Document of The World Bank Report No: ICR00003156 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-H1520 TF096672) ON A GRANT AND AN ADDITIONAL TRUST FUND GRANT IN THE AMOUNT OF US$72.04 MILLION TO THE SOCIALIST REPUBLIC OF VIETNAM FOR THE VIETNAM HIV/AIDS PREVENTION PROJECT June 24, 2014 Human Development Department (EASHH) Health, Nutrition and Population East Asia and Pacific Region CURRENCY EQUIVALENTS Currency Unit = GBP GBP 1.00 = US$ 1.5152 (effective when IDA grant was negotiated) GBP 1.00 = US$ 1.45 (effective in 2010 with first Additional Financing paper) GBP 1.00 = US$ 1.6079 (effective in 2013 with second Additional Financing paper) SDR 1.00 = US$1.42169 (effective when IDA grant was negotiated) FISCAL YEAR 1 January to 31 December ABBREVIATIONS AND ACRONYMS AF Additional Financing IEC Information, Education and Communication AIDS Acquired Immuno-Deficiency Syndrome ISR Implementation Status and Results Report ART Antiretroviral Therapy M&E Monitoring and Evaluation BCC Behavioral Change Communications MOF Ministry of Finance BSS Behavioral Sentinel Surveillance MOH Ministry of Health Centers for Disease Control and Prevention CDC MOLISA Ministry of Labor, Invalids and Social Affairs (United States) CPMU Central Project Management Unit MSM Men who have Sex with Men CSW Commercial Sex Worker/s NTP National Targeted Program Department for International Development DFID PAC Provincial AIDS Committee (United Kingdom) DGA Development Grant Agreement NSP Needle and Syringe Program Ethnic Minority Policy Framework for HIV EMPF Prevention amongst Ethnic Minorities in PAD Project Appraisal Document Vietnam FSW Female Sex Worker/s PDO Project Development Objective Global Fund for AIDS, Tuberculosis and GF PE Peer Educator Malaria United States President’s Emergency Plan For HCMC Ho Chi Minh City PEPFAR AIDS Relief People Living With HIV/AIDS (new HCT HIV Counselling and Testing PLWHA terminology often used: PLHIV HIV Human Immunodeficiency Virus PPMU Provincial Project Management Unit HIV Sentinel Surveillance, plus core sexual HSS+ STI Sexually Transmitted Infection behavior questions Vietnam Administration for HIV and AIDS IBBS Integrated Bio-Behavioral Survey VAAC Control Implementation Completion and Results ICR VCT Voluntary Counseling and Testing Report IDA International Development Association WB World Bank IDU Injecting Drug User/s WHO World Health Organization Vice President: Axel Von Trotsenburg Country Director: Victoria Kwakwa Sector Manager: Toomas Palu Project Team Leader: Mai Thi Nguyen ICR Team Leader: Marelize Gorgens VIETNAM HIV/AIDS Prevention Project CONTENTS Data Sheet A. Basic Information .........................................................................................................................i B. Key Dates......................................................................................................................................i C. Ratings Summary..........................................................................................................................i D. Sector and Theme Codes ............................................................................................................ ii E. Bank Staff ................................................................................................................................... ii F. Results Framework Analysis ...................................................................................................... iii G. Ratings of Project Performance in ISRs ...................................................................................xiv H. Restructurings and Additional Financing .................................................................................xiv I. Disbursement Graph .................................................................................................................xvi 1. Project Context, Development Objectives and Design ................................................................ 1 2. Key Factors Affecting Implementation and Outcomes .............................................................. 10 3. Assessment of Outcomes ............................................................................................................ 20 4. Assessment of Risk to Development Outcome .......................................................................... 27 5. Assessment of Bank and Borrower Performance ....................................................................... 29 6. Lessons Learned ......................................................................................................................... 33 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners ............................ 35 Annex 1. Project Costs and Financing........................................................................................ 37 Annex 2. Outputs by Component ............................................................................................... 39 Annex 3. Economic and Financial Analysis ............................................................................... 59 Annex 4. Bank Lending and Implementation Support/Supervision Processes .......................... 66 Annex 5. Project Risks and Project Transition Arrangements ................................................... 68 Annex 6. Comments on Draft ICR from the Borrower .............................................................. 75 Annex 7. Comments of Cofinanciers and Other Partners/Stakeholders ..................................... 76 Annex 8. List of Documents Consulted ..................................................................................... 77 Annex 9. List of Publications Generated from data that the Project financed............................ 79 Annex 10. Report on review of implementation of project’s transition arrangements ................. 80 Annex 11. Summary of Return on Investment Calculations (from UNSW report) ..................... 91 Annex 12: Details of correlation between project provinces achieving their IDU and FSW coverage and concomitant reductions in HIV prevalence. ......................................... 93 Annex 13. Project provinces, additional Project provinces after DFID AF and the overlap with PEPFAR provinces in Vietnam .......................................................................... 95 DATA SHEET A. Basic Information Country: Vietnam Project Name: HIV/AIDS Prevention International Development Association (IDA) grant: P082604 L/C/TF Project ID: IDA-H1520,TF-96672 Additional financing 2010: P143540 Number(s): Additional financing 2013: P144537 ICR Date: 06/20/2014 ICR Type: Core ICR SOCIALIST Lending SIL Borrower: REPUBLIC OF Instrument: VIETNAM Original Total Disbursed XDR 23.10M XDR 23.08M Commitment: Amount: Revised US$72.04M Amount: Environmental Category: B Implementing Agencies: Ministry of Health Co-financiers and Other External Partners: Co-financier DFID; external partners WHO, US Government, and the Global Fund to Fund AIDS, TB and Malaria B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 12/18/2003 Effectiveness: 10/17/2005 10/17/2005 3/16/2010 Appraisal: 12/06/2004 Restructuring(s): 1/23/2013 Approval: 03/29/2005 Mid-term Review: 12/08/2008 12/08/2008 Additional 10/03/2009 Financing open IDA Project Closing: 12/31/2011 12/31/2012 Additional 12/31/2013 12/31/2013 Financing closing C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Moderately Satisfactory Risk to Development Outcome: Substantial Bank Performance: Satisfactory Borrower Performance: Moderately Satisfactory i C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Satisfactory Implementing Quality of Supervision: Satisfactory Moderately Satisfactory Agency/Agencies: Overall Bank Overall Borrower 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 No None at any time (Yes/No): (QEA): Problem Project at any Quality of No None time (Yes/No): Supervision (QSA): DO rating before Moderately Closing/Inactive status: Satisfactory D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 15 15 Health 70 70 Other social services 5 5 Sub-national government administration 10 10 Theme Code (as % of total Bank financing) Decentralization 25 20 HIV/AIDS 50 60 Other social development 25 20 E. Bank Staff Positions At ICR At Approval Vice President: Axel van Trotsenburg Jemal-ud-din Kassum Country Director: Victoria Kwakwa Klaus Rohland Sector Manager: Toomas Palu Fadia M. Saadah Project Team Leader: Mai Thi Nguyen Maryam Salim ICR Team Leader: Marelize Gorgens n/a ICR Primary Author: Marelize Gorgens n/a ii F. Results Framework Analysis Project Development Objective (from Project Appraisal Document) According to the originally-signed Development Grant Agreement (DGA), the development objective (PDO) of this project was to assist the Recipient in establishing and maintaining national, provincial and local policies and capacity to design, implement and evaluate information and service delivery programs designed to halt the transmission of HIV/AIDS among vulnerable populations and between vulnerable populations and the general population, thereby assisting the Recipient in the implementation of its National Strategy on HIV/AIDS Prevention and Control. A shorter version of this PDO was used in the Project Appraisal Document (PAD), the results framework and project Implementation Status and Results Reports (ISRs), and added some specificity to the definition of vulnerable populations (based on epidemiological evidence): To support programs designed to halt transmission of HIV/AIDS among vulnerable populations (people living with HIV/AIDS (PLWHA), Injecting Drug Users (IDU), Commercial Sex Workers (CSW)1, and their clients and sexual partners) and between these vulnerable populations and the general population. In this shorter version of the PDO, the vulnerable populations were specified, which was not the case in the PDO. (a) PDO Indicators Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Percent of vulnerable groups in participating provinces reporting safer injection practices (from an estimated 20% at baseline to 70% at project end), as measured Indicator 1 : by the following indicator in the national Monitoring and Evaluation (M&E) framework, 2007: Percentage of injecting drug users who constantly use clean syringes and needles in the last month. Value (quantitative or 20% (estimate) 70% 85% 82.7% qualitative) Date achieved 1 Mar 2005 31 Dec 2010 15 Jan 2013 20122 97% achieved. This indicator was substantially achieved, even after the project Comments (incl. % target was revised from 80% to 85% at the start of 2013. Baseline values were not achievement) measured, but estimated through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. 1 During the course of the project, the terminology of CSW was replaced with FSW (Female Sex Workers) so as to be more specific about who was targeted. The terminology FSW is therefore used throughout this ICR, except when the PDO is quoted from the PAD. 2 2013 data not available; 2013 target revised prior to 2012 results being available iii Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Percent of vulnerable groups in participating provinces reporting condom use in Indicator 2 sexual intercourse (from an estimated 40% at baseline to 80% at project end), as measured by the following 2 indicators in the national M&E framework, 2007: Percentage of female sex workers reporting condom use with their most recent Indicator 2(a) casual clients Value (quantitative or 40% (estimate) 80% 85% 85.2% qualitative) Date achieved 1 Mar 2005 31 Dec 2010 15 Jan 2013 2012 100% achieved. This indicator was achieved, even after the project target was Comments (incl. % revised from 80% to 85% at the start of 2013. Baseline values were not measured, achievement) but estimated through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. Percentage of female sex workers reporting condom use with their most recent Indicator 2(b) regular clients Value (quantitative or 40% (estimate) 80% 85% 84.8% qualitative) Date achieved 1 Mar 2005 31 Dec 2010 15 Jan 2013 2010 100% achieved. This indicator was substantially achieved, but at the time of the survey, data for 2013 (from the national expanded HIV Surveillance Survey Comments (incl. % (HSS+)) was not yet available. Baseline values were not measured, but estimated achievement) through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. (b) Intermediate Outcome Indicators: Component 1 It should be noted for all these Outcome Indicators that data are only available as of the end of 2012. It was therefore challenging to measure project achievements in 2013, during the second round of Additional Financing (AF). Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years An increased percentage of vulnerable groups (from an estimate 10% at baseline to Indicator 3: 80% by project end) are reached by provincial services, as measured by these indicators in the national M&E framework, 2007: iv Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Indicator 3(a) Percentage of female sex workers receiving condom in the last 6 months Value (quantitative or 10% (estimate) 80% No revision 63.2% qualitative) Date achieved 1 March 2005 31 Dec 2010 2012 79% achieved. This indicator was substantially achieved. Baseline values were not measured, but estimated through program data at the start of the project. This Comments (incl. % was the most reported component of the harm reduction program that female sex achievement) workers (FSW) were exposed to – see Annex 2. Source: Government‘s Project Implementation Results Report, October 2013. Percentage of female sex workers receiving Information, Education and Indicator 3(b) Communication (IEC) materials in the last 6 months Value (quantitative or 10% (estimate) 80% No revision 66.0% qualitative) Date achieved 1 March 2005 31 Dec 2010 2012 83% achieved. This indicator was substantially achieved. Baseline values were not measured, but estimated through program data at the start of the project. This Comments (incl. % was the second most regular component of the harm reduction program that FSW achievement) were exposed to – see Annex 2. Source: Government‘s Project Implementation Results Report, October 2013. Percentage of female sex workers receiving introduction of Sexually Transmitted Indicator 3(c) Infections (STI) treatment in the last 6 months Value (quantitative or 10% (estimate) 80% No revision 52.3% qualitative) Date achieved 1 March 2005 31 Dec 2010 2012 65% achieved. This indicator was partially achieved. Baseline values were not measured, but estimated through program data at the start of the project. This lower percentage reflects the lower number of FSW taking up the STI services Comments (incl. % offered, and the small number of STI clinics relevant to the project, as well as the achievement) fact that the STI component of the project was transferred to other funders in 2010 – see Annex 2. Source: Government‘s Project Implementation Results Report, October 2013. v Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Percentage of female sex workers receiving information from peer educators in Indicator 3(d) the last 6 months Value (quantitative or 10% (estimate) 80% No revision 74.0% qualitative) Date achieved 1 March 2005 31 Dec 2010 2012 93% achieved. This indicator was substantively achieved. Baseline values were not measured, but estimated through program data at the start of the project. This Comments (incl. % achievement reflects the consistent presence of peer educators as the backbone of achievement) the Government’s harm reduction program – See Annex 2. Source: Government‘s Project Implementation Results Report, October 2013. Indicator 3(e) Percentage of female sex workers receiving HIV testing in the last 6 months Value (quantitative or 10% (estimate) 80% 60% 47.8% qualitative) Date achieved 1 March 2005 31 Dec 2010 16 Mar 2010 2012 80% achieved. This indicator was substantively achieved, after the project target Comments (incl. % was revised downwards in 2010. Baseline values were not measured, but estimated achievement) through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. Indicator 3(f) Percentage of IDU receiving clean needles and syringes in the last 6 months Value (quantitative or 10% (estimate) 80% No revision 65.3% qualitative) Date achieved 1 March 2005 31 Dec 2010 2012 82% achieved. This indicator was substantively achieved. Baseline values were not measured, but estimated through program data at the start of the project. Annex Comments (incl. % 2 details the consistent performance of the Government in this area. achievement) Source: Government‘s Project Implementation Results Report, October 2013. Indicator 3(g) Percentage of IDU receiving IEC materials in the last 6 months Value (quantitative or 10% (estimate) 80% No revision 59.1% qualitative) Date achieved 1 March 2005 31 Dec 2010 2012 74% achieved. This indicator was partially achieved. Baseline values were not measured, but estimated through program data at the start of the project. The Comments (incl. % indicator value reduced from 69.7% in 2010 to 59% in 2012, partially because of achievement) the shift away from IEC materials in the project. Source: Government‘s Project Implementation Results Report, October 2013. vi Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Indicator 3(h) Percentage of IDU receiving information from peer educators in the last 6 months Value (quantitative or 10% (estimate) 80% No revision 74.6% qualitative) Date achieved 1 March 2005 31 Dec 2010 2010 93% achieved. This indicator was substantially achieved as of 2010, and Comments (incl. % information after this date was not available. Baseline values were not measured, achievement) but estimated through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. Indicator 3(i) Percentage of IDU receiving condoms in the last 6 months Value (quantitative or 10% (estimate) 80% 60% 28.3% qualitative) Date achieved 1 March 2005 31 Dec 2010 16 Mar 2010 2012 47% achieved. This indicator was only moderately achieved. The reason is that condom distribution was not seen by IDU peer educators as a core part of their responsibility: their focus was on outreach with peers, on needle and syringe Comments (incl. % distribution and pick up – see Annex 2 for details about the focus of the harm achievement) reduction program for IDU. Baseline values were not measured, but estimated through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. Indicator 3(j) Percentage of IDU receiving HIV testing in the last 6 months Value (quantitative or 10% (estimate) 80% 50% 40.0% qualitative) Date achieved 1 March 2005 31 Dec 2010 16 Mar 2010 2012 80% achieved. This indicator was substantially achieved, after the indicator target was revised downwards in 2010, recognizing the challenges of increasing HIV Comments (incl. % counselling and testing (HCT) uptake amongst IDU and the handover of the HCT achievement) sites to other donors – see Annex 2 for details. Baseline values were not measured, but estimated through program data at the start of the project. Source: Government‘s Project Implementation Results Report, October 2013. Number of lubricant sachets distributed to Men having Sex with Men (MSM) in Indicator 3(k) the last 6 months Value (quantitative or Information not 0 480 000 No revision qualitative) available Date achieved 16 Jan 2013 31 Dec 2013 Indicator information not available for the last year of the project. Data about the Comments (incl. % number of lubricant sachets purchased was available, but not the number achievement) distributed. vii Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years An increased percentage (for an estimated 10% at baseline to 90% at project end) Indicator 4 of provinces prepare high quality provincial plans Value 10% (estimate) 90% No revision 91% Date achieved 1 March 2005 31 Dec 2010 2012 100% achieved. This indicator was achieved and reflects the ongoing technical Comments (incl. % assistance provided to the provinces by Central Project Management Unit (CPMU) achievement) and the Regional Institutes – see Annex 2 for details. Source: Government‘s Project Implementation Results Report, October 2013.. An increased percentage of provinces (from an estimated 10% at baseline to 80% Indicator 5 at project end) meet targets specified in provincial plans Value 10% (estimate) 80% 90% 85% Date achieved 1 March 2005 31 Dec 2010 Jan 2013 2010 94% achieved. This indicator was substantively achieved and reflects the Government’s focus on technical support to and capacity building of the Provincial Comments (incl. % Project Management Units (PPMUs) throughout the project life cycle – see Annex achievement) 2 for details. Source: Government‘s Project Implementation Results Report, October 2013. An increased percentage of provinces (from 0 to 90% at project end) establish Indicator 6 adequate M&E systems, including appropriate biological and behavioral surveillance, effectiveness research and program activity and financial monitoring. Value 0% 90% No revision 100% Date achieved 1 March 2005 31 Dec 2010 2012 111% achieved. This indicator was achieved, and reflects the support provided by Comments (incl. % CPMU’s M&E team, as well as the Regional Institutes – see Annex 2 for details. achievement) Source: Government‘s Project Implementation Results Report, October 2013. Increased percentage of provinces offering Methadone maintenance therapy Indicator 7 services (from an estimated 0 at baseline to 6 by project end) Value 0 6 No revision 7 Date achieved 16 March 2010 31 Dec 2012 2012 117% achieved. The indicator achieved its target. This indicator was added after Comments (incl. % the Midterm review, when the methadone maintenance therapy pilot program was achievement) introduced. Source: Government‘s Project Implementation Results Report, October 2013. Increased number of injecting drug users enrolled in Methadone maintenance Indicator 8 therapy services (from an estimated 0 at baseline to 2,000 by project end Value 0 2000 No revision 1927 Date achieved 16 March 2010 31 Dec 2012 2012 96% achieved. The indicator substantively achieved its target. This indicator was Comments (incl. % added after the Midterm review, when the Methadone maintenance therapy pilot achievement) program was introduced. Source: Government‘s Project Implementation Results Report, October 2013. viii (c) Intermediate Outcome Indicators: Component 2 Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Capacity building needs assessment undertaken and capacity building plan Indicator 9 prepared and adopted Indicator Value (quantitative or No plan in place Plan developed dropped during Plan developed qualitative) MTR Date achieved March 2005 Dec 2008 Oct 2008 This is a qualitative indicator, and was dropped during the project Level II restructuring in March 2010, because the indicator has been achieved and the plans Comments were developed. Source: March 2010 Project Restructuring Project Paper Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Key capacity building products, including provincial study tours, policy papers Indicator 10 guidelines, training curricula and courses, research protocols and completed studies Study tours undertaken, 83 Value (quantitative or Qualitative indicator N/A N/A persons trained, and qualitative) 12 academic papers published Date achieved 2010 This indicator’s performance was assessed qualitatively throughout project Comments implementation; achievements are detailed in Annex 2. Key outcomes, including improved policy and contextual environment for Indicator 11 interventions amongst vulnerable groups achieved Value (quantitative or Policy activities Qualitative indicator N/A N/A qualitative) undertaken Date achieved 2010 Achieved -- see details in Annex 2. Comments Source: Government‘s Project Implementation Results Report, October 2013. Indicator 12 At least 60 Innovation Grants awarded Value (quantitative or 0 60 N/A 61 qualitative) Date achieved 2005 2010 2010 102% achieved. All planned Innovation Grants were awarded. See Annex 2 for Comments (incl. % details. achievement) Source: Government‘s Project Implementation Results Report, October 2013. ix Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years At least 20 activities with innovation grants are subsequently sustained with Indicator 13 support from other sources Value (quantitative or 0 20 N/A 33 out of 61 qualitative) Date achieved 2005 2010 2010 165% achieved. This indicator was achieved as several activities received funding Comments (incl. % from external sources. achievement) Source: Government‘s Project Implementation Results Report, October 2013. At least 10 innovative prevention, care support and treatment models identified, Indicator 14 evaluated and promoted Value (quantitative or 0 10 N/A Not measured qualitative) Date achieved 2005 2010 Not measured. According to Government’s report, an assessment of the Comments (incl. % sustainability of the Innovation Projects was not done and this indicator was achievement) therefore not measured. Source: Government‘s Project Implementation Results Report, October 2013. Health providers in pilot receive appropriate training and supervision in Indicator 15 comprehensive treatment, care and support Value (quantitative or Indicator not Qualitative indicator N/A N/A qualitative) measured Date achieved 2010 Achieved. Training took place – see Annex 2 for details. The indicator was dropped in 2010 when the care and treatment sub-component of the project was Comments dropped. Source: Government‘s Project Implementation Results Report, October 2013. Clients in pilot receive appropriate comprehensive treatment, care and support, Indicator 16 including integrated harm reduction services Value (quantitative or Indicator not Qualitative indicator N/A N/A qualitative) measured Date achieved 2010 Not measured. According to the Government’s report, AIDS patients in 05-06 Centers received care, treatment and support. The indicator was dropped in 2010 Comments when the care and treatment sub-component of the project was dropped. Source: Government‘s Project Implementation Results Report, October 2013. Indicator 17 National M&E Framework, operational plan and budget adopted by year 1 Value (quantitative or Qualitative indicator N/A N/A Satisfactory qualitative) Date achieved 2007 Achieved. The project and national M&E system was developed and is still in use. Because it was achieved, this indicator was dropped in 2008 after the Midterm Comments Review confirmed the suitability of the M&E system developed. Source: 2010 AF Project Paper x Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years National and regional institutions strengthened through establishment of AIDS Indicator 18 M&E nuclei and recruitment and training of AIDS M&E staff Value (quantitative or Qualitative indicator N/A N/A Satisfactory qualitative) Date achieved 2013 Achieved. Regional institutes seem to have built capacity and strengthened the Comments technical quality of PPMU annual plans, as per Government report data. Source: Government‘s Project Implementation Results Report, October 2013. Biological, behavioral and health facility surveillance strengthened, effectiveness Indicator 19 research initiated and functioning provincial activity and financial monitoring system established by year 1 Value (quantitative or Qualitative indicator N/A N/A Partially Satisfactory qualitative) Date achieved 2013 Partially Achieved. Several rounds of Integrated Bio0Behavioral Surveys (IBBSs) were undertaken with the support of the regional institutes. 