those interviewed in the field that research in key areas would have provided local managers and implementers with information that would enable sound strategic decision-making for enhancing performance. On the other hand, an assessment of orphan support (para. 5.21) does not seem to have been used to improve the effectiveness and efficiency of efforts in this regard. 5. Outputs and Outcomes by Objective 5.1 In assessing attribution of outcomes to GARFUND, it is important to note that GARFUND (a) represented about 30 percent of all public and private HIV/AIDS financing in 2003 (its first full year of implementation),58 (b) excluded the support of HIV/AIDS activities carried out by MoH;59 and (c) supported non-health implementing agencies (public and private) that were also receiving financial support from other partners in development. There is incomplete information on the full costs and financing of the GoG's fight against HIV/AIDS by programmatic activity, by geographical region, by implementing agency/recipient of financing, and by financier.60 However, during GARFUND implementation (2002-2005) an estimated US$10.6 million in external assistance was earmarked for HIV/AIDS activities carried out under MoH,61 and an additional US$91.3 million in external assistance was channeled directly to key sectors (such as Education), NGOs, CBOs and other local-level agencies).62 Many civil society organizations interviewed stated that, in 57 For example, knowledge, behaviors, access to and utilization of care and support services. 58 UNAIDS and GAC, National HIV/AIDS Accounts, 2002-2003 (Annex F, Table F.5). 59 Activities to reduce HIV transmission (promotion, provision and expansion of VCT and PMTCT services); improvements to the quality and coverage of care and services for PLWHA (ART and treatment of OIs, STI treatment, lab services, support to PLWHA associations for income generation and other activities); epidemiological surveillance; and production/dissemination of information and guidelines on prevention and care. 60 The National HIV/AIDS Accounts (2002-2003) present of total expenditure by program intervention and by type of financier, but expenditures by individual financier, implementing agency, and region/district are not itemized. 61 Global Fund for AIDS, Tuberculosis and Malaria (GFATM), UNAIDS, EU and Danish Embassy earmarked these funds, over and above allocations to the NACP within the health budget including basket funding provided by selected donors (on which specific data could not be obtained) (Annex F, Table F.5 ). The following year an additional US$14.9 million IDA grant was approved for Ghana under the US$60 million Regional HIV/AIDS Treatment Acceleration Project (TAP), managed by MoH, to improve quality of life and quality of care of PLWHA; prevent mother-to-child transmission and decrease the risk of sexual transmission; and diminish the stigma of HIV/AIDS. 62 DFID, USAID, Netherlands, WHO, UNICEF, UNDP, CIDA, UNFPA, IPPF, FBOs and private corporate entities 19 addition to GARFUND, they were receiving financial support from other sources, but were not able to present an overview of costs and financing of their activities. Annex G presents a matrix summarizing this evaluation's attempt to show program and project outcomes against targets and indicators by project objective and subobjective. For reasons noted in Chapter 4, this was a difficult undertaking, which has produced an incomplete patchwork of data that must be interpreted cautiously. Reduce the spread of HIV infection 5.2 The GARFUND rapidly increased the availability of financial resources to support Ghana's fight against HIV/AIDS. Between 2002 (the year of GARFUND start-up) and 2003 (its first full year implementation), total annual expenditures on HIV/AIDS in Ghana more than doubled, rising from 108 billion cedis (US$13.5 million) to 237 billion cedis (US$28.1 million).63 Forty-eight percent of the increase was attributable to GARFUND, and 47 percent was attributable to other foreign assistance, especially GFATM's first disbursement to Ghana in 2003 (Annex F Table F.5). These expenditure levels are likely underestimates, as they do not fully capture out-of-pocket expenditures for traditional medicine or the considerable non-monetary transactions and contributions of families and communities. 5.3 However, the efficiency of overall HIV/AIDS expenditures has not improved. While the mobilization of financial resources for non-health activities was the intention of GARFUND, health sector investments essential to the achievement of adequate coverage and quality of HIV/AIDS-related services (health personnel, reagents, consumables, condoms, drugs, etc.) are noted by GAC and UNAIDS to be underfinanced. National HIV/AIDS Accounts analysis for 2002-2003 show that expenditures on health activities increased only marginally, relative to overall increases, while non-health expenditures increased almost eightfold (Figure 1). In addition, expenditure (all financing sources included) on high-transmission groups and other target groups, specified in the NSF and in GARFUND design documents as high-priority, is extremely low at barely 1 percent of total expenditures for 2003 (Annex F, Table F.6), and low coverage of target groups persisted throughout implementation. Expenditure on CSW, which is known to be highly effective and efficient in averting new infections, was only 0.2 percent of all expenditure in 2003, or 1 percent of total GARFUND expenditure for that year, despite nationally-set and project-specific targets to reduce infections among this group. 63Source: GAC, UNAIDS, et al., 2004 ; exchange rates used for conversion into US$: 8050 (July1, 2002); 8450 (July1, 2003) (World Bank Loan Department) 20 Figure 1. Ghana 2002-2003 HIV/AIDS Expenditures by Category (in billions of Cedis) 99.8 Nonhealth expenditures 1/ 12.8 Public health and 91.9 prevention 2/ 64.5 Personal health 30.1 services 3/ 21.4 Maintenance and 15.4 strengthening 4/ 9.5 0 10 20 30 40 50 60 70 80 90 100 Billions of CEDIS 2002 2003 Source: National HIV/AIDS Accounts, 2005. 1. Services and activities out of the health sector classification of functions, but relevant as a part of the HIV/AIDS country response, such as research and development, staff training, advocacy, organization and empowerment, policy dialogue. 2. Heath services whose benefits are public in nature, for example: epidemiological surveillance, IFC, condom use promotion and distribution, blood banks, among others. 3. Health sector services that benefit the individual client, including treatment, medicines, small equipment, diagnostics and laboratory services and supplies. 4. Including administration costs, and investments in equipment and infrastructure. 5.4 GARFUND has mobilized a broad range of actors and sectors for a more expanded and expansive response and supported a wide array of HIV/AIDS interventions. The financing of subprojects in the public sector engaged and supported numerous actors in the preparation and implementation of work place policies, including key sector ministries,64 research institutions, Regional Coordinating Councils, District Assemblies, Parliamentarians, traditional councils and chiefs, and other public sector agencies.65 Through its financing of civil society subprojects, hundreds of NGOs/FBOs and thousands of CBOs were mobilized into action in all of Ghana's districts. Thirteen PLWHA associations in six regions were provided with technical and financial support with a view to increasing and enhancing their involvement. Within the private sector 11 health-related facilities and nine private enterprises were brought into the fight. Figures 2 and 3 provide an overview of the types of interventions supported under subprojects designed and implemented by public sector agencies and NGOs/FBOs, respectively. Public sector interventions supported the establishment of workplace interventions and interventions aimed at the clients of these agencies. CBO subprojects were too numerous to analyze in detail, but a quick review of tallies indicates that the vast majority focused their efforts on peer education, advocacy, awareness creation, with far fewer subprojects focusing on care and support. Private health facilities concentrated their efforts on VCT, care and support, capacity building and IEC, while private enterprises supported workplace policies and interventions, focusing primarily on IEC and, to a lesser extent, peer education, advocacy and capacity building (Annex E, Figure E.3). The exact nature of activities undertaken within these broad categories of interventions (for each group 64 Education; Justice; Transport; Food and Agriculture; Manpower Development and Employment; Local Government and Rrual Development; and Women's and Children's Affairs. 65 Some 26 subprojects were executed by other public sector agencies, including, among others: the Food and Drugs Board, the National Fire Service, the Civil Servants Association, the National Commission for Civic Education, the National Population Council, the National Commission on Culture, regional branches of the National Disaster Management Organization (NADMO), Customs, Excise and Preventive Services and the Social Security and National Insurance Trust. 21 of implementing agencies) is not known, nor is the coverage of interventions documented. Figure 2. Percent of MDA Subprojects Supporting HIV/AIDS Interventions (n=70 subprojects) Advocacy 79 Capacity Building 60 IEC 43 VCT 26 Counseling 23 Other 11 Condom Distribution 11 Peer Education 6 OVC 4 Care/Support 4 0 20 40 60 80 100 Percent Figure 3. Percent of NGO Subprojects Supporting HIV/AIDS Interventions (n=501 Subprojects) Peer Education 48 IEC 46 Capacity Building 41 Counseling 37 Condom Distribution 34 Advocacy 34 Care/Support 27 VCT 6 Social Support 3 Awareness 2 Behavior Change 1 0 20 40 60 80 100 Percent Source: GAC, December 2005 5.5 GARFUND has contributed to national capacity building efforts for HIV/AIDS in Ghana. It supported the establishment and operations of the GAC and its Secretariat and provided technical and financial support to the review and refinement of the national strategy with other partners. GARFUND has also supported the decentralization of the fight against HIV/AIDS. Regional and District AIDS Committees (RACs and DACs) are now operating in Ghana's 10 regions and 110 districts and have benefited, along with civil society actors in every region, from management and technical training. Guidelines for subproject proposal preparation, review and implementation also supported capacity building of local actors. Nevertheless, the institutional framework for HIV/AIDS remained rather centralized in its functions and understaffed at the decentralized levels and roles and responsibilities between the central, regional and district levels were reported by many informants to be unclear and not fully reconciled with the DRI being piloted at the time of project design. 5.6 Prevention subprojects financed by GARFUND were geared for the most part around the transmission of basic knowledge about the disease through IEC and peer education, instead of a more concerted effort to address and change 22 specific behaviors. Inclusion by most subprojects of advocacy, peer education, IEC, sensitization is likely to have contributed (along with other GARFUND-financed efforts66) to improved knowledge of the main modes of transmission and prevention methods. Subprojects with condom promotion and distribution were small in scale and irregular, and thus not likely to have contributed substantially to ensuring sustainable access or use of condoms in the regions and districts. Regional and district focal persons did not have specific information on the total numbers and destinations of condoms distributed in their respective catchment areas under the 2001-05 NSF, or on GARFUND's particular contributions to this end. 5.7 Most subprojects were targeted to the general population and youth (in and out of school), with relatively few instances of interventions for specific high-risk groups that would be expected to have the largest impact on new infections (subproject documentation and field interviews). Interventions for high-risk groups are very modest in terms of (a) the level of effort of the ongoing targeted interventions; and (b) the small number of targeted interventions, indicating low coverage of some high-risk groups and no coverage of others.67 5.8 The quality and likely efficacy of subprojects financed under GARFUND suffered because of insufficient rigor in the review of subproject proposals and NGO/CBO capacity, thus undermining their results orientation and their potential development impact. Both development partners and regional- and district-level actors noted that selection/approval of proposals was not sufficiently rigorous. Development partners noted that many technically inadequate proposals were approved. District-level actors noted that while GAC approval of NGO subproject proposals was contingent on endorsement by the districts, in actual fact districts were encouraged to dispatch proposals immediately to GAC for their rapid approval, without reviewing or commenting on them.68 GAC and UNAIDS documented that the selection of NGOs and CBOs "...is done each year only considering the organizations that have not received support in previous years (GAC, 2005)." Many, if not most, NGOs and CBOs were inexperienced in designing and implementing HIV/AIDS prevention activities, but this practice of approving subprojects submitted by implementing agencies that had not previously received GARFUND financing encouraged them to join the effort.69 5.9 Prevention activities did not receive the requisite technical guidance and support. District focal persons had neither the time nor the expertise to provide technical guidance70. The consequences of this were evident in some billboards, 66 GARFUND financed development of 17 productions for TV (a few comprised of several episodes); 13 radio productions; and 6 music productions on various HIV/AIDS topics which aimed to increase knowledge about the risks and methods of protection; stigma and discrimination; the national policy and program; and NGO efforts (GAC, 2006). 67 Coverage estimates are notional at best because mapping and counting of high-risk groups were not undertaken. Evidence of extremely low coverage of high transmission groups by Ghana's AIDS efforts more generally is also documented in the 2002- 2003 National HIV/AIDS Accounts (Figure 4). Recently, development partners have incorporated targets of increased coverage of high-risk groups and geographical areas into the policy matrix of the multi-donor budget support initiative. 68 The Borrower, in its comments (Annex H-1), has noted that District AIDS Committees (DACS) reviewed and recommended proposals for the 3rd and 4th calls but not for the 1st and 2nd calls during which time DAC capacity was considered to be weak. 69 Development partners and local-level actors alike noted that the availability of financing under GARFUND brought a significant proliferation of NGOs and CBOs, some of them created for noble reasons, but others established simply to access funds. 70 The Borrower, in its comments (Annex H-1), cites some training and guidance provided with other financing, but also notes considerable turnover of focal persons. 23 posters and other material that were inappropriate, considered offensive to certain groups, technically incorrect and/or risked further stimulating fear, stigma and discrimination. The failure to fill the Technical Director post in the GAC Secretariat and the exclusion of MoH from the GARFUND combined to undermine the provision of needed and requested71 technical assistance and supervision or regional and district-level activities. 5.10 The opportunity to improve the effectiveness of activities over time was undermined by the failure to undertake essential operational research. Very little operational research was supported by GARFUND at the regional and district levels in order to improve Figure 4. Expenditures by Target Group as a Percent of Total the effectiveness of HIV/AIDS Program Expenditures, 2003 prevention interventions. The 70% design and targeting 58% 60% of prevention 50% activities were thus 40% not systematically 30% tailored to local- 15% 17% 20% level evidence. 5% 10% Rather, prevention 1% 0% activities, including High Vulnerable PLWHA Health facilities Non-targeted transmission groups** staff expenditures*** those carried out by groups* local-level Source: National HIV/AIDS Accounts, Ghana, 2002-03, UNAIDS and GAC implementers * High Transmission Groups include: CSW, men who have sex with men, blood remained very donors, migrant workers, uniformed population, prison inmates general, drawing ** Vulnerable Groups include: school children, workers, children at risk of heavily on generic vertical transmission, pregnant women, children and youth in social risk *** Includes administrative overhead and non-targeted activities materials made available by GAC and general knowledge. Local-level actors interviewed were emphatic in their call for the support of relevant operational research for more appropriate and more effective interventions.72 5.11 There are no documented trends in awareness, knowledge, attitudes, and behaviors among the general population or among high-risk groups over the period of project implementation, data which might have suggested plausible impact on new infections. Evidence gathered through field visits provides some insight, although it must be interpreted very carefully. 5.12 Field visits indicate a likely increase in the knowledge of modes of HIV transmission and prevention methods among the general population and among youth. Informants in the regions and districts visited noted that the creation and work of the RACs and DACs, and the expanded prevention activities of civil society and 71Regional- and district-level actors widely noted the absence of sound, pedagogical supervision and their need for guidance in the design and implementation of their activities. 72This was a strong recurring theme of IEG's field discussions and corroborated by central level actors and development partners, alike. A wide range of topics for operational research was suggested by respondents, including (but not limited to): the tendency to hide rather than bring rape cases to justice; the behaviors and role of parents in protecting youth; the belief that sex with virgins will cure AIDS; the practice of wife inheritance; effective messages for Muslim youth; behavioral underpinnings of specific groups and the dynamics of the epidemic (causes and consequences of behaviors of street children, customs/border workers; port workers; migrant workers, etc.). 24 other sectors in the regions and districts (financed by GARFUND and by other sources) have stepped up considerably the open discussion of HIV/AIDS, its risks and ways and means of preventing its spread, at the local, community and individual level. They note that, as a consequence of the increased detail and frequency with which HIV/AIDS prevention messages are disseminated, knowledge has increased in their respective districts. Because a great majority of these discussions and activities were targeted at the general adult population and youth, both in and out of school, it is plausible to assume that levels of knowledge within these groups would have increased during the life of GARFUND. On the other hand, respondents did not cite any perceptible changes in knowledge among any of the high-risk groups nor were any trend data available.73 5.13 Field visits provided no indication of significant changes in levels of fear, stigma and discrimination associated with HIV/AIDS during the period of GARFUND implementation. Informants from virtually all regions/districts visited, including PLWHA, noted that fear, stigma discrimination have not improved in the last several years. The demand to be tested is still low, despite the fact that VCT services are increasingly available.74 Discussion with RAC members and implementers in one region raised the issue of counselors violating the confidentiality of their relationship with those being tested. Few who test positive seek support, as this would mean disclosing their status. It was pointed out that the decision not to disclose one's status is not limited to those who are uneducated; even civil servants testing positive choose not to disclose their status. A discussion with RAC members and implementers revealed that families often present their PLWHA family members as victims of food poisoning. 5.14 Numerous associations of PLWHA were formed over the period 2002-200575 and more are reported to be reaching out to populations to share their experiences. A number of respondents observed that the holding of monthly meetings of these groups at the regional and district levels has the potential for combating stigma and discrimination. However, current membership represents a small fraction of the estimated number of PLWHAs. Discussions in the field revealed cases of mothers prohibiting their children from attending PLWHA association meetings for fear of family dishonor. PLWHAs have indicated that despite meetings and outreach and income generation activities of associations, they have observed no change in stigma and discrimination over the past several years. 