Document of The World Bank Report No: ICR00002554 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-H3280) ON A GRANT IN THE AMOUNT OF SDR, EURO 6.60 MILLION (US$10.0 MILLION EQUIVALENT) TO THE ISLAMIC REPUBLIC OF AFGHANISTAN FOR AN AFGHANISTAN HIV/AIDS PREVENTION PROJECT (P101502) June 21, 2013 Human Development Unit South Asia Region CURRENCY EQUIVALENTS (Exchange Rate Effective January 28, 2013) Currency Unit = Afghani 1.00 = US$0.01934 US$ 1.00 = 51.7000 FISCAL YEAR March 21 – March 20 ABBREVIATIONS AND ACRONYMS AHAPP Afghanistan HIV/AIDS Prevention Project AIDS Acquired Immunodeficiency Syndrome ART Antiretroviral Treatment (Therapy) BCC Behavior Change Communication BPHS Basic Package of Health Services EC European Commission FSW Female Sex Worker GoIRA Government of Islamic Republic of Afghanistan HACCA HIV/AIDS Coordinating Committee of Afghanistan HIV Human Immunodeficiency Virus HMIS Health Management Information System HR Harm Reduction I-ANDS Interim Afghanistan National Development Strategy IBBS Integrated Biological and Behavioral Surveys IBRD International Bank for Reconstruction and Development IDA International Development Association IDU(s) Injecting Drug User(s) IEC Information, Education and Communication KAP Knowledge, Attitudes and Practices M&E Monitoring and Evaluation MDGs Millennium Development goals MOCN Ministry of Counter-Narcotics MOF Ministry of Finance MOJ Ministry of Justice MOPH Ministry of Public Health MOU Memorandum of Understanding MDM Medecins du Monde MSM Men who have Sex with Men MSW Male Sex Worker NACP National AIDS Control Program NGOs Non-Government Organizations OECD Organization for Economic Co-operation and Development OST Opioid Substitution therapy PLWHA People Living with HIV and AIDS STI Sexually Transmitted Infection SW Sex Worker TIs Targeted Interventions TOR Terms of Reference UNAIDS Joint United Nations Program on HIV and AIDS UNODC United Nations Office on Drugs and Crime VCT Voluntary Counseling and Testing WHO World Health Organization Vice President: Isabel Guerrero Country Director: Robert Saum Sector Manager: Julie McLaughlin Project Team Leader: Mohammad Tawab Hashemi ICR Team Leader & Author: Robert Oelrichs ISLAMIC REPUBLIC OF AFGHANISTAN HIV/AIDS PREVENTION PROJECT (P101502) CONTENTS Data Sheet A. Basic Information B. Key Dates C. Ratings Summary D. Sector and Theme Codes E. Bank Staff F. Results Framework Analysis G. Ratings of Project Performance in ISRs H. Restructuring I. Disbursement Graph 1. Project Context, Development Objectives and Design ............................................... 1 2. Key Factors Affecting Implementation and Outcomes .............................................. 5 3. Assessment of Outcomes .......................................................................................... 13 4. Assessment of Risk to Development Outcome......................................................... 23 5. Assessment of Bank and Borrower Performance ..................................................... 24 6. Lessons Learned ....................................................................................................... 27 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners .......... 28 Annex 1. Project Costs and Financing .......................................................................... 29 Annex 2. Outputs by Component ................................................................................. 30 Annex 3. Economic and Financial Analysis ................................................................ 35 Annex 4. Bank Lending and Implementation Support/Supervision Processes ............ 36 Annex 5. Summary of Borrower's ICR and/or Comments on Draft ICR ..................... 38 Annex 6. List of Supporting Documents ...................................................................... 52 Annex 7. Figures ........................................................................................................... 55 A. Basic Information Afghanistan HIV/AIDS Country: Afghanistan Project Name: Prevention Project Project ID: P101502 L/C/TF Number(s): IDA-H3280 ICR Date: 05/03/2013 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: AFGHANISTAN Original Total XDR 6.60M Disbursed Amount: XDR 6.40M Commitment: Revised Amount: XDR 6.60M Environmental Category: B Implementing Agencies: Ministry of Public Health Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 10/31/2006 Effectiveness: 09/17/2007 09/17/2007 Appraisal: 04/16/2007 Restructuring(s): Approval: 07/31/2007 Mid-term Review: 02/03/2010 02/11/2010 Closing: 12/31/2010 11/30/2012 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Substantial 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: Satisfactory Implementing Quality of Supervision: Satisfactory 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 Indicators Rating Performance (if any) Potential Problem Project Quality at Entry Yes None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes None time (Yes/No): Supervision (QSA): DO rating before Moderately Closing/Inactive status: Unsatisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 26 26 Health 70 70 Other social services 4 4 Theme Code (as % of total Bank financing) HIV/AIDS 67 67 Health system performance 33 33 E. Bank Staff Positions At ICR At Approval Vice President: Isabel M. Guerrero Praful C. Patel Country Director: Robert J. Saum Alastair J. McKechnie Sector Manager: Julie McLaughlin Julian F. Schweitzer Project Team Leader: Mohammad Tawab Hashemi Mariam Claeson ICR Team Leader: Robert Oelrichs ICR Primary Author: Robert Oelrichs F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The project development objective is to slow down the spread of HIV and build up the national capacity to respond to the epidemic by i) behavior change among vulnerable groups at high risk, and ii) improving knowledge of HIV prevention and reducing stigma related to HIV and AIDS in the general population. Revised Project Development Objectives (as approved by original approving authority) Not revised (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Indicator 1 : HIV prevalence in the general population at less than 0.5% Value quantitative or <0.5% <0.5% <0.1% Qualitative) Date achieved 06/30/2007 12/31/2010 11/30/2012 Comments (incl. % 100% of target achieved. achievement) Indicator 2 : HIV prevalence amongst high risk groups (IDU and SW) less than 5% IDU = 4.4%, FSW Value IDU:7.1%, SW: 0% = 0.3%, MSM = quantitative or <5% (IBBS 2009) 0.4%, Prisoners = Qualitative) 0.7% Date achieved 06/30/2007 12/31/2010 11/30/2012 Comments (incl. % 100% of target achieved. achievement) Percentage of health workers who correctly identify ways of preventing HIV Indicator 3 : transmission Value 19.3% (Health facility quantitative or 75% Data not available assessment 2011) Qualitative) Date achieved 12/30/2011 06/30/2012 Comments Data were to be collected via health facility assessment by March 2013. The draft (incl. % version available at the time of this ICR did not include HIV indicators. achievement) Among health workers who have heard of HIV, the percentage who are willing Indicator 4 : to care for patients with HIV Value 24.4% (Health facility quantitative or 80% Data not available assessment 2011) Qualitative) Date achieved 12/30/2011 06/30/2012 Comments Data were to be collected via health facility assessment by March 2013. The draft (incl. % version available at the time of this ICR did not include HIV indicators. achievement) Indicator 5 : Percentage of SW reporting use of condom with their most recent client Value quantitative or 58% (IBBS 2009) 50% >60% 52.3% (IBBS 2012) Qualitative) Date achieved 06/30/2009 12/31/2010 06/30/2012 11/30/2012 Comments (incl. % 87% of target achieved. (In IBBS 2012 the indicator is “Condom use at last sex�) achievement) Percentage of IDUs reporting use of sterile injection equipment at last time of Indicator 6 : injection Value quantitative or 95% (IBBS 2009) 30% increase >80% 92.9% Qualitative) Date achieved 06/30/2009 12/31/2010 06/30/2012 11/30/2012 Target exceeded by 15%. The validity of the 2009 IBBS data on this indicator was questioned at MTR and judged to be probably inaccurate. This indicator was Comments not assessed in the full IDU IBBS sample in 2012, only in two prison sites, (incl. % giving a rate of 41%. The figure used here is instead derived from IBBS indicator achievement) of “Used non-sterile injecting equipment at least once in the past 3 months� is used, which is used as a proxy for national data. Indicator 7 : Percentage of IDU reporting use of condom at last time sex Value 23.9% (crude quantitative or 27% IBBS 2009 20% increase >40% average of IBBS Qualitative) 2012) Date achieved 06/30/2009 12/31/2010 06/30/2012 11/30/2012 Comments (incl. % 60% of target achieved. achievement) (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Indicator 1 : IBBS maintained and expanded to southern and eastern sites of the country Value (quantitative 4 6 6 or Qualitative) Date achieved 06/30/2009 06/30/2012 11/30/2012 Comments (incl. % 100% of target achieved. achievement) Indicator 2 : Number of Drug Users enrolled in OST. Value (quantitative 0 200 71 or Qualitative) Date achieved 06/30/2007 06/30/2012 11/30/2012 Comments (incl. % 36% of target achieved. achievement) Indicator 3 : Number of urban centers which have done high risk group mapping Value (quantitative 3 cities 33% of provinces 4 cities 6 cities or Qualitative) Date achieved 06/30/2007 12/31/2010 06/30/2012 10/04/2012 Comments Target was exceeded by 50%. (incl. % achievement) Percentage of most at risk populations (e.g., IDUs, FSWs, prisoners) reached by Indicator 4 : HIV prevention program. IDU: 62.4% IDU: 60% IDU: 70% IDU: 90.2% Value FSW: 1.4% FSW: 50% FSW: 10% FSW: 8.3% (quantitative Prisoners: NA Trucker: 25% Prisoners: Prisoners: NA or Qualitative) (IBBS 2009) Prisoners: 80% 80% (IBBS 2012) Date achieved 06/30/2009 12/31/2010 06/30/2012 11/30/2012 Target partially achieved (129% in IDU, 83% in SW and data not available in Comments prisoners). The indicator is based on the question in the IBBS survey “Use of risk (incl. % reduction service among those corresponding who have heard about the achievement) services�. Percentage of most at risk populations (e.g. IDUs, FSWs, prisoners) who Indicator 5 : correctly identify two ways of preventing HIV transmission IDU: 71.4% IDU: 50% IDU: 75% IDU: 22.1% Value FSW: 12.8% FSW: 50% FSW: 20% FSW: 15.3% (quantitative Prisoners: 25.4% Trucker: 50% Prisoners: Prisoners: 6.6% or Qualitative) (IBBS 2009) Prisoners: 75% 30% (IBBS 2012) Date achieved 06/30/2009 12/31/2010 06/30/2012 11/30/2012 Comments 29% of target achieved among IDUs, 76% among FSWs and 22% among (incl. % prisoners. achievement) Annual NACP reports and action plans informed by surveillance and routine Indicator 6 : monitoring data Value Annual reports and Annual reports and (quantitative 0 actions actions or Qualitative) plans available plans available Date achieved 06/30/2007 12/31/2010 10/04/2012 Comments (incl. % Target 100% achieved. achievement) Indicator 7 : TI sites meeting service quality standards IDU: 5 of 6 IDU: 7 of 7 Value >75% of sites FSW:1 of 1 FSW: 1 of 1 (quantitative 0 Truckers Trucker: 1 of 1 Data for prisoners or Qualitative) dropped Prisoners: 1 of 1 available Date achieved 06/30/2007 12/31/2010 06/30/2012 10/04/2012 Comments (incl. % Target exceeded by 25%. achievement) Indicator 8 : Financial Management reports include physical progress report Financial Management Value Financial Management reports include (quantitative reports did not include Done physical or Qualitative) physical progress reports progress reports Date achieved 06/28/2009 06/30/2012 10/04/2012 Comments (incl. % Target was 100% achieved. achievement) G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 01/14/2008 Satisfactory Moderately Satisfactory 0.13 Moderately 2 07/13/2008 Moderately Satisfactory 0.60 Unsatisfactory 3 01/18/2009 Moderately Satisfactory Moderately Satisfactory 1.54 4 08/02/2009 Moderately Satisfactory Moderately Satisfactory 3.26 5 04/14/2010 Satisfactory Moderately Satisfactory 5.43 6 01/15/2011 Satisfactory Moderately Satisfactory 6.58 7 09/07/2011 Satisfactory Moderately Satisfactory 8.07 8 06/05/2012 Moderately Satisfactory Moderately Satisfactory 9.07 Moderately 9 11/20/2012 Moderately Satisfactory 9.99 Unsatisfactory H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions Extension of the closing date by 18 months to June 30, 2012. 06/28/2010 N S MS 5.43 Cancellation of part 4.2 in Schedule 1 to the FA (the Innovation Initiative). Extension of the closing date by 06/28/2012 N MS MS 9.38 5 months to 11/30/2012 I. Disbursement Profile 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal In 2007, an assessment of the transmission of the human immunodeficiency virus (HIV) in Afghanistan revealed a disturbing picture. The levels of high risk behavior in key populations indicated there was fertile ground for an explosive epidemic within these high risk groups, as had been seen in other countries of the region. Little national capacity or resources were present to mitigate this risk, as the country grappled with the consequences of decades of war and instability. Alarmingly, a survey in 2007 had shown a sudden emergence of HIV infection in injection drug users (IDU) 1 , and fertile ground for transmission to accelerate in other key populations. A paper published in the Lancet (Saif- ur-Rehman, 2007) by representatives of the Government of Islamic Republic of Afghanistan (GoIRA) and the World Bank highlighted international concern that, without swift action, an epidemic of HIV in IDU would be precipitated, with risk of significant spread to the general population. The purpose of the Afghanistan HIV/AIDS Prevention Project (AHAPP) was to assist the Government of Afghanistan in achieving the national development goal of maintaining HIV prevalence below 0.5 percent in the general population and below 5 percent among vulnerable groups at high risk. HIV Epidemic and Response At appraisal, in 2007, the HIV epidemic was judged to be at an early stage in Afghanistan, concentrated among high risk groups, mainly IDU and their partners. Although HIV prevalence was low, there was high potential for rapid spread due to increases in injecting (as opposed to non-injecting) drug use. A study among IDU in Kabul city (Todd, 2007a) had found that three percent of the IDU were HIV positive. In Quetta, a town in Pakistan bordering Afghanistan, a 24% prevalence of HIV infection among IDU was reported (Achakzai, 2007). UNAIDS and WHO estimated that there were potentially already 1,000 to 2,000 HIV-positive cases in Afghanistan at this time. Specific data were lacking – especially the population size, degree of risk and accurate HIV prevalence in key populations, including sex workers (SW) and men who have sex with men (MSM). However, global and regional experience at that time indicated that each of these groups were likely to be vulnerable to a rapidly expanding epidemic – and that this might in turn lead to increased HIV transmission in the general population of Afghanistan. IDU-driven epidemics in particular had been shown to ignite suddenly, sometimes after long periods of latency. For example, between 1995 and 1997, HIV-1 seroprevalence among the IDU of Kathmandu, Nepal had risen from 0% to between 40-50% (Oelrichs, 2000). 1 Note – During the preparation and implementation of this project, the term “Injecting Drug Users� (IDU) was most frequently employed by the Bank and UN agencies to denote those individuals at high risk of HIV infection, as is reflected in this document. The term “People Who Inject Drugs� (PWID) is now preferred (UNAIDS). 1 Despite the absence of definitive data, other key populations were thought likely to be vulnerable to HIV infection - including inmates of Afghan prisons, long distance truck drivers and street children. Several structural amplifiers were identified, that potentially heightened the national risk for rapid spread of HIV. These included chronic war and conflict and attendant migration, displacement and poverty. At that time, almost four million Afghan refugees still lived in Pakistan and Iran, countries which both had rapidly growing IDU-driven HIV epidemics. In addition, the literacy rate in the general population was very low, with little awareness about HIV/AIDS and almost no condom use. Unsafe blood transfusion added to the risk of spread to general population, with only 30 percent of transfused blood being tested for HIV. A particular contextual risk was presented by the local production of opium – which exceeded 6,000 tons in 2006, representing 92 percent of the global total. Increasing use of heroin in Afghanistan and the focus of counter-narcotic efforts on supply reduction, was thought to increase the risk of opium users turning to injecting heroin, thus amplifying the risk of HIV transmission. The project was thus proposed and planned in a context of urgency, recognizing that action was needed to avert a much larger epidemic – and putting acquisition of categorical epidemiological data at the heart of project activities. A critical consideration at appraisal was the perceived very low capacity of the GoIRA. Although the Ministry of Public Health (MOPH) had developed a national strategic plan, it had very little capacity or resources to implement an effective HIV prevention program at any scale. This was identified as a major risk and therefore it was decided to implement project activities mainly through experienced international and national Non-Government Organizations (NGOs). At that time, the Basic Package of Health Services (BPHS), which was aimed by MOPH to cover 90 percent of the population, was largely delivered by contracted NGOs. A transparent and efficient mechanism for contracting out of services to NGOs had already been established; hence, it was decided to build on this experience – also contracting out the important surveillance and communications components to firms with proven track-records in building capacity and implementing these activities, and in working with the National AIDS Control Program (NACP). Rationale for Bank Assistance Prior to this project, HIV/AIDS prevention programs had been fragmented and small scale in Afghanistan. A few local and international NGOs and development partners provided prevention services to high-risk and vulnerable populations - mainly HIV prevention interventions for IDU, including demand reduction activities. A limited number of interventions had also been designed and launched among SWs, MSM, truck drivers, and police and prison staff. Government policies and strategies were favorable - one of the pillars of the Interim Afghanistan National Development Strategy was the goal to keep HIV prevalence in the general population at less than 0.5 percent. The Bank had been a lead player in Afghanistan’s health sector since 2003 and was playing an instrumental role in assisting the MOPH to rebuild the basic health service system in Afghanistan through policy dialogue, technical assistance and funding support for the delivery of the BPHS in 11 provinces. Through these efforts, the Bank was perceived by 2 development partners to play a critical role in the health sector – as it had gained extensive experience and knowledge on delivery of health services in this fragile country context. At the time that the GoIRA requested financial assistance for HIV/AIDS Prevention, the Bank had already provided some support to build the capacity of the NACP. The Bank had also supported some formative analytic work – notably the social mapping of groups at high risk of HIV infection. At appraisal, the main rationale for Bank involvement in supporting the national HIV/AIDS response was the lack of sufficient financial support for HIV/AIDS programming from other sources as well as the Bank’s considerable experience in assisting countries in South Asia to deal with HIV systematically. The role of other development partners in financing the HIV response was also taken into account in estimating the overall Bank support for the operation. The IDA grant was calculated to provide a third of the budgeted national program over a three year period, with minor support for blood bank safety secured from French Cooperation and more anticipated from the Global Fund for AIDS, TB and Malaria (GFTAM). 