Document of The World Bank Report No: ICR2268 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-34160 IDA-3416A) ON A GRANT IN THE AMOUNT OF SDR 20 MILLION (US$30 MILLION EQUIVALENT) TO THE FEDERAL DEMOCRATIC REPUBLIC OF ETHIOPIA FOR A SECOND MULTI-SECTORAL HIV/AIDS PROJECT March, 29, 2012 AFTHE Country Department AFCE3 Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective 09/31/2011) Currency Unit= Birr Birr 1.00= 0.058US$ US$ 1.00= 17.30 Fiscal Year July 1 - June 30 ABBREVIATIONS AND ACRONYMS ACTAfrica AIDS Campaign Team for Africa IFR Interim Financial Report AIDS Acquired Immune Deficiency M&E Monitoring and Evaluation Syndrome ARV Antiretroviral (drugs) MAP Multi-country HIV/AIDS Program BCC Behavior Change Communication MARP Most at Risk Population BPR Business Process Re-Engineering MDG Millennium Development Goals CBO Community-Based Organization MSM Men who have Sex with Men CIS Community Information System NGO Non-Governmental Organization DHS Demographic and Health Survey PAD Project Appraisal Document DO Development Objective PDO Project Development Objective EAF Emergency HIV/AIDS Fund PCU Project Coordination Unit EMSAP Ethiopia Multisectoral HIV/AIDS PEPFAR President's Emergency Plan for AIDS Relief Project (US) EFY Ethiopian Fiscal Year PIM Project Implementation Manual FSW Female Sex Worker PLWHA People Living with HIV/AIDS (F)HAPCO (Federal) HIV/AIDS Prevention SoE Statement of Expenses and Control Office GFATM Global Fund for AIDS, SPM Strategic Plan for Management Tuberculosis and Malaria HCT HIV Counseling and Testing SW Sex Worker HMIS Health Management Information STI Sexually Transmitted Infection System IDA International Development TTL Task Team Leader Association IEC Information, Education and VCT Voluntary Counseling and Testing Communication Vice President: Obiageli Katryn Ezekwesili Guang Country Director: Zhe Chen Sector Manager: Jean J. De St Antoine Project Team Leader: Feng Zhao ICR Team Leader: Mohamed Ali Kamil Ethiopia Multi-Sectoral HIV/AIDS Project II Table of Contents A. Basic Information ...................................................................................................................... iii B. Key Dates .................................................................................................................................. iii C. Ratings Summary ...................................................................................................................... iii D. Sector and Theme Codes........................................................................................................... iv E. Bank Staff ................................................................................................................................... v F. Results Framework Analysis ...................................................................................................... v G. Ratings of Project Performance in ISRs .................................................................................. xii H. Restructuring (if any) ............................................................................................................... xii I. Disbursement Profile ............................................................................................................... xiii 1. Project Context, Development Objectives and Design ............................................................... 1 2. Key Factors Affecting Implementation and Outcomes ............................................................... 4 3. Assessment of Outcomes ............................................................................................................ 8 4. Assessment of Risk to Development Outcome ......................................................................... 15 5. Assessment of Bank and Borrower Performance...................................................................... 15 6. Lessons Learned ........................................................................................................................ 17 Annex 1. Project Costs and Financing .......................................................................................... 19 Annex 2. Outputs by Component................................................................................................. 22 Annex 3. Economic and Financial Analysis (including assumptions in the analysis) .................. 25 Annex 4. Bank Lending and Implementation Support/Supervision Processes............................. 28 Annex 5. Summary of Borrower's ICR ......................................................................................... 31 Annex 6. List of Supporting Documents ...................................................................................... 37 Annex 7. Methodology, Results, and Conclusions of the Special KAP Survey Commissioned by HAPCO on HIV/AIDS. ........................................................................................................... 38 IBRD Map 35214 ii A. Basic Information Second Multi-sectoral Country: Ethiopia Project Name: HIV/AIDS Project Project ID: P098031 L/C/TF Number(s): IDA-H2790 ICR Date: 03/29/2012 ICR Type: Core ICR FEDERAL DEMOCRATIC Lending Instrument: SIL Borrower: REP. OF ETHIOPIA Original Total XDR 20.00M Disbursed Amount: XDR 16.81M Commitment: Revised Amount: XDR 16.81M Environmental Category: B Implementing Agencies: Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 03/17/2006 Effectiveness: 06/28/2007 06/28/2007 Appraisal: 01/11/2007 Restructuring(s): Approval: 03/08/2007 Mid-term Review: 03/21/2009 03/30/2009 Closing: 06/30/2010 09/30/2011 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Satisfactory Risk to Development Outcome: High iii Bank Performance: Moderately Satisfactory Borrower Performance: Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Moderately Satisfactory Government: Moderately Satisfactory Implementing Quality of Supervision: Satisfactory Moderately Satisfactory Agency/Agencies: Overall Bank Overall Borrower Moderately Satisfactory Moderately Satisfactory Performance: Performance: C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments (if Indicators Rating Performance any) Potential Problem Project No Quality at Entry (QEA): None at any time (Yes/No): Problem Project at any Quality of Supervision Yes None time (Yes/No): (QSA): DO rating before Moderately Closing/Inactive status: Satisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) General public administration sector 24 15 Health 26 80 iv Other social services 50 5 Theme Code (as % of total Bank financing) Gender 17 HIV/AIDS 33 Other social development 17 Participation and civic engagement 16 Population and reproductive health 17 E. Bank Staff Positions At ICR At Approval Vice President: Obiageli Katryn Ezekwesili Hartwig Schafer Country Director: Guang Zhe Chen Ishac Diwan Sector Manager: Jean J. De St Antoine Laura Frigenti Project Team Leader: Mohamed Ali Kamil Gebreselassie Okubagzhi ICR Team Leader: Mohamed Ali Kamil ICR Primary Author: Michael O'Dwyer F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The main objectives are to: a) increase access to prevention services for youth, in particular females aged 15-24, and other most-at-risk groups; and b) sustain access to care and support for PLWHA and orphans undertaken in EMSAP I. v Revised Project Development Objectives (as approved by original approving authority) (a) PDO Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Percentage of young people aged 15-24 who both correctly identify ways of Indicator 1 : preventing the sexual transmission of HIV and who reject major misconceptions about HIV transmission Value F 79% F 67.3%, M 54.3% >75% for both quantitative or M 83% (DHS 2005) sex Qualitative) (FHAPCO 2011) Date achieved 04/13/2005 09/30/2011 09/15/2011 Comments The actual value achieved is drawn from a survey conducted by FHAPCO (Survey on (incl. % the Status of Sexual Behavior, Knowledge about HIV/AIDS and Condom Utilization in Thirty-Five Woredas, FHAPCO, Government of Ethiopia, 2011) achievement) Percentage of young women and men aged 15-24 reporting the use of a condom during (last) sexual intercourse with a non-regular partner (of those reporting sexual Indicator 2 : intercourse with a non-regular partner in the last 12 months Value F Not measured F 88% F >50% quantitative or M 49.5% M 88% M >50% Qualitative) (DHS 2005) (FHAPCO 2011) Date achieved 04/13/2005 09/30/2011 09/15/2011 Comments The actual value achieved is drawn from a survey conducted by FHAPCO (Survey on the Status of Sexual Behavior, Knowledge about HIV/AIDS and Condom Utilization in (incl. % Thirty-Five Woredas, FHAPCO, Government of Ethiopia, 2011) vi achievement) Increased percentage of sex workers who report using a condom with their most Indicator 3 : recent client (of those surveyed having sex with any clients in the last 12 months) Value 96.6% 98% quantitative or >98% (BSS 2002) FHAPCO 2011) Qualitative) Date achieved 05/21/2002 09/30/2011 09/15/2011 Comments The actual value achieved is drawn from a survey conducted by FHAPCO (Survey on (incl. % the Status of Sexual Behavior, Knowledge about HIV/AIDS and Condom Utilization in Thirty-Five Woredas, FHAPCO, Government of Ethiopia, 2011) achievement) Number of people aged 15 and older undergoing voluntary HIV counseling and Indicator 4 : testing (depending on expected changes to the National Policy, otherwise 18 and older) Value 564,000 9,448,880 quantitative or 700,000 (FHAPCO) (FHAPCO) Qualitative) Date achieved 06/18/2005 09/30/2011 09/15/2011 Comments Target exceeded: actual achievement is far higher than the project target, reflecting (incl. % an massive government push to expand testing (Millenium Development Campaign). achievement) Percentage of woredas that are supporting effective PLWHA groups and associations Indicator 5 : and implementing community programs to support orphans Value New indicator 98% quantitative or 95% No baseline (GoE ICR) Qualitative) Date achieved 06/28/2007 09/30/2011 09/15/2011 Comments Fully achieved (incl. % vii achievement) Establishment of an HIV/AIDS information and M&E system that supports improved Indicator 6 : policy and program decision making. Value New indicator quantitative or 100% 100% No baseline Qualitative) Date achieved 06/28/2007 09/30/2011 09/15/2011 Comments At project completion an M&E system is established which provides comprehensive (incl. % information on the HIV/AIDS response evidenced by availability of manuals and guidelines, annual and quarterly M&E reports. This information informed developm achievement) (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Percentage of project woredas implementing HIV/AIDS health and non-health plans Indicator 1 : in the last 12 months Value Number of project (quantitative 100% 100% woredas in pilot program or Qualitative) Date achieved 02/15/2007 09/30/2011 09/15/2011 Comments All project woredas implemented health and non-health HIV/AIDS plans during the (incl. % project period. achievement) Percentage of woredas submitting completed program monitoring forms for Indicator 2 : community-based indicators to HAPCO in the last 12 months viii Value (quantitative Forms being developed 100% 100% or Qualitative) Date achieved 02/15/2007 09/30/2011 09/15/2011 Comments All project woredas submitted these report to their respective regions or zones and (incl. % through them to FHAPCO in 2011. achievement) Ability of M&E system to collect data and measure: (i)ART treatment (Men and women); and Indicator 3 : (ii) HIV-infected pregnant women receiving a complete course of antiretroviral prophylaxis to reduce the r Value ART 42195 (quantitative YES YES PMTCT 2036 or Qualitative) Date achieved 02/15/2007 09/30/2011 09/15/2011 Comments The M&E system completely captures information on ART and PMTCT programs (incl. % nationally. achievement) Number of priority non-health public sector actions plans targeting most-at-risk Indicator 4 : groups Value New measure for M&E (quantitative 100% 100% system or Qualitative) Date achieved 02/15/2007 09/30/2011 09/15/2011 Comments 17 Major sector offices at federal level and additional different sector bureaus at (incl. % regional and woreda level accessed fund at each level and implemented HIV prevention interventions targeting most at risk groups. achievement) ix Percentage of funds allocated to support most at-risk groups by public sector Indicator 5 : agencies Value (quantitative 0 70% 100% or Qualitative) Date achieved 02/15/2007 09/30/2011 09/15/2011 Comments (incl. % Target fully achieved. achievement) Indicator 6 : Increased number of male and female condoms distributed nationally Value (quantitative 80 million 140 million 174 million or Qualitative) Date achieved 06/18/2005 09/30/2011 09/15/2011 Comments Target exceeded. Refers to male condoms distributed (no information available for (incl. % female condoms). achievement) Indicator 7 : Percentage of grant financing to NGOs/CBOs through woredas in the last 12 months Value 29.7% (quantitative 45% 15% NA (FHAPCO) or Qualitative) Date achieved 02/15/2007 09/30/2011 09/30/2011 09/15/2011 Comments (incl. % Most recent report not yet available achievement) Indicator 8 : At least 75% percent of grants finance targeted prevention related services Value 70% 75% 75% x (quantitative (FHAPCO) or Qualitative) Date achieved 02/15/2006 09/30/2011 09/15/2011 Comments (incl. % achievement) All participating woredas have implemented their approved action plans for Indicator 9 : programs targeting most at- risk groups within three years Value (quantitative New measure for M&E 95% 100% or Qualitative) Date achieved 02/15/2007 09/30/2011 09/15/2011 Comments All 417 project woredas implemented HIV prevention programs focusing on most at (incl. % risk groups achievement) Number of PLWHA and orphans who have received care and support through Indicator 10 : woredas Value PLWHA 35,000 PLWHA 103,659 PLWHA 100,000 (quantitative Orphans 80,000 Orphans 354,660 Orphans 200,000 or Qualitative) (FHAPCO) (FHAPCO) Date achieved 06/05/2005 09/30/2011 09/15/2011 Comments (incl. % achievement) xi G. Ratings of Project Performance in ISRs Date ISR Actual Disbursements No. DO IP Archived (USD millions) 1 12/03/2007 Satisfactory Satisfactory 4.00 2 01/25/2008 Satisfactory Satisfactory 4.22 3 06/28/2008 Satisfactory Moderately Unsatisfactory 6.38 4 02/20/2009 Moderately Unsatisfactory Moderately Unsatisfactory 11.91 5 06/30/2009 Moderately Satisfactory Moderately Satisfactory 13.36 6 12/20/2009 Moderately Satisfactory Moderately Satisfactory 15.17 7 06/23/2010 Moderately Satisfactory Moderately Satisfactory 18.97 8 03/26/2011 Moderately Satisfactory Moderately Satisfactory 24.32 9 10/30/2011 Moderately Satisfactory Moderately Satisfactory 26.00 H. Restructuring (if any) Not Applicable xii I. Disbursement Profile xiii 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal 1. Country and Sector Background: At the time of project appraisal Ethiopia had a population of over 72 million, 45 % of whom were under 15 years old. The first AIDS case was reported in 1986 and by 2005 adult HIV prevalence had reached 1.4% (DHS+, 2005) – 5.5% urban, 0.7% rural. In 2007, an estimated 1.32 million people were infected, with over 130,000 new infections a year. The country had a mixed epidemic with HIV transmission among both Most At-Risk Populations (MARPs), particularly Sex Workers (SWs) and their clients, and the general population. Findings from an epidemiological synthesis study, carried out with Bank support, highlighted the importance of addressing the epidemic in MARPs, and were available to inform project design1. Life expectancy was falling as a result of the epidemic, and was expected to drop to 50 years by 2010 (from a previous estimate of 59). 