5.15 Field visits and data on condom sales indicate an increase in condom use during GARFUND implementation. Respondents in various regions/districts visited reported that demand for condoms seems to be rising, along with the availability and sales of condoms. One region (Upper East) reported a likely increase in use, based on the fact that more and more criminals arrested were found to have condoms in their pockets. Indeed, the annual sale of condoms in Ghana rose from 24 million to 30 million between 2002 and 2005. Sales had increased more rapidly in the years prior to 73This is not to state conclusive evidence that knowledge did not increase among any of these groups, only to note that there was no indication of any changes that IEG could find. 74The RAC of Bolgatanga estimated that 10 percent of adults have been tested, and most of those are women. 75Statistics on total number of PLWHA associations in the country is not available, but GARFUND financed the subprojects of 13 and NACP/MoH with GFATM financing supported 27. 25 GARFUND rising from 10 million in 1998 to 24 million in 2002. A major social marketing program supported by USAID is likely to have been instrumental in achieving these sales. While it is plausible that GARFUND interventions may also have contributed, such contributions are likely to have been modest because (a) condom promotion and distribution made up a very small part of all subproject activities and (b) the bulk of prevention efforts were focused more on IEC than on behavior change. Other regions (in the north, as well as in the south/port areas) reported that young men are resisting condom use. They are hesitant to access condoms where they would be available publicly because they would be labeled as a "bad boy." Informants in various districts visited and at the national level also reported that religious leaders are uncomfortable with condom promotion and encourage abstinence (for youth) and fidelity (for married couples). 5.16 It is not possible to document trends in new HIV infections in Ghana during the life of GARFUND, either in the general population, or among high-risk groups. At the time of project design (2000), an estimated 3.0 percent of ANC women were HIV positive and a target was set to maintain HIV prevalence in this group at 6 percent or below.76 While the prevalence of HIV in Ghana fell, trends in HIV prevalence are not a valid indicator of trends in new infections. A decline in HIV prevalence could conceivably be the result of high AIDS mortality or unsuccessful treatment efforts,77 while an increase could signal more widespread and successful treatment. 5.17 Trends in HIV prevalence in the youngest age cohort of pregnant women (15- 19 years) can serve as a proxy for infection rates, as this group has not been sexually active as long as the older groups and is less likely to experience AIDS mortality. HIV prevalence rates in this age group in Ghana show significant decline, from 2.3 percent in 2002 to 0.8 percent in 2005. However, this trend must also be interpreted with caution. First, it is a proxy for infection rates in that particular group only (pregnant women, attending ANC clinics, aged 15-19), and should not interpreted as an indication of infection trends in other groups (all pregnant women, all age groups, all men and women, for example). Second, changes in the composition of sentinel sites over time, as well as changes in fertility as a result of infections with other STDs can lead to spurious trends, which may not be reflective of the actual situation. 5.18 Data on HIV prevalence among commercial sex workers were collected in the cities of Accra (2001) and Kumasi (2002), documenting high rates among seaters (76 percent and 54 percent, respectively) and somewhat lower rates among roamers (23 percent and 15 percent, respectively).78 In 2006 data were collected again for the cities of Accra/Tema and Kumasi documenting rates among seaters (52 percent in Accra/Tema and 39 percent in Kumasi) and roamers (37 percent in Accra/Tema and 24 percent in Kumasi).79 It is not clear to what extent these trends are attributable to AIDS mortality, successful treatment and/or high turnover within each cohort. No 76This target was set on the basis of information available in 1999 when it was assumed that (a) 4.6 percent of the adult population was infected; and (b) prevalence was rising (Measures Project and GAC, 2003). 77This is particularly relevant to the Ghana context as ART was rolled out in the country during the life of GARFUND. 78NACP/West African Project to combat AIDS- (WAPCAS)Canadian CIDA/ Strengthening HIV/AIDS Response Project (SHARP) of Academy for Educational Devt of USAID. Seaters are CSW operating from a residence; roamers are mobile CSW, operating in bars, truck stops, market places and other areas of high frequency. 79"Female Sex Worker HIV/STI Behavioral Surveillance Survey (2006)," Strengthening HIV? AIDS Response Partnerships (SHARP) Project, Accra, Ghana, 2007. 26 trends are available on new infections or HIV prevalence rates during the time of GARFUND implementation among other high-risk groups identified during project preparation (truck drivers, prisoners, military, border, port/harbor workers, miners, migrants, other). Neither baseline data nor end-of-project data were collected on any of these groups. Reduce the impact of AIDS on those who are infected and their families 5.19 GARFUND's contribution to this objective was measured in terms of the supply of two services: (a) home-/community-based care of PLWHA; and (b) care of orphans. The quality, appropriateness and effectiveness of these services in mitigating the impact of AIDS on PLWHA, their families, and their orphans, remain important, but unexplored questions. 5.20 The availability of home- and community-based services for the care and support of PLWHA is reported by GAC and by most informants in most regions/districts to have increased. Half of all districts now provide community- based care for PLWHA, fully satisfying the 50 percent target set during the design stage. According to GAC statistics, the project has supported home-based and community-based care services for more than 30,000 PLWHA nationwide or about 8 percent of the total estimated number of PLWHA. It is difficult to appreciate this coverage in the absence of (a) information on support to PLWHA financed by other sources; and (b) the number of PLWHA that sought medical treatment (the denominator).80 5.21 The absence of a functional referral system between the health system and community-based care likely reduced the efficacy of home-based care interventions. The project target was to achieve functional referral systems in 40 percent of districts by the project's closing. While they have not been systematically measured, the quality, effectiveness and sustainability of community-based services are likely to be deficient in the absence of established links with the formal health system, through which these services should receive referral and technical support and guidance.81 A GAC-commissioned study in 2003 documented the low number of home-based care initiatives in 2003. In that year, of the 40 home-based care programs, 21 were managed by FBOs, 13 by NGOs, 5 by government and 1 by a CBO. This report noted that both the quality and coverage of existing home-based care programs were low.82 5.22 The geographic coverage of orphan support activities met GARFUND targets, but the efficacy and efficiency of these efforts are uncertain. By the end of GARFUND, 35 percent of all districts were reported to be providing care for about 17,500 orphans, exceeding the project target of 30 percent. But the spectrum and combinations of support are wide and varied, including payment of school fees, 80 The project target was to achieve a coverage of 30 percent of all PLWHA that sought medical treatment. 81 NACP/MoH support for care and treatment consists of treatment of OIs and ART and supporting activities (ART management, capacity building of service personnel). In 2004, 4,504 patients were receiving HIV clinical care through the public health system, of which 2001 were receiving ART. There was no indication in its annual report of the provision of technical support and guidance to home and community-based care. Guidelines for managing OIs and on managing the side effects of ART were produced, but it is not clear if these were made available to care-givers outside the health system (NACP 2004). 82 GAC, 2003b 27 provision of nutritional supplements, medical assistance, housing, etc. A UNICEF study, commissioned and financed by the GAC in 2004, provides insights into the nature and effectiveness of GARFUND assistance to AIDS orphans through a community foster care program developed by the Queen Mothers of Manya Krobo district in Eastern Region (Box 3).83 While Queen Mothers had traditionally financed the care of orphans with monthly dues paid by members of their association, starting around 2000 they sought and received training, financial and other solicited support from NGOs and international agencies. In 2002-2003 they received a GARFUND grant of about US$30,000, supplemented by US$20,000 from other sources.84 Box 3. Queen Mother's Community Foster Program for Children Affected by HIV/AIDS, Manya Krobo District, Eastern Region: Findings of a 2004 Review The Queen Mothers Association supports a total of 600 children, including those whose parent(s) die or become too ill to care for them. About 10 percent of foster parents subcontracted the care of their foster children to persons who were unknown to the program manager of the Queen Mothers Association and therefore unsupervised. Feeding costs were calculated on the basis of three meals, but most children ate only twice a day and foster children from Upper Manya were found to be less well nourished overall. Only 52 percent of foster children were in school at the time of the study because GARFUND support had not been received for 2004. The health needs of the orphans were poorly addressed, with only two of the children tested for HIV and none of the children receiving vitamins, malaria prevention treatment or insecticide treated bed nets for sleeping. Only four seropositive orphans had been taken to a health facility in the year of the study. Yet, foster parents claimed an average of US$15 in health expenditures per child per month. The psychosocial needs were found not to be adequately addressed for many of the children who suffer from anxiety, depression, discrimination and stigmatization, although orphans were found to be better integrated into their foster families than those in other programs with less stigmatization and discrimination within the community. Over 80 percent of children were related to their foster parents. Nevertheless, about two-thirds of 50 orphans interviewed reported being subjected to emotional abuse, 24 percent to physical abuse, 10 percent to neglect, and 4 percent to sexual abuse. Other forms of maltreatment were reported by foster children, including denial of food, insults and beatings. Over 80 percent were found to be engaged in income generating activities, ranging from daily hawking of wares to hard labor (crushing stones in a quarry), 44 percent of whom were under 15 years old. There were no formal selection criteria for foster parents or foster children and no limits on the number of foster children per household, with some foster parents caring for as many as six children, netting considerably more than the average household income in the district. Only 25 percent of parents had been trained in the care and support of PLWHA. The Manya Krobo District Social Welfare office was found to be inadequately staffed and under-resourced, with just two field officers, already overstretched with a wide range of responsibilities, and lacking basic training in the care of orphans and vulnerable children, and in HIV counseling and care. Neither did this office have the basic infrastructure and logistics to carry out its critical role of counseling, oversight and supervision. (Source: UNICEF et al., 2004). 6. Ratings 6.1 This PPAR evaluates performance against objectives and expected outcomes documented in Ghana's own national strategy and in the project design document. The outcome of the GARFUND project is unsatisfactory, based on its modest relevance, modest efficacy and negligible efficiency (Table 5). 83 Each traditional area in the Manya Krobo district has a Queen Mother, who is responsible for the welfare of women and children of the area and promotes and preserves traditional cultural norms. 84 World Bank/IFC, UNAIDS, UNDP ­ MOWAC, Queen Mother subscriptions/year and others (Soroptimist Society, Calvary University, Catholic Relief Services and World Vision). 28 Table 5. Summary IEG Ratings* by Objective Development Objective Relevance Efficacy Efficiency Outcome Reduce the spread of HIV infection. Modest Modest Negligible Unsatisfactory Reduce the impact of AIDS on those infected and their Modest Modest Negligible Unsatisfactory families Overall Project Unsatisfactory * See inside cover of this report for definition of relevance, efficacy, and outcome. 6.2 The overall relevance of the project is modest. The relevance of project objectives is substantial. They are responsive both to the Government's past (2001- 2005) and current (2006-2011) HIV/AIDS Strategic Frameworks, and to the Bank's previous and current CAS, which emphasize poverty reduction and human development, with specific focus on the prevention and control of HIV/AIDS and the mitigation of its impact. 6.3 The relevance of project design, on the other hand, was modest. The design did not ensure that activities implemented would be the most responsive to the priorities of Ghana's particular epidemic, which, as noted in the national strategy and the design document, was concentrated in certain groups and in certain regions. The project's drive to mobilize and support as broad a range of actors working in as many communities as possible diverted attention away from high-impact interventions specified in the design document and from the strategic use of civil society organizations and various sector agencies to address hard-to-reach groups, in line with their comparative advantages. The project logframe was not systematic in developing a coherent results chain with strong links between inputs, outputs, outcomes and impact. 6.4 The institutional framework for program/project management and implementation did not build on existing capacities and frameworks at central and decentralized levels. The adequacy of existing institutional arrangements for HIV/AIDS was not sufficiently assessed. Over and above the exclusion of MoH activities from GARFUND financing (World Bank, 2000b), the organizational structure and mandate of the GAC and the GAC Secretariat (Act 613 of Parliament, January 2002) significantly reduced the role of the MoH, curtailing its potential to provide needed technical guidance to efforts at national, regional and district levels (Box 4).85 M&E focal persons operating at the regional and district levels had neither the adequate time, nor the proper training to carry out their responsibilities. Decentralization of the HIV/AIDS effort was also hampered by GARFUND arrangements whereby NGOs and FBOs were contracted centrally by the GAC, reported directly to the GAC and, as a consequence, did not feel accountable to RACs and DACs, which were supposed to supervise them. 6.5 Overall the efficacy of the objective to reduce new HIV infections was modest. HIV prevalence rates among young pregnant women (15-19 years) attending ANC clinics indicate a possible decline in new infections in that particular group. The lack of trend data on knowledge attitudes and behavior is one reason for this rating. Another is that support is likely to have had modest impact, at best, on any changes in 85The Borrower, in its comments (Annex H-2) notes that the poor relationship between MoH and the Ghana Health Service (GHS) made it difficult to implement GARFUND. However, field visits and interviews attributed these issues to poor coordination and unclear division of roles and responsibilities between GAC and the health sector at national, regional and district levels, and not to issues between MoH and the GHS. 29 knowledge, attitude and behaviors, given: inadequate focus on behavior change and high-risk groups despite nationally-set and project-specific targets to this end, lack of operational research to assess and enhance the cost-effectiveness of interventions, lack of technical rigor in selection/approval of proposals, low capacity of implementers, and inadequate technical guidance and support to implementing agencies during subproject implementation. 6.6 Efficacy of the objective to reduce the impact of AIDS on those already infected and affected was also modest. The availability of home- and community- based services for the care and support of PLWHA and orphans and vulnerable children increased over the life of GARFUND. However, field visits and interviews indicate that the quality, effectiveness and sustainability of these services suffer from the absence of established links with the formal health system. These services do not receive referral and technical support and guidance, which is essential to the quality of care, and which clearly falls under the mandate of the MoH. Likewise, available evidence suggests that the additional care and support for OVC financed by GARFUND was lacking in quality. Civil society organizations delivering services to orphans and PLWHA were unanimous in expressing their concern about the sustainability of these services. 6.7 The efficiency with which project objectives were achieved is negligible. There was a poor fit of project indicators and design with the epidemiology of HIV/AIDS and the sociology and economics of behaviors particular to Ghana and its regions and subregions, which were reflected in Ghana's own strategy. As a consequence, there was insufficient targeting and prioritization of activities around: high risk groups, high-risk regions, high-risk behaviors, and places of high-risk activity or high vulnerability, as a complement to Ghana's effort to undertake prevention activities that would benefit the general population. Much effort and expenditure were devoted to IEC activities, even though (a) awareness was already universal at the project's outset (para. 1.4) and (b) Ghana's strategy advocated a shift away from IEC and towards behavior change interventions. More attention to targeted behavior change would have significantly enhanced the potential impact of subprojects. These were neglected, despite a call for increased attention on the occasion of the MTR. 6.8 At the time of project design MoH was responsible for leading the fight against HIV/AIDS in Ghana. This effort was multi-sectoral and decentralized to the regional level, with promising pilots ongoing at the district level (DRI). A review of HIV/AIDS efforts at that time suggested that MoH could be more effective with increases in its staffing and financing and by raising the NACP sufficiently in the hierarchy. The failure to build on existing efforts, experience, and initiatives created inefficiencies in implementation, as: (a) a new institution and new staff, not involved in national strategy formulation or project preparation, took up their responsibilities at the time of project launch; and (b) the MoH, which had led and continues to play a critical role, was alienated. This was especially problematic since the Technical Director position in the GAC Secretariat was never filled.86 The push to expand 86The Borrower in its comments (Annex H-2) reports that it did undertake efforts to fill this position, but these efforts were unsuccessful. 30 rapidly the national response seems to have come at the expense of quality and impact of the 3000+ subprojects, some of which were carried out by inexperienced civil society and public sector organizations, a number of which were not completed, and few (if any) of which were evaluated for results. GARFUND's practice to approve and finance subprojects proposed by new NGOs/CBOs at each round and not to provide more financing of NGOs/CBOs that had received funding in previous rounds may have been effective in expanding the number and range of implementers of HIV/AIDS activities in the country. However, it was a very inefficient approach for achieving sustainable results and building capacity. The failure to utilize and support GHANET in its mandate to coordinate the work of over 100 NGOs in carrying out HIV/AIDS activities was a missed opportunity both to utilize existing capacity and to expand capacity of civil society.87 6.9 Risks to Development Outcome. The risk that development outcomes will not maintained or realized is significant. The technical resilience (and efficacy) of efforts is likely to suffer unless the Technical Director position in the GAC Secretariat is filled and coordination/collaboration with MoH improves. Continual financing of HIV/AIDS prevention and mitigation activities by thousands of civil society organizations is unlikely to be sustained unless external financing is secured for the long-run. Ownership of and commitment to HIV/AIDS activities by public sector agencies, and their willingness to provide counterpart funding for HIV/AIDS will continue to be fragile unless agencies are selectively supported and trained and develop a full understanding of the dynamics and impacts of the epidemic as they relate to their particular sector and their role and comparative advantage to contribute to the fight. Governance issues, raised in GAC meetings and in many public arenas, including the press, will continue to undermine development outcome unless and until mechanisms are refined for ensuring accountability for results and proper use of financial resources. 