Document of The World Bank Report No: ICR00001792 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-H1500) ON A GRANT IN THE AMOUNT OF SDR 13.7 MILLION (US$ 20 MILLION EQUIVALENT) TO THE GREAT LAKES INITIATIVE ON HIV/AIDS SUPPORT PROJECT (GLIA) June 20, 2011 Human Development Department Regional Integration Department Africa Region 1 Member countries include: Burundi, Democratic Republic of Congo, Kenya, Rwanda, Tanzania and Uganda. CURRENCY EQUIVALENTS (Exchange Rate Effective 00000000) Currency Unit = SDR SDR .63 = US$1 US$ 1.59 = SDR 1 FISCAL YEAR January 1- December 31 ABBREVIATIONS AND ACRONYMS AIDS Acquired Immune Deficiency Syndrome ARCAN African Regional Capacity Building Network for HIV/AIDS Prevention, Treatment and Care ARV Anti-Retroviral (Drugs) ART Anti-Retroviral Therapy BCC Behavior Change Communications BP British Petroleum BSS Behavioral Surveillance Survey CAS Country Assistance Strategy CM Council of Ministers DGA Development Grant Agreement DRC Democratic Republic of Congo EAC East African Community FMA Fiduciary Management Agent GAMET Global AIDS Monitoring and Evaluation Team (WB) GLIA Great Lakes Initiative on HIV/AIDS EC Executive Committee ES Executive Secretariat HIV Human Immune Deficiency Virus HIV/AIDS Human Immune Deficiency Virus/ Acquired Immune Deficiency Syndrome IOM International Organization for Migration IDA International Development Association IDP Internally Displaced People IGA Income Generating Activities ISR Implementation Status Report MAP Multi-Country HIV/AIDS Program for Africa M&E Monitoring and Evaluation MOU Memorandum of Understanding MTR Mid-Term Review NAC National AIDS Commission OI Opportunistic Infections OVC Orphans and Vulnerable Children PAD Project Appraisal Document PCG Partnership Consultative Group PDO Project Development Objective PLWHA People Living With HIV/AIDS PMTCT Prevention of Mother-To-Child Transmission QCBS Quality and Cost Based Selection QER Quality Enhancement Review QAG Quality Assurance Group SALT Support, Appreciate, Listen, and Transfer SGBV Sexual and Gender Based Violence STI Sexually Transmitted Infection TB Tuberculosis TOR Terms of Reference TOWA Total War Against HIV and AIDS TTL Task Team Leader UNAIDS United Nations Program on HIV/AIDS UNGASS United Nations General Assembly Special Session UNHCR United Nations High Commissioner for Refugees UNITAR United Nations Institute for Training and Research USAID United States Agency for International Development VCT Voluntary Counseling and Testing WHO World Health Organization Vice President: Obiageli Katryn Ezekwesili Country Director: Yusupha B. Crookes Sector Manager: Eva Jarawan Project Team Leader: Pamphile Kantabaze ICR Team Leader: Miriam Schneidman GREAT LAKES INITIATIVE ON HIV/AIDS SUPPORT PROJECT (GLIA) TABLE OF CONTENTS Document of The World Bank ........................................................................................... i  DATA SHEET ................................................................................................................... 1  A. Basic Information .......................................................................................................... 1  B. Key Dates ...................................................................................................................... 1  C. Ratings Summary .......................................................................................................... 1  D. Sector and Theme Codes............................................................................................... 2  E. Bank Staff ...................................................................................................................... 2  F. Results Framework Analysis ......................................................................................... 3  G. Ratings of Project Performance in ISRs ....................................................................... 7  H. Restructuring (if any) .................................................................................................... 8  I. Disbursement Profile ...................................................................................................... 8  1. Project Context, Development Objectives and Design .................................................. 9  2. Key Factors Affecting Implementation and Outcomes ............................................... 12  3. Assessment of Outcomes ............................................................................................. 16  4. Assessment of Risk to Development Outcome ............................................................ 24  5. Assessment of Bank and Borrower Performance ........................................................ 26  6. Lessons Learned........................................................................................................... 28  7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners.............. 30  Annex 1. Project Costs and Financing ............................................................................. 31  Annex 2. Outputs by Component..................................................................................... 32  Annex 3. Economic and Financial Analysis .................................................................... 51  Annex 4. Bank Lending and Implementation Support/Supervision Processes................ 52  Annex 5. Beneficiary Survey Results .............................................................................. 55  Annex 6. Stakeholder Workshop Report and Results ...................................................... 57  Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ........................ 58  Annex 8. Comments of Co financiers and Other Partners/Stakeholders ......................... 62  Annex 9. List of Supporting Documents ......................................................................... 63  MAP ................................................................................................................................. 64  AFRICA GREAT LAKES INITIATIVE ON HIV/AIDS SUPPORT PROJECT(GLIA) DATA SHEET A. Basic Information Great Lakes Initiative on HIV/AIDS Country: Africa Project Name: Support Project (GLIA) Project ID: P080413 L/C/TF Number(s): IDA-H1500 ICR Date: 6/7/2011 ICR Type: Core ICR GREAT LAKES Lending Instrument: SIL Borrower: INITIATIVE ON HIV/AIDS Original Total XDR 13.7M Disbursed Amount: XDR 13.7M Commitment: Revised Amount: XDR 13.7M Environmental Category: B Implementing Agencies: GLIA Executive Secretariat Co financiers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept 07/10/2003 Effectiveness: 08/12/2005 03/14/2006 Review: Appraisal: 10/05/2004 Restructuring(s): Mid-term Approval: 03/15/2005 09/30/2007 12/23/2008 Review: Closing: 03/31/2009 12/31/2010 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately unsatisfactory Risk to Development Outcome: Significant Bank Performance: Moderately unsatisfactory Borrower Performance: Moderately satisfactory 1 C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Moderately Moderately Quality at Entry: Government: satisfactory satisfactory Quality of Moderately Implementing Moderately Supervision: unsatisfactory Agency/Agencies: satisfactory Overall Bank Moderately Overall Borrower Moderately Performance: unsatisfactory Performance: satisfactory C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Quality at Entry Project at any time Yes NA (QEA): (Yes/No): Quality of Problem Project at any Yes Supervision (QSA): DO: Unsatisfactory time (Yes/No): 12/2009 DO rating before Moderately Closing/Inactive status: Unsatisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 40 40 Other social services 60 60 Theme Code (as % of total Bank financing) Gender 17 17 HIV/AIDS 33 33 Health system performance 17 17 Population and reproductive health 16 16 Tuberculosis 17 17 E. Bank Staff Positions At ICR At Approval Vice President: Obiageli Katryn Ezekwesili Callisto Madavo Country Director: Yusupha B. Crookes Mark D. Tomlinson Sector Manager: Eva Jarawan Keith E. Hansen 2 Richard Seifman Project Team Leader: Pamphile Kantabaze Pamphile Kantabaze ICR Team Leader/ Miriam Schneidman Primary Author F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The development objectives of the project are to facilitate: (i) establishment of HIV/AIDS prevention, care, treatment, and mitigation programs for mobile and vulnerable groups such as refugees, transport sector workers, and highly affected/infected populations in each of the GLIA Member States, and (ii) enhancement of prospects for coordinated approaches addressing HIV/AIDS prevention, care, treatment and mitigation among the GLIA Member States. Revised Project Development Objectives (as approved by original approving authority) NA (a) PDO Indicator(s) Formally Original Revised Actual Value Achieved at Completion Indicator Baseline Value Target Target or Target Years Values Values Reduction in HIV prevalence in the refugee populations and surrounding communities identified by GLIA— Indicator recognized as inappropriate and no data was collected. Increase in the Kenya Refugee 31.4% Kenya Refugee camps 63.0% number of youth (aged (Kakuma) camps (Kakuma) 15-24) in the target Surrounding 24.6% Surrounding 47.2% populations reporting communities communities NA No use of a condom Uganda Refugee 9.7% Uganda Refugee camps 18.2% during sexual (Nakivale) camps (Nakivale) intercourse with a Surrounding 18.0% Surrounding 34.4% non-regular partner communities communities Tanzania Refugee 39.0% Tanzania Refugee camps 39.0% (Lugufu/ camps (Lugufu/ Nyarugusu) Surrounding 28.0% Nyarugusu) Surrounding 35.4% communities communities 3 Increase in the Kenya Refugee 45.1% Kenya Refugee camps 35.4% number of (Kakuma) camps (Kakuma) respondents 15-24 years of age within the target population who Surrounding 18.3% Surrounding 34.2% both correctly identify communities NA No communities ways of preventing sexual transmission of HIV and who reject Uganda Refugee 32.6% Uganda Refugee camps 33.9% major misconceptions (Nakivale) camps (Nakivale) about HIV transmission or prevention Surrounding 38.5% Surrounding 45.7% communities communities Tanzania Refugee 25.7% Tanzania Refugee camps 46.9% (Lugufu/ camps (Lugufu/ Nyarugusu) Nyarugusu) Surrounding 34.1% Surrounding 59.4% communities communities Increase in social and gender conditions within the target populations in refugee and surrounding populations -Percent of women aged Kenya Refugee Kenya Refugee 15-49 who were forced (Kakuma) camps 5.2 (Kakuma) camps 0.9 No to have sex in the past NA 12months (*) Surrounding Surrounding communities 9.2 communities 2.0 Uganda Refugee Uganda Refugee (Nakivale) camps 1.6 (Nakivale) camps 1.4 Surrounding Surrounding communities 2.4 communities 0.2 Tanzania Refugee Tanzania Refugee (Lugufu/ camps 3.2 (Lugufu/ camps 2.4 Nyarugusu) Nyarugusu) Surrounding Surrounding communities communities 1.5 0.2 -Sex with a Kenya Refugee 1.3 NA No Refugee 1.0 Kenya transactional partner in (Kakuma) camps camps (Kakuma) the last 12 months among men and women Surrounding 1.6 Surrounding 1.2 aged 15-49 (*) communities communities 4 Uganda Refugee 1.1 Uganda Refugee 0.8 (Nakivale) camps (Nakivale) camps Surrounding 3.5 Surrounding 1.7 communities communities Tanzania Refugee 14.0 Tanzania Refugee 12.0 (Lugufu/ camps (Lugufu/ camps Nyarugusu) Nyarugusu) Surrounding 4.3 Surrounding 1.6 communities communities Networks have been strengthened and there is improved management and AIDS learning capacities of networks and member organizations- See intermediate outcome indicators below Increased knowledge transfer between GLIA countries and uptake of health services by mobile populations -- See intermediate outcome indicators below NA=Not Available (*) Outcome indicators from BSS which were not included in the PAD (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Component 1 Prevention 100% of sites have Refugee camps: 0 9 (100%) No 9 (100%) uninterrupted and sufficient supplies of male condoms over the previous 6 month period Surrounding: TBD 9 (100%) No 9 (100%) 100% of sites have HIV Refugee camps: 0 9 (100%) No 9 (100%) posters and billboards in appropriate language Surrounding: TBD 9 (100%) No 9 (100%) >75% have functioning peer Refugee camps: 0 7 (75%) No 9 (100%) educator programs Surrounding: TBD 7 (75%) No 9 (100%) 100% sites have access to Refugee camps: 0 9 (100%) No 9 (100%) functioning VCT services* Surrounding: TBD 9 (100%) No 9 (100%) 100% of refugee, surrounding Refugee camps: 0 9 (100%) No 9 (100%) and returnee sites have access to functioning PMTCT programs Surrounding: TBD 9 (100%) No 9 (100%) 5 Treatment 100% of health workers who Refugee camps: 0 9 (100%) No 9 (100%) treat STIs in health clinics have been trained in syndromic management a/ Surrounding: TBD 9 (100%) No 9 (100%) 100% of health workers who Refugee camps: 0 9 (100%) No 9 (100%) treat common opportunistic infections (OI) in health clinics have been trained a/ Surrounding: TBD 9 (100%) No 9 (100%) Care >75% of sites have trained Refugee camps: 0 7 (75%) No 9 (100%) community health workers who are actively providing home-based care* Surrounding: TBD 7 (75%) No 9 (100%) Component 2 For PLWHA networks Each identified network has 0 6 No 7 adopted a strategy and developed an action plan for support and strengthening * Each identified network has 0 6 No 7 carried out an annual AIDS Self Assessment* % of organizations that meet 0 90% No 100% the AIDS Self Assessment target b/ Number of formal sub 0 3 No 3 regional meetings between networks of the 6 countries to exchange information, good practices and experiences Number of management 0 2 2 training sessions per country organized for member organizations For transport sector network Each identified PLWHA network 0 6 No 7 has adopted a strategy and developed an action plan for support and strengthening * Each identified PLWHA network 0 6 No 7 has carried out an annual AIDS Self Assessment * % of organizations that meet the 0 70% No 100% AIDS Self Assessment target Number of formal sub regional 0 2 No 2 meetings between PLWHA networks for the 6 countries to exchange information, good practices and experiences 6 Number of management training 0 2 No 2 sessions per country organized for member organizations Component 3 Number of formal health 0 2 No 4 sector interaction meetings to exchange information, country experiences and effective practices between the GLIA member states* Number of protocols 0 3 No 6 harmonized and adopted Number of formal interactions 0 2 No 2 between GLIA member states to exchange information on HIV/AIDS prevention, care and treatment for refugees, IDPs, returnees, and surrounding communities* A strategy with adapted health 0 Yes No Yes services for truck drivers has been adopted through consensus during a formal meeting Number of health sites with 0 22 No 21 adapted services to improve access of transport sector workers at selected sites* Component 4 GLIA has a functioning M&E None Yes No No system where data flows to a central level, is captured, analyzed and disseminated to stakeholders* The GLIA Secretariat has 0 8 No 6 appointed all 8 core staff members c/ *Indicators included in the DGA. a/ Indicators are combined into one indicator in the DGA. b/ This indicator had a target of 75% in the DGA. c/ Indicator replaced in the DGA by one related to fiduciary management. G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 05/25/2005 Satisfactory Satisfactory 0.00 2 12/28/2005 Satisfactory Moderately Satisfactory 0.00 3 06/29/2006 Satisfactory Satisfactory 2.05 4 02/14/2007 Moderately Moderately 4.78 7 Unsatisfactory Unsatisfactory 5 07/13/2007 Moderately Satisfactory Moderately Satisfactory 6.87 6 03/03/2008 Moderately Satisfactory Moderately Satisfactory 8.24 7 12/27/2008 Moderately Satisfactory Moderately Satisfactory 10.11 Moderately 8 06/30/2009 Moderately Satisfactory 13.37 Unsatisfactory Moderately 9 12/30/2009 Moderately Satisfactory 17.86 Unsatisfactory Moderately 10 06/29/2010 Moderately Satisfactory 20.53 Unsatisfactory 11 01/11/2011 Unsatisfactory a/ Moderately Satisfactory 20.82 a/ Prior to the preparation of the ICR, the PDO was down rated to unsatisfactory since the end of project survey results were not yet available. H. Restructuring (if any) NA I. Disbursement Profile 8 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal 1. The context at appraisal was characterized as follows: (i) a difficult socio- economic regional situation stemming from conflicts, genocides, and natural disasters which fueled vulnerability and massive displacement of populations; (ii) a dire epidemiological situation which called for urgent action with the Great Lakes sub-region estimated to have a high HIV burden; and (iii) emerging experience with regional approaches that were viewed as adding value to national level investments. 2. As noted in the Project Appraisal Document (PAD) the cumulative number of refugees and displaced persons was estimated by the United Nations High Commissioner for Refugees (UNHCR) at over 6.5 million. The Great Lakes countries were estimated to have over 6 million people living with HIV/AIDS out of 25 million in Sub-Saharan Africa (UNAIDS, 2004). HIV adult (15-49) infection rates in the GLIA member countries ranged from about 4.0 to close to 9.0 percent.1 Four of the countries in the region (Kenya, DR Congo, Tanzania and Uganda) were amongst the nine highest tuberculosis (TB) burden countries in sub-Saharan Africa with up to 75 percent of TB patients in some countries co-infected. 3. In light of high levels of mobility and vulnerability the proposed regional approach was critical to providing complementary support to ongoing country initiatives to address the HIV epidemic. The Great Lakes Initiative on HIV/AIDS (GLIA), established in 1998, was piloting a number of promising cross-border interventions in the transport sector, and supporting cross fertilization of experiences among member states. There was a consensus among the countries that they needed to go beyond these modest efforts to create the critical mass of regional programs and policies that could accelerate their collective effort to stem the pandemic. Recognizing the potential mutual benefit of harmonizing policies and programs, member states came together to broaden GLIA’s institutional mandate and strengthen its capacity to effectively coordinate HIV/AIDS interventions in the context of the Bank-funded GLIA Support Project. 