158 surveys were undertaken during project life cycle. But, there was some inconsistency and a lack of uniformity between project surveillance and national surveillance. This was raised in 2008 in the Midterm Review Aide Memoire and again in 2013 in the Comments project transition arrangements assessment, but has not yet been addressed. Because a national coordinated system of surveillance is now being put in place, this indicator is rated as partially satisfactory. Sources: Project Transition Implementation Assessment 2013; Midterm Review Aide Memoire 2008. Mechanisms developed to regularly disseminate M&E results and to use data for Indicator 20 program improvement Value (quantitative or Qualitative indicator N/A N/A Partially Satisfactory qualitative) Date achieved 2013 M&E meetings and national conferences for data dissemination took place. According to a 2008 Government assessment, feedback loops and data use was not Comments optimal. Source: Government’s Midterm Review Report 2008 Health workers and communities report reduced stigma and discrimination Indicator 21 towards PLHWA, IDU, SW and communities Value (quantitative or Baseline not Indicator not Target not set N/A qualitative) measured measured Date achieved 2010 This indicator was not measured and was dropped in 2010. Comments Source: March 2010 AF Paper xi Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Increased number of published or aired (TV, radio) materials on HIV/AIDS by Indicator 22 trained journalists from 2005 Target not set; only Value (quantitative or Indicator not 0 an increase to be N/A qualitative) measured noted Date achieved 2005 2010 This indicator was not measured and was dropped in 2010. Comments Source: March 2010 AF Paper Increased number of people living with HIV/AIDS (PLWHA) who appeared on Indicator 23 radio, TV or print to share their stories from 2005 Target not set; only Value (quantitative or Indicator not 0 an increase to be N/A qualitative) measured noted Date achieved 2005 2010 Increases were recorded during the first three years of implementation, although Comments quantitative data does not exist. This indicator was dropped in 2010. Source: March 2010 AF Paper Increased number of PLWHA who are members of PLWHA associations in Indicator 24 participating provinces Value (quantitative or Indicator not Qualitative indicator N/A N/A qualitative) measured Date achieved 2010 This indicator was not measured and was dropped in 2010. Comments Source: March 2010 AF Paper (d) Intermediate Outcome Indicators: Component 3 Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Project management unit and administrative and procurement sub units established, bank account opened, staff and consultants recruited, implementation Indicator 25 plan prepared and improved, coordination and technical support provided to provinces and coordination services provided to implementation partners Value (quantitative or Qualitative indicator N/A N/A Satisfactory qualitative) Date achieved 2013 This process indicator was measured through procurement and process reviews. Delays were experienced early on, but the CPMU recruited all staff to support the Comments project management of the project. Source: World Bank ISRs xii Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised Target approval Completion or Values documents) Target Years Project management units prepare adequate plans, meet annual implementation Indicator 26 targets and provide timely financial and activity reporting Value (quantitative or Qualitative indicator N/A N/A Satisfactory qualitative) Date achieved 2013 This process indicator was measured through procurement and process reviews. Comments Source: World Bank ISRs Baseline, mid-term and end-of-project management and capacity reviews demonstrate enhanced project management / institutional capacity, evidencing an Indicator 27 increase in the number of provinces judged to have adequate managerial and institutional capacity (0 to 80% by year 3) Value (quantitative or 0 80% N/A Not measured qualitative) Date achieved 2005 2010 2013 Not measured. Baseline capacity reviews were undertaken, but follow up Comments assessments were not done so as to assess changes in capacity. Source: World Bank ISRs Government demonstrate increased capacity to effectively coordinate development Indicator 28 partners’ contribution to AIDS response Value (quantitative or Qualitative indicator N/A N/A Satisfactory qualitative) Date achieved 2013 This process indicator was assessed on an ongoing basis through interaction with development partners. During the ICR mission, discussions with donors took place. There was mixed reviews about the level of interaction and quality of Comments coordination, particularly at the provincial level with the parallel PPMU / Provincial AIDS Committee (PAC) structure. Source: World Bank ISRs xiii G. Ratings of Project Performance in ISRs H. Restructuring and Additional Financing ISR Ratings at Amount Board Restruc Restructuring Disbursed at Approved Reason for Restructuring & Key turing Restructuring PDO Changes Made Date(s) DO IP in USD Change millions Definition of vulnerable populations: Throughout the project, the PDO remained unchanged. However, in the shorter version of the PDO (used in the ISRs, results framework, and WB Operations Portal), October 0.00 N S S ‘vulnerable populations’ were defined. In 2005 (IDA only) the results framework of the IDA grant, the vulnerable populations were defined as being PLHWA, IDU, CSW, and their clients and sexual partners). xiv ISR Ratings at Amount Board Restruc Restructuring Disbursed at Approved Reason for Restructuring & Key turing Restructuring PDO Changes Made Date(s) DO IP in USD Change millions After Vietnam’s National HIV M&E indicators were agreed on in 2005, specific indicators from the M&E framework were selected to disaggregate the key performance indicators that were of a bundled/composite nature. This was needed because both key outcome (PDO) indicators and one of the intermediate outcome indicators were ‘composite’ indicators; i.e. indicators that referred to more than one population or more than one service. In this case, these indicators referred to ‘percent of vulnerable groups’ without disaggregating by vulnerable group and to ‘provincial services’ without disaggregating by which HIV service. This was problematic, as it would have resulted in indicator values that would be difficult to calculate (one would need to sum up across population groups and across HIV services to find an aggregate figure) and aggregated indicator values that would make it difficult to March 7.56 N S S understand variations in scaling up the 2007 (IDA only) coverage of specific HIV services for specific vulnerable populations. Therefore, indicators had to be unpacked so that they could accurately be measured. This deeper definition/disaggregation of relevant key performance indicators did not deter from the initial intent of what the indicator was meant to measure, and was used consistently throughout the project implementation period. This way of aligning the project’s ‘composite’ indicators to the agreed outcome indicators, resulted in all the disaggregated indicators having the same baseline values and the same target values. Although it is not problematic for the indicators to have had the same target value, having the same baseline value is problematic, as this was estimated, but never measured for each of the disaggregated indicators. xv ISR Ratings at Amount Board Restruc Restructuring Disbursed at Approved Reason for Restructuring & Key turing Restructuring PDO Changes Made Date(s) DO IP in USD Change millions This Level II restructuring extended the closing date by 12 months from December 31, 2010 to December 31, 2012 to enable 28.65 the implementation of an expanded version March N S S of the project to 12 more provinces, with 2010 (IDA only) additional financing from DFID. Certain project sub-components were dropped, and indicators amended to accommodate the changes in project sub components. This Level II restructuring extended the closing date by 12 months from December 35.00 31, 2012 to December 31, 2013 to enable (IDA) improved sustainability and implementation January of transition arrangements, and to focus on N MS S 25.60 2013 a new vulnerable population, financed (AF) through second round of additional financing from DFID. One project indicator was added to reflect the new focus on MSM in the project. I. Disbursement Profile 80 70 60 50 IDA Planned 40 IDA Actual 30 AF TF Actual 20 TOTAL Actual 10 0 2005 2005 2006 2007 2008 2009 2010 2011 2012 2013 2013 2013 2013 Q2 Q4 Q4 Q1 Q1 Q1 Q1 Q1 Q1 Q1 Q2 Q3 Q4 xvi 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal 1. Country background at the time of appraisal: By any measure, Vietnam’s meteoric economic and social development in the last two decades is an unprecedented development success story (see Figure 1). The political and economic reforms (Doi Moi) launched in 1986 have transformed Vietnam from one of the poorest countries in the world, with per capita income below US$100 in 1990, to – 25 years later – a lower middle income country, with per capita income of US$1,755. The proportion of the population in poverty fell from 58% in 1993 to 11% in 2011, and indicators of welfare have improved: under-5 mortality has decreased almost three- fold from 67 to 23 per 1000 live births, and the lifetime risk of maternal death has decreased more than 6-fold from 1:180 in 1980 to 1:1100 in 2012. Figure 1: Vietnam’s meteoric economic and human development the last 25 years 2000 120 Health Expenditure Per Capita and NET 1800 GDP per capita (current US$) ODA REceived Per Capita (US$) 1600 Health expenditure per capita (current US$) 100 GDP Per Capita (US$) 1400 Net ODA received per capita (current US$) 80 1200 1000 60 Project Appraisal 800 600 40 400 20 200 0 0 1992 1990 1991 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 60 100 Infant deaths / 1000 live births % births attended by skilled 90 50 80 40 70 Project Appraisal 60 staff 30 50 40 20 30 10 20 10 0 0 2006 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2007 2008 2009 2010 2011 2012 Mortality rate, infant (per 1,000 live births) Births attended by skilled health staff (% of total) Source: World Development Indicators, www.worldbank.org 2. Vietnam’s economy emerged from a deep crisis in the 1980s, and a subsequent period of vigorous growth through the 1990s. At the time of appraisal, in 2002, Vietnam had already seen 1 good results from the policy changes of the previous decade. Growth of national income had averaged 6% over the preceding 4 years. Around the time of appraisal, the Government finalized its first poverty reduction strategy -- the Comprehensive Poverty Reduction and Growth Strategy. Concerns about this Strategy’s implementation – relevant for the implementation of any development effort in the country at the time – were about the lack of a strong coordination unit to drive implementation; the weakness of arrangements for including stakeholders in implementation; weak M&E, notably through poor access to data; lack of alignment of the country’s national growth outcome targets with resource allocation decisions at the central and provincial levels; and better information about how the strategy would be cross-sectorally implemented. 3. Sector background at the time of appraisal: At the time of appraisal, surveillance data showed that Vietnam’s HIV epidemic was rapidly increasing amongst certain populations. As Figure 2a, 2b and 2c illustrate, the rapid rise in the HIV epidemic was concentrated primarily among two sub populations: (a) young IDU – of which approximately 85% were men – with HIV prevalence of 26.8%; and (b) FSW, with HIV prevalence estimated at 4.2% -- both much higher than HIV prevalence of 0.4% among the general population. Figure 2a. Changes in estimated HIV infections, reported AIDS cases and AIDS deaths in Vietnam, 1993 to 2011 Project Appraisal Source: Vietnam Country HIV Progress Report, 2012 Figure 2b. Trends in HIV prevalence among men who inject drugs in Vietnam, 1994 to 2011 Project Appraisal Source: Vietnam Country HIV Progress Report, 2012 2 Figure 2c. Trends in HIV prevalence among female sex workers in Vietnam, 1994 to 2011 Project Appraisal Source: Vietnam Country HIV Progress Report, 2012 4. Amongst these two populations, HIV prevalence at the time was the highest in Vietnam’s two economic hubs, the Haiphong – Hanoi corridor of the Red River delta and the Mekong River delta (which includes Ho Chi Minh City (HCMC)). Injecting drug use was (and remains) central to these epidemics, accounting directly for about 60% of reported cases and contributing indirectly to an even higher percentage, through the nexus between IDU and FSW. Behavioral surveillance survey data confirmed these trends of overlapping HIV transmission risk amongst these vulnerable populations: a quarter to a third of FSW clients are IDU, while FSW report low levels of consistent condom use. The widely-accepted epidemiological wisdom at the time was that the HIV epidemics of some countries in Asia had the potential to explode into the general population: Vietnam, China and India were three countries about which such concerns abounded. 5. What fuelled the concern about the rising epidemics amongst IDU and FSW, was the high proportion of IDU and FSW who were young (62% are young people aged 20 to 29). With 51% of the Vietnamese population being younger than 24 at the time, there was a large population at risk of becoming drug users, changing their drug use habit to injecting drug use, and thus of HIV transmission. 6. Another concern at the time was the profile of drug users in Vietnam. As Figure 3 illustrates, most drug users were IDU and the most common form of drug being used, was heroin. Heroin users typically inject two to five times a day, and this meant an increased risk of sharing needles if clean needles were not available. This, in turn, increases the risk of HIV acquisition. Figure 3. Profile of Drug Users in Vietnam, 2005 Opium 10.2% 0.8% Sniffing Cannabis Smoking 0.4% ATS Injecting Heroin 13.5% 23.0% 63.5% 88.1% Source: UNODC Fact Sheet about Vietnam, 2007 3 7. HIV was highly stigmatized and known as a ‘Social Evil,’ not a medical condition that could be treated. 8. In terms of Vietnam’s health and HIV service delivery systems in place during appraisal, the health sector had the infrastructure to deliver services down to commune level and has had a ‘tradition of caring for all of its citizens.’ There were also significant efforts underway to support decentralized planning, budgeting and service delivery in the health sector. 9. Despite the progress made in terms of health service delivery, there was also, at the time of appraisal, recognition that health care access was unequal in Vietnam, and that the quality of health care at public facilities were not always good. This is evidenced by the fact that almost 61% of Vietnam’s total health expenditures were private, out of pocket spending, and over 85% of health service contacts were with the private sector, most notably through the plethora of pharmacies that sprung up after the deregulation of the pharmaceutical sector. Making use of this architecture of health service delivery – through non-traditional outlets and making use of pharmacy networks – was relevant in terms of the project’s implementation arrangements. 10. In terms of HIV service delivery, the provision of HIV prevention services started in the health sector in Vietnam. But, experience and know-how of how to go about implementing harm reduction HIV programs for vulnerable populations was lacking. Given that the project intended to provide services for hard-to-reach and vulnerable populations, it was recognized at appraisal that HIV harm reduction program design and delivery capacity would need to be strengthened throughout the project. 11. Rationale for WB assistance: The appraisal documents, preceding discussions and missions elucidated five reasons as a rationale for World Bank (WB) involvement in the HIV response in Vietnam, despite the fact that Vietnam was at the time scheduled to receive funds from the Global Fund and from the US Government (their PEPFAR program) as one of the 15 PEPFAR focus countries, and was already receiving some HIV financial and technical support from several bilateral donors such as DFID, JICA and Australia:  First, the epidemiological importance: As stated earlier, Vietnam at the time seemed to be on the precipice of a potentially-exploding HIV epidemic.  Second, the Government’s readiness: The Government’s first comprehensive AIDS prevention strategy – Vietnam’s National Strategy on HIV/AIDS Prevention and Control for the period 2004 to 2010 was just finalized; the Strategy was the first with a focus on harm reduction, as opposed to a punitive approach. A legal framework was developed to support implementation of the national program, that AIDS committees were set up in each province, HIV testing sites had been set up in all provinces, blood screening processes for HIV were been adopted, routine collection of sentinel surveillance data (behavioral and sero- prevalence) was established, and locally-executed prevention activities were commencing, as local HIV prevention efforts was a bedrock of the Government’s new strategy. Financing this bold new approach to comprehensive HIV prevention was seen to be catalytic in changing the course of the HIV response in Vietnam.  Third, the complementary nature of the WB HIV financing, despite other development partners being involved: The WB investment would support interventions and approaches that were complementary to that of other development partners’ support to HIV activities: it would focus on HIV prevention amongst vulnerable populations whereas PEPFAR and the GF focused on financing HIV treatment (this was during the time that the world was trying to 4 scale up HIV treatment to 3 million people by 2005: the global ‘3x5’ target); and it would procure injecting paraphernalia for the needle and syringe programs, something which the US Governments could not support.  Fourth, the WB’s expertise: It was generally seen that the WB brought ‘more than money’ – it had cutting edge technical assistance and unique analytical expertise to help ensure that the HIV investments were the most cost effective.  Fifth, the WB’s experience in managing funding for HIV responses: This project was designed in the hey-day of the MAP program and the WB was generally seen to have unique knowledge of and experience in the implementation of large scale HIV programs whilst maintaining transparent and fair procurement and financial processes. 1.2 Original PDO and Key Indicators 12. According to the originally-signed DGA, the PDO of this project was to assist the Recipient in establishing and maintaining national, provincial and local policies and capacity to design, implement and evaluate information and service delivery programs designed to halt the transmission of HIV/AIDS among vulnerable populations and between vulnerable populations and the general population, thereby assisting the Recipient in the implementation of its National Strategy on HIV/AIDS Prevention and Control. 13. A shorter version of this PDO was used in the PAD, the results framework and project ISRs, and added some specificity to the definition of vulnerable populations (based on epidemiological evidence): To support programs designed to halt transmission of HIV/AIDS among vulnerable populations (PLWHA, IDU, CSWs, and their clients and sexual partners) and between these vulnerable populations and the general population. In this shorter version of the PDO, the vulnerable populations were specified, which was not the case in the PDO. 14. Key outcome indicators (PDO indicators) agreed during project negotiations, documented in the PAD, and used consistently throughout the project until its closure, are:  Percent of vulnerable groups in participating provinces reporting safer injection practices (from an estimated 20% at baseline to 70% at project end); and  Percent of vulnerable groups in participating provinces reporting condom use in sexual intercourse (from an estimated 40% at baseline to 80% at project end). 15. In terms of the targets for these outcome indicators, it was noted in the PAD that province- specific targets would be set once baseline data were available; however, this was not realized during implementation. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 16. Definition of vulnerable groups referenced in the PDO and key outcome indicators in 2005: Throughout the project, the PDO remained unchanged. However, in the PDO (used in the ISRs, results framework, and WB Operations Portal), ‘vulnerable populations’ were defined. In the results framework of the IDA grant, the vulnerable populations were defined as being PLHWA, IDU, CSW, and their clients and sexual partners). 