2. Economic and social context: The economy is predominantly agricultural, 80% of the population is rural (60 million), and in 2005 around 39% of Ethiopians were living below the poverty line. AIDS had become the leading cause of mortality in 15-49 year olds (43% of all adult deaths), and was seen as a major threat to economic growth and poverty alleviation. Major factors thought to contribute to the epidemic included poverty, illiteracy, widespread transactional sex, gender disparity, population movement and harmful traditional practices such as Female Genital Mutilation. There was a significant difference in infection rates between men and women (seven times higher in 15-19 year old girls). 3. Rationale for Bank assistance: This project built upon successful completion of the first Ethiopian Multi-Sectoral HIV/AIDS Project (EMSAP1) launched in 2001 and closed in December 2006. There was a recognized need to further scale up appropriate interventions for a number of years, and to increase the focus on MARPs, in order to see declines in prevalence and incidence and to reduce the impact on life expectancy and economic growth. EMSAP1 had effectively supported general awareness of HIV and AIDS, and some change in sexual behavior (such as condom use in high risk settings) but not at a sufficiently high level. 4. EMSAP1 had played a key role in building an effective national response and creating an enabling environment for decentralized HIV activities. Financial resources for the response had increased substantially by 2007 with significant amounts of Global Fund and PEPFAR funding (GF and PEPFAR commitments for 2007 totaled $250 million, of which $130m was for ART and $109m for OVC support, condoms and PMTCT). However, there remained a crucial gap in support for comprehensive prevention programs2. Bank support was needed to partially cover this gap, and improve national program delivery and utilization of all resources. The Bank therefore decided to continue to finance prevention interventions and support to improve public and private sector institutional capacity. Particular emphasis was given to targeting communities not already reached and MARPs, and helping government implement prioritized, targeted and focused interventions. 1 HIV/AIDS in Ethiopia: An Epidemiological Synthesis; Ethiopia HIV/AIDS Prevention & Control Office and Global HIV/AIDS Monitoring and Evaluation Team (GAMET), The Global HIV/AIDS Program, World Bank, April 2008 2 FHAPCO estimated that US$19.5 billion would be needed to reach the HIV/AIDS MDG by 2015. 5. The project was designed in a context of strong government commitment, based on a multisectoral Strategic Plan for Management (SPM - revised in 2004), supported by all development partners. Federal and Regional HIV/AIDS Prevention and Control Offices ((F)HAPCOs) were established to coordinate and guide the national response with oversight provided by a Management Board chaired by the Minister of Health. There was a good working relationship between FHAPCO and the MoH (in 2005, the MoH integrated its HIV/AIDS and other STIs Prevention and Control Unit into HAPCO, further strengthening overall national program coordination within one institution). The country complied with the ―Multi-country AIDS Programs repeater requirements‖3 with the exception of one comprehensive M&E system. EMSAP1 had, however, helped establish an M&E unit in FHAPCO, and it was stated that the project would support improvements to the system. 1.2 Original Project Development Objectives (PDO) and Key Indicators 6. The project development objectives were to: (i) increase access to prevention services for youth, in particular females aged 15-24, and other most-at-risk groups and (ii) sustain access to care and support for PLWHA and orphans undertaken in EMSAP1. 7. PDO indicators were: (i) increased percentage of young people aged 15-24 who both correctly identify ways of preventing the sexual transmission of HIV and who reject major misconceptions about HIV transmission; (ii) increased percentage of young women and men aged 15-24 reporting the use of a condom during (last) sexual intercourse with a non-regular partner (of those reporting sexual intercourse with a non-regular partner in the last 12 months); (iii) increased percentage of sex workers who report using a condom with their most recent client (of those surveyed having sex with any clients in the last 12 months); (iv) increased percentage of people aged 15 and older undergoing voluntary HIV counseling and testing (depending on expected changes to the National Policy, otherwise 18 and older); (v) increased percentage of woredas that are supporting effective PLWHA groups and associations and implementing community programs to support orphans; and (vi) establishment of an HIV/AIDS information and M&E system that supports improved policy and program decision making. 1.3 Revised PDO and Key Indicators (as approved by original approving authority), and reasons/justification 8. The original project objectives were not revised. 3 (i) One national HIV/AIDS Strategy and one national HIV/AIDS authority; (ii) a comprehensive national M&E system for HIV/AIDS; (iii) willingness to use simplified implementation procedures; and (iv) willingness to integrate performance considerations into fund allocation and disbursement decisions. 2 1.4 Main Beneficiaries 9. The project targeted youth aged 15-24 years, particularly young women (who had significantly higher rates of infection than young men), and other MARPs, especially SWs and their clients, farmers, migrant workers, and military personnel. It was stated that there would be a focus on rural and peri-urban areas and specific ―hot-spots‖. 10. It was noted at the time that there was little information (in what was a difficult cultural, political and religious context) about Men who have Sex with Men (MSM). There were plans to use PEPFAR funds to carry out an assessment of the HIV status of this group. Since project design, some small scale studies have been published: - a small ethnographic study published in 2009 suggested that MSM are an at-risk population needing policy and program attention4; - another study, quoted in an MoH/UNFPA desk review (2009) shared these concerns pointing out that ―[a]part from the high-risk behavior of MSM, … [there is] the possibility of cross bridging of the HIV epidemic to the heterosexual community because of dual sexual orientation of some gay people (bisexuality)5;‖ - the Epidemiological Synthesis [referred to above] noted that ―… this population is essentially unstudied in Ethiopia. After many years of denial, evidence is emerging about MSM populations in other East African countries, so it is likely that they exist in Ethiopia as well, and are probably at even greater risk because of stigma and lack of access to services‖. 1.5 Original Components 11. Component 1: National Program Coordination and Institutional Strengthening (Appraisal estimate US$2.5 million): The objective of this component was to further improve the coordination and delivery of the national HIV/AIDS strategy and action plan at all levels more effectively. Activities carried out under this component sought to improve the implementation capacity of public sector entities as well as private sector and civil society organizations to ensure effective coordination and to achieve better results on the ground. Project support aimed to address key capacity gaps in line with the national capacity building program. Particular emphasis was placed on strengthening the monitoring and evaluation system. Expected results included establishment of restructured federal and regional HAPCOs and a fully functional M&E system. 12. Component 2: Multi-Sectoral Prevention Fund (Appraisal estimate US$3.5 million): This component aimed to support a broad range of public sector interventions to prevent HIV spread including: peer education, counseling and testing, condom distribution, and mechanisms for reduction of harmful traditional practices. Specific attention was to be paid to integration of HIV/AIDS in reproductive health interventions. Financial support would be provided through a competitive process based on submission of proposals by the respective ministries and agencies. 4 Assessment of HIV/AIDS-related risks among men having sex with men (MSM) in Addis Ababa, Ethiopia, Gebreyesus & Mariam, 2009, Journal of Public Health Policy Vol. 30, 3, 269–279 5 Assessment of HIV/AIDS related risks among MSM in Addis Ababa. Hagos S. MPH Thesis: Dept. of community Health, Addis Ababa University, 2006 3 13. Component 3: Local Response Fund (Appraisal estimate US$15 million): to support community and woreda-based initiatives delivering evidence-based HIV prevention and mitigation activities targeting specific key at-risk groups. Appropriate targeting was to be based on the epidemiological synthesis study published in April 2008. Activities included: HIV/AIDS awareness creation; behavioral change (reduction of traditional harmful practices; safer sexual behavior by both SWs and their clients; delayed first sexual intercourse; reduction in early marriage; reduced sex with non-regular partners); condom distribution; and counseling and testing services. 1.6 Revised Components 14. The original components were not revised. 1.7 Other significant changes 15. The project was restructured twice: (i) $5 million of the total credit was cancelled in December 2008. These funds were reallocated to the Food Crisis Response Program. This was in response to a request from Government in October 2008, but also took note of the slow disbursement rate; (ii) the closing date was extended, at no additional cost, from 07/02/2011 to 09/30/2011, to allow for completion of an ongoing survey to assess the status of the PDO indicators (necessary because of an unexpected delay in publication of the DHS results). Both restructurings were second level, and were approved by the Country Director. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry Preparation 16. A project identification mission in June 2006 undertook a detailed assessment of the epidemiological situation, critical response needs, contributions from other development partners, and the Bank’s comparative advantage. The mission team included a full range of development partners (including the Donor Group Co-chair, UNAIDS and key co-sponsors, and DFID). Consultations were held with government agencies (FHAPCO; sector ministries), other donors, and civil society. Project design reflected a request by the Minister of Health that the project should: support multi-sectoral interventions, capacity building, and creation of an enabling environment for the proper implementation of treatment, care and support; and give priority to prevention among high risk groups with particular emphasis to rural areas. The Project Appraisal Document (PAD) highlighted the importance of addressing stigma and discrimination as a key element of an enabling environment, but it is noted that this would be a long-term undertaking. 17. It was clear that the Bank could not close the huge financing gap but should focus on programmatic areas in which incremental finances could result in critical or significant additional benefits. The evidence to guide prevention related interventions was based on a well documented assessment of effectiveness. It was noted that many current support 4 programs had a rather short timeframe (subject to annual review in some cases) for their financial commitments. The project was designed to provide flexible support over three years, in order to mitigate the risk of any abrupt changes in the resource envelope. 18. Lessons learnt and reflected in the project design: The project design reflected important lessons from EMSAP1, including the need for: a) continued targeting of high-risk areas and most-at-risk groups, with renewed focus on prevention activities; b) interventions to promote greater male responsibility and to reduce inter-generational sex; c) a strong national coordination entity, with further capacity support to FHAPCO; d) partner coordination: it was agreed that FHAPCO would review the effectiveness of the various fora and move towards one national Partnership Forum; e) need for an annual joint review and budgeting process enabling Government to plan and allocate resources according to the country’s needs; f) integration of health and HAPCO M&E systems; g) a rigorous assessment of the major transmission dynamics and potential evolution of the epidemic; and analysis of the current responses in relation to this; h) capacity building of communities, woredas and civil society and i) strengthened financial planning and reporting mechanisms. There is undeniable evidence that the male circumcision reduce the risk of heterosexual transmission of HIV infection. Since over 90% of Ethiopian men are circumcised, male circumcision was not considered as a critical element of HIV/AIDS prevention in Ethiopia and therefore was not included among the agreed activities. 