1.2 Original Project Development Objectives (PDO) and Key Indicators 4. The development objectives of the project were to facilitate: (i) establishment of HIV/AIDS prevention, care, treatment, and mitigation programs for mobile and vulnerable groups such as refugees, transport sector workers, and highly affected/infected populations in each of the GLIA member states; and (ii) enhancement of prospects for coordinated approaches addressing HIV/AIDS prevention, care, treatment and mitigation among the GLIA member states. 1 HIV adult infection rates were estimated as follows: Burundi (6.0%); DRC (4.2%); Kenya (6.7%); Rwanda (5.1%); Tanzania (8.8%); and Uganda (4.1%), December 2003 data provided in 2004, UNAIDS. 9 5. The key performance indicators for the GLIA supported areas were set out in the PAD as follows: (i) a reduction in HIV prevalence in the refugee and surrounding populations; (ii) a reduction in the number of infected infants born to HIV infected mothers; (iii) an increase in the number of youth (aged 15-24) reporting condom use during sexual intercourse with a non-regular partner (adapted UNGASS indicator), 2 (iv) improved gender sensitivity in the target population; and (v) an increase in the number of youth respondents (aged 15-24) who correctly identify ways of preventing HIV transmission and who reject major misconceptions about HIV transmission or prevention (adapted UNGASS indicator). 1.3 Revised PDO and Key Indicators, and reasons/justification N/A 1.4 Main Beneficiaries 6. The primary beneficiaries of the project were: (i) vulnerable groups, including refugees, returnees, internally displaced people (IDP), residents of the surrounding communities, including youth (15-24); (ii) mobile groups, such as long distance transport workers as well as the groups and communities with whom they interact; and (iii) people living with HIV/AIDS (PLWHA) and their network organizations. 1.5 Original Components 7. Component l: HIV/AIDS support to refugees, affected areas surrounding the refugee communities, internally displaced people, returnees (US$8.0 million). This component aimed to provide services to a limited number of targeted populations, and would include the full range of prevention, care, treatment and mitigation. Program content would be determined by what is learned from refugee programs and by the need to ensure synergies with national programs. UNHCR would be responsible for scaling up HIV activities in refugee camps (US$5 million), and the national AIDS Commissions would coordinate activities in the surrounding communities (US$3.0 million) with each country selecting two priority catchment areas. 8. Component 2: Support to HIV/AIDS related networks (US$3.0 million). The second component aimed to enhance “AIDS Competence” of long-haul transportation workers and PLWHA by developing: (i) management capacity of the networks and lead member organizations in planning, financial management, resource mobilization, and M&E; and (ii) HIV/AIDS learning capacity and transfer of knowledge between network members. Transport networking would focus on two principal transmission corridors: (i) Mombasa-Nairobi-Kampala-Kigali-Bujumbura-Bukavu-Goma, and (ii) Dar es Salaam- Dodoma-Kigali-Bujumbura-Bukavu-Goma, covering both the truckers and the communities and groups with whom they interact. Sub-regional networking of PLWHA groups would: (i) provide important reinforcement of national and under-funded regional 2 All countries committed to reporting on core HIV indicators during the 2001 United Nations General Assembly Special Session (UNGASS). 10 advocacy efforts; and (ii) engage in the sharing good practices. Intermediary institutions would provide management training, and support AIDS competence activities, including piloting knowledge rooms along the transport corridors. 9. Component 3: Support to Regional health-sector collaboration (US$3.0 million). The regional health sector component provided support for five key activities: (i) an inventory of effective interventions and sharing of information, (ii) review of protocols, materials, and training opportunities for prevention and treatment, (iii) information exchange on refugee, IDP, returnee HIV/AIDS health-related programs, (iv) transport sector HIV/AIDS strategy coordination and piloting of targeted transport packages along two main regional corridors, and (v) information exchange on drug policies and procurement. 10. Component 4: Management, capacity strengthening, monitoring and evaluation, and reporting (US$6.0 million). This component covered three activities aimed at strengthening GLIA’s capacities: (i) administration and management, including support for core professional staff, a fiduciary management agent, GLIA country focal points, and operating expenses; (ii) capacity strengthening and policy/technical support, including fiduciary, management, and advocacy training; and policy discussion, development, and technical support for other cross-cutting issues (e.g. gender sensitive HIV/AIDS services for mobile population programs); and (iii) monitoring and evaluation, and reporting. 1.6 Revised Components 11. The components were not revised. 1.7 Other significant changes 12. While the December 2008 Mid-Term Review (MTR) joint assessment recommended a restructuring to formalize proposed changes to the results framework, this was not done for various reasons (e.g. inability of task team to process the restructuring quickly, difficulties in reaching consensus expeditiously among six countries, insufficient Bank management support). The restructuring would have consisted of two critical changes: (i) dropping prevalence and social/gender outcome indicators; (ii) refining other outcome indicators and intermediate indicators to better reflect the PDOs. Even though the results framework was not formally restructured, GLIA stakeholders and the Bank team were tracking and reporting on several new indicators. The ICR team noted that not restructuring the project was a missed opportunity to take into account early lessons, and recalibrate the results framework. The MTR recommendation to extend the time frame for implementation was taken into account with a 21-month extension (i.e. from March 2009 to December 2010). 11 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 13. Preparation: The project was prepared by a strong Bank team in close collaboration with the GLIA member states and several key partners, including UNHCR and UNAIDS (United Nations Program on HIV/AIDS). The task team successfully addressed various challenges during the preparation phase and facilitated a process of developing a common vision and a joint regional agenda among six countries which was well appreciated by all parties. The task team conducted an assessment of regional institutions early in the preparation process and selected the GLIA Secretariat based on its unique mandate and ongoing collaboration with UNAIDS targeting vulnerable groups. Management support was strong as evidenced by the level of resources provided, and by the quality of the technical advice. As noted during the August 2004 Quality Enhancement Review (QER) there was recognition of the substantial progress made by the task team in a difficult work and political environment where several of the member states did not have diplomatic relations. The panel provided appropriate advice on technical, operational, and safeguards issues. It endorsed the proposed AIDS competence methodology as an innovative tool of potential use to the targeted transport and PLWHA networks. Both the panel and management correctly identified the institutional arrangements as a key issue, and recommended that the GLIA Secretariat be reinforced, and key fiduciary functions be initially outsourced. 14. The readiness of project activities was discussed and the task team provided assurances that work plans and detailed budgets had been developed with GLIA country representatives for each of the components. The task team pointed out that UNHCR worked closely with member states during project preparation, and provided support in drafting the work plans for refugee sites. Activities supported by UNHCR got off the ground quickly but that did not appear to be the case for other components, suggesting different levels of readiness. 15. Design: The project objectives and activities were in line with the Country Assistance Strategies (CAS) of the six GLIA member countries. 3 The rationale for a regional project was well presented in terms of tackling cross border issues which were not adequately addressed by national programs. Risks and mitigation measures were generally well identified. For example, the risks that the fiduciary environment was not tested and there was insufficient capacity were well mitigated by retaining a Fiduciary Management Agency (FMA) for the first year. Use of specialized agencies, such as UNHCR and its sub-contractors, which were well placed to address the needs of refugees, proved effective in reaching underserved groups, and mitigating the risk that the GLIA Secretariat would have difficulties carrying out the full range of responsibilities. 3 All six CAS identified HIV prevention and control as key strategic priorities, and most identified regional integration as a means for controlling the spread of communicable diseases. 12 16. A Social and Gender Assessment (recognized as best practice by the Bank’s gender group), was conducted in the DRC, Tanzania and Uganda, and informed the design of the project in a number of key areas, including the emphasis placed on empowerment of mobile populations to address the family’s basic needs and minimize transactional sex which is a key factor in HIV transmission; and the strategy of targeting both refugees and the surrounding communities correctly aimed to promote equity in service provision, foster solidarity, and minimize the risk of strained relations. The use of Behavioral Surveillance Surveys (BSS) at inception and completion was best practice, generating important information about the key drivers of the HIV epidemic, and facilitating an assessment of progress on key performance indicators. With the exception of the BSS other Monitoring & Evaluation (M&E) activities were not contracted out. 17. The task team drew key lessons from the successful Abidjan/Lagos HIV/AIDS Transport Corridor which inspired the institutional arrangements for the GLIA Support Project. While a new organization was established for the Abidjan/Lagos operation in the case of GLIA the Secretariat existed since the late 90s. Moreover, the institution benefited from strong support from the Government of Rwanda which was providing accommodations to host the Secretariat, and was willing to request a PPF on behalf of the other member states and run the risk of having to repay back the funds in case the project did not get off the ground. While the choice of the GLIA Secretariat as the institutional home for the project appears sensible, an explicit capacity building phase should have been identified and simpler flow of fund mechanism and Results Framework could have been adapted. On balance, given the unique and bold nature of the operation the design was reasonable. 18. Quality at Entry: The quality at entry could have been stronger, to facilitate a more rapid start up. Despite having a US$0.6 million Project Preparation Facility to facilitate recruitment of the FMA and preparation of detailed operational manuals and work plans readiness to jump start the project activities was tenuous, as reflected in the 1- year delay in effectiveness due to difficulties and delays in meeting the conditions of effectiveness, setting up project accounts, and selecting intermediary agencies. 4 2.2 Implementation 19. Three main factors slowed down implementation: (i) inadequate management capacity of the GLIA Secretariat in relation to its revamped mandate; (ii) cumbersome flow of funds mechanisms, which resulted in protracted delays in channeling funds to beneficiary organizations; and (iii) weak M&E capacity which rendered difficult the 4 The conditions of effectiveness included: signature of Headquarters Agreement relating to GLIA privileges and immunities; adoption of the Operations Manual by the Council of Ministers; appointment of five GLIA Secretariat staff; selection of Financial Management Agent; establishment of procurement and financial management system; and development of terms of reference for an independent auditor. 13 monitoring of project activities, validation of the data, and assessment of impact. Even though the GLIA Secretariat was not a new structure it was assigned substantial new roles and responsibilities. The Secretariat needed to recruit core staff, set up fiduciary systems, and sub-contract the financial management functions. Its governance structure had some limitations, including weak accountability relationships (e.g. difficulties of the Secretariat to use its authority over member states to facilitate the systematic introduction of harmonized protocols), and a policy of recruiting staff based on nationality and not merit. The Secretariat encountered problems recruiting and retaining staff, which hindered performance. 20. The flow of funds mechanism was complex with: (i) centralized process of channeling funds to countries; and (ii) cumbersome procedures at national level in some countries which contributed to lengthy delays (i.e. Kenya and Tanzania did not receive funds until 2008 as they had difficulties providing timely financial reports). Moreover, there were delays in providing country contributions (e.g. DR Congo) to the GLIA Secretariat. The M&E shortcomings continued to plague the implementation phase. 21. As a result of these various issues, the project was first rated at risk one year after effectiveness, rebounded by the December 2008 MTR, but several key ratings (e.g. PDO) deteriorated by project completion in December 2010, as shown in Annex 4. The Bank should have moved swiftly to restructure the project and extend the closing date immediately following the MTR. Performance was picking up, 50 percent of the funds were still available, and hence there was time to recalibrate the PDO, performance indicators, and implementation arrangements. A restructuring could have made it easier to monitor progress and assess impact. There was insufficient leadership from the Bank in guiding the restructuring process which required extensive consultations and consensus building among the six GLIA member states. A December 2009 Quality Assurance Group (QAG) review of the project identified many of the same issues raised in the ICR but was particularly critical of Bank management for providing insufficient support, and of the task team for creating a new institution, and targeting the wrong population.5 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 22. The design of the project’s M&E system had major shortcomings as recognized by the MTR, implementation was weak, and there was limited evidence of data utilization. The major issues can be summarized as follows: 5 The ICR mission disagreed with the QAG panel that the project targeted the “wrong population”. This conclusion stems from a rapid epidemiological assessment which found that fishermen, long distance truck drivers, military, and female sex workers were at increased risk of HIV transmission and were relatively neglected. The focus on refugees seemed sensible as these groups were not the responsibility of any individual country. The project did target long distance truck drivers and their clients (female sex workers), and other groups (military, prisoners) were clearly the responsibility of national governments. 14  Inappropriate outcome indicators which were qualitative in nature (e.g. increase in social and gender conditions) or could not be easily attributable to the project (i.e. reduction in HIV prevalence rates).  Incomplete baseline data for performance indicators with two important exceptions ---condom use and correct HIV knowledge. In addition, the Development Grant Agreement (DGA) included only a subset of the PAD performance indicators and there were minor discrepancies between the two documents.  Partial behavioral data which limits the evidence for assessing impact. While use of Behavioral Surveillance Surveys for tracking changes in knowledge and behavior is one of the single most innovative aspects of this operation and makes the GLIA project unique among the MAP operations, coverage was incomplete. 6  Limited capacity at national and regional levels, with GLIA country focal points not devoting sufficient time to validating data, and the Secretariat having difficulties establishing an effective regional M&E system. In spite of the Bank’s efforts to assist the Secretariat and the countries to introduce procedures for quality assurance there was generally limited progress with only a couple of data audits and spot checks performed. 2.4 Safeguard and Fiduciary Compliance 23. Environmental issues were adequately addressed, including appropriate disposal of waste in refugee camps. Participating countries had ongoing HIV/AIDS Projects, as part of the Multi-Country HIV/AIDS Program for Africa (MAP), which required preparation of Waste Management Plans that governed all national activities, including those funded under the GLIA project. Gender issues received adequate attention during supervision missions. 24. Fiduciary issues were generally well handled by the GLIA Secretariat. The institution maintained sound financial management systems, ensured key personnel (including an internal auditor) were in place, and produced timely interim financial and audit reports in compliance with the provisions of the DGA. In spite of some initial challenges with the FMA (e.g. perceived high cost, insufficient capacity building) the Secretariat ensured a smooth transition to building in house capacity with the Bank fiduciary team providing support from the Country Office in Rwanda where the GLIA Secretariat was based. As in the case of other MAP operations, the major financial issues which impeded implementation were delays by governments in submitting financial reports, and in providing government counterpart contributions to the GLIA Secretariat. At the time the last ISR was finalized there were significant amounts of unaccounted 6 Baseline BSS surveys were conducted in Kenya and Rwanda (2004), Tanzania (2005), and Uganda (2006) with the follow up surveys carried out in 2010. Across the seven sites, close to 7,300 participants were sampled at baseline and slightly more than 6,000 at follow up. Surveys were not conducted in Burundi and DRC due to difficult country conditions and the end of project survey was not carried out in Rwanda due to disagreements on modalities for conducting the survey between UNHCR and the Government of Rwanda. 