5 17. Disaggregation of key performance indicators in 2005, and subsequent reporting on disaggregated indicators in 2007: After Vietnam’s National HIV M&E indicators were agreed on in 2005, specific indicators from the M&E framework were selected to disaggregate the key performance indicators that were bundled/composite. This was needed because both key outcome (PDO) indicators and one of the intermediate outcome indicators were ‘composite’ indicators; i.e. indicators that referred to more than one population or more than one service. In this case, these indicators referred to ‘percent of vulnerable groups’ without disaggregating by vulnerable group and to ‘provincial services’ without disaggregating by which HIV service. This was problematic, as it would have resulted in indicator values that would be difficult to calculate (one would need to sum up across population groups and across HIV services to find an aggregate figure) and aggregated indicator values that would make it difficult to understand variations in scaling up the coverage of specific HIV services for specific vulnerable populations. Therefore, indicators had to be unpacked so that they could accurately be measured; the unpacked indicators are detailed in the results framework reporting and summarized in Table 3 in Annex 2. 18. Ultimately, this deeper definition/disaggregation of relevant key performance indicators did not deter from the initial intent of what the indicator was meant to measure, and was used consistently throughout the project implementation period. This way of aligning the project’s ‘composite’ indicators to the agreed outcome indicators, resulted in all the disaggregated indicators having the same baseline values and the same target values. Although it is not problematic for the indicators to have had the same target value, having the same baseline value is problematic, as this was estimated, but never measured for each of the disaggregated indicators. 19. Changes to project components and intermediate outcome indicators: During the December 2008 project midterm review, changes were made to two project components, which resulted in changes to the intermediate results and accompanying indicators: a) Dropping Behavior Change Communication (BCC) sub-component indicators: the Bank and the Government agreed that BCC activities should be decentralized to provinces - therefore, BCC activities would be integrated in all provincial action plans and no longer measured centrally (BCC indicators were therefore dropped from the results framework). It was also agreed at the time that the innovative integrated treatment component at rehabilitation centers would be dropped for three reasons: (a) a decrease in the number of detainees; (b) a reduction in the number of sites under the auspices of the project from three to one; and (c) a hand-over of activities to the Global Fund, who would continue the antiretroviral (ARV) program within the scope of a community-based approach. b) Dropping and adding to care and treatment sub-component indicators: The original Care and Treatment subcomponent was designed to promote the development of innovative, effective prevention and treatment approaches among vulnerable groups through integrated demonstration sites. After piloting in three sites, a number of new changes called for project restructuring. The first was a decrease in the number of detainees; the second was a reduction in the number of sites under the auspices of the project from three to one and the third was a hand-over of activities to the Global Fund, who would continue the ARV program within the scope of a community-based approach. Also, methadone pilot sites were introduced as a new part of the care and treatment component of the project. Therefore, with respect to the indicators on care and treatment, it was agreed that the care and treatment indicators on treatment would be dropped because these models are now supported by other donors. It was also agreed that a new indicator will be added to reflect the increased support for methadone substitution treatment. 6 20. These changes were cemented in 2010, when a first round of AF in a form of a Trust Fund Grant from DFID was approved. The original PDO remained unchanged (as per the 2010 DGA amendment agreement), but there were changes to: a) the definition of vulnerable groups referred to in the PDO was changed to reflect IDU, CSWs, and their clients and sexual partners. The 2010 definitional change took place because of a change in project focus to move away from BCC and treatment at the national level; and b) some of the Key Performance Indicators, which were revised and documented in a Supplemental Letter (4 June 2010), as summarized in Annex 2. 21. The definition of vulnerable populations PDO was included in the DGA amendment in 2013 when a second round of AF was agreed on. MSM was included as a vulnerable population in the last year of implementation because of the new strong epidemiological evidence that HIV prevalence amongst MSM was high and increasing. The 2013 PDO read as follows: To assist the Government of Vietnam in establishing and maintaining national, provincial and local policies and capacity to design, implement and evaluate information and service delivery programs designed to halt the transmission of HIV/AIDS among vulnerable populations (including injecting drug users, commercial sex workers and their clients and sexual partners, and men having MSM) and between vulnerable populations and the general population, thereby assisting the Recipient in the implementation of its National Strategy on HIV/AIDS Prevention and Control. 22. MSM were added in 2013 as a target population because of new epidemiological information about HIV prevalence amongst MSM that only came to light in 2009 when sentinel surveillance for MSM was established. Based on limited data, the estimated HIV prevalence in MSM increased from 4.1% in 2003 to 6.5% in 2012. The rising trend is consistent with other findings. Vietnam’s MOH reported that HIV prevalence among MSM increased markedly from 5.3% to 16.7% in HCMC and from 9.4% to 17.4% in Hanoi between 2006 and 2009. Additionally, a review in 2011 had indicated increasing HIV prevalence among MSM in urban Vietnam. 23. When the definition of vulnerable groups was added to the PDO in 2013, the key performance indicators were also revised (see Table 3 in Annex 2). This revision reflected the reality of what had been reported on in the project throughout its lifespan. 24. Because of the addition of MSM as a vulnerable population in 2013, one additional intermediate outcome indicator was added to reflect the new target population being included in the PDO definition: “Number of lubricant sachets distributed to MSM in the last 6 months.” 1.4 Main Beneficiaries 25. Primary project beneficiaries according to the PAD: In the results framework of the IDA Grant, the vulnerable populations were defined – based on available epidemiological data and disease burden – as being PLHWA, IDU, CSW, and their clients and sexual partners. In 2013, the definition of vulnerable populations was adjusted to include MSM. 26. Secondary beneficiaries: Because of the strong capacity building component, direct project beneficiaries were also project managers, government staff and other staff who were trained on public health, harm reduction planning and implementation during the project. Because of the 7 Environmental Safeguards, there was also a specific focus in the Project on providing comprehensive HIV prevention services for ethnic minorities in Vietnam, as per the Ethnic Minority Policy Framework (EMPF). 27. Indirect beneficiaries: Given the epidemiological impact of halting HIV amongst these targeted populations, it was also anticipated during appraisal that the project would also have a positive effect on the general population in that it would reduce new HIV infections amongst these populations. Therefore, the general population is an indirect beneficiary of the project’s impact. 1.5 Original Components (as approved in 2005) 28. Component 1: Implementation of Provincial HIV/AIDS Action Plans (appraisal estimate was US$21.616 million (WB IDA grant)/62% of estimated cost; actual cost US$42.9 million (IDA grant and 2 rounds of AF)/61% of actual cost). This was the main component of the IDA grant and the additional financing and focused on providing sub-grants to 18 provinces and 2 cities to support the design and implementation of Annual Provincial HIV Action Plans that was based on the approved national HIV strategy and adjusted for the local context. Although each provincial action plan was customized for the needs and implementation modalities that would most suit that province, the content of provincial action plans typically consisted of four components: (a) behavior change communication; (b) harm reduction intervention; (c) M&E and (d) capacity building, thus mirroring the overall project objectives. 29. Component 2: National HIV/AIDS Policy and Program Support (appraisal estimate was US$ 8.2million/23% of estimated cost; actual cost US$ 15.1million (IDA grant and 2 rounds of AF)/21% of actual cost): consisted of three sub-components: a) Subcomponent 1: This subcomponent strengthened capacity at national and provincial levels and promoted the development of innovative, effective prevention and treatment approaches and models among vulnerable groups through: (i) demonstration sites that was meant to demonstrate a community-based treatment model to integrate harm reduction with comprehensive HIV treatment and care; (ii) policy studies and research; (iii) knowledge sharing and training; and (iv) other innovations. b) Subcomponent 2: This subcomponent supported the development of a national HIV M&E system and results framework, operational plan and system with which to monitor the country’s HIV response progress and achievements. c) Subcomponent 3: This subcomponent focused on reducing stigma and discrimination was planned to support a 5-year nationwide campaign to reduce stigma and discrimination against vulnerable groups. 30. Component 3: Project Management (appraisal estimate was US$ 5.184million/15% of estimated cost; actual cost US$ 12.6million (IDA grant and 2 rounds of AF)/18% of actual cost). This component was included to bolster management capacity in the Ministry, given the multitude of sub-grants anticipated, the complexity of operations, and the concern at project inception that one of the main challenges with implementing the project was the lack of capacity in designing and implementing HIV harm reduction programs at decentralized levels. As such, this project component was included to provide support to the set up and management of the CPMU as well as the PPMU located in the 18 provinces and 2cities where the project would initially be implemented. 8 1.6 Revised Components 31. Approval of AF in 2010 to scale-up to more Project provinces and fund additional Project activities: In 2010, the United Kingdom’s Department for International Development (DFID) and the WB negotiated AF (‘2010 AF’) through a Trust Fund Grant. The value of the 2010 AF managed by the Government through a Recipient Executed Trust Fund (TF096672) was GBP18 million (US$25.46 million). The activities financed under the AF were similar to that of the IDA grant and thus consistent with the PDO. The AF was intended to complement ongoing activities by mitigating the risk of contracting HIV for vulnerable groups and increase aid effectiveness through donor coordination. This co-financing enabled an increase in the number of Project provinces from 20 to 32 (of the total 63) provinces in Vietnam. During the 2010 AF negotiations, DFID and the WB agreed that at least 60% of project funds would be allocated to harm reduction activities. 32. The Project components were revised as follows, the results framework was adjusted to reflect changes in project indicators, the project closing date was extended to 31 December 2012, and component 2 was restructured, as follows. a) Expansion of Component 1: Implementation of provincial HIV/AIDS Action Plans (US$ 12.34 million). Support was provided to the 20 provinces and cities currently covered, and would be expanded to an additional 12 provinces. The rationale for it being expanded to 32 provinces was as follows: DFID’s first project supported 22 provinces, and WB supported 20. Eight of these overlapped, which meant that after the co-financing, the total number of provinces was 32 to cover both DFID and WB-financed provinces. The provinces were allocated block grants based on specific criteria to determine the size of each year's base allocation. Activities were expected to reflect the diversity of provincial needs as well as the different responses needed in each location. b) Expansion of Component 2: National HIV/AIDS Policy and Program (US$ 5.6 million). This component strengthened capacity at the national and provincial levels, and promote the development of innovative, effective prevention through condoms promotion and needle exchange programs. Opiate substitution therapy in the form of methadone maintenance therapy was also be scaled up with the technical support of the World Health Organization. Simultaneously, the financing of the 06 rehabilitation centers was dropped and the financing of the BCC activities was transferred to the provinces as part of the provincial HIV/AIDS Action Plans, as earlier described. c) Expansion of Component 3: Project Management (US$ 7.0 million). This component will provide support to establish 12 PPMUs in the 12 additional project provinces. It was also designed to provide support to the institutions that will manage and implement the project, such as the Central Project Management Unit (in the Ministry of Health), Provincial AIDS Steering Committees and other implementing agencies. 33. To deepen the impact of the Project, enhance sustainability and focus on HIV prevention service provision for MSM: In January 2013, approval was obtained for further AF of US$9.38 million equivalent (‘AF 2013’) from DFID through an extension of the DFID Trust Fund Grant. This AF resulted in sustaining core project activities whilst financing the distribution of lubricants for MSM. The second round of AF led to the following changes in project components: 9 a) Expansion of Component 1: Implementation of provincial HIV/AIDS Action Plans (GBP 3.6 million/US$5.79 million). One more vulnerable population (MSM) was provided with intervention support in the form of lubricant distribution. b) Changes to Component 2: National HIV/AIDS Policy and Program (GBP1.26 million / US$2.03 million). This component was redesigned to strengthen capacity at the national and provincial levels, and promote the development of innovative, effective prevention through condom promotion and needle exchange programs. c) Expansion of Component 3: Project Management (GBP 0.971 million/US$ 1.56 million). This component provided support to the institutions that would manage, implement, transition and close out the Project. 1.7 Other significant changes 34. Changes in Project Completion Date: The project closing date was extended twice: from 31 December 2010 to 31 December 2012 to accommodate the first round of AF, and from 31 December 2012 to 31 December 2013 to accommodate the second round of AF. 35. Changes to Procurement Category Allocations as a result of AF: The two rounds of AF resulted in revised procurement allocations for the project, as follows: Table 1. Changes to Procurement Category Allocations, 2004 to 2013 (US$) Original Project Category Costs financed DFID AF 2010 DFID AF 2013 Revised Cost by IDA Grant (1) Provincial Sub-Grants 22,600,000 14,583,653 5,840,000 43,023,653 under Component 1, including 960,000 - Goods 630,000 - Consulting Service (2) Innovation Sub-Grants 500,000 64,316 0 564,316 under Component 2.1 (d) (3) Civil works 70,000 128,632 0 198,632 (4) Goods 1,930,000 3,344,432 430,000 5,704,432 (5) Consultants’ services 5,200,000 8,119,895 1,620,000 14,939,895 (6) Training 1,700,000 1,157,688 1,200,000 4,057,688 (7) Incremental Operating 3,000,000 257,264 290,000 3,547,264 Costs (8) Unallocated 0 0 0 0 TOTAL AMOUNT 35,000,000 27,655,880 9,380,000 72,035,880 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry 36. Soundness of background analysis and incorporation of lessons learned: Detailed project preparations took place over 22 months (Reconnaissance Mission was in May 2003 and the project was approved in March 2005). During these preparations, detailed epidemiological 10 analysis was undertaken to understand trends in epidemic trends, determine project provinces (with highest disease burden) and determine which populations to focus on. The project design reflected generic lessons learned from international experience in responding to HIV/AIDS, from the overall development challenges as identified in the CGPRS, and from the implementation of other projects in Vietnam, namely (a) the need for extensive supervision; (b) client/partner commitment; (c) solid M&E; (d) support for the newly formed decentralized government structures; and (e) the management of a “demand-driven” fund for civil society. 37. Assessment of project design (objectives, components and organization): The project objective was and remains relevant given the nature of Vietnam’s HIV epidemic. Project components were typical that of projects financed under the WB Multi-country AIDS Program with an important difference: less funding for civil society organizations and more decentralized implementation by local government entities. There was also not a public sector component for internal HIV mainstreaming activities (which were a most inefficient part of the MAPs). 38. Project organization at design was adequate for the institutional and policy environment: a focus on decentralized implementation and extensive capacity building, given the newness of the harm reduction approach. Whilst the project design did respond to the urgent need to exclusively focus on most at risk populations and provided a clear directive to provinces as to which types of harm reduction activities should be focused on service delivery standards for harm reduction activity implementation were not developed. For example, there were no guidelines as to how many times a week an FSW peer educator should see their clients, or the time spent per peer educator. Peer educators (PEs) were also not given micro planning tools to avoid overlap and duplication of efforts with other peer educators – the concept of a unique identifier code for each peer educator and peer educator client was only piloted in 2011. Further, the project did create large, stand-alone bureaucratic structures that were not linked to the PACs, which were created in 2010. 39. The financing of the IDU rehabilitation centers (‘06 centers’) was the most challenging part of the project design. The Task Team defended their design as follows: “Prior to approving the grant for the Project, the WB did consider the “legal due process issues” around the involuntary nature and lengthy periods of treatment in the drug rehabilitation centers. However, the WB did also consider that avoiding any involvement with such rehabilitation centers would result in the Project missing a crucial segment of the population at risk and a substantial proportion of the national epidemic because: (i) prevalence of HIV was highest among injecting drug users that was driving the epidemic; (ii) a large number of accessible IDU are residents of these rehabilitation centers; and (iii) the management of HIV within these centers is failing to prevent transmission. In addition, the Project sought to demonstrate alternative best practice models for AIDS treatment and prevention among vulnerable populations, for example, by introducing the an extension of community-based care and HIV voluntary counseling and treatment centers and supporting the community based peer-educator networks.” In 2011, after this subcomponent was dropped from project, Human Rights Watch raised a concern about the financing of HIV services for these centers, suggesting that financing the centers with HIV programs would be perceived as a de facto approval of their existence. 40. Co-financiers: The US Government and the Global Fund were two new HIV financiers on the horizon at the time of project design. Discussions with these HIV financiers took place, but the nature of their support could was not yet defined and could, as such, not be further aligned than what was the case. 11 41. Results of Quality Enhancement Review (QER) process: The QER, undertaken midway through project design, found that the Project was in alignment with the CAS, that it was timely in the context of the emerging HIV situation in Vietnam, that the epidemiological context was well articulated to justify the Bank supporting an HIV project, and that the main elements of design were appropriate. The Panel recommended that improvements in M&E and implementation arrangements were necessary before the project could be presented to the Board for approval. 42. Risks and their mitigations: The PAD rated the overall risk of the project as Substantial. The ICR team agrees with this assessment, given the fact that a harm reduction program, at the time, was a new undertaking in Vietnam. The PAD identified important risks, many of which materialized during implementation. As Table 5 in Annex 5 shows, risk was generally well managed in this project: key project risks were identified, the risk mitigation measures were generally appropriate, well anticipated and implemented. 43. Adequacy of government commitment, stakeholder involvement and/or participatory processes during appraisal: At the time of appraisal, Government commitment was perceived (and remained) high during implementation. The Government had just issued a new directive to expand HIV services to the provinces, adopted a harm reduction strategy, and the new Comprehensive Plan for HIV Control (2004 – 2010) was in the process of being developed. Government commitment at Central level was perceived to be high. Given the newness of programs for vulnerable populations and the difference in approach (a harm reduction approach as opposed to a punitive approach), government commitment at provincial level was seen to be an ongoing implementation challenge that needed to be overcome during project implementation, partially through ongoing advocacy with provincial structures and their gaining the experience of implementing these new programs. There were significant discussions and consultations on project design with the UN system, bilateral donors such as US Government with their PEPFAR program and the Global Fund to Fight AIDS TB and Malaria, DFID and NGOs. It was agreed, for example, that where there were WB provinces and PEPFAR provinces that overlapped, implementation would focus on different communes within the province. The participatory nature of planning was validated through independent focus groups with an external evaluator at the end of the project. From their perspective, the design and planning of the project appears to have been systematic and efficient. 2.2 Implementation Changes to the project context during project implementation 44. Changing economic situation during the project implementation period: This project was implemented over almost a decade, during which significant changes took place in the Vietnamese economy: growth averaged 7.5% and Gross Domestic Product rose by 44%. These changes resulted in Vietnam attaining Lower Middle Income Country status in 2010. Although these economic developments are excellent for Vietnamese people and will make a significant difference in their everyday lives, it is a time to be ever more vigilant in terms of HIV transmission: HIV disease pathologies have shown that HIV tracks economic development and is fostered in highly inequitable societies; periods of growth and development have often been associated with more disposable income, higher risk sexual behavior and therefore increased risk of HIV epidemic upswings. 45. Evolution of harm reduction legislation in Vietnam, impact on project implementation and potential impact on future harm reduction activities: Over the decade of DFID/WB supported 12 HIV/AIDS programs (2003-2013), Vietnam has seen a shift in HIV prevention policies and legislations from repressive and punitive control measures to a more pragmatic approach to HIV prevention and control, in line with internationally recognized legislative and policy reform. This is a milestone development in a region of the world where the implementation of harm reduction policies has been marred by political kowtowing. The new policies have been built on the success of the project, and made the implementation of HIV-related harm reduction activities incrementally easier during the project period. In future, it should lead to an even more pragmatic implementation of harm reduction activities and the curtailing of IDU-related incarcerations, provided that policy directives are implemented as planned. See Annex 2 for more details. 