19. Risks and their mitigation: PAD identified important risks6 and proposed measures to mitigate these. These measures were generally appropriate, and effective risk management reduced the impact. FHAPCO, as implementing agency, carried out annual financial reviews and gap analysis to ensure implementation of HIV/AIDS related interventions within the resources available. FHAPCO also provided leadership and coordinated HIV activities throughout the country in accordance with its mandate as the single national authority on HIV/AIDS in Ethiopia. However, it proved difficult to maintain trained and experienced staff, particularly in the light of an ongoing civil service reform process (the BPR, see below); scale up to an additional 150 woredas added to the difficulty. This risk was partially mitigated through support to training and capacity building using project funds, but staffing constraints did impact on project performance. The PAD rated overall risk of the project as substantial. The ICR team agrees with this assessment. 20. Adequacy of government’s commitment, stakeholder involvement and participatory processes: Government, as represented by FHAPCO, was closely involved in identification of project objectives, and had accorded HIV and AIDS a high level of priority at the time of project preparation. There was broad stakeholder consultation on project design, including UN agencies, multilateral and bilateral donors, NGOs and CSOs. 6 (i) failure to sustain and expand government commitment; (ii) program resources may be manipulated for personal ends or special interests; (iii) maintaining trained and experienced staff; (iv) unpredictable flow of donor finance; (v) complacency [about the seriousness of the epidemic]; (vi) Ministry of Health Leadership; (vii) gap filling nature of the Project; (viii) scaling up decentralization of the HIV program would increase the risk of monitoring results and having relevant skills mix. 5 21. There was no formal Quality at Entry Review. 2.2 Implementation 22. Positive factors affecting implementation included: the short time interval between closure of the previous project (EMSAP 1) and project effectiveness (six months); effective coordination structures, with a fully operational FHAPCO and plans to strengthen the staffing of regional HAPCOs; and a single national plan, the Strategic Plan for Management, providing the framework for all interventions. An effective social mobilization strategy, including the ―community conversations‖ program7, provided a sound base for delivery of the Local Response Fund (LRF). Continuity of Bank staff (the same Task Team Leader since development of EMSAP 1) ensured that institutional knowledge was maintained. 23. Several factors impacted negatively on project implementation, particularly in the early years: there was still limited understanding of the epidemic with respect to MARPS (with lack of clarity as to which populations were most at risk, or whether community organisations were best placed to provide services to them) making it difficult for HAPCO to prioritize, plan and manage prevention interventions; completion of a Business Process Re- Engineering (BPR) exercise to identify key structural weaknesses in the HAPCO setup was delayed by a year, resulting in high staff turnover and sub-optimal FHAPCO performance; there was limited capacity within project teams particularly for procurement and financial management, and in the newly added woredas (exacerbated by high rates of staff turnover); and the project documents had not made it clear that community-based disbursement procedures applied to the LRF, resulting in overly cumbersome procurement procedures and slow disbursement. 24. It was agreed to increase already high EMSAP1 coverage levels, adding a further 150 woredas8 to the 266 woredas already supported. The project covered approximately 50% of the country’s population. All woredas were included in five key regions (Addis Ababa, Afar, Dire Dawa, Harari, Tigray). Project scale up slowed implementation (see Risks, #18 above) but increase in coverage is likely to have contributed significantly to project impact. Completion of the BPR resulted in an increase in delegation of authority to decentralized levels, which is likely to have strengthened community and woreda initiatives. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization M&E Design 25. The PAD included a set of 6 PDO and 10 Intermediate Outcome indicators. It was noted that some of these could be measured at least annually and others only at the end of the project. An appropriate mix of outcome and output level indicators was chosen to support and assess project and program management. It was stated that three of the PDO indicators (VCT 7 Meetings of groups of up to 50 households to discuss topics such as risk factors, vulnerable groups and prevention measures; discussions led by trained community facilitators and informing development of kebele HIV/AIDS action plans. 8 Ethiopia is a federation of 9 Regions, subdivided into 800 Woredas (districts) 6 uptake; numbers of woredas supporting PLWHA groups and OVCs; establishment of an information and M&E system) were measured regularly. Good baseline data was available for the other three measures (relating to knowledge of HIV transmission, and condom use) required population based surveys. M&E Implementation and Utilization 26. It was recognized that considerable inputs would be needed to support the national M&E system: integration of the HAPCO and health management information systems; system decentralization to enable regions and woredas to monitor performance; capacity building; and Community Information System (CIS) development to address the needs of vulnerable communities. The project design underestimated the level of inputs and time which would be needed to have an effectively functioning and fully integrated system. The system could not generate data for 8 out of the 16 indicators until the mid-point of the project. 27. FHAPCO was expected to develop, pilot test and implement a system to collect community-based and program monitoring information. A CIS manual and format to capture non- health indicators was designed and pilot tested but was not fully implemented. 28. The project planned to support and develop national and regional M&E centers of excellence (to provide capacity building and collect data for FHAPCO) but this was not fully implemented, limiting the capacity of the M&E system to generate essential monitoring data. Donors had different M&E requirements and approaches to capacity building, and lack of harmonization created difficulties for FHAPCO. However, by the time of the Mid-Term Review it was noted that FHAPCO had made progress in capacity strengthening and development of systems to monitor and evaluate progress towards the project objectives, as well as the wider national AIDS response. Development of the Health Management Information System (HMIS), the Manual for Project Monitoring and Evaluation, and the FHAPCO database and website were recognized as important steps towards the ―one national M&E system‖. Completion of the BPR process guided efforts to strengthen FHAPCO's HR capacity for M&E, and staff were in post at federal, regional and woreda levels. 2.4 Safeguard and Fiduciary Compliance 29. Environment. The project did not trigger any major concerns related to the Bank’s safeguards policy. It was noted that health facilities had provision for disposal of medical waste, and no significant negative impacts were anticipated. ISRs consistently rated the project satisfactory for overall safeguards compliance and in respect of operational policy OP4.01 (Environmental Assessment). 30. Financial management and procurement. There was ongoing concern over the financial management of the project for the first two years of implementation, and it was consistently scored Moderately Unsatisfactory. The main issues were recurrent delays in the submission of Interim Financial Reports (IFRs), audit reports and Statements of Expenses (SoEs). The issue of perennially late audits was recognized to be a government-wide issue, well beyond the remit of this particular project. The audit for the fiscal year ending July 2007 was qualified due to the substantial amount of unjustified expenditure. The situation 7 improved significantly by mid 2009: the government submitted a plan of action to respond to issues raised by auditors over the 2006/2007 audit, and the 2008/2009 audit report was being prepared. Financial management performance was rated Satisfactory by the end of the project. Procurement weaknesses and absence of federal and regional procurement plans were noted during the first year of implementation, but performance improved over time. 2.5 Post-completion Operation/Next Phase 31. There is concern that current levels of financing, particularly from the Global Fund and PEPFAR, will not be sustained in the longer term. The government and civil society have limited resources and capacity to sustain the levels of inputs and services which will be needed for prevention and the growing demand for treatment. There is an urgent need to identify other sources of funding, including increased government appropriations and community contributions. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation 32. Relevance of objectives: The project objectives focused on provision of HIV prevention interventions to MARPs, and particularly young women. This was highly appropriate for several reasons: there was a continued high incidence of new infections in the country, particularly in young women (who were seven times more likely than young men to become infected); there was also a concentrated epidemic in sex workers; and rising infection rates in rural areas (home to over 85% of the population) raised a valid concern that there could be a rapid increase in the burden of disease and a heavy socio-economic burden. Internationally recognized best practice supported prioritization in those high-risk populations identified as the drivers of the epidemic. However, the PDOs could have been more specific in definition of the target groups, including: the subset of young women who reported having multiple sex partners; sex workers as the primary MARP of concern; and older men (both as clients of sex workers, and as the source of the excess of infections in young women). This would have strengthened targeting and led to the selection of more focused PDO indicators. The PDOs were consistent with the government’s development priorities as reflected in the Poverty Reduction Strategy Paper and the Bank’s Country Assistance Strategy. On this basis the relevance of objectives is rated Substantial. 33. Relevance of design and implementation: The design built upon that of EMSAP1, and incorporated current international experience on HIV and AIDS with its focus on most-at- risk groups in higher risk areas. It also reflected significant inputs from other development partners, particularly the Global Fund and PEPFAR, who were supporting rapid treatment access expansion. HAPCO, for example, received more than $300 million for HIV/AIDS from the Global Fund, which meant that the project had to be particularly strategic in its choices. The project aimed to partially cover the gap in prevention, and to improve national program delivery and utilization of all resources, through improved public and private sector institutional capacity. Particular emphasis was given to partner coordination and a well- functioning M&E system integrating the HAPCO and health systems. Design also noted the 8 need for better epidemiological analysis and use of data for decision-making; and for strengthened financial reporting. 34. The choice of project components addressing program coordination and institutional strengthening, multi-sectoral prevention, and community and woreda-based initiatives, provided a relevant design and implementation framework. The PAD proposed a broad list of types of activities to be financed by the LRF, but did not provide any guidance as to the choice of activities nor on how these should be implemented. Work with Woreda and Kebele HAPCOs was appropriate, allowing for implementation to be managed at a highly decentralized level. 417 out of 800 Woredas (approximately 50% of the population) were covered, with an effective balance between urban centers (containing many higher risk ―hot- spots‖) and rural areas. Alternative designs were rejected for well-considered reasons (including the need to respond to a changing epidemiological situation, and to maximize impact of prevention through the most efficient use of limited funds). Providing considerable autonomy to local stakeholders to design programs that were culturally sensitive was a reasonable approach in the context. 35. On the basis of the above the relevance of design and implementation would have been rated high. However, there was insufficient emphasis put on reaching those most-at-risk populations identified at the project appraisal. The time and levels of input needed to establish a fully integrated M&E system were also underestimated. The relevance of design and implementation is rated substantial. 36. Based on high relevance of objectives and substantial relevance of design and implementation, overall relevance is rated substantial. 