1.2 Original Project Development Objectives (PDO) and Key Indicators The project was intended to “contribute to the national development goals of the Interim Afghanistan National Development Strategy of maintaining HIV prevalence below 0.5 percent in the general population and below 5 percent among vulnerable groups at high risk of infection� (PDO Indicators 1 and 2). The project’s development objectives were to slow down the spread of HIV and build up the national capacity to respond to the epidemic. This would be accomplished by: (a) behavior change among vulnerable groups at high risk; and (b) improving knowledge of HIV prevention and reducing stigma related to HIV and AIDS in the general population�. The key performance indicators for the project, as presented in the Appraisal Document were:  The percent of injecting drug users who have adopted behaviors that reduce transmission of HIV, that is, who use clean injecting equipment at last time injecting;  The percent of sex workers who report using a condom with their most recent client; and  The percent of young people aged 15-24, in areas covered by the project, who correctly identify ways of preventing HIV transmission. 1.3 Revised PDO and Key Indicators The PDO was not revised. At Mid-Term review (February 2010), the Results Framework (RF) was suggested to be revised, reflecting the finding that several indicators of the original Framework had proven difficult to measure, or provided little added value in understanding the dynamics of the highly concentrated HIV epidemic in Afghanistan. The indicator pertaining to knowledge in the general population (original Intermediate Indicator 9) was dropped, as was Indicator 6 on condom use in truckers, as HIV prevalence had been found to be zero in this group. New indicators were added on knowledge and attitude 3 amongst health workers, patient enrollment in the pilot opioid substitution therapy (OST – also referred to as methadone maintenance therapy) site and financial management. The revised Results Framework was formalized in the project restructuring paper (June 28, 2010) and is reflected in the ICR Datasheet. The resulting RF monitored five of the original seven PDO indicators and four of the original eight Intermediate Outcome indicators. 1.4 Main Beneficiaries The primary target groups of this project were to be the key populations in Afghanistan at risk of HIV infection, hence primarily the socially excluded groups of sex workers and IDU – also including interventions for truck drivers, and prison inmates. However it was the intention of the project to allow the focus of activities to be guided by evidence. Once sound epidemiological data on risk and HIV prevalence had been obtained from the project’s first Integrated Bio-behavioral Survey (IBBS), the focus of interventions would be adjusted. Much of the activity in Communication and Advocacy (Component 1) was also focused on promoting a policy environment that enabled these interventions in key populations to proceed. As stated in the PDO, activities directed to key populations were also anticipated to control the spread of HIV in the general population – which by extension was a stated beneficiary. The first PDO indicator (population HIV prevalence) emphasized the importance of this goal, and the original intermediate outcome indicator quantifying young people’s knowledge on preventing HIV transmission made it specific. 1.5 Original Components (as approved in Technical Annex: Report No: T7698 - AF) (See full details under Annex 2) Component 1: Communications and Advocacy (US$1.06 million – total estimated baseline costs). This component was intended to strengthen the advocacy and communication capacity of the NACP to create a policy environment that enabled scaling up of targeted interventions (TIs) for vulnerable populations at high risk, and reduced stigma and discrimination related to high risk behaviors. The component included Communications activities, Behavior change communications and Advocacy activities. Component 2: Strengthening of HIV Surveillance (US$1.61 million – total estimated baseline costs). This component was intended to build the evidence base for HIV and AIDS planning and help monitor the effectiveness of the program. It included separate mapping and size estimation of at risk populations, collection of biological data from representative samples of the populations, and tracking the behaviors that contribute the most to HIV transmission dynamics). Component 3: Targeted Interventions for High Risk Behaviors (US$4.06 million – total estimated baseline costs). This component was intended to support TIs to prevent further spread of HIV among vulnerable groups at highest risk, promoting safe practices and reducing risky ones. TIs for IDU, sex workers and their clients, truckers and prisoners included a) gathering of baseline data to estimate the size of the populations to be covered; b) establishment and operation of appropriate services such as, in the case of HIV prevention among IDU, drop in centers (DIC), social support, needle and syringe exchange, 4 OST (in one package in Kabul and among prisoners), condom distribution, sexually transmitted infections (STI) care, other medical support, and Voluntary Counseling and Testing (VCT); and c) reporting. Component 4: Program Management, Capacity Building, Monitoring and Innovation Activities (US$1.80 million – total estimated baseline costs). This component was intended to strengthen the core functions of the national HIV/AIDS prevention program, including program management, capacity development, program monitoring by NACP, and oversight of the HIV/AIDS Coordinating Committee of Afghanistan (HACCA). 1.6 Revised Components At the MTR mission in February 2010, it was decided that Component 4 should be revised, which was undertaken in the June 2010 restructuring. Specifically, due to a failed procurement for the Innovation Activities, and the low prospects for an acceptable bid, this sub-component was dropped and the funds were re-programmed to additional interventions for IDU and institutionalization of surveillance. 1.7 Other significant changes During the February 2010 MTR, an agreement was reached to extend the Closing Date of the project by 18 months. The reasons for this were the delayed start to the project, which had been rapidly prepared under Emergency Procedures. As a result, it was projected that there would be US$30 million unspent and a number of critical activities uncompleted, if the original closing date has been kept. A second extension of a further five months was granted in June, 2012 – extending the project until the end of November 2012.This extension was to allow completion of the second round of the IBBS survey – an important surveillance activity under the project. During the MTR, the MOPH and the Bank also reached agreement on a set of priorities for the remainder of the project term. These were informed by the first round IBBS, which had indicated that MSM should be included as a key population, and that interventions for truck drivers should be de-emphasized. The priorities specifically covered a) expansion of high quality harm reduction services among IDU in Herat and possibly other western provinces, b) extension of current harm reduction contracts for IDU and prevention efforts among CSW by 18 months and c) postponement of the second round of IBBS for 1 year, while expanding it to eastern and southern provinces and including MSM, besides other high risk groups. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry The preparation of this project was shaped by three broad considerations. Firstly, in 2006/7, knowledge of the diversity of global HIV epidemics was changing. Although it was already clear that many epidemics outside Africa were contributed to importantly by transmission 5 amongst key populations at risk, it was not yet understood to what extent national epidemic spread in the general population might be limited in such concentrated epidemics. It remained a fear that any significant transmission within a country was a harbinger of an incipient generalized epidemic. This project was at the vanguard of evidence-driven design, placing interventions directed to key populations at the center – but the development objective and key indicators also included general population impact. Secondly, the Afghanistan epidemic was approached with a sense of urgency. The two main reasons for this were firstly, the recent regional experience of explosive epidemic spread in high risk groups, and secondly, alarming (albeit very limited) epidemiological data showing high rates of infection and large vulnerable populations in Afghanistan. Coupled with this was the fear that rapid, unchecked spread between IDU would lead to significant spread in the general population – perhaps with the attendant social and economic sequelae being seen in African countries. Urgently stopping a national epidemic was the objective of this project. Thirdly, the particular national context of Afghanistan as a fragile and conflict-affected state underscored all aspects of project design. The risks attached to an intervention in an insecure and politically unstable situation, were amplified by very limited capacity of the implementing agency. These risks were well recognized by the Bank. The project team also recognized the particular challenges attached to providing services for marginalized and stigmatized populations (sex workers, drug users, prisoners) within a society with little traditional interest in these groups. The proposed interventions themselves – OST especially – were also contentious and, if not experimental, then at least still on the margins of AIDS control programming in many countries. Importantly, the project placed significant emphasis on advocacy and communication activities that would facilitate the development of a favorable policy environment for accessing these populations with this ambitious plan of services. In the end, much of the progress in implementation hinged on critical policy and regulatory bottlenecks, some of which proved intractable to advocacy activities. Given the above mentioned considerations, the following key factors stand out as influencing project preparation and design. a) Fragile security environment - the risks that security posed to the project were clearly and reasonably assessed during project preparation. The country health team was well aware of the impact of uncertain security on operations, through experience with the Health Sector Emergency Reconstruction and Development project. It is worth stating first, and with emphasis, that to pursue this project to deliver services for some of the most disenfranchised populations, within this most fractured society, was a highly commendable aim, consistent with the best aspirations of the Bank. The project was designed to focus its targeted interventions geographically, in four cities: Kabul, Jalalabad, Mazar and Herat, where security had remained more stable. Given that security concerns did not affect this project during implementation to a degree disproportionate to other projects, this judgment was well founded. b) Cultural and political sensitivities surrounding Harm Reduction - Drug use clearly has particular cultural and political connotations in Afghanistan. It was recognized during preparation that the project would need to invest considerable effort in advocacy to 6 sway political opinion towards permitting (if not actively supporting) implementation of needle and syringe services and OST. Although OST was to be only one element of services provided under the project, it was a high-profile intervention due to its innovative approach – and received international attention. The team was probably encouraged by some success of the Bank in catalyzing policy change in China, Viet Nam and elsewhere – as well the spectacular success of harm reduction approaches in Australia (from where consultants contributed expert advice). It is not clear, however, what process was followed to ascertain the critical point of how feasible it might be to bring about a more permissive policy environment in Afghanistan. It is possible that had more time been spent on consultation, with a broader range of stakeholders, the intractable political (and ideological) opposition to OST – which differed from general resistance to providing services for IDU - may have become apparent earlier. c) NGO contracting - Opting to design the project with heavy reliance on NGOs to deliver activities was based on solid experience. At the time, it was recognized as the best option to deal with a situation of extremely limited capacity of Government in providing services or high-risk groups. A transparent and efficient mechanism for contracting out of services to NGOs had been established under the previous Bank supported health project in Afghanistan. Highly capable firms with strong country track-records were available - for example Johns Hopkins University, which undertook the important surveillance components. What does not seem to have been anticipated during design was the degree to which reliance on NGOs placed the timeline of implementation at risk. Although the contracts did contain standard linkage of NGO payments to deliverables, a more finely structured set of incentives for performance may have given the project more leverage to ensure timely provision of services. d) Emphasis on generating epidemiological evidence - It was correct to address the dearth of rigorous HIV monitoring data available in Afghanistan. There was sufficient epidemiological basis from outside the country also to have concern about HIV spread to other vulnerable groups - particularly sex workers, where high risk behavior had been documented. The project very appropriately anticipated that the emphasis of activities should be revised, based on evidence generated of the HIV prevalence and risks of these populations. This did, however, make the project especially vulnerable to delay in timely production of these data. At the design phase it seems clear that the primary purpose of supporting surveillance was programmatic – with the data to be made immediately available and owned by the GoIRA. A quite different emphasis is apparent shortly into the project life, when a very elaborate study was envisaged by the firm, rigorous enough for research quality publication, but commensurately costly and time consuming. This even led to some confusion between the firm and NACP, as to the ownership of data gathered for research purposes. AHAPP project documents clearly note the substantial risks that were assessed during preparation in 2007. Largely the same team was responsible for the concurrent Afghanistan health project financed by the Bank, Strengthening Health Activities for the Rural Poor (SHARP) - in which the team was faced with many of the same contextual implementation risks as in AHAPP. These are nicely summarized in the Emergency Project Paper for SHARP (February 2009), which rated most of them as High. They include a) Overall country inherent risk, b) Growing insecurity disrupting provision of services in certain 7 areas as well as hampering monitoring and evaluation of activities, c) Perceived corruption, d) Political opposition to contracting, e) Weak implementation capacity of MOPH – particularly in financial management and procurement. 