46. Rapid increases in external financing for HIV for programs in Vietnam. In 2008 – 2010, Vietnam undertook its first comprehensive National AIDS Spending Assessment (NASA). This analysis showed that 40% of the country’s HIV response was financed by the Government, 12% through out-of-pocket expenses, and 48% through international donors, and that AIDS expenditure rapidly increased during the study period (see Figure 4). Using the 2009 MOH medium estimate of the number of PLHIV in Vietnam, the NASA determined that AIDS expenditure per person living with HIV in the country increased from US$301 in 2007 to US$416 in 2008, US$520 in 2009 and US$607 in 2010. This is high compared to other countries with concentrated epidemics. Figure 4: Changes in AIDS expenditure from 2008 to 2010 in Vietnam Source: Vietnam NASA 2008 - 2010 47. Change in approach to procuring, promoting and distributing condoms: In 2010, the Ministry of Health (MOH) decided to change their approach to purchasing, marketing and distributing condoms3. Instead of purchasing and distributing condoms for free, the Government decided to venture into condom social marketing. This would imply that the distributor of the 3 It should be noted that this decision was made after the co-financing arrangement was agreed with the World Bank. The social marketing of condoms to vulnerable populations has been started through DFID funding project in 2005-2009 (£17.5 million project) by an NGO - DKT. But since 2010, the Government decided to own a brand and conduct the social marketing of condoms by themselves. This may relate to the sustainability issue of this activity, as clearly without donor funding, the Government cannot afford to sustain the brand and the system to conduct social marketing of condoms. 13 condoms would get an amount per condom back from the distributor/provider of the condoms. This amount would be paid over to the PAC, who would send the money back to CPMU for repurchasing new condoms. The FSW peer educators were used to market and distribute the 4 condoms by selling them either individually to other FSW, or to ‘entertainment establishments ’ that are frequented by SW and their clients. This change in the condom distribution process and market was done in 2009, and was a significant change to how the project was implemented. 48. Change to request for the types of injecting equipment purchased: During the course of the project, low dead-space syringes became available on the market. This injecting equipment poses a lower HIV transmission risk, because less fluid is retained in the needle and syringe. “High - dead-space needles retain fluid in the tip of the syringe, hub of the needle and in the needle. Low- dead-space needles, on the other hand, do not have a detachable needle, and after the plunger is fully depressed they only retain fluid in the needle. On average, low-dead-space syringes retain only two microliters of fluid, compared with 84 microliters in high-dead-space syringes.”5 Efforts are underway by several partners to promote the use of low-dead-space syringes (more expensive) in Vietnam because of the lower transmission risk, even if shared. 49. Creation of PACs / Commissions, and the relationship with the PPMUs. Throughout the project duration, PACs were created and strengthened. The PACs are responsible for the coordination of HIV activities in the province, were financed by (and managed the financing of) the Global Fund funding at local level. In 25 Project provinces, the PPMUs were merged with the PACs and in 7 Project provinces, the Provincial Department of Health maintained the PPMU for the World Bank project separate from the PAC. The separate nature of the PPMUs and their perceived role as managing the WB project meant that the collaboration between these two entities was neither automatic nor always positive. In one Project province, the relationship could be described as rocky at best. It is relevant to keep in mind in future project implementation and in discussion with Provincial authorities, to minimize these parallel program management arrangements so as to reduce the potential for conflict as well as to increase the overall efficiency, sustainability and effectiveness of the program management. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization 50. M&E design of the project: A full set of performance indicators was included in the PAD; in some cases, these indicators were ‘composite’ indicators referring to several sub population and several HIV services. After the national HIV M&E system was developed and indicators agreed on, the indicators in the PAD were ‘aligned’ to the indicators in the national HIV M&E system and, subsequently, reported on throughout program implementation. These HIV M&E system indicators were documented officially in the 2013 project AF request, and included in the accompanying DGA supplemental letter. There was, however, a challenge with the selection of baseline values for each of these indicators in the results framework: the assumed ‘umbrella’ indicator baseline value was also attributed to the HIV M&E system indicator, which was not necessarily the measured baseline value of the HIV M&E system indicator. Indicator targets were set aspirationally high, with 80% coverage of harm reduction services and 80% of vulnerable persons expected to adhere to low risk behaviors. 4 This is the local terminology used in Vietnam to refer to local motels or karaoke bars where casual or commercial sex is typically sought – non-traditional outlets for condom distribution. 5 http://www.aidsmeds.com/articles/dead_space_1667_23582.shtml 14 51. Contribution to national HIV M&E system design: Vietnam’s first national HIV M&E system was initiated in 2005 with the agreement on indicators. Once the national HIV M&E indicators were agreed on, the actual indicators used to track performance for the PDO and intermediate outcomes were selected and used throughout. In September 2006, Vietnam’s government issued Decision 26, which required medical clinics throughout the country to collect and report data on HIV/AIDS patients and prevention programs, creating a national M&E framework and system for HIV surveillance. The latest version of the system is referred to as Decision 28, or simply “D-28.” This system, which consists of an online form, ensures that the Vietnam Administration for AIDS Control (VAAC) receives quarterly data on the progress with implementing the HIV response in Vietnam. One of the unique features about HIV M&E in Vietnam is the plethora of behavioral surveillance data available. The Project contributed to the design of Vietnam’s first national M&E framework which would be used by donors and the Government. In 2007, the M&E framework was finalized with Project support. 52. M&E implementation: The Project’s M&E efforts were dovetailed with the Government’s overall M&E efforts. VAAC and its Department for HIV/STI Surveillance and M&E is the lead national HIV M&E institution. Through its four regional institutes, it offers national guidance and M&E technical assistance to all 63 Provincial AIDS Centers. The National Strategic Information and Monitoring and Evaluation Technical Working Group, chaired by VAAC, brings together practitioners from central and provincial level national institutions, international partners and the UN to share resources and experiences. 53. To ensure sufficient human resources for M&E, the CPMU and every PPMU allocated funding for an M&E officer position to ensure that sufficient resources were in place. Several of the provincial M&E officers were trained and received master’s degrees in public health – either through local training efforts at the Hanoi School of Public Health, or internationally at Mahidol University in Thailand. 54. Evaluation of reach of M&E support: In 2008, the CPMU evaluated the effectiveness of the M&E support provided to provinces. They asked questions about the equipment, staff and capacity of provincial M&E units. Results of the evaluation are provided in Annex 2 and pointed to the focus on provision of equipment and supervision. This resulted in the CPMU scaling up M&E support to some provinces. 55. The four regional Institutes of Public Health provided M&E support for the provinces throughout the Project period. The regional Institutes provided the technical assistance and capacity building to the Project provinces on (a) survey and data use; (b) supervision; (c) reporting system; and (d) best practice in how to undertake IBBSs. 56. The number of HIV-related surveys over the last decade has ballooned, with the HIV Sentinel Surveillance+ (HSS+), as well as provincial specific IBBSs being undertaken. The Project has significantly invested in these IBBSs, and the capacity to undertake them. However, despite the technical assistance provided, these surveys have not always been carried out using comparable sampling methods and have not been done at equal time intervals, hindering comparisons and trend analyses. The revised National Plan on HIV Surveillance in future would help to standardize methods and approaches. 57. Reliability of M&E data: In the reports that have been provided to the WB and during discussions with the WB team, some discrepancies in the condom distribution figures were observed. The implementation review team recommended that a condom routine data quality 15 audit be done to verify the condom distribution figures, but this was not executed. One of the other concerns raised during project appraisal and reiterated during the midterm review, was about the reliability of the self-reported behavioral data: “As quality [of the IBBSs] improves, reported coverage and behavior change may decline: It must be stressed that as the project improves quality and critical analysis, coverage and behavior change results may decline. Such as trend should be encouraged and not seen as a result of faltering project implementation.” 58. Utilization of M&E information: There was evidence during implementation, including field visits to several provinces, that IBBSs results were used to inform the populations being focused on. Evidence for using the D28 form results as a basis for program management and corrections, was scarce. The program routine data, for example, clearly showed that there were some changes in the time that peer educators spent with FSW as opposed to time they spent on marketing condoms, but this data did not lead to a change in how the program was implemented. There were some positive examples of M&E data utilization, with the four regional institutes responsible for IBBSs quality assurance, producing 17 published academic papers to better understand the HIV epidemic in Vietnam (papers attached as Annex 10). 2.4 Safeguard and Fiduciary Compliance 59. Environmental Safeguards. The project triggered OP 4.10 (Indigenous People). Although the project had no adverse impacts on ethnic minority communities, several issues were identified during appraisal that were specific to ethnic minorities and HIV transmission: (a) HIV is prevalent and spreading quickly among injecting drug users and commercial sex workers in border areas; and (b) awareness of these health risks is still limited among some ethnic minority communities. As a result, an EMPF 6 was developed, reviewed and cleared by the WB. The framework provides guidance in the development of provincial/city action plans to ensure that, under the project, ethnic minority groups are informed, consulted and mobilized to participate in its project activities. 60. EMPF review. In 2009, a review of the implementation of the EMPF revealed several weaknesses: a) The quality of the Vietnamese translation of the EMPF was questionable, and it was not properly disseminated in project districts or communes. b) The EMPF was not referenced in the Project Operations Manual, and a representative of the Committee for Ethnic Minority Affairs was not included as a required member of the provincial Steering Committee. c) In addition, in the Government’s guideline for project cost norms for this projects, there was no mention of an ethnic minority position at PPMU level to ensure the implementation of the provisions of the EMPF. As such, there was no staff in the CPMU or PPMU responsible for 6 “Ethnic Minority Framework” means the policy framework dated February 2005, adopted by the Recipient’s Ministry of Health through Decision No. 431/QD-BYT dated February 23, 2005, satisfactory to the Bank, which sets out: (a) the policies and procedures to ensure meaningful consultation with, and the informed participation of, ethnic minorities within the Participating Cities and Participating Provinces who are affected by the Project; and (bi) the principles for the preparation of ethnic minorities development plans as may be required during the implementation of the Project, as said Framework may be revised from time to time with the prior concurrence of the Bank. 16 overseeing that the ethnic minority safeguards were implemented, even in provinces with a high proportion of ethnic minority population, like Lai Chau. d) CPMU and PPMU staff had not been trained on the contents of the EMPF; this is exacerbated by the frequent staff turnover. e) The CPMU had not prepared detailed guideline and template on how to present data concerning ethnic minority issues in PAPs and the D28 routine monitoring form. 61. Addressing these recommendations, the Government took the following remedial actions: a) Reviewed the translation of EMPF to ensure understandable languages were used. b) Added a representative from the Committee for Ethnic Minority Affairs Provincial Committee to Provincial Steering committee in those provinces where minority groups reside. c) Appointed a CPMU staff member to be in charge of supporting and monitoring the implementation of the EMPF. d) Provided training to PPMUs on data collection and reporting using the revised D28 form. Additionally, the Bank included a Social Safeguard Specialist to support implementation of the EMPF in every mission undertaken beginning in 2010. 62. Accomplishments in relation to ethnic minorities. Despite the weaknesses identified, the ethnic minority effort in the Project has achieved positive results. As Annex 2 details, surveillance amongst ethnic population was undertaken, and larger proportions of ethnic minority persons in Vietnam have knowledge of HIV, have positive attitudes towards persons living with HIV, and been tested for HIV. Concomitantly, a reduction in HIV prevalence amongst the ethnic minorities has been observed. Although this cannot be attributed to project outcomes, there were no other significant projects implemented amongst ethnic minority people during the project period. 63. Health Care Waste Management Plan. Given the distribution of injecting equipment, the project design required that a health care waste management plan7 be prepared and implemented. The VAAC, MOH enhanced their guidelines to ensure that there was safe management of healthcare waste generated from HIV/AIDS prevention and control program. The guideline for implementing Harm Reduction Program focused on the collection and treatment of used needles and syringes. It was promulgated by VAAC and disseminated to all the provinces. At the provincial level, PPMUs have staff (M&E staff or project officers) for monitoring and supervising health waste management, and to ensure that funding for health care waste management is included in the annual Provincial Action Plans. Guidance for healthcare waste generated from harm reduction program was integrated into most of the training courses for collaborators and Peer Educators (PEs). The PEs are provided with sufficient equipment for collecting used needles and syringes such as safe boxes and pincers as well as personal protective 7 “Health Care Waste Management Plan” means the plan dated January 2005, adopted by the Recipient’s Ministry of Health through Decision No. 431/QD-BYT dated February 23, 2005, satisfactory to the Bank, which sets out the environmental mitigation measures in respect of the treatment of the health care waste generated at the rehabilitation centers and community-based clinics at the pilot sites under Part A and B of the Project. 17 equipment (uniforms, gloves, shoes, raincoats etc.). Once collected, safe boxes containing used needles and syringes were transported to the commune health stations or higher level health facilities for treatment. Against a service standard of 70%, by the end of the project an estimated 64% of distributed needles and syringes were collected, transported and treated. 64. Financial Management and procurement: The financial management aspect of the Project, which was managed by CPMU at MOH and PPMUs at all provincial implementing agencies (Provincial Departments of Health or Provincial HIV/AIDS Control Centers), has performed satisfactorily throughout the project life cycle and is in full compliance with the WB’s financial management policies and procedures. Procurement progress at central and provincial levels was relatively slow at the project start-up in 2006, and then progressively improved over 2007-2008. During end 2008 – 2009, the project progress was slow again due to high rotation of procurement staff. Procurement rating for this period was downgraded to ‘Moderately Satisfactory’ and then was picked up to “Satisfactory” started from mid-2010. Since then the procurement progress had been almost continuously rated at Satisfactory until project closure. There were no cases of mis- procurement, and 100% of procurement packages had been processed by the end of the project period. 2.5 Post-completion Operation/Next Phase 65. Given the dramatically-different financing situation in Vietnam a decade after the project was first discussed, a follow-on operation is not planned. This is also in line with the new WB Country Partnership Strategy, which has shifted focus away from disease-specific financing so as to support the Government with sustaining economic growth on the country’s path to middle income country status in equitable ways and to supporting universal health coverage through social health insurance and other mechanisms to ensure equity and eradication of extreme poverty in Vietnam. 66. Sustainability of the country’s harm reduction program after the project has ended. This project is not the only one supporting harm reduction activities in Vietnam. In June 2012, detailed discussions, led by the Government, took place with all relevant stakeholders to agree on the sustainability arrangements for the harm reduction activities financed under project. Future harm reduction activities would be financed from three sources: (a) the Government’s own allocations through the National Targeted Program (NTP) (although funding levels are much lower than the World Bank project funding); (b) LifeGap; and (c) through Global Fund allocations. Table 6 in Annex 5 details the transition arrangements, and challenges with these arrangements. 67. Sustaining investments in M&E. The project has significantly invested in M&E, more than any other of the WB’s standalone HIV projects. These investments have yielded benefits beyond the project itself, and will be sustained through a detailed assessment of HIV surveillance in the country, and the development of a harmonized HIV surveillance system. 68. Sustaining the availability and distribution of needles and syringes. The lack of a steady supply and a stable distribution network of clean N&E would have dire consequences for Vietnam’s HIV epidemic. As the transmission efficiency during shared needle use is estimated at 50% -- large increases in sharing of injecting equipment could therefore lead to rapid increases in HIV epidemics, similar to the rates seen in the late 1990s. 69. Future of the VIP Plus socially marketed condoms. When the concept of socially marketing VIP Plus condoms was raised and implemented for the VIP Plus brand in Vietnam, some market research was done. Whereas the condom social marketing program clearly worked (in a condom 18 retail outlet survey, it was shown that the availability of VIP Plus condoms increased dramatically at hotels and guest houses – from being available at 8% of hotels and guest houses in 2012 to being available at 26% of hotels and guest houses in 2013 – see Figure below), it is unclear whether this represents a substitution effect (Figure 5 clearly shows that the availability of other types of condoms at hotels and guest houses over the same period has decreased) or a real increase in total condoms. With over 300 brands of condoms available in Vietnam, of which 6 brands are socially marketed, it is unclear as to whether the total condom market can sustain VIP Plus as a socially marketed brand – irrespective of whether the Government continues to subsidize it or whether the condom social marketing companies for VIP Plus run it. Figure 5. Changes in the availability of condoms at hotels and guest houses in Vietnam 70% 58% 60% 50% 44% 40% 26% 30% 17% 2012 20% 13% 8% 9% 6% 7% 7% 10% 2013 2% 3% 3% 2% 2% 1% 0% Number VIP Plus VIP Hello Happy OK Protector Yes One Plus SM SM FS SM FS SM FS SM SM = Socially marketed; FS = Fully subsidised Source: 2013 PSI Condom Outlet Survey Findings 70. In October 2013, CPMU commissioned a review of the implementation of the transition arrangements (see Annex 11 and Table 2 below): Table 2. Summary of progress with sustainability/transition arrangements as of June 2014 Aspect of Harm Transition / Sustainability Progress as of June 2014 Reduction Program arrangements PE and outreach Peer educators would be 64% of PE for IDU and 71% of PE for FSW transitioned to LifeGap and were transferred Global Fund Needle-syringe Government to take over the With current stock of injecting equipment, programs provision of clean injecting there is sufficient stock for 1 injecting paraphernalia (WB project equipment per IDU per day until June 2014. financed 23 million clean Procurement arrangements after this stock injecting equipment in 2012) has been depleted, had not been made when the ICR review was done. Condoms and condom Government to continue with Government informed the WB team that social marketing condom social marketing of VIP Revenue from condom social marketing will Plus, increasing the prices and be used to purchase OST. thus making it self-sustaining 19 Aspect of Harm Transition / Sustainability Progress as of June 2014 Reduction Program arrangements HCT uptake 29 HIV counselling and testing Transition of 19 GF sites has progress well, sites would be transitioned to the but the transition of the 10 NTP sites has Global Fund (10 sites) and to the been somewhat problematic. NTP (19 sites) Sexually transmitted This would be transitioned to the STI treatment has continued, albeit not infections treatment Government through the mobile clinic model that IDU and FSW preferred Methadone This would be transitioned to the All 7 sites have been transitioned maintenance treatment Global Fund 71. Closure of CPMU and PPMU offices. It was agreed that the CPMU and PPMUs would be dissolved; future coordination functions would continue to rest with VAAC and with PACs. CPMU had implemented specific arrangements for the closure of CPMU and the PPMUs. These steps focused on human resources, asset handling, final project report, saving documents and hand over accounting documents, managing financing after the project ends, counterpart funding, settlement of completion report, project audit, closing the tax code, close the project account, and handling the project seal. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Overall Relevance Rating: Substantial 72. Relevance of objectives from a regional epidemiological perspective: Asia-Pacific is the second-most HIV-affected region in the world. In HIV epidemics in the region, the majority of new infections have occurred amongst most at risk populations, such as IDU and FSW, through sharing of injecting equipment and unprotected sex. Evidence from other countries with concentrated epidemics have shown that a drastic scale–up of comprehensive, harm-reduction- based HIV prevention programs for these populations have strong positive effects on preventing HIV amongst these populations and therefore arresting epidemic spread to the general population (see Wilson et al, 2014). Given that the project objectives focus on scaling up programs for these populations in Vietnam, which has a concentrated epidemic, the project objectives were relevant and remained relevant for the duration of the project: the objectives and design focused on the right populations in the most affected areas with comprehensive interventions. 