3.2 Achievement of Project Development Objectives 37. Assessment of PDO achievements has been made difficult because of unforeseen and continued delays in availability of the full DHS report. The Task Team took appropriate steps to fill the information gaps, by supporting FHAPCO to commission a dedicated survey9. A preliminary DHS report has recently been made available, providing some of the required information (on young women and men’s knowledge of transmission and use of condoms in high-risk settings). Access to prevention services 38. The preliminary DHS report provides good evidence that levels of knowledge of sexual transmission of HIV have been maintained or increased amongst young people: the proportion of women aged 15-24 who state that HIV infection can be prevented by using condoms has increased by over a third during the project period (from 47% to 62%); 81% of young men are aware of this (up from 66% in 2005); the percentages who know that infection can be prevented by remaining faithful to one uninfected partner have not changed significantly (for young women an increase from 66% to 68%; for young men a small 9 Survey on the Status of Sexual Behavior, Knowledge about HIV/AIDS and Condom Utilization in Thirty-Five Woredas, FHAPCO, Government of Ethiopia, 2011 9 decrease from 76% to 74%). Overall, knowledge of HIV prevention methods has increased since 2005, especially among women. According to the 2005 EDHS, 35 percent of women knew that HIV could be prevented by using a condom and by limiting sexual partners; this compares with 43 percent in 2011. Among men age 15-49, this percentage increased from 57 percent in 2005 to 64 percent in 2011. The same groups have also maintained or increased rates of condom use in high-risk sex situations: of particular note is that the reported use of condoms by women in high risk settings has more than doubled from 27% to 62% (but it should be noted that this is based on very small sample sizes – only 0.5% of women said that they had had two or more partners in the last 12 months). Support for these positive findings comes from the dedicated survey referred to above: 83% of young men and 79% of young women mentioned abstinence, faithfulness to one partner and condom use as effective prevention methods and rejected common misconceptions about infection transmission; 86% of those young men reporting transactional sex stated that they had used a condom during the last such encounter. A summary of the methodology and results from the study carried out in 35 woredas is included in Annex 7. 39. The percentage of female sex workers reporting condom use with their most recent client was already very high at baseline (96.6%). The survey referred to above reported that the target of 98% of condom use was met. This does not tally with the percentage of young men reporting condom use in transactional sex, and the result should be viewed with caution. However, there is good evidence of increase in condom use rates overall, with an increase in the number of condoms distributed from 70 million (in the year ended July 2008) to 173 million in the year to July 2011 (exceeding the agreed target of 140 million). 40. Underlying these broadly positive results (and based on the assumption that awareness of one’s HIV status will act to reinforce positive prevention behaviors) has been a huge increase in the percentage of people aged 15 and older undergoing voluntary HIV counseling and testing. The baseline number tested in 2005 was 564,000 (FHAPCO Annual HIV/AIDS M&E Report); a target of 700,000 was set for year three of implementation; by the end of the project in 2011, an estimated 9.8 million tests had been carried out (it should be noted that there is no information on how many of these were repeat tests). This was the result of a massive push by the government to expand testing throughout the country (the millennium development campaign) using EMSAP2, Global Fund and PEPFAR resources. Access to care and support 41. At project completion over 95% of woredas reported that they were supporting effective PLWHA groups and associations and implementing community programs to support orphans. FHAPCO reported that assistance had been provided to over 100,000 PLWHA and 350,000 orphans. This included provision of food and shelter, income generating training and start up capital, and educational support for orphans. This support occurred in the context of a marked reduction in stigma and discrimination10(though it should be noted that high levels of stigma and discrimination persist). 10 Ethiopian Stigma and Discrimination Survey, Network of Networks of HIV Positives in Ethiopia (NEP+): presentation, Annual review meeting, Hawassa, November 2011 10 42. The sixth PDO indicator was ―establishment of an HIV/AIDS information and M&E system that supports improved policy and program decision making‖. This took a considerable amount of time to achieve: at the time of the project Mid-Term Review (May 2009) it was noted that the system was clearly defined in the Project Implementation Manual and the ―Final Manual for Monitoring and Evaluation of EMSAP2 and Global Fund Projects‖. Staff and funding were in place, and there was evidence of use of data in strategic decision making, but weaknesses were also apparent, including poor aggregation of data, weak data quality verification, and lack of system capacity to assess MARP program effectiveness. However, it was judged that this indicator had been fully met by mid 2010, following strengthening and roll out of the Health Management and Community Information Systems. 43. The PAD acknowledged that whilst the World Bank would not continue to be ―the major funding agency in the fight against HIV/AIDS, its continued involvement is still believed to be critical …. particularly in the areas of prevention...‖. Support to prevention had previously been mainly limited to urban areas, but there was growing evidence that the epidemic was increasingly affecting rural populations. The project made a substantial contribution to the national response in key areas by: (i) maintaining the focus on MARPs, building on the Epidemiological Synthesis (and helping to overcome a perceived reluctance on the part of government to work with these groups); (ii) strengthening and expanding institutional capacity, particularly within HAPCOs (which continued to function during the lengthy BPR reorganization process); (iii) strengthening a policy and institutional framework to enable effective and rapid mobilization of funds from other sources (the Global Fund and PEPFAR); (iv) providing funds to civil society through the LRF to, inter alia, sustain the Community Conversations (an effective mechanism to spread prevention messages, and address stigma and discrimination) and ensure that the program reached rural areas. 44. The limitations of HIV prevalence as an indicator of program achievement are well recognized, and the PDO rightly focused on access to prevention services. However, it should be noted that ―[a]t the national level, the epidemiologic trend over the past eight years has been stable… [and] HIV prevalence appears to be declining in urban areas, according to analysis of data from ANC sites that collected data consistently for more than ten years‖11. This would suggest that prevention related interventions are having a significant impact at the population level, given the significant increase in access to treatment (which serves to maintain prevalence rates) during the same period. The possibility remains that this positive national level trend is masking continued high transmission rates and increased prevalence in MARPs, underlying the need for careful monitoring of the epidemic and sustained prevention efforts. 45. On the basis of the available evidence we conclude that the project has successfully achieved the targets set for the PDO indicators. The achievement of project development objectives is rated Substantial. 11 Report on progress towards implementation of the UN Declaration of Commitment on HIV/AIDS 2010; Federal Democratic Republic of Ethiopia, Federal HIV/AIDS Prevention and Control Office, March 2010 11 3.3 Efficiency 46. Highly cost-effective interventions were selected for implementation using project funds12. Several key interventions had a cost-effectiveness of US$100 per HIV infection averted or less (a commonly-used threshold for cost-effectiveness) including: voluntary testing and counseling (US$34-161 per DALY averted); peer-based programs to educate higher-risk groups, including sex workers and youth (US$11-74 per DALY averted); and promotion, and distribution of condoms (US$39-305 per DALY averted). 47. Allocative efficiency was increased by the use of comprehensive and up to date epidemiological information derived from the Epidemiological Synthesis Study. Based on this knowledge of the drivers of the epidemic emphasis was placed on reaching youth, particularly young women, seen as highly vulnerable, and sex workers in locations previously not recognized as epidemiologically important (including small towns and rural areas). The operating costs for the project increased about five times over the initial allocation to about 20% of the disbursed credit. This was partly due to the increase of the number of woredas from 226 to about 416 as well as an initial under-estimate of the resources required to carry out decentralized HIV activities. 48. There are no recent studies or data available to allow for a meaningful a posteriori economic and financial analysis of project implementation. However, it is recorded that the project has satisfactorily achieved its targets. In this regard even if it could not be quantified, most of the activities carried out by the communities had a potential for reducing the cost for prevention. An economic cost and benefit analysis was performed during the Project appraisal process, focusing on HIV prevention activities. Even though microeconomic benefits are difficult to quantify, on the basis of the existing evidence, stabilization and subsequent fall in the HIV prevalence rate will be likely to increase the growth rate of GDP per capita, and a reduction in AIDS-related deaths and morbidity will increase the country’s labor force, which in turn would contribute to increasing GDP growth. A recent analysis of the economic benefits of community-based preventive activities supported through the project considered: (a) behavior change interventions at community level; (b) counseling and testing; and (c) condom distribution. The increase in condom use and consequent reduction in unprotected sex in the general population was found to be the main source of the estimated economic benefits derived from the project implementation, with a benefit/cost ratio of between 1.99 and 2.43. [See Annex 3] 49. Based on the emphasis placed on highly cost-effective interventions, and increasing allocative efficiency (see OED 's analysis regarding the principal elements contributing to the efficiency of the MAPs13), project efficiency is rated as Substantial. 12 Disease Control Priorities in Developing Countries, Jamison et al, 2006 13 Operations Evaluation Department, Committing to Results: Improving the Effectiveness of HIV/AIDS Assistance (2005), see especially Table 4.1, p. 45. 12 3.4 Justification of Overall Outcome Rating 50. Rating: Based on the above presented data on the relevance of the project’s objectives, design, and implementation (Substantial), achievement of PDOs (Substantial) and efficiency (Substantial), the overall outcome rating of the project is Satisfactory. 3.5 Overarching Themes, Other Outcomes and Impacts a) Poverty impacts, gender aspects and social development 51. Poverty Impacts. The project was a major contributor to efforts to prevent spread of the epidemic and reduce the number of new infections. Although there were no formal studies of the Socio-Economic Impact of the epidemic, project support is likely to have led to a reduced burden of AIDS-related illness, particularly in adults of working age, and it can be expected that this will have had a positive impact on earning capacity and household wealth. 52. Gender Aspects. The PAD highlighted the particular risk of infection facing young women and SWs, identified the main factors underlying this, and set out ways in which the issues would be addressed. As a result, the project had a major focus on young women and female sex workers, as set out in the PDO and reflected in the PDO indicators. Although there continue to be gender differences in project outcomes (as seen, for example, in levels of knowledge of the epidemic) improvements in these outcomes have been seen for both women and men. The government’s ICR states that poor women were targeted to receive income generating support (in the form of training, start-up working capital and links to micro- finance, though there is no information available on the extent to which these women were able to participate in design and management of such support programs. 53. Social Development. The project built upon a highly effective social mobilization strategy, the ―community conversations‖ program, which was considered to have resulted in increased awareness of HIV/AIDS, improved participation of women, greater support for PLWHAs and OVCs, and a decrease in tolerance of customary practices such as multiple sexual relationships. There was at least one conversation group in each kebele where 50 participants met twice per month to discuss responses to the epidemic. The Local Response Fund (component 3) helped to expand these community level initiatives, which significantly reduced stigma and discrimination, and reached marginalized segments of the population. 