15 funds which were reflected in an MU rating. By project completion the GLIA Secretariat and the member states made a concerted effort to address this issue with all funds accounted for, with roughly US$0.6 million remaining unspent and to be returned to the World Bank. Furthermore, the Secretariat’s perseverance paid off by project completion with all arrears cleaned up, demonstrating strong ownership of member states. As a result of these various factors, the overall FM rating for the project is rated as marginally satisfactory. 25. On procurement the ICR mission found that the mix of methods and the management of procurement were well done. The bulk of the procurement was through UN agencies (roughly 48 percent), the second most important method was individual consultants (about 37 percent), which included salaries for the GLIA Secretariat, followed by the competitive recruitment of firms through Quality and Cost Based Selection, QCBS, (about 13 percent). Use of direct contracting was less than 1 percent, which is reasonable given that this procurement method is less competitive and should be used only on an exceptional basis. Based on the assessment conducted the procurement aspects are rated as satisfactory. 2.5 Post-completion Operation/Next Phase 26. A formal request for a follow up operation and a detailed concept note for a second phase were submitted to the World Bank by the chair of the GLIA Council of Ministers (CM) on March 16, 2010, reflecting continued strong commitment and ownership. The Bank suggested that the GLIA Secretariat submit a proposal to the Global Fund which was done during Rounds 8 and 10 but rejected. 7 27. The Bank’s response not to proceed with a follow up operation must be understood in the rapidly evolving regional and global context. The GLIA member states are progressively moving towards broader health systems strengthening, focusing on consolidation of national HIV/AIDS efforts, and increasingly relying on financing from other partners (e.g. Global Fund, United States). There has been substantial progress in containing the HIV epidemic in the GLIA member states. The political and economic context in East Africa has evolved substantially since project inception with the overriding priority shifting towards strengthening the East African Community (EAC). The global economic downturn has made selectivity increasingly important for both Ministries of Finance and the World Bank. Currently, efforts are underway to ensure that activities initiated under the project are incorporated into national HIV/AIDS Strategic Plans and/or transferred to other agencies and institutions. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation 7 The main reasons for the rejection stemmed from concerns over potential duplication with the NACs and the excessive focus on technical support rather than service delivery. 16 28. The development objectives of the project (i.e. to establish HIV/AIDS programs for mobile and vulnerable groups and to enhance prospects for coordinated HIV/AIDS programs) were highly relevant to country and regional priorities at project preparation even though the second objective lacked specificity. The objectives remain broadly consistent with the CAS and Poverty Reduction Strategies. Some mobile and vulnerable populations, particularly refugees, clearly require special attention as hosting governments may not have the incentives and financial resources to address their needs. The provision of a comprehensive package of HIV prevention, care and treatment and empowerment related activities responded to a clear unmet need in a timely and efficient manner operating through strong, specialized agencies, including UNHCR and some 12 seasoned sub-contracting agencies with a strong regional (e.g. Africa Humanitarian Action) or international (e.g. GTZ, International Red Cross) track record serving refugee groups. 29. While the project was designed like other MAPs with a broad range of eligible interventions, in practice, many of the interventions funded (e.g. HIV testing, PMTCT, ART, early pregnancy and gender based violence prevention) are considered good value for money, high impact activities. Targeting both the refugee groups and the surrounding communities proved to be a good strategy for fostering solidarity and facilitating equity in service provision. Similarly, standardizing service provision across the main transport corridors for long distance truck drivers and their clients remains highly relevant and efforts are underway to consolidate these investments and ensure continuity. Finally, exchange of information, epidemiological data, and good practices are critical to strengthening coordination at the regional level with discussions underway with the EAC on an appropriate way forward. 8 Even though the HIV adult infection rates have been revised downwards in most countries (as discussed below) the estimated overall burden of disease remains high with roughly 4.5 million people living with HIV/AIDS in the sub-region (2007). On balance, the relevance of the objectives, design, and implementation was strong and hence is rated as substantial. 3.2 Achievement of Project Development Objectives 30. The achievements of the project are presented for each Project Development Objective (PDO), providing evidence for: (i) establishment of HIV/AIDS prevention, care, treatment, and mitigation programs; and (ii) enhancement of prospects for coordinated HIV/AIDS approaches. The project results are presented schematically in a Results Chain in Annex 2 to better illustrate the links between the activities funded and the outcomes attained. With respect to the first PDO, the analysis covers a discussion of access, utilization, and knowledge and sexual practices, disaggregated by target group (i.e. refugees, surrounding communities) for the three countries with two BSS rounds. To 8 During a recent regional meeting of health practitioners from the EAC member countries the GLIA Secretariat team initiated discussions with the EAC health desk on transferring knowledge and expertise and ensuring that the harmonized protocols will be effectively used. 17 address the influx of new refugees, the BSS authors conducted a sensitivity analysis and found that the results presented below were robust. 9 31. The BSS found consistent declines in risky sexual behaviors, whether multiple, non-regular, or transactional as well as an increase in abstinence among youths and condom use with non-regular partners. The counterfactual (i.e. what would have happened in the absence of the project) cannot be easily established nor can these improvements be attributed to a single project. Nevertheless, to the extent that there were virtually no HIV services in refugee camps prior to the project, and only limited services in the underserved, surrounding communities, it is clear the GLIA Support Project made a contribution towards the outcomes presented below. The latest rounds of Demographic and Health Surveys in the GLIA member states found similar patterns of improvements in self-reported sexual behavior, so the trends cited below appear credible. 10 Interviews with beneficiaries suggest that most of the HIV related interventions would not have taken place without the GLIA Support Project (see Annex 5). 32. There have been substantial improvements in both HIV knowledge levels and condom use among youth at GLIA supported sites. As seen from the figures below, knowledge of HIV/AIDS among 15-24 year olds (project outcome indicator) has nearly doubled in Tanzania, and Kenya (surrounding communities), and improved slightly in Uganda. Condom use among 15-24 year olds with non-regular partner (project outcome indicator) also increased at most sites, reaching 18-34 percent in Uganda, 35-39 percent in Tanzania, and over 47 percent in Kenya (surrounding community). In addition, the surveys found large increases in HIV testing during the past 12 months across all sites. Percent of men and women aged 15-24 with comprehensive Condom use at last sex with a non-regular partner correct knowledge of HIV/AIDS among men and women aged 15-24 Baseline End of Project Baseline End of Project 70 70 63.0 59.4 60 60 50 45.1 45.7 46.9 50 47.2 Percentage Percentage 40 38.5 39.0 39.0 35.4 34.2 34.1 40 35.4 32.6 33.9 31.4 34.4 30 25.7 30 28.0 24.6 20 18.3 18.2 18.0 20 9.7 10 10 0 0 Refugee Surrounding Refugee Surrounding Refugee Surrounding Refugee Surrounding Refugee Surrounding Refugee Surrounding Camps Areas Camps Areas Camps Areas Camps Areas Camps Areas Camps Areas KENYA Kenya UGANDA Uganda TANZANIA Tanzania KENYA Kenya UGANDA Uganda TANZANIA Tanzania (Kakuma)* (Nakivale) (Lugufu/Nyarugusu) (Kakuma)* (Nakivale) (Lugufu/Nyarugusu) *Major demographic change in refugee camp population *Major demographic change in refugee camp population 9 To adjust for demographic changes a sample comparable to baseline was recruited for Tanzania. This was not feasible for Kenya and Uganda. A sensitivity analysis was therefore conducted (excluding new arrivals) which found that the direction of change, if not the absolute levels, for most indicators was comparable for older residents and new arrivals. 10 Looking back and moving forward: Report on GLIA context, the results achieved by the GLIA Support Project to date, and the arrangements that need to be put in place in 2009 and 2010 to improving results monitoring for the GSP, 21 October 2009. 18 33. The surveys also found an increase in abstinence and a notable reduction in high risk sexual activity at virtually all sites supported by the GLIA Support Project. Even though these indicators were not part of the project results framework, they provide additional evidence of reduction in risky sexual behavior. The proportion of never married 15-24 year olds who never had sex remained virtually unchanged in Uganda, increased by over 8 percent in Kenya (surrounding community), and rose by over 20 percent at all sites in Tanzania. The patterns of multiple sexual partners were consistent across sites with males reporting higher levels of sexual activity than their female counterparts, and all respondents reporting fewer sexual partners in the past year with reductions ranging from 39-58 percent in Uganda, 38-68 percent in Tanzania, and 42 percent in Kenya (surrounding community). 34. Key findings from the BSS are generally encouraging with respect to gender related outcomes and broadly consistent with service statistics. 11 The surveys found a reduction in transactional sex at all sites and a systematic drop in forced sex.12 Service statistics tracked a 1.5 fold increase in the number of sexual and gender based violence cases, which key stakeholders noted reflects an improvement in reporting as women overcame their fears and reported abuses with greater frequency. At the same time, the increase in the percentage who received appropriate medical attention increased 2.5 times, reflecting improved access to urgently required services. Against these generally positive trends, the BSS found mixed trends in accepting attitudes towards PLWHA. Inside the refugee camps attitudes towards PLWHA remained unchanged in Uganda and improved slightly in Tanzania. In the surrounding communities there was a dramatic three-fold improvement in Uganda but steep drops in both Kenya and Tanzania (48-60 percent), highlighting remaining challenges to fighting stigma and discrimination. 35. These broad trends are encouraging and consistent with the types of HIV programs which were established under the project. Access to a comprehensive package of prevention, care, and treatment services was virtually universal at all sites supported under the project (Annex 2). Notable results included: 5.2 million condoms distributed to the GLIA funded target groups; 1.8 million persons reached through Information Education Communications/ Behavioral Communications Change programs; roughly 14,500 trained peer educators disseminating messages to their counterparts at youth centers constructed under the project; about 139,000 persons over 15 years of age benefited from VCT programs; roughly 65,000 pregnant women attending antenatal clinics received their HIV test results; close to 1,800 benefited from PMTCT programs; and an estimated 10,200 placed on ART. In addition, the project funded a wide range of activities to: (i) reduce gender based violence, (ii) mitigate sexual and economic 11 The ICR team assessed progress on gender related outcomes (which were broadly worded in the PAD) in terms of what could be inferred from key associated outcome indicators, in accordance with the OPCS Implementation Completion and Results Report Guidelines (November 10, 2010 update). 12 Drops in transactional sex ranged from 14-63 percent in Tanzania, 27-51 percent in Uganda, and 25 percent in Kenya (surrounding communities); and for forced sex among 15-49 year old females from an initial range of 1.5-9.2 percent to .2-2 percent by project completion. 19 vulnerability through income generating activities; (iii) improve conditions of vulnerable groups through provision of home based care for over 21,000 individuals, and school supplies and basic support for roughly 33,600 orphans and vulnerable children. 36. Notable progress has also been made in strengthening two key networks-- long distance truckers, and people living with HIV/AIDS-- and facilitating knowledge sharing among member states. In spite of the difficult conditions in some member states the GLIA project attained important results, even in fragile (e.g. Burundi, DRC) or post conflict (Rwanda) states. Given the vaguely worded nature of the second PDO and the related outcome indicators the ICR team focused more on results indicators and on feedback from beneficiaries. What is clear is that 14 networks of civil society organizations, across the six countries were systematically trained in two important areas where key gaps existed: (i) fiduciary, planning, and managerial skills of networks of civil society organizations; and (ii) AIDS competence and self assessment to enhance awareness and promote empowerment. All civil society networks and member organizations of long distance truck drivers and people living with HIV/AIDS conducted annual AIDS self assessments, adopted strategies, and implemented annual action plans to boost their capacities. There were about a dozen regional events among network organizations and health professionals to exchange information and good practices, and support the harmonization of six protocols. 37. While it is not easy to assess the quality or impact of the training it is important to note that feedback from key stakeholders was generally very positive. Representatives of the targeted networks interviewed noted appreciation with the empowering aspects of planning and implementing their own activities, and the learning opportunities at the regional level. Views of beneficiaries were consistently positive, noting that the project allowed them to come up with their own creative solutions (e.g. Family Day Initiative in Rwanda; Moon Light VCT in Kenya), and jointly design income generating activities. According to beneficiaries and partners interviewed, national HIV/AIDS related health sector information exchange, and harmonization of protocols resulted (or will result) in better services for mobile populations. 38. Use of a systematic approach across the six countries resulted in the establishment of 21 Wellness Centers along the main corridors, which responded to an important gap in service provision for long distance truck drivers and sex workers as well as neighboring communities. It is noteworthy that all the GLIA member states succeeded in establishing these centers. In the absence of regional funding it is unlikely that the fragile states would have attained the same results. The roadside Wellness Centers were able to adapt services and hours of operation to the needs of these mobile and vulnerable groups, operate out of modest containers, generate support from local councils and governments, and establish referral mechanisms to neighboring health centers to improve access to STI and HIV diagnostic and treatment services. Based on reports from the GLIA focal points at project completion virtually all the Wellness Centers were operational but as most were completed towards the end of the project their level of functionality varied, as detailed in Annex 2. 20 39. On the basis of the solid evidence presented above, particularly for the measurable PDO, the ICR team considered the overall effectiveness substantial. However, as data on key outcomes is only available for three countries, the results cannot be readily extended to the other member countries. Consequently, the overall effectiveness is rated Modest. 3.3 Efficiency 40. As was the case for other MAP operations no formal economic or financial efficiency analysis was conducted at appraisal. Efficiency is discussed primarily in qualitative terms. Important efficiency gains were made by:  Working with strong, experienced institutions with good absorptive capacities, such as UNHCR (which managed about 27 percent of the total funds) and its sub- contractors; these institutions had a good knowledge of the refugee population; strong capacities on the ground, which allowed them to hit the ground running quickly and efficiently; unit cost analyses of the number of beneficiaries reached are complicated by the fact that UNHCR pooled IDA resources with other sources of funding; important investments in physical infrastructure (e.g. health facilities; maternity wards; feeding centers; laboratories, youth centers) in the refugee camps were made by UNHCR which in some cases served both the camp populations and neighboring communities, maximizing the returns on these investments. In total, UNHCR successfully constructed/renovated 10 health facilities, 14 youth centers, and trained over 6,000 health personnel, serving a population of roughly .4 million refugees (and part of the 1.2 million in the surrounding communities) with a full range of evidence-based interventions. These activities were delivered on time and within budget.  