6.10 The Bank's performance during preparation was moderately unsatisfactory. Insufficient attention was paid to the analysis and integration of country-specific evidence, experience, lessons, ongoing efforts and initiatives, institutional arrangements and capacity, and outlook.88 Furthermore, the strategic relevance and approach of the project were undermined both by the decision to eliminate MoH from the GARFUND design and by the failure to underpin the importance of mapping and effectively covering key target populations. In essence, project design focused on fiduciary, institutional and implementation aspects and failed to analyze or develop sufficiently the epidemiological and strategic foundations of program and project design. Weak project design was also reflected in a logframe which did not establish links between inputs, outputs, outcomes and impact and which was deficient in its choice of indicators and in the establishment of baseline data on key indicators. Even in the context of a rapid preparation process, the Bank could have (a) integrated more 87The Borrower, in its comments (Annex H-1), notes that GHANET was supported by other sources to establish its secretariat, but the fact remains that its capacity for networking outreach was not utilized under GARFUND. 88Both officials and technical staff involved in GARFUND preparation and development partners noted that the Bank's team did not listen to their opinions, concerns and advice on the Bank's proposed support and on ideas and opportunities to coordinate and collaborate. 31 fully into the design the technical and economic aspects; and (b) reflected more the experience and perspectives of Government and development partners alike. 6.11 The Bank's performance during supervision was unsatisfactory. While supervision inputs were adequate, the Bank's supervision missions were devoted primarily to implementation issues, ensuring that the review, approval and financing of subprojects were carried out expeditiously so as to accelerate both the mobilization of actors and the disbursement of funds for the implementation of subprojects. Development partners raised concern about what they considered to be the Bank's almost exclusive focus on rapid disbursement and their unavailability to engage in a meaningful dialogue with them, especially to listen to their concerns about the development effectiveness of HIV/AIDS efforts in Ghana. They noted that the mid- term review was a missed opportunity to correct certain issues of project design and focus that were raised, but not effectively addressed. They also observed that the Implementation Completion Report (ICR) mission marked the first time during implementation that the Bank was sufficiently disciplined in its focus on the project development objectives. These observations are corroborated by a review of internal Bank reports, which reported GARFUND's success on the basis of trends in HIV prevalence, even though this is a meaningless indicator with respect to either objective and, which paid particular attention to the monitoring of fiduciary requirements, implementation issues and process. Technical/public health and quality aspects were not sufficiently reviewed or supported. Field-based actors and implementers interviewed at decentralized levels expressed their wish that the Bank missions would have engaged in more substantive dialogue with them and given them more technical feedback and advice during their visits on how to improve the development effectiveness of their efforts. On the other hand, the GAC Secretariat expressed strong appreciation of the Bank's performance during supervision. 6.12 Performance issues were raised on a number of occasions during implementation, but not rigorously followed up on. The interim review of GARFUND, conducted in 2004 by ACTafrica89 as a part of a broader interim review of MAP investments to date, raised salient issues about performance and outcome.90 Bank sector management comments on supervision reports also highlighted areas of concern.91 However, the thrust of reviews were, by and large, positive and focused primarily on implementation issues and much less on results. The drive to mobilize actors and disburse funds was accorded top priority. 6.13 Borrower Performance was moderately unsatisfactory overall. After initial delays, due in part to a change in government and in part to the establishment of the 89 AIDS Campaign Team for Africa (ACTafrica) was established in 1999 in the Africa Region to provide resources and technical support to country teams to mainstream HIV/AIDS activities in all sectors and, the following year, to support the implementation of MAPs. 90 Among the issues/concerns raised in this interim review were: NSF not being evidence-based, exclusion of MoH from GARFUND, inadequate coordination among partners, low capacity/lack of accountability of CSOs, weak performance on knowledge management and capacity building, weak M&E, and need for stronger technical support provided by the Bank. 91 The lead specialist, on behalf of the sector manager, did raise a number of areas of concern for follow-up action over a series of ISR, including the importance of : carrying out second generation surveillance, a focus on development objectives in spite of pressure for fast implementation, stronger linkages with MoH, stronger partnerships with donors, GAC transcending its PIU role to take on program coordination, collecting data on key indicators, a results-based MTR, moving away from awareness raising to behavior change. 32 GAC, the performance of the GAC Secretariat in its capacity as project implementation unit (PIU) for the GARFUND was moderately satisfactory. It fulfilled the basic functions of a PIU, especially those on which the Bank placed great emphasis: implementation of a new, innovative project in the space of in three and one half years; accelerated disbursements; the engagement of a broad range of civil society actors for a multi-sectoral response; and the satisfaction of fiduciary and reporting requirements specified in the development credit agreement. However, the GAC Secretariat's performance was weaker with regard to the technical (public health) and quality aspects of GARFUND. Its failure to monitor and evaluate project performance as initially envisaged undermined both its learning-by-doing approach and the accountability for results of the broad range of stakeholders supported under the project. 6.14 The performance of the GAC (including its Secretariat) in fulfillment of its mandate to coordinate and manage a multi-sectoral response was moderately unsatisfactory. First, it was noted by a number of development partners that during the life of GARFUND the GAC Secretariat devoted itself almost exclusively to the role of PIU for the Bank and, as a consequence, neglected some of its core responsibilities to enable and facilitate the work of the GAC, which oversees the entire national HIV/AIDS program. Second, the failure to fill the position of Technical Director of the GAC Secretariat (despite GAC efforts to this end) undermined the ability of the GAC to strategically manage the technical content and quality of HIV/AIDS program efforts. The GAC Secretariat has indicated that this void was filled in part by their efforts to solicit the technical support of MoH. However, (a) there were tensions between MoH and GAC that persisted throughout the life of GARFUND that affected both the quality and quantity of MoH support; (b) the interface between GAC and MoH was weak at national, regional and district levels; and (c) as a consequence, GARFUND subprojects continued to suffer from poor technical quality and inadequate public health content. Third, program monitoring and evaluation arrangements were inadequately implemented (e.g., baseline mapping and assessment of coverage of high-risk groups). GAC's failure to ensure the refinement and implementation of its draft M&E plan for 2001-2006, as well as to utilize data for tracking program performance and strategic decision- making, constitutes a significant shortcoming in Government's performance. Development partners noted that their dialogue with the GAC emphasized the importance of harmonized and disciplined M&E for improved strategic direction and management, but without much success. Finally, the failure of (most) non-health public sector agencies financed under GARFUND to budget adequate counterpart financing for their HIV/AIDS plans, for which they received GARFUND financing (World Bank internal reporting), indicates that commitment and conviction within the Borrower's public sector agencies are still in need of improvement. 6.15 For reasons presented in Chapter 4, the design and implementation of arrangements for GARFUND monitoring and evaluation and for the utilization of data produced for decision-making were negligible. 33 7. Lessons and Challenges Lessons 7.1 The drive to accelerate disbursements and to spread the financing across a wide range of implementers, activities and geographic areas poses the risk that these process goals could take precedence over the quality of interventions, as well as the strategic selection and prioritization of the highest-impact activities. Ghana's HIV epidemic remains largely concentrated in high-risk groups; it is essential to reduce the highest-risk behavior to prevent the spread to the general population. However, simply putting the high-priority interventions on the menu of eligible activities was not sufficient to ensure that they would get done, or that adequate coverage of the highest-risk groups and areas would be ensured. 7.2 A more strategic choice and sustained use of implementers with the experience and capacity to carry out the highest impact interventions likely would have resulted in stronger and more sustainable results. GARFUND supported over 70 public sector agencies and over 3,000 civil society organizations. The prioritization and more in-depth support of fewer, high-impact sectors would likely have enhanced considerably GARFUND's contribution to national prevention objectives and, as a consequence, its impact on the epidemic. Moreover, the practice of awarding GARFUND financing to new agencies and organizations with each new call for proposals, rather than providing sustained support to experienced ones, is likely to have mitigated GARFUND's potential impact. 7.3 Efforts to strengthen a multi-sectoral response would likely have been more successful had synergies been created with the MoH, capitalizing on technical and operational expertise. GARFUND's experience has shown that limiting health sector's role primarily to the clinical and medical, and failure to build these synergies risks compromising the quality and effectiveness of the prevention and care services being undertaken by non-health sector implementers. A project design, which would have fully engaged the MoH's technical capacity to ensure quality in key areas at national, regional and district levels, might have facilitated the fulfillment of the GAC's key role of coordination, which was found by partners to be neglected. This issue was raised during the MTR, but not adequately addressed. 7.4 "Learning-by-doing" has the potential to improve effectiveness, but it is unlikely to take place in the absence of strong preparation and incentives for systematic monitoring and evaluation. In GARFUND this approach was to be supported by a strong M&E component to track program performance and outcome, and operational research to document and refine the effectiveness of interventions in the various local contexts in Ghana, but these were never well defined during preparation. Incentives for GAC to disburse quickly and for implementers to manage and account for financial resources were much stronger than incentives to learn or to be accountable for the effectiveness of interventions and for results. There were no incentives for evaluation and improved performance in the project design. 34 Box 4. Role of MoH in the Context of the GAC The GAC, when it was established in 2002,assumed numerous responsibilities that had previously been assigned to NACP/MoH, notably: coordination of the national response; consensus building; policy development; technical support to other stakeholders; resource mobilization and allocation; research; donor coordination; and monitoring and evaluation.92 NACP/MoH retained responsibility for implementation of HIV/AIDS activities falling within a very narrowly defined mandate for MoH: the more clinical aspects of prevention, voluntary counseling and testing, and treatment and care of PLWHA.93 Among the 47 members of the Commission, two are from MoH: the Minister of Health and the Program Manager of the NACP. The MoH is also represented on the Steering Committee of the GAC and on the "Prevention Care" and "Care and Support" Committees. A Technical Services Division, created within the GAC Secretariat was assigned responsibilities for which MoH had the expertise, experience and the tradition to execute: "...technical guidance, supervision and coordination of technical services, including prevention and control of HIV/AIDS, clinical care and support of people living with HIV/AIDS." IEC, also traditionally a responsibility of MoH, became the responsibility of the GAC Secretariat's Policy, Planning, Research, Monitoring and Evaluation Division (p. 36 of the Act). This institutional set-up, for which GARFUND financed the salaries and most operating costs, translated into the suboptimal involvement of the lead technical ministry in the new multi- sectoral framework. Source: Act 613 of Parliament of January 2002, creating GAC. Challenges 7.5 A follow-on HIV/AIDS operation, the Multi-Sectoral HIV/AIDS Project (M- SHAP), was approved on November 2005 and became effective on March 15, 2006. Co-financed by IDA (US$20 million), DFID (US$8.3 million) and GoG (US$7.0 million), this project supports the development objectives of NSF II (2006-2011) to: reduce new infection among vulnerable groups and the general population; mitigate the impact of the epidemic; and promote healthy life-styles. 7.6 The design of M-SHAP reviewed and attempted to incorporate key lessons of experience gained from GARFUND design and implementation,94which point to the importance of: according priority to high-risk groups and high-risk areas; assisting the GAC to assume its role of coordination, facilitation and oversight; routine site visits to ensure accountability of subprojects and other decentralized activities for proper use of resources and for delivering on contract commitments; stronger, more collaborative linkages between GAC and MoH;95 strengthened partnerships (among development partners and between development partners and GAC) in favor of a program approach; and further strengthening of M&E systems. The Bank's efforts to compile these lessons and to address them under the M-SHAP hold promise for improving the development effectiveness of the Bank's support to HIV/AIDS in Ghana. 7.7 The findings and lessons from this PPAR indicate additional opportunities to improve the effectiveness and efficiency of the Bank's support to national HIV/AIDS efforts. These include: being more strategic in the choice and support of public sector and civil society actors; strengthening the technical capacity of the institutional 92 Act 613 of Parliament of January 2002, creating GAC 93 The 2001-2005 NSF describes MoH's role as "...facilitating the development of programmes on institutional care for PLWHA, STD management, blood products and PMTCT." 94 World Bank PAD, October 18, 2005. 95 A Memorandum of Understanding that clarifies the respective roles of GAC and MoH was signed as a condition of effectiveness. 35 framework by drawing on health sector expertise and experience for policy-making and for technical oversight and support; the strengthening of M&E, operational research and incentives to underpin a learning-by-doing approach and to ensure accountability of all implementers for the delivery of results. The following components of sound M&E cannot be overemphasized: (a) including targets and indicators for systematic tracking of coverage and trends among high-risk groups, as well as the general population; (b) defining and tracking accountabilities for results; and (c) designing a research agenda that will address key programmatic issues and underpin efforts to enhance the cost-effectiveness of interventions specific to the multiple contexts of Ghana. 37 References Adjei, Sam, D. Attipoe, E. Addai, M. Bleboo, B. Heloo, L. Agbe, P. Wondergem, and B. Garshong. 2000. "Ghana HIV/AIDS Response Analysis." Ghana: Health Research Unit of the Ghana Health Service. Adu-Oppong, Ahmed. 2005. Social and behavioral determinants of consistent condom use among female commercial sex workers (FCSWs) in Ghana. Publication Number AAT 3180787. ETD collection for Houston Academy of Medicine-Texas Medical Center. Adomako Ampofo, Akosua. 1999. "Nice guys, condoms, and other forms of STDs protection: sex workers and AIDS protection in West Africa." In Becker, Charles Dozon J.-P., C. Obbo and M. Touré. (eds.). Vivre et Penser le Sida en Afrique / Experiencing and Understanding AIDS in Africa. Paris: Codesria, IRD, Karthala, PNLS. Akwara, Priscilla A., G. B. Fosu, P. Govindasamy, S. Alayón, and A. Hyslop. 2005. "An In- Depth Analysis of HIV Prevalence in Ghana. Further Analysis of Demographic and Health Surveys Data." Calverton, Maryland: OCR Macro. Amoa, Sakyi A. 2005a. Facing the Challenge. A Serious Dialogue About HIV/AIDS. Ghana: Ghana AIDS Commission, and Step Publishers (step@ghana.com). Amoa, Sakyi A. 2005b. Facing the Deadly Virus. Urgent Talks About HIV/AIDS. Ghana: Ghana AIDS Commission, and Step Publishers (step@ghana.com). Amoa, Sakyi A. 2005c. Facing the Threat. A Critical Discourse on HIV/AIDS. Ghana: Ghana AIDS Commission, and Step Publishers (step@ghana.com). Amoa, Sakyi A. 2006. "Progress Report of the Ghana AIDS Commission for the First & Second Quarter of the Year (January-June 2006) in Coordination of the National Response." Presented to the Second Business Meeting of the Partnership Forum on 28 June 2006. Ghana. Ernst & Young. 2003. "IDA Credit 3458-GH (Volume II) Audit of 75 NGOs and CBOs for the Year ended 31 December 2002." Report to the Ghana AIDS Commission on the Utilisation of Ghana AIDS Response Fund (GARFUND). Family Health International (FHI) with Research International Ghana and the NACP, Ghana Health Services, Ghana. 2002. "Behavioral Surveillance Survey, Ghana 2002, Round II and Rounds I and II Comparison Report." Accra, Ghana. Family Health International (FHI). 2005. "The Final Report for the Implementing AIDS Prevention and Care (IMPACT) Project in Ghana. September 1998 to June 2004." www.fhi.org Ghana AIDS Commission. 2001. "Ghana HIV/AIDS Strategic Framework 2001-2005." Accra, Ghana. Ghana AIDS Commission. 2002. "The National Monitoring and Evaluation Plan for HIV/AIDS in Ghana 2001-2005." Accra, Ghana. Ghana AIDS Commission. 2003a. "Minutes of the Third General Meeting of the Ghana AIDS Commission Held at the Speaker's Conference Room, Parliament House, Accra on 9 July 2002." Ghana. Ghana AIDS Commission. 2003b. "Study of the Status of Home-Based Care, PLWHA Associations and Support Networks in Ghana." repared by: HMC Ltd., Tema; JSA Consultants Ltd., La Accra; and Pro-Link Org., East Airport Accra. Ghana. 38 Ghana AIDS Commission. 2004a. "Guidelines for the Utilisation of Ghana AIDS Response Fund (GARFUND) and Sub Project Proposal Format." January. Accra, Ghana. Ghana AIDS Commission. 2004b. "Ghana HIV/AIDS Response Fund (GARFUND) Operational Manual." May. Accra, Ghana. Ghana AIDS Commission. 2004. "The National Monitoring and Evaluation Plan 2001-2005 for HIV/AIDS in Ghana. 2004 Progress Report. Ghana. Ghana AIDS Commission. 2005a. "The National Monitoring and Evaluation Plan 2001- 2005 for HIV/AIDS in Ghana: 22004 Progress Report." May. Accra, Ghana. Ghana AIDS Commission. 2005b. National HIV/AIDS Strategic Framework II 2006-2010. Ghana. Ghana AIDS Commission. 2005c. "National Integrated IEC/BCC Strategic Framework (HIV and AIDS) 2006-2010." October. Accra, Ghana. Ghana AIDS Commission. 2005d. "National Report on the Follow-up to the United Nations General Assembly Special Session (UNGASS) Declaration of Commitment on HIV/AIDS. Reporting Period: 2004-2005." Ghana. Ghana AIDS Commission. 