54. A large study of PLWHA, carried out by the Network of Networks of HIV Positives in Ethiopia found ―.. a general feeling among [respondents] that the level of stigma is declining and less virulent than the situation five years ago‖. The study also found relatively low levels of discrimination against those affected and their children in school settings, and survey respondents expressed higher levels of confidence in health personnel than other groups. There remain major areas of concern: the high levels of poverty reported by those living with HIV and AIDS (95% of women and 93% of men); the ―institutionalization‖ of discrimination (eg the widely reported charging of higher rents to those known to be HIV positive; 13 discriminatory employment practices, including loss of work; bureaucratic obstacles to legal action to enforce rights); and low levels of awareness of basic rights amongst PLWHA. (b) Institutional Change/Strengthening 55. The strengthening of HAPCOs established with the support of EMSAP1 has been a significant institutional development, laying the foundations for a sustained national response. Community involvement was broadened and deepened, through LRF support to creation and development of civil society organizations, and further institutionalization of the ―community conversations‖. Essential instruments for coordination, management and implementation were developed (including the Financial and Procurement Manuals, and technical guidelines and manuals for eg Anti-Retroviral Treatment, HIV Counseling and Testing (HCT), Prevention of Mother to Child Transmission). The EMSAP1 and Global Fund Project Coordination Units (PCUs) were formally closed by FHAPCO but provided the basis for the current unified system of grant management, guided by the grant management manual developed with Bank support. (c) Focus on Most At Risk Populations 56. A major achievement of the project was to ensure a sustained focus on Most At Risk Populations. This was based on a sound epidemiological assessment as part of project preparation (and drawing on EMSAP1 experience), further supported by an important epidemiological study ―HIV/AIDS in Ethiopia: An Epidemiological Synthesis‖ carried out with project support in 2007. The study confirmed that ―young populations, especially never- married sexually active females have the greatest risk of HIV infection in the country‖ and noted that ―discordant couples are also a concern, pointing to a clear need for couple counseling services‖. The study also noted that ―traditional high-risk groups such as sex workers seem to be reducing some of their risky behaviors‖. This was a clear endorsement of the project approaches and development objectives: it confirmed the value of the emphasis placed on expansion of HCT; and also underlined the importance of interventions addressing sex-workers and their clients. 57. An evaluation of the second SPM14 completed in Feb 2009 reaffirmed the focus on MARPs, observing that ―high risk groups are increasing in number and type: college students and MSM are increasingly recognized as most-at-risk population groups‖. The project Mid- Term Review carried out in March 2009 noted the importance of assessing the effectiveness of interventions for these groups, the need to increase prevention activities targeted to MARPs relative to those for the general population, and agreed that a national survey of MARPs (updating the Epidemiological Synthesis study carried out in 2007) would be carried out that year (there is no record that this survey took place). This reflected some initial concerns over government reluctance to discuss and provide targeted prevention services for sex workers (most notably condoms). The project made a major contribution to discussion of this issue and provision of these services. 14 Ethiopia Strategic Plan for the Multi-Sectoral Response for 2004-2008 14 4. Assessment of Risk to Development Outcome 58. The major risk to the Development Outcome is continued financing of Ethiopia’s response to the epidemic at levels sufficient to ensure that prevention efforts will be maintained. EMSAP2 was not a major source of funding in the context of overall funding, but played a very important role in ensuring an ongoing focus on prevention. Wider funding constraints (particularly the uncertainty over future levels of resources from the Global Fund) could impact negatively on the national response and sustainability of the progress achieved. The risk to Development Outcome is therefore rated High. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry 59. Project design built upon lessons learned from the first project (EMSAP1) and on other lessons learned from the implementation of similar projects in other countries. The project appraisal process was inclusive of a wide range of stakeholders, including Government, Civil Society and a full range of development partners. The design team had confirmed that the project was recognized as a priority by government, that government was fully committed to implementation, and that the design was fully consistent with national policies and strategies (as encapsulated in the Strategic Plan for Management). 60. Institutional arrangements were agreed with the Government: program management offices (HAPCOs) had been established at all levels (Federal, Woreda and Kebele), and training had been provided in project planning and implementation. Project manuals, including the Project Implementation Manual (PIM), Financial Manual and Procurement Manual were prepared. These documents fully set out the arrangements for release of funds to Regions and Woredas. They were disseminated and discussed at a national project launch workshop held in July 2007, and follow up launch meetings for regions, woredas and NGOs. 61. However, although over 60% of project funds were allocated to support community and woreda-based initiatives through the LRF, the Financial Manual did not make it clear that community-based disbursement procedures would apply. The resulting cumbersome procedures required to justify community based expenditures resulted in slow disbursement and impeded effective implementation during project start-up. Project design underestimated the inputs and time needed to develop a fully effective M&E system (leading to lack of data for half the project indicators until project mid-point). 62. The many positive design features are noted. Slow disbursement of the LRF had a major impact in the early part of the project, and there were issues with M&E throughout. On this basis, Bank performance in ensuring quality at entry is rated Moderately Satisfactory. 15 (b) Quality of Supervision 63. The Bank’s country office had a full project team led by an experienced Task Team Leader (who had also led the team on design and implementation of EMSAP1). This allowed for effective and supportive supervision from the start of the project. Additional support was provided by the Bank’s headquarters, particularly the AIDS Campaign Team for Africa (ACTafrica) and the Global HIV/AIDS Monitoring and Evaluation Team (GAMET), who provided valuable assistance during the course of project execution, including high-level field level participation in the project Mid-Term Review. 64. No formal Quality of Supervision assessment was carried out. However, a review of project documentation confirms that the task team undertook regular supervision missions, prepared Aide-Memoires, alerted government and FHAPCO to problems with project execution, and facilitated remedies in a timely manner, in conformity with Bank procedures. The Implementation Status and Results reports (ISRs) realistically rated project performance. The task team also monitored safeguard and fiduciary compliance. Financial management and audit issues were identified and brought to the attention of government, and were eventually resolved through the persistent and patient efforts of the country team. 65. Rating: on the basis of the above the Bank performance on quality of supervision is rated Satisfactory. (c) Justification of Rating for Overall Bank Performance 66. Based on the moderately satisfactory rating for quality at entry and satisfactory rating for quality of supervision, overll Bank Performance is rated as Moderately Satisfactory. 5.2 Borrower Performance (a) Government Performance 67. Federal and Regional governments had accorded HIV and AIDS a high level of priority at the time of project preparation. They continued to build upon and support the institutional frameworks (legislation, policies and strategies) and structures established during the previous Bank project (EMSAP1). In 2009, Federal Government prepared a second SPM to guide implementation of all HIV/AIDS activities up to 2011. The National AIDS Council (chaired by the President), FHAPCO Management Board (chaired by the Minister of Health) and Review Board met regularly and discharged their responsibilities fully. Regional AIDS Councils (chaired by the regional presidents), Management Boards (chaired by the heads of the Regional Health Boards) and Review Boards also met regularly. Governments at all levels supported project implementation, provided timely approval of HAPCO Annual Action Plans, and collaborated and cooperated fully with the project team. 68. There were however, two major areas of concern: (i) the BPR exercise, a management change process across the civil service, was delayed by over a year; this resulted in high staff 16 turnover and sub-optimal HAPCO performance during the period of change; (ii) there were recurrent long delays in delivery of satisfactory audit reports, IFRs and SoEs; this was recognized as a cross-government problem and was not addressed until the time of the Mid- Term Review. The project was formally flagged as a problem project in mid 2008, but the situation improved greatly over the following year. 69. Rating: Moderately Satisfactory (b) Implementing Agency Performance 70. FHAPCO, as implementing agency, effectively managed planning, approval, and review processes, and ensured effective coordination between the different sectors and implementation levels. However, financial and procurement management was a major concern, particularly during the first two years of implementation. This partly resulted from the lengthy restructuring of the agency as part of the BPR (discussed above), which contributed in part to high staff turnover (though it must be questioned whether more effective FHAPCO management could have mitigated this). There was insufficient FHAPCO capacity to manage a large influx of funding from other sources (notably PEPFAR and the Global Fund). Lower priority was given to EMSAP2 activities, and implementation was slow during the first two years of the project. 71. Performance improved considerably once the agency had completed the BPR process and finalized its internal structuring and staffing. All collaborative projects were managed by a unified directorate, ensuring appropriate levels of prioritization and oversight. FHAPCO also increased its technical staff (with some project support). Project funds were fully disbursed by the closing date. 72. Rating: Moderately Satisfactory (c) Justification of Rating for Overall Borrower Performance 73. In the light of the Moderately Satisfactory Government performance and FHAPCO’s Moderately Satisfactory performance as implementing agency, the overall performance of the Borrower is rated Moderately Satisfactory. 6. Lessons Learned 74. Maintaining strong national and lower level coordination entities is a critical component of an effective and sustained response: The establishment and strengthening of HAPCOs at the federal, regional and woreda levels was a major undertaking that has proven effective. The time, staffing and other resources allotted for the continued strengthening of the system are important in project implementation. The project supported effective coordination of an impressively wide range of public sector agencies. 75. Ensuring harmonization of development partner inputs requires agreement to adopt shared processes and systems, and move beyond “consensus rhetoric�: eventual agreement 17 to use a single M&E framework, and impressive moves towards joint review, planning and budgeting, helped to reduce the administrative burden on the country, and mitigate the serious capacity constraints. 76. Management capacity should be proportionate to levels of funding and program expectations: Particular care is needed to ensure that the levels of capacity are sufficient to manage large influxes of donor funding; in the absence of this, smaller projects are likely to receive insufficient attention. 77. Availability of the country based TTL is critical for ongoing supervision of the project and timely identification of bottlenecks and appropriate solutions. 78. Relatively small amounts of funding can have a catalytic effect if designed and delivered strategically: EMSAP2 was instrumental in re-directing national efforts towards effective strategies (clearly defining and focusing on MARPs) and catalyzing community involvement and responses. 79. Development of a comprehensive M&E framework should be one of the first project interventions: Such a framework needs to reflect the different M&E requirements and approaches to capacity building of different partners, and to ensure timely integration of different components of the system; in the case of EMSAP2, delays in establishment of the CIS, and in integrating the HMIS and HIV/AIDS systems, led to lack of timely information to report on program performance. The project design underestimated the level of inputs and time which would be needed to have an effectively functioning and fully integrated system. 80. A Heterogeneous and diversified epidemic requires detailed and regularly updated information to guide interventions: The Epidemiological Synthesis study, carried out with Bank support in 2008, was an important contribution to national program development and updating, and confirmed the importance of focusing on prevention and targeting the most-at- risk population and hot spot areas, but continued follow up was (and still is) required for a fully effective response. A later survey15 identifying Institutes of higher learning as emerging ―hot spots‖ for HIV transmission and identified university students as a previously unrecognized group at risk of acquiring HIV infection. 81. Successful HIV/AIDS prevention programs need to respond to a full analysis of the epidemiological situation, including an honest assessment of those populations most likely to contribute to new infections: In the case of Ethiopia there was good reason to be concerned at the increased infection risks faced by MSM, and the possibility that new infections in this group were an important driver of the epidemic, but there was widespread reluctance to address this. A formal Political Economy Analysis could have informed a strategy to overcome this reluctance. 