Standardizing approaches and sharing good practices at the regional level rather than operating on a country by country basis; considering the resources made available for this purpose, the process was quite efficient in having a regionally coordinated effort to harmonize approaches; examples include design, development and establishment of Wellness Centers which used low cost infrastructure (i.e. containers which cost about 20K on average) to serve mobile groups, allowing truckers to get the same care and treatment in every country during their long haul trips; organization of standardized training; harmonization of protocols to ensure the provision of standardized care; and sharing good practices and experiences to promote innovative approaches.  Scaling up key interventions at an accelerated pace. In spite of the delayed start up and delays in disbursements, following the MTR the results attained were impressive, such as full coverage with static VCT centers, a seven-fold increase in the number of condom outlets established, and roughly 1.8 million persons reached through IEC/BCC programs.  Maximizing resources devoted to service delivery. The bulk of the project resources (i.e. over 72 percent) were devoted to the provision of services or strengthening of institutional capacities with the GLIA Secretariat absorbing roughly 16 percent of the total and the balance used for the GLIA governance structures, M&E, and various consultancies and preparatory activities. 21 41. On the basis of the evidence presented above efficiency is rated as Modest. 3.4 Justification of Overall Outcome Rating 42. The achievements of the GLIA Support Project must be viewed in light of the difficult socio-political context in which the operation was executed, its ambitious goals in light of modest institutional capacities, and its relatively meager resources (US$20 million) for a regional project covering six countries. Taking into consideration these various aspects, results on the ground are relatively strong both in terms of expanded access to a full range of HIV/AIDS services for the targeted populations (particularly the refugees and the surrounding communities), and enhanced capacity of civil society organizations and the GLIA Secretariat. Moreover, substantial investments in infrastructure and human resources have been made and will be left behind for the refugee camps, surrounding communities, and two main road corridors in East Africa. One of the most important contributions of the project was the production of Behavioral Surveillance Surveys which provide a stronger basis for planning future investments. While difficult to quantify, the project played a catalytic role in supporting innovative approaches and cross country sharing of experiences and good practices. On the basis of the substantial relevance, modest effectiveness and efficiency, the overall project outcome is rated as moderately unsatisfactory. Breakdown of project outcome rating Objectives Relevance Efficacy Efficiency Project Outcome Rating Establishment of HIV/AIDs Substantial Modest Modest Moderately programs for mobile & vulnerable Unsatisfactory groups Enhancement of prospects for coordinated HIV/AIDS approaches 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 43. The project effectively addressed economic and sexual vulnerability of refugees and mobile populations, using HIV programs as an entry point. Vulnerable groups benefitted from income generating activities which provided them with a source of livelihood. For PLWHA who were on antiretroviral therapy, the income generating activities were critical to addressing their nutritional needs. Income generating activities also necessitated establishment of support groups and associations which promoted solidarity and contributed to building social capital in some communities that were ravaged by conflicts. Beneficiaries of these activities included orphans and vulnerable children, women who engaged in sex work, and PLWHA. 44. Numerous promising schemes have been initiated under the GLIA Support Project. The scope and scale of many of these activities was limited because of the level of funding but the demonstration effect was critical with important lessons for future investments. In Burundi, the East African Food Restaurant was set up by the spouses’ 22 association of truck drivers; located adjacent to a Wellness Center the association has a booming business serving long distance truckers arriving in Bujumbura. In the Kigoma district of Tanzania, two centers have been established for youth, women, and members of HIV/AIDS clubs to develop income generating activities and promote HIV counseling and condom use. The National Council of Churches in Kenya integrated HIV activities into reproductive health services and combined them with income generating activities (e.g. hairdressing, small restaurants) for vulnerable groups. Similar interventions were funded in the refugee camps and surrounding communities in Uganda with a broad range of preventive and mitigation activities, including vocational training for vulnerable children. Rwanda rolled out a broad range of income generating activities (e.g. mushroom production, hair salons, tailoring associations, kitchen gardens, rabbit and chicken raising, soap making) both for the refugee camps (Gihembe, Kiziba, Nyabiheke) and surrounding communities (Gicumbi, Karongi, Gatsibo), which reflects the close integration of GLIA activities within the national HIV/AIDS program. (b) Institutional Change/Strengthening 45. Institutional strengthening has been an implicit goal of this regional project, both in terms of the GLIA Secretariat and the targeted civil society networks. The GLIA Secretariat was founded in the late 90s and formally established through the GLIA Convention (July 27, 2004), signed by representatives of the six countries. The Convention spelled out the roles and responsibilities of the organization, its governance structures, and the financial arrangements. 13 The Secretariat was given an ambitious mandate in relation to its initial capacity. Institutional strengthening efforts had mixed results. In the area of financial management and fiduciary matters, including internal audit functions, the Secretariat performed well, ensuring a smooth transition from a FMA to performing these functions in house. The fact that there were no governance problems reflects good management practices. With respect to procurement, the Secretariat also put in place adequate capacity and conducted procurement with few difficulties. 46. On technical and operational issues, the Secretariat managed to coordinate the project activities reasonably well in spite of several impediments (e.g. difficulties recruiting and retaining staff; limited authority over national agencies). The Secretariat’s performance on M&E was particularly weak, not succeeding to put in place a reliable system for monitoring project activities or establishing capacity at the country level. Only towards the end of the project the agency established the Partnership Consultative Group, which could have provided the platform for effective partner coordination and additional resource mobilization beyond the contributions from the member states. In spite of these shortcomings and its continued dependence on member state contributions and IDA, the GLIA Secretariat played an important role and continues to represent the 13 The key governance structures included: GLIA Council of Ministers, the policy making body consisting of Ministers in charge of HIV/AIDS which met annually to approve annual program and budget; Executive Committee consisting of executive secretaries of the NACs which provided oversight and technical guidance; Partnership Consultative Group which is the advisory body to the Ministers; and GLIA Focal Points who were responsible for ensuring coordination between NACs and the Secretariat. 23 institutional memory of this regional initiative. Discussions are underway with the East African Community health desk to determine the best ways to ensure continuity of key activities, particularly in terms of protocol harmonization. Civil society organizations supported under the project were able to build their management and fiduciary capacities as well as their AIDS competency, and many will be in a position to submit requests to other partners and to better manage their resources. The ICR team was unable to make an assessment of the 14 strengthened networks but feedback was overwhelmingly positive, and the fiduciary training in particular was unique and filled a key gap. (c) Other Unintended Outcomes and Impacts (positive or negative) 47. Several unintended positive outcomes are worth noting. First, the project brought together officials from countries which had tense diplomatic and political relations. They rallied together in the fight against a major public health threat and succeeded in establishing a regional initiative endorsed by the parliaments of six countries, showing how collaboration around HIV/AIDS contributed to broader engagement and improved understanding among the GLIA member states, as noted by numerous stakeholders (Annex 5). One Congolese stakeholder echoed the views of others, noting: "We overcome challenges at the political level and crossed new frontiers, the GLIA brought us into new territory, and enabled us to overcome political barriers, and enabled people to come together and reflect on a health problem that is currently engulfing our region”. Second, some of the project activities (e.g. income generating activities, AIDS competence training) contributed to promoting collaboration between refugees and surrounding communities, building trust, and strengthening social capital in impoverished communities. Stakeholders interviewed noted that one of the positive spillover effects of the AIDS competence training was that it “allowed individuals of different tribal affiliations to work together”, thus contributing to reconciliation efforts. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops N/A 4. Assessment of Risk to Development Outcome 48. Contextual Risks. There are a number of contextual or external risks which may affect the development outcomes. First, there is a global shift towards broader health systems strengthening rather than disease specific ones with the bulk of the global financing targeted for national (and not regional) programs. Second, the sense of urgency about addressing HIV/AIDS has diminished and there is a risk that complacency will set in; this is in part due to progress in containing the epidemic in the region during the past five years, and to more reliable recent estimates which show lower levels of HIV infection in most countries. 14 Third, there is a move towards greater selectivity as 14 HIV adult infection rates for the period of the 2005-2007 were revised downwards for Burundi (3.0%); Rwanda (3.0%); DR Congo (1.3%); and Tanzania (6.4%), while Kenya (7.1%) and Uganda (6.4%) were revised upwards; data are from the last round of Demographic and Health Surveys or other national surveys. 24 donors grapple with the global economic downturn. Finally, the emergence in recent years of numerous regional institutions and the increasing attention being given by member countries to economic, and trade issues through regional economic blocks, may dilute the attention of Ministries of Finance to single theme initiatives. 49. Internal Risks. Project or institutional specific risks can be summarized as follows:  Unpredictable financing: The Secretariat’s inability to diversify funding sources places the institution in a precarious financial situation. To preserve the institutional memory and consolidate accomplishments the GLIA Secretariat has adopted a two-pronged approach: (i) conducting discussions with the EAC to determine the prospects for continuity of key activities which appears like a sensible strategy as it will preserve the institutional memory and continue to raise the profile of these issues in high-level policy and ministerial forums;15 and (ii) exploring possibility of member states maintaining their financial contributions. In spite of difficulties encountered, the Secretariat is viewed by many stakeholders as having built capacities that need to be preserved and GLIA’s 2008-2012 Strategic Plan provides a sound basis for moving forward.  Lack of continuity: The capacity of countries to sustain the levels of activities supported under the project hinges on their: (i) ability to integrate activities into the national HIV/AIDS Strategic Plans and/or transfer them to other donors; all countries are committed to taking over these activities and some countries are reasonably confident that they can do this but others may struggle; 16 (ii) interest and financial capacities of localities or associations to assume the running costs of the Wellness Centers, as many were completed immediately prior to project completion; 17 and (iii) ability of UNHCR to identify other sources of funding for the HIV/AIDS activities. 50. The overall risk to development outcome is rated as significant. 15 A Memorandum of Understanding (MOU) between the EAC and GLIA was signed in February 2010 to promote collaboration and this could serve as the basis for ensuring sustainability of a number of key activities, either by having GLIA serve as a key sub-recipient of EAC funds or transferring the activities to the EAC with special provision for DRC which is not a member of the EAC. 16 Rwanda has a Global Fund grant which generously funds its HIV/AIDS Strategic Plan; in Kenya there is strong ownership of the GLIA activities and a commitment to their continuity, and a possibility that some of the network activities can be picked up through the grant mechanism under the Bank-funded TOWA Project); Burundi and DRC may struggle as MAP operations are closing and other agencies do not necessarily fund the same types of activities or provide comparable levels of financing. 17 Innovative solutions are being identified to consolidate these investments, such as establishing partnerships with the private sector to support their operational costs, as is taking place in Uganda with the association of long distance truck drivers which has assumed management of some facilities; using income generating activities to sustain the centers, and working closely with district and health authorities to ensure effective coordination, such as is taking place in Tanzania. 25 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry 51. The project was generally well designed and responded in a timely fashion to a key public health problem. It was prepared with the support of several partners and benefited from strong managerial support, as evidenced by the budget (608K) provided which was slightly above the average (531K) for a group of similar regional projects. The task team should be commended for a bold vision, and for bringing six countries together to agree on a common set of activities, which is not an easy task. In spite of a concerted effort to ensure readiness, there were lengthy delays in effectiveness and disbursements as capacities needed to be put in place. As was the case for similar projects, the results framework provided only a broad framework with baseline data and targets not always available, and with several loosely worded outcome indicators. Bank performance in ensuring quality at entry is rated as moderately satisfactory. (b) Quality of Supervision 52. There was good continuity in task management with one field-based task team leader from inception to completion. Missions were generally well staffed, and there was good use of field based staff. Specialists were added as needed in the areas of pharmaceuticals, health, M&E, social development and gender. Continuity of sector specialists (particularly on health) could have been stronger. Oversight of fiduciary aspects was particularly strong with good attention to safeguards policies and periodic attention to gender issues. Advice was generally timely and appropriate and by and large well appreciated by the clients, based on feedback received during the ICR mission. The task team identified implementation problems in a timely manner but the response was strongest on the FM and procurement issues. Governance and anti-corruption issues were well handled during supervision. The task team made a concerted effort to address the QAG recommendations, including providing additional support on M&E, ensuring that the second round of BSS was completed, and engaging national authorities to mainstream priority activities into national programs. 53. The MTR was well planned with a clear issues paper and a comprehensive Aide Memoire which laid out the main issues and way forward. In spite of a concerted effort the task team did not succeed in restructuring the operation. The Bank did not provide the leadership and sustained support which was needed to rectify the weaknesses in the results framework and to recalibrate the development objectives, and outcomes. The Bank also missed an opportunity to include the GLIA project in the generic restructuring of all MAP operations to remove reference to reduction in HIV prevalence rates. The supervision budget (540K) slipped below a regional average (614K) for the same group of regional projects. The candor, realism, and consistency in Implementation Status Report (ISR) ratings were mixed. In some cases, supervision ratings tend to be more optimistic than warranted by the detailed component-specific description in the aide 26 memoires, and in other cases ratings may have been too negative. As noted by the QAG panel, management attention should have been stronger. Management feedback on ISRs was not systematic, particularly for the Regional Integration Department which commented only on the last few ISRs. 54. In spite of a generally good supervision record, there were a number of key lapses which have resulted in a rating of moderately unsatisfactory for the quality of supervision. (c) Justification of Rating for Overall Bank Performance 55. Based on the moderately satisfactory quality at entry and the moderately unsatisfactory quality of supervision, overall Bank performance is rates as moderately unsatisfactory. 