73. Relevance of objectives from a WB strategy perspective. Given the severity of the increase in HIV preceding project appraisal, the WB Country Assistance Strategies for 2002-2004 and for 2004-2006 both emphasized – as a key development priority – the importance of decisively dealing with the HIV epidemic. The PDO responded to these strategies and to the 2007-2011 Vietnam Country Partnership Strategy (Report No. 38236-VN), which aims for "an increase, in targeted provinces, in the proportion of HIV/AIDS vulnerable groups who use safer injection practices, and reduced stigma and discrimination against people living with HIV/AIDS." 74. Relevance of objectives from a country perspective: When the project was designed, HIV prevalence amongst IDU and FSW had increased dramatically in the five years prior to the mission. With HIV at the time on the increase among tuberculosis patients (3.2%) and military recruits (1.31%), as well as a steady rise of infection among pregnant women to 0.4% in 2002, there was widespread concern that the transmission of HIV infection was poised to move from a concentrated to a generalized level of epidemic, with infection spreading into the general 20 population. The PDO was consistent with both national epidemic trends and with the changes in policy environment8. The PDO was supportive of and helped to showcase the Government’s strong focus on decentralized harm reduction programs. This was the first project, together with DFID assistance from 2003 to 2009, to focus on harm reduction. In a wider social context, reducing the risk of HIV among most at risk populations helped to address related social challenges, including poverty, homelessness, unemployment, and violence. On this basis, the relevance of the PDO is rated high. 75. Relevance of design and implementation from a global and regional perspective. The project focus was consistent with global and regional guidance and best practice about the basic components of harm reduction programs9. In terms of implementation, the project adopted some of the regional implementation lessons, but perhaps not all lessons early enough in project implementation10. 76. Relevance of design and implementation from a country perspective. The project design responded to Vietnam’s key HIV epidemiological considerations and Government implementation challenges11. Project resources were focused geographically on the provinces with the greatest HIV burden and risk. These programs have been implemented effectively through a range of mechanisms including the use of peer-based distribution. Innovative approaches were also employed to improve commodity distribution (IDU, for example, obtained free sterile needle-syringes through fixed boxes at ‘secret’ venues, tea stalls and by redeeming vouchers distributed by peer educators at pharmacies incentivized by the project). FSW sourced condoms through a variety of mechanisms including peer educators and social marketing outlets at guest houses/hotels, cafes and pharmacies and at some entertainment venues. These mechanisms resulted in substantial increases in the coverage of key commodities over the program period and relatively good coverage levels. Project implementation was phased and incremental with strong capacity building elements as new skills were transferred for implementation of harm reduction programs. Methadone maintenance therapy is a good example of this incremental approach: methadone maintenance treatment was piloted and is now nationally available in Vietnam. On this basis, the relevance of the design is rated substantial and the relevance of implementation is rated mostly substantial. Based on the high relevance of objectives, the substantial relevance of design and the mostly substantial relevance of implementation, the overall relevance is substantial. 8 Most prominently the 2006 Law on HIV/AIDS Prevention and Control (no. 64) 9 See Vickerman et al. 2014 for a detailed analysis of the minimum required elements of an FSW harm reduction program and Wilson et al, 2012 for a detailed analysis of the required components of an IDU harm reduction program . 10 See Sgaier et al, 2013, for what was done in India (using micro-planning principles and private sector approaches) to ensure the dramatic increases in coverage of programs for most at risk populations. 11 Government implementation challenges were: (a) lack of M&E data and a national system for HIV/AIDS surveillance; (b) stigma and discrimination being a pervasive issue in all discourse surrounding HIV/AIDS and risk behaviors; and (c) capacity constraints at both national and provincial levels. 21 3.2 Achievement of Project Development Objectives Project Outcome Rating: Satisfactory 77. The project outcome is rated as satisfactory because, as anticipated in the PDO, the government has implemented and sustained core components of a comprehensive harm reduction program for vulnerable populations. The project achieved 2 of the 3 project outcome indicator targets, and substantially achieved the third one. Key populations had access to and used comprehensive harm reduction services and rapidly started to use harm reduction commodities. Harm reduction commodities consisted of the needle and syringe program (NSP) for IDU and condom distribution programs for FSW. Estimated NSP coverage among IDU increased from 22.5% in 2006 to 70.4% in 2011 then declined slightly to 59% in 2012. Estimated condom coverage among FSW increased from 58.2% in 2006 to 89.2% in 2012. Across the 32 project provinces, an estimated 168 000 IDU and 52 000 FSW were reached, accounting for over 80% of the estimated numbers of FSW and PWID in Vietnam. It is estimated that each IDU in the targeted provinces received an average of 152 clean needles and syringes per year. This is considered to be mid-level coverage according to global technical guidelines, but does falls short of the high threshold target of 200 needle-syringes per IDU per year. It is estimated that FSW received an average of 326 free condoms each year under the 100% condom use programs funded by DFID/WB. Condom distribution among FSW achieved the WB target of 240 condoms per year per FSW. 78. In contrast, other components of harm reduction programs for most at risk populations were not implemented to the same levels, and most intermediate outcome indicators were only partially achieved (see Annex 2 for details). Although surveillance, financed by the project, elucidated the emergence of MSM as an epidemiological priority, the project did not achieve its target of distributing 468 000 lubricant sachets to MSM. 79. Because of the successful scaling up of the core components of a harm reduction program, the HIV epidemics amongst IDU and FSW have substantially declined in Vietnam. HIV prevalence among IDU at the national level declined from 21.3% in 2003 to 9.6% in 2012 and among FSW declined from 3.7% to 2.6%. Project-financed activities have resulted in new HIV infections having been averted: during the period 2003-2012, an estimated 33,054 HIV infections, 924 HIV-related deaths and 17,392 Disability Adjusted Life Years were averted because of the joint WB/DFID-financed activities. If the project had not been implemented, significant increases in HIV incidence and prevalence amongst IDU and FSW would likely have been observed. The greatest increase would have been among IDU and FSW with anticipated elevated prevalence of 18.1% and 3.4% in 2012, respectively (see Figure 6). Figure 6. Anticipated HIV infections amongst FW and IDU with and without the intervention Source: UNSW report, 2013 22 80. These evaluation results (that both programs have averted new infections, but that IDU programs have been more successful than FSW programs) triangulate well with a 2010 observational. This analysis showed that there was a stronger correlation between project provinces achieving IDU coverage targets and having declining HIV prevalence amongst IDU, than was the case for FSW. See Annex 13 for details. 81. Comparison of project outcomes to the original economic and financial analysis estimates. The project averted more HIV infections than anticipated at appraisal (estimated 33,000 new HIV infections averted with IDA grant financing and 2 rounds of AF, compared to the anticipated 22,000 HIV infections averted with IDA grant financing): this is mainly due to the AF available and thus the increase in the number of FSW and IDU reached with the programs. The cost per infection averted is estimated to be US $1,007 (for both NSP and FSW programs), which is slightly less expensive than the ex-ante estimate of an US$1,285 per HIV infection averted. This difference is mainly as a result of the same impacts being achieved with lower-than-expected service coverage levels. 82. Potential contribution of other development partners to these outcomes. Most project provinces relied exclusively on this project for harm reduction financing. Other HIV programs implemented in these provinces funded by other international sources such as PEPFAR and GF had different interventions that targeted different populations. Because the project did not have substantial overlaps with other domestic and international programs12 and because there was good complementarity between different organizations investing in HIV responses in Vietnam, the outcomes described here can reasonably be attributed to the project. 83. Institutional, system and individual capacity building. In terms of capacity building, the project has also achieved measurable outcomes: a) Systems and institutional capacity building: With project funds, the national HIV M&E system was created and strengthened. Further, the project-specific system of annual provincial (decentralized) planning for province-relevant HIV activities was not just applied in the project, but extended to be used as the accepted national system for decentralized Government planning of the country’s HIV response, thus more broadly contributing to the country’s decentralization of health care delivery goals. b) Individual capacity building: During the project’s lifespan, 56 government staff was sent for overseas training and 27 for in-country training; these 83 staff greatly contributed to epidemiological analyses, annual provincial action plan design and implementation. In an independent evaluation of the capacity building efforts that were part of the WB/DFID efforts, “service providers reported that training offered by the project had increased organizational capacity and resulted in a more skilled and professional prevention workforce. They reported acquiring new skills through project training activities in planning, project management and coordination, report writing and monitoring and evaluation.” Through the assessment, it was clear that the training increased the ability of Government 12 This was possible because of close collaboration between CPMU, PPMUs, MOH and development partners. Based on a matrix developed when planning HIV/AIDS provincial prevention activities, provincial authorities were able to clearly identify activities funded by various agencies. This coordinating mechanism, created and facilitated by CPMU, provided them with insights into existing program coverage and prevented duplication of services. 23 staff to perform well in their jobs and therefore the increased ability of PPMUs and CPMU staff (and units) to perform well. c) Advocacy with provincial level leaders: High level advocacy undertaken consistently by CPMU also ensure that perceptions amongst Government officials about HIV and PLHWA changed. Anecdotal evidence shared during the field visits suggested to the ICR review team that the project helped to change the perception that HIV was a ‘Social Evil’. 84. Results in terms of ethnic minorities: In terms of ethnic minorities, the project has also reported positive trends in HIV knowledge, attitudes, uptake of HIV testing and concomitant reductions in HIV prevalence. Whereas attribution to the Project could not be established, no other development partners (or the Government) mentioned other HIV activities for ethnic minorities during the ICR processes. 85. Contribution to national-level HIV strategy goals. These project-specific achievements have contributed to the continued achievement of the Government of Vietnam’s HIV strategy, which is to ensure that HIV prevalence in the general population remains below 0.3%. 3.3 Efficiency Overall Efficiency Rating: Moderately Satisfactory Technical Efficiency of Services Delivered 86. Comparative unit costs of outputs delivered. It is estimated that an annual per-capita investment of US$ 25.40 per IDU was made (with resultant 152 injecting equipment per IDU distributed). The cost of reaching FSW with 326 condoms per year, was US$34.50. 87. Comparison of unit costs of commodities to Global Fund procurement average unit costs shows that commodities bought under the project were more expensive than average GF prices . As Table 4 in Annex 3 points out, the unit costs of two of the most common commodities procured by the project – condoms and HIV test kits – is higher than the commensurate average unit costs from GF grants in the region (Vietnam unit costs for condoms $0.069/GF $0.03; Vietnam unit costs for HIV test kits: $2.90/GF $1.30). It might have been technically efficient to investigate pooled procurement options with the local Global Fund grant in Vietnam. 88. High set-up and project management costs have reduced technical efficiency, but it helped to provide substantial knowledge and implementation know-how: An independent evaluation in 2012 revealed that the general perception was that project was generally well-managed by the CPMU and PPMUs. The program management and administration costs amounted to US$ 24.8 million, accounting for 28% of the total project cost. Over half (55.7%) of the total US$ 89.5 million joint DFID and WB investment (from 2003 to 2012) was spent on indirect costs (all costs not directly intended for service delivery) including program management, maintenance, M&E and capacity building. This represents a substantial investment and is could be considered to be excessive. However, given the investments that the Government made in setting up a harm reduction program, in building capacity for it, in financing all provincial units to manage the grants, and in M&E, including 31 IBBSs, some of these investments can be justified. However, when the PACs were created in 2010 (at the same time as the AF was approved), it could have been technically efficient to amalgamate these units and create one provincial AIDS structure – this possibility was not investigated at the time and the status quo in terms of implementation arrangements was maintained during 2010. 24 89. Isolated program interruptions took place in 2005 and 2013 that might have reduced implementation efficiency: Occasional discontinuities in the transfer of funding from central to provincial to district levels took place. This resulted in delays in paying the peer educator networks, which might have interrupted service delivery. Also, the supply of condoms in 2013 arrived 8 months late, causing peer educators not to have any condoms to sell through their social marketing from January to October 2013. Allocative Efficiency of Objectives Achieved – Cost Effectiveness of Interventions 90. The project component focusing on IDU was cost-effective in the short-term and programs for FSW may yield long-term benefits. a) The project component focusing on IDU was deemed to have been cost-effective. Overall, program costs amounted to an estimated US$1,007 per HIV infection averted during the period 2003-2102. The costs for averting one HIV-related death and one Disability Adjusted Life Year were US$36,020 and US$1,914 respectively. The life-time cost per Disability Adjusted Life Year averted was estimated to be US$522. According to standard willingness to pay thresholds, these values indicate that the programs are good value for money. For every dollar spent on NSP, the estimated rate of return in healthcare costs saved (not required to be expended for treating infections) was US$1.93. Thus, every dollar spent was returned and provided close to an additional US dollar not required to be spent in the future. The cost of averting one Disability Adjusted Life Year was US$917. If the life-time impacts of NSP are considered, the return-on-investment ratio was 42.82 and the cost required for one Disability Adjusted Life Year aversion was US$270. In comparison to NSP in other international contexts, NSP implemented by the project in Vietnam were typical. The estimated required cost of US$486 to avert one new infection through NSP is comparable to findings reported in developing country settings, such as China (US $560-810) and Belarus (US$359, $234-1,054). Notably, this cost amount is much lower than the amount in developed country settings (typical cost of US$3,000-20,000 per infection averted). b) The epidemic benefits derived from condom distribution programs for FSW were unlikely to have been cost-effective during the implementation period (2003-2012). However, when the life-time impacts of the condom distribution program are considered, the return-on- investment ratio was 4.53 and the cost required to avert one Disability Adjusted Life Year was US$425, demonstrating good long-term cost-effectiveness. The STI prevention program also demonstrated moderate cost-effectiveness in the long-term. Behavior change communication was deemed not to be cost-effective and the decision to drop this activity was an economically-efficient decision. 3.4 Justification of Overall Outcome Rating Rating: Moderately Satisfactory 91. The project was rated moderately satisfactory despite it having achieved two of its three outcomes and having substantially achieved the third outcome. Reasons are as follows: a) On the positive side, the PDO remained relevant throughout the project. A substantial number of HIV infections have been averted because of the project. Positive outcomes were also observed amongst ethnic minorities, and capacity building efforts have yielded some positive results in terms of provincial and national government officials’ ability to design and 25 implement programs. Staff training provided by the project was perceived – in focus group discussions with training participants – as having facilitated effective program design and implementation. The project design and implementation also remained substantially relevant throughout implementation. b) But, all project components and subcomponents were not efficiently implemented: the FSW harm reduction program was not as effective as the IDU harm reduction program, and project commodities were more expensive to purchase than the regional norms. Also, the EMPF was not fully implemented and environmental safeguard standards were not met (fewer needles and syringes returned than the target of 70%). 3.5 Overarching Themes, Other Outcomes and Impacts (a) Institutional Change/Strengthening 92. As articulated in the PDO, the project aimed to build capacity at national, provincial and local levels to design, plan and implement harm reduction programs. An independent qualitative assessment indicated that the project has significantly increased organizational capacity and human resources ability: 60 staff from provinces, regional institutions and VAAC received master degrees in public health at Mahidol University and Hanoi Public Health University. Health workers appreciated new skills they obtained from attending training activities and on-the job application of these skills in planning, project management and coordination, report writing and M&E. (b) Other Unintended Outcomes and Impacts (positive or negative) 93. Provincial planning model expanded to other provinces: The project was significant in demonstrating the value of harm reduction to the Vietnamese government and contributed to changes in the legal and policy frameworks for harm reduction interventions in Vietnam. This is exemplified through recent laws which enable HIV/AIDS prevention and ongoing implementation of harm reduction activities (specifically 64/2006/QH11 and associated decree 108/2007 ND-CP). This provincial planning process is now replicated in all 63 provinces in Vietnam, showing the longer-term institutional impact of the project. 94. Demonstrated that harm reduction can work. The project has contributed to alleviating stigma and discrimination against key affected populations and PLHWA. Harm reduction interventions delivered by the programs were well-aligned with intensive advocacy activities in Vietnam during the past 10 years, challenging the use of the ‘social evils’ approach to drug use and sex work. The project established credibility with key stakeholders at all levels of the government and health system. Collaboration with local communities and authorities was essential to successful program implementation. 95. Private-public partnership potential demonstrated. The social marketing of condoms, albeit somewhat controversial in terms of its longer term sustainability or its need from a total condom marketing point of view, has contributed to the realization of what can be done in public-private partnerships. As a result of this effort, some peer educators have set up their own small business such as café shop (where they can also sell condoms) and condoms shops. 96. Catalytic effect of project efforts: Based in part on the successful experience of the small Methadone Maintenance Therapy pilot sites that were financed through the project, the national 26 Methadone Maintenance Therapy program was initiated by the Government. It has been expanded to 11 provinces and treats more than 6,900 people in 41 clinics, with an adherence rate of 96%. The Government has planned to continue to expand this service to 245 clinics in 30 provinces and 80,000 patients by 2015. There are also plans to offer Methadone Maintenance Therapy in prisons. This type of treatment (Methadone maintenance therapy) has broader positive impacts too: it facilitates the reintegration of IDU into society by enabling them to bypass the negative physical and psychological consequences of withdrawal. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops 97. Although this was not a learning ICR, the ICR team recommended a series of lessons learned workshops because of the sheer volume of stakeholders involved and the duration of the project (8 years of implementation). Three workshops were held in October 2013 (see Annex 5 for more detailed information about the outputs of the stakeholders’ workshops). The following key messages were communicated about the project during these workshops: a) Behavior Change Communication: There was a general view that BCC activities have led to reductions in stigma and discrimination against those with HIV. Lessons were also learned in terms of how the implement BCC activities, the different types of media that existed, and involving the vulnerable groups in designing appropriate messages. b) Harm reduction services & activities: Harm reduction network of peer educators was established and provincial officers gained knowledge in how to maintain such a network over time. It was shown that approaches different to punitive and rehabilitative measures to curb HIV prevention amongst IDU can and does work. c) M&E: The importance of regular reporting, feedback and the IBBSs were listed as key lessons learned. d) Capacity building & training: After training, it was felt that trainees needed to commit to working in the field for at least one year, and not to seek alternative (often higher paying) employment as a result of having undergone the training course and professional development. e) Management & coordination: Better coordination between PPMU and PAC was a necessary lesson for future implementation, as well as the time that it takes to have procurement plans approved. 