2.2 Implementation The assessment of risk during AHAPP project preparation was largely focused on specific implementation challenges to the project. The following mitigation measures were proposed during preparation (as outlined in the Memorandum of the President), resulting in the following experience during implementation: Project Risk Mitigation proposed Experience during Implementation Envisaged Weak Project activities to be The model proved largely successful – although implementation carried out mainly by complexities in procurement resulted in delays with capacity of the experienced NGOs, under a NGO contracting. executing transparent and efficient agency contracting mechanism Security Target interventions Significant delays were experienced in hiring situation may geographically, in four cities: consultants to key positions. The security situation discourage Kabul, Jalalabad, Mazar and was identified as a key constraint to attracting and NGOs and Herat where security to date retaining international consultants. The constrained consultants has remained more stable site of project activities, although essential for from security considerations, made flexible allocation of participating in project resources in response to epidemiological the country, or data more difficult. outside of Kabul The general National AIDS policy will be The policy dialogue proved fruitful at several levels lack of prepared by the MOPH, – including resulting in the development of key awareness and supported by key ministries policies consistent with project objectives. widespread and endorsed by parliament. However, barriers to OST at the policy and service stigma Communications component delivery level proved intractable. The contribution associated with of the proposed project will of the communications components (as distinct from HIV and risky contribute to creating an advocacy activities) is difficult to establish. practices. enabling environment. There is little evidence that any significant risk to AHAPP went unnoticed. Significant risks did receive attention, as was noted in ISRs and other mission documents. There are several additional issues which had an impact on implementation progress, as follows: a) Delays in the IBBS rounds. The data collection activity for the first IBBS of IDU, SWs, and prisoners was initially planned to be completed within the first 6-9 months of the project. As noted above, the IBBS implemented by the consultant firm took considerably longer than this (completed 2009) and suffered procedural delays due to its elaborate and research-focused approach. Despite the expense and delays associated with the first round IBBS, the MTR recommended that future rounds should be expanded, and should be done 8 every two to three years. As it transpired, the second round proved to be even more delayed than the first, faced procurement challenges, and was the main reason for the final five months extension of the project Closing Date (Level two Approved on 28 June 2012). The final report was only submitted five months after the project Closing Date. As a consequence of these delays in the survey, project ISRs (including the last ISR) reflect unchanged data on indicators over much of the project life, as the definitive data to inform these were serially delayed. This was unfortunate as the IBBS was specifically conceived to be a tool to increase efficiency of the project by providing timely data and informing a more accurate tailoring of its response. Instead, the delayed and infrequent data did not facilitate responsive implementation - and in the end delayed the project itself. The use of a much less elaborate but frequent method of surveillance – using a sampling methodology chosen with the priority of generating rapid program-ready data from multiple sites, may have provided more timely data, with little loss of accuracy. In future, the team could consider more closely supporting the client to obtain a package of surveillance services in line with its program needs. b) Contracting of services to local and international NGOs. The advantages of the mechanism of contracting services to NGOs have been discussed above. The proof of the success of this approach is its subsequent adoption in the newly approved SEHAT project. Despite the overall positive results of using this methodology, there were delays in the contracting process - primarily due to lack of experience of the implementing agency (but also of the Bank), amplified by the restricted field of potential contractors. c) Uneven response to evidence. Although delayed, the first IBBS provided categorical evidence on the Afghanistan epidemic – clearly demonstrating for the first time the absence of HIV infection in several groups thought to be at increased risk – notably road transport workers and female sex workers. It also highlighted extreme regional variation in the epidemic, with significantly higher HIV rates in western Afghanistan. However, these findings were unevenly translated into an operational response. Although services for road transport workers were de-emphasized after the MTR, those for sex workers were continued. (At the end of the project term, surveillance had identified just three HIV positive sex workers in the entire country). On the other hand, the MTR rightly proposed an increase in services to those regions most affected by the epidemic – but this was slowly and partially implemented in the remainder of the project. d) Balance between evidence and economy. Throughout the life of the project, OST (and other harm reduction services) were provided by an iNGO at a single clinic in Kabul. The quality of services was excellent by international standards – as several independent technical reviews attested. However, as a model for expansion, the site was completely unsustainable. The large number of medical, pharmacy and other professional staff, together with the costs of the services themselves resulted in a large annual operating budget – despite the fact that a maximum of 70 patients was being serviced at any one time. That alternative, lower-cost models were not explored over the life of the project illustrates two issues. Firstly, as noted in b) above, contracting these services out to an international NGO with its own set of stakeholders and standards may have reduced the incentives to explore adaptions in service provision specific to the Afghanistan context. Secondly, the OST pilot operated as a demonstration project with a dominant purpose to convince policy makers that international evidence of the success of this approach could be relevant in the country. Innovation in this political context was very difficult – as the mandate was to 9 follow global best practice examples, not to explore potential efficiencies which may have entailed risks. e) Human resources. Implementation was challenged throughout the project life by difficultly in recruiting national consultants which, due to the very limited government capacity, filled the majority of posts in the NACP. The same difficulties pertained to the international HIV/AIDS consultant position – in which retention also proved to be a challenge. These delays were at least in part due to the security situation, which made the position of international technical advisor unattractive. Several candidates to this position withdrew their application after recruitment had commenced. At least one incumbent in the position left earlier than contracted. These difficulties are noted in ISRs and other project documents and were neither specific to this project nor within the control of the Bank to effectively alter. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization The major points concerning the design and use of surveillance have been made above. These are expanded on below: a) Focusing on generation of epidemiological evidence was a great strength. The institution of analytic work to capture data from marginalized populations (IDU, MSM and sex workers) was an enormously progressive step, especially in the political and social context of Afghanistan. As has been seen elsewhere in the region, the Bank’s focus on program efficiency led logically to acknowledgement of the importance of populations at risk in the epidemic, and supported generation and rigorous analysis of evidence of the burden of disease and the efficacy of interventions targeted to these groups. This project is one piece of a broader effort that has fundamentally changed the way international assistance for AIDS is delivered. The first round of the IBBS survey (2009) proved the importance of this investment. The categorical data on HIV in key populations allowed the project to focus its efforts on those most affected by the epidemic – IDU – and to strengthen provision of services in the geographic areas where prevalence was highest. Interventions and monitoring in truckers were dropped when the HIV prevalence was found to be zero. Although a rate of 0% was also found in sex workers, interventions continued in this group – a reasonable decision given the high risk behavior that had been documented. b) Paradoxically, the ambitious aim of the surveillance work led to the choice of unwieldy, expensive methods which produced infrequent and delayed data. For example the choice of respondent driven sampling (RDS), the method used for the IBBS among IDU and sex workers, was recognized to be “cutting edge technology, subject to new statistical developments and evolving implementation issues� in 20082. Project progress was closely tied to implementation of surveillance from the beginning – indeed initiation of IBBS was a legal covenant of the financing agreement. This choice of surveillance approach had an effect on the team’s ability to monitor project impact – with ISRs relying 2 Implementation Support Mission (March 15-20, 2008). Aide–mémoire 10 on the same indicator data points for some indicators between 2010 and 2012. The use of a more rapid, frequent method of monitoring (for example expanded annual sentinel surveillance, convenience sampling or case reporting) may have made supervision more effective. c) In retrospect, the Afghanistan context in fact makes it unlikely that there would be significant spread of HIV from IDU to the general population. Although sexual risk behavior and sexually transmitted diseases are present in other groups, HIV remains almost completely confined to IDU. Together with estimation of a maximum of 20,000 IDU in the country, this made it essentially impossible that the 0.5% general population prevalence chosen for the first PDO indicator would be exceeded. At the MTR, and at the project restructuring (June 2010), it was recognized that the general population indicator (knowledge in 15-24 year olds) was of “little added value in understanding the dynamics of the HIV epidemic in…Afghanistan� and it was dropped. The 0.5% general population prevalence was retained – probably because of its status as a goal in the Interim National Development Strategy. d) At the same time, the RF was enhanced by the addition of indicators around the coverage and quality of service delivery, i.e., the knowledge and attitude of health workers to HIV and enrollment under the pilot OST program. The latter of these proved critically important, as it documented in real time the difficulties in implementation faced by the OST clinic from political opposition to the importation and use of methadone. It may also have been useful to have included an additional indicator of the OST site that reflected the difficulty and great value of having established and maintained such a facility in a challenging political and security environment. This ICR review is constrained to quantify the success of the OST pilot in terms of patients enrolled only – however the quality of the achievement itself was excellent, as is noted in several places in this review. e) The baseline values for the health worker indicators were assessed in a health facility assessment conducted in December 2011. The second health facility assessment was scheduled to be available by March 2013. The draft of this report has been shared with the MOPH by the contractor; however the HIV related indicators were not included as part of the main report. These will be analyzed separately, and should be completed before the end of June 2013; however at the time of writing this ICR report, the data are not available for review. 2.4 Safeguard and Fiduciary Compliance Financial Management: A significant number of financial management risks were assessed as high at appraisal and the financial management (FM) arrangements were continuously monitored and reviewed during the project. However the MTR noted only a number of relatively minor fiduciary issues – including delays in undertaking audits of NGOs and internal project activities. At this time, ongoing issues with payments to NGO contractors were noted - the project disbursement stalled for the periods of July to November 2008 and September to December 2009. This was subsequently rectified and financial management was assessed as moderately satisfactory by following implementation support missions. After MTR, a new Intermediate Result Indicator was proposed – “FM Reports include physical project report�. Compliance with this by the quarterly reports is noted in subsequent ISRs, and this indicator is rated as “achieved� at ICR. 11 Procurement is rated consistently as moderately satisfactory. A consistent theme in earlier supervision missions (2008 and 2009) is the slow process of contracting firms and individual consultants, due to the low capacity of the NACP procurement unit. However no serious issues of non-compliance were noted. Safeguards: Overall safeguard compliance is consistently noted to be satisfactory in project ISRs. Early in the project (2008), some issues were noted around medical waste management, as syringes were disposed in areas accessible by drug users and others. This was addressed by increased supervision and training of NGOs by NACP. 2.5 Post-completion Operation/Next Phase The project paper for the recently approved Bank-financed US$100 million health project in Afghanistan - System Enhancement for Health Action in Transition (SEHAT) sets out the health priorities for the country - especially Maternal and Child Health, and expanding secure provision of a basic package of health services. It explains how Afghanistan’s heavy reliance on finance from multiple donors, the country’s upcoming political tr ansition and continuing weak government capacity all support a model of delivering sectoral support for health. The Bank intends to shift its support from a series of emergency responses - of which AHAPP was an example - to a flexible and coordinated systems-focused approach. SEHAT will form a platform for harmonizing support to the health sector from IDA, the EU and other donors contributing to the Afghanistan Reconstruction Trust Fund (ARTF).The strengthening effects of the Bank’s previous health interventions in Afghanistan, including AHAPP, are also mentioned. Within SEHAT, several elements built under AHAPP are proposed to be continued, using the successful modality of contracting NGOs for service delivery - notably, the provision of services for marginalized populations such as prisoners, and HIV/AIDS prevention services for targeted population sub-groups who are at an elevated risk for HIV-infection. It is planned to close the single site delivering OST. SEHAT allocates US$10 million for the provision of HIV/AIDS targeted interventions and the project results and monitoring framework includes an indicator for HIV in IDU: “Percentage of IDU reached by Needle Syringe Program, to be monitored by IBBS, every 2 years�. Details are not given on how this indicator will be monitored. Although no specific mention is made of further rounds of IBBS in the project’s support for Integrated Disease Surveillance, it is understood that the MOPH plans to continue this within the project. There is no indication of plans to continue support for the advocacy and communications elements of AHAPP. Particularly, the US$90 million component of SEHAT which will support stewardship and systems building does not mention specifically work in HIV/AIDS – despite having a sub-component to strengthen MOPH’s health promotion unit work in MCH, malnutrition, TB, hygiene promotion and other areas. However the structure of SEHAT is flexible and the country team would have the capacity to finance such activities should the utility of these be agreed with the client. Dropping the less useful general population IEC activities and incorporating those for vulnerable populations in contracts specific to those groups will be an efficient way to sustain the most effective achievements 12 of AHAPP. The degree to which these other activities previously supported under AHAPP will be sustained therefore depends on the availability of finance from other donors. From interviews with GFATM and USA-funded program representatives, it is understood that their support is likely to be limited to AIDS treatment services and harm reduction. SEHAT retains the capacity to finance additional interventions, should the GoIRA request this – although the small scale of the epidemic in Afghanistan makes this improbable. There are no plans to continue support to the pilot site for OST. The iNGO which had implemented this facility has ceased its operations. Plans to transition the facility to a local NGO have been problematic, with no clear national counterpart available to assume delivery of services. This, coupled with continued opposition from the GoIRA (Ministry of Counter-narcotics) to the expansion of service beyond 71 enrolled clients, has led to the decision to close the facility, in a controlled fashion that protects the current clients from any harm. In summary, SEHAT will continue to support those elements of the AHAPP which were demonstrably most likely to efficiently deliver HIV outcomes – harm reduction services for IDU and regular surveillance of HIV in key populations. Additionally, support will continue for the pioneering and unique work providing health services for prison inmates. The demise of the only site delivering methadone therapy in Afghanistan is regrettable, but unavoidable given the intractable political barriers to its expansion, and the high unit cost of services. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Objectives: At appraisal, the PDO of this project was directly responsive to the global and country priorities – as articulated in MDG6: to “have halted by 2015 and begun to reverse the spread of HIV/AIDS� and the Afghanistan National Development Strategy 2006 – 2013: “To have halted and begun the reverse the spread of HIV/AIDS by 2020�. The project was also relevant to the Bank’s regional strategy of assistance in HIV/AIDS – projects of a similar overall objective were supported in countries throughout South Asia. As noted in the Afghanistan Country Program Evaluation undertaken by the Bank’s Independent Evaluation Group (IEG), Bank assistance during FY02–11 was guided by two Transitional Support Strategies and two Interim Strategy Notes. At the time of AHAPP appraisal, the Bank Group’s strategy for Afghanistan was organized around three “pillars�:  Building the capacity of the state and its accountability to its citizens  Promoting growth of the rural economy and improving rural livelihoods  Supporting growth of the formal private sector, including through infrastructure development. The IEG report notes the investments to address HIV/AIDS (contracting NGOs to provide harm-reduction services) as being appropriate components of this capacity building approach (Pillar One). A major review of the health program (Belay 2010) places the sector in its macroeconomic context. Afghanistan’s first National HIV and AIDS Policy (2012) clearly articulates the importance of focusing on populations at highest risk – closely consistent with the objectives of AHAPP. 