15 ―Risky Sexual Behavior and Predisposing Factors Among Ethiopian University Students� 18 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies/ 82. No issues raised. (b) Co-financiers Not applicable. (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) 83. No issues raised. Annex 1. Project Costs and Financing 1. According to the signed financial agreement, the initial project fund of EMSAP-II was SDR 20 million (US$ 30 million). This was reduced to USD 25 million after USD 5 million had been moved to the Country’s Food Security Response. However due to gains from exchange rate changes, the total amount of fund utilized was US$25,997,472.12. This is 104% of the signed amount. a) Project Cost by Component Components Appraisal Estimate Actual (latest Percentage of (in Million USD) estimate) appraised Component 1 2.5 2.5 100% Component 2 3.5 1.5 42.8% Component 3 15.0 15.0 100% Total Baseline cost 21.0 19.0 90.47 Physical 4.3 4.5 Contingencies Price Contingencies 4.7 1.5 Total Project Cost 30.0 Front- end Fee PPE 00 Front- end Fee 00 IBRD Total Financing 30.0 25.0 83.3% Required 19 b) Financing Source of Fund Type of Appraisal Actual Latest Percentage of Cofinancing estimate(US$ estimate (US$ appraisal Millions) Millions) Borrower US$0.0M US$0.0M International USD30M USD25M Development 100% Association c) Disbursement Profile d) 2. As per the report of the last Supervisory mission, the disbursement from IDA to the project was SDR 16.8m from the allocated budget of SDR 16.8m after cancellation of SDR 3 million. This brings the rate of disbursement to 99.96%. Table 1. Final Disbursement Status by Category Category Description Amount Discharged SDR 1(a) Grants under2(a) of the project 2,204,109.82 1(b) Grants under Part 3 of the Project 10,241,140.93 2 Civil Works under 2(b) of the project 582,567.62 3 Goods, equipment, drugs& medical supplies 195,826.53 4 Training, Consultants, services,& audits 221,213.24 5 Operating Costs 3,354,540.89 6 Exchange rate arrangements 10,823.54 7 Cancellation effective February 03, 2012 6645.48 8 Grant Amount 16,816,868.05 9 Cancellation effective September 30 , 2008 3,183,131.95 Original Grant Amount 20,000,000.00 Source: World Bank Ethiopia Office Feb 2012 20 Table 2. Consolidated Disbursement, Settlements and Unsettled Balances (HAPCO) S/No Transfer Implementers COMPONENT I II III TOTAL 1 Regions 48,917,034.00 44,489,270.00 192,885,046.00 286,291,350.00 2 GOs - 18,749,328.00 - 18,749,328.00 NGOs, 3 FBOs,Ass. 229,000.00 - 4,090,628.00 4,319,628.00 4 FHAPCO 10,951,669.00 1,221,198.00 - 12,172,867.00 Total Advances 60,097,703.00 64,459,796.00 196,975,674.00 321,533,173.00 SETTLEMEMNTS Implementers COMPONENT S/No I II III 1 Regions 48,533,491.23 44,489,270.00 191,857,846.83 2 Gos 18,730,602.44 - NGOs, 3 FBOs,Ass. 229,000.00 3,620,495.72 4 FHAPCO 10,951,669.00 1,221,198.00 Total Settlement 59,714,160.23 64,441,070.44 195,478,342.55 UNSETTELED BALANCE 383,542.77 18,725.56 1,497,331.45 ( HAPCO Jan 2012) 21 Annex 2. Outputs by Component Component 1: National Program Coordination and Institutional Strengthening 1. The project provided financial, material and technical inputs to strengthen governance and coordination mechanisms at all levels. High level coordination outputs included regular meetings of the National AIDS Council, (chaired by the President), FHAPCO Management Board (chaired by the Minister of Health) and Program Review Board, and regular meetings of the equivalent bodies at regional level (Regional AIDS Councils, chaired by the regional president; HAPCO Management Board meetings, chaired by Regional Health Board Heads; and regional Program Review Boards). 2. A notable output was the holding of quarterly Joint Review meetings at all levels. These involved a wide range of stakeholders and development partners and were an effective mechanism for coordination and monitoring of program performance. 3. Capacity building of government agencies and civil society was strengthened through support to recruitment of key officials and staff in all parts of the system. Examples include: - recruitment of regional finance officers (29), woreda focal persons (415) and regional auditors (18), who filled critical Human Resource gaps. - financial management staff in all regions benefitted from the special training programs on the Bank`s financial management systems, organized by HAPCO. - procurement management staff received two rounds of training in Ghana. - the national M&E system was expanded and strengthened, and is now used by implementing partners; this included design and implementation of the Community Information system, designed to capture key non-health performance results. - ―Training of Trainers‖ training on program management provided to government staff from a wide range of different sectors; and on guidelines for provision of care and support to Orphans and Vulnerable Children (provided to 66 staff). - Two days training on HIV and AIDS to members of the House of Representatives. - Drafting and distribution of policy and program manuals and guidelines (eg 15,000 HIV Testing and Counseling manuals). 4. At the time of project closure the M&E system was producing regular reports using developed formats on a quarterly, biannual and annual basis. Monitoring and supportive supervision missions were conducted by FHAPCO directorates on a quarterly basis, to provide significant strengthening of the national response. 5. Special note should be made of the epidemiological synthesis study carried out in 2007. This had a significant impact upon policy, acting to refocus attention on the importance of addressing Most At-Risk Populations. (HIV/AIDS in Ethiopia: An Epidemiological Synthesis; Ethiopia HIV/AIDS Prevention & Control Office and Global HIV/AIDS Monitoring and Evaluation Team (GAMET), The Global HIV/AIDS Program, World Bank, April 2008). 22 Component 2: Multi-Sectoral Prevention Fund 6. This component supported a broad range of prevention interventions in the public sector, particularly targeting the most high-risk and vulnerable populations. At federal level 17 government public offices designed and implemented intervention programs targeting their staff in the workplace, and their client groups. In addition, eleven regional government bureaus and offices accessed the Multi-Sectoral Prevention Fund. Table 6 summarizes the main outputs produced by these different agencies. Table 6 Summary of Outputs by Federal level public sector Ministries, Departments and Agencies implementing project activities Sector Major Outputs Ministry of Federal Affairs  IEC/BCC; Condom promotion and distribution National Lottery Administration  IEC/BCC; Condom promotion and distribution Ethiopian Mapping Agency  Workshop and training to staff; IEC, BCC; Condom Distribution; Preparation of peer education manual; Conducting Knowledge, Attitudes and Practices (KAP) survey; HCT promotion Ministry of Finance & Economic 7. HIV/AIDS work place policy /guideline development Development 8. KAP survey; HIV/AIDS impact assessment Trainings; IEC/BCC materials production; HCT promotion; Condom distribution Ministry of Water Resources  Trainings on Mainstreaming; Peer education trainings 9. HIV/AIDS work place policy /guideline development 10. IEC/BCC; HCT promotion; Condom distribution Ethiopian Environmental Protection  HIV/AIDS work place policy; Condoms promotion & distribution; Agency IEC materials production; Sensitization workshop ; Training on mainstreaming for HIV/AIDS Task Force; HCT promotion National Archives & Library Agency  IEC/BCC; Peer education Trainings  IEC/BCC; Condom promotion and distribution Transport Authority  HCT promotion; Impact assessment  Ethiopian Electric Power Cooperation IEC/BCC; training on peer education; mainstreaming & M&E; KAP survey; Preparation of guideline  Ethiopian Rural Energy Development IEC/BCC; Condom promotion and Promotion Center Ministry of youth and Sports  IEC/BCC; Developing policy and guideline; Trainings and advocacy; Condom promotion and distribution; youth centers established and equipped.  IEC/BCC; Condom promotion and distribution Ethiopian Roads Authority  Trainings and advocacy; Ministry of Mines & Energy  IEC/BCC;Condom promotion and distribution Secretariat Office of the House of  HIV impact assessment; HIV/AIDS work place policy People's Representatives  HCT; Trainings and advocacy Food, Medicine and Health  Condom Survey Administration and Control Authority St. Peter’s TB Specialist Referral  Completion of construction (started during EMSAP1) Hospital 23 Component 3: Local Response Fund 11. The Local Response Fund was established as Component 3 to support community and woreda-based initiatives delivering evidence-based HIV prevention and mitigation activities targeting specific key at-risk groups. The focus of activities was at kebele (local council) level, including: ―Community Conversations‖ (meetings of up to 50 participants to discuss HIV and AIDS related topics); dialogue with young people; support to PLHIV and OVC; condom promotion and distribution; and uptake of counseling and testing. Some Kebeles installed condom dispensing boxes providing easy access to condoms. 12. These activities contributed to important Outputs, most notably: the huge increase in VCT uptake (voluntary HIV testing before marriage became the norm in much of the country); a significant reduction in stigma and discrimination (as noted by those directly affected); and support to marginalized groups such as prisoners, housemaids, the elderly and the disabled. 13. Specific Outputs included: - Food support to over 230,000 OVCs and shelter to 23,000 (community members in some rural areas established grain banks at kebele level to support orphans; others contributed cash) - IGA training to 43,000 OVCs, with business startup funds provided to 37,000 children and their guardians - Educational support provided to over 350,000 OVCs. - Number of Kebeles conducting community conversations 17,575 - Schools implementing Community conversation 15,305 24 Annex 3. Economic and Financial Analysis (including assumptions in the analysis) 1. This annex presents estimates of economic analysis done for the project. The analysis focuses on the economic benefits on the following categories of community-based preventive activities supported through the project: (a) behavior change interventions at community level; (b) counseling and testing; and (c) condom distribution. These activities are part of Component 2 of the project which focuses on preventive activities, with strong emphasis on community-based intervention delivery channels. The total expenditure under this component was $3.5 million. The analysis below provides estimates of the total economic benefits of these three categories of activities financed by the project. 2. It is worth noting that this analysis does not take into account the economic impact of activities aimed at strengthening the capacity among agencies working on HIV/AIDS in the country, which were carried out under Component 1 (The National Program Coordination and Institutional Strengthening), a key component of the project. As noted in the PAD, the World Bank was seen as donor of last resort in the context of EMSAP 2, ―filling the gaps and funding shortfalls that are not being addressed by other donors.‖ This is a critical function to increase the efficiency of the national response and promote resilience to sudden changes in donor funding priorities, by allowing more flexibility to focus on medium to long term planning capabilities. This set of activities was not taken into under this valuation exercise in light of the major difficulties to estimate the amount of economic benefits that they generate. As a result the estimates of economic impact reported here represent a conservative, lower- bound estimate of the actual benefits provided by the project activities. 3. The analysis presented in this annex extends and updates the economic analysis that was conducted during the project appraisal process. The estimation of the economic impact of the project under this approach is a combination of two main measures: (a) a valuation of the loss in years of the victims’ productive life and (b) the increase in treatment costs due to HIV/AIDS. The first measure takes into account that, without access to life-saving ARV treatment, HIV patients tend to becomes less productive over time and die at an earlier age than they otherwise would in the absence of disease. The second measure recognizes that HIV patients often incur excess health care associated with the treatment of symptoms and opportunistic infections. These need to be valued at market reference rates to reflect the opportunity cost of providing treatment to the disease at the expense of other health priorities. In addition, as noted in the PAD, ―ART itself can incur benefits by extending the number of productive years of HIV/AIDS victims.‖ An objective measure will therefore need to focus on incorporating the positive benefits of ART in the calculation of the net benefit of intervention. 4. Additional measures of impact, such as through reduction in capital stock and 25 accumulated GDP per capita as well as wider macroeconomic impact, although plausible, are not taken into consideration explicitly in this analysis. Again, the main reasons are (a) the lack of necessary information to allow a coherent and consistent estimation; and (b) preference to provide a conservative estimate of the lower bound estimate. Key assumptions of the analysis Table 1: Key Parameters Used in the Analysis Parameter Value Discount rate 3% Ethiopia project target population (15-24 years old) 28,144,798 Low case High case 16 Ethiopia sex worker population 19,701 1,210,226 HIV Prevalence in general population 5% HIV Prevalence among sex workers17 25% Risk of infection per episode of unprotected sex 1 in 2,000 Reduction in the risk above when using a condom 89% Average annual number of sexual episodes, general 10 population Average annual number of sexual episodes, per sex 281 worker Measures of project impact Value at Value at baseline completion Condom use, general population 50% 88% Condom use among sex workers 97% 98% Results of the analysis 5. The results are driven mainly by the size of the target population and the project effectiveness in increasing the key outcome indicator, namely the rate of condom use per sexual episode including both the general population and the sex workers groups targeted by interventions. Evidence from the ICR suggests the project was highly successful in increasing the rate of condom use in the general population, with a 76% increase observed between the baseline and project completion. 