4. Assessment of Risk to Development Outcome Rating: Substantial 98. The project has generated both short-term and long-term health and financial benefits among populations of IDU and FSW in Vietnam. Sustaining these harm reduction programs is expected to have accumulative population impacts on the HIV epidemic in Vietnam. There are substantial risks to sustaining the project’s impact: a) Financial risk to sustainability: Vietnam’s current HIV response is highly dependent on foreign aid. With the gradual withdraw of foreign investment in the health sector, domestic allocations from the Vietnamese government to sustain the harm reduction program will be increasingly important to cover the gap. 27 b) Programmatic risk to sustainability. If the harm reduction HIV prevention programs are not sustained into the future then there could be a significant increase in the number of new infections by 2020 (4,698 extra infections), mostly attributable to IDU (4,061), FSW (59) and their clients (327). c) The HIV epidemic amongst MSM is increasing, but substantive harm reduction efforts for them are not yet in place. HIV prevalence data in MSM in Vietnam are scarce as sentinel surveillance for MSM was only established in 2009 (with financing from the project). Based on limited data, the estimated HIV prevalence in MSM increased from 4.1% in 2003 to 6.5% in 2012 (Figure 7). The rising trend is consistent with other findings. The Vietnam’s MOH reported that HIV prevalence among MSM increased markedly from 5.3% to 16.7% in HCMC and from 9.4% to 17.4% in Hanoi between 2006 and 2009. Additionally, a recent review indicated increasing HIV prevalence among MSM in urban Vietnam. HIV prevalence among the general Vietnamese population remained at ~0.3%. This trend is expected to be maintained until 2020, with HIV prevalence in FSW and IDU stabilizing at low (1.3% and 6.2% respectively) levels. HIV epidemics attributable to heterosexual transmission and sharing of injection equipment are decreasing, whereas new infections attributable to homosexual transmission are increasing (Figure 7). By 2020, up to two-thirds of new infections could be attributable to sex between men. Yet, sufficient programs for these populations are not in place. Figure 7. Estimated sources of new HIV infections in Vietnam, 2000-2020 Source: UNSW report, 2013 d) Risks in continuing the ethnic minority program. The continuation of the ethnic minority program was not included in the transition arrangements, and the staff for implementing the EMPF was not in place for the duration of the project. Therefore, the sustainability of this particular project component beyond the life of the project is questionable. 28 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Bank Performance in ensuring Quality at Entry Rating: Satisfactory 99. The WB took all necessary steps to ensure the best possible project design – relevant epidemiological and implementation data were analyzed, implementation arrangements were clarified upfront, lessons from previous projects were taken into account, discussions with development partners took place to ensure that overall implementation was harmonized with their efforts, and investments in M&E and capacity were made. (b) Quality of Supervision Quality of Supervision Rating: Satisfactory 100. Supervision took place at least once a year (13 ISRs were recorded during the project’s 8 year life span). Recommendations and actions from ISRs were detailed, clear and followed-up on. Having the co-Task Team Leader (who took over later as Task Team Leader) in-country for the entire duration of the project ensured consistency in approach and institutional knowledge that would be hard to have achieved with a Task Team Leader not locally resident. Complaints about delays with providing No Objections was not recorded, and the WB also provided high quality technical support throughout project implementation. Coordination with other donors was good: they were involved in transition arrangements and although implementation processes and payment structures (for peer educators) differed, implementation was coordinated, where feasible. 101. Although the ICR review team felt that the WB task team could have improved on the oversight of implementing recommendations to change the EMPF weaknesses, the design of the condom social marketing strategy and could have made more of an effort to better understand the Total Condom Market and the sustainability of VIP Plus beyond the project itself, and that sustainability of the harm reduction network of peer educators and harm reduction commodities for IDU could have been improved, overall performance was consistent and as such, is rated satisfactory. (c) Justification of Rating for Overall Bank Performance Overall Bank Performance Rating: Satisfactory 102. Justification of WB rating: Overall WB performance is rated Satisfactory. The WB team clearly identified a development challenge in the health sector, sought to understand it better, was not afraid to either take risks (trying out innovative approaches such as condom social marketing and community-based continuous harm reduction models), be a trail blazer (together with DFID for the harm reduction approaches in Vietnam) or change direction if new evidence so required, coordinated and collaborated with other development partners, and sought to support the Government to ensure that Project objectives were achieved. 29 5.2 Borrower Performance (a) Government Performance Government Rating: Satisfactory 103. Justification of Government rating: the Government of Vietnam at national level has shown the political will to implement the country’s first comprehensive harm reduction program, a first for the East Asia region at the time of its implementation. Further, the national Government has also embraced local level planning processes and local solutions to implementation challenges. As a result, stigma was greatly reduced and HIV (and the people living with it) were more accepted in society).Where data pointed to weaknesses, the Government has changed course in the project, as the various course directions have illustrated. Where capacity was lacking, the Government invested in capacity building particularly at lower levels of Government. . At provincial level, commitment was varied. Some were very committed, whilst there was some resistance in a few project provinces to implement activities for these vulnerable populations. 104. What remain of concern, are the sustainability arrangements in terms of continued supply of funds for implementation of the program and continued supply of commodities that form the backbone of the harm reduction program. The extent to which the harm reduction program for IDU, for FSW, and for the ethnic minority groups will be maintained now that the project has ended. The IDU harm reduction program is particularly vulnerable due to uncertainty about the future consistent supply of injecting equipment and the maintenance of the supply chain network through the peer educators. (b) Implementing Agency or Agencies Performance Overall Implementing Agency Rating: Moderately Satisfactory 105. Justification of Implementing Agency Rating at national level: This rating is moderately satisfactory because the CPMU was a committed and skilled group of persons who demonstrated participatory leadership. Flexible participatory annual planning processes, satisfactory financial management and procurement management ratings, and moderately satisfactory social safeguards and environmental safeguard ratings justifies this rating. 106. At provincial level, the performance of PPMUs was varied. Some performed exceptionally well, whilst others needed more capacity building to sustain activities and to change direction to accommodate harm reduction approaches. Focus group participants, for example, reported interruptions in needle-syringe distribution due to interruptions in funding which resulted in problems with commodity procurement and staff attrition. 107. Flexible and participatory annual planning processes: Through this funding, the project developed an efficient and flexible management framework. The project established a wide collaborative network with all levels of government bodies and research institutions in Vietnam, as well as with international health organizations. The project management was flexible and allowed annual project planning to be adjusted to suit actual local needs. This was identified as one of the project’s strengths. While overseen by VAAC at the central level, project provinces were allowed to design their own approaches and activities for harm reduction programs that were feasible and responsive to local needs. 30 Overall Financial Management Rating: Satisfactory 108. Financial Management arrangements: The Project financial management arrangements, as well as the Project implementation arrangements, were complicated and decentralized by design. While having CPMU at VAAC as the focal point for Project budget approval, financial reporting and audit, all other financial management areas (such as planning and budgeting, contract and expenditures management, expenditures approval and accounting records maintenance) were decentralized to 20 provinces (which increased to 32 provinces in 2008 when DFID pooled more funds). CPMU managed the Project designated account opened at a commercial bank. Each PPMU also opened a bank account in local currency to receive the Project funds transferred by CPMU based on the approved annual operational plan. Report on expenditures between PPMUs and CPMU was done monthly online. Interim financial reports were submitted by CPMU to the Bank quarterly. Financial audits by an independent firm were conducted annually and audit reports were submitted to the Bank before end of June of the following year in accordance with the Financing Agreement. 109. Financial Management ratings over the project period: The Project financial management performance improved from Moderately Satisfactory in late 2006, 2007 and 2008 to Satisfactory in 2009, 2010, 2011 and 2012 and 2013. The areas that had significant improvement are: plannng and budgeting; financial reporting at provincial and district levels; maintenance of accounting records and assets management at provincial and district levels. Because of the high decentralization of Project activities, to districts and even community levels, and the high number of participating provinces and implementing agencies, the financial management performance was variable among provinces. CPMU took a lot of effort in training to PPMUs and districts, including co-operators and peer educators, enforcing the detailed financial management manual, and regular supervision activities at field level. At the end of the Project, all provinces have satisfactory performance. Overall Procurement Rating: Satisfactory 110. Procurement Arrangements and Execution: Procurement progress at central and provincial levels was relatively slow at the project start-up in 2006, and then was progressively improved over the next two years 2007-2008. Between the end of 2008 and 2009, the progress was slow again due to high rotation of procurement staff. Procurement rating for such period was downgraded to ‘Moderately Satisfactory’ and then was raised to “Satisfactory” from mid- 2010. Since then the procurement progress had been almost continuously rated at Satisfactory until project closure. Key procurement activities included purchase of condoms (via ICB and UN agencies), vehicles (via UN agency), syringes and needles, consumable medical material, office and IEC equipment, medical equipment, training, social marketing services, various consulting services with firms and individuals, etc. Large packages were procured under the ICB procedure by the CPMU at central level, included procurement via UN agencies for the purchase of vehicles and large quantities of condoms. At provincial level, most packages were procured under the NCB and Shopping procedure for Goods, while CQS and IC are dominant method for selection of consultants. The two-year extension for the project was also based on the project satisfactory performance that included one year extension (2012) for AF from IDA Grant and another year extension (2013) for AF from DFID TF Grant. 111. Procurement Challenges: A solid project management structure was designed with the intention of implementing procurement arrangements. As part of broader project management, procurement was also negatively affected by (a) high rotation of procurement staff at central level, and (b) human resource capacity constraints at local levels. Further procurement problems 31 experienced included the geographic challenges of project areas making transportation difficult, delays to execution of contracts, and late delivery of goods to districts and communes, all of which negatively affected activity implementation. Despite these challenges, by the project closing date, no planned procurement activity was dropped. Therefore, there was 100% completion of all procurement packages. Overall Environmental Safeguards Rating: Moderately Satisfactory 112. Environmental Safeguard Arrangements: The VAAC, MOH enhanced guidelines and trained staff in safe management of healthcare waste generated from HIV/AIDS prevention and control program. The guideline for implementing Harm Reduction Program which covered content of collection and treatment of used injecting equipment was promulgated by VAAC and disseminated to all the provinces. Guidance for healthcare waste generated from harm reduction program was integrated into almost all of the training courses for collaborators and Peer Educators. At the provincial level, PPMUs have staff (M&E staff or project officers) for monitoring and supervising health waste management to assure that this activity will get adequate concerns and budget integrated with Annual Provincial Action Plans. The PEs are provided with sufficient equipment for collecting used needles and syringes such as safe boxes and pincers as well as personal protective equipment (uniforms, gloves, shoes, raincoats etc.). Once collected, safe boxes containing used injecting equipment were transported to the commune health statins or higher level health facilities for treatment. About 70% of distributed needles and syringes in 20 old provinces were collected, transported and treated (meeting a service standard of 70%); however, this percentage in the 12 newly included provinces is lower. In general, nearly 64% of distributed needles and syringes were collected, transported and treated. Overall Social Safeguards Rating: Moderately Satisfactory 113. Social Safeguard Arrangements: The project triggered OP 4.10 Indigenous People. Although the project has no adverse impacts on ethnic minority communities, several issues required particular attention during the project implementation phase, including: (a) HIV is prevalent and spreading quickly among injecting drug users and commercial sex workers in border towns and cross border areas; (b) awareness of these health risks is still limited among some ethnic minority communities. During the project preparation, an EMPF was developed, reviewed and cleared by the WB. The framework provides guidance in the development of provincial/city action plans to ensure that under the project ethnic minority groups are informed, consulted and mobilized to participate in project activities. 114. Justification for Social Safeguard Rating: Despite the implementation challenges pointed out in the EMPF evaluation in 2009, proposed activities for ethnic minority communities were implemented. Several BCC efforts have been specifically designed for ethnic minority people including harm reduction interventions, HIV counseling and testing, and IEC materials. Implementing agencies paid particular attention to ensure that proposed activities are culturally appropriate such as: (a) using ethnic minority IDU as peer educators; (b) communication message in ethnic minority languages; and (c) mobilizing chief village or elder as influencer and communicator. This component achieved measurable results and increases in HIV knowledge, positive attitudes, HIV testing and reductions in HIV prevalence were observed over the project period, although attribution to the WB efforts have not been established. The Government has noted that it views the ethnic minority aspects of this project as good practice. But, it is unclear to the ICR team how the EMPF implementation challenges noted in 2009 were addressed during project implementation. This raised some concerns over the sustainability of this effort, given the lack of staff to oversee EMPF implementation. Therefore, the rating is Moderately Satisfactory. 32 (c) Justification of Rating for Overall Borrower Performance Overall Borrower Rating: Moderately Satisfactory 115. Throughout the project, the Government has shown that it had the political will, the implementation wherewithal and the courage of their convictions to implement the project and make changes to project implementation. Taking the step towards embracing harm reduction approaches is brave, and the Government of Vietnam has embraced it. The Government has also used the national and provincial organs of state to implement the program successfully by achieving most of the project outcomes. 116. However, the environmental safeguard standards and the social safeguard standards were not met, and as a result, the overall implementing agency rating is moderately satisfactory. 6. Lessons Learned 117. What key lessons should Vietnam learn about the implementation of its HIV/AIDS harm reduction/prevention activities going forward? a) Current models of needle-syringe distribution programs delivery are not innovative or adaptive enough. The almost exclusive reliance on the peer educator distribution model – albeit necessary – is unnecessarily narrow and restrictive in scope. New models, such as integrated methadone maintenance therapy-NSP service delivery, the use of social media and treatment as prevention for FSW, should be considered b) Secret spots/hidden locations as a strategy for NSP were viewed by service providers and IDU as complementary and part of a range of strategies designed to maximize access and availability of needle-syringes. However, the close collaboration of local communities and authorities was instrumental to the success of this model. c) The pharmacy sector in Vietnam has a potentially important role in the distribution of sterile injecting equipment. There are a number of models internationally and include incentivizing pharmacies to distribute needle-syringes to IDU which should be explored as a means of increasing needle-syringe coverage for IDU. d) Decisions about future condom social marketing should be made with the total condom market in mind. e) Ongoing focus on STI treatment access for FSW and MSM are necessary. Examination and testing collaborations between government health staff, project peer workers and venue owners may be an effective means of increasing uptake of STI screening by venue-based FSW. The development of future STI initiatives requires an assessment of the effectiveness of collaborations with the private sector to screen and treat both venue and street-based FSW. 118. What should Development Partners and Vietnam learn about implementing Development Assistance operations going forward? Lessons were learned in terms of procurement, financial management and transition arrangements. a) Late approval of annual procurement plans at both central and provincial levels delayed procurement implementation. It is strongly recommended that a procurement plan for the subsequent period be well prepared and approved by relevant authorities (MOH at central 33 level or PPC/DOH at provincial level) by the end of the preceding period, so that the planning for procurement activities can be started early in the next period. b) High turnover of staff in charge of procurement, as well as other key/leading staff at implementing units, weakens capacity of the affected implementing agencies, and should be avoided. c) Project Implementation Units as duplicate structures should be avoided. Tendency to establish a new implementation unit for the new project at both MOH and provincial level may dilute the pool of staff/experts experienced in project implementation. d) Decentralization of appraisal/approval authorities should be made clear before actual project implementation, so that to minimize delays in the Borrower’s internal appraisal/approval process. e) Decentralization of activities to provinces and districts helped to increase their capacity in Project management and also their ownership in Project activities implementation. The human resource left by the Project after it closes is a very good resource for the Government to implement similar activities. The planning and budgeting of HIV/AIDS prevention activities were local-based rather than central-based, which made the plan more realistic and the implementation more efficient. f) With such high decentralization and high number of implementing agencies, a detailed Financial Management manual and its enforcement are critical for Project funds to be used for intended purposes efficiently and economically. g) The Project’s approach to get the Ministry of Finance’s approval on expenditures cost norms higher than regular cost norms applied for all ODAs and Government budget has both pro and cons effect. The Project staff, consultants and beneficiaries have more incentives to conduct Project activities with high payments. Performance, staff continuity and transparency were maintained at high level. The favorable cost norms however created a discrepancy among projects in the health sector, among donor-funded projects and even among WB- funded projects. This is also not a sustainable approach for the Government to improve performance when the Project is closed and the Government has very limited budget which can afford similarly high cost norms. h) Finally, role of CPMU in financial management particularly, and in Project implementation generally, is essential to the success of the Project. i) Transition arrangements can be made and are possible: The project invested US$24.8 million for setting up an efficient management framework for project implementation at both national and local levels in the past decade. During this process, the project has actively involved the Vietnamese government and relevant HIV prevention bodies to ensure a transferrable and sustainable management model after the withdrawal of the programs. 119. What should other countries and Development Partners consider for developing and implementing large scale Harms Reduction / Prevention programs? (a) HIV prevention interventions need to be aligned (and re-aligned) with changing epidemic trends. The delay in instituting HIV surveillance among MSM meant that there was no evidence on whether there was a need for MSM-focused HIV prevention programs or where they should be implemented. 34 (b) A supportive policy environment is not only essential, but rapid changes are possible if it is implemented well. The experiences of more than a decade of HIV prevention programs in Vietnam indicate that a supportive legal and policy environment is essential for the effective implementation of harm reduction programs. (c) To make harm reduction work, a sustainable commodity distribution models and channels are needed. The project was the biggest suppliers of condoms for FSW and clean needle- syringes for IDU and other preventive services across all regions of Vietnam, contributing to the large reduction in risk behaviors and HIV prevalence in FSW and IDU. The project set up the first innovative models for commodity distribution through multiple channels including social marketing, peer-educators, fixed-boxes at hotspots/venues, and the pharmacy sector. (d) Social marketing may increase condom distribution to FSW. The targeting of venues where sex work occurs was prioritized at the provincial level as a way to increase condom availability. The development of programs which increase condom use in an environment where sex work remains illegal has required multi-sectoral collaborations which have been driven by PACs. (e) But, condom promotion programs may not work as well as is commonly accepted. Consistent with numerous other international settings, NSP in Vietnam have been shown to be cost- effective. However, condom distribution programs for FSW have only shown moderate impacts on high-risk sexual behaviors and HIV prevalence. There is not strong evidence of epidemiological impact associated with condom distribution programs. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies 120. The Government commended the WB for a detailed report and thanked the WB for excellent support provided during supervision. It also reiterated the importance of the timing of the project coinciding with the Government’s new strategy, and the important contributions of the project to the policy advocacy, especially policies related to implementation and expansion of harm reduction intervention program. It requested the WB to add information about Government commitment to harm reduction, and to mention that the decentralized planning instituted by the project, is now being used nationally. It also requested that the WB document the extent of capacity building in the document, and that it makes it clear that the project financed the development of the national M&E system, which is still in place. (b) Co-financiers 121. DFID noted that overall the project has been well delivered and achieved the original set targets. DFID co-financed the WB project since early 2010 as a follow-up to the successfully completed project on HIV prevention funded by DFID during 2005-2009. DFID finds the cooperation with WB highly effective, complementary and take full advantage of both agencies’ comparative advantages. The coordination has reduced transaction costs for both the Government and financing agencies. The WB team has been flexible in adjusting and adapting to new epidemic trends as well as partner’s priorities. The adjustment in 2010 in project design has been an example. The Task Team Leader has been highly approachable, consultative and effective in dealing with complex situations to lead the project to a successful completion. This project can be showcased as a good example of effective cooperation between development partners. 35 (c) Other partners and stakeholders 122. Two issues that were particularly raised during ICR discussions with partners were related to the commodities procured under the project – ensuring the future availability of injecting equipment, and the future of the VIP brand condoms. 123. The notion of using a separate PPMU and CPMU structure (separate from VAAC and the PACs) also came under scrutiny, as it was felt that these separate structures did not always foster collaboration, integration and efficiency of service delivery. 36 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Project Cost by (USD million) Original Additional Additional Revised Appraisal Percentage Project Project Projects Costs Total estimate of Costs Costs financed by Project (as per Appraisal financed by financed by additional funds Cost PAD) Component IDA Grant DFID in DFID Trust (Board Trust Fund Fund Grant Approval in Grant (January 2013) March (March D= (E) (D as a 2005) 2010) (A + B + percentage (C) C) of E) (A) (B) 1. Provincial 21.60 13.26 5.79 40.65 21.616 188% Implementation of HIV/AIDS Action Plans (as Sub-Grant to be distributed for implementation at provincial level) 2. National 7.22 4.45 2.03 13.70 8.20 167% HIV/AIDS Policy and Program 2.1 Policy and Program 2.86 4.45 1.45 8.76 Development and Implementation 2.2 Monitoring and 2.10 0 0.58 2.68 Evaluation 2.3 Behavior 2.26 0 0 2.26 Change Communication Focused on Reducing Stigma and Discrimination 3. Project 5.18 6.33 1.56 13.07 5.184 252% Management Total Baseline 34.00 24.04 9.38 67.42 35.00 193% Costs Contingencies 1.00 0.9 0 1.9 0 Total Project Costs 35.00 24.94 9.38 69.32 35.00 [Exchange rate of DFID Trust Fund 2010: 1 GBP = 1.45 US$] [Exchange rate with AF 2013 as of 29 October 2012: 1 GBP = 1.6079 US$] 37 (b) Financing Appraisal Actual/Latest Type of Co- Estimate Estimate Percentage of Source of Funds financing (USD (USD Appraisal millions) millions) Borrower 3.50 3.50 IDA GRANT FOR HIV/AIDS 35.00 35.00 100% Additional financing from DFID TF Grant 24.94 24.94 100% Additional financing from DFID TF Grant 9.38 9.38 100% 38 Annex 2. Outputs by Component 1. Component 1: Support for Provincial HIV Action Plans Support for the provincial HIV action plans focused on the following activities: harm reduction activities for FSW and IDU; behavior change efforts; mobile STI services (mostly for FSW); care and support for PLHWA; HCT services (mostly for FSW and IDU); and project management, and M&E. The harm reduction programs for both FSW and IDU followed a similar peer outreach, peer education and peer distribution network system: peer educators were recruited and paid a stipend to both provide outreach services to their peers (behavior change communication), to encourage safer behavior, to support and link the target populations to health services where needed, and to distribute commodities (clean injecting equipment for IDU and condoms for FSW). IDU peer educators were also asked to collect used needles and syringes by placing disposal containers and picking up used needles and syringes. The role of FSW peer educators was extended as of 2012 to also market and distributed socially-marketed condoms, in addition to distributing free condoms. The basic measure of adequate program coverage used in this analysis is whether commodity distribution exceeded 300 condoms per FSW per year and whether it exceeded 200 needles/syringes per IDU per year (keeping in mind that with heroin, the most used drug in Vietnam, and one has to inject at least once a day). The achievement of provinces in terms of the implementation of the Provincial Action Plans is described below. In looking at these data, three facts need to be borne in mind: (a) The data up to 2009 only includes the 20 project provinces, whereas the data from 2010 onwards include the outputs achieved in all 32 project provinces. Therefore, in most cases, the graphs reflect dramatic increases between 2009 and 2010; (b) Also, the data are from the Government’s database, are were not consistently reported on in a routine or standardized manner in the periodic Aide Memoires or ISRs. These data were never data audited, so their veracity could not be independently confirmed during the ICR review; and (c) Data for implementation in 2013 was not available and therefore all data presented here, is only up to December 2012. 1.1 Harm Reduction Program for FSW 1.1.1 Geographic coverage of the program, and temporal changes in the number of program clients (FSW) and program service providers (FSW peer educators) The figures below illustrate temporal trends in terms of the number of service clients (FSW) and service providers (peer educators), as well as the geographic distribution of project activities. These figures confirm that there was some variation, but that in general, project activities were implemented in the majority of communes in project provinces (Figure A). It also shows that the number of peer educators and FSW in project provinces rapidly increased, as expected, over the duration of the project and stabilized through the end of the project (Figure B). In line with the data in Figure B, the number of peer educators per 1,000 FSW steadily increased and stabilized towards the end of the project with a slight downward trajectory, partially due to the lack of effort to recruit new peer educators into the program (Figure C). The distribution of the number of FSW 39 peer educators per FSW was narrow, with most provinces having either 20 or 30 peer educators per 1,000 FSW (Figure D). Figure A: Percentage of FSW in communes where project-funded activities took place 120% 93% 96% 100% 90% 84% 80% 74% 65% 60% 40% 34% 24% 20% 0% TOTAL Region 1 Region 2 Region 3 Region 4 Region 5 Region 6 Region 7 Figure B: Temporal trends in the estimated # of FSWs and # of FSW peer educators 90,000 1,200 # of FSW 80,000 Number of FSW peer educators 1,000 70,000 Number of FSW 60,000 800 # of peer educators for 50,000 600 FSW group 40,000 30,000 400 20,000 200 10,000 - 0 2006 2007 2008 2009 2010 2011 2012 40 Figure C: Temporal trends in the number of peer educators per 1,000 FSW 18 16 14 12 10 8 6 4 2 0 2006 2007 2008 2009 2010 2011 2012 Figure D: Frequency distribution of number of peer educators per 1,000 FSW in 2012 12 10 Number 0f provinces 8 6 4 2 0 Number of peer educator per 1,000 FSW 1.1.2 HIV prevention commodities (condoms) distributed to FSW Condoms were distributed throughout the program to FSW: up to 2010, it was distributed for free and from 2011 onwards mainly through condom social marketing. Figure E shows temporal trends in condoms distributed for free, either to FSW (the majority of them), and to a much lesser extent to IDU and others (such as MSM). Figure F shows how rapidly socially marketed condom distribution increased towards the end of the project, whereas free distribution, as expected, dramatically decreased. In Figure G, the average number of condoms distributed per year is illustrated, and compared against the benchmark of 300 condoms per FSW per year. It shows that since the condom social marketing campaign was introduced, the number of condoms dramatically increased in the last 2 years of the project, and that condom distribution reached the benchmark of 300 condoms per FSW per year in 3 of the project years (2008, 2009 and 2011). 41 Figure H provides an overview of the frequency distribution of condoms per FSW across all the provinces in 2012. It shows that in 2012, most provinces distributed 150 condoms per FSW per year, that only 2 provinces achieved the benchmark of 300 condoms per FSW per year, and that 8 of the 32 province provinces achieved close to that – by distributing either 200 or 250 condoms per FSW per year. Figure E: Temporal trands in the Number of Condoms Distributed to FSW and other populations for free 25 Millions 20.1 20 17.7 16.8 Condom for others 15 12.5 13.1 Condom for IDU 9.7 Condom for FSW 10 8.4 6.0 5 0 2005 2006 2007 2008 2009 2010 2011 2012 Figure F: Trends in Condom Distribution to FSW 18,000,000 16,000,000 # of VIP Plus condom sold through social marketing 14,000,000 program Number of Condoms 12,000,000 # of VIP Plus condom sold via 10,000,000 Traditional Outlet (like Pharmacy) 8,000,000 # of VIP Plus condom sold via 6,000,000 Non Traditional Outlet (NTO) 4,000,000 2,000,000 # of condoms distributed to FSW for free 0 2006 2007 2008 2009 2010 2011 2012 42 Figure G: Temporal trends in the number of condoms distributed per FSW per year 1200.0 1000.0 Condoms per FSW in the 800.0 project provinces Condoms per FSW in the 600.0 communes where project was implemented 400.0 Threshold for condom distribution to FSW (300 per FSW per year) 200.0 0.0 2005 2006 2007 2008 2009 2010 2011 2012 Figure H: Frequency distribution of annual number of condoms distributed to FSW, 2012 12 10 Benchmark of 300 Number of province condoms per FSW 8 per year 6 4 2 0 50 150 200 250 300 350 400 450 500 More Number of condoms distributed to each FSW 1.1.3 HCT services provided to FSW Figure I shows the number and percentage of FSW reached with the program that underwent HCT financed by the project. It clearly shows that uptake was slow, low and never reached higher than 24%. It also showed that in 2011, when the HCT centers were handed over to the GF and PAC as part of the project transition arrangements, the project ceased to finance HCT sessions for FSW. 43 Figure I: Number and percentage of FSW reached by the project, who had an HIV test financed by the project 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2005 2006 2007 2008 2009 2010 2011 2012 FSW that have been reached 44617 45153 45659 46163 46,633 47382 47810 48244 FSW undergoing VCT 0 1,284 4,860 3,799 10,899 14,136 16,172 0 1.1.4 Summary: Harm reduction program for FSW As the data in sections 1.1.1 to 1.1.4 corroborate, the main focus of the harm reduction program for FSW was on condom distribution and information exchange either through IEC materials or counselling from peers. This was also revealed in the project’s midterm review: the Government’s own analysis confirmed that the bulk of FSW accessed condoms and IEC materials (see Figure J). 44 Figure J: Summary of kinds of harm reduction services that FSW reached by the project received Receiving condoms 80 69.8% 60 HCT 40 IEC materials 32.9% 20 60.9% Counseling from 0 Health workers STI treatment 36.5% 38% Counseling from 56.8% peer educators Source: Government Midterm Review Report, 2008 1.2 Harm Reduction Program for IDU 1.2.1 Geographic coverage of the program As Figures K and L show, the geographic coverage of the IDU harm reduction program increased significantly over the life of the project, helping to achieve the Government’s objective of harm reduction programs for vulnerable populations being rolled out at decentralized levels. The rapid change between 2009 and 2010 is because of the addition of 12 provinces to the project after DFID co-financing. Figure M confirms that the harm reduction program for IDU was implemented in the majority of communes in project provinces, with the exception of project provinces Thai Nguyen, Yen Bai and Ho Chi Minh City. 45 Figure K: Temporal changes in the number of districts with harm reduction interventions for IDUs 250 200 150 100 50 - 2006 2007 2008 2009 2010 2011 2012 Figure L: Temporal changes in the number of commumes implementing interventions among IDU 2,500 2,000 1,500 1,000 500 - 2006 2007 2008 2009 2010 2011 2012 46 Figure M: Percentage of FSW in communes where project-funded activities took place 100% 90% 90% 86% 79% 80% 69% 70% 60% 49% 50% 42% 40% 34% 30% 20% 14% 10% 0% TOTAL Region 1 Region 2 Region 3 Region 4 Region 5 Region 6 Region 7 1.2.2. Temporal changes in the number of program clients (IDU) and program service providers (IDU peer educators) Figure N below confirms that the number of IDU supported and the number of IDU peer educators recruited steadily increased throughout the project life cycle. There is also confirmation (Figure O) that the majority of IDU were supported by the project and that as the project recruited more IDU PEs, the number of IDU that needed to be served by one IDU PE reduced threefold from over almost 340 IDU per IDU PE to just under 120 IDU per IDU PE. Figure O: % of estimated IDU in project provinces covered with project's IDU harm reduction program 100% 90% 90% 86% 79% 80% 69% 70% % of IDUs 60% 50% 42% 40% 34% 30% 20% 14% 10% 0% Region 1 Region 2 Region 3 Region 4 Region 5 Region 6 Region 7 47 Figure P: Changes in number of IDUs per PE and IDU PEs per 1000 IDU in project provinces 400 12 339 350 10 300 287 IDU PE per 1000 IDUs 227 8 IDUs per IDU PE 250 Number of IDUs per IDU peer 200 6 educator (PE) 140 115 119 Number of IDU per 1000 IDU 150 102 4 peer educators (IDU PEs) 100 2 50 - 0 2006 2007 2008 2009 2010 2011 2012 1.2.3 Commodities (injecting equipment) distributed to IDU for free As Figure Q shows, the number of clean needles and syringes distributed rapidly increased in the early years of the project, as peer educators were identified, trained and the injecting equipment procurement was finalized. Figure R shows that from 2008 to 2012, annually more than the threshold (benchmark) of 200 needles and syringes were distributed per IDU reached by the program (keeping in mind that not all IDU in a project province were reached by the program – see Figure O). Figure S confirms this trend by showing that, in 2012 in several provinces, the number of needles and syringes distributed exceeded the threshold of 200 needles and syringes per year per IDU. Figure Q: Number of Needles and Syringes distributed throughout the project 35,000,000 Number of neeldes distributed 30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 2005 2006 2007 2008 2009 2010 2011 2012 48 Figure R: Temporal Trends in Needles and Syringes Distributed per IDU per Year by the HIV Project in Vietnam 400.0 Neeldes and Syringes per IDU per Year 350.0 N&S per IDU across country 300.0 N&S Per IDU in project 250.0 communes 200.0 150.0 100.0 50.0 0.0 2005 2006 2007 2008 2009 2010 2011 2012 Figure S: Frequency distribution of injecting equipment provided to each IDU last year, 2012 6 Bench mark of 200 N&ES per IDU per year 5 Nunmber of province 4 3 2 1 0 50 150 200 250 300 350 400 450 500 More # of injecting equipment distributed per IDU per year, in 2012 1.2.4 HCT services provided to IDU Figure T shows the number and percentage of IDU reached with the project that underwent HCT financed by the project. It clearly shows that uptake was slow, low and never reached higher than 13%. It also shows that in 2011, when the HCT centers were handed over to the GF and PAC as part of the project transition arrangements, the project ceased to finance HCT sessions for IDU. 49 Figure T: Number and percentage of IDUs reached by the project, who had an HIV test financed by the project 100% 80% 60% 40% 20% 0% 2005 2006 2007 2008 2009 2010 2011 2012 IDUs reached by the project 77843 78777 79662 80589 81508 84213 85080 85956 IDU undergoing VCT funded by the 0 1,812 6,627 10,490 13,640 16,189 17,263 0 project 1.3 Summary of HCT provided by the project Figure U provides a snapshot of the HCT sessions financed by the project. As the Figure shows, the number of HCT sessions steadily increased by the project and focused primarily on getting FSW and IDU tested. But, HCT uptake amongst IDU and FSW was low throughout the project, as Figures I and S earlier in this Annex illustrate. HCT site support ended in 2011, when the 29 HCT sites financed by the project were handed over to the Global Fund (10 sites) and to the Government-financed NTP, run by the PACs (19 sites). In 2013, an evaluation of the handed- over sites showed that the GF-run sites were more successful than the other sites – they were better stocked, trained and understood the HCT curriculum and service delivery system better than the HCT sites run by the NTP. Figure U: Summary of HCT services financed by the project 50 43.9 45 36.7 40 31.7 35 26.0 Thousands VCT at 05-06 Centres 30 23.5 Other persons undergoing VCT 25 FSW undergoing VCT 20 IDU undergoing VCT funded by the project 15 8.1 10 5 0.0 0.0 0 2005 2006 2007 2008 2009 2010 2011 2012 50 1.4 STI treatment provided STI treatment in the form of mobile clinics was provided until 2011 (Table below summarizes the progress), when a decision was made by the Government for all STI treatment financed by the project to be handed over to the Global fund project (ST treatment in 8 provinces) and to the NTP (STI treatment in 18 provinces). Also, training was done for 100 private clinics in 10 provinces to provide STI services for men, particularly focusing on MSM. Results from these 100 clinics suggest that stigma and discrimination amongst MSM is high. In 2012, of the 5843 patients who were seen in these clinics, 2,547 had STIs, but only 119 self- identified as MSM. # of IDU STI # of FSW STI cases # of other STI cases Total Number of STI cases treated treated treated cases treated 2006 786 2,838 4,037 7,661 2007 1,118 10,138 4,051 15,307 2008 450 7,789 3,802 12,041 2009 1,050 20,583 8,352 29,985 2010 1,122 27,535 6,353 35,010 2011 30,706 905 4,423 36,034 Total 35,232 69,788 31,018 136,038 Source: From CPMU database 1.5 Regional level M&E and PPMU management Supervision and M&E support by Regional Institutes, PPMUs and District Collaborators: At regional level, M&E support was provided through the four regional Institutes of Public Health). These Institutes carried out meetings amongst PPMU staff, undertook field visits to PPMUs and to district collaborators, and held meetings with collaborators and with peer educators, and, in a cascade way, PPMUs visited district collaborators who, in turn, visit peer educators. Results from their work in 2012 are as follows: Type of supervision Activity 2006 2007 2008 2009 2010 2011 2012 Reviewing meeting at provincial level 273 425 392 514 552 581 447 (reported by PPMUs) Field supervision of PPMUs (by 1,028 2,321 2,490 2,317 2,691 3,581 3,246 regional institutes) Supervision of activities of 3,661 8,465 11,826 12,739 20,749 24,187 22,215 collaborators (by PPMU staff) Review meetings between 2,968 4,420 7,689 14,010 17,285 10,881 9,276 collaborators and PEs Source: From CPMU database. Note: Data after 2011 not available from CPMU 1.6 Capacity building at PPMU and lower levels Several short courses by PPMU staff for peer educators and district collaborators took place annually (details available in table below). In addition, long-term public health training was 51 offered to PPMU staff: either in Hanoi at the Hanoi School of Public Health (45 students) or at Mahidol University in Thailand (27 students). All staff members, once they completed their training, were obliged to return to their posts for a minimum of two years. 2006 2007 2008 2009 2010 2011 TOTAL Number of courses for collaborators 72 54 40 42 71 32 311 Total number of trainees 1,379 1,538 1,220 1,496 2,128 978 8,739 Of which, number of women 316 658 315 477 872 336 2,974 Of which number of ethnic minority people NM NM NM NM 221 82 303 Number of courses for peer educators 107 51 51 104 92 91 496 Total number of trainees 2,638 1,449 1,514 1,382 2,760 2,310 12,053 Of which, number of women 436 480 345 463 1,297 887 3,908 Of which number of ethnic minority people NM NM NM NM 300 203 503 Number of training courses for health workers 174 77 48 34 38 45 416 Total number of trainees 2,558 1,813 1,355 1,058 1,114 1,310 9,208 of which, number of women 833 652 566 359 353 412 3,175 Of which number of ethnic minority people NM NM NM NM 1 97 98 Number of other courses 52 58 97 30 43 60 340 Total number of trainees 1,599 2,214 3,855 1,059 1,364 2,554 12,645 of which, number of women 398 527 737 345 389 343 2,739 Of which number of ethnic minority people NM NM NM NM 200 589 789 Note: (1) NM = Not Measured; (2) Data after 2011 not available from CPMU. 2. Component 2: National level support Subcomponent 2.1: Policy and Program Development and Implementation When designed, this sub-component was meant to finance Policy and Program Development, and Implementation, which would strengthen capacity at national and provincial levels and promote the development of innovative, effective prevention and treatment approaches and models among vulnerable groups through: (a) demonstration sites that would explore a community-based treatment model to integrate harm reduction with comprehensive HIV treatment and care; (b) policy studies and research; (c) knowledge sharing and training; and (d) other innovations. (a) Demonstration Sites for Integrating Harm Reduction with Treatment and Care at the 05-06 Centers13. The original Care and Treatment subcomponent was designed to promote the development of innovative, effective prevention and treatment approaches among vulnerable groups through integrated demonstration sites. After piloting period in three sites, a number of new changes called for project restructuring. The first was a decrease in the number of detainees; the second was a reduction in the number of sites under the auspices of 13 05 centers are government-run rehabilitation centers for FSW and 06 centers are government-run rehabilitation centers for IDU. 52 the project from three to one and the third was a hand-over of activities to the Global Fund, who would continue the ARV program within the scope of a community-based approach. (b) Policy Research and Studies. This subcomponent has provided support for a number of initiatives including, but not limited to, support for the development of the Law on HIV/AIDS Prevention which was adopted by the National Assembly in 2006. In 2007, the project continued to finance a number of policy workshops to develop regulations. The decree No 108/2007/ND-CP was issued to provide guidance on how to implement the Law. The development of a number of important technical guidelines has also been supported, including the following: (i) HIV/AIDS Prevention Law (ii) Decree 108/2007/ND-CP on guiding implementation of the HIV Prevention Law (iii) Direction No 32/2008/CT-TTg permitting the implementation of harm reduction activities (iv) Decision 5073/QD-BYT approving the Methadone maintenance therapy therapy pilots Source: UNSW report, 2013 (c) Knowledge Sharing and Training. Under this subcomponent a national scientific conference was organized in 2005, and again in 2010 and in 2012. Support has also been provided for high ranking government officers at both central and provincial levels to attend international conferences on HIV/AIDS Prevention and also conferences on Harm Reduction. Study tours have also taken place in Thailand, China and Australia for participants to learn more about methadone programs. In the last two years, national workshops have been organized for peer educators to allow for an opportunity to exchange lessons. Extensive training has taken place, including for doctors, nurses and medical staff for 05-06 centers, district health centers/dermatology hospital, as well as two training courses for medical staff about the diagnosis, care and treatment of PLWHA and ARV treatment initiation. A total of 16 nurses were also trained by National Institute of Infectious and Tropical Diseases to equip them in how to care for PLWHA. (d) Innovation Support. The first Innovation Day was successfully organized in 2007 to coincide with World AIDS Day. The theme of the event was "Keep promise to stop AIDS", and it focused on (i) effective intervention models using BCC approach to prevent HIV/AIDS transmission; (ii) care and support for PLWHAs, reducing stigma and discrimination; and (iii) commitment of the policy makers and social organizations to fight against the HIV/AIDS epidemic. More than 800 proposals were received from the general public, PLWHA, professionals, private sector, mass organization, NGOs and schools, of which 670 proposals were submitted by students from the secondary and high schools around the country. The jury 53 board awarded 31 proposals, of which 21 focused on effective intervention models, 4 proposals were on reducing stigma and discrimination, 2 proposals were on treatment models, and 4 proposals were on care and support for PLWHAs. A second Innovation Day was organized in 2009, during which 29 innovation projects were awarded. (e) Addition to the project: pilot sites for methadone maintenance therapy: In 2010, the project also commenced the financing of pilot methadone maintenance therapy sites. This was only possible in 2010 because at the time the project was designed, the environment was not yet conducive for supporting drug substitution, however, the PAD indicated “should drug substitution become part of the program, a separate facility might be funded for the provision of methadone maintenance therapy, in order to separate the two client populations.” In early 2008 the Ministry finally approved the Decision to begin a pilot on methadone treatment. The mission re-affirmed the principle that a key criterion for selection of a province remains its demonstrated political commitment to implementation of an Methadone maintenance therapy program, as demonstrated by an (i) official request from provinces; (ii) official approval from the Provincial People’s Committee and the People’s Council; and (iii) official approval from MOH. Good progress was made throughout the project with this piloting, with 11 pilot sites having been established by the end of the project, and 1,927 of the planned 2,000 persons having been on Methadone maintenance therapy therapy by the end of the project. Sub-component 2.2: M&E Not only was the national HIV M&E framework developed, but the project indicators were aligned to that framework, as Table 3 summarizes. Also, 11 sets of technical guidelines for HIV M&E and program implementation were issued. 