13 Design: The aim of providing world class HIV prevention services to marginalized, vulnerable populations in Afghanistan was excellent from several perspectives - being human rights affirming, efficient and pro-poor. At the time AHAPP was approved, the greatly increased efficiency of focusing investments on populations at risk in concentrated HIV epidemics was becoming apparent. Evidence of the effectiveness of targeted interventions in populations at risk – for example opioid substitution therapy for IDU - was also maturing. The project focus, therefore, was cutting edge and highly appropriate – as experience in many countries and populations has subsequently borne out. The use of NGO contracting was based on sound experience in the country, but was nevertheless an innovative solution to the problems of low capacity and fragile governance. The advantages of this approach have been examined in detail (Belay 2010), and the new health project SEHAT will further build on this experience. In the difficult context of Afghanistan since 2007, these project innovations are also directly relevant to the Bank’s policy in more effective engagement in fragile and conflict-affected situations (see: The World Development Report 2011: Conflict, Security and Development). The project design was also very responsive to the limited amount of epidemiological evidence available to guide project allocations by risk or by geography. It placed an emphasis on generating these important data, and explicitly tied them to future reviews of implementation. This had the unintended effect of building delay into project implementation – as decisions on program allocations awaited the outcome of elaborate sampling and analysis methods, also subject to lengthy international ethical review because of their experimental focus. The project’s monitoring component may have been better designed to emphasize the generation of rapid, serial, program-relevant data – and keeping the research epidemiology (although of great value) less directly tied to implementation progress. Also, the effect of advocacy on entrenched opposition to methadone therapy may have been over-estimated. 3.2 Achievement of Project Development Objectives The project’s development objectives were to slow down the spread of HIV and build up the national capacity to respond to the epidemic. This would be accomplished by: (a) behavior change among vulnerable groups at high risk; and (b) improving knowledge of HIV prevention and reducing stigma related to HIV and AIDS in the general population. The PDO indicator targets at project end were assessed as follows: Indicator 1. HIV prevalence in the general population at less than 0.5%. The rate of HIV in the general population in Afghanistan is currently estimated to be less than 0.1% (UNAIDS 2011 estimate). Therefore this target was surpassed. Indicator 2. HIV prevalence amongst high risk groups (IDU and SW) less than 5%: a) HIV rates in IDU in Kabul, Herat and Mazar-i-Sharif in IBBS 2012 showed no statistically significant change from IBBS 2009. The RDS-adjusted HIV prevalence rates from highest to lowest were as follows: Herat (13.3%), Kabul (2.4%), Jalalabad 1% and 14 Mazar 0.3%. The overall HIV prevalence among IDU in all 5 cities was 4.4% (95% CI, 3.3-5.8) in IBBS 2012. b) HIV rates in SW in IBBS 2012 were 0.0% in Kabul and 0.9% in Herat, with no statistically significant change from IBBS 2009. Therefore, the target of maintaining prevalence to less than 5% was achieved under this indicator. Indicator 3. Percentage of health workers who correctly identify ways of preventing HIV transmission: This indicator was adopted after the mid-term review, and formalized in the Results Framework annexed to the restructuring paper June 28, 2010, with the target value set as 75%. The baseline data were collected at a health facility assessment in December 2011 and measured at 19.3%. At the time of writing this report, the 2013 data were not available to the ICR. Indicator 4. Among health workers who have heard of HIV, the percentage who are willing to care for patients with HIV: This indicator was adopted after the mid-term review, and formalized in the Results Framework annexed to the restructuring paper June 28, 2010, with the target value set as 80%. The baseline data were collected at a health facility assessment in December 2011 and measured at 24.4%. At the time of writing this report, the 2013 data were not available to the ICR. Indicator 5. Percentage of SW reporting use of condom with their most recent client: A target of 50% was provisionally set at project approval, recognizing that this would require adjustment when the results of the IBBS 2009 survey became available. The IBBS 2009 found a rate of 58%, and the target for the indicator was set at 60% after MTR and project extension. The IBBS 2012 found a rate of 52.3%, therefore this target was not achieved, as the rate of reporting condom use was not found to have increased between 2009 and 2012 and was less than the end target. (In IBBS 2012 the indicator is phrased “Condom use at last sex�). Indicator 6. Percentage of IDU reporting use of sterile injection equipment at last time of injection. As for Indicator 5, it was intended to adjust targets when the results of the first IBBS survey became available. The validity of the IBBS 2009 data on this indicator (a rate of 95%) was questioned at MTR. Data at project approval had indicated a rate of 45% might be likely (University of Manitoba, 2007). Therefore an end target of >80% was chosen. The IBBS 2012 did not assess this indicator as written in the broader IDU population. Assessment of the indicator was undertaken among the smaller, less representative sample of IDU in prisons, where the level was found to be 41%. The IBBS indicator “Used non-sterile injecting equipment at least once in the past 3 months� is used as a proxy for national data for this indicator, yielding a crude average of 92.9%, well above the credible baseline figure and exceeding the project target. Indicator 7. Percentage of IDU reporting use of condom at last time sex: As for Indicators 5 and 6, it was intended to adjust targets when the results of the first IBBS survey became available. A target of 20% increase was provisionally set at project approval. The IBBS 2009 found a rate of 27%, and the target for the indicator was set at 15 >40% after MTR and project extension. The IBBS 2012 found a rate of 23.9% (as a crude average of reported data). Therefore this target was not achieved as the 2012 IBBS finding was lower than the target and the baseline values. The Intermediate Indicator targets at project end were assessed as follows: Indicator 1. IBBS maintained and expanded to southern and eastern sites of the country: A baseline of four sites was expanded to six after the MTR agreed to drop Component 4, subcomponent 2 (the Innovation Initiative) and to divert these resources to expanded surveillance. Six sites were successfully included in IBBS 2012, a signal achievement in providing the rigorous national surveillance data across a broad geographic scope, and including MSM for the first time in Afghanistan. Therefore this target was achieved. Indicator 2. Number of Drug Users enrolled in OST: This new indicator was adopted at MTR, after the OST pilot site had been successfully initiated. A target of 200 enrolled patients was set. By project end only 71 patients (36% of the target) had been enrolled, due to political constraints outside of the control of the project. Therefore this target was partially achieved. Indicator 3. Number of urban centers which have done high risk group mapping: At baseline, three cities had conducted high risk group mapping and it was aimed to increase this to more than 33% of urban centers. At MTR, this target was adjusted to four cities – and a total of six has been completed by the end of the project. Therefore this target was exceeded. Indicator 4. Percentage of most at risk populations (e.g., IDU, FSWs, prisoners) reached by HIV prevention program. A target of IDU: 60%, FSW: 50%, Truckers: 25% and Prisoners: 80% was set at project approval. The IBBS 2009 found a rate of 0% of HIV amongst truckers – so this group was dropped from the indicator. The other targets were adjusted based on the findings of the IBBS 2009 after MTR and project extension. The target for IDU was increased to 70%, that for SW was decreased to 10% and the target for prisoners was not changed. The indicator is calculated based on the question in the IBBS survey “Use of risk reduction service among those corresponding who have heard about the services�. The IBBS 2012 found a rate of IDU: 90.2% and FSW: 8.3%, but data from prisoners were not available. Therefore this target was exceeded in the most important population (IDU), partially achieved in SW and unknown in prisoners. Indicator 5. Percentage of most at risk populations (e.g. IDU, FSWs, prisoners) who correctly identify two ways of preventing HIV transmission: A target of IDU: 50%, FSW: 50%, Truckers: 50% and Prisoners: 75% was set at project approval. The IBBS 2009 found a rate of 0% of HIV amongst truckers – so this group was dropped from the indicator. The other targets were adjusted based on the findings of the IBBS 2009 after MTR and project extension. The target for IDU was increased to 75%, that for SW was decreased to 20% and the target for prisoners was decreased to 30%. The IBBS 2012 found a rate of IDU: 22.1% (down from 71.4% in 2009), FSW: 15.3% (up from 12.8%) and Prisoners: 6.6% 16 (down from 25.4%). Therefore this target was not achieved in the most important population (IDU), partially achieved in SW and not achieved in prisoners. Indicator 6. Annual NACP reports and action plans informed by surveillance and routine monitoring data. By the end of the project implementation, annual NACP reports and action plans were informed by surveillance and routine monitoring data. Therefore this target was fully achieved. Indicator 7. TI sites meeting service quality standards: The initial goal was for each of the service sites for SW, truckers and prisoners to meet service quality standards, and five of the six (subsequently six of the seven) for IDU. This was revised after MTR to >75% of sites meeting standards, with discontinuation of the TI site for truckers. The project achieved quality standards in 100% of the sites for which data was available. Therefore this target was exceeded. Indicator 8. Financial Management reports include physical progress report. By the end of the project implementation, all financial management reports included a physical progress report. Therefore this target was fully achieved. The following PDO Indicator was dropped after MTR: Indicator 6 (in the original Project Results Framework): Percentage of truckers reporting use of a condom last time with a sex worker: This indicator was dropped as the HIV prevalence in truckers was found to be 0% in the IBBS 2009. The following Intermediate Indicators were dropped after MTR: Indicator 8 (in the original Project Results Framework): Score on subset of National Composite Policy Index. Indicator 9 (in the original Project Results Framework): Percentage of those aged 15-24 who correctly identify ways of preventing HIV transmission: The HIV prevalence in the general population was modeled as extremely low <0.1% following IBBS 2009. Indicator 15 (in the original Project Results Framework): Percentage of contracted/sanctioned implementation units with uninterrupted fund flow and service delivery: This indicator was designed to monitor component 4, which was restructured. Linkage between outputs and outcomes: Coverage: The project was able to exceed coverage targets for access to prevention services in the most important population, IDU. The IBBS 2012 survey found an average 90.2% coverage, exceeding the project target of 70%. By the closure of the project, 3,479 non-imprisoned IDU had been reached. Access to services was high in those four areas where TIs were supported by the project (86-99%) but very low (<1%) in Charikar - where no AHAPP intervention was active. This indicates that one project output (TIs for preventive services) was strongly linked to one outcome indicator, i.e., IDU access to 17 services. Although international evidence leaves little doubt about the efficacy of the preventive interventions deployed, there is mixed evidence for impact on the ultimate outcomes of HIV knowledge, risk behavior and HIV prevalence. The PDO indicators for condom use in sex workers, and condom use by IDU were not achieved based on the data available. The comparison of the key behavior variables among IDU in Kabul, Herat and Mazar-i-Sharif from 2009 to 2012 showed a significant decrease in the proportion of IDU who had adequate knowledge about HIV prevention in Kabul (p<0.001) and a significant increase in Mazar-i-Sharif. The critical PDO Indicator 6 of “Percentage of IDU reporting use of sterile equipment at time of last injection� was not assessed in the IDU population at large. The figure available for the indicator, of 41%, is drawn from the prison population – which is heavily biased and does not have access to the range of harm reduction services available to broader population. However, two findings in the IBBS 2012 survey indicate that shared needle use was likely quite low in IDU. The results to the questions “Ever shared non-sterile injecting equipment� and “[Did not use] non-sterile injecting equipment at least once in the past 3 months�, yielded answers of 0.4-33.2% and 0-24.5%, respectively across four cities. Due to the consistency of these indicators, it is reasonable to adopt the latter as a proxy for national data for PDO Indicator 6, and a crude average yields a rate of 92.9%. This figure is a substantial increase on the most credible baseline data (45% in 2007) and significantly exceeds the project target. High coverage with services at these sites is consistent with an HIV prevalence rate remaining below 5%. National coverage of the IDU population with preventive services under this and other projects (mainly supported by GFATM) was not at a level that was likely to have significantly inflected the national trajectory of the HIV epidemic. For example, the MTR mission identified that of 19,000 estimated IDU nationally 3,250 (17%) had been reached through existing projects at end 2010. The mission noted that WHO, UNODC UNAIDS Technical Guidance defined high coverage for Needle Syringe Programs as over 60% IDU reached. The maximum enrolment at the OST clinic was 70 patients, in an estimated population of 12,541 IDU in Kabul city. Several factors contributed to this. Firstly, the population size of IDU in Afghanistan remains highly uncertain – this figure is dependent on accessing sensitive data concerning drug use in general, and difficult estimates of the much smaller drug-injecting proportion. As the project progressed, it became clear that the population size of IDU was much larger than had been initially thought. Secondly, the project was geographically constrained by security reasons to a limited number of sites – which prevented easy scale-up in those areas with highest HIV prevalence. Thirdly, despite the best efforts of the project staff, regulatory and policy barriers proved intractable to expansion of services at the OST site. Surveillance data and epidemiological modeling generated by the project helped to clarify the nature of the Afghanistan epidemic as highly concentrated in IDU. The first round of IBBS disclosed zero HIV in either transport workers or sex workers. As noted elsewhere, the general population figure of HIV prevalence was unlikely to have approached the PDO target 0.5% unless there had been a catastrophic increase in IDU, with significant spread to other risk groups and bridge populations. 18 Without a counterfactual comparator, it is impossible to judge categorically whether interventions targeting these groups materially contributed to HIV control in Afghanistan. However, an examination of other populations and regional countries is instructive. Figure A, Annex 7 (UNAIDS, 2004) demonstrates the explosive increase in HIV prevalence that had been seen in many IDU populations in Asia by the time this project was designed. There was good reason to anticipate a similar increase amongst the IDU of Afghanistan. The data from neighboring Pakistan (Figure B Annex 7) show steady increases in populations at risk over the same period as AHAPP implementation (the data from IDU are on the left). By contrast with these historical and regional examples, Pakistan’s IDU epidemic has remained static – or indeed has decreased slightly, although not within statistical significance. This fact requires some explanation. One possibility is simply that the IBBS data were not reliable, but this seems unlikely given the proficiency of the team which conducted the analysis and the adequate sampling size and frame. No other large scale HIV preventive interventions were undertaken in Afghanistan over this period. It is probable that the broader efforts of the project – the communication, advocacy and surveillance over five years, as well as the targeted interventions described - had a controlling effect on a national epidemic that may have otherwise expanded. Other factors also probably contributed, particularly the unique epidemiology of drug use in Afghanistan, including the historically low rates of injecting, network structure and the influence of neighboring epidemics on a mobile population. One study (Todd 2007b) found 76% of IDU had ever lived outside Afghanistan in the past ten years – a majority in Iran. This apparently static trend should be interpreted with caution. A serious HIV epidemic continues amongst IDU in Afghanistan – particularly in Herat (Table 28). The sharp increase there in prevalence of hepatitis C (another virus transmitted by shared injecting) underscores the ongoing vulnerability in this population. Stable HIV prevalence might be also explained by high mortality among IDU - especially during the winter from exposure. Despite the continuing low prevalence of HIV levels in sex workers, the risk to this population also remains substantial, and should be carefully monitored. 