16 Vandepitte et al (2006) Estimates of the number of female sex workers in different regions of the world, Sex Transm Infect. 82(Suppl 3): iii18–iii25. 17 UNAIDS (2010). Report on progress towards implementation of the UN Declaration of Commitment on HIV/AIDS 26 6. The increase in condom use and consequent reduction in unprotected sex in the general population was key to the project success and the main source of the estimated economic benefits derived from the project implementation. 7. As seen in Table 2, there is a sizeable estimate of the economic benefit of Component 2. The benefit/cost ratio is estimated between 1.99 (low case for sex worker population size estimate) and 2.43 (high case for sex worker population size estimate). Table 2: Estimated Economic Benefits of Project - Component 2 Estimated Economic Benefits of Project - Component 2 Low Case High Case Total Benefits 6,976,954 8,499,100 Net Benefits 3,476,954 4,999,100 Benefit/Cost Ratio 1.99 2.43 27 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Name Title Unit Gebreselassie Okubagzhi TTL AFTH3 Maryanne Sharp Sr. Operations Officer AFTH3 Nadeem Mohammad Sr. Operations Officer AFTHV Jody Kusek Lead M&E Specialist HDNGA S. Omar Fye Sr. Environmental Specialist AFTS2 Eshetu Yimer Sr. Financial Management Specialist AFTFM Frode Davanger Operations Officer AFTHV Myrina Mccullough Consultant AFTH3 Eleni Albejo Program Assistant AFTH3 Albertus Voetberg Lead Health Specialist AFTHV Samuel Haile Selassie Procurement Specialist AFTPC Majorie Mpundu Counsel LEGAF Rajat Narula Sr. Finance Officer LOAG2 Tafesse Freminatos Financial Management Specialist AFTM Richard Olowo Senior Procurement Specialist AFTPC Monserrat Meiro-Lorenzo TTL and Team Leader AFTH3 Abiy Demissie Financial Management Analyst AFTFM Donald Mphande Senior FM Specialist and Cluster Leader Africa I, Tesfaye Ayele Procurement Specialist, AFTPC Endeshaw Tadesse Sr. operations officer AFTFM Mercy Mataro Sabia Sr. Financial Management specialist AFTFM Meseret Markos Procurement assistant (b) Ratings of Project Performance in ISRs No.Date ISR Archived DO IP Actual Disbursements (XDR million) 1 12/03/2007 S S 2.61 2 01/25/2008 S S 2.75 3 06/28/2008 S MU 4.07 4 02/20/2009 MU MU 7.72 5 06/30/2009 MS MS 8.69 6 12/20/2009 MS MS 11.72 7 06/23/2010 MS MS 12.29 8 10/27/2010 MS MS 15.77 28 9 05/10/2011 MS MS (100%) 16.81 29 (c) Staff Time and Cost Staff Time and Cost (Bank Budget Only) US$000 (including Stage of Project Cycle No. of staff weeks travel and consultant costs) Lending FY07 – FY08 18.54 104,558 Total: 18.54 104,558 Supervision/ICR FY08 – FY 12 68.05 252,317.84 Total: 86.59 356,875.84 30 Annex 5. Summary of Borrower's ICR HIV/AIDS PREVENTION & CONTROL OFFICE (HAPCO) Background 1. In response to HIV/AIDS pandemic the government of FDRE and partners are implementing various projects and programs for more than a decade. One of such partners is World Bank. World Bank was pioneer to initiate multi-sectoral HIV/AIDS prevention and control Project in the Ethiopia. The first project supported by the Bank was MAP-I or EMSAP-I. The total amount of fund allocated for this project was US$ 61.7 Million. The Project became effective in 2001 and closed in December 2006. 2. Based on lessons learned from implementation of EMSAP-I and other donor activities, as well as on evidences from other countries the importance of implementing interventions focusing most-at risk groups (MARPs) and in ―hotspots‖ became evident. Cognizant of this fact World Bank continued to support HIV/AIDS prevention and control response of the country by developing EMSAP-II project. The Bank continued its support with the following rationales.  At the time of the initiation of EMSAP2 the epidemic was progressing at a slower rate, but it will take number of years to see noticeable declines in both prevalence and incidence.  To reduce the incidence as well as prevalence it needs consorted effort of all actors.  If the potential impacts of the epidemic are not controlled, it would have devastating effects on economic growth, social wellbeing, life expectancy etc.  Government and its many partners acknowledge the need to maintain the gains and further scale-up interventions through well-coordinated planning and better program implementation. 3. When World Bank decided to continue supporting the project, it ensured that the country complies with the following Bank Multi- country AIDS Program (MAP) repeater requirements:  The existence of one national HIV/AIDS strategy and one national HIV/AIDS authority.  The consolidation of a comprehensive national monitoring and evaluation (M&E) system for HIV/AIDS.  Willingness to use simplified implementation procedures to facilitate a satisfactory disbursement pace.  Willingness to integrate performance considerations into fund allocation and disbursement decisions. 4. EMSAP-II was signed on March 30, 2007 between IDA and Ministry of Finance and Economic Development of FDRE. It became effective in July 2007 and it was closed on September 30/2011. The main objectives of the project were:  To Increase access to prevention services for youth, in particular females aged 15-24, and other most-at-risk groups; and  To sustain access to care and support for PLHIV and orphans undertaken in EMSAP-I. 31 5. To complement increasing prevention services for these vulnerable and at risk groups, the project designed to support community-based interventions focused on changing behavior, such as partner reduction, fight against Female Genital Mutilation, increased condom utilization, and decrease in gender-based violence. The project also focused on rural and peri- urban areas where the majority of the population lives and on specific ―hotspots‖ such as market towns, secondary towns, and transportation hubs. 6. EMSAP-II had the following three components. Component 1: National Program Coordination and Institutional Strengthening: The objective of this component was to further improve the coordination and delivery of the national HIV/AIDS strategy and action plan at all levels. Project support in this area was in line with the national capacity building program of the Government and addresses key areas of capacity gaps in the implementation of the national HIV/AIDS strategy. In addition, it strengthened the monitoring and evaluation system at all levels, supported specific operations research and other impact evaluation and assessment activities Component 2: Multi-Sectoral Prevention Fund: This component was aimed at supporting a broad range of interventions in the public sector to prevent the spread of HIV. The main activities supported included behavioral change targeted activities, peer education, HCT, condom distribution, and mechanisms for reduction of harmful traditional practices. The support e targeted the most important high-risk and vulnerable population. Financial support was provided through a competitive process based on submission of proposals by the respective ministries and agencies. Component 3: Local Response Fund: This component supported community and woreda- based initiatives focused on primarily HIV prevention and mitigation activities targeting specific key at-risk groups. The types of activities financed under this component included HIV/AIDS awareness creation at community level; behavioral change (reduction of traditional harmful practices; safer sexual behavior by both SWs and their clients; delayed first sexual intercourse; reduction in early marriage; reduced sex with non-regular partners); condom distribution; and counseling and testing services. 7. According to the signed project agreement, the initial project fund of EMSAP-II was SDR 20 million (US$ 30 million). This was reduced to 25 million US$ after 5 million US$ had been moved to the Country’s Food Security Response. However due to gains from exchange rate, the total amount of fund utilized was 25,997,472.12 US$. This is 104% from the signed amount and almost 100% of the project fund. This suggests that there was good performance in implementing project activities and utilizing allocated project fund. 8. The project was implemented in 417 (52% of the total) woredas drawn from 9 regional states and 2 city administrations. The project reached about 8000 kebeles from the selected project woredas. Besides at federal level the project was implemented by 17government offices and 7 Civil Society organizations. 32 Project Implementation and achievement status of each component A. Component-I: Capacity Building and Coordination 9. Component-I fund was allocated for capacity building and enhancing coordination The objective was to improve the capacity of implementers at all level by supporting the establishment of and strengthening relevant structure and deploying able human resources. 10. The total amount of fund utilized for financing this component was Birr 74,689,085, and mainly covered the cost of goods, trainings and consultancy and operating cost including salary for staff. 11. This facilitated implementation process of the project. Other major activities included  training to regions on financial management focusing on settlement of advances by disbursement based on the Bank modification of financial management . Support was alo provided for two rounds training on procurement management in Ghana.  Joint Review meetings were conducted quarterly in harmonized manner  AIDS council, management board and review board were functional having regular meetings according to their schedules.  The M&E system has been established and became operational. Accordingly implementers were submitting their reports using developed formats on quarterly, biannually and annually basis. Supportive supervisions were conducted by FHAPCOs different directorates on quarterly basis.  All regions hired finance officers and auditors with the support from the project. Few regions employed additional program and M&E officers by the project at regional and zonal levels.  In all regions the necessary institutional arrangements that have been established. ( like Regional AIDS councils, management boards and review boards ) were supported and strengthened. B. Component 2: Multi-Sectoral Prevention Fund 12. This component was aimed at supporting a broad range of interventions in the public sector. The main supported activities included behavioral change targeted activities, peer education, HCT, condom distribution, and mechanisms for reduction of harmful traditional practices. The support targeted the most important high-risk and vulnerable groups. The amount of fund utilized for this component was Birr 57,178,527. At federal level 17 government public offices after submitting their proposals and won the competitive process accessed fund from this component. They assigned focal persons and conducted trainings on mainstreaming and other HIV/AIDS related issues. Few conducted researches to find out the impact of HIV/AIDS in their organizations. interventions targeting MARPs. The major sector offices include Ministry of Federal Affairs, National Lottery Administration, Ethiopian Mapping Agency, Ministry of Finance and Economic Development, Ministry of Water Resources, Environmental Protection Agency, National Archives & Library Agency, Transport Authority, Ethiopian Electric Power Corporation, Rural Energy Development& Promotion Center, Ministry of Youth and Sports and Ethiopian Road Authority Ministry of 33 Mines & Energy, House of Representatives Secretariat ,Food Medicine and Health Administration and Control Authority 13. Almost in all regions major MARPs focused government sectors were given due attention. The sector offices were selected through the review process since their proposals address behavioral change targeted activities, peer education, counseling and testing, condom promotion and distribution and other related activities which focus on vulnerable groups and most at risk populations. 14. The most important partners were bureaus of education, bureaus of women, youth and children affairs and bureaus of youth affairs implemented strategic preventive interventions. . 15. The component fund was used to strengthen mainstreaming of HIV/AIDS prevention and programs in key public sector offices. It helped to establish AIDS fund through allocation of 2% from recurrent budget, assigning of focal persons and preparation and execution of sector specific HIV/AIDS annual plans. 16. Activities implemented in schools strengthened HIV/AIDS prevention responses among the in- school youth. Youth centers were established and equipped with the project fund.IGA intervention implemented by some bureaus has helped to economically empower poor women and older orphans. C. Component 3: Local Response Fund 17. This component mainly supported community and woreda as well as kebele based initiatives focused on primarily HIV prevention and impact mitigation activities targeting specific key at-risk population group. Civil society organizations (CSOs), NGOs and FBOs were also supported by this component. The total amount of fund utilized for this component was Birr 189, 920,139. At federal level 7 CSOs, FBOs and NGOs were supported. 18. Major activities carried out at kebele level included community conversation, , youth dialogue, support to PLWHIV and OVC, condom promotion and distributions. 19. In each kebele there was at least one CC site while in some they had up to 4 CC sites. In each site 50 participants conducted community conversations twice per month on various HIV/AIDS related issues. Condoms procured by the component fund were also distributed at woreda and kebele levels. 