2006a. The National Response to HIV and AIDS. Five Year Programme of Work. Ghana. Ghana AIDS Commission. 2006b. "GARFUND Implementation Completion Report." Submitted to: Implementation Support/Implementation Completion (ICR) Mission. March. Accra, Ghana. Ghana AIDS Commission (GAC), Joint United Nations Program on AIDS (UNAIDS), Mexican Health Foundation (FUNSALUD)/Regional AIDS Initiative for Latin America and the Caribbean (SIDALAC). 2004. "National HIV/AIDS Accounts, Ghana, 2002-2003: Level and Flow of Resources and Expenditures to Confront AIDS." Accra, Ghana. Ghana Health Service. 2004. Annual Report of the National AIDS Control Programme. Ghana. Ghana Poverty Reduction Strategy (GPRS). 2003. An Agenda for Growth and Prosperity. Volume I: Analysis and Policy Statement. Ghana. Ghana Statistical Service (GSS) and Macro International Inc. (MI). 1999. Ghana Demographic and Health Survey 1998. Calverton, Maryland: GSS and MI. Ghana Statistical Service (GSS), Noguchi Memorial Institute for Medical Research (NMIMR), and ORC Macro. 2004. Ghana Demographic and Health Survey 2003. Calverton, Maryland: GSS, NMIMR, and ORC Macro. Ghana Statistical Services, Ministry of Health, USAID, UNICEF, 2007. "Ghana Multiple Indicator Cluster Survey 2006," Preliminary Report. February. Governments of Denmark and Ghana. 2005. "Agreement regarding Support to the National Strategic Framework II for HIV/AIDS in Ghana, 2006 Programme of Work. Accra, Ghana. Government of Ghana. 2005. "Guidelines for Utilization of District Assemblies' Common Fund." In collaboration with the Ministry of Finance and in accordance with the decision of Cabinet. Ghana. IEG­World Bank (Independent Evaluation Group­World Bank). 2003. India National AIDS Control Project. Project Performance Assessment Report. Washington, DC: World Bank. 39 IEG-World Bank (Independent Evaluation Group-World Bank). 2005. Committing to Results: Improving the Effectiveness of HIV/AIDS Assistance, An OED Evaluation of World Bank's Assistance for HIV/AIDS Control. Washington, D.C. Jamison, Dean T., W. H. Mosley, A. R. Measham, J.L. Bobadilla. (eds.) 1993. Disease Control Priorities in Developing Countries. Washington, DC: World Bank. KPMG Auditors. 2006. "Ghana AIDS Response Fund of Ghana AIDS Commission." Management Report for the Year Ended 31 December 2005. Accra, Ghana. Kwame Nkrumah University of Science and Technology (KNUST). Oxidative Stress in HIV Infection: The Role of Antiretroviral Therapy and Antioxidant Supplementation as Combined Therapy in the Management of People Living with HIV/AIDS (PLWHAs). Section A. Department of Molecular Medicine, School of Medical Sciences, KNUST, Kumasi, Ghana. Martin, Gayle H., and D. Z. Logan. 2005. Facilitative Studies to Inform the Study of the Social and Economic Impact of HIV/AIDS in Ghana. MEASURE Project and Ghana AIDS Commission. 2003. AIDS in Africa During the Nineties: Ghana. A Review and Analysis of Survey and Research Results. Carolina Population Center, University of North Carolina at Chapel Hill. Mullen, Patrick. 2005. "Review of National HIV/AIDS Strategies for Countries Participating in the World Bank's Africa Multi-Country AIDS Program (MAP)." Background Paper for the IEG Evaluation of the World Bank's Assistance for HIV/AIDS Control. Washington, DC: World Bank. NACP. 2005. "HIV Sentinel Survey. 2004 Report." National AIDS/STI Control Programme Ghana Health Service. Ghana. OED (Operations Evaluation Department) Renamed Independent Evaluation Group in December 2005. Please see IEG for publications. Outlook. 2004. "An Update from the AIDS Prevention Initiative in Nigeria." Nigeria AIDS. Boston, MA. Phillips, James F., A. A. Bawah, F. N. Binka. 2005. "Accelerating Reproductive and Child Health Program Development: The Navrongo Initiative in Ghana." Working Paper no. 208. New York: The Population Council, Inc. Putzel, James. 2003. "Institutionalising an Emergency Response: HIV/AIDS and Governance in Uganda and Senegal." HIV/AIDS and Governance in Uganda and Senegal. Report submitted to the Department for International Development. Rutenberg, C. Baek, S. Kalibala, and J. Rosen. 2003. "Evaluation of United Nations- Supported Pilot Projects for the Prevention of Mother-to-Child Transmission of HIV." Overview of Findings. HIV/AIDS Working Paper. New York: United Nations Children's Fund (UNICEF). Special Report. 2006. "An Unhappy Anniversary for AIDS. Twenty-five years of AIDS." The Economist, June 3, 24-25. Tuakli-Ghartey. J. M. 2004. "Manya Krobo Queen Mothers' Foster Care Program for AIDS Orphans." A Review Funded by UNICEF and Ghana AIDS Commission. Ghana. UNAIDS. 1999. "Joint Country Mission Report." International Partnership against AIDS in Africa. Ghana 12 ­ 16 October. UNAIDS/WHO. 2004. "Epidemiological Fact Sheets on HIV/AIDS and Sexually Transmitted Infections -- 2004 Update." 40 World Bank. 1989. Ghana: Population Health and Nutrition Sector Review. World Bank. 1997. Confronting AIDS: Public Priorities in a Global Epidemic. New York: Oxford University Press. World Bank. 2000a. Intensifying Action against HIV/AIDS in Africa: Responding to a Development Crisis. Africa Region, Washington, D.C. World Bank. 2000b. AIDS Response Project (GARFUND). Project Appraisal Document. Human Development III, Africa Regional Office. Report no. 21458-GH. December. Washington, DC: World Bank World Bank, 2000c. Memorandum of the President of the International Development Association to the Executive Directors on a Country Asssistance Strategy of the World Bank Group for the Republic of Ghana. Country Department IV, Africa Region. June. Washington, D.C. World Bank. 2003. Health Sector Support Project Implementation Completion Report. Human Development 2, Africa Region. Report No. 25543 World Bank. 2005a. "HIV/AIDS Program. A Guidance Note on Disbursement Procedures." Global HIV/AIDS Program (HDNGA). November. Washington, DC: World Bank. World Bank. 2005b. 2005. Multi-Sectoral HIV/AIDS Project (M-SHAP). Project Appraisal Document. Human Development II, Africa Regional Office. Report no. 33917-GH. Washington, DC: World Bank. World Bank. 2006. AIDS Response Project (GARFUND). Implementation Completion Report. AFTH2. Report no. 36119. 41 Annex A. Basic Data Sheet GHANA AIDS RESPONSE PROJECT (GARFUND) (CREDIT 3458) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project cost (US$ million) 27.8 26.6 96% Loan amount (SDR million) 19.6 18.1 92% Cancellation (SDR million) 1.5 Project Dates Original Actual Board approval 12/12/2000 12/28/2000 Signing 11/23/2001 Effectiveness 05/08/2002 05/08/2002 Closing date 06/30/2005 12/31/2005 Staff Inputs (staff weeks) Actual/Latest Estimate No Staff weeks US$US$(`000) Identification/Preparation 0.34 784.20 Appraisal/ Negotiations 21.90 51,982.28 Supervision 62.02 146,098.71 ICR 8.23 17,826.19 Total 92.49 216,691.38 Mission Data Date No. Specializations represented Performance Rating (month/year) of persons Implementation Development Progress Objective Identification 11/27/2000 5 Task Team Leader (1); Sr. Counsel S S /Preparation (1); Counsel (1); Lead Financial Management Specialist (1); Health Specialist (1); Procurement Specialist (1); Financial Management Specialist (1) Appraisal/ 10/22/2000 10 Task Team Leader (2); Health S S Negotiation Specialist (1); Consultant (2); Financial Supervision 07/07/2001 2 Task Team Leader (1) Health S S Specialist (1) 03/01/2002 7 TTL (!); Health Specialist (1); Social S S Protection (1); Financial Management (1); Procurement (1); Communications (1); Education (1) 06/07/2002 5 Task Team Leader (1); Health S S Specialist (1); Education Specialist (1); Implementation (1); HIV/AIDS Strat. UNAIDS (1) 11/08/2002 7 Task Team Leader (1); Health S S Specialist (1); Senior Operations Officer (1); Financial Management (1); Procurement (1); Team Assistant (1); M&E (1) 42 Date No. Specializations represented Performance Rating (month/year) of persons Implementation Development Progress Objective 03/14/2003 6 Task Team Leader (1); Procurement S S Specialist (1); Operations Officer (1); Implementation Specialist (1); Financial Management (1); HIV/AIDS Programs (1) 10/17/2003 7 Task Team Leader (1); Operations S S Officer (1); Education Specialist (1); Procurement Specialist (1); Financial Management Specialist (1); M&E Specialist (1); IEC Specialist(1) 04/20/2004 7 Task Team Leader (1); Healthy/HIV S S (1); Financial Management (1); Procurement (1); Education (1); M&E Specialist (1); Implementation Consultant (1) 11/11/2004 6 Task Team Leader (1); Operations S S Officer (1); Lead Health Specialist on HIV/AIDS (1); Financial Management Specialists (1) Procurement Specialist (1); Educ. Specialist (1) 04/16/2005 6 Task Team Leader (1)l Lead Health S S Specialist (1); Operations Officer (1); Operations Analyst (1); Financial Management Specialist (1); Program Assistant (1) 12/14/2005 7 Task Team Leader (1); Lead Health S S Specialist on HIV/AIDS (1); Operations Officer (1); Sr. Health Economist (1); Sr. Procurement Specialist (1); Financial Management Specialist (1); Program Assistant (1) ICR 03/07/2006 6 Task Team Leader (1); Lead Health Specialist (1); Procurement Specialist (1); Senior Financial Management Specialist (1); Consultant (1); Program Assistant (1) 43 Annex B. Persons and Organizations Consulted GHANA, Accra Government of Ghana Ghana AIDS Commission Professor Sakyi A. Amoa, Director General Mr. Emmanuel Larbi, Monitoring and Evaluation Coordinator Dr. Sylvia J. Anie, Policy Planning Research, Monitoring and Evaluation Mrs. Vera Ouaye, Capacity Building Coordinator Mr. Anthony Boateng, Accounts Manager Mr. Eric Pwadura, Communications Manager Mr. Kyeremeh Atuahene, Research Coordinator Mr. Abu Fuseini, Procurement Manager Mr. Maxwell Addo, Director, Finance Ministry of Health Dr. Edward Addai, Director, Policy Planning, Monitoring and Evaluation, Ministry of Health Mrs, Joyceline Azeez, Head, Procurement & Supplies Ghana Health Service Dr. Frank Bonsu, Disease Control Department Mr. Armah, Head, Nutrition Department Mrs. Hanna Adjei, Nutrition Department Mrs. Rosana Agble, Former Head, Nutrition Department Mr. Daniel Darko, Chief Biostatistics Officer, Head Centre for Health Information System Mr. Emmanuel T. Tidakbi, AG Dir. Health Administration & Support Services Dr. Nii Addo, Head, NACP Mr. Sam Boateng, GMS Procurement & Supplies Ministry of Tourism and Diaspora Relations Mrs. Bridget J. Katsriku, Chief Director Bilateral and Multilateral Development Partners Mr. Jacobsen, Embassy of Denmark Mrs. Helen K. Dzikunu, Senior Programme Advisor, DANIDA Dr. Koma S. Jehu-Appiah, Embassy of the Kingdom of the Netherlands Matilda Owusu-Ansah, HIV/AIDS Adviser, DFID Peter Wondergen, HIV/AIDS Specialist, USAID Ute Moehring, Programme Officer, Governance and Civil Society, Delegation of European Union Commission in Ghana Dr. Morkor Newman, National Professional Officer HIV/AIDS, WHO Dr. Rosalinda Herrondel, Country Officer, HIV/AIDS, WHO Dr. Warren Naamara, UNAIDS Country Coordinator, UNAIDS Taari Erkkoka, M & E Adviser, UNAIDS Ms. Liv Elden, UNICEF Greater Accra Region Tema Municipal Assembly Mr. Chris Azumah, Deputy Director Tema Municipal AIDS Committee Mrs. Lilian Baker, HIV/AIDS Focal Person (New) Assam de graft, Social Welfare Officer Ms. Esther Sintim, CBO Representative Mathias Ble, National Youth Coordinator Glastone Attipo, President, TASA 44 Upper West Region Wa Regional AIDS Committee Yacubu David, Director Regional Coordination. (Representing Reg. Minister on mission to Accra) Alhaji Tamimu Zaidu, Regional Focal Person on HV/AIDS Jamila Issa, Women Affairs Mark Abughaba, National Population Council Abdul Samed, Information & Public Campaign Official Ahmed Yacub Sidiki, Jamiat Al-Hadiyat Nadowli District AIDS Committee Helen Tanye, District Focal Person Eric Dankurah, Wing Commander, GAF (Retired) Upper East Region Bolgatanga Regional AIDS Committee Mr. M. Billey, Deputy Director Regional Coordinating Council Mr. Samuel Angyodem, RFP Mr. Yusif Akuduyu, Regional House of Chiefs Ms. Lydia Charity Domalane, Ghana Education Services Mr. Anthony A. Awiah, Department of Social Welfare Mr. Alosibah Azam, Population Office Mr. Victor Nti-Berkom, NAP+Ghana Regional Chairman Mr. Gregory Derg, AAIG (Program Officer) Mr. Akwasi Amankwaa, Regional Information Officer Kassena Nankana District (KND) Assembly, Navrongo Sr. Augusta T. Azupia, FASTRADS Anderson Pwadura, FASTRADS Mary K. Atiye, GHS KND Emelia Talata Yaro, Information Service Collins Ohene Gyan, District Planning Officer Amadu Abubakari, District Budget Officer Comfort Agasasa, Gender Officer Augustine G. Ayirezany, Ghana Education Service Basugu Mumuni, Ghana Police Service Peter Anyawie, MOFA Fangaje Robert B., Community Development Officer Rafina Asuru, Ghana Health service Gilbert Nuuriteg, District Assembly Dr. Osafroadu Amankwah, District Assembly Northern Region Tamale Regional AIDS Committee Sulemana B Saaka, Regional Coordination Committee, Regional Focal Person Alhassan I. Amadu, Natioan Population Council S. S. Mahama, Ghana Red Cross Anaba Nabila K., Action AID-Ghana J. L. Ndego, NCWD J.J. Babinah, NRHC Tolon District AIDS Committee Ekow White, District AIDS Focal Person Metan Toudo, AMR Zion Church Ashanti Region Kumasi Metropolitan Authority 45 Regional AIDS Committee Mrs. Felicia Dapaah, Regional Focal person on HIV/AIDS Mr. B. K. Gwasi Kumasi Metropolitan AIDS Committee Mrs. Modesta Bokuma, Metropolitan AIDS Focal Person Mr. Yaw Otchere Baffour Mr. Eric Yeboah-Ntiamoah Mr. Issa Uttman Dango Mr. Nti Berko Edward Mr. W. K. Yeboah Mr. Anthony Agemang Mrs. Christina D. Dural Kyes Mr. Enoch Andoh Mr. Enoch Andoh Mrs. Helen Asante Mr. Syvester Gerden Mr. Umer Faruuk Mr. Blevi Richard Mr. H.O. Amankwah Dr. Joseph Oduro Mr. Samuel Kissi Mr. Frank Duff Memo Mrs. Agnes Aua Opoku Mrs. Agnes Opoku, Africa hope Foundation of Ghana Mr. Gabral O. Mansah Frank Duff Memorial Kwame Nkrumah University of Science and Technology Kumasi Dr. Francis A. Yeboah, Senior Lecturer & Consultant Chemical Pathologist Kumasi Metropolitan Authority NGOs/CBOs/FBOs Agnes Afua Opoku, Africa Hope Gabriel O. Mensah, Frank Duff Memorial Elder Sackey Ebenezer, Kwadaso Estate Youth Association Fredrick Asamaah, Providence Centre for Humanity International Rebecca Anopong, Young Women's Christian Association (YMCA) Anthony Akwasi Amoaten, Christian Social Action Movement Lawford B. Acheamfuor, Centre for family Care & Development Asafo Adjei Twum, Faith for Ghana Ministry Dan Wantungo, Social Action for development Rev. Kofi Boateng, African Christain Homes & Rehabilitation Center Esther Mensah, Healthcare Christian Fellowship (HCF) Ghana Stephen Osei Taleyi, African Christian Home & Rehabilitation Center Justice Bashir Yakubu, Youth Network Centre Lewis Billy Bonsu, Teens Net Foundation Elliot Yin Samunah, Life Center of Ghana Alhaji Ali Salifu, Friends Assistance Global Spread of HIV/AIDS, street Children & Crime Olivia Oppong Fosu, Grassroots Initiatives programme World Bank Washington D.C. Alexandre Abrantes, Former Sector Manager for the GARFUND Sandra Rosenhouse, Task team Leader, GARFUND Preparation Jonathan Brown, Operations Adviser Alex Preker, Lead Specialist David Peters, Senior Public Health Specialist (HDNHE) 46 World Bank Country Office Accra, Ghana Mats Karlsson, Country Director Laura Rose, Task Team Leader MSHAP Evelyn Awittor, Senior Operations Officer Fred Yankey, Senior Financial Management Specialist World Bank Office Beirut, Brussels by e-mail and by telephone Eileen Murray, Former Task Manager for GARFUND Implementation Francois Decaillet, Lead Health Sector Specialist, Former Task Manager for Ghana Health World Bank Office Dakar, Senegal Dr. Aissatou Diack, Senior Public Health Specialist/Team Leader for ICR Other interviews by phone or E-Mail UNDP New York by Telephone Dr. Joe Annan, Former Head of JSA Consultants Ltd. Accra FHI Arlington by telephone Dr. Kweku Yeboah, Former NACP Head UK Scotland by telephone Dr. Sam Adjei, Former Deputy Director at Ministry of Health Akwamu Traditional Council by telephone on August 23, 2006 Mr. Moro Seidu Coordinator 47 Annex C. Supplemental Background and Contextual Information Figure C.1 Reported cases of AIDS by year 1986 - 2002 Reported AIDS cases by Year 1986 - 2002 12000 10528 10476 10000 8000 7752 r 6289 6000 Numbe 4854 4000 3833 3295 2442 2699 2334 2578 2000 2010 2371 2330 42 112 646 0 1986 1987 1988 1989* 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year No. Source: NACP. Annual Report, National AIDS Control Programme, 2003 p. 16 Figure C.2. Reported AIDS by Region 2002. Figure 3: Reported AIDS Cases By Region 2002 3500 3261 3000 2500 2406 2000 asesC of.oN 1419 Series1 1500 1000 890 669 471 442 414 500 267 237 0 ti lta st Ashan stern afo ral Vo erAccra We Ea Cent rthern r East No Western per Great Brong Ah Uppe Up Region Source: Figure from NACP Report for 2003 p. 18. 48 Box C.1. Ghana's HIV/AIDS Response as of 2000 Promotion of Safer Sex through ABC strategy · Information: mass media campaigns (print and electronic) since 1986; talk shows, films and theatre; school curriculum programmes; peer education (youth, commercial sex workers, work place), with involvement of: Ministries of Health, Communications, Education; NGOs, FBOs civil society groups; private for-profit sector · Condom distribution: in clinical settings through MoH, private sector distribution (commercial and social marketing) and peer distribution Clinical Response · Preventing and managing STDs: Strengthening STD Management within the Ghana Health Service (GHS) and in the private sector; : strengthening STD management in basic health care service; integration of STD management into family planning services; and training of nurse tutors; targeting high-risk groups (e.g., specialized clinics providing preventive and curative services for CSW with involvement of police) · Preventing HIV Transmission through Blood Transfusion and : supplying safe blood for transfusion; programs to minimize blood transfusion · Reducing other iatrogenic transmissions: a number of policies and protocols in place for improved infection control in the health care setting and in other traditional practices (traditional surgical procedures, barbers and hairdressers' practices) · Reducing mother-to-child transmission: No official policy; nothing being done. · Institutional care of PLWHA: virtually non-existant (protocol exists, but fear among health workers); many PLWHAs isolated or sent away. · Voluntary counseling and testing: not available in Ghana, but several NGO proposals under development for its introduction Community and social response · Community and social response: most AIDS cases are cared for at home, but home-based care is the least developed component of the continuum of care within the HIV/AIDS program, weak linkage with health services and no training/supervision; community-level support of PLWHA and families and intensive home- based care is carried out under a number of initiatives and pilots ; in 2000 about 5 functional associations of PLWHA · District response initiative: a strategy to decentralize the national HIV/AIDS response and to improve its management at the district level. since late 90s, adopted DRI to move towards a more integrated, multisectoral and development-oriented approach through a partnership between Ministry of Employment and Social Welfare, 10 administrative districts and UNAIDS/WHO, including support of management capacity building and activities implementation by communities, local NGOs and institutions. Ashanti region first poised for full implementation. Legal and human rights response: · There are no HIV/AIDS specific laws to protect the rights of PLWHA. HIV/AIDS policy provides guidance on rights (e.g., informed consent for testing, counseling and confidentiality of status), but there are problems with its implementation. IEC on human rights and destigmatization have been limited. No coordination/capacity building of major stakeholders: Ministry of Employment and social welfare, Chiefs, Police, CHRAJ, Judiciary, MoH, District assemblies, NGOs. Management and Coordination: · @ national level: NACP mandate, a technical working group chaired by deputy Minister of Health, including Ministries of Ed, social welfare, youth and sports, communications, donor agencies, platform for exchange, HIV/AIDS Thematic Group under UNAIDS, CHAG coordinates all mission hospitals and clinics; MoE Task force for AIDS prevention, MESW for DRI; GHANET coordinating over 100 NGOs. · @ regional level: NACP has part time regional coordinators; few regional AIDS Advisory Committees are functional · @ district level: DRI so new that results are yet to be known by promising. Monitoring, Evaluation and Research: · M&E weak with some exceptions: Red Cross peer education programme; PPAG activities, national HIV epidemiological surveillance; too little effort spent on M&E; too few instances where programme effectiveness was critically reviewed. · Research: undertaken but influence not clear; absence of a central coordination mechanism for research and no coherent research agenda exist Source: "Ghana HIV/AIDS Response Analysis, "June 2000. 