5.2 Borrower Performance (a) Government Performance 56. Member states demonstrated strong commitment to GLIA’s mission and objectives and managed to overcome political differences to reach a consensus on a regional HIV agenda. The GLIA Council of Ministers met regularly to guide the institution and resolve key issues. While governments did not always provide funds on a timely basis to the GLIA Secretariat, in part due to the precarious financial and political conditions, by project completion all arrears were cleaned up. The delayed start up and difficulties in some countries in channeling funds expeditiously to implementing entities resulted in some activities finalized only towards project completion. Nevertheless, by project completion only US$0.6 million remained unspent and had to be returned to the World Bank. (b) Implementing Agency or Agencies Performance 57. Given the complexities and challenges of providing oversight over six countries and the need to build capacity, the GLIA Secretariat performed as well as could have been expected under these circumstances. No major problems of misprocurement or mismanagement were found by the audits. During the first year of implementation, a FMA was hired to help build GLIA capacity and to test the newly designed management tools. Internal controls were put in place by recruitment of an internal auditor and by adherence to the GLIA convention and regulations in the operations manual. There were some inevitable communication issues between the Anglophone and Francophone countries with some stakeholders noting that the Secretariat could have managed certain processes (e.g. staff recruitment) in a more transparent manner. There were also concerns over the practice of ensuring geographical representation at the Secretariat that did not always result in the selection of the best candidates. Finally, the GLIA Secretariat had an important responsibility with respect to monitoring and evaluation of project activities that was not adequately performed. 27 58. The performance of other key agencies was generally strong. The performance of the UNHCR with respect to component one was solid even though this might have led to weakened country ownership. It sub-contracted organizations with a long standing track record in working with refugees, and carried out activities on a timely basis, and with reasonably good results. UNHCR worked closely with local and district health authorities but the coordination with the NACs (which were mandated to carry out activities in surrounding communities) could have been stronger in some cases to facilitate resolution of key issues. For example, UNHCR’s disagreement with the Government of Rwanda about the modalities for conducting the 2010 BSS were not resolved in a satisfactory manner, which implied that the end of project BSS was not conducted. Use of the adapted BSS gold standard survey is considered best practice in terms of monitoring trends in behavior among the targeted groups but there was limited evidence of effective use of the data. While the co-mingling of Bank funds minimized transaction costs for UNHCR, it limited the Secretariat’s capacity to monitor and report on results. The performance of the FMA was satisfactory on balance. The performance of the two intermediary agencies used for capacity building was generally good, even though some stakeholders argued that they had a limited role in determining the work of these agencies, and the training activities limited direct financial support to the networks. (c) Justification of Rating for Overall Borrower Performance 59. Based on the performance of the member states, GLIA Secretariat, and key implementing agencies the Borrower’s performance is rated as moderately satisfactory. 6. Lessons Learned 60. Key lessons learned are summarized below, broken down into generic lessons on regional health investments, and those specifically related to Bank-funded operations. Regional Health Investments  Regional investments can add value to national investments and maximize impact, especially for mobile and vulnerable groups who tend to be neglected. The main beneficiaries of this regional project---refugees and long distance truckers--- as well as the communities and individuals they interact with (e.g. commercial sex workers) were clearly the appropriate groups to target. Individual countries are faced with enormous challenges in addressing the rapidly evolving needs of refugees camped on their national territories, and may not have the incentives and resources to address their needs.  Addressing both the needs of mobile and vulnerable groups and surrounding communities in an integrated manner avoids creation of distortions. Refugee camps can serve as poles of development for the provision of health and related social services. The GLIA Support Project demonstrated how a regional project can raise the bar in terms of access to services for groups in remote border areas which were lagging behind.  Ensuring activities funded are well integrated into health district plans early on to promote synergies and to facilitate continuity. In countries where the GLIA 28 activities were well integrated into health district or local government plans from the outset, this enhanced the chances of sustainability at project completion.  Working with existing regional institutions and beefing up their capacity through an initial capacity building phase. The Bank can play a critical role in supporting regional institutions, particularly as funding from the Global Fund and bilateral donors is focused primarily on country programs.  Developing a strong Monitoring and Evaluation framework and a clear Results Chain. Four key lessons are noteworthy:  Establishing and institutionalizing a robust M&E system from the onset improves the documentation process, provides solid basis for review of progress, and motivates project stakeholders to achieve the set targets.  Conducting Behavioral Surveillance Surveys enhances understanding of the key drivers of the HIV epidemic and provides a sounder basis for planning future investments. 18  Ensuring that data generated from the BSS are effectively used to inform the design of interventions. The end of project BSS underscored the need for stepping up HIV prevention activities among residents who frequently visit surrounding communities, and among boys and young men to reduce early sexual debut and casual sex, and among adult men to reduce multiple partnerships.  Conducting surveys more frequently, particularly given the rapid demographic changes in most refugee camps and the evolving needs of new arrivals. Surveys that generate “just in time information” would be particularly useful to both implementers and policymakers. Funding permitting, it would also be useful to include modules for assessing HIV infection levels. World Bank Task Teams and Management  Ensuring realism and specificity in project design. Given the inherently complex nature of regional projects it is critical to ensure simplicity in design (i.e. sequence and phase in activities, adopt simple flow of fund mechanisms, focus on a few activities, a limited geographical area). A related lesson is the need to improve readiness through: (a) stronger preparation up front; (b) reliance on established institutions which can hit the ground running quickly and efficiently; and (c) explicit capacity building initial phase. Finally, making the experimental nature of an operation explicit from the outset (as was done for the Treatment Acceleration Project) is helpful and reinforces the need for learning in order to scale up in a subsequent phase. 18 Prior to conducting the BSS in the GLIA member states there was a broad based view that sexual behavior among highly mobile and displaced groups was associated with higher risks; the survey results showed that the situation was much more complex and that these groups did not always have the highest risk levels. The end of project BSS found that refugees had lower levels of risky sex than surrounding communities in Kenya and Uganda but not in Tanzania. 29  Adopting a proactive approach to restructuring projects to enhance chances of success. Bank teams with the support of management, and in collaboration with clients, need to be assertive about restructuring operations. The time and effort required to restructure a regional operation and the perceived low benefits may serve as a disincentive to task teams. Management needs to provide strong support and guidance, and to create an environment which rewards task teams that successfully restructure operations.  Including an Interim Review. Task teams should consider including an Interim Review at 18 months to take stock of progress and identify remedial actions early rather than wait until a Mid-Term Review.  Leveraging field-based staff and maximizing links to national-level investments; working through field-based staff maximizes ownership, ensures clear links to other Bank-funded national health investments, and promotes efficiencies and economies in project preparation and supervision.  Planning for sustainability early in the project cycle, given that regional projects often involve multiple institutions and countries. The project team, in collaboration with the clients, needs to design an exit strategy early in the project cycle. Assisting beneficiary networks or other local groups to take over management of facilities supported by the project should be given priority and agreed upon during the Mid-Term Review.  Staying the course. When embarking on challenging operations the World Bank needs to “stay the course” which was one of the main recommendations of the MAP Interim Review, recognizing the time, effort, and resources needed to undertake a regional project. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies (b) Co-financiers (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) 30 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD (USD millions) Appraisal millions) Refugees 8.00 8.2 102.5 HIV/AIDS Networks 3.00 3.6 120.0 Health sector collaboration 3.00 3.3 110.0 Management and capacity 6.00 5.7 95.0 building Total Baseline Cost 20.00 20.8 104.0 Physical Contingencies 0.00 0.00 Price Contingencies 0.00 0.00 Total Project Costs 20.00 20.8 Front-end fee PPF [.6] [.6]0 Front-end fee IBRD 0.00 0.00 Total Financing Required 20.00 20.8 a/ Differences are due primarily to exchange rate fluctuations. b/ The PPF is included in the figures above. (b) Financing Appraisal Actual/Latest Type of Estimate Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 0.00 0.00 .00 IDA GRANT FOR HIV/AIDS 20.00 20.8 104.00 31 Annex 2. Outputs by Component I. Introduction 1. This annex presents the status of the main outcome indicators, and analyzes the outputs by component. The results are also presented schematically in the Results Chain below as it provides a sounder framework for understanding the links between activities and results. The Results Chain takes into account key activities or outputs (Level IV), intermediate outcomes such as availability and access (Level III and II), and coverage outcomes, including knowledge levels and condom utilization (Level I). Virtually all the indicators in the Results Chain are derived from the Results Framework, the Behavioral Surveillance Surveys, or the GLIA service statistics. Level 0 refers to impacts that are usually beyond the influence of a specific project, such as changes in HIV prevalence rates. This annex presents systematically the outcome and output indicators included in the Project Appraisal Document and the DGA. To better understand and interpret the data, it is important to point out the following caveats, weaknesses and issues. 2. First, there were no baseline data for the majority of the performance indicators and targets set in the PAD. The baseline Behavior Surveillance Survey (BSS) were conducted in only 4 out 6 GLIA countries (Kenya, Rwanda, Tanzania, Uganda). There were no BSS conducted in Burundi or DRC because of the difficult country conditions and in the case of Rwanda no end of project BSS was conducted due to disagreements between UNHCR and the government on the modalities for conducting the survey. 3. Second, data from the BSS need to be interpreted with caution given their self reported nature, and the dynamic changes in the population composition (e.g. for example, the Kakuma camp in Kenya accommodated predominantly refugees from South Sudan at the beginning of the project while by completion most refugees were from Somalia). Sensitivity analyses conducted on the BSS data from Kenya and Uganda found that the direction of change was similar for older residents and new arrivals. Finally, it should be noted that no BSS included truck drivers, which were an important target group. 4. Third, the PAD had too many indicators to report on. A maximum of 2 to 3 project outcome indicators and one or two intermediate results indicators for each of the 4 project components would have been enough. Some indicators (e.g. outcome indicator on decreasing HIV prevalence in intervention areas) were unrealistic and difficult to measure, given the scope and duration of the GLIA Project. Baseline data and targets for some of the outcome indicators were qualitative in nature and difficult to monitor. 5. Fourth, at the MTR a project restructuring of the results framework was proposed but the restructuring could not take place. Nonetheless, countries continued to report on some of the new indicators in the proposed results framework following the MTR. 32 6. Fifth, the roles and responsibilities of GLIA country focal points were not clearly defined. In some cases these focal points had other responsibilities in the NACs which diluted their responsibilities towards the GLIA Project, as they were unable to spend sufficient time to review data quality in reports of implementing agencies. 7. The Development Grant Agreement (DGA) included a subset of the results indicators which were in the PAD. The DGA stated clearly in Schedule 6 that “recognizing that further indicators are set out on in the Results Framework of the Project, the recipient shall implement the Project in accordance with the performance indicators set forth in the DGA.” So, even though from a legal point of view the client was only accountable for the DGA performance indicators (*), the ICR team chose to analyze and report on the entire set of indicators in the PAD. 8. Finally, it should be noted that while broad health and social support activities conducted in refugee camps under the leadership of UNHCR are funded from many funding sources, HIV interventions were not part of the core package prior to GLIA project or were provided on a very small scale. The GLIA Support Project helped to scale up HIV interventions, contributing to the observed trends in outcomes discussed below. 33 Figure 1: Results Chain Level 0 Level IV Level III Level II Level I Health and Development Outcomes Activities Capacity Outcomes Service delivery Outcomes Coverage Outcomes  (Impact) Constructing/Rehabilitating Facilities Improved Availability of Infrastructure Effective Access to Health Services - Health facilities constructed, renovated, and/or -Sites with functioning health facility (%) - People with access to a basic package of health, PDO 1: Establishment of HIV/AIDS prevention, equipped (number). [CORE] - Sites with functioning Wellness Centers (%)* nutrition, or population services (percent increase). care, treatment and mitigation programs for - Wellness Centers constructed (number)* - Sies with Functioning Youth Centers (%) [CORE] mobile and vulnerable groups - Youth centers constructed (number) - Refugee, surrounding and returnee sites with - Pregnant women attending ANC services who access to functioning VCT services (%) * received HIV test results (number)* Prevention - Refugee, surrounding and returnee sites with - Persons who are 15 yrs and above who undergo - Youth (aged 15-24) reporting condom use during sexual access to functioning PMTCTservices (%) * VCT and know their HIV results (number)* intercourse with a non-regular partner (%)* - Health sites with adapted services to improve - Youth (aged 15-24) who correctly identify ways of accessibility of transport sector workers (number)* preventing HIV transmission and who reject major misconceptions about HIV transmission or prevention (%)* Procuring Condoms, Drugs, Consumables & Improved Supply of Condoms, Drugs, Effective Availability of Condoms, Drugs, - Pregnant women who are HIV positive in GLIA target Educational Materials Consumables and Educational Materials Consumables and Educational Materials populations who received a complete course of ARV to reduce MTCT (number)* - Condoms procured (US$) IMPROVED HEALTH OUTCOMES - Condoms distributed in the GLIA funded target - Sites having uninterrupted and sufficient supplies - STI/OI/TB drugs procured (US$) popultations (number)* Care and Treatement of male condoms over the previous 6 months (%) * - Diagnostic tests procured (US$) - People with advanced HIV infection in GLIA target population who receive ART (number)* - HIV prevalence in the refugee and surrounding population* Mitigation - Infected infants born to HIV infected mothers* -PLHIV in GLIA target population that were enrolled in Improved Availability of Skilled Human homebased care (number)* Training and Retraining Health Workers - OVCs in GLIA target population who received basic Resources external support (number)* - Health personnel receiving training (number) - Health workers who treat STIs in health clinics that have been trained in syndromic management (%)* Gender sensitization [CORE] - Women 15-49 forced to have sex (%)* - Community health workers trained (number) - Health workers who treat common OI sin health clinics that have been trained (%)* -Never married 15-24 who never had sex (%)* - Sites have trained community health workers who -More than one sexual partner among 15-49 yr olds (%)* are providing home-based care (%)* -Sex with a transactional partner among 15-49 yr olds (%)* -Sexual and gender based violence cases in target population (number)* - Sexual and gender based violence cases reported that have received appropriate medical support according to Training in IEC/BCC Improved Information, Education, and Access to Information, Education, and guidelines (number)* Communication/Behavior Change Programs Communication/Behavior Change Programs - Sites with functioning peer education programs -Active peer educators ( number )* - People reached with IEC/BCC programs (%)* -IEC materials procured (US$) ‐Functioning peer educator programs (%)* (number) * - Sites withHIV posters and billboards in appropriate languages (%)* -IEC Materials produced & distributed (number) PDO 2: Enhancement of prospects for Developing GLIA Secretariat & Member State Improved Monitoring, Coordination and coordinated approaches addressing HIV/AIDS