3.3 Efficiency An explicit economic evaluation of the project was not undertaken, nor was there any specific analysis supported under the four project components. In assessing project efficiency, the following factors should be considered: a) As described in detail above, the design of the project focusing on interventions in key populations at risk, was at the forefront of evidence-based AIDS programing and ensured an increased return on investment by this allocation, compared to projects in which the general population was addressed. The efficiency gains from using this project design extended not just to this project, but to those subsequently supported by the Bank and others which have drawn lessons from its implementation. b) Some interventions were of high cost, with small returns. The most notable example was the OST site, which provided services to a maximum of 71 clients and which was closed at the end of the project. However, it is difficult to assess the true returns of this pilot intervention – and to fully account for the opportunities lost with its closure. This model 19 service was the only example in Afghanistan of best international practice in preventive and medical services for IDU. Its implementation drove policy and regulatory dialogue in the country and, despite its eventual closure, the clinic was a striking demonstration that provision of OST services to the very poor and marginalized population of IDU in a most challenging political environment is not only possible, but can be highly effective. c) The cost-benefit analysis undertaken during project preparation3 was predicated on an incipient increase in transmission of HIV in the general population that, in the light of subsequent epidemiological data, does not seem have been inevitable and was not realized. However, some of the key principles outlined remain relevant, and were borne out by the experience of implementation: i) Harm Reduction interventions targeted to the populations at highest risk of HIV are efficient, pro-poor and found to be highly effective in a diverse range of international experience. ii) Strengthening HIV surveillance delivered distinct benefits, primarily through increasing allocative efficiency by enhanced targeting of harm reduction programs. Several dependent processes were identified at approval as necessary for any economic benefits to arise. These included proper assignment of qualified personnel, collection of accurate data and the production of actionable, disseminated reports. From project experience, to this might be added the importance of tailoring surveillance to a design that can be implemented in a timely manner, with a focus on producing program-relevant data. iii) The chief economic benefit of the OST pilot, as with any pilot, arose from the externalities of learning-by-doing. 3.4 Justification of Overall Outcome Rating Rating: Moderately Satisfactory The relevance of the project to the country, the region and to the World Bank’s engagement with Afghanistan was very high. Of the seven project development objectives indicators, the two original HIV prevalence goals – in the general population and in key populations - were achieved (Indicators 1 and 2). The critically important indicator of preventive service effectiveness - reflecting use of clean injecting equipment in IDU - was surpassed (Indicator 6). However, two PDO indicators were only partially achieved (Sex workers and IDU use of condoms – Indicators 5 and 7), and data are not available to assess the remaining two (Indicators 3 on 4 on Health Workers attitudes to HIV). The causal link between project activities and the first PDO indicator (population prevalence) was not logically clear. The decrease in the second PDO indicator (HIV prevalence in IDU) was not within statistical significance, but there is a logical causal pathway between the implementation of effective targeted interventions, the documented increase in the percentage of IDU being served by contracted NGOs, the documented substantial decrease in risk behavior in IDU (use of non-sterile equipment) and the maintenance of an HIV prevalence of <5% in this group. Although there are no data on the counterfactual (which would have been difficult, if not impossible to experimentally establish in Afghanistan at 3 Technical Annex, July 2007 20 this time), regional epidemics in IDU indicate that an increase in HIV prevalence in those IDU covered by services was probably averted by project interventions. Of the Intermediate Outcome Indicators, one was partially achieved (Percentage of most at risk populations reached by the HIV prevention program). Five were fully achieved or exceeded, encompassing important national gains in surveillance, policy development and management capacity. However, the two indicators relating to OST service provision and impact of prevention services on knowledge were not achieved. Although this pilot was only one of several interventions, it was a high profile activity of the project, was the focus of much of the advocacy efforts undertaken and received significant international attention. The design of the project, which focused on populations at highest risk, was a leading global example of increasing the efficiency of HIV interventions. However, the high cost of some interventions for small returns (especially OST), the small impact demonstrated on HIV prevalence in sex workers and the delays in implementing the IBBS surveys decreased the project’s efficiency. The revision of the result targets at restructuring may have been somewhat ambitious, since a number of them were partially achieved. Therefore, on balance, given the High relevance, Moderate/Substantial efficacy and Moderate/Substantial efficiency, the overall outcome rating is Moderately Satisfactory. It should be noted that the “Progress towards achievement of PDO� rating was lowered from “MS� to “MU� in the last ISR of the project (Report No: ISR8536). The main reason for this was the delay in receiving the IBBS 2012 survey results, which meant that assessment of performance under several indicators was not possible by the team. The data in this ICR report reflect the final version of the IBBS 2012, which became available in April 2013. 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development Poverty: The greatest impact of this project is potentially from its example as model of HIV service delivery that accessed vulnerable and marginalized populations, and was demonstrably more efficient because of this. The efficiency extended along two dimensions: i) the allocation of resources to populations most affected by HIV, who were also amongst the poorest in the country; and ii) the choice of interventions that were evidence-based and effective. AHAPP was also able to institute these services in a highly challenging country context, characterized by low counterpart capacity, intense security challenges and social fragmentation. Gender Aspects: The World Bank’s conceptualization of gender as a development issue is one that is evolving to greater inclusion of sexual minorities. In this respect, the work done under the project to characterize the HIV vulnerability of MSM in Afghanistan was progressive, timely and met an important gap in evidence which other work in the Bank has recently explored in more detail (Beyrer, 2011a; Kerrigan, 2012). (b) Institutional Change/Strengthening 21 NGO Contracting: As outlined above, and as explored in more detail in Belay (2010), the Emergency Project Paper of SEHAT (2013) and Loevinsohn (2005) the system for use of private, non-state providers to deliver health services in a situation of low counterpart capacity built on prior experience and was largely successful under the AHAPP. The structures for NGO contracting are well established within the MOPH and are an integral part of the MOPH organogram. AHAPP experience further laid the ground for expansion of the approach in subsequent Bank engagement with Afghanistan. These projects demonstrate that contracting of NGOs by government to deliver basic health services can be successful in improving service coverage in the most difficult of circumstances. This approach is now widely used in World Bank projects in fragile states. Policy Environment: Advancing the advocacy and policy dialogue which was supported under AHAPP faced significant hurdles. The construction of an enabling policy environment for provision of services for drug users, sex workers and homosexual men is difficult to progress in any country. In Afghanistan, which faces a range of serious challenges in health and beyond, the degree to which the project was able to bring these issues to political attention and gather political support from some key stakeholders was remarkable. The concrete outcome of these efforts is seen in the policy documents developed and approved by the government (particularly the National HIV/AIDS Policy and the Afghanistan National HIV/AIDS Strategic Framework 2011-2015) and the harm reduction services for IDU, which are established and will continue under SEHAT. The provision of general medical services for prisoners – an enormously progressive and pro- poor intervention – was another result. It was unfortunate that the first successful demonstration of OST in Afghanistan was not able to be continued. NACP: The project built and supported a highly functional NACP within the MOPH, staffed with experts in the field and supported by an international specialist adviser. As with all activities in Afghanistan today, the sustainability of this institutional strengthening is uncertain – as is the case across ministries and projects. The majority of the staff were consultants hired under the project, rather than regular MOPH staff. NACP will continue to be supported in this fashion under SEHAT for a period until civil service reform is fully implemented. The MOPH is engaged with the Capacity Building for Results Facility (CBR), a separate civil service reform project supported by IDA and ARTF. The aim of CBR project is to assist performance of line ministries including the MOPH. CBR finances the recruitment of managerial and professional staff, as civil servants, for key positions and support targeted training programs. The MOPH is developing a CBR program proposal to the Ministry of Finance with its service delivery priorities and system development reforms and related staffing needs including those for the NACP. (c) Other Unintended Outcomes and Impacts (positive or negative) Global Knowledge of HIV program efficiency: Since the inception of AHAPP, the expansion of rigorous population-based surveys and better surveillance generally, led to an appreciation that all HIV epidemics did not have a southern African scenario as their natural end point. The data and analysis generated by Johns Hopkins University under this project improved that understanding. The use of this information by the project to 22 efficiently prioritize activities and monitor success was an early example of the evidence- based HIV programming which now guides global practice. Hepatitis C (HCV) and other Blood Borne Viruses: It is worth highlighting the alarmingly high prevalence of HCV among IDU that was found by the IBBS surveys in all regions of Afghanistan. The only significant changes that were detected between the 2009 and 2012 IBBS rounds were the increases in HCV sero-prevalence in Herat and Kabul – in the former of these cities the rate in IDU is now 71%. It is remarkable that the high levels in other cities are not considerably worse, as the transmission of HCV in IDU populations is much more efficient than that of HIV. It is probable that the provision of clean injecting equipment under the project has done something also to blunt HCV spread. 4. Assessment of Risk to Development Outcome Rating: Substantial Country context: Security in Afghanistan is a critical challenge, especially in the southern and eastern provinces. The large majority of international military forces are expected to leave Afghanistan by the end of 2014. This poses not only potential additional security risks, but also the possibility of reduction in donor funding to HIV and health programs. Any displacement of populations or interruption of the provision of services in this context will have a negative impact on HIV control. Epidemiology: In the 2012 IBBS report it is asserted that Afghanistan is in an "early phase" of the epidemic. This is not epidemiologically accurate. HIV has been in the country doubtless for many years - it has been slow to spread as expected in vulnerable populations, in part possibly due to the prevention efforts under this project. Imagining a perennially early epidemic takes away this important epidemiological conclusion. However, it would be unwise to assume that HIV will remain at the current level with unchanged (or reduced) levels of investment in prevention. IDU epidemics characteristically progress to high prevalence – often with an explosive trajectory which may be delayed. Without ongoing and expanded interventions for IDU, the risk that HIV prevalence will increase above 5% in this group is high. A similar risk does not necessarily attach to sex workers – due to the very low levels of HIV and the plans to carefully continue surveillance in this population. Political Support: Much of the success of the project hinged on the political support built, which provided the permissive environment for harm reduction activities for IDU and health services in prisons. The lack of support for and closure of the OST pilot site demonstrates that blanket political support for all IDU interventions cannot be taken for granted, and may change over time. Without the policy and advocacy activities that were supported under the project, it is possible that support for HIV prevention activities in key populations may wane, with a negative impact on HIV control. The likelihood of any of these risks eventuating is moderate to high. The impact on the PDO objective of slowing down the spread of HIV from any of them would be significant – as the current epidemic trend is static, and only a small increase in transmission would act to increase the prevalence level past the PDO indicator level of 5% in IDU. 23 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Satisfactory The strategic relevance of an HIV prevention project in Afghanistan was high at the time of approval. The relevance of this investment in the context of competing health priorities (especially maternal and child health, and the provision of a Basic Package of Health Services) might be questioned post hoc; but at the time of approval the focus was clearly consistent with regional, country and institutional priorities. The design of the project – to focus on key populations most at risk - was highly appropriate and efficient, even though the choice of general population prevalence as a development indicator was not consistent with the focus on key populations. The focus on generating epidemiological data was a wise priority. This provided critical information to inform national policy and guide efficient project implementation. However the complexity, time and expense of the IBBS was under-estimated during project design. This led to delays in establishing baselines for indicators and delays in project implementation and review while these important data were awaited. The choice of NGO contracting as a key mode of service delivery was appropriate and largely successful. The risk assessment was thorough and consistent with the country context. In terms of poverty, gender and social development, the design of the project was exemplary. The targeted populations were amongst the poorest and most socially marginalized in the country. The inclusion of services for sex workers and MSM as key populations at risk was extremely progressive at the time, particularly in the social context of Afghanistan. The team made a calculated risk by inclusion of the OST pilot facility, recognizing the potentially very high gains from implementation of this approach, despite understanding the political barriers and difficulties of translating international experience into locally acceptable, cost-effective services. (b) Quality of Supervision Rating: Satisfactory It is important to note that project supervision in Afghanistan was conducted in an environment with some of the most serious logistical and security challenges facing any Bank operation. Nevertheless the team managed consistent, professional and high quality supervision over the five years of the project. The absence of data points for many of the key performance indicators during the first years of the project, then the availability of only one point until the second round IBBS data were available (four months after the closure of the project) meant that supervision was inevitably focused more on process than on development impact. This was compounded by the delays faced in NGO contracting and slow recruitment and difficulties in retaining staff at NACP. Once the critical IBBS 2009 data were available, the team conducted an MTR mission which was highly responsive to the evidence, focused activities and thoroughly revised the results framework. This 24 significantly improved project efficiency. There were five TTLs during the term of project supervision, a turn over not unique considering the difficulty in retaining staff working in fragile and conflict states; however there was substantial continuity provided by in-country team members. The project ISRs and the MTR report were of extremely high quality, with a candid and solution-focused assessment of all aspects of project implementation. The team’s efforts in progressing this challenging project were consistently recognized by management. The Progress towards PDO ratings were assessed as Satisfactory from 2010 until 2012, when this was downgraded to Moderately Satisfactory (June) and then Moderately Unsatisfactory in the final ISR (November). The main reasons for this were the delayed receipt of the IBB 2012 report until after the last ISR, and the difficult progress with methadone importation for the OST pilot site. The supervision of fiduciary and safeguard aspects was appropriate in dealing with the minor issues which arose. The post-closure continuity of supported activities was carefully managed by the project team, especially the selection of those which would be supported under the new health project SEHAT. This selection was managed by the Bank in close consultation with the GoIRA and other development partners. It is unfortunate that the OST services were not able to be continued once the contract of the iNGO delivering these had ended. In the absence of clear political support for continuation or expansion of the pilot, this was an unavoidable decision. The cessation of clinical services will be managed in a controlled manner to minimize any harm to patients enrolled in the program. Improved understanding of the epidemic (much of it derived from analytic work supported by the project) revealed that the epidemic was confined almost exclusively to the IDU population. Reconsideration of the PDO indicator on general population HIV level at the time of MTR might have better reflected the focus of the project. However, as previously noted, this indicator was also selected for its mandate as part of Afghanistan’s national development strategy. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Satisfactory With hindsight, there were moderate shortcomings in preparation of the project, particularly in the selection of the PDO indicators, and in the design of the surveillance component. There were minor shortcomings in the proactive identification of opportunities and resolution of issues during project supervision. 5.2 Borrower Performance (a) Government Performance Rating: Moderately Satisfactory The GoIRA had a strong commitment to the objective of controlling HIV in populations at risk - evinced from project preparation through to closure. This was reflected in the high- level political attention which HIV prevention enjoyed (including discussion in the national 25 parliament) and the policy documents developed and endorsed – resulting in the adoption of the Afghanistan HIV Code of Ethics, an important institutional protection for people living with HIV in Afghanistan. However, despite the development of an OST policy that was approved and endorsed by the Minister of Public Health, and the development of a significant and important relationship between NACP and the Ministry of Counter Narcotics, it was not possible during the project to implement the OST pilot to the full quota of patients – mainly due to difficulties in negotiating the approval and importation of methadone. This political barrier proved intractable, despite consistent efforts by the MOPH, the Bank and other development partners to present evidence of the efficiency and effectiveness of the OST approach. An important part of this advocacy effort was the OST pilot itself, which demonstrated highly successful implementation in the Afghanistan context and which was assessed very favorably by several independent reviews. The Government’s opposition to methadone importation and use was a significant barrier to exploration and expansion of what was planned to be a key intervention under the project. The GoIRA has worked closely with the Bank team, and in consultation with other development partners, to ensure adequacy of the transition arrangements for supported activities which will continue under the new health project SEHAT. (b) Implementing Agency Performance Rating: Moderately Satisfactory The implementing agency (NACP) was housed at the MOPH. Alongside the coordination and management of AHAPP, the NACP was the principal recipient for HIV activities funded under GFATM – resulting in close coordination between these two sources of finance. Several challenges were faced by the NACP during the initial stages of project implementation. The NACP had difficulties in establishing and maintaining input from the target groups themselves. IDU and sex workers were particularly challenging for the NACP to work with, due to stigma against these groups and the illegality of risk behaviors. It was difficult for NACP to find and engage community leaders from these groups who were able to participate in meetings and events. Low capacity in the MOPH, poor security and timing of the presidential elections slowed initial project implementation. As noted in supervision missions, six months after effectiveness of the project, the key issues experienced were a) Relocation of the NACP to a remote and ill equipped premises, b) High number of vacancies in key staff positions – particularly technical specialist, c) Procurement – especially delays and technical issues with tender processes and contract negotiations, d) Delayed payments to contracted NGOs and e) Lack of some essential Financial Management capacities. However, the majority of these issues were resolved in a timely manner, including fiduciary issues, and subsequent to the MTR it is noted that the project’s efforts in institutional capacity building had resulted in improvements in the standard and timing of each of these functions. 26 The delays experienced with implementing surveillance were most significant both because they were prolonged and because so much of the project monitoring and revised programming depended upon the data to be generated. Although this is due in part to the ambitious scope of the surveys undertaken, it was probably exacerbated by conflicting expectations between the contractor and the implementing agency. As had been noted during project preparation, the surveillance activities supported were envisaged to be owned by and readily available to the NACP, routine and not subject to research review processes. The contractor clearly felt that their priority in the exercise was the highest possible quality epidemiological research, safeguarded by thorough independent review. The capacity of NACP to counterbalance the ambitions of the “global experts� and to manage this international institution was understandably limited. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately Satisfactory Government performance was rated Moderately Satisfactory, Implementing Agency performance was rated Moderately Satisfactory. The overall rating is Moderately Satisfactory. 6. Lessons Learned a) Preparation and effective implementation of an HIV/AIDS project in Afghanistan demonstrated that innovative programming is possible even in situations of extreme fragility and low counterpart capacity. Neither these circumstances, nor a challenging social context, should pose a barrier to the implementation of projects delivering interventions to marginalized populations at high risk from HIV. b) The use of NGO contracting for service delivery, although challenged by delays in defining and implementing a selection process in a limited field of potential candidates, built on solid experience in this innovative approach and helped form the model for future projects. It provided a flexible means to supplement and build capacity in the implementing partner, and to accelerate the provision of services. The lessons learned in contracting NGOs from this and previous projects are exemplified in the design of SEHAT, as described in the Emergency Project Paper (February, 2013). In brief, this includes:  Strengthening of the capacity of implementing agency to procure, monitor and manage the contracts of NGOs delivering services  Ensuring adequate coverage of Health Management Information Systems, supplemented by verification and by independent third party assessments, to assess NGO performance  The institution of quarterly implementation progress workshops with the participation of all implementers to review progress, identify implementation bottlenecks and prepare action plans. c) The efficiency of HIV prevention programs is greatly enhanced by the generation of epidemiological data to guide allocations to those populations and regions at greatest risk. However, surveillance components supported by the project should have a primary purpose to guide government policy and project focus. Implementation progress should not be made contingent on the results of research-focused activities. 27 d) The effectiveness of HIV prevention programs depends on technical strength of evidence-based interventions, adequate coverage of target populations with services and a permissive legal and policy framework. Each of these should be rigorously assessed during preparation as each critically and independently influences program impact. e) The World Bank has been a global leader in providing services to vulnerable and marginalized populations in the HIV epidemic. The ethical propriety of this project – which provided services for some of the most poor and disenfranchised people in Afghanistan, should be highlighted and commended. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies See Annex 5 for the Borrower ICR and official Government comments on the ICR. (b) Co-financiers There were no co-financiers of this project. (c) Other partners and stakeholders None. 28 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Latest Estimate Appraisal including Percentage of Components Estimate (USD contingencies (USD Appraisal millions) millions) Communication and 1.06 1.06 84% Advocacy Strengthening HIV 1.61 1.53 85% Surveillance Targeted Interventions 4.06 5.73 117% Program management, capacity building and 1.80 1.60 81% monitoring and evaluation Total Baseline Cost 8.53 Physical contingencies 1.00 Price contingencies 0.47 Total Project Costs 10.00 9.92 99.2% (b) Financing Appraisal Actual/Latest Percentage of Source of Funds Estimate Estimate Appraisal (USD millions) (USD millions) Borrower 0.00 0.00 0.00 IDA Grant 10.00 9.92 99.2% 29 Annex 2. Outputs by Component Component 1: Communications and Advocacy (US$1.06 million – total actual costs) This component was intended to strengthen the advocacy and communication capacity of the NACP to create a policy environment that enabled scaling up of targeted interventions (TIs) for vulnerable populations at high risk, and reduced stigma and discrimination related to high risk behaviors. Specifically, informed advocacy among policymakers and opinion leaders was aimed to reduce stigma and harassment of vulnerable populations and have leaders: (a) appreciate the threat of HIV to Afghanistan; (b) better understand the actions that could prevent a full-blown HIV epidemic; (c) become effective sources of information for the rest of the community; (d) take actions themselves to assist their communities to avoid HIV; and (e) provide continuous support for the AIDS control program. It was agreed that a national policy statement would be in place at the start of the project and a substitution treatment policy for IDU would be developed. The component included the following activities: 1) Communications activities were intended to increase access to information and appropriate knowledge of HIV prevention and other related health issues for vulnerable groups at high risk, including sex workers and their clients, truckers, and youth in areas covered by the project. 2) Behavior change interventions were intended to include a focus on the use of clean needles and syringes, and on condom use among IDU and their partners; on consistent condom use among sex workers and their clients; and, on appropriate knowledge about HIV transmission and ways of preventing it among all vulnerable populations at risk. 3) Advocacy activities were intended to include (a) target audience research; (b) development of an advocacy strategy; (c) development of core training and information packages for use by service providers; (d) specific communication training for NGO service providers and others; (e) gender- and culturally- appropriate communication media targeting vulnerable persons at risk; (f) audience surveys; and (g) reporting. To ensure a coherent approach and complementarity of activities, one organization was to be contracted to develop the plan and implement all elements of the communication and advocacy component under the coordination and guidance of the NACP. Component 2: Strengthening of HIV Surveillance (US$1.53 million – total actual costs) This component was intended to build the evidence base for HIV and AIDS planning and help monitor the effectiveness of the program. It planned to build the NACP’s capacity to track the magnitude and trajectory of the HIV epidemic in Afghanistan through the development of second generation surveillance (SGS) system. (This included separate mapping and size estimation of at risk populations, collection of biological data from representative samples of the populations, and tracking the behaviors that contribute the most to HIV transmission dynamics). The data collected through surveillance was 30 intended to provide the basis for future resource allocation and program management decisions. All the SGS data were, therefore, to be “owned by, and readily available to, the NACP, and generated on an intermittent or routine basis, and not subject to the same internal review processes put in place for research and special studies� 4 . Epidemiologically relevant data in Afghanistan were scarce at the time of appraisal; however some recent studies of at risk populations were chosen to refine the selection of survey groups and locations for the baseline round of the surveillance activities. The surveillance activities were to be implemented by an organization selected through international competitive bidding. This organization would work with the NACP, guided by an HIV Surveillance Working Group, to train its staff to conduct surveillance, obtain all necessary test kits and other materials, provide timely reports, and disseminate findings. The terms of reference (TOR) for surveillance included a) size estimation of at- risk populations in urban centers; (b) integrated biological and behavioral surveys (IBBS) of IDU, female sex worker (FSW), prisoners, and truckers conducted twice during the Program Implementation Plan period; (c) two rounds of knowledge, attitudes and belief (KAB) surveys of mainstream opinion leaders to assess the effectiveness of the advocacy and communication component and (d) annual data synthesis and epidemic modeling of available surveillance and other data to project the impact of the epidemic on the overall country scenario. Due to the importance of obtaining baseline measures for the core program indicators in order to measure project results, the data collection activity for the IBBS of IDU, SWs, and prisoners was to be completed within the first 6-9 months of the project. Component 3: Targeted Interventions for High Risk Behaviors (US$5.73 million – total actual costs) This component was meant to support TIs to prevent further spread of HIV among vulnerable groups at highest risk, promoting safe practices and reducing risky ones. The targeted intervention for HIV prevention among IDU was intended to provide a comprehensive harm reduction package, including needle and syringe exchange, peer counseling and education, and condom promotion, delivered at drop in centers and through other outreach services. The services were to be delivered where HIV transmission was most likely to occur, based on mapping of high risk populations and sero prevalence survey data, with flexibility to expand to new sites as they were identified. Proposals were to be invited from NGOs for work with identified groups, such as IDU and their partners, sex workers and their clients, and others as appropriate in selected sites starting with the major urban areas. In Kabul, due to the estimated size and spread of mainly street-based IDU, there were to be three contracts covering different parts of the city and surrounding areas, while in the other three cities with surrounding areas (Herat, Jalalabad and Mazar) there would be one contract for each area. NGOs were to provide the services under an MOPH contract. 4 Technical Annex July 5, 2007 31 The TOR for targeted interventions for IDU, sex workers and their clients, truckers and prisoners included (a) gathering of baseline data to estimate the size of the populations to be covered; (b) establishment and operation of appropriate services such as, in the case of HIV prevention among IDU, drop in centers (DIC), social support, needle and syringe exchange, OST (in one package in Kabul and among prisoners), condom distribution, sexually transmitted infections (STI) care, other medical support, and Voluntary Counseling and Testing (VCT); and (c) reporting. The contracts for IDU were to target groups in four cities and their surrounding areas with flexibility to expand to new sites each following year. An ethnographic study was to be conducted to learn more about MSM networks and how to design appropriate strategies for reaching men who have sex with men (MSM) with HIV prevention interventions. Component 4: Program Management, Capacity Building, Monitoring and Innovation Activities (US$1.60 million – total actual costs) This component was intended to strengthen the core functions of the national HIV/AIDS prevention program, including program management, capacity development, and monitoring by NACP, and the multi sector HIV oversight of the HIV/AIDS Coordinating Committee of Afghanistan (HACCA). This component was also intended to provide funding of innovative approaches – an “Innovation Initiative� for mainstreaming of the national multisector HIV response. Membership of HACCA was to be extensive and include NGOs and civil society representatives as well as ministries. Capacity building activities in this component would facilitate learning and include exposure visits, short term TA, training, including training of health services providers, and conferences leading up to policy development and review. The NACP was to be supported through six national advisors, the program manager, and an international advisor – two of the national advisors would provide support in financing and procurement. This component was also to support the development of routine monitoring and service quality assessment, closely aligned with the existing monitoring and evaluation (M&E) and health management information systems (HMIS) of the MOPH, and including an action plan, timeline, and assigned responsibilities to the M&E advisor and other NACP staff. Standardized, routine monitoring indicators and formats were to be developed for each program area. The principles and operations of routine monitoring for NACP were intended to be consistent with established practices of the larger MOPH M&E and HMIS system for Basic Package of Health Services/Emergency Project Health Services facilities. Outputs of the Component: Costs Projected Actual Costs at approval (USD): (USD): Component One – Communications and Advocacy  Target audience research $1,263,200 $1,063,200  Gender and culturally appropriate communication media targeting vulnerable persons at risk  Audience surveys 32  Development of an advocacy strategy and activities  Development of core training and information packages for target audiences for use by service providers  Specific communication training for NGO service providers, journalists, private practitioners and others Component Two – Strengthening of HIV Surveillance  Mapping and size estimation of at risk populations $1,799,700 $1,532,171 in urban areas  Implementation of two rounds of Integrated Biological and Behavioral Surveillance (IBBS) surveys among PWIDs, FSWs, prisoners, and truck drivers  Two rounds of Knowledge, Attitude, and Practice (KAP) surveys among policy makers and their influencers.  