20. Some of the major challenges encountered are inadequate and irregular fund disbursement; high turnover of staff; limited opportunity of capacity building trainings, marginal involvement of zones and occasional changes in woreda boundaries 21. In conclusion, based on the project targets indicators, all set targets have been achieved and this was confirmed by the last supportive supervision mission of the World Bank. The project has significantly contributed for the overall national response . Even though the 34 project fund compared with others was small, the results are far reaching and instrumental. The project was catalyst in effectively implementing others programs and strengthening community response. . 22. It is recommended to  Exert all efforts to maintain and strengthen the already established institutional arrangements at all level.  Give due attention to public sectors. to strengthen mainstreaming and establish fund raising mechanisms  foster community response. 35 36 Annex 6. List of Supporting Documents  Project Appraisal Document World Bank 2007 (Report number 37790-ET)  Additional Financing Project Paper  Aides Memoires (numbers 1-5)  Implementation Status Reports (number 1-9)  Mid Term Review Report, March 2009  Interim Financial Reports and Annual Audit reports  EMSAP I, Development Credit Agreement between Federal Democratic Republic of Ethiopia and IDA  Annual HIV/AIDS Monitoring and Evaluation Report July 2010 – June 2011  Antenatal Care Sentinel HIV Surveillance in Ethiopia, 2009 Round (Ethiopian Health and Nutrition Research Institute, 2011)  HIV/AIDS Home Care Handbook - Supporting primary care givers in Ethiopia - Ministry of Health - January 2002.  Guideline for implementation of antiretroviral therapy in Ethiopia - Ministry of Health - January 2005.  Ethiopian HIV/AIDS National Response 2001-2005 - Consolidated National Report of the Joint Mid-Term Review, HAPCO, March 2003  National Monitoring and Evaluation Guideline for the Multi-Sectoral Response to HIV/AIDS in Ethiopia - FHAPCO.  Survey on the Status of Sexual Behaviour , Knowledge about HIV/AIDS and Condom Utilization in Thirty Five Woredas (Babu General Trading- Consultancy Service Division)  Sexual Behaviours and Predisposing Factors among Ethiopian University Students, May 2011 (Universities of Gondar,Haram ya, Awasa, Jimma, Mekele)  Growth and Transformation Plan-2010/11—2014/2015 Volume I: Main Text MoFED 2010 Addis Ababa  2009ANC Based HIV Surveillance-EHNRI – paper presented HAPCO Review Meeting Awasa September 21-23; 2010  HIV Prevention Package MARPs and Vulnerable Groups-HIV Prevention Package – MARPs and Vulnerable Groups FHAPCO September 2011  Tool Kit For Positive Change - for Orphans and Vulnerable Children  Community Health Information System – Data Recording and Reporting - User`s Manual FMoH 2011  Ethiopia Young Adults & Gender Survey Addis Abeba UNFPA / Population Council 2010  People Living with HIV Stigma NEP +2010 ( draft)a INDEX—Ethiopian Stigma and Discrimination Survey Report  Road Map for Intensifying Multi-Sectoral HIV/AIDS Response in Ethiopia 2010-2014 FHAPCO August 2010 37 Annex 7. Methodology, Results, and Conclusions of the Special KAP Survey Commissioned by HAPCO on HIV/AIDS. A. Background 1. Due to unanticipated delays in conducting a demographic and health survey (DHS), FHAPCO commissioned a special survey on ―The Status of Sexual Behavior, Knowledge about HIV/AIDS and Condom Utilization In Thirty-Five Woredas.‖ This knowledge, attitudes and practices (KAP) survey was intended to obtain information on three outcome indicators of EMSAP-II project. FHAPCO commissioned a consulting firm, Babu General Trading PLC, to conduct this KAP survey. 2. The general purpose of this survey was to assess the knowledge, attitude and practice of the target groups regarding HIV and AIDS. Specifically, the survey was to 1) assess the comprehensive knowledge of the target groups on HIV and AIDS; 2) ascertain the target groups’ knowledge on HIV prevention methods and HIV transmission modes; 3) appraise the prevailing misconceptions of the target group about HIV and AIDS; and 4) estimating the level of condom utilization among the target groups. B. Methodology Primary data collection through structured surveys Survey coverage 3. The assessment survey was designed and conducted to maximize its representativeness in terms of both respondents and variables. Maximum effort has been exerted to make its coverage and scope as much technically inclusive as was possible. Being so, it covered all the areas enumerated by the client. 4. In determining the coverage and scope of the KAP survey, the geographic, demographic and sectoral diversities of the regions were the primary considerations. The survey covered localities that are within urban and pastoralist settings. Taking the lifespan of localities as a factor, it equally embraced sites that are relatively old and new given the presumed difference in levels of urbanization that derive from the length of period in which the locality has been settled. 5. Considering all these factors, 35 woredas were selected from nine regions and two city administrations of FDRE. Sampling frame and designs 6. The target groups of the survey were youth in the age group of 15 – 24 years, including both in- school and out-of school youth. Since there are two types of target groups, in- and out-of school youth sub-groups, two sampling frames were used. The school roaster constituted the sampling universe for the in-school youth while the list of the localities in the KAs was the source from which the out-of-school youth were drawn after a mapping exercise which identified congregation sites and times. 7. In designing the sample for the determination of the sample size for the two sub-groups, the prime consideration has been the factor of access to information services on reproductive health. Equally 38 important has been the assumption that each segment has a significant diversity within it deriving from socio-economic status. 8. A two-stage sampling procedure was employed. While the first procedure involves the determination of enumeration areas or KAs, the second stage was the selection of the ultimate individual respondents using standard probability sampling. Sample Size: 9. In consultation with the client, a total of 1,225 sample size was set. This total sample size was equally distributed among the thirty-five Woredas. The survey group was also equally divided between male and female respondents with the latter category subsuming FSWs. Survey Instruments 10. Appropriate instruments for the collection of data from various sources were developed. Two types of formats were used. The first is the data collection format, which had been used during the interaction with the target audience at the field level in the participatory approach. The second is semi- structured format. This format was used as a checklist to guide the qualitative appraisal in the conduct of FGDs and panel interviews. 11. The survey instrument included the variables most relevant to revealing the status of KAP towards HIV and AIDS as well as the evaluation of past performance. There have been other instruments geared towards capturing the available (local) response capabilities to HIV/AIDS. These (local) response capabilities in terms of prevention include the establishment of anti-AIDS clubs, the proliferation of associations of PLWAs, Community Conversation (CC) increasing and sustaining care and support services as well as the institutionalization of monitoring and evaluation systems. Participatory Appraisal through non-structured survey 12. The participatory approach helped to generate such important information that could not be easily captured through the structured survey and also provided greater flexibility to the experts to investigate new conditions that may not have been anticipated in the planning of the study and designing the instruments. 13. FGDs were undertaken with different segments of the target youth groups in order to have a fair representation of the various cross-sections of the age cohort. The consulting team held one FGD in each woreda selected for the structured survey. On average, about ten youth participated in these FGDs. Supervisors purposively selected the participants with sensitivity to gender and social station. Secondary data collection from stakeholders 14. These quantitative and qualitative sources of data were further enriched by the information provided by active stakeholders working in the front lines, including local HAPCO, the anti-AIDS clubs in schools, the local chapters of the national and regional associations of PLWAs as well as front line peer educators. They provided crucial information about the factors that influence the sexual behavior of the local youth, the possible sources of misconception, and the degree of acceptance of HIV/AIDS not only as a healthcare issue but more importantly as a socio-economic phenomenon as well as the available care and support services within the local context. Equally important has been the information provided by these stakeholders regarding the nature and effectiveness of the intervention currently underway their respective localities. Data Validation 15. The consulting team reviewed the data collected from the different sources to verify their validity at the field level. This helped track data inconsistencies and allow timely corrections. 39 C. Results Percentage of Young People aged 15-24 who both correctly identify ways of preventing the sexual transmission of HIV and who reject major misconceptions about HIV transmission (All Respondents N = 1223 ) Male Female Total Region ABC+NMC Total % ABC+NMC Total % ABC+NMC Total % Tigray 55 71 77% 55 69 80% 110 140 79% Afar 30 35 86% 25 35 71% 55 70 79% Amhara 93 106 88% 90 104 87% 183 210 87% Oromiya 145 175 83% 140 176 80% 285 351 81% Somali 25 34 74% 25 36 69% 50 70 71% B_Gumuz 15 17 88% 13 18 72% 28 35 80% SNNPRS 81 99 82% 82 108 76% 163 207 79% Gambella 13 17 76% 14 18 78% 27 35 77% Harari 15 17 88% 14 18 78% 29 35 83% Addis Ababa 18 18 100% 15 17 88% 33 35 94% Dire Dawa 15 17 88% 14 18 78% 29 35 83% Total 505 606 83% 487 617 79% 992 1223 81% Percentage of Young People aged 15-24 who reported using condom during last sexual encoumter (last 12 months N = 930) Yes No Total Region N % N % N % Tigray 100 93% 8 7% 108 100% Afar 38 72% 15 28% 53 100% Amhara 150 91% 15 9% 165 100% Oromiya 240 92% 20 8% 260 100% Somali 50 89% 6 11% 56 100% B_Gumuz 24 92% 2 8% 26 100% SNNPRS 125 83% 26 17% 151 100% Gambella 20 80% 5 20% 25 100% Harari 28 85% 5 15% 33 100% Addis Ababa 20 87% 3 13% 23 100% Dire Dawa 24 80% 6 20% 30 100% 40 Grand Total 819 88% 111 12% 930 100% Percentage of female sexual workers who reported use of condom during last sexual encounter(last 12 months N = 557) Yes No Total Region N % N % N % Tigray 58 98% 1 0% 59 100% Afar 32 97% 1 0% 33 100% Amhara 103 99% 1 2% 104 100% Oromiya 150 98% 3 3% 153 100% Somali 34 94% 2 0% 36 100% B_Gumuz 14 93% 1 0% 15 100% SNNPRS 93 98% 2 5% 95 100% Gambella 13 100% 0 8% 13 100% Harari 17 100% 0 0% 17 100% Addis Ababa 14 100% 0 0% 14 100% Dire Dawa 17 94% 1 0% 18 100% Grand Total 545 98% 12 2% 557 100% 16. The 3 tables above provide some of the key findings of the survey. The first table suggests that knowledge about prevention of HIV and rejection of common misconceptions is widespread and that there is little difference between men and women and between the regions. The second table suggests that among the sexually active population the use of condoms is widespread and with little regional variation except for Afar. The last table provides data on the practices of more than 500 commercial sex workers and indicates that condom use is widespread and consistent. D. Conclusions 17. The methodology used for the rapid KAP survey commissioned by HAPCO was in keeping with standard practices and the sampling appeared to be reasonably done. The results about knowledge appear somewhat higher than the preliminary data from the 2011 DHS (see para. 38 in the main text) but do not seem wildly out of keeping with the broader national findings. There is room for legitimate skepticism about the results about condom use among sex workers which was also reported to be very high at baseline. Findings such as these are improbably high and likely reflect that CSWs know the right answer and provide it to surveyors. 41 IBRD 35214 32°E 36°E 40°E 44°E ETHIOPIA ERITREA R PRODUCTIVE SAFETY NET REP. PROGRAM e OF d ETHIOPIA YEMEN C C C C FC FC Adigrat SAFETY NET PROGRAM WOREDAS* Humera Axum C F S C C F C TYPES OF SAFETY NET PROGRAMS: 14°N C C FC C 14°N e C F F TIGRAY C FC F F FOOD ONLY C a C FC Mekele C CASH ONLY C C C C F F FC FOOD AND CASH C C C F F F F F FC DISTRIBUTION RESPONSIBILTY, BY WOREDA: Gonder F FC C FC F C C C F WORLD FOOD PROGRAM AMHARA FC F F NON-GOVERNMENTAL ORGANIZATIONS C C F FC Lake C FC F AFAR Tana GOVERNMENT FC FC F C Debra Weldiya Bahir Dar Tabor FC FC C C FC F F DJIBOUTI SELECTED CITIES AND TOWNS FC C C F 48°E F C F NATIONAL CAPITAL C Dese F F C C C C REGION BOUNDARIES C C SUDAN C C C F C F F INTERNATIONAL BOUNDARIES Debre C C F F BENSHANGUL Markos C C C F F *The Bank project will support certain woredas within C C the overall Government program. Data source: World Food Program, April 17, 2006. 10°N F 10°N C DIRE DAWA F Dire Dawa FC F F F C F F This map was produced by the Map Design Unit of The World Bank. C C C F F Jijiga The boundaries, colors, denominations and any other information 48°E ADDIS F C F F Harer C shown on this map do not imply, on the part of The World Bank Gimbi ABABA F C Group, any judgment on the legal status of any territory, or any Nekemte F Awash HARARI endorsement or acceptance of such boundaries. ADDIS ABABA F C F C C F FC C C C C F C F C Nazret C F C SOMALIA Gambela F C C Degeh Bur Aware Gore Welkite C F C C C GAMBELA C Asela F Domo Jima C C C C C F F Hosaina C C C OROMIYA C C C Bonga C C C C F Shashemene F C C Goba C SOMALI C Awasa FC Dodola Warder C C C FC F C C C FC Sodo F Kebri Dehar C F FC Wendo C FC C C F F Imi C C F C FC F F C C C C Arba F 6°N SNNPR C F Minch C 6°N C F F C C Negele C Ferfer F Yavello F F F C Dolo Odo Lake C Mega 0 50 100 150 200 Kilometers Turkana F INDIAN Moyale 0 50 100 150 Miles UGANDA KENYA OCEAN 32°E 36°E 40°E 44°E 48°E DECEMBER 2006