54 Table 3. Changes to Intermediate Results indicators of the Vietnam HIV Prevention Project in 2010 and in 2013 Indicators agreed to during negotiations Indicators in 2010 during first Level II Indicators in 2013 during second Level II restructuring and additional (2005) restructuring and additional financing financing (2013) (2010) Percent of vulnerable groups in participating Percentage of vulnerable groups in Percent of vulnerable groups in participating provinces reporting safer provinces reporting safer injection practices participating provinces reporting safer injection practices (from 20% -70%), as measured by these indicators in the (from an estimated 20% at baseline to 70% at injection practice (from an estimated 20% at Government of Vietnam’s national HIV M&E system: (a) Percentage of project end) baseline to 70% at project end) injecting drug users who constantly use clean syringes and needles in the last month Percent of vulnerable groups in participating Percent of vulnerable groups in Percent of vulnerable groups in participating provinces reporting condom use provinces reporting condom use in sexual participating provinces reporting condom in sexual intercourse (from 40% -80%), as measured by these indicators in intercourse (from 40% to 80%) use in sexual intercourse (from an estimated the Government of Vietnam’s national HIV M&E system: 40% at baseline to 80% at project end) a) Percentage of female sex workers reporting condom use with their most recent casual clients b) Percentage of female sex workers reporting condom use with their most recent regular clients An increased percentage of vulnerable groups An increased percentage of vulnerable An increased percentage of vulnerable groups (from an estimated 10% at (from an estimate 10% at baseline to 80% by groups (from an estimate 10% at baseline to baseline to 80% by project end) are reached by provincial services, as project end) are reached by provincial 80% by project end) are reached by measured by these indicators in the Government of Vietnam’s national HIV services provincial services M&E system: a) % of female sex workers receiving condom in the last 6 months b) % of female sex workers receiving IEC materials in the last 6 months% of female sex workers receiving introduction of STI treatment in the last 6 months c) % of female sex workers receiving information from peer educators in the last 6 months d) % of female sex workers receiving HIV testing in the last 6 months e) % of IDU receiving clean needles and syringes in the last 6 months f) % of IDU receiving IEC materials in the last 6 months g) % of IDU receiving information from PEs in the last 6 months h) % of IDU receiving condom in the last 6 months 55 Indicators agreed to during negotiations Indicators in 2010 during first Level II Indicators in 2013 during second Level II restructuring and additional (2005) restructuring and additional financing financing (2013) (2010) i) % of IDU receiving HIV testing in the last 6 months j) Number of lubricant sachets distributed to MSM in the last 6 months (480 000 distributed by the end of 2013) [this was a new indicator added at the time] An increased percentage (for an estimated An increased percentage (for an estimate Increased percentage (from an estimated 10% at baseline to 90% at project 10% at baseline to 90% at project end) of 10% at baseline to 90% at project end) of end) of provinces prepare high quality provincial plans provinces prepare high quality provincial provinces prepare high quality provincial plans plans An increased percentage of provinces (from 0 An increased percentage of provinces (from An increased percentage of provinces (from 0 to 90% at project end) to 90% at project end) establish adequate 0 to 90% at project end) establish adequate establish adequate M&E systems M&E systems M&E systems An increased percentage of provinces (from An increased percentage of provinces (from An increased percentage of provinces (from an estimated 10% at baseline to an estimated 10% at baseline to 80% at an estimated 10% at baseline to 80% at 80% at project end) meet targets specified in provincial plans project end) meet targets specified in project end) meet targets specified in provincial plans provincial plans Health providers in pilot receive appropriate Dropped Dropped training and supervision in comprehensive treatment, care and support Clients in pilot receive appropriate Dropped Dropped comprehensive treatment, care and support, including integrated harm reduction activities National and regional research institutions' National and regional research institutions' Dropped strengthened, through establishment of AIDS strengthened, through establishment of M&E departments and recruitment and AIDS M&E departments and recruitment training of AIDS M&E staff and training of AIDS M&E staff Mechanisms developed to regularly Mechanisms developed to regularly Dropped disseminate M&E results and to use data for disseminate M&E results and to use data for program improvement program improvement 56 Indicators agreed to during negotiations Indicators in 2010 during first Level II Indicators in 2013 during second Level II restructuring and additional (2005) restructuring and additional financing financing (2013) (2010) Health workers and communities report Dropped Dropped reduced stigma and discrimination towards PLWA, IDU, SW and communities Increased percentage offering methadone Increased number of provinces offering Methadone services (from an services (from an estimated 0 at baseline to estimated 0 at baseline to 6 by project end) 6 by project end) [this is a new indicator added at the time] Increased number of injecting drug users Increased number of injecting drug users enrolled in Methadone services enrolled in methadone services (from an (from an estimated 0 at baseline to 2000 by project end) estimated o at baseline to 2,000 by project end) [this is a new indicator added at the time] 57 Subcomponent 2.3: Behavior Change Communication for and Amongst PLHIV to Reduce Stigma and Discrimination. Up to 2008, a total of 895 BCC initiatives to reduce stigma and discrimination were implemented. These BCC efforts used a variety of communication channels, including television, radio, newspapers, publication and conferences/workshops/talk shows, with the latter being the most popular method. In 2008, during the project’s midterm review, this project component was dropped as a national level activity for the following reasons stated in the Midterm review report: (a) lack of performance; (b) delays in commencing with the sub components; and (c) procurement challenges. As an alternative, provinces were asked to include BCC activities in their provincial action plans. In addition, in 2010, the CPMU added the establishment of a social marketing network for the VIP brand of condoms (‘VIP Plus’) to this sub component. This subcomponent was reasonably implemented, although the ICR review found that (a) not all funds from the provinces for the socially-marketed condoms already distributed, had been paid over to the CPMU (see table below, from the Oct 2013 progress report); and (b) the socially-marketed condoms were expensive and not self- sustaining. 58 Annex 3. Economic and Financial Analysis Financial and economic analysis data were obtained from several sources: (a) DFID Evaluation of a Decade of DFID/WB HIV Support for Vietnam; (b) Global Fund pooled procurement report; (c) global study data on the unit costs of delivering HIV prevention services for female sex workers; (d) Bill and Melinda Gates Foundation study on the costs of delivering HCT and ART services; (e) unit cost data from different provinces provided by PPMU; (f) relevant HIV epidemic data from the HIV Landscape in Asia report; and (g) Vietnam HIV country progress report. These data were compared with the original economic and financial analysis information summarized in the PAD in relation to four key analysis areas, as follows: I. Project Relation to Vietnamese Development Context The project remained relevant throughout its eight year implementation cycle – from both an epidemiological perspective as well as the Government prioritization perspective. In terms of the HIV epidemic trends in the region and in Vietnam, it is clear that the disease burden in East Asia is still growing (Figure 3A), that Vietnam shoulders a significant burden of disease in the region (Figure 3B), and that while HIV is not the leading cause of disease burden in Vietnam, it does contribute to almost 5% of years of life lost’ the third largest cause of years of life lost after strokes and road traffic accidents, (see Figure 3C). Figure 3A. HIV disease burden and epidemic trends in HIV in the Asia and Pacific region, 2000 to 2020 59 Figure 3B. Estimated share of PLHIV in East Asia and Pacific that are in Vietnam, 2012 Figure 3C. Years of Life Lost in Vietnam, 1990 to 2010 In terms of the Government’s response to HIV: When the project was designed, the Government had just approved its national HIV strategy (for the period 2004 to 2010). Therefore, the project was designed, project activities developed, and project budget determined based on what would 60 be required to support the Government to achieve its strategic objective of ensuring that HIV prevalence remained at 0.3%. This would be done by implementing – through the project and other sources of funds, including Government funds – the Government’s new direction and commitment to a harm reduction approach (as opposed to a punitive approach) to addressing HIV in Vietnam. This commitment was followed through by actual laws aimed at strengthening harm reduction approaches to HIV prevention implementation passed in 2006 and 2007. These laws, specifically 64/2006/QH11 and associated decree 108/2007 ND-CP, reinforced the implementation of the Government’s commitment to a harm reduction approach to HIV prevention. The further approval and implementation of the national HIV M&E framework in 2007 and subsequent Decision 28 to standardize routine reporting on HIV activities, as well as the Government’s new HIV strategy development in 2011, all reinforced the Government’s ongoing commitment to responding to HIV with evidence-based approaches. As further evidence of the sustained relevance of the project in terms of helping to address Vietnam’s development challenges and find evidence-based solutions for it, it is worthwhile to mention that the DFID/WB programs were significant in demonstrating the value of harm reduction to the Vietnamese government and contributed to changes in the legal and policy frameworks for harm reduction interventions in Vietnam. Examples include: (a) DFID/WB programs have contributed to alleviating stigma and discrimination against key affected populations and people living with HIV (PLHIV). Harm reduction interventions delivered by the programs were well-aligned with intensive advocacy activities in Vietnam during the past 10 years, challenging the use of t he ‘social evils’ approach to drug use and sex work. The project established credibility with key stakeholders at all levels of the government and the health system. Collaboration with local communities and authorities was essential to successful program implementation. (b) DFID/WB programs significantly increased organizational capacity and human resource skills. In particular, the project was the first to build the capacity of staff working in HIV prevention using a harm reduction and evidence-based approach. Health workers appreciated new skills they obtained from attending training activities and on-the job application of these skills in planning, project management and coordination, report writing, and M&E. From these three sets of information – continued epidemiological relevance of HIV at population level and in the region, continued Government commitment, and the project’s successes in demonstrating evidence-based approaches, and building capacity for its implementation-- it is clear that responding to HIV in Vietnam is an ongoing development priority in Vietnam, that the project was appropriate at the time and that the project has contributed to significant changes in policy and strengthened implementation. Whilst currently being a success story in terms of the ongoing relevance of the project objectives for Vietnam’s development context, it is also clear that reductions in financing or lack of future continuation of Vietnam’s harm reduction program (amongst IDU, MSM and FSW) could reverse some of the gains made. This project is thus not only important, but its future sustainability is of continued relevance to the overall development objectives that Vietnam is striving to achieve. II. Justification for Government Financing Preventing injecting drug use transmission of HIV – with its 50% risk of transmission with shared injecting equipment use – is one of the most significant public health priorities in countries 61 without harm reduction programs, with significant HIV disease burdens, and with significant IDU populations who share injecting equipment. Government financing at the time of appraisal was justified by the public health importance of HIV programs. Government financing was deemed to be important as neither the private sector nor civil society organizations could assume this responsibility. Ongoing Government financing also remains important as the potential for increases remain (Figure 3D). Figure 3D: Categorization of countries in Asia-Pacific by the trend in their HIV prevalence from 2000 to 2013 III. Economic Analysis A detailed qualitative and quantitative assessment, with extensive political economy, mathematical, epidemiological and economic analysis and modelling, was carried out in 2013 to determine the impact of the WB/DFID financing for harm reduction programs in Vietnam. The key results of this assessment are summarized below; as relevant, comparisons to the results anticipated when the economic analysis in 2004 was carried out are made to assess the veracity of the economic and financial assessment at the time of the project. Conclusion 1: The Vietnam HIV Prevention Program was allocatively efficient DFID/WB HIV prevention programs have been appropriated designed and the resources available have focused geographically on the provinces with the greatest HIV burden or numbers of people at greatest risk of HIV, namely, IDU and FSW. Historically, DFID/WB programs have largely targeted IDU and FSW and have been implemented in 32 Vietnamese provinces, covering all 5 geographical regions and the 2 major Vietnamese cities, Hanoi and HCMC. The 32 project provinces include an estimated 167,541 IDU and 51,844 FSW, accounting for over 80% of the estimated numbers of FSW and IDU in Vietnam. Conclusion 2: The Vietnam HIV Prevention Project was partially technically efficient Although program coverage levels during implementation were not as high as anticipated during appraisal (see Annex 2), program coverage was sufficient to achieve the anticipated HIV infections averted during the program period. It is estimated that IDU in the targeted DFID/WB provinces received an average of 152 clean needle/syringes per year from the NSP funded by DFID/WB, at an annual per-capita investment of US$25.40 per IDU. This is considered to be mid-level coverage according to technical guidelines of the World Health Organization, United Nations Office on Drugs and Crime, and Joint United Nations Program on HIV/AIDS and although very good compared to most countries in the region, and around the world, it falls short 62 of the high threshold target of 200 needle-syringes per IDU per year. It is estimated that FSW received an average of 326 free condoms each year under the 100% condom use programs funded by DFID/WB, at an annual per capita investment on FSW of US$34.50. Condom distribution among FSW achieved the WB target of 240 condoms per year per FSW. The unit costs of reaching each IDU or FSW, at these levels of commodity distribution, are deemed to be acceptable when benchmarked against costs of programs in other settings, although commodity costs were higher than the average costs (see Table 4 below). Table 4: Median and inter-quartile costs (2005-2012) for commodities by Global Fund region GF procurement cost GF procurement cost median values: median values: Condoms HIV test kits (range in brackets) (range in brackets) Eastern Europe / Central Asia $0.07 (.05-.15) $1.50 (.86-3.5) Latin America / Caribbean $0.04 (.03-.06) $1.30 (.95-1.8) North Africa / Middle East $0.02 (.02-.03) $0.94 (.80-1.8) South Asia $0.02 (.02-.03) $0.23 (.22-.80) SSA: East Africa $0.03 (.03-.03) $1.30 (.90-1.9) SSA: Southern Africa $0.04 (.03-.05) $0.84 (.76-1.0) SSA: West and Central Africa $0.02 (.01-.04) $1.40 (.83-2.3) East Asia / Pacific $0.03 (02-.04) $1.30 (.80-1.6) Unit costs for this Project in $ 0.069 $2.90 Vietnam Source: Regional comparative data from AIDSpan Report on Global Fund Commodity Procurement Costs; Vietnam data from project information provided by Government Conclusion 3: Innovative program implementation helped to increase the technical efficiency of the HIV prevention project implementation The DFID/WB funded activities consisted primarily of harm reduction programs. These programs have been implemented effectively through a range of mechanisms including the use of peer-based distribution relying on peer educators. Innovative approaches were also employed to improve commodity distribution. IDU also obtained free sterile needle-syringes through fixed boxes at ‘secret’ venues, tea stalls and by redeeming vouchers distributed by peer ed ucators at pharmacies incentivized by the project. FSW sourced condoms through a variety of mechanisms including peer educators and social marketing outlets at guest houses/hotels, cafes and pharmacies and at some entertainment venues. These mechanisms resulted in substantial increases in the coverage of key commodities over the program period and relatively good coverage levels. Estimated NSP coverage among IDU increased from 22.5% in 2006 to 70.4% in 2011 then declined to 59% in 2012. Estimated condom coverage among FSW increased from 58.2% in 2006 to 89.2% in 2012. Conclusion 4: The anticipated decreases in HIV disease burden before the program and the estimated actual decreases in HIV disease burden (after the project) are comparable Our of the 32 project provinces, 26 showed a declining trend in HIV prevalence among IDU, 6 had stabilized prevalence, and, importantly, none experienced an increasing trend. NSP appeared to be more effective than condom programs in averting new infections. In 8 out of 32 project 63 provinces, there was an increasing trend in HIV prevalence among FSW, a decreasing trend in 16 provinces and stable prevalence in the remaining 8 provinces. Overall, the ex poste analysis showed that the DFID/WB funded programs in Vietnam were estimated to have had a significant impact on preventing HIV infections. If DFID/WB programs had not been implemented, significant increases in HIV incidence and prevalence would likely have been observed. Through modelling, it was estimated that in the absence of DFID/WB funded harm reduction programs, HIV prevalence would have increased ~18.1% among IDU and ~3.4% among FSW by 2012. We estimated that between 2003 and 2012 the DFID/WB programs have reduced the disease burden of HIV/AIDS by preventing ~33,000 HIV infections, 924 HIV- related deaths, and 17,392 disability adjusted life years (Disability Adjusted Life Years). The vast majority of these health benefits were attributed to NSP for IDU. NSP alone averted an estimated 31,000 infections, 872 HIV-related deaths and 16,395 Disability Adjusted Life Years during the period 2003-2012. Condom distribution programs for FSW averted an estimated 1,585 infections, 42 HIV-related deaths, 788 Disability Adjusted Life Years. Further, the benefits derived to date will accrue to greater benefits in the longer term. These results are comparable to the ex ante estimates that the original project (IDA grant of US$35 million) would avert approximately 22,000 new infections (range from 10,555 to 30,449). After the project ended, it was estimated that with the program coverage levels (% of vulnerable populations reached with services) realized during the program, keeping the expanded geographic scope in mind with the addition of the 12 more project provinces in 2010 supported through DFID additional financing, the project averted 33,000 new HIV infections. Conclusion 5: The cost per infection averted was less expensive than anticipated, and the benefit-cost ratio larger than anticipated DFID/WB funded programs, especially NSP for IDU, were deemed to have been cost-effective. Overall, program costs amounted to an estimated US$1,007 per HIV infection averted during the period 2003-2012 for the IDU and FSW programmes. The estimate of US$1007 per infection averted in the ex poste analysis is slightly less expensive than the ex ante estimate of US$1,285 per HIV infection averted. The reason is that the program achieved comparable impacts to what was anticipated, even though program coverage levels were lower during project implementation than what was initially anticipated (although it has not affected the program impact, in general – supporting Vickerman et al’s analysis that reaching more sex workers with fewer interventions increases impacts the quickest: expanding geographic scope is more important than expanding program intensity for persons already reached with the programme). The estimated required cost of averting a new infection through the NSP is less expensive -- approximately US$486 to avert one new infection through NSP. For every dollar spent on NSP, the estimated rate of return in healthcare costs saved (not required to be expended for treating infections) was US$1.93. Thus, every dollar spent