49 Box C.2. The Way Forward Promotion of safer sex: · Refine promotion of safer sex to be better targeted and more effective, more oriented around behavior change (versus IEC) · Increase availability and access to and targeting of quality condoms by building on and expanding various, non-clinical distribution mechanisms; studies on condom utilization to monitor change in safer sex behavior Clinical Response · Improve the quality, targeting, coverage and monitoring of STD services · Further reduce transmission in health clinic and other settings (updating guidelines, monitor performance, improving quality assurance at decentralized levels, training, outreach) · Develop guidelines for MTCT and counseling and testing services for HIV+ women, pregnant and lactating mothers. Ensure perinatal anti-viral therapy · Develop VCT in urban and major centers/high-prevalence areas. · Upgrade guidelines, train staff, develop professional code of conduct, publicise exemptions policy · Support/expand PLWHA associations. Community and district response · Link home based care to health services through referral, supervision, training · Strengthen community involvement/ownership and provide financial and material support · Accelerate and expand DRI and strengthen NACP visibility/technical support, which is critical to success Legal and human rights response · Urgent review of the existing human rights laws; develop laws, policies, programs, plans; sensitize; modify laws that encourage harassment of sex workers, inhibit sex-worker interventions; campaigns to reduce discrimination, and stigma; actions to fight against discrimination: employment, treatment, representation, legal system/network. . Management/Coordination & M&E · Study 3 options to improve overall national coordination: second other actors to NACP and raise its hierarchy in MoH; a supra-ministerial body representing different sectors of society in office of VP, with NACP as secretariat; group of cabinet Ministers ensures coordination of activities with strong management bodies w/in each sector. · A number of HIV/AIDS coordination mechanisms do exist and multisectoral coordination needs to be investigated and improved at the district, regional and national levels. · Accelerate and simplify and expand the DRI · All parties need to keep track of process and results, for which national guidelines, including a selection of indicators, needs to be developed. Capacities for M&E, better mechanisms for reporting · High political commitment must be sustained and permeated down to all pol and adm levels in the country. Source: "Ghana HIV/AIDS Response Analysis," June 2000. 50 Table C.1 World Bank Lending for Health and HIV/AIDS at the Time of Project Design Project Approval Closing Objectives HIV/AIDS Date Health and Education 01/86 12/31/91 To provide emergency No specific support provided. Rehabilitation Project rehabilitation and a fast (Cr. 1653-GH) injection of supplies to meet urgent needs of the health and education systems Second Health and 12/13/90 12/31/97 To improve the quality and No specific support provided. Population Project (Cr. coverage of health services 2193-GH) and increase the availability and accessibility of family planning services, with an emphasis on primary health care in the three under- served northern regions Health Sector Support 10/21/97 06/30/02 To: improve health status This support covered MOH's NACP AIDS Project (Cr. 2994-GH) through enhanced quality, activities (management of sexually transmitted access and efficiency of infections (STI), clinical care for persons living (Sector-Wide basic health services; with HIV/AIDS (PLWHA), a TB program, a Approach [SWAp] increase health financing; program to prevent mother to child HIV investment in Ghana's reduce population growth; transmission (PMTCT), voluntary counseling and five-year health reduce malnutrition; testing for HIV (VCT) and IEC for public program improve access to water and awareness and targeting high risk groups and sanitation; and reduce promoting condom use). Evaluation of this project1 poverty. pointed to the fact that the magnitude of HIV/AIDS epidemic was underestimated, and its importance within the PoW was underemphasized at the project's outset. HIV/AIDS program performance was assessed to be unsatisfactory and a call for greater focus on HIV/AIDS was made for the follow-on PoW. Table C.2. Development Partners' Support to HIV/AIDS at the Time of Project Design Development Areas of Support Partner USAID IEC, STD management training, MTCT, VCT, condom distribution, NGO support, advocacy, workplace programs, surveillance CIDA Support to CSW, IEC, STD management, support for PLWHA GTZ Technical assistance to integrate HIV/AIDS into district primary health care DANIDA Support to PLWHA DFID Reproductive health and (planned) capacity building of non-health sector ministries EU Support for NGOs, STD management UNAIDS Advocacy, resource mobilization, training of traditional healers and physicians, planning UNICEF IEC, peer education to adolescents, support to CSW UNDP Home-based care UNFPA Training, support for NGOs, peer education, IEC, advocacy WHO Surveillance, training on STDs management and counseling, IEC, home-based care, support to NACP in planning, policy Source: World Bank PAD, December 8, 2000 1 World Bank, Implementation Completion Report, June 4, 2003. 51 Annex D. GARFUND: Planned vs. Actual Support by Component Planned (source: SAR and DCA) Implemented? Comments 1. Prevention and Care Services Communications campaign to inform potential executing entities about the objectives of Garfund, how to access funds and activities eligible for financing Prevention (awareness raising, behavior change, reduction in stigma) Activities to be carried out by different implementers through contracts of one year or less Process indicators (PAD) % of funds disbursed under Prevention and Care Component going to No Potential of contracting GHANET for this purpose was underexploited. existing affiliations (umbrellas) that are able to manage smaller civil society organizations to provide services Baseline: n.a. Target: 30 percent Actual: 1 percent % of funds awarded to CBO submitted projects (through windows B Yes and C) of all subproject funding Baseline: n.a. Target: 20 percent Actual: 23 percent % of subprojects that are meeting their stated objectives in a satisfactory No GAC reports 91% for first call (although not based on systematic evaluations); other calls reported manner to be unsatisfactory. Baseline: n.a. Target: 10% annual IEG found that subproject reporting did not focus on achievement of objectives, but rather increase implementation progress and financial statements (disbursements and expenditures). Actual: mixed Window A: proposals from line ministries for activities Very modestly 70 subprojects were financed, of which 54 (72 percent) were completed by GARFUND closing. targeted at: (GAC) No evidence of prioritization of ministries/interventions in line with key strategic documents: Joint Partnership Mission on HIV/AIDS, 1999; Analysis of Ghana's Response, 2001; and National Strategic Framework 2001-2006. Demand for GARFUND assistance was very sluggish (MAP Interim Evaluation). · Internal clients GAC tallies of MDA subprojects do not indicate whether focus is on internal or external clients. · External clients Top three most cited activities in list of 70 MDA contracts were: advocacy (79 percent), capacity building (60 percent), and IEC (43 percent). Menu of Group A activities: Advocacy to increase political support and mobilize the community Most MDA subprojects were geared around advocacy and awareness raising activities, but, as Awareness raising: Target messages to captive homogeneous audiences noted above, implementation was very modest, in part due to agencies' failure to budget for their such as schools, unions, other organizations, with large membership, on counterparts. promotion of save practices and to dispel misconceptions about HIV/AIDS Policy development No evidence of policy development emanating from these activities. Preparation of guidelines and manuals related to confronting AIDS in Other than manual prepared by GAC, no evidence of sector-specific guidelines and manuals on their sector HIV/AIDS. Sensitization and training of ministerial staff centrally and training of Substantially But impact/use of training not assessed. trainers at district level on HIV/AIDS Preparation of strategic plans at the district level to combat AIDS All districts have HIV/AIDS plans. Technical assistance required to provide that service more effectively Details not available. 52 Planned (source: SAR and DCA) Implemented? Comments WindowB: proposals from civil society organizations, Very modestly 760 subprojects were financed, of which 535 (70 percent) were completed by GARFUND closing. (GAC) including: · NGOs · CBOs · Trade and professional associations · Associations of PLWHAs · districts · affiliation networks of such groups GHANET, an established network of over 100 NGOs working on HIV/AIDS, did not receive substantial support from GARFUND, even though it was documented to have significant experience and potential for guiding and building the capacity of NGOs for improved effectiveness. Window C: activities conducted by very small CBOs Substantially, with 2174 subprojects were financed, all of which were completed by GARFUND closing. (GAC) and associations, based on letter of intent caveats on content Policy of financing different, new NGOs and CBOs every round undermined the sustainability, effectiveness and credibility of subproject implementers. Menu of Group B and Group C activities: Disturbing picture of multiple subprojects within the same district financing the same activity (e.g., multiple peer education projects in Builsa (8), Dormaa (18) and Akuapen North (31). Lends some credence to information provided by DACs on NGOs/CBOs'practice of changing cover page of other proposals to apply for funding. At best there is considerable scope for overlap due to lack of coordination and oversight. Prevention Quality of interventions was compromised by the facts that (a) IEC messages were not vetted by a knowledgeable authority such as MoH for its technical accuracy and appropriateness; (b) they were not pre-tested for finetuning and improved effectiveness; and (b) prevention efforts were not based on operational research that would have facilitated a better understanding of the sociology of risks and behaviors and more effective crafting and targeting of messages. Some billboards on roadside during IEG mission conveyed messages that would enable persistent and widespread fear and stigma (e.g., "AIDS kills.") Advocacy to increase political support and mobilize the community Awareness creation and peer education dominated subprojects funded, but no quantitative data on Awareness raising: Local campaigns to raise awareness and outputs/outcomes are available. understanding about HIV/AIDS, its mode of transmission and its consequences Interventions targeting high-risk groups to encourage low-risk behavior Information about coverage of high risk groups was unavailable, as there were no data on the total and voluntary counseling and testing of both HIV and STDs. High risk populations of these groups (denominator) and limited data on number of persons covered. There groups include: is risk that there might have been overlaps in target groups of different subprojects. Furthermore, · commercial sex workers and their clients reviews of contracts and interviews indicated that focus on risky groups, and risky behaviors were · truck drivers very modest. · prison population · the army · youth Targeted messages to captive homogeneous audiences such as schools, Youth in schools constituted an important target group for Ghana's HIV/AIDS program, but these unions and other organizations with large membership on promotion of activities were financed in large part by DFID. PPAR field visits did encounter one CBO financed safe practices and to dispel misconceptions about HIV/AIDS by Ghana that supported HIV/AIDS interventions in schools. Peer education in the community, schools, workplace, etc. to promote Many, if not most subprojects, included a range of activities, encompassing advocacy, peer safe practices, encourage voluntary counseling and testing for both HIV education, reduction of stigma, and care. and STDS, and to encourage community-based care of PLWHAs Condom distribution combined with information about safe practices Very modest Neither GAC, regions, districts, nor NGOs/CBOs met could quantify condoms purchased and distributed with GARFUND support. While other donors were providing significant support to this effort, including social marketing of condoms, GARFUND supported very modest and ad hoc efforts. NGOs/CBOs funded for condom distribution activities: 10 of 98 subprojects under 1st call (2002); 22 of 49 subprojects under 2nd call (2002). Only 9 of 188 MDA subprojects distributed condoms. (GAC tallies) 53 Planned (source: SAR and DCA) Implemented? Comments Care Legal advice and information for protection of PLWHA No evidence of Thirteen associations of PLWHA in 10 districts were established and funded under the project, but implementation no quantitative or qualitative data was provided on type of support provided, outputs and Income generation activities for PLWHAs and their families outcomes. (Thirty-two associations with 2,289 members were supported under the Global Fund.) Support included food, medication, and counseling. There was no defined package of services, rather each subproject did what it thought was needed. While some support aimed at building the confidence and capacity of PLWHA to live productive lives and to earn incomes, others supported substantial monthly handouts, payments for attending meetings and fees for transport fares to meeting locations. This latter type of support caused widespread public perception of GARFUND assistance to be generous handouts, leading to public comments that one had to contract HIV to get easy money from government (field interviews). PLWHA believed GARFUND money to be their entitlement and some sent errand boys/girls to collect their shares. Activities to prepare legislation to protect PLWHAs No evidence of Reports on work by Legal and Ethics Committee of GAC not made available to IEG. implementation Community care for orphans Substantially No compilation/aggregation of the type (nutrition, school fees, uniforms, housing, other), coverage implemented, with (total number of orphans supported under GARFUND divided by total estimate of orphans) and caveats costs of the various components of community care. Neither is there information on total support/coverage for orphans (financed by other sources). Study on Queen Mothers' Foster Care Program for AIDS Orphans found that this program was not based on a critical assessment of existing community-managed OVC support activities. No formal selection criteria for foster parents were used and the absence of any disabled or handicapped children among the OVC in the program suggested discrimination and exclusion. Cases of OVC being subcontracted by queen mothers to other foster parents raised questions about risk of child labor and exploitation as well as the potential for fraud. The study also found that health and nutritional needs were not addressed and a high percentage of OVCs were made to do hard labor to supplement household income. This study put into question the Bank's assertion that this program should be considered best practice. Home-based care for the PLWHA, including nursing care and support Substantially Home-based care was supported through numerous NGO/CBO interventions, but type of support, in daily routine activities. implemented, with quality, coverage, cost and outputs were not quantified. Neither was information available on total caveats coverage (all financing included) and links with health services. Elimination of MoH from GARFUND undermined environment and opportunities for establishing referral between health facilities and communities and technical support. Prioritization/targeting: · first cycle: greater Accra and Kumasi, and the Eastern Substantially This prioritization made good sense and was followed in the first cycle of subproject region, areas of the highest incidence of HIV/AIDS. proposals/support. · high-risk groups and vulnerable groups: Some local actors exhibited a reluctance to work with marginalized groups (IEG field visits) and in fact the large majority of projects were targeted to the general population and youth. Opportunities to work with high-risk groups were significantly underexploited. o preparation of careful mapping of the location of Not implemented these groups when Garfund is established o assessment of coverage attained prior to each Not implemented subsequent funding cycle to define funding priorities for the following call for proposals. · Garfund in charge of the definition of broad priorities for No evidence of Priority setting was to emanate from mapping and coverage exercises for high risk groups, which action. implementation. never occurred. Window D: activities conducted by private sector Very modestly 21 subprojects were financed, 14 (67 percent) of which were completed by GARFUND closing. entities (added as a separate window during implemented. implementation). 54 Planned (source: SAR and DCA) Implemented? Comments 2. Strengthening Public/Private Institutions for HIV/AIDS Control and Care Giving To be subcontracted to NGOs and line ministries with expertise on the GAC estimated that No evidence of efforts to screen/rate the expertise and comparative advantages of line ministries issue in question. some 20,000 and NGOs with the potential to serve as subcontractors for capacity building. individuals benefited from capacity building Underutilization of GHANET (network of 100+ NGOs working on HIV/AIDS) for activities under assessing/building capacity of civil society organizations. GARFUND, but training designed and Capacity building activities (indicated in this section) were not based on capacity assessments of delivered was not the various entities slated to be strengthened, nor is there evidence that any of these activities were based on capacity evaluated. There is no complete itemization of capacity building activities supported under assessments and was GARFUND, and no quantification of inputs and outputs by type of agency. DFID through the never evaluated. SIPAA project provides significant parallel financing of capacity building initiatives for HIV/AIDS activity design, management and implementation. The relative contributions and complementarities of GARFUND and DFID's support have not been assessed. Internal Bank reporting indicates that a significant amount of training through a subcontract to ActionAID-2000 for NGOs, CBOs and line ministries was carried out with the help of DFID-SIPAA funding. The policy of financing new, different NGOs/CBOs every round to spread GARFUND financing more widely and to extend the national response actually undermined the capacity and credibility of the better established, more capable NGOs and CBOs. Failure to exploit the capacity, experience and mandate of GHANET (coordinating network of NGOs working on HIV/AIDS) by contracting this entity to further build capacity of its 100+ members was a missed opportunity. Capacity building is included in most MDA, NGO and CBO subprojects as a discreet intervention (among many others). But there is no compilation of the nature of capacity building interventions, or of their outputs and outcomes. Over and above inadequate counterpart funds, MDA capacity was also undermined by limited motivation and dynamism of some ministerial focal points, who have multiple responsibilities and are not dedicated full-time to HIV/AIDS. Technical/practical aspects of working with HIV/AIDS · Production of a technical manual on HIV/AIDS related issues (ICR) Training · Focal persons in 74 MDAs (and other MDA staff) trained to prepare and implement Technical assistance HIV/AIDS plans (ICR) · Training of training at district level by all line ministries (ICR). However, utility, application, outputs and outcomes of this training were not assessed. Community participation · Organization of leadership training for beneficiary organizations (ICR) Training Technical assistance Project management · Training of field investigators on supervision, research and analysis (ICR) Training · Training of civil society institutions in budgeting and accounting (ICR) Technical assistance · Training in proposal writing for NGOs and CBOs through workshops held in all regions (parallel financial support provided by DFID-SIPAA to this end) (ICR). Limited capacity at RAC and DAC levels: criteria for appointment of focal/M&E person not evident, capacities (in terms of skills and availability) are limited. Supervision and technical backstopping provided or commissioned by GAC were noted to be deficient by field-based informants. Feedback from beneficiaries will be obtained on a regular basis to IEG found no evidence of systematic feedback from beneficiaries on the quality of services monitor the quality of the services provided. provided. Process Indicators (PAD) % line ministries with trained trainers at district level on HIV/AIDS Achieved, with caveats GAC reports 