Capacity for Regional Integration Management prevention, care, treatment and mitigation among GLIA member states -- Networks with strategy and action plan for - PLWA and transport sector networks that have support and strengthening (number)* - Organizations that meet the AIDS Self carried out an annual AIDS Self Assessment* - Management training sessions organized for Assessment target (%)* (PLWA networks and - Strategy with adapted health services for truck member organizations (number) * Transport sector networks) system workers has been adopted through - AIDS Competence training for member - Protocols harmonized and adopted (number)* consensus during a formal meeting* organizations (number)* - GLIA has a functioning M&E system where data - Subregional meetings between networks to flows to a central leve is captured, analyzed and exchange good practices (number)* disseminated* -- Formal health sector interaction meetings to exchange information (number)* -GLIA has appointed all 8 core staff members* * Data for these indicators are presented in the tables below 34 II. Outcome indicators 9. According to the PAD, data on six outcome indicators were to be collected at the start and end of the project. For two outcome indicators (measured in four countries through the BSS) there were baseline data available against which achievements were measured at the end of the project. There was no baseline or targets set for the remaining four outcome indicators. The outcomes discussed below can, in part, be attributed to the results achieved on outputs, as shown under the section below. Outcome indicator #1: Reduction in HIV prevalence in the refugee populations and surrounding communities identified by GLIA. 10. This indicator was part of the core indicators that were commonly used by the international community (UNGASS) at the time the GLIA Support Project was being prepared. It became subsequently clear that a project like GLIA, with limited resources and activities reaching a relatively small number of people, could not achieve a reduction in HIV prevalence rates (i.e. a Level 0 Outcome Indicator). The MTR team recognized this and no data were collected for this indicator. As noted earlier, overall trends in the GLIA countries are generally encouraging with a decline in the number of new infections in at least three countries (i.e. Rwanda, Kenya, Tanzania), reflecting the stepped up HIV efforts funded by numerous partners. Graph 1 Estimated new HIV infections (All ages) 140000 120000 100000 80000 Tanzania Kenya 60000 Uganda Rwanda 40000 20000 0 2003 2004 2005 2006 2007 2008 2009 Source: UNAIDS Estimate 2010 Outcome indicator #2: Increase in the number of youth (aged 15-24) in the target populations reporting condom use during sexual intercourse with a non-regular sexual partner. 11. The data below show that there has been significant increase in condom use between baseline and end of project in the 3 GLIA countries where BSS data are available. The wide availability of condoms accompanied by aggressive campaigns from youth peer educators to promote condom use combined with the uninterrupted supply of 35 condom supplies contributed to the results achieved. The data on condom distribution and availability in refugee camps and surrounding communities are presented below under the section on results. Graph 2: Outcome indicator #3: Increase in the number of respondents 15-24 years of age within the target population who both correctly identify ways of preventing the sexual transmission of HIV and who reject major misconceptions about HIV transmission or prevention. 12. Graph 3 below shows the trend in comprehensive correct knowledge of HIV transmission and prevention among 15-24 years. The data suggest that there have been overall increases in knowledge of HIV/AIDS in both refugee camps and in surrounding communities served by the GLIA Project over the period of the project. Part of the decline in the Kakuma refugee camp reflects the change in the refugee population rather than a true decline in comprehensive level of knowledge about HIV/AIDS. 13. The two extreme values in comprehensive knowledge relative to baseline ranged from a low of 18.3 percent in the Kenya hosting community at baseline to 59.4 percent in the Tanzania hosting community at project completion. In contrast to the increase in HIV knowledge, accepting attitudes towards people living with HIV/AIDS remained virtually unchanged in one site, declined at three sites, and improved at two locations (Graph 4), highlighting remaining challenges to fighting stigma and discrimination in some countries. Further analysis would be needed to better understand the underlying factors behind these mixed trends and develop appropriate responses. 36 Graph 3: Graph 4: Outcome indicator #4: Increase in the social and gender conditions within targeted populations in refugee sites and surrounding populations. 14. It is difficult to measure such a broadly worded outcome indicator. Achievement of this indicator may be measured by a combination of indicators such as the number (or percent) of pregnant women placed on ARVs, number of income generating activities benefiting women and orphans and vulnerable children (OVCs), drop in sexual and gender-based violence, and the decline in transactional sex with expanded access to alternative sources of livelihood. The project has supported various activities which may have contributed to the trends discussed below. 15. The number of reported sexual and gender based violence (SGBV) cases in the GLIA-funded target population increased from 1,196 at the MTR to 1, 772 by end of project. Although seemingly an unfavorable outcome, this may also be interpreted asa positive development as more women were able to overcome fears and come out to report abuses. In parallel the number of sexual and gender-based violence cases that received 37 appropriate medical support (according to guidelines) also increased during this period (from 418 at the midterm review to 952 at end of project), reflecting improved access to medical care. Data from the two repeat BSS show positive trends on several social and gender indicators. For example, forced sex towards women decreased at six sites, with one of the highest declines in the surrounding communities in Kenya. Graph 5: 16. Progress was also made on abstinence among never married 15-24 year olds which increased at four sites, remaining virtually unchanged in the Uganda refugee camps and hosting community (Graph 6). Graph 6: 17. The BSS also found declines in multiple sexual partnerships (Graph 7). The decrease was highest in the Tanzania refugee camp and hosting community. There has been a relative decrease in transactional sex at all the sites between baseline and end of project (Graph 8). 38 Graph 7: Graph 8: Outcome indicator #5: Networks have been strengthened and there is improved management and AIDS learning capacities of networks and member organizations 18. This is a qualitative indicator for which there was no baseline data or targets. The ICR mission found positive views among beneficiaries with the training received on planning, financial management, and networking. In terms of learning capacity of networks, progress has been made through training members of the networks in each GLIA country. 19. In total, the number of facilitators that have been trained in the AIDS self assessment process was 21 in Tanzania, 21 in Kenya and 4 in Uganda. On the managerial capacity building side, the number of trainers that have been trained was 6 in DRC, 15 in 39 Kenya, 18 in Rwanda, 10 in Tanzania and 6 in Uganda. Rwanda mobilized another 20 persons who received training in managerial capacity prior to GLIA. Outcome indicator #6: Increased knowledge transfer between GLIA countries and uptake of health services by mobile populations 20. This is another qualitative indicator for which there is no baseline or target data. Increased knowledge transfer occurred through various regional workshops and study visits between GLIA member countries where good practices were exchanged. It is during one of these visits, for example, that Burundi learned about the existence of an association of spouses of truck drivers in Rwanda which led to the establishment of a similar association in Burundi (Association de Femmes des Chauffeurs et Conducteurs Agrees au Burundi, AFCHACABU). Another example is the VCT Moonlighting where commercial sex workers and truck drivers benefited from HIV awareness activities and free testing. Kenya copied this experience after an exchange visit with their counterparts in Uganda. 21. Network knowledge was increased by facilitating discussions among network members through an AIDS competence course and Self Assessment and Learning Transfer (SALT) strategy. In total, 36 SALT visits were facilitated for the targeted networks. Although late in the project cycle, 21 Wellness Centers were established by project completion with support from the project across the 6 participating countries along 2 main transport corridors. While many of these centers provided primarily recreational activities and HIV awareness support for network members, in some cases they also provided VCT and other health services. At most sites individuals were referred to nearby health centers, some of which were constructed under the GLIA project. III. Results indicators 22. Contrary to the outcome indicators, baseline data and targets were set for all results indicators which consisted of a combination of process and output indicators. The baseline data were set to 0 for all indicators to reflect the general lack of availability of these services at project inception. The targets for results indicators have been set in terms of the projected number of refugee sites and surrounding communities to be supported by the GLIA Support Project. The targets for all results indicators for component 1 for each of the two communities (refugees and surrounding populations) were set at 9, or a percentage of that number. Table 1 below show the size of the refugee camps and surrounding areas and, the implementing agencies in the refugee camps. In total, the population of the GLIA targeted areas in the six countries was close to 1.6 million. 40 Table 1: Population size of GLIA supported refugee sites and surrounding communities with respective implementing partners Country Refugees sites Population Implementing Surrounding Population Implementing as of June, partners communities as of June, partners 2010 2010 Burundi Gasorwe 9,532 Stop SIDA, TPO Gasorwe 86, 130 District, MOH Musasa 6,504 Stop Sida, TPO, Musasa 95,710 District, MOH AHA Bwagiriza 3,082 AHA - - District, MOH DRC Libenge 37,778 AHA - - Kenya Kakuma 72,000 LWF, NCCK, Kakuma 150,000 NCCK, MOH CWS, IRC Rwanda Gihembe 19,488 ARC Gicumbi 86,444 District, MOH Kiziba 18,827 AHA Karongi 84,942 District, MOH Nyabiheke 14,247 ARC Nyabiheke/ 68,045 District, MOH Gatsibo Tanzania Mtabila 35,130 JRS-Radio Ngara 162,320 District, MOH Kwizera, (Nyamahwa/ Norwegian Kasulo) Church Aid, RUDESO Nyarugusu 61,130 TRC Kigoma - 158,869 District, MOH Kazura Mimba/ Uvinza Kasulu 149,774 District, MOH Uganda Nakivale/ 21,136 Akton Afrika Isingiro 111,632 District, MOH Oruchinga Hilfe, Germany District Kyangwali 54,000 GTZ Hoima 61,500 District, MOH District/ Kyangwali TOTAL 352,854 TOTAL 1,215,366 Component 1: HIV/AIDS support to refugees, affected areas surrounding the refugee communities, internally displaced people, returnees. 23. GLIA support boosted VCT, PMTCT, STI, SBGV, post test clubs, and outreach activities, training, and funded renovation/equipment of health, nutrition, and youth centers. As a result, over 20 upgraded/rehabilitated/constructed health centers could offer a broader range of services, including curative and preventive services, reproductive health, HIV counseling and testing, HIV/AIDS comprehensive care, and home-based care. Additional activities included income generating activities (IGAs), support to orphans and vulnerable children (OVC), youth centers, nutritional support, behavior change communication (BCC), and condom distribution. UNHCR was responsible for activities in the refugee camps, as stipulated in a Memorandum of Understanding and management contract between the GLIA Secretariat and UNHCR, and in turn subcontracted specialized NGOs to implement the activities. With respect to the host 41 communities the NACs were responsible for implementing these activities. All results indicators for component 1 were fully achieved with improved access to a full range of HIV/AIDS prevention, care and treatment services. However, these figures do not capture the quality, utilization or coverage of these services. Table 2: Baseline data, targets, midterm and end of project results indicator achievements for component 1 Intermediate Results Indicators Baseline Target Mid End of Results Term project Prevention Component One: 100% of sites have Refugee: 0 9 (100%) 9 (100%) 9 (100%) The 9 refugee sites, uninterrupted and Surrounding: 9 (100%) 9 (100%) 9 (100%) 9 surrounding sufficient supplies of TBD populations and 1 male condoms over the returnee site have previous 6 month period been provided with 100% of sites have HIV Refugee:: 0 9 (100%) 9 (100%) 9 (100%) increased access to posters and billboards in Surrounding: 9 (100%) 9 (100%) 9 (100%) functioning and appropriate language TBD appropriate >75% have functioning Refugee:: 0 7 (75%) 9 (100%) 9 (100%) HIVIAIDS peer educator programs Surrounding: 7 (75%) 9 (100%) 9 (100%) prevention, care TBD and treatment 100% sites have access Refugee:: 0 9 (100%) 9 (100%) 9 (100%) services  to functioning VCT Surrounding: 9 (100%) 9 (100%) 9 (100%) services* TBD 100% of refugee, Refugee:: 0 9 (100%) 9 (100%) 9 (100%) surrounding and returnee Surrounding: 9 (100%) 9 (100%) 9 (100%) sites have access to TBD functioning PMTCT programs Treatment 100% of health workers Refugee:: 0 9 (100%) 9 (100%) 9 (100%) who treat STIs in health Surrounding: 9 (100%) 9 (100%) 9 (100%) clinics have been trained TBD in syndromic management** 100% of health workers Refugee:: 0 9 (100%) 9 (100%) 9 (100%) who treat common Surrounding: 9 (100%) 9 (100%) 9 (100%) opportunistic infections TBD (OI) in health clinics have been trained** Care >75% of sites have Refugee:: 0 7 (75%) 9 (100%) 9 (100%) trained community Surrounding: 7 (75%) 9 (100%) 9 (100%) health workers who are TBD actively providing home- based care* *Indicators included in the DGA. ** Indicators are combined into one indicator in the DGA. 42 24. The availability of programs for HIV/AIDS prevention, care and treatment, and mitigation of social impacts in refugee camps, and surrounding areas and returnees/IDPs, targeting underserved populations in the Great Lakes region improved substantially. In part due to the GLIA Support Project, by project completion refugee populations have been integrated for the first time into some national ART scale up programs of hosting governments. In the area of prevention, notable accomplishments included: an expansion in the production of IEC materials (i.e. that rose from 1,125,473 at the MTR to 1,783,399 by project completion) and a major boost in the number of people reached with HIV/AIDS messages. 25. The number of outlets for condom distribution increased as well from 43 at MTR to 307 by the end of the project, and the number of condoms distributed increased from 3,407,441 at MTR to 5,241,773 at the end of the project, which contributed to the reported improvements in condom utilization among youth presented above. Youth centers have been constructed in each refugee camp and in surrounding communities and are used for recreational and HIV prevention activities, such as drama groups for post HIV test clubs and HIV awareness raising. In addition, HIV positive or negative youth/adults have formed clubs to raise awareness and fight stigma and discrimination which remain problems as seen from the variable improvements in attitudes towards PLWHA. At each refugee site and surrounding community, there were training sessions for youth peer groups (in- and out-of-school) and drama groups (formed by peer groups). 26. There was 100 percent coverage of sites with static VCT centers. Reproductive health/FP services have improved through training of midwives, rehabilitation of maternity wards, and provision of commodities for safe delivery. Sexual and gender- based violence survivors were provided with a comprehensive HIV prevention package, consisting in STI, VCT, and counseling and post exposure prophylaxis (PEP) services which contributed to the reported improvements discussed above in the number of sexual and gender-based violence cases that received appropriate medical support. 27. In the area of care and treatment, all refugee sites had at least one ART center which was used by both refugees and host community populations. Access to treatment for STIs, OIs, and TB has been scaled up and was virtually universal with these services reaching increasing numbers of people. All sites (refugees and surrounding communities) were equipped to provide home-based care. The number of PLWHAs receiving home based care increased gradually over time with the expansion in outreach activities supported under the project (Table 3). 28. In the area of impact mitigation, support has been provided to OVCs by supplying school materials, uniforms and vocational training (e.g., tailoring, masonry, driving, and gardening. Support has also been provided to PLWHA through the provision of grants for IGAs (e.g. piggery, bee keeping, clothes dying, tailoring, and retail shops). 