Annual data synthesis and epidemic modeling of available surveillance and other data to project the impact of the epidemic on the country scenario.  Rapid assessment of PWIDs in Western provinces of Afghanistan Component Three – Targeted Interventions for High Risk Behaviors  Regular IDU serviced: 3,479 $4,939,700 $5,725,579  Distribution of Safe-Injecting kits: 3,351,348 units  Collection of used needles and syringes: 3,170,396 units  Opioid Substitution therapy (at one site): 70 patients enrolled  VCT for HIV: 13,896 tests  Hepatitis C/B and Syphilis testing: 33,516 tests  Condoms distributed: 1,034,984 Other services provided – not quantified  Syndromic management of STIs  Abscess management  Overdose management  Referral for TB services  Referral to ARV center for HIV+ clients  Referral for drug detoxification and abstinence based treatment  Primary health check ups  Hygiene kits  IEC  Social services Component Four – Program Management, Capacity Building, Monitoring and Innovation Activities 33  Staff positions at NACP: number local? An $1,977,400 1,601,000 international HIV advisor since April 2012  NACP coordination mechanisms developed at different levels  HIV technical groups convened on regular basis: surveillance, monitoring and evaluation.  Development of new HIV testing and counseling guidelines  Development of National HIV/AIDS policy  Management of national HIV/AIDS prevention program  Oversight and support of HIV/AIDS Coordinating Committee of Afghanistan (HACCA)  Assessments conducted: o 1) Assessment of the Islamic Republic of Afghanistan HIV & AIDS Prevention Programme (2010) o 2) Assessment of OST pilot – Johns Hopkins University (2011) 34 Annex 3. Economic and Financial Analysis An explicit economic evaluation of the project was not undertaken, nor was there any specific analysis supported under the four project components. The cost-benefit analysis undertaken during project preparation 5 was predicated on an incipient increase in transmission of HIV in the general population that was not realized. See a further consideration of this in the main ICR text (3.3 Efficiency, Page 24). 5 Technical Annex, July 2007 35 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Names Title Unit Silvia M. Albert Temporary WBITI Mariam Claeson Program Coordinator SASHN Deepal Fernando Senior Procurement Specialist ECSO2 Phoebe M. Folger Operations Officer SASHN Sundararajan Srinivasa Gopalan Senior PHN Specialist SASHN Palwasha Lena Kakar Consultant SASDI Hasib Karimzada Program Assistant SASHD Laura M. Kiang Consultant SASED Benjamin P. Loevinsohn Lead Public Health Specialist AFTHW Kenneth O. Okpara Sr. Financial Management Specialist SARFM Mohammad Arif Rasuli Senior Environmental Specialist SASDI Ghulam Dastagir Sayed Senior Health Specialist SASHN Rahimullah Wardak Procurement Specialist SARPS Alexander David Wodak Consultant SASHD Supervision/ICR Muhammad Wali Ahmadzai Financial Management Analyst SARFM Silvia M. Albert Temporary WBITI Abdul Mohammad Durani Social Development Specialist SASDS Emanuale Capabianco Senior Health Specialist SASHN Deepal Fernando Senior Procurement Specialist ECSO2 Celine Ferre Consultant SASSP Mohammad Tawab Hashemi Health Specialist SASHN Hasib Karimzada Program Assistant SASHD Laura M. Kiang Consultant SASED Cornelis P. Kostermans Lead Public Health Specialist SASHN Asha Narayan Sr. Financial Management Specialist SARFM Homira G. Nassery Operations Officer HDNHE Mohammad Yasin Noori Social Development Specialist SASDS Robert Oelrichs Senior Health Specialist HDNHE Kenneth O. Okpara Sr. Financial Management Specialist SARFM Asif Qurishi Team Assistant SASHD Mohammad Arif Rasuli Senior Environmental Specialist SASDI Ghulam Dastagir Sayed Senior Health Specialist SASHN Rahimullah Wardak Procurement Specialist SARPS David Wilson Program Director HDNVP Debrework Zewdie Director HDNHE 36 (b) Staff Time and Cost (PE-P101502-LEN-BB and PE-P101502-SPN-BB) Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY07 58.71 114,892.77 FY08 3.39 343.58 Total: 62.10 115,236.35 Supervision/ICR FY07 0 0 FY08 40.08 56,579.44 FY09 35.12 39,580.19 FY10 29.50 64,617.74 FY11 30.53 25,757.66 FY12 30.72 11,604.74 FY13 13.53 10,379.15 Total: 179.48 208,518.92 37 Annex 5. Summary of Borrower's ICR and/or Comments on Draft ICR Islamic Republic of Afghanistan Ministry of Public Health DG of Preventive Medicine Communicable Disease Control Directorate National AIDS Control Program Implementation Completion and Result Report Afghanistan HIV/AIDS Prevention Project Prepared by NACP March, 2013 38 List of acronyms AHAPP Afghanistan HIV/AIDS Prevention Project AIDS Acquired Immune Deficiency Syndrome FSWs Female Sex Workers GCMU Grants and Contract Management Unit HACCA HIV/AIDS Coordinating Committee of Afghanistan HIV Human Immunodeficiency Virus IBBS Integrated Biological and Behavioral Surveillance ICR Implementation Completion and Result INGOs International Non-Governmental Organizations M&E Monitoring and Evaluation NGOs Non-Governmental Organizations PWIDs People Who Inject Drugs SGS Second Generation Survey TI Targeted Interventions WB World Bank 39 Introduction Afghanistan HIV/AIDS Prevention Program (AHAPP) commenced in 2007 as a World Bank (WB) grant with a total of 10 million USD to support the Government of Afghanistan’s response to HIV epidemic. It was designed to strengthen the national capacity by scaling up prevention programs targeting populations at higher risk of HIV infection including People Who Inject Drugs (PWIDs), Female Sex Workers (FSWs), truck drivers and prisoners. In addition, AHAPP also aimed at strengthening the capacity of the National AIDS Control Program (NACP) in areas like advocacy and communication, HIV surveillance and Monitoring and Evaluation (M&E). AHAPP was initially designed for 33 months starting July 2007 through March 2010 with an overall objective to slow down the spread of HIV and build up the national capacity to respond to the epidemic through (i) targeted interventions for Key Affected Populations (KAPs) and (ii) Improving knowledge of HIV prevention and reducing stigma related to HIV in the general population. The following were the project’s components to achieve the stated objective: Component 1: Communications & Advocacy Component 2: Strengthening of HIV Surveillance Component 3: Targeted Interventions for High Risk Behaviours Component 4: Programme Management, Capacity Building, Monitoring & Innovation activities This Implementation Completion Report (ICR) was prepared by NACP, in response to the requirement by the WB, as an evaluation on the project’s execution, its cost and the benefits derived from it. I. Basic information This section includes basic information about: A. The country and project name (title and ID): Project: P101502 -Afghanistan HIV/AIDS Prevention Project B. Donor: World Bank IDA C. Original total commitment (USD) and disbursed amount: Original total commitment 10,000,000 USD Disbursed USD 9,917,127.95 D. Key dates:  Grant agreement: July 2007  Original contracts: 12 contracts started from 2008 till February 2011  Revised contracts: 9 contracts from February 2011 till June 2012. II. Project objectives, description and results framework This section includes information about objectives, targets and indicators of the project: A. Objectives (general and specific) of the project: The project’s development objectives are to slow down the spread of HIV and build up the national capacity to respond to the epidemic. This will be accomplished by (i) Targeted Interventions for most at risk populations (MARPs), and (ii) Improving 40 knowledge of HIV prevention and reducing stigma related to HIV in the general population. The key performance indicators that will be used to track the project development objectives are: o The percent of injecting drug users who have adopted behaviors that reduce transmission of HIV, that is, who use clean injecting equipment at last time injecting o The percent of sex workers who report using a condom with their most recent client o The percent of young people aged 15-24, in areas covered by the project, who correctly identify ways of preventing HIV transmission. In order to achieve the overall objective of the project, AHAPP was designed to operate through four component areas described in the following section: B. Components (project areas): 1) Advocacy and communication To facilitate an effective and smooth implementation of HIV-related activities by creating an enabling environment to scale up targeted interventions to populations at higher risk of HIV as well as to reduce the level of stigma and discrimination around HIV and key affected populations. The following were the activities proposed for this component:  Target audience  Gender- and culturally  Audience surveys research appropriate communication media targeting vulnerable persons at risk  Development of an  Development of core  Specific advocacy strategy and training and information communication activities (e.g. packages for target training for NGO meetings, training, and audiences for use by service providers, dissemination of service providers journalists, private advocacy packages) practitioners and others  Reporting 2) Strengthening of HIV surveillance This component was designed with the purpose to make HIV biological and behavioral data available through the development of second generation surveillance (SGS) system in order to monitor HIV epidemic profile in Afghanistan. The terms of reference for surveillance include:  Desk review of available information about location and size estimation of risk population, 41  Mapping and size estimation of at risk populations in urban centers (up to 8 sites every second year or twice during the duration of the project),  Design/implement two rounds of Integrated Biological and Behavioral Surveillance (IBBS) surveys among PWIDs, FSWs, prisoners, and truck drivers  Conduct two rounds of Knowledge, Attitude, and Practice (KAP) surveys among policy makers and their influencers.  Annual data synthesis and epidemic modeling of available surveillance and other data to project the impact of the epidemic on the country scenario. 3) Targeted interventions (TIs) Interventions were directed to populations at higher risk of HIV infection (PWIDs, SWs, truckers) comprised of harm reduction package6. In some of the TIs, particularly for prison services, a component of operations research has been included. In light with the conventional harm reduction package, MoPH/NACP has defined a list of services to be included in TIs that takes into account the country context (annex 3). These services were provided to target populations through outreach teams and at drop-in- centers. 4) Program management and capacity building: This component was designed to strengthen the core technical and managerial functions of the NACP and the multi-sector oversight of HIV/AIDS Coordinating Committee of Afghanistan (HACCA) through development of routine monitoring and service quality assessment. In addition, this component included the innovation initiative for multi sector mainstreaming to enable other line ministries, NGOs, and the private . However, this component was cancelled due to long governmental procurement process. C. Project’s results framework (annex 1): III. Implementation This section includes information implementing agencies, technical and financial management: A. Technical, financial, and procurement management of AHAPP The overall project management was coordinated by a multi-sector review panel led by NACP and including HACCA and Grants and Contract Management Unit (GCMU). 6 WHO/UNODC IDU target guide (2009). 42 NACP was playing the leadership role in managing AHAPP through its coordination with other MOPH departments, and other key-non-health sectors. NACP is staffed with a number of qualified national officers and consultants in areas like advocacy and communication, targeted interventions, surveillance, and monitoring and evaluation. In addition, NACP is capacitated with an international advisor. Through its various HIV- technical working groups (e.g. surveillance, M&E, targeted interventions), NACP was providing the technical support to the implementing partners through regular technical committees in different HIV areas. HACCA has been established by MOPH as a multisector committee and playing a critical role in ensuring a broad-based and inclusive oversight mechanism to HIV programming in the country. GCMU is the MOPH department that is responsible for overall financial and procurement management of AHAPP in coordination with other involved line ministries e.g. Ministry of Finance. To strengthen its capacity, GCMU was, additionally, staffed with qualified and experienced procurement specialist and accountant. B. Target populations, implementing agencies and provinces: Populations at higher risk of HIV infection were targeted by AHAPP interventions. They are PWIDs, SWs, prisoners, and truck drivers. AHAPP was operating in four cities (Kabul, Mazar-i-Sharif, Hirat, and Jalalabad) and activities under AHAPP contracts were implemented by nine NGOs; fiver local NGOs andfour INGOs. Please refer to annex 2 for details. IV. Results framework analysis This section includes assessment of outcome indicators of the project as well as the performance/output indicators: 1) Advocacy and communication This component was implemented by Futures Group in partnership with NACP over 30 months (July 15, 2008 – January 15, 2011). Activities were implemented as per the proposed plan and outcomes for this component were achieved (detailed reports are available upon request). 2) Strengthening surveillance  Two rounds of IBBS were conducted during the life time of AHAPP. Results show signs of concentrated HIV epidemic among PWIDs in Hirat province. The first round was conducted among four target populations in three cities and the second round was designed to cover five target populations across five locations in the country.  Two rounds of KAP surveys among policy makers and their influencers were completed during the three years of AHAPP projects. Findings, conclusion, and recommendations were recently presented to CGHN /MoPH for further programing of HIV interventions in the country (full reports are available upon request)  As per the epidemic modeling to estimate the HIV prevalence among general population, IBBS data were not sufficient to conduct such a modeling. 43 However, Data from two rounds of IBBS represent a strong evidence to describe the HIV pattern among populations at higher risk of HIV and can be used as a tool to report on progress indicators e.g. UNGASS.  Rapid assessment of PWIDs in Western provinces of Afghanistan was also conducted and provided a detailed picture of this key affected population in terms of HIV epidemic profile and needed/recommended interventions to be considered in future HR planning. N.B. the end of project report summarizing the achievement done so far in this component is yet to be received by the implementing agency (Johns Hopkins University). 3) Targeted interventions These interventions were amongst the most successful HIV prevention programs in the country with standardized protocols and data collection tools. The following progress through target intervention implementation. Condom and NSP distribution among PWIDs through six Targeted Interventions 2008- 2012 Contract/location Total Regular NSP NSP % of NSP Condom 2008-2012 Users IDUs distribution collection collection distributed Con 3 & 4 Kabul city 1564 989 855,424 814,523 95% 302609 Con 5 Kabul city 541 460 628,781 633,288 101% 119357 Con 6 Mazar-e-sharif city 613 449 367,667 368,464 100% 252043 Con 7 Herat city 1109 1039 1,039,822 967,614 93% 225247 Con 8 Jalalabad city 352 327 330,095 265,812 81% 107889 Con 10 Pul-e-charkhi Prison 166 144 - - 162 Con 23 OST Kabul city 93 71 129,559 120,695 93% 27677 Total 4438 3479 3351348 3170396 95% 1034984 Prevalence of HIV, HBV, HCV and Syphilis among PWIDs in the six Targeted Interventions Contract/location HIV HIV + HBs HBs + HCV HCV Syphilis Syphilis 2008-2012 tested tested tested + tested + Con 3 & 4 Kabul city 3432 53 3482 216 3470 634 3122 136 Con 5 Kabul city 1059 10 1059 53 1059 135 1049 18 Con 6 Mazar-e-sharif city 1141 14 1139 33 1138 102 1139 43 Con 7 Heart city 3770 41 1805 45 1717 411 1322 8 Con 8 Jalalabad city 684 18 797 49 787 74 662 44 Con 10 Pul-e-charkhi Prison 2407 10 2329 133 2310 139 1729 51 Con 23 OST Kabul city 1403 18 1403 63 1396 171 1396 32 Total 13896 164 12014 592 11877 1666 10419 332 The following progress was made against the pre-set target indicators of AHAPP Indicator Baseline Mid project End of project Remarks 44 progress progress 1 HIV prevalence in general <0.5% ? No population intervention was able to measure this indicator 2 HIV prevalence among high <=5% (sample (population risk groups (IDU , SW) estimate) estimate) IDU=7.1% IDU=4.4% Prisoner=1.6% Prisoner=0.7% FSW=0% FSW=0.3% MSM=NA MSM=0.4% RTW=0% RTW=0% 3 Percentage of IDUs reporting 30 % age 69 % 90.0% Sterile use of sterile injection point injecting equipment at last time injected increase equipment at least once in the past 3 months 4 Percentage of IDUs reporting 20 % age 26.8% 23.9% use of condom at last time sex point increase 5 Percentage of SW reporting 50% 58.1% 52.3% Condom use use of a condom with their at last sex most recent client 6 Percentage of truckers 20 % age 24.6% 12% reporting use of a condom last point Ever used Used condom time with sex worker increase condom last last time you time had sex had sex with FSW 7 Percentage of prisoners who 20 % age NSP is not yet report access to sterile point implemented injecting equipment. increase in prisons 8 Score on subset of National Not available Composite Policy Index (UNGASS)** 9 Percentage of those aged 15- NA NA 17.7% AMICS: the 24 who correctly identify ways indicator was of preventing HIV transmission measured – measured in geographic only among areas specific to program. women aged 15-24 10 Percentage of urban centers 3 cities 33% of Annual NGO reports which have done high risk provinces 45 group mapping 11 Number of annual NACP -- 1 2 reports and action plans informed by surveillance and routine monitoring data. 12 Percentage of most at risk -- IDU=62.4% IDU=90.8% Use of risk populations (e.g., IDUs, FSWs, Prisoner=NA Prisoner=NA reduction truckers, prisoners) reached by FSW=1.4% FSW=8.3% services. HIV prevention program MSM=NA MSM=10.8% RTW=1.64% RTW=55.6% 13 Percentage of most at risk -- IDU=NA IDU=22.1% Adequate populations (e.g., IDUs, FSWs, Prisoner=NA Prisoner=6.6% knowledge truckers, prisoners) who FSW=NA FSW=15.3% about HIV correctly identify 2 ways of MSM=NA MSM=4.7% prevention preventing HIV transmission (UNGASS RTW=NA RTW=11.8% definition) 14 Percentage of TI sites meeting Not available service quality standard 15 Percentage of contracted/ NA 92% 92% sanctioned implementation units w/ uninterrupted fund flow and service delivery. Opioid Substitution Therapy (OST): by the end of AHAPP, a national OST workshop was conducted in order to create a national consensus for scaling up OST to cover more of the target population and widen its geographical availability in the country. This workshop was based on the successful and effective implementation of OST pilot during the life time of AHAPP. As per the request from MoCN, the harm reduction guidelines were drafted and updated to be used as a starting step towards scaling up OST in three more cities in addition to Kabul. 