100%, but the nature, utility and utilization of the training has not been assessed. Ministries are not proactive in managing and implementing HIV/AIDS programs within their respective mandates. # districts that have prepared and are implementing strategic plans to Achieved Baseline: n.a. confront HIV/AIDS Target: 100% 55 Planned (source: SAR and DCA) Implemented? Comments Actual: 100% 3. Knowledge Management Field visits revealed a strong, unmet demand for best practice, exchange of experience and lessons among implementers, and systematic pedagogical supervision for corrective feedback and guidance. Establishment of a mechanism/strategy to collect, organize and There is no evidence of the existence of a mechanism or strategy to collect, organize and disseminate up-to-date information on: disseminate knowledge. Process Indicator (PAD) Not achieved Bank and GAC report the documentation and dissemination of best practices on HIV/AIDS % districts which receive regularly information on best practice Target: 100% interventions and programs throughout the country, but (a) it is not clear to what extent and on examples of HIV/AIDS interventions what basis best practices are assessed and how effectively and how often they are disseminated; and (b) field interviews revealed a significant lack of information and guidance on good practices, and great demand for technical support, guidance and information. Research results Research and studies supported include: Situation Appraisal of HIV/AIDS, June 2002; Study on OVCs, June 2003; National Assessment of Home-Based Care and Research on Associations of PLWHA, June 2003. (MAP interim evaluation) These were reported to have been disseminated at a national research conference on HIV/AIDS, which took place in early 2004. (WB Internal reporting) Other studies which were undertaken and reported to be disseminated include: a study on HIV, HCV, TB and Syphilis among prisoners; a review of HIV seroprevalence among health workers; and a study on knowledge, attitudes and perception of traditional healers on HIV/AIDS. (GAC's contribution to the GARFUND ICR, March 31, 2006) However, GAC was unable to provide IEG with the major findings of these studies and indication of how they were incorporated into policy and practice. Evidence suggests that the OVC report may not have been actively disseminated. Other reports/initiatives GAC newsletter (HIV/AIDS update, 2004) Sharing experiences across subprojects with similar orientations IEG found no evidence of sharing across subprojects. Regional AIDS committees, once established, to serve as focal points RACs visited seem to lack the time and knowledge base to fill this role. RACs and DACs were ill- for dissemination placed to disseminate good practices and other knowledge to field-based implementers, as they did not have direct oversight responsibilities for NGOs/FBOs, contracted centrally by GAC. 4. Project Management, including GAC Secretariat regularly organized and supported GAC meetings, and annual joint reviews. However, partners were emphatic in their observations that GAC Secretariat was devoting most of Monitoring and Evaluation its energy and efforts serving as PIU for GARFUND, especially the prevention and care grants component, and at the expense of its official mandate of policy formulation, strategic management, and coordination and facilitation of a broad-based, multisectoral response. Absence of documentation on total costs and financing of all HIV/AIDS efforts, their implementation (inputs and outcomes) and their effectiveness (baselines, outcomes, impacts) corroborate this observation. Project management and coordination secretariat to be Substantially achieved, established and maintained w/in GAC until project but with important shortcomings (in right- completion hand column) Secretariat of the Ghana AIDS Commission Established in 2002 · Staffing: coordinator, two financial management specialists, Recruited in 2002 one procurement specialist, and at least three technical experts (DCA, schedule 4) However, Director of Technical Services position was never filled, which undermined the technical quality and capacity of the GAC. · Technical advisory services, vehicles, equipment Technical/advisory services were limited. Field visits revealed a strong unmet demand for technical support and advice. · Activities and operational costs: contract management, Substantially achieved training, monitoring of technical and financial performance of implementers, knowledge management, report preparation/submission on finance, audit and project performance Setting up structures for HIV/AIDS at the decentralized level New AIDS committees were set up with support from district and regional authorities and funded by GARFUND. 56 Planned (source: SAR and DCA) Implemented? Comments Monitoring and evaluation Institutional framework for M&E established at central, regional and district levels suffered from very weak capacity, weak incentives and did not support a results-based, learning-by-doing approach. Stronger emphasis was placed on tracking disbursement and execution rates than on performance and results. Intro of sound financial monitoring procedures and a simple, credible application process for beneficiaries (Ritchie) Joint reviews discussed M&E Framework, but it was never rendered functional and statistics on all HIV/AIDS activities (within GARFUND project, and for program as a whole) remain patchy and incoherent. System not yet in place that fleshes out and defines results chain and links between inputs, outputs, outcomes, impact. Assess progress in achieving targets by monitoring of indicators Limited data on knowledge and behaviors (DHS not enough), limited bio- or second generation surveillance of priority groups, other ehan sex workers in two cities (Ritchie) Emphasis more on disbursements, in line with signals of Bank supervision missions. Monitor implementation of activities All districts produced strategic plans, but capacity of districts varied and many did not receive strong support and supervision in plan implementation and M&E. No M&E reports readily available in regions and districts visited. Not evident to what extent district M&E focal persons verify reports of NGOs and CBOs. M&E focal persons lack the time, capacity, guidance, tools and technical support to carry out their responsibilities effectively. 57 Annex E. Subproject Data Table E.1 Summary of Criteria for Assessing Subproject Proposals Contract Type Institutional Eligibility Technical Eligibility Window A: · An account designated for the project · Work plan identifying areas where the MDA plans to develop Contracts up to US$100,000 · A designated office responsible for implementing, in order to contribute to the multi-sectoral response (MDAs) supervising, monitoring and reporting to the GARFUND · Technically and financially viable proposal secretariat · Indication of how funds will be channeled to the districts, · Matching funds/budget line where applicable Window B: · An account designated for the project · Experience in HIV/AIDS or STDs Contracts up to US$25,000 · Staff with technical capabilities required by the · Duly registered with the Registrar General (NGOs, Networks, District subproject · Availability of administrative and accounting staff Assemblies, Private Sector) · Technical viability of proposal in light of NGO capacity · Availability of office equipment · District endorsement · Office location and address · Sound financial management, as evidenced by financial statements · Proven competence in project management, as evidenced by previous program reports, particularly in the district concerned · Proposal developed through participatory processes with potential beneficiaries Window C: · Community ownership · Relevance to HIV/AIDS local needs Contracts up to US$2,500 · Bank account (CBOs) · Joint liability declaration by 5 people Source: GAC, GARFUND Operational Manual, May 2004 Figure E.1 Subproject Completion Rate as of 2005, by Implementer and Date of Approval 100 100 100 100 89 84 84 80 76 67 69 65 67 60 t 60 cen erP 40 20 0 0 0 0 0 1st call (June 2002) 2nd call (November 3rd call (June 2003) 4th call (April 2004) 2002) MDAs NGO/FBO/DA CBO Private Source: GAC, December 2006 58 Figure E.2 Distribution of Subprojects by Region, March 2006 MDAs 71 Western 208 Ashanti 557 Upper West 73 Volta 289 Upper East 96 Brong Ahafo 313 Northern 207 Central 257 Greater Accra 317 Eastern 638 Source: GAC, 2006b Figure E.3 Percent of Private Sector Subprojects Supporting HIV/AIDS Interventions VCT 11 91 Care/Support 11 82 Capacity Building 22 64 100 IEC 55 Treatment 0 9 Home Care 0 9 Condom Distribution 0 9 Peer Education 44 0 Advocacy 33 0 Counseling 11 0 0 20 40 60 80 100 Percent Other Non-Health Enterprises (n = 9) Health Care Enterprises (n = 11) 59 Table E.2. Financial Data on Subprojects, based on GAC/GARFUND Tallies for Selected Groups of Implementers, December 2005 (in millions of cedis and million of US$) Subproject Total # Total Approved Average Contract Total Disbursements Range of Contract Amounts Proportion Category Subproject Contract Amount Amount Highest Lowest submitting Contracts completion reports at project closing: 12/05 cedis US$ cedis US$ Cedis US$ As proportion of cedis US$ cedi US$ contract amounts s Window A MDAs 70 21,545 2.375 308 0.034 18,265 2.0 85% 9001 0.099 362 0.004 Conflicting data3 Window B PLWHA Groups 13 2,400 0.265 185 0.027 2,4004 0.265 100% 2505 0.028 1206 0.013 n/a NGOs/FBOs 501 91,512 10.086 183 0.020 91,512 10.086 100% 2507 0.028 178 0.002 57% Parliamentarians against HIV/AIDS 158 3,160 0.348 20 0.002 3,160 0.348 100% 20 0.002 20 0.002 43% District Assemblies n/a n/a n/a n/a 10 0.001 n/a CBOs9 2464 20 0.002 82% (ann. ceiling) Private health facilities/entities 11 2,004 0.220 182 0.002 1,932 0.214 96% 30010 0.033 3711 0.004 73% Non-health sector private enterprises 9 1,450 0.160 161 0.018 1,315 0.145 91% 30012 0.033 5013 0.006 67% Source: Calculated on the basis of end-of-project data provided by GAC/GARFUND, December 2005 1National Commission on Culture, and Ministry of Women and Children's Affairs 2National Commission for Civil Education, New Juaben Municipal Assembly 3GARFUND tally of MDA subprojects reports no project completion reports as of 12/05 4No specific data on disbursements to PLWHA associations, but assumed to be 100% disbursed 5Wisdom Association, Korle-bu Fevers Unit, Accra 6Women's AIDS Control Organization, Half-Assini, Jomoro, Western Region 7GHANET/Accra Metropolitan Assembly (counseling, advocacy) 8Trees for the Future/North Tongu District (awareness creation, condom distribution) 9Since IEG did not have access to the full list of CBO projects, summary information on CBOs is incomplete and based on information gotten from GAC while on mission 10Wenchi Methodist Hospital, Wenchi District 11Anfoega Catholic Hospital, Kpando Distsrict 12Benson Educational Games, Ltd., Accra 13Spinnet Textile & Garment Cluster 60 Table E.3 GAC Secretariat's Efforts to Recover Subproject Funds Misused or Unaccounted For: A Status Report as of February 2007 (millions of cedis) Groups of Civil Society Recipients of Total amount of Expenditures reported Expenditures Amount Remaining GARFUND Subproject Financing reimbursements by GAC Secretariat to reported to be recovered balance of requested* be validated as accounted for on the funds to be legitimate upon basis of reception of reimbursed further investigation a report** List of NGOs/CBOs whose pending 248.162 0 60.064 188.098 reimbursements (raised in previous (100%) (24%) (76%) audits) were published in the newspaper Civil society organizations that were 556.587 0 556.587 0 sanctioned by financial monitoring (100%) (100%) supervisions in the districts and requested to reimburse ineligible expenditures 158 members of Parliament 3,160.000 0 2,180.000 980.000 (100%) (69%) (31%) Ineligible expenditures/unaccounted 236.252 63.732 0 172.520 funds identified in the audit of selected (100%) (27%) (73%) accounts for 2005 Total 4,201.001 63.732 2,180.000 616.651 1,340.618 (2%) (52%) (14%) (32%) US$ equivalent (@ 9,073 cedis to the $463,022 $7,024 $240,273 $308,238 $148,000 dollar) Source: Tallies sent by GAC to the Bank on March 1, 2007. Accompanying financial documentation was not forwarded with these tallies. * It is important to note that reimbursements requested are based on audits and field reviews which aimed to cover some 20 percent of all subprojects. This means that the remaining 80 percent (over 2,400 subprojects) have not been audited or reviewed. ** GAC tallies of 159 subprojects show that expenditures were considered legitimate because a report was received. While in the field, IEG requested that samples of reports from Parliamentarians but never received any, on which basis it could have gotten information on the type of expenditures incurred, the types of activities undertaken and outputs. In the absence of this information, IEG reserves judgment on the legitimacy of these expenditures. It is important to note that at the time of IEG's mission only 43 of the 159 Parliamentarians submitted reports, already a year after project completion. The number of reports received increased 109 as of March 1, 1007, nearly two years after completion. 61 Annex F. Costs and Financing Table F.1. Planned versus Actual Use of IDA Credit by Disbursement Category (millions of SDR) Disbursement Category Initial Actual as % of Initial Allocation Actual Disbursements Allocation (1) Grants for subprojects 14.00 14.60 104 (2) Goods (a) Motor vehicles 0.07 0.06 86 (b) other 0.63 0.16 25 (3) consultants' services, training and audits 3.50 2.23 64 (4) Incremental operating costs 0.90 0.80 89 (5) Unallocated 0.50 Reconciliation of IDA special account 0.23 Total 19.60 18.08 92 Amount cancelled: 1.52 million SDRs, or 7.75% of original credit amount Source: World Bank Loan Department, September 12, 2006 62 Table F.2 Summary of all HIV/AIDS Activities Financed during GARFUND Implementation by Funding Source and Managing Agency Funding Source Amount ($) Period Purpose Funding Channel Funds Manager Global Fund 4,965,478.00 2002-2004 HIV/AIDS activities Ministry of Health Ministry of Health/Country coordinating mechanism World Bank 25,000,000.00 2001-2005 GARFUND GARFUND Ghana Aids Commission International Development Association DFID 36,000,000.00 2001-2004 AIDS-related activities Direct to implementers DFID USAID 22,519,000.00 2001-2004 Prevention,care, and support, CAS USAID information, education and communication lab services, infrastructure, etc. Royal Netherlands 3,750,000.00 2002-2005 Care and support Direct to implementers Royal Netherlands Embassy Embassy WHO 1,246,000.00 2002-2005 HIV/AIDS activities ADMIN. support Direct to implementers World Health Organization to UNAIDS and support to GARFUND UNAIDS 1,464,000.00 2002-2005 Prevention, care, and support, and Direct to implementers UNAIDS/ Ministry of Health research EU 1,879,680.00 2002-2005 Support for STIs and HIV/AIDS Ministry of Health Direct to implementers UNICEF 271,923.00 2002-2005 HIV/AIDS activities Direct to implementers UNICEF UNDP 636,000.00 2002-2005 AIDS-related activities, research and Direct to implementers UNDP support to Ghana AIDS Commission CIDA 1,500,000.00 2002-2005 Sex workers intervention project Direct to implementers Canadian International Association Agency Bill & Melinda 7,523,041.00 2002-2005 Prevention advocacy capacity dev and Country Assistance Strategy Gates (African research Youth Alliance Danish Embassy 2,217,752.00 2003 HIV/AIDS activities Ministry of Health Ministry of Health UNFPA 5,204,784.00 2002-2003 Prevention, care, and support Direct to implementers UNFPA information education and communication, research IPPF 1,534,825.00 2002-2006 HIV/AIDS activities Planned Parenthood Association of Planned Parenthood Association Ghana of Ghana Japan Trust Fund 71,802.00 2003 HIV/AIDS activities Planned Parenthood Association of Planned Parenthood Association Ghana of Ghana Japan International 101,067.00 2004-2006 HIV/AIDS activities Planned Parenthood Association of Planned Parenthood Association Cooperation Ghana of Ghana Agency JOICFP 17,325.00 2002 HIV/AIDS activities Planned Parenthood Association of Planned Parenthood Association 63 Funding Source Amount ($) Period Purpose Funding Channel Funds Manager Ghana of Ghana Government of 1,102,035.00 2001-2004 Counterpart funding for GARFUND GARFUND Ghana Aids Commission Ghana- 1(counterpart funding for Ghana AIDS Commission GARFUND) Government of - 2002-2004 HIV/AIDS activities Ministry of Health Ministry of Health Ghana-2 (counterpart funding for health sector wide approach) Government of - 2002-2004 Budgetary allocations by ministries, Ministries, departments and agencies Ministries departments and Ghana-3 departments and agencies for HIV/AIDs agencies (budgetary activities allocation by ministries departments and agencies) Government of 2,170,875.00 2002-2004 1% budgetary allocation from district District assemblies District assemblies Ghana 4 assembly common Funds Faith-based 10,000,000.00 2002-2004 HIV/AIDS activities Direct to implementers Faith-based organisation organisations Private sector 10,000,000.00 2002-2004 Workplace HIV/AIDS Activities Companies Companies corporate entities Total 139,175,587.00 64 Table F.3. HIV/AIDS Expenditures by Target Group (in millions of cedis) Target Groups 2002 2003 Differences Amount % Amount % Absolute Relative Total Expenditures 108,310.91 100 237,133.11 100 128,822.20 119 Non targeted 63,220.48 58.4 137,177.45 57.8 73,956.97 117 School children 9,613.91 8,9 14,441.76 6.1 4,827.85 50 Workers 1,271.41 1.2 17,240.34 7.3 15,968.93 1256 Commercial Sex Workers 24.00 0.0 477.00 0.2 453.00 18888 Men who have sex with men 0 0 0 0 Children in risk of vertical transmission N?A 7.30 0.0 Blood donors 550.77 0.5 549.41 0.2 (1.36) 0 Migrant workers N/A 0.0 305.00 0.1 305.00 Uniformed population 19.00 0.0 381.60 0.2 362.60 1908 Prison inmates 107.00 0.1 547.60 0.2 440.60 412 Pregnant women 7,282.10 6.7 459.89 0.2 (6,822.21) -94 Children and youth in social risk 798.05 0.7 2,399.81 1.0 1,601.76 201 Health Facilities Staff 1,002.23 0.9 22,142.32 9.3 21,140.09 2109 People living with HIV 222.00 0.2 543.80 0.2 321.80 145 People living with AIDS 24,199.96 22.3 40,459.83 17.1 16,259.87 67 Source: HIV/AIDS National Accounts, 2004 (UNAIDS, GAC) Table F.4. HIV/AIDS Expenditures by Budgetary Item (in millions of cedis) Budgetary Items 2002 2003 Differences Amount % Amount % Absolute Relative Total Expenditures 108,310.60 100 237,133.14 100 128,822.54 119 Health Personnel 2,615.70 2.4 4,051.90 1.7 1,436.20 55 Non Health Personnel 3,159.13 2.9 10,894.93 4.6 7,735.80 245 Pharmaceutical products 2,855.95 2.6 13,366.57 5.6 10,510.62 368 Medical Supplies 1,571.30 1.5 2,455.60 1.0 884.30 56 Condoms 37,545.66 34.7 22,121.94 9.3 (15,423.72) -41 Reagents and materials 15,087.66 13.9 10,515.60 4.4 (4,572.06) -30 Food 7,710.38 7.1 4,085.19 1.7 (3,625.190) -47 Other materials 8,729.68 8.1 36,448.56 15.4 27,718.88 318 Building construction and refurbishing 11.21 0.0 694.23 0.3 683.02 6093 Medical equipment 4,783.82 4.4 3,436.80 1.4 (1,347.02) -28 Non Medical equipment 284.05 0.3 2,182.46 0.9 1,898.41 668 Administration 4,525.74 4.2 11,055.55 4.7 6,529.81 144 Research and consultancy 5,397.11 5.0 31,004.63 13.1 25,607.52 474 Maintenance 0.0 100.00 0.0 100.00 Hotels and rented vehicles 4,029.44 3.7 17,822.89 7.5 13,793.45 342 Other services 5,208.92 4.8 38,873.92 16.4 33,665.00 646 Per diem 4,774.47 4.4 27,630.68 11.7 22,856.21 479 Monetary benefits 20.40 0.0 391.69 0.2 371.29 1820 Source: HIV/AIDS National Accounts, 2004 (UNAIDS, GAC) Table F.5. National HIV/AIDS Expenditures by Source (millions of cedis) Budgetary Items 2002 2003 Differences Amount % Amount % Absolute Relative Public Sources 17,484.73 16.1 79,628.79 33.6 62,144.06 355.4 Ministry of Health 7,852.45 7.2 8,247.23 3.5 394.78 5.0 Other Central Government (including 9,632.28 8.9 71,381.56 30.1 61,749.28 641.1 GARFUND) Private Sources 10,462.50 9.7 17,460.59 7.4 6,998.09 66.9 Corporations - 0.0 34.17 0.0 34.17 Non-profit Organizations 290.0 0.3 387.00 0.2 97.00 33.4 Households 10,172.50 9.4 17,039.42 7.2 6,866.92 67.5 Foreign Sources 80,363.37 74.2 140,043.76 59.1 59,680.39 74.3 Multilateral Agencies 1,273.30 1.2 36,004.99 15.2 34,731.69 2727.7 Bilateral Agencies 66,177.70 61.1 93,962.40 39.6 27,784.70 42.0 Non-profit International Organizations 12,912.37 11.9 10,076.37 4.2 (2,836.00) -22.0 Total Expenditures 108,310.60 100.0 237,133.14 100.0 128,822.54 119 Source: HIV/AIDS National Accounts, 2004 (UNAIDS, GAC) 65 Annex G. GARFUND: Project Outcomes by Objectives and Targets Objective/subobjective/indicator Baseline data Targets Actual Achievements GARFUND Contribution to Achievements Objective #1: Reduce the spread of HIV infections Reduce the spread of infections · In the general population Not possible to assess Not possible to assess · Prevalence of HIV among pregnant 3.9% in 20021 Maintain at 2.7% in 2005 (MoH/NACP, 2005) women or below 6% Not a viable measure of program/project success for prevention, as it is a function of mortality and new infections. A decline could conceivably be the result of high mortality or decreased infection. · Proxy