29. The expanded access to these services has contributed to the improved outcomes of the GLIA Support Project (e.g. improvements in HIV/AIDS knowledge levels, increases in condom use, reduction in high risk sex). Additional highlights of component 43 1 achievements are presented below for the three countries (Burundi, DRC, and Rwanda) which lack data on project outcomes from two repeat BSS. Table 3: Selected GLIA output indicators on prevention, care and treatment, and impact mitigation by the MTR and end of project Outputs MTR 2008 End of project Prevention Number of persons reached with IEC/BCC programs (type 1, 125, 473 1, 783, 399 specified) in the GLIA funded target populations Number of condoms distributed in the GLIA funded target 3,047,441 5,241,773 populations Number of active peer educators in GLIA-funded target 5,903 14,516 populations Number of sexual and gender based violence (SGBV) cases 1,196 1,772 reported in GLIA-funded target population Number of sexual and gender-based violence cases reported that 952 have received appropriate medical support according to guidelines 418 Number of pregnant women attending antenatal clinic services in GLIA-funded target populations who received their HIV test 33,241 64,988 results Number of pregnant women who are HIV positive in GLIA funded target populations who received a complete course of ARV 1,493 1,793 prophylaxis to reduce the risk of mother-to-child transmission of HIV Number of persons in the GLIA funded target populations who are 15 years old and above who undergo HIV voluntary 98,974 138,811 counseling and testing who know their HIV results Care and Treatment Number of people with advanced HIV infection in the GLIA funded target population who qualify for ART according to 3,594 10,211 national agreed protocols or WHO/UNAIDS standards who receive it Impact mitigation Number of PLWHA in the GLIA funded target populations that 8,803 21,281 were enrolled in home based care program Number of orphans and other vulnerable children (OVC) in the GLIA funded target populations who received of basic external 17,952 33,608 support Burundi 30. In Burundi, GLIA funds supported comprehensive HIV activities in Bwagiriza, Gasorwe and Musasa refugee camps and the surrounding communities. The bulk of the interventions focused on behavior change communications, promotion of condom use, voluntary HIV counseling and testing, and PMTCT services. The number of persons reached through the IEC/BCC program was 79,823 at MTR and 192,217 by the end of the project. There were only 2 active peer educators by the MTR and this number skyrocketed to 210 by the end of the project. Given that the entire target population of the refugee camps and surrounding communities consisted of roughly 201,000 inhabitants this suggests that these interventions reached a considerable share of the target population. Burundi reported 12 SGBV cases at the MTR and 37 at the end of the project. Around 733 pregnant women at MTR and 1, 156 at the end of the project 44 accessed HIV counseling and testing services, of which 9 (1 at MTR and 8 at the end) HIV positive patients in advanced stage of the disease were put on antiretroviral therapy. DR Congo 31. Concerted efforts were made to provide support to the returnees at the Libenge camp. About 25 health care workers underwent 3-day training on HIV capacity building, coordination, planning, and monitoring and evaluation. In addition, an HIV focal person, a nurse supervisor, a logistics specialist/accountant, and a driver were recruited which considerably boosted the camp’s human resource capacity. 32. To support HIV prevention activities for 37,778 returnees, 1,500 IEC HIV prevention materials and 3,457 (leaflets, posters, video tapes, rolls banners and T-shirts) were procured and distributed by the MTR and by the end of project, respectively. The number of persons reached with IEC/BCC program was 39, 854 at the MTR and 62,895 by the end of the project, suggesting that some people were exposed to multiple activities. There were a total of 11 active peer educators working with the returnees. The number of SGBV cases reported by the returnees was the lowest among GLIA countries with 9 cumulative cases reported by MTR and 18 by the end of the project. 33. Nevirapine tablets and boxes of Nevirapine syrup were procured and distributed in government health facilities to support PMTCT services with 1, 761 women attending prenatal care clinics benefiting from HIV counseling and testing services and receiving their test results. Of those tested, 20 were HIV positive and were given a complete course of ARV prophylaxis, per the national guidelines. Rwanda 34. GLIA funds were used to provide PMTCT services in refugee camps and surrounding communities (9,195 cumulative number of pregnant women were tested and received their result by the end of the project, of which 268 tested positive with 185 receiving a complete course of ARV prophylaxis to reduce the risk of PMTCT). Income generating activities start up kits were given to 5,978 OVCs in the refugee camps and numerous other income generating activities were scaled up in the surrounding communities in line with the government’s policy of supporting vulnerable groups such as PLWHA to find an alternative form of livelihood. The GLIA Support Project supported a wide range of income generating activities (e.g. mushroom production, hair salons, tailoring associations, kitchen gardens, rabbit and chicken raising, soap making) both for the refugee camps (Gihembe, Kiziba, Nyabiheke) and surrounding communities (Gicumbi, Karongi, Gatsibo), which reflects the close integration of GLIA activities within the national HIV/AIDS program. Component 2: Support to HIV/AIDS related networks 35. GLIA targeted efforts around two principal transport corridors: (i) Mombasa- Nairobi-Kampala- Kigali-Bujumbura-Bukavu-Goma; and (ii) Dar-es-Salam-Dodoma- 45 Kigali-Bujumbura-Bukavu-Goma. In collaboration with member states, the GLIA Secretariat conducted an inventory of health services along the main road axes and selected sites strategically by prioritizing hot spots for disease transmission in proximity to truck stops. In total, 21 Wellness Centers were established at these sites (Table 4). 36. Health services (prevention, care, and treatment) for truckers and the communities the groups (commercial sex workers) they interact with were gradually expanded. The district health teams were the main providers of these services. In some centers (e.g. Mlolongo Wellness Center in Kenya) a full range of health services were provided while in others (e.g. Kigali based Magerwa Wellness Center) only recreational and educational services were delivered. Training of counselors on HIV pre-test, post- test counseling and psychosocial support to transport workers took place. Member states also conducted planning meetings to inform national partners about HIV/AIDS services targeted to long-haul transport workers and to foster improved coordination across countries. Stakeholders from member states worked together to design the Wellness Centers and facilitate access of long-distance truck drivers to health services along the two main corridors. Table 4: Wellness Centers supported by GLIA Countries Location End of Project Status a/ Burundi Kanyaru, Kobero, Bujumbura Centers are completed with different degrees of functionality. DRC Uvira, Goma, Bukavu Centers are operational. Kenya Mlolongo, Busia, Malaba, All centers are operational. Efforts are Mariakani underway to transfer the Busia center to IOM/IRAPP Rwanda Magerwa (Kigali), Huye, Centers operating with various degrees of Rusumo (Kirehe) Rubavu functionality. Tanzania Zanzibar (Pemba), Kibaigwa, Pemba is operational with the other two Chalinze, Kagongwa completed and equipped after project completion and expected to be operational by mid 2011. Uganda Mbuya, Rubaale and All three centers are fully functional. Naluwerere a/ End of project status as reported by GLIA Focal Points 37. Two intermediary institutions provided management training for trainers-of-trainers from the networks supported under the project (i.e. long distance truckers, PLWHA) in techniques of organizational self-assessment, and management. As a result of this support, existing networks were strengthened and new ones Source: www.aidscompetence.org promoted by: (i) developing their management capacity in the area of planning, financial management, resource mobilization and M&E; and (ii) promoting 46 networking and developing HIV/AIDS learning capacity. The learning approach was based on BP/UNITAR methodology known as the “AIDS Competence Program”, developed by British Petroleum Company and UNITAR. Both direct beneficiaries and other partners interviewed during the ICR mission noted that truck drivers’ networks and associations of PLWHAs increased their capacities at individual and organizational level to plan and implement HIV projects and manage resources. 38. As table 5 shows, all targets set forth in the DGA for component 2 were achieved or in some cases exceeded. By establishing roadside Wellness Centers and linking them to selected health facilities along the major traffic corridors, GLIA made it possible to provide HIV prevention, care and treatment to an otherwise underserved populations group of long distance truck drivers and the groups they interact with (i.e. commercial sex workers). In some cases (e.g. Uganda) the establishment of Wellness Center led local communities to lobby for the creation of health facilities at nearby locations. Most of the Wellness Centers were established towards the end of the project and network members did not have enough time to implement all planned activities, hence they are now operating with various degrees of functionality. Table 5: Baseline data, targets and results indicators for component 2 Intermediate Results Indicators Baseline Target Mid End of Results Term project Networks For PLWAs networks (transport sector Each identified network have 0 6 6 7 and PLWHA) in adopted the strategy and developed the Great Lakes an action plan for support and Region (GLR) strengthening support and have been strengthening* identified, strengthened and Each identified network has carried 0 6 6 7 supported in a out an annual AIDS Self sustainable way assessment* Networks % of organizations that meet the 0 90% 90 100% (transport sector AIDS Self Assessment target** and PLWHA) in the GLR have Number of formal sub regional 0 3 2 3 improved their meetings between networks of the own HIV 6 countries to exchange prevention, care information, good practices and and treatment experiences activities. Number of management trainings 0 2 1 2 sessions per country organized for member organizations For transport sector network Each identified PLWA network 0 6 7 7 have adopted the strategy and developed an action plan for support and strengthening support and strengthening Each identified PLWA network 0 6 7 7 has carried out an annual AIDS Self assessment 47 % of organizations that meet the 0 70% 70% 100% AIDS Self Assessment target Number of formal sub regional 0 2 1 2 meetings between PLWA networks the 6 countries to exchange information, good practices and experiences Number of management trainings 0 2 1 2 sessions per country organized for member organizations *These indicators are in the DGA **This indicator is in the DGA with a target of 75% Component 3: Support to Regional health sector collaboration 39. The GLIA Secretariat and member states have achieved tangible results in terms of harmonizing HIV approaches and programs. This was one of the first major initiatives to bring together NACs and Ministries of Health at the regional level, enhancing collaboration and cooperation. According to the beneficiaries and partners interviewed during the ICR mission, national HIV/AIDS related health-sector information exchange, and harmonization of protocols resulted (or will result) in better services for mobile population (e.g. long distance truck drivers). 40. GLIA also facilitated the exchange of country experiences and good practices between the NACs and implementers (MOH, PLWHA and transport networks), in collaboration with other partners (WHO, EAC, IOM). Two regional workshops with NACs and UNHCR on HIV/AIDS service delivery for refugees and the surrounding communities were organized as well as two sub-regional workshops for the comparative analysis and harmonization of 6 HIV service delivery protocols at sub-regional level. 41. It is also important to mention GLIA’s contribution to having NACs reorient their HIV/AIDS programs in line with their epidemiological profile, by focusing on high risk groups, and integrating these activities into national HIV/AIDS Strategic Plans. The 2008-2012 GLIA HIV/AIDS Strategic Plan focused attention on these high risk groups based on the work conducted under the project as well as the rapid epidemiological review conducted with support of the World Bank. As noted below, the targets were achieved, and to some extent exceeded (Table 6). Table 6: Baseline data, targets, midterm and end of project results indicators for component 3 Intermediate Results Indicators Baseline Target Mid End of Results Term project Component Number of formal health sector interaction 0 2 1 4 Three: meetings to exchange information, country There is experiences and effective practices between increased the GLIA member states.* synchronization Number of protocols harmonized and 0 3 6 6 and adopted harmonization of Number of formal interaction between GLIA 0 2 1 2 health policies member states to exchange o f information 48 on HIV/AIDS prevention, care and treatment for refugees, IDPs, returnees, and surrounding communities* A strategy with adapted health services for 0 yes No Yes truck system (TS) workers has been adopted through consensus during a formal meeting Number o f health sites with adapted services 0 22 21 21 to improve accessibility of transport sector workers in selected sites* *These indicators are in the DGA, although worded in a slightly different way. Component 4: Management, capacity strengthening, monitoring and evaluation, and reporting 42. The GLIA Secretariat took over from the FMA in August 2007 following one year of technical assistance. Financial management and reporting tools (including internal auditing tools) were developed and in-country training sessions were conducted to improve transparency and efficiency in the management of GLIA funds. GLIA internal audits were conducted to verify the status of financial management systems put in place by the member states. 43. The GLIA M&E system was meant to function symbiotically with the six national HIV M&E systems, to avoid duplicative efforts. The GLIA M&E Focal Points (based at the NACs) were meant to be the key interface between the GLIA Secretariat and the NACs, feeding information to the Secretariat, disseminating data to inform planning of future HIV activities, and ensuring that all routine GLIA data were of good quality. An assessment of the quality of reported routine GLIA data indicated that data quality was weak, as the data quality assurance processes were not institutionalized or implemented as planned in the GLIA M&E Implementation Manual. Only a couple of data audits/spot checks were conducted, and the audit did not ‘trace and verify’ the reported data against source documents at the GLIA implementation sites. Moreover, reporting was not done on a quarterly basis as planned. Finally, the existence of multiple M&E frameworks reflects the overall weaknesses which persisted throughout the project. 44. Notwithstanding these shortcomings on M&E, the project has allowed member countries to operationalize their organizational structures while at the same time involving the countries in implementing activities at the grass root levels, and targeting vulnerable groups which were otherwise left out from national HIV prevention activities. Furthermore, the GLIA Secretariat, with support from GAMET completed an HIV epidemiological and response analysis for the 6 countries. Based on this analysis, GLIA formulated its 2008-2012 Strategic Plan that included priority activities to be scaled up for mobile and at-risk populations. 49 Table 7: Baseline data, targets, midterm, and end of project results indicators for component 4 Intermediate Results Results Indicators Baseline Target Mid End of Term project The GLIA Convention The GLIA has a functioning None Yes No No has been operationalized M&E system where data by the GLIA secretariat, flows to a central level, is who is able to plan, captured, analyzed and implement, monitor and disseminated to stakeholders* evaluate activities that The GLIA Secretariat has 0 8 7 6 will lead to the appointed all 8 core staff achievement o f the members** GLIA mission *This indicator is in the DGA **This indicator has been replaced by an indicator on fiduciary management in the DGA. 50 Annex 3. Economic and Financial Analysis 1. No specific economic analysis was carried out as part of the project preparation for the GLIA Support Project or the ICR mission. The economic rationale for the proposed project was based on the Multi-Country HIV/AIDS Program for the Africa Region (Report No.20727-AFR) which included a cost-benefit analysis of HIV/AIDS interventions and an assessment of the overall impact of HIV/AIDS on the economy. 2. The main findings of the assessment were as follows: (i) HIV/AIDS has negative effects on productivity, domestic savings and overall economic growth; (ii) HIV/AIDS increases health costs and runs the risk of crowding out other key public health programs, such as immunization, maternal and child health, malaria and parasitic diseases; (iii) care and treatment of AIDS patients imposes high costs on families and reduces their earning power; and (iv) family coping strategies may result in children abandoning school prematurely to look after relatives or families cutting other health or social expenditures to unacceptable levels. 