4) Program management and capacity building NACP is well staffed by qualified personnel in almost all technical areas of HIV programing including harm reduction, surveillance, and monitoring and evaluation. An international HIV advisor has joined NACP since April 2012 to help capacitate technical and management aspects of NACP. NACP has a number of coordination mechanisms at different levels. Internal staff meetings are weekly held to discuss and plan for specific emerging activities and events. Quarterly workshops are also organized with all AHAPP implementers to discuss progress of different interventions and exchange experiences across all NGOs. In addition, HIV technical groups are invited on regular basis in areas like surveillance and monitoring and evaluation. A new HIV testing and counseling guidelines were drafted and shared with technical groups for further processing and implementation of up-to-date principles of HIV testing in VCTs and DICs for better quality of the service. 46 Furthermore, HACCA is coordinating other HIV topics and issues among high profile stakeholders for the purpose of mainstreaming HIV and AIDS in other governmental and private sectors. V. Project expenditure Please refer to annex 2 for expenditures of individual contracts. VI. Lessons learned A number of lessons were learnt throughout the period of this project:  The project was successful, for the first time in Afghanistan, to provide a baseline data for HIV epidemic pattern of the country through the first two rounds of IBBS surveys.  In addition to the HIV biological component of the IBBS, a number of behavioral indicators are now available describing risk behaviors among populations at higher risk of HIV infection in five cities of Afghanistan and should be used for further elaboration on the dynamics of HIV in the country.  The developed expertise of the implementing partners (e.g. national NGOs) in the area of targeted intervention should be further scaled-up and continued through technical and financial support.  When scaling up a pilot program such as OST, it is important to make sure that a national consensus and strategy are in place, and not simply a budget.  Harm Reduction was new practice in Afghanistan, so the problem of police community leaders with PWIDs could be solved through trainings and meetings.  An effective decrease in the stigma and discrimination cannot be expected in the short term.  Harm Reduction program challenges can be overcome through outreach and advocacy meetings, which continue to be held with local authorities throughout the project lifetime.  Access to PWIDs is possible through a “Peer Network� outreach workers  In the absence of targeted interventions, a large number of Heroin smokers would shift to injection.  An effective counseling could change the mode of drugs use i.e. from injecting to non-injecting drug use  Identification of hot spots and establishment of safe places for PWIDs is very successful ways access to clients  The role of Peer Educators has proven to be most important in targeted intervention projects. VII. Annexes Summary of Borrower's ICR Annex 1: Project results framework 47 Project Indicators Baseline Target End Frequency Data Yr 0 Yr 3* and Collection Reports Instruments Program Development Objectives: 1. HIV prevalence in general <0.5% <0.5% Every 2 Modeling population years 2. HIV prevalence among high risk IDU: 3% <=5% Every 2 IBBS groups (IDU , SW) years Project Key Performance Indicators: 3. Percentage of IDUs reporting use 30 % age point of sterile injection equipment at last --% Every 2 yrs IBBS increase time injected 4. Percentage of IDUs reporting use 20 % age point --% Every 2 yrs IBBS of condom at last time sex increase 5. Percentage of SW reporting use of --% 50% Every 2 yrs IBBS a condom with their most recent client 6. Percentage of truckers reporting 20 % age point use of a condom last time with sex --% Every 2 yrs IBBS increase worker 7. Percentage of prisoners who report 20 % age point --% Every 2 yrs IBBS access to sterile injecting equipment. increase 1. Project Intermediate Outcomes & Outputs by component: Component 1. Advocacy and Communications 8. Score on subset of National -- Every 2 Policy Composite Policy Index years assessment (UNGASS)** 9. Percentage of those aged 15-24 --% 20 % age point Every 2-3 NRVA who correctly identify ways of increase years preventing HIV transmission – measured in geographic areas specific to program. Component 2. Strengthening Surveillance 10. Percentage of urban centers which 3 cities 33% of Annual NGO reports have done high risk group mapping provinces 11. Number of annual NACP reports -- Every year Annual Annual and action plans informed by report and surveillance and routine monitoring action plan data. Component 3. Targeted Interventions for at risk groups 12. Percentage of most at risk -- IDU: 60% Every 2 yrs; IBBS populations (e.g., IDUs, FSWs, truckers, prisoners) reached by HIV FSW: 50% prevention program Trucker: 25% Annual Routine monitoring Prisoner: 80% 13. Percentage of most at risk IDU: 50% --% Every 2 yrs IBBS populations (e.g., IDUs, FSWs, 48 truckers, prisoners) who correctly FSW: 50% identify 2 ways of preventing HIV transmission Trucker: 50% Prisoner: 75% 14. Percentage of TI sites meeting IDU: 5 of 6 service quality standard FSW 1 of 1 Balanced -- Annual score card Trucker: 1 of 1 assessment Prisoner: 1 of 1 Component 4. Program management and Capacity building 15. Percentage of contracted/ sanctioned implementation units w/ GCMU -- 85% Annual uninterrupted fund flow and service reports delivery. 49 Summary of Borrower's ICR Annex 2: AHAPP service contracts Contract Implementer Contract type Location Time frame Expenditu Number From To res (USD) 1 AFG/MOPH/ Futures Group HIV Advocacy & All AHAPP 15 July 2008 15 Jan 2011 1,063,200 HIV/001 International Communication provinces (FGI) 2 AFG/MOPH/ Johns Hopkins HIV Surveillance All AHAPP 26 May 2008 30 Nov 2012 1,532,171 HIV/002 University provinces (JHU) 3 AFG/MOPH/ Zendagi Nawin- TI for IDU Kabul West 7 Aug 2008 30 June 2012 953,022 HIV/003 OTCD 4 AFG/MOPH/ Zendagi Nawin- TI for IDU Kabul East 12 Feb 2009 30 June 2012 341,079 HIV/004 OTCD 5 AFG/MOPH/ Solidarity for TI for IDU Kabul 10 Aug 2008 30 June 2012 827,242.6 H Afghan V/005 Families (SAF) 6 AFG/MOPH/ Solidarity for TI for IDU Mazari- 10 Aug 2008 30 June 2012 814,219.7 HIV/006 Afghan Sharif Families (SAF) 7 AFG/MOPH/ Shahamat TI for IDU Herat 10 Aug 2008 30 June 2012 855,288 HIV/007 Health & Rehabilitation Organization (SHRO) 8 AFG/MOPH/ Zendagi Nawin TI for IDU Jalalabad 10 Aug 2008 30 June 2012 623,565.7 HIV/008 - OTCD 9 AFG/MOPH/ JACK TI for sex workers Kabul 20 Sep 2008 30 June 2012 427,119.3 HIV/009 & clients 10 AFG/MOPH/ Samayee TI for prisoners Kabul 7 Aug 2008 30 June 2012 402,791.0 HIV/0010 Development inc. operations Organization research (SDO) 11 AFG/MOPH/ Action Aid (AA) TI for truckers & Kabul 7 Aug 2008 30 June 2011 492,699.5 HIV/0011 partners 12 AFG/MOPH/ Medicins De Piloting OST for Kabul 19 Jul 2009 30 June 2012 275,551.7 HIV/0023 Monde (MDM) IDU 50 Summary of Borrower's ICR Annex 3: conventional and MoPH package for harm reduction services for TIs MOPH/NACP TI Package of Services Conventional international package 1. Distribution of Safe-Injecting kits 1. Needle and syringe programmes 2. Collection of used needles and 2. Opioid substitution therapy & other drug syringes dependence treatment 3. Counselling for blood borne diseases 3. HIV testing & counselling including VCT for HIV 4. Anti-retroviral therapy 4. Hepatitis C/B testing 5. Prevention and treatment of STIs 5. Condom promotion 6. Condom promotion programmes 6. Syndromic management of STIs 7. Targeted information, education and 7. Abscess management communication for IDUs and their sexual 8. Overdose management partners 9. Referral for TB services 8. Vaccination, diagnosis and treatment of viral 10. Referral to ARV centre for HIV+ hepatitis clients 9. Prevention, diagnosis and treatment of TB 11. Referral for drug detoxification and abstinence based treatment 12. Primary health check ups 13. Hygiene kits 14. IEC 15. Social services 51 Annex 6. List of Supporting Documents Journal Articles and Studies 1. Responding to HIV in Afghanistan. Saif-ur-Rehman, Rasoul MZ, Wodak A, Claeson M, Friedman J, Sayed GD., Lancet, 2007. 2. HIV, hepatitis C, and hepatitis B infections and associated risk behavior in injection drug users, Kabul, Afghanistan. Todd CS, Abed AM, Strathdee SA, Scott PT, Botros BA, Safi N, Earhart KC. Emerg Infect Dis. 2007a. 3. Seroprevalences and co-infections of HIV, hepatitis C virus and hepatitis B virus in injecting drug users in Quetta, Pakistan. Achakzai M, Kassi M, Kasi PM. Trop Doct. 2007. 4. The explosive human immunodeficiency virus type 1 epidemic among injecting drug users of Kathmandu, Nepal, is caused by a subtype C virus of restricted genetic diversity. Oelrichs RB, Shrestha IL, Anderson DA, Deacon NJ. J Virol. 2000. 5. Association between expatriation and HIV awareness and knowledge among injecting drug users in Kabul, Afghanistan. Todd CS, Abed AM, Strathdee SA, Scott PT, Botros BA, Safi N and Earhart KC. Confl Health. 2007b. 6. The Global HIV Epidemics among Men who have Sex with Men . Beyrer C. et al. Directions in Development Series. World Bank, 2011a. 7. The Global HIV Epidemics among Sex Workers. Kerrigan D, Wirtz A, Baral S, Decker M, Murray L, Poteat T, Pretorius C, Sherman S, Sweat M, Semini I, N'Jie N, Stanciole A, Butler J, Osornprasop S, Oelrichs R, and Beyrer C. Directions in Development Series, World Bank, 2012. 8. Buying results? Contracting for health service delivery in developing countries. Loevinsohn B, Harding A. Lancet. 2005. 9. Contracting for health services: Effects of utilization and quality on the costs of the Basic Package of Health Services in Afghanistan. Ameli, O. and Newbrander, W. Bulletin of the World Health Organization, 2008. 10. The Global HIV Epidemics among People Who Inject Drugs . Dutta, Wirtz, Stanciole, Oelrichs, Semini and Cleghorn., Directions in Development Series, World Bank, 2012. 11. Effects of duration of injection drug use and age at first injection on HCV among IDU in Kabul, Afghanistan. Bautista CT, Todd CS, Abed AM, Botros BA, Strathdee SA, Earhart KC, Safi N, Scott PT. J Public Health (Oxf). 2010. 12. The Golden Crescent and HIV/AIDS in Central Asia: Deadly interactions . Beyrer C. Glob Public Health. 2011b. 13. Implementing harm reduction for heroin users in Afghanistan, the worldwide opium supplier . Maguet O, Majeed M. Int J Drug Policy. 2010. 14. Prevalence of HIV, hepatitis B and hepatitis C and associated risk behaviours amongst injecting drug users in three Afghan cities. Nasir A. et al. Int J of Drug Policy, 2011. 15. Concentrated epidemics of HIV, HCV, and HBV among Afghan refugees . Khanani MR, Ansari AS, Khan S, Somani M, Kazmi SU, Ali SH. J Infect. 2010. 16. HIV awareness and condom use among female sex workers in Afghanistan: implications for intervention. Todd CS, Nasir A, Stanekzai MR, Scott PT, Close NC, Botros BA, Strathdee SA, Tjaden J. AIDS Care. 2011. 17. HIV, hepatitis B, and hepatitis C prevalence and associated risk behaviors among female sex workers in three Afghan cities. Todd CS, Nasir A, Stanekzai MR, Bautista CT, Botros BA, Scott PT, Strathdee SA, Tjaden J. AIDS. 2010. 18. Seroprevalence and correlates of HIV, syphilis, and hepatitis B and C virus among intrapartum patients in Kabul, Afghanistan. Todd CS, Ahmadzai M, Atiqzai F, Miller S, Smith JM, Ghazanfar SA, Strathdee SA. BMC Infect Dis. 2008. 19. Molecular epidemiology of HIV type 1 infection in Iran: genomic evidence of CRF35_AD predominance and CRF01_AE infection among individuals associated with injection drug use . Jahanbakhsh F, Ibe S, Hattori J, Monavari SH, Matsuda M, Maejima M, Iwatani Y, Memarnejadian A, Keyvani H, Azadmanesh K, Sugiura W. AIDS Res Hum Retroviruses. 2013. 52 Reports 1. Evaluation of Afghanistan OST Pilot Project. Emran Razaghi. World Health Organization, Eastern Mediterranean Regional Office, April 2011. 2. Opiate Substitution Therapy Evaluation. Kabul, Afghanistan, Recommendations . Johns Hopkins Bloomberg School of Public Health. June 2012. 3. Integrated Behavioral & Biological Surveillance (IBBS) in Afghanistan. Findings of 2009 IBBS survey. Johns Hopkins Bloomberg School of Public Health, 2010. 4. Integrated Behavioral & Biological Surveillance (IBBS) in Afghanistan. Findings of 2012 IBBS survey and comparison to 2009. Johns Hopkins Bloomberg School of Public Health, 2012. 5. Assessment of the Islamic Republic of Afghanistan HIV & AIDS Prevention Programme . Gordon Mortimor. Submitted to: The World Bank Afghanistan, 2010. 6. Observing the World AIDS Day in Afghanistan, December 03, 2011. A Brief Report, Islamic Republic of Afghanistan, National AIDS Control Program. 2011. 7. Annual Project Report: July 2009 to June 2010. Afghanistan HIV&AIDS Prevention Project (AHAPP), Grant No. H-328 AF, NACP, August 2010. 8. National AIDS Control Program Semi- annual Report 1st Saratan-30th (22 June-22 Nov 2011). Ministry of Public Health, Afghanistan, December, 2011. 9. Building on early gains in Afghanistan’s Health Nutrition and Population Sector: challenges and options. Belay, T. World Bank, Washington DC, 2010. 10. The Afghanistan Health Sector – A Decade of Tremendous Progress. Submission to the United States Department of the Treasury, Development Impact Honors 2012. By: World Bank, Afghanistan Health, Nutrition and Population Team. C. Kostermans, 2012. 11. Afghanistan Country Program Evaluation 2002-2011. Independent Evaluation Group (IEG), World Bank. Washington DC, 2011 12. World Development Report 2011: Conflict, Security, and Development . World Bank, Washington DC, 2011. 13. Report of Findings of an Ethnographic Study of Injecting Drug Use in Afghanistan Context - Specific Implications for HIV Prevention. Report by: F. Mansoor & F. Claudio. World Vision International. 2011. 14. AIDS epidemic update, UNAIDS, Geneva. 2004. Bank Documents 1. Project Concept Note, Afghanistan HIV/AIDS Prevention Project (P101502), 2006 2. Afghanistan HIV/AIDS Prevention Project (P101502), Implementation Status & Results Reports and associated mission aides–mémoire: January 2007 – November 2012. 3. Afghanistan Health Sector Portfolio Review Mission, October 3-11, 2009: Mission Aide–mémoire. 4. Afghanistan HIV/AIDS Prevention Project (P101502) Mid Term Review, February 2010: Mission aide–mémoire 5. Afghanistan HIV/AIDS Prevention Project (P101502) Pre-appraisal January 2007: Mission aide– mémoire 6. Technical Annex (Report No: T7698 – AF) – Proposed Afghanistan HIV/AIDS Prevention Project, July 5, 2007 7. Integrated Safeguards Data Sheet (#43940) Afghanistan HIV/AIDS Prevention Project (P101502) Appraisal Stage, Date prepared/updated: 05/08/2007. 8. Memorandum and Recommendation of the President (Report No: P7698 – AF) for a Proposed Afghanistan HIV/AIDS Prevention Project, July 5, 2007 9. Emergency Project Paper on a Proposed Grant to the Islamic Republic of Afghanistan (Report No. 47144-F) for Strengthening Health Activities for the Rural Poor (SHARP) Project February 18,2009 53 10. Project Information Document (#AB2839). Afghanistan HIV/AIDS Prevention Project (P101502) Appraisal Stage, July 2007. 11. Restructuring Paper Afghanistan HIV/AIDS Prevention Project (P101502), June 28, 2010. 12. Emergency Project Paper on a Proposed Grant to the Islamic Republic of Afghanistan (Report No. 73645-AF) for System Enhancement for Health Action in Transition (SEHAT) Project February 9, 2013 13. Implementation Completion and Results Report (IDA-H0430 IDA-H2060 IDA-H3840) on Grants to the Islamic Republic of Afghanistan for a Health Sector Emergency Reconstruction and Development Project (Report No. ICR00001294), March 31, 2010 Government Documents 1. Afghanistan National HIV/AIDS Strategic Framework 2011-2015 (NSF II), Government of the Islamic Republic of Afghanistan, Kabul, Afghanistan, December, 2011. 2. National HIV and AIDS Policy. Government of the Islamic Republic of Afghanistan, Ministry of Public Health, National AIDS Control Program, 2012. 3. Afghanistan National Development Strategy – Summary Report. An Interim Strategy for Security, Governance, Economic Growth & Poverty Reduction. Government of the Islamic Republic of Afghanistan. 2005. 4. Afghanistan National Development Strategy 1387 – 1391 (2008 – 2013) A Strategy for Security, Governance, Economic Growth & Poverty Reduction , Government of the Islamic Republic of Afghanistan, 2008. 5. Implementation Completion and Result Report, Afghanistan HIV/AIDS Prevention Project, Islamic Republic of Afghanistan, Ministry of Public Health, DG of Preventive Medicine, Communicable Disease Control Directorate, National AIDS Control Program. 2013. 6. Afghanistan National HIV Code of Ethics. Islamic Republic of Afghanistan, Ministry of Public Health, June 2007. 54 Annex 7. Figures Figure A. Regional HIV prevalence data in IDU. UNAIDS, 2004 40 6 Round 4 Round 4 35 5 HIV Prevalence - percent HIV Prevalence - percent 30 4 Round 3 25 Round 3 Young 20 PWID 3 15 Round 2 2 10 Round 2 Round 1 1 5 Round 1 0 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Modeled male PWID prevalence in Goals Modeled MSW/HSW prevalence in Goals National PWID prevalence from surveys National weighted average MSW/HSW prevalence from surveys Figure B Comparison of model and survey-based HIV prevalence estimates in Pakistan over 2001-11 for PWID (left) and for MSW/HSW (weighted average) (right) Source: The Global HIV Epidemics among People Who Inject Drugs. Dutta, Wirtz, Stanciole, Oelrichs, Semini and Cleghorn. World Bank, Directions in Development Series, 2012. (Authors’ calculations from referenced national data). 55