indicator: prevalence rates among 2.3 in 20022 0.8% in 2005 (sentinel surveillance) younger women **Important to disaggregate by sentinel o 15-19 years site, as addition of new, lower-prevalence (rural) sites could cause the overall rate for this age group to go down. · In high-risk groups Not possible to assess because baseline GARFUND impact on any changes in infection and actual data on prevalence are on rates among CSW is likely to be negligible because different population sets. expenditure and effort on CSW and other high- transmission groups were minimal (Source: HIV/AIDS National Accounts for 2002-03; Field visits; and GAC reports on subprojects) · Prevalence of HIV infections among 82% Reduce to CSW in Kumasi 75% · Proxy indicator: prevalence rates among Trends in prevalence rates among CSW are difficult CSW/seaters to interpret because (a) changes are a function of both o Accra 76% in 20013 n/a new infections and mortality, on which there is no o Accra/Tema n/a 52% in 20064 information; and (b) the members of these groups are o Kumasi 45% in 20025 39% in 20066 transcient, with significant turnover likely in the course of five years. GARFUND targeted support to CSW in Kumasi and · Proxy indicator: prevalence rate among elsewhere was very modest; and there is no CSW/roamers 1Source: MoH/NACP, 2005 2Source: Sentinel Surveillance 66 Objective/subobjective/indicator Baseline data Targets Actual Achievements GARFUND Contribution to Achievements o Accra 23% in 20027 n/a compilation of the type, range, coverage or o Accra/Tema n/a 37% in 20068 outputs/outcomes of subproject interventions. These o Kumasi 15% in 20029 24% in 200610 facts, combined with the fact that a number of other interventions targeted at CSW were ongoing with other financing sources, makes it difficult to document any attribution to GARFUND of trends in HIV infections among CSW. Trends in infection/prevalence rates among No baseline No end-of-project (or program) data at GARFUND also financed a few activities targeted at other high-risk groups identified for targeting established in project's closing. other high-risk groups, identified during project during project preparation (truck drivers, 2001 preparation, including truck drivers, prison prison population, the military, and others) population, the military, and others, but (as the case for CSW), there is no compilation of the type, range, coverage or outputs/outcomes of subproject interventions. Increase knowledge No end-of-project data available to Information gleaned from field visits and project document trends among general performance reporting indicated that prevention population or other target groups. activities did increase discussion of HIV/AIDS issues and might have increased knowledge as a consequence. On the other hand, field visits and document reviews revealed little indication of any perceptible changes in knowledge among any of the high-risk groups and negligible investment by GARFUND in increasing their knowledge. · % of males/females who know they can Two data sets Trends documented between GDHS 98 and 03 cannot avoid HIV by using condoms incomparable11 be attributed to GARFUND. GARFUND became o Males 40% in 1998 75% n/a effective in mid-2002, making GDHS 03 data more of 82 % in 2003 a baseline than an end-point for the project. o Females 22% in 1998 50% n/a Knowledge levels among the general population or 50% in 2003 any other target populations were not measured at the · % of males/females who know they can Differences in 90% n/a end of the project. avoid HIV by restricting sex to one the way these Anecdotal evidence from the field indicates that there 3Source: WAPCAS 4Source: WAPCAS 5Source: WAPCAS 6Source: WAPCAS 7Source: WAPCAS 8Source: WAPCAS 9Source: WAPCAS 10Source: WAPCAS 11Source: GDHS 1998: unprompted question and GDHS 2003: prompted question 67 Objective/subobjective/indicator Baseline data Targets Actual Achievements GARFUND Contribution to Achievements partner questions were have been no significant changes in the high levels of posed make fear, stigma and discrimination associated with these data sets HIV/AIDS during the period of GARFUND incomparable12. implementation. It is not clear from GAC reporting o Males 62% in 1998 n/a what was the content and coverage of interventions 89% in 2003 aimed at reducing these negative attitudes. o Females 63% in 1998 n/a 86 % in 2003 Field discussions revealed some perceptions of increased availability of condoms, and some reported perceptions of increased use. Still others in selected regions/districts reported resistance to Reduce risky behaviors condom use because of strong social and religious · % of men and women in a union using stigmas. Annual condom sales increased from 24 condoms during last intercourse million to 30 million during the project's life (2002- o Males 15% in 1998 30% n/a 2005) but this is largely attributable to large o Females 6% in 1998 20% n/a investments in social marketing of condoms with USAID support. Project attribution is low because the number of condoms distributed and condom use was not documented, but GAC reports and field discussions indicated that GARFUND investment in condom distribution was very modest and the bulk of prevention efforts were focused more on IEC than on behavior change. · % of men and women who have 60% in 199813 80% n/a reduced the number of sexual partners in response to perceived risk · Median age at first intercourse GDHS data: 18 Trends documented between GDHS 98 GARFUND supported activities aimed at youth, along and 03 cannot be attributed to with significant DFID investments in HIV/AIDS GARFUND. GARFUND became education in schools. effective in mid-2002, making GDHS 03 data more of a baseline than an end-point for the project. o Males 20.2 in 2003 n/a o Females 18.3 in 2003 n/a Project investment/activities · % of all subprojects that develop IEC n/a 30% 45% (GAC statistics) (target exceeded). Project failed to develop quality materials that would materials designed specifically for rural Not a viable measure. IEC not properly be more appropriately targeted towards changes in 12GDHS 1998 and 2003 13Time period of behavior change not specified in questionnaire 68 Objective/subobjective/indicator Baseline data Targets Actual Achievements GARFUND Contribution to Achievements populations (in local dialects) researched, coordinated, or vetted for specific behaviors by specific populations, based on quality. BCC was a more appropriate research and vetted by MoH or another qualified intervention in the context of Ghana, with technical authority. awareness levels was as high as 98% at the time of project design.14 · % of 1st and 2nd cycle schools n/a 50% 50% for 1st cycle (target met) Project attribution not clear, given large investments providing HIV/AIDS education 100% for 2nd cycle (target exceeded) by DFID in this intervention during the life of GARFUND relative to smaller investments under GARFUND. Objective #2: Reduce the impact of AIDS on those infected and their families Care of PLWHA GARFUND supported the increase in the quantity of home/community care, but the quality and sustainability of such care is lacking. · PLWHA receiving home/community n/a Increase to Outcome not calculated because of Project has provided substantial home-based and care among PLWHA that have sought 30% absence of viable data on (a) the number community-based care services for more than 30,000 medical treatment through MoH network of PLWHA that sought medical treatment PLWHA nationwide (GAC). The quality and (the denominator); and (b) support to sustainability of this care has not been assessed, but PLWHA financed by other sources. they are questionable given that (a) links with the health system (for referral and support) were not established; and (b) care and support financed by others, including that provided within the health system, has not been quantified. · # districts that provide community- n/a Increase to 50% achieved Project attribution is strong, as many subprojects based care for PLWHA 50% supported the provision of community-based care. · % of districts where an adequate n/a 40% n/a; not achieved Due in large part to its exclusion of MoH activities, referral system between home-based inherent in the project design, the GARFUND failed and institutional care has been to establish and render functional a referral system established and is functioning between the health system and community- and home-based care. Care of orphans Quantitative target met, but serious concerns about quality, effectiveness and sustainability. 14Source: GDHS 1998 69 Objective/subobjective/indicator Baseline data Targets Actual Achievements GARFUND Contribution to Achievements · % of districts that have organized n/a 30% 35% (Source: GAC) (target exceeded), Project attribution is strong. GAC statistics reveal care for AIDS orphans that under GARFUND approximately 17,500 orphans have benefited from different kinds of support ranging from payment of school fees, provision of nutritional supplements, and medical assistance. However, there is no data on quality or effective coverage of OVCs or on the total number of orphans covered by all sources of financing. A UNICEF study on the quality and effectiveness of GARFUND-supported orphans care through a Queen Mothers initiative to provide foster care revealed serious concerns about the quality, effectiveness and sustainability of this model of orphan care and about the low capacity for supervision, quality control and financial accountability. 71 Annex H. Borrower's Comments Annex H-1 - Comments of the Ghana Aids Commission (GAC) on Draft GARFUND (Cr 3458-GH) Project Performance Assessment Report (PPAR) Submitted by Professor Awuku Amoa, Director General, GAC Secretariat General Comments a. We note a major fundamental flaw in the procedure used by the Independent Evaluation Group in collecting data for the report. The evaluators showed that field data formed the basis of most of their conclusions and therefore, the rating of GAC's performance. However, in standard practice data collected from the field should be validated in order to produce a report reflecting the true situation. Thus data obtained from the field, the development partners and other stakeholders should have been validated by checking with GAC before the draft report was prepared. This would have ensured fairness and justice to GAC, but was not done resulting in several inaccuracies in the report.1 b. Contrary to their stated methodology as on p. i, para. 2 of the Report, viz. ".To prepare PPARs, IEGWB staff examine project files and other documents, interview operational staff.........." the Evaluators never interviewed any of the project management staff of the GAC Secretariat on the issues they examined. Neither did they review nor cross-check findings from their field visits with the staff as would have been expected. Besides the introductory meeting they held with the staff on the first day of the exercise, they never met with the staff again, not even for a debriefing.2 c. The above inadequacies in methodology account for the factual inaccuracies and misrepresentations that we identify in the Report. d. It is noted that in many instances, the Evaluators attribute GAC's performance rating of unsatisfactory, as they assigned, to the absence of a Technical Director within the GAC Secretariat.3 It is unfair to indict GAC on that since there are justifiable reasons for the situation that existed (as provided on p.4 herein) and which the Evaluators could have obtained from GAC Management so as to avoid the distortions in the Report. e. Assessment is generally unfair to GAC for the fundamental reason that criteria used were unknown to GAC at the onset of the project. It is illogical to use indicators 1 IEG notes that its conclusions are based on triangulation of evidence from many sources: interviews with stakeholders (implementers, beneficiaries, officials, policymakers, and development partners), direct field observation, review of project and program documentation and research. 2 IEG notes that the evaluation team met with the head and staff of the GAC Secretariat at the outset of the mission to explain its purpose and methodology and a debriefing was held at the mission's end. Additional discussions with selected GAC Secretariat staff were held on specific issues and topics. A liaison person from the GAC Secretariat facilitated the work of the mission, contacts with staff and the provision of supplemental information from the GAC Secretariat after the team's return to Washington. Annex B lists the staff of the GAC Secretariat with whom the IEG team met. 3 IEG notes that the PPAR rates the performance of GAC as the implementing agency as moderately satisfactory (not unsatisfactory) and in its role of coordinator and manager of the national multisectoral HIV/AIDS effort as moderately unsatisfactory. 72 that were not agreed upon and incorporated in the original project design for assessment.4 f. The Evaluators admit themselves (as they state in para. 6.3 of the Report) that the project design was deficient in several respects. It should be noted that the design was largely the blueprint of the World Bank which GAC was to implement. What then is the justification for GAC's indictment whereas the project foundation itself was poorly designed? 1. Background and Context · p. 1, para 1.1, line 3: use of the word rampant is inaccurate. What is the justification? 2. Objectives and Design · p. 7, para 2.2, last sentence: `Because the ongoing Health SWAp was assumed to be providing adequate financial support to all of MoH's programs, including HIV/AIDS, GARFUND's design specifically excluded the support of HIV/AIDS activities falling within MOH's mandate.' i. MOH was funded to undertake its HIV/AIDS activities and in a number of ways. Funds were provided from the GARFUND to the Ghana Medical Association, Nurses Association, Pharmacy Association, Laboratory Technicians all within MOH for intervention activities. This is a clear indication that GARFUND did not exclude support for MOH HIV/AIDS activities. ii. GARFUND provided support to numerous quasi government and private hospitals. Activities funded included VCT, support to PLWHA and opportunistic infections.'5 · p. 10, footnote 28. Statement that: ".... two factors might have discouraged follow- on contracts; first was the general policy of GAC to mobilize as many NGOs and CBOs as possible....; second was the reference to CBO support as `seed money' implying a one-time investment to stimulate action" is incorrect.6 i. The inability of most of the NGOs and CBOs to carry out repeater projects was due to the policy that an implementing agency should complete its sub-project before 4In line with its objectives-based methodology, IEG evaluated the extent to which GARFUND achieved the objectives stated in the design document and logframe (World Bank, 2000b). It assessed performance against project indicators for which trend data were available, and assessed additional trends and information relevant to project objectives to compensate for the absence of trend data on other project indicators. 5 This is clearly documented in the design document (World Bank, 2000b). According to GAC subproject tallies, the only MoH program which received GARFUND financing under the public sector (MDA) window was the Onchocerciasis Program in the amount of US$28,000 for capacity building, advocacy and VCT. No other direct financial support to MoH agencies was provided under GARFUND. A small number of non-governmental entities did receive financing for health-related activities, and are itemized in this report. 6 The evidence base for this statement was derived from: interviews with the former Bank Task Team Leaders and with GAC staff (first factor); and the design document (World Bank 2000b) (second factor). 73 it could qualify for further funding. Most of them could not satisfy this requirement by the time subsequent calls were advertised to solicit for proposals. ii. GAC did not consider support to CBOs as "seed money". Funds provided to CBOs were based on contracts approved for sub-projects implementation within the communities. 3. Implementation and Costs · p.13, para. 3.5: Average size of sub projects The basis for determination of the figures quoted here is unclear. Upon the presumption that they are based on selected sub-projects, one would like to know how representative the samples examined for the various categories of implementers are. We question the correctness of the figures.7 Under GARFUND and as agreed upon with the World Bank project supervisors and consultants (and documented in the Operational Manual), contract sums were as follows: i. MDAs were awarded cedi equivalent of $100,000 ii. NGOs were awarded cedi equivalent of $25,000 or $20,000 dependent on the NGOs sub project budget. iii. CBOs - initially were awarded $2,500 sub-project contract but when NGOs/CBOs contracts were decentralized to the District Assemblies, each District was awarded $15,000. iv. Private sector organizations on the average were awarded $25,000 per sub project. v. PLWHA Associations: These associations sprang up after the end of GARFUND project. We wonder how they could feature prominently in the evaluators' report. · p. 13, paras. 3.6 & 3.7: GARFUND financing new, different NGOs/CBOs...... ........GARFUND's practice of financing new, different NGOs/CBOs for every round of proposals undermined the effectiveness and sustainability of capacity building efforts.' i. It is inaccurate that GARFUND always sought to finance new, different NGOs/CBOs for every round (call for proposals). The correct situation was that at each call for proposals, old beneficiaries who had completed sub-projects previously awarded and had reported and accounted on their funds could apply to implement new sub-projects. Those with uncompleted projects at the time of the next call for 7IEG notes that data were calculated on the basis of end-of-project data provided by GAC/GARFUND, December 2005. A new Table (E.2) has been added to Annex E to document calculations more fully. 74 proposals could not apply for new sub projects. New NGOs/CBOs applied and accessed the funds alongside old ones that had completed their sub-projects. It is incorrect to state that GAC policy was to mobilize as many NGOs and CBOs as possible. That was not its policy, hence not documented in the Operational Manual. No criteria were established to exempt old NGO's from applying for new funding. The District Assemblies evaluated, selected and awarded contracts to NGOs/CBOs based on their District Strategic Plans. These were repeater contracts. We wonder where the independent evaluators reviewed those sub projects. ii. p.13, para 3.7: Statement that "Training was not based on capacity assessment......" is incorrect. In 2002 through the SIPAA Project capacity assessment of implementing agencies was conducted. (The following reports on GARFUND Implementing Entities Needs Assessment are available for study: 1. A Rapid Appraisal of Implementation Status of the HIV/AIDS Sector Plans by MDAs ­ conducted by GAC, ActionAid SIPAA Project, July 2003. 2. Strategic Framework for Supporting and Building the Capacity of CSOs, NGOs, FBOs, CBOs Implementing HIV/AIDS Programmes in Ghana ­ conducted by GAC, ActionAid SIPAA Project, August 2003. · p. 14, para. 3.8: Statement that (a) `it is not clear to what extent and on what basis best practices were determined and how effectively and often they were disseminated"; is incorrect. i. There is evidence to show that best practices were shared and disseminated to groups during workshops and conferences, e.g. National HIV/AIDS Research Conference (NHARCON) and M & E conferences. ii. No matter what dissemination is done, one cannot get all respondents to indicate positive change in attitudes. · p.14, para. 3.8 Statement that (b) `field interviews reveal a strong, unmet demand for technical advice, guidance and support of relevant operational research.' i. Statement must be supported with facts. ii. Under GARFUND support for operational research could be applied for. · p. 14, para 3.10: Statement that "One crucial position, Director of Technical Services was never filled". It is acknowledged that the position was not filled but for the following reasons: i. attempt made to recruit in 2002 did not succeed in getting a suitable person for the position; none of the candidates interviewed was successful. 