51 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Pamphile Kantabaze Senior Operations Officer AFTHE Otieno Ayany Financial Management Specialist AFTFM Anne Marie Bodo Pharmaceutical Specialist, Consultant AFTHE Frode Davanger Operations Officer AFCRI Sameena Dost Senior Counsel LEGES Diego Garrido Martin E T Consultant AFTRL Marelize Gorgens Monitoring & Evaluation Specialist HDNGA Fabrice Houdart Country Officer ECCU8 Wacuka W. Ikua Senior Operations Officer AFTHE Sylvie Ingabire Team Assistant AFMRW Chantal Kajangwe Procurement Specialist AFTPC Antoinette Kamanzi Procurement Assistant AFMRW Alex Kamurase Social Protection Specialist AFTSP Mohammad Javed Program Assistant AFTHE Karimullah Carl Adam Per Lagerstedt Senior Health Specialist AFTH1 - HIS Luc Lapointe Procurement Specialist, Consultant AFTPC Emmanuel G. Malangalila Senior Health Specialist, Consultant AFTHE Jean-Pierre Manshande Senior Health Specialist, Consultant HDNGA Myrina D. McCullough Consultant AFTSP Montserrat Meiro-Lorenzo Senior Public Health Spec. AFTHE Michael Mills Consultant AFTHE Joseph Kizito Mubiru Senior Financial Management Specialist LCSFM Shimwaayi Muntemba Social Dev. Specialist, Consultant QLP Deo Ndikumana Senior Operations Officer AFCRI Prosper Nindorera Senior Procurement Specialist AFTPC Seraphine Nsabimana E T Temporary AFRVP John Nyaga Senior Financial Management Specialist EAPFM Peter Okwero Senior Health Specialist AFTHE Clarette Rwagatore Team Assistant AFMBI Richard M. Seifman Consultant AFTHE Souleymane Sow Consultant AFTH3 - HIS Julie Tumbo Odhiambo Consultant HDNGA 52 b) ICR Names Title Unit Miriam Schneidman Lead Health Specialist, TTL AFTHE Enias Baganizi Senior Health Specialist AFTHE Jean Jacques Frere Senior Health Specialist AFTHE G.N.V. Ramana Lead Health Specialist AFTHE Alexander Ritter Consultant AFTHE Yvette Atkins Senior Program Assistant AFTHE Otieno Ayany Financial Management Specialist AFTFM Chantal Kajangwe Procurement Specialist AFTPC Antoinette Kamanzi Procurement Assistant AFMRW Josiane Niyonkuru Acting Team Assistant AFMRW Lucy Musira Team Assistant AFCE2 (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY03 2 36.01 FY04 37 271.04 FY05 51 301.01 FY06 0.00 FY07 0.00 FY08 0.00 Total: 90 608.06 Supervision/ICR FY03 0.00 FY04 0.00 FY05 3 4.45 FY06 18 110.44 FY07 18 54.44 FY08 42 134.15 FY09 31 125.8 FY10 35 110.34 Total: 147 539.62 53 (c) ISR Performance Ratings ISR Number 1 2 3 4 5 6 7 8 9 10 11 Date 5/25/05 12/28/05 6/29/06 2/14/07 7/13/07 3/3/08 12/27/08 6/30/09 12/30/09 6/29/10 12/29/10 Progress toward achievement of PDO: S S S MU MS MS MS MS MU MU U Project Component  Refugees S S S MS MS S S S MS MS MS HIV/AIDS S MS S MS MS MS S MS MS MS MS Health Sector  collaboration S S S MS MS MS S MS MS MS MS Management and  capacity building S U S MS MS S S MU MU MU MU Summary  Implementation  Development  Performance Ratings Objectives Rating S S S MU MS MS MS MS MU MU U Overall  Implementation  Progress S MS S MU MS MS MS MU MS MS MS Financial Management S S S MS MS MS S MS MS MS MU Project Management S U S MU MS S S MS MS MS MS Counterpart Funding S S S MS MS MS MS MU MU MU MS Procurement S S S MU MS MS S MS MS MS MS M&E S S S MU MS MS MS MU MU MU U Overall Safeguard  Overall Safeguard  Compliance Compliance S S S S S S S S S S S Environmental  Assessment (OD 4.01) S S S S S S S S S S S Involumntary  Resttlement (OP 4.12) S S S S S S S S S S S 54 Annex 5. Beneficiary Survey Results While no formal Beneficiary Survey was conducted views and attitudes of GLIA stakeholders and beneficiaries were often solicited and documented, using the Most Significant Change methodology, whereby individuals or groups were asked how the project activities impacted their lives. While these individual stories are anecdotal in nature a few stories are presented below as they give an overall sense of how beneficiaries and stakeholders viewed the project and how they benefited from the activities supported by the GLIA Support Project.  Empowering truck-drivers to obtain HIV testing. During a SALT visit to Katma it was discovered that truck drivers were willing to have an HIV test but they were not available at times which were convenient to them. As noted by one truck driver, “The health center where tests are available is only four 4 kms away, but it is only open from 8:00 a.m to 17:00 p.m, while we never get to Katma (Uganda) before 19:00 p.m.” Soon after the community group with the health center came up with a solution of establishing mobile teams to serve truck drivers in the evenings. A team from the health center now comes every Thursday night to Katma with tests kits. [Story told by M. MUGENI Ouma Naphtal, from the Uganda Truck Drivers Network].  Coming out on HIV/AIDS--two visits and a small revolution. While visiting Kibaigwa, an area known to have high infection rates, a GLIA team asked locals if they knew anyone living with AIDS. They all answered they did not. But when GLIA facilitators living with AIDS gave their testimony, people opened up. A mother of two children, who works in a NGO dealing with AIDS, said that she was HIV positive. After the meeting, truck-drivers introduced the team to some sex workers in a nearby restaurant. During a second visit, the team found out that these people had created a group of people living with HIV: 70 people had openly declared their status and had joined. One sex worker said: “when you came, last time, the way you listened and talked to us stimulated us. You made us believe that we can defeat the virus and stop spreading it”.  Involving local communities and actors. Local communities, National AIDS Commissions, and individuals started taking the issue of HIV in their own hands. In Tanzania, before establishing the Wellness Centers, the steering committee held a meeting with the NAC. The result was that NAC sponsored visits to 3 sites and connected the steering committee to the district government which discussed the location of the centers. Local government allocated plots for the centers. By involving local authorities the “response is sustainable”, reported one stakeholder. In Kenya, the NAC brought together different actors to mobilize resources to launch Wellness Centers. The local governments donated land, a transport company provided containers and the NAC coordinated the set up. Kenyan truck drivers invited Ministry of Health officials to commemorate their colleague who died of AIDS. The long term problem of lack of treatment for STIs, and lack of 55 access to ART and VCT services was presented to them and mobile services were subsequently established. In North-East Uganda, SALT visits stimulated some 300 sex workers to launch an association to stop the spread of HIV in their community, after doing their self-assessment.  Promoting regional collaboration. The MSC methodology was also applied to determine view of stakeholders on the extent to which collaboration around HIV/AIDS was a vehicle for broader engagement and increased understanding between countries. In the words of some stakeholders, policymakers "did everything in their power so that the GLIA would go ahead, even overcoming political differences. The technical experts broke through these political barriers to really advance the initiative. Another stakeholder said, “When the parliaments, the Cabinets, all ratified the convention in six months, we were shocked! ….For me it was like a transformation… Because even getting three countries to work together at the highest political level, at the Ministerial level, would have been difficult two or three years ago, even impossible. So when they all ratified the convention, people were saying, “this is a real success, this is a major step forward.” One Bank stakeholder emphasized the strong commitment of the GLIA ministers who scrutinized the GLIA Convention carefully and who “wanted to make sure this would be a success… they clearly took ownership at every step of the process, which takes time, but which is also crucial for such an initiative”. 56 Annex 6. Stakeholder Workshop Report and Results N/A 57 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Summary of Borrower’s ICR 1. This GLIA Completion Summary Report includes: (i) an assessment of the project objectives, design, implementation, and outcomes, (ii) assessment of the borrower`s performance and lessons learned, (iii) performance of the World Bank as well as other partners who were involved in the design and implementation of the GLIA operation, (iv) description of the proposed arrangements for sustaining the achievements of the Project. Assessment of the operation’s objectives, design, implementation, and outcomes 2. The development objectives of the Project were and remain relevant. The regional aspects of the project are unique. The design was consistent with the objectives. The GLIA Support Project was an exciting experience and at the same time challenging. The Project was the first major regional activity of the GLIA Secretariat, therefore it helped the institution to pursue its mandate, and add value to national investments. However, for some project components (2 &3), the design did not take into account sufficiently the variable national and capacities and procedures. 3. Regarding implementation, GLIA developed a generic Operations Manual which specified clearly the implementation modalities. In spite of having clear procedures and guidelines a number of implementation problems emerged, notably difficulties in ensuring timely transfer of funds, and difficulties in addressing the rapidly changing needs of refugees and returnees. In terms of timeliness of disbursements, problems emerged due to the lengthy and cumbersome procedures for fund transfers, and the different funding cycles in each country. As a result of the problems in the initial years, implementation was concentrated in the final years following the Mid-Term Review with some activities completed only towards the end. 4. With respect to the Monitoring & Evaluation System, the Project faced serious shortcomings. Lack of ownership of the M&E system was one of the key impediments; moreover the duration of the Project was too short to develop and sustain broad based ownership by all the NACs. 5. Regarding operational experience, the GLIA Secretariat gained considerable experience with the Bank-funded Project. It added value to GLIA as an organization. GLIA gained operational experience in project design, proposal development, capacity strengthening, financial management, M&E, advocacy, research, coordination and oversight of regional and national projects. The Project was viewed as a pilot operation that would lead to a follow up WBSP2 operation. 6. The outcome of the operation against its objectives is generally satisfactory. GLIA has initiated key activities that are regional in nature and considered high-value in 58 the HIV/AIDS response. Though achievements are registered, impact could not be measured due to the nature and length of the project. Evaluation of the borrower’s own performance during the preparation and implementation of the operation, with special emphasis on lessons learned. 7. The performance of the borrower was generally satisfactory. GLIA’s institutional arrangements were well designed. Tremendous results have been attained despite the weaknesses noted above. GLIA as an organization was, however, too dependent on the WB financing. Nevertheless, the ownership and commitment of countries to implement the project were demonstrated by: i) Issuing letters of commitment to participate in a regional programme. ii) Signing/ratifying the Convention establishing GLIA. iii) Setting up governance structures, and recruiting staff for the GSP Coordination. iii) Participating actively in the GLIA governance structures (i.e. Council of Ministers, Executive Committee). (iv) Meeting the financial contributions to the GLIA Secretariat with all Member States releasing their financial contributions by project completion. Lessons Learned 8. Regarding lessons learned, delivery of services on a regional level was challenging, especially in the following areas: (i) establishment of effective regional mechanisms of coordination, (ii) effective use of the harmonized HIV/AIDS related protocols, (iii) operationalization of a regional network of Long Distance Truck Drivers, (iv) sustained collaboration between various actors (NACs, Secretariat, World Bank, MOH, Intermediary Agencies, other implementing agencies). 9. The prospects for sustainability are challenged by the regional political context (i.e. articulation with the EAC, with 5 out of the 6 GLIA member states represented in the EAC), the collaboration with other regional projects, and linkages with national programmes. Evaluation of the performance of the Bank and other partners during the preparation and implementation of the operation, including the effectiveness of their relationships 10. The Bank played a key role during the Project design, preparation, and implementation. Its performance was satisfactory. 11. During Project preparation: The WB discussions with GLIA countries started in 2002 at a very critical time for the organization due to lack of resources to run a programme related to its mandate, five years after the creation of GLIA. The Bank provided well appreciated support to comply with the conditions of effectiveness. 59 12. Regarding the Project Implementation: The WB also provided technical support to GLIA during the implementation phase through regular supervision missions; participated in the GLIA statutory meetings (COM, EC); provided quality assurance; supported the process of mobilizing additional resources; provided continuous support on procurement procedures; and assisted with Monitoring & Evaluation. 13. The UNAIDS played a key role of facilitation during the preparation phase and provided technical support during design and implementation by organizing meetings, developing key project documents, and participating in supervision missions, mainly on M&E aspects. 14. The UNHCR signed a MOU with the Council of Ministers and Management Contract with the GLIA Secretariat. UNHCR implemented project activities through strong NGOs which functioned as sub implementers. The performance of sub implementers was satisfactory despite some delays in carrying out the UNHCR activities. 15. Project Results Framework The M&E framework was weak, not only in terms of baseline, targets, and PDO data but also the support provided to GLIA to adjust the framework and implement it. GLIA was also not capacitated to handle professionally the M&E framework. The lesson to draw is that there was a joint responsibility on this very important component of the project. Proposed arrangements for sustaining achievements of the Project: 16. To ensure the sustainability of the Project achievements, the following actions are being considered:  NACs to take over some project activities, particularly those in the surrounding communities and in some countries (e.g. Rwanda) in refugee camps as well.  GLIA Secretariat to provide support for proposal development.  Member states to continue paying their membership fees and the Rwandan government to continue providing office space for the Secretariat.  NAC Focal Point concept to be realigned to the current national and regional context.  GLIA Secretariat to be empowered/supported to play coordination and convening roles.  NACs to use sub-implementers for projects so they can provide oversight functions.  GLIA to be affiliated to a solid regional entity (i.e. EAC). Comments on ICR 17. Detailed comments were received from several National AIDS Commissions as summarized below. 18. The National AIDS Commission of Rwanda emphasized: (i) GLIA’s uniqueness as an institution in the sub region with a strong mandate to fight HIV/AIDS; (ii) GLIA’s 60 outstanding experience gained from the implementation of the Bank-funded project; (iii) pilot nature of the 5-year project which yielded impressive results; (iv) positive impact of project which is highly visible among beneficiaries, especially in terms of mitigation measures which addressed the underlying vulnerability of those affected by HIV/AIDS; (v) GLIA’s ability to bring together PLWHA and long distance truckers who typically would not have interacted; (vi) need to consolidate achievements of civil society groups, such as the long distance truck drivers and the drivers’ wives associations; (vii) need for broader consultation with the Council of Ministers and the Executive Committee at project completion to assist GLIA to develop an exit strategy, including mobilizing additional resources; and (viii) importance of mobilizing additional resources from the Bank in order to leverage other sources of funding. 19. The Tanzanian National AIDS Commission underscored two issues related to sustainability which are part of the lessons learned highlighted in this report: (i) early and sustained involvement of a wide range of stakeholders (e.g. central government, local government authorities, civil society groups) improved monitoring, increased ownership of the GLIA activities, and enhanced prospects for sustainability; and (ii) strong linkages between the GLIA-funded activities and local health services, including use of local staff in implementing the project activities, has led to the inclusion of these activities into local health plans and budgets, hence promoting sustainability. 20. The Ugandan National AIDS Commission underscored five key points: (i) a considerable amount of time was allocated to intermediary agencies under component two, leaving the networks insufficient time and resources to implement their activities; (ii) GLIA’s survival relies heavily on donor funding given modest country contributions, which places the institution in a precarious situation; (iii) EAC health desk has agreed to have GLIA as one of its implementing partners to ensure continuity of the Bank-funded activities, particularly with respect to harmonization of protocols and operations of the Wellness Centers; (iv) key planned activities were not implemented that could have helped to better measure and discuss the performance of the project, including the Most Significant Change methodology, and a project completion workshop to take stock of achievements; and (v) monitoring and evaluation was lacking and the structures and procedures for M&E were not effectively utilized. 61 Annex 8. Comments of Co financiers and Other Partners/Stakeholders NA 62 Annex 9. List of Supporting Documents  Project Concept Note, [March, 2005]  Project Appraisal Document, [February, 2005]  Six Country CASes (Burundi, Kenya, Rwanda, Uganda, DRC, Tanzania)  GLIA Support Project Work Plan  Aide Memoire [December, 2008]  ISRs (1-11)  QAG Report [December, 2008]  QER Minutes [August, 2004]  Development Grant Agreement [April, 2005]  IDA15 Mid-Term Review [October, 2009]  GLIA Stories of Change  GLIA Progress Reports  GLIA External Audit Reports [2009, 2010]  GLIA Country Behavioral Surveillance Surveys (BSS) 1 and 2  UNHCR Technical Reports [2007-2010]  GLIA Annual Reports [2007-2010]  GLIA Mombasa Workshop Final Report [August, 2010]  GLIA Nairobi Information Sharing Workshop Report [November, 2010] 63 IBRD 35794 15˚E 20˚E 30˚E 35˚E 40˚E GREAT LAKES REGION This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank RÉGION DES GRANDS LACS SUDAN Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. CENTRAL AFRICAN REPUBLIC SOUDAN RÉPUBLIQUE CENTRAFRICAINE ETHIOPIA 5˚N ÉTHIOPIE 5˚N Lake CAMEROON Rudolf CAMEROUN Area of Zone de la carte Map SOMALIA SOMALIE UGANDA OUGANDA KENYA 0˚ KÉNYA 0˚ Lake GABON Victoria RWANDA DEM. REP. CONGO OF CONGO RÉP. DÉM. BURUNDI DU CONGO 5˚S Lake 5˚S Tanganika TANZANIA TANZANIE INDIAN OCEAN Lake Mweru 10˚S AT L A N T I C ANGOLA MALAWI OCEAN Lake INTERNATIONAL BOUNDARIES FRONTIÈRES INTERNATIONALES Malawi ZAMBIA ZAMBIE MOZAMBIQUE 15˚S 0 100 200 300 400 Kilometers 15˚S 0 100 200 300 400 Miles ZIMBABWE 15˚E 20˚E 25˚E 30˚E 35˚E 40˚E NOVEMBER 2007