75 ii. consequently, and as NACP already existed as a key partner and to avoid duplication of functions, a deliberate policy decision was taken by he Commission to utilize its technical services. · p.14, para. 3.11: Statements in the paragraph are mixed up. They should be corrected as follows: The Commission met regularly thrice a year to review progress towards the implementation of Ghana's Strategic Framework. The GAC Project Review and Appraisal Committee (PRAC) (whose membership was made public and rotated once every two years) met twice a year to review and recommend to the GAC financing of subproject proposals..................... · p. 15, para. 3.14: Financial Management8 i. It is incorrect that duplication of payment and relatively large sums were transferred from sub-project account to CBO accounts. CBOs accounts are the same as sub project accounts. It is, therefore, unclear as to which sub projects transfers were made to CBO accounts as the Report states. ii. Statement that ".......some NGOs proposals were funded without a budget.....". All proposals that were submitted included budgets otherwise they would not qualify to be evaluated. The statement should, therefore, be corrected. iii. Statement that "CBOs made changes in signatories to bank accounts without GAC prior approval........" This is an outcome of an external audit. It should be acknowledged that audits normally would reveal some deficiencies in systems and procedures that need to be corrected and not a flawless situation. iv. It is acknowledged that at the time of the evaluation, funds granted to Parliamentarians had not been fully accounted for. GAC Management had made several efforts including seeking the intervention of the leadership of Parliament to get the accounting done and there has been considerable improvement in the situation as at now. It is unfortunate that the Evaluators did not enquire from GAC Management to obtain such information for their report. As at date, 105 out of 158 Parliamentarians had accounted for funds granted. · p. 16, para. 3.16: Disbursements It is inaccurate to suggest that delays were due to the steep learning curve of GAC Secretariat. Delays in releasing tranches were the result of the World Bank's requirement that reports on disbursed funds by NGOs/CBOs should be evaluated by external groups before any further releases. That is not a steep learning curve but World Banks' own engineered bottlenecks in disbursements. 8Weaknesses in financial management were cited in an audit of 75 CBOs financed under Window C (Ernst and Young, 2003). 76 4. Monitoring and Evaluation · p. 18, para. 4.4, "Roles and responsibilities for M/E were not clearly articulated in the project design document"......Incentives for evidence-based implementation (as opposed to disbursement-based) implementation were absent from GARFUND's M & E Plan. i. Statements should be clarified. ii. Roles and responsibilities for entities such as DACs, RACs, M/E focal persons etc were defined under GARFUND as in the Operational Manual and the M & E Framework 2001-2005. · page 18, 4.5: Statement that: `Development partners noted that throughout the life of GARFUND, GAC's M & E plan for the NSF was not operational and their dialogue with GAC to strengthen and operationalize it became a source of tension. Respondents were emphatic that M & E seemed more a concern of the partners outside of the project than of the Bank or GAC.' This observation is untrue and judgemental. i. Did the evaluators deem it significant to talk to other stakeholders apart from development partners? We need evidence of the so called tension which in our view is non-existent. ii. The operationalisation of the M & E plan was first raised by GAC at a meeting of the Research, Monitoring & Evaluation technical committee. GAC served as a catalyst in proactively steering the process for the development of M & E Framework 2001 ­ 2005, Annual M & E Report 2002 etc. · p. 19, para 4.7 contradicts p. 13, para 3.7 viz. `Considerable amount of training and workshops were financed....' There is a discrepancy. Statement that: `while the institutional and organizational framework and staffing for M&E envisaged under the national M&E plan were clear, implementing agencies did not have the proper training or tools for setting targets and measuring their outcomes' is invalid for the following reasons: i. The Support to the International Partnership Against AIDS in Africa (SIPAA) project between 2002 and 2005 provided support to GAC. The main target of the SIPAA project during this period were capacity building of implementing partners of GAC including District Focal Persons. ii. During the period of the project a comprehensive capacity building programme was developed and implemented in Proposal Development, Project Management, Financial Management, Counseling and Care for HIV, Peer Counseling and Advocacy, HIV Competency among others. 77 iii. Manuals developed for project management training covered issues of target setting and measuring of project outcomes among others. iv. In 2002 and 2003 a total of 406 implementing partners benefited from training on project management. These were not captured during the assessment as funding was not drawn from GARFUND. v. At the district level, a total of 327 GARFUND beneficiary organizations benefited project management training. · p. 19, para. 4.8. Statement that: Key information was not collected and thus could not be used for decision-making. More fundamentally, the incentives to collect and analyze data to improve program, project impact were weak. This is incorrect. A significant number of end of project data on the indicators was available in cases where surveys, sentinel surveillance etc had been completed. 5. Outputs and Outcomes by Objective · para 5.3, p 22: Statement that "Expenditure on high transmission groups and other groups specified in NSF and in the GARFUND design documents as high priority is extremely low at barely 1 % of total expenditures for 2003". i. The high transmission groups were part of the general population and were reached by some NGOs and WAPCAS with direct funding from CIDA.9 ii. issue of high-risk groups was not prominent in the original project design and of global focus.10 · para. 5.6, p. 23: Statement that " "Subprojects with condom promotion and distribution were small in scale and irregular........' The project design did not allow for procurement of condoms. GAC could, therefore, not promote it as should have been the case.11 · p. 25, para 5.8: paragraph is untrue. Line 27 `District level actors noted that while GAC approval of NGO subprojects was contingent on endorsement by the district, in actual fact districts were encouraged to dispatch proposals immediately to GAC for their rapid approval without reviewing and commenting on them' This is absolutely incorrect. 9These sources of financing are included in the 1 percent (National HIV/AIDS Accounts, 2005). 10 Both Ghana's National Strategic Framework for HIV/AIDS (2001-2005) and the GARFUND design document (World Bank 2000b) specified the targeting of high-risk/high-transmission groups as an important element of prevention, in addition to interventions aimed at the general population and vulnerable groups therein. 11 Included among the eligible activities for GARFUND financing are: interventions targeting high-risk groups to encourage low-risk behavior; and condom distribution (World Bank 2000b). 78 i. Proposals have always been evaluated by independent evaluators and not GAC staff. ii. District AIDS Committees reviewed and recommended proposals for funding during the 3rd and 4th calls but not the 1st and 2nd calls when GAC did that through the independent evaluators as capacity at the district level during the initial calls was totally lacking. · p. 25, para 5.9: i. District M & E Focal Persons have been given training on best practices in the production of IEC materials including bill boards. ii. The BCC strategy which was widely disseminated provides a blueprint for targeted IEC and BCC materials production. iii. Training was given to the focal persons but the turnover as a result of frequent transfers of trained Focal Persons affected the capacity of new ones to deliver. · p. 27, para 5.13: i. The observation in this paragraph is rather strange. Where in the world has "fear, stigmatization and discrimination" been completely dealt with. ii. There has been significant change in the attitudes of the general public. iii. It is unfair to make a conclusion on only one (1) reported case of fear and stigma. It is a fallacy of hasty generalisation. 6. Ratings · p. 31, para 6.4: Many of the observations in the paragraph are incorrect. i. It is not true that MOH was excluded from GARFUND because of the mandate of GAC. MOH was given finance to undertake its HIV/AIDS activities. In addition GARFUND financed MOH in a number of ways. Finances were made available to Ghana Medical Association, Nurses Association, Pharmacy Association, Laboratory Technicians all within MOH for intervention activities. ii. Statement that: "Over and above the exclusion of MOH activities from GARFUND financing, the organizational structure and mandate of GAC and the GAC Secretariat reduce the role of the MOH, curtailing its potential to provide needed technical guidance............" is ambiguous. iii. In what respect did the mandate of the GAC reduce the role of the MOH? We do not find GAC's mandate to be in conflict with the role of MOH. If MOH did not have the opportunity to provide technical guidance as the Evaluators believe it might 79 be due to MOH's own staffing and logistical problems rather than an undefined role for it under the GARFUND by GAC. iv. In one breadth the Evaluators found the establishment of the position of Technical Director within the set up of the Secretariat appropriate and even goes on incessantly to lament the fact that it was never filled and, therefore, technical guidance of implementing agencies was inadequate. At the same time they expected the MOH to have provided the ".....needed technical guidance .........". If the Technical Director was recruited and MOH was to undertake this function wouldn't that have been tantamount to duplication of functions, which was one of the unintended reasons why the position was not filled? v. Decentralization efforts were rather enhanced by GARFUND arrangement. It is not correct to say that GARFUND arrangement whereby NGOs and FBOs were centrally contracted affected their accountability to RAC's. vi. It is incorrect to state that the Secretariat devoted itself almost exclusively to the Role of PIU for the Bank. GAC was not created as project management unit (PIU) and therefore did consider itself as such. GAC was created by an Act of Parliament 613 and therefore its functions were enshrined in the Act. These were what GAC facilitated to get the national response on track. vii. It is also incorrect that Non-health public sector agencies financed under Garfund failed to budget for counterpart funds for their HIV/AIDS plans. GAC required that Ministries, Departments and Agencies (MDAs) demonstrated evidence of payment of their own counterpart contribution to Garfund sub project before funds were released to the agencies. This was the reason why most MDAs failed to access the Garfund because the payment of counterpart funding was a problem. So all those non-health public sector agencies that accessed the GARFUND indeed paid up their contributions in cash as their commitment. The evaluators got it all wrong. · p. 32, para 6.7. The true picture is not presented here. i. At the beginning of GARFUND, awareness was not universal. It is incredible for anyone to say that awareness was universal at the onset of the project. If that was the case the project would not have had an objective of increasing awareness.12 ii. Ghana should be given the credit for enormous work done to achieve universal awareness within three years of GARFUND. iii. Targeting high risk regions, high risk behaviours and places of high risk activity became an issue at the tail end of the project. · p. 32, para 6.8, last sentence. 12HIV/AIDS awareness was already universal at the start of GARFUND. In 2003, the first full year of GARFUND implementation, 98 percent of women and 99 percent of men were aware of HIV/AIDS. 80 GHANET was fully supported. GAC through its SIPAA projects provided funds to GHANET to establish its secretariat and provided enough funds to NGOs for their activities. What is the evidence that GHANET was not supported? · p. 36, Box 4: Role of MOH: i. Totally wrong information is presented on GAC and MOH.13 ii. It is incorrect to say that MOH was not given any special status on the Commission. iii. The MOH is represented on the Commission by two (2) key officials of MOH, namely the Minister of Health and the Programme Manager of NACP. iv. Both serve on the Steering Committee of GAC; the Minister is the Chairman of the Care and Support Committee and the Programme Manager serves on the Prevention & Care Committee. NACP staff also serve on the GAC Research, Monitoring and Evaluation Committee. What other special status could have been given the MOH? v. The decision even not to recruit the Technical Director after the initial effort had failed, was partly due to the recognition GAC gave to need to utilize the technical services of the NACP/MOH. vi. GAC did not take over IEC responsibility from MOH. This is totally wrong. vii. The whole content of Box 4 is misinformation and should be removed. · Comment on Overall Rating On the basis of the foregoing comments, GAC's position is that: We do not accept the rating of Unsatisfactory on the Outcome and Overall Project. We believe they are both Satisfactory as the Implementation Completion Report assessed. 13IEG notes that this information is from Act 613 of Parliament of January 2002, creating GAC. 81 ANNEX H-2 ­ Comments of the President's Office on Draft GARFUND (Cr 3458-GH) Project Performance Assessment Report Submitted by Dr. Fred Sai, Special Adviser to the President on HIV/AIDS I am yet to have a satisfactory reason as to why the assessors found it impossible to contact me during their visit to Ghana.1 A project of this type always has political, administrative and legal dimensions which should have considerable weight in comprehensive assessment. It is not clear if the assessors really appreciated the roles of the Ghana AIDS Commission and its technical committees. They obviously did not appreciate the fact that a Commission of this nature cannot undertake the work needed for approving proposals. As the person with direct political and administrative oversight, on behalf of the President, a good interview with me should have been a must for any evaluation. Although I had no part in the development of the project I played a major role in the establishment of the secretariat and the reconstitution of the Commission itself. The Commission decided at the beginning of GARFUND that the Ghana epidemic was already generalized enough for a very broad education campaign aimed first and foremost to prevent the over 96% of the population, not infected, from getting infected. Secondly to provide care, support etc. for those infected and affected. Special risk groups such as female sex workers and men having sex with men received attention from specific NGOs, and rather quietly, since the current laws of Ghana have to be respected. In fact efforts have been made to get the police to be gentler in the interpretation of the laws and for the judiciary to respect the human rights of those who come to court. Whilst these are going on, flagrant challenges will not help.2 It must also be pointed out that the GAC could not fund activities which included condom purchase and distribution because the donors were against the GAC being involved in those.3 The assessors would have been told this if only they had cared to involve the GAC staff much more closely in their work I find the approach to the assessment rather strange. How did the assessors validate the information they received from those they interviewed.4 Some of the NGO and MDA representatives have gripes and knowledge gaps which could be profound. A senior representative of a ministry, for example, came to a business meeting of the partners to ask why the ministry was not being supported by GAC. In fact the particular ministry was being supported and had reported on the very good work 1IEG notes that it was unable to meet with Dr. Sai while in Ghana due to the mission's travel to the interior and Dr. Sai's unavailability. He was reached by phone upon the mission's return to Washington, but preferred to participate by sending comments on the draft report. 2Both Ghana's National Strategic Framework (2001-2005) and the design document specified the need for more interventions aimed at high-risk, high-transmission groups, as a complement to interventions aimed at the general population and vulnerable groups therein. 3IEG notes that the GARFUND design document specifies behavior change interventions, including condom purchase and distribution, among the activities eligible for financing (World Bank 2000b). 4IEG notes that it triangulates evidence across many sources: interviews, direct field observation and review of relevant documents, and research. 82 being done to the Commission in full session. Some of the donor representatives have their "pet priorities" and I wonder how the assessors managed to separate objective views from biased opinions.5 Pressure to do more work with sex workers and men having sex with men is one such problem issue. There was an occasion when an important donor representative even queried the priority being given to youth. At times I could not help feeling that the assessors were not judging by the state of knowledge and practice at the time the project was developed but by what is going on currently. The section on incidence and tracking of the epidemic is a case in point. Nobody in this field or any other medical area would use prevalence only to track an epidemic. The sentinel survey had been well conducted in Ghana. The age range, being used by many as proxy for incidence, is showing a good trend. The DHS is scheduled to be repeated next year. Between the 2003 and the 2008 findings we will get some dependable information of nation-wide relevance. I cannot therefore see any justification for spending money and resources on other surveys at this time. I am the first to admit that the Evaluation work done by the GAC is not doing them credit. A lot of data exist but these have not been collated and worked to produce usable information. This is already being addressed. It is not clear if the assessors appreciated the intra-MOH problem of relationships between the MOH and the Ghana Health Service (GHS); nor how these relate to the decentralized administration being pursued by the Ghana Government. I am almost certain that a chat with me and the Manager of the NACP6 would have been of use in appreciating how closely the GAC and the NACP had been working together. The GAC has been trying, for the purpose of sustainability, to work through the district administration. The absence of a technical director is not the sole or even the main source of any perceived lack of collaboration with the MOH. It is unfair to give the impression that the post of technical director did not receive proper attention. The first round of interviews for the post was attended by several well qualified individuals. The one selected was receiving better pay from a donor so he turned down the offer. Practically all the others were better off in the Government service. Efforts to get a joint appointment with the GHS were not successful. In fact just this week another set of interviews have been conducted. Out of five candidates short- listed only two turned up. One was completely below standard. The one being considered for appointment now did not have the number of years post-qualification experience required but it was agreed that that individual had enough specialized experience to make up for this. Hopefully he will accept an offer. I would like to make it clear too that the project was developed before the secretariat was completely established. Practically all concerned with its development disappeared before implementation. The overall unsatisfactory rating of this project disturbs me greatly. Despite what the assessors say about awareness surely everyone knows that there is a difference between being aware of something and knowing what it means or its implications. Thanks to a strong media practically all Ghanaians heard of HIV/AIDS 5Objectives-based evaluation methodology assessed performance against targets and objectives established in Ghana's National Strategic Framework for HIV/AIDS (2001-2005) and in the project document. 6IEG notes that Dr. Addo, Head, National AIDS Control Programme, was interviewed (see Annex B). 83 almost as soon as the first cases were identified in the country. But what did they know about it? This is what the project tackled so well.7 To expect massive behavior change in a three year life of a project seems to me to be overoptimistic. Right now one of the most troublesome areas is how to get health staff working closely with HIV to make it clear by their body language and behavior that the infection is not easy to catch. Although one cannot prove that the favorable trend changes in the Ghana epidemic are attributable solely or even mainly to the work don with the GARFUND, it cannot be proved that it has made no contribution either. To me an unsatisfactory rating means the Ghana situation has changed for the better despite GARFUND. How can such a conclusion be justified? The GAC staff is submitting its considered views on specific items in the report. Fred Sai, Advisor on Reproductive Health HIV/AIDS Accra March 28, 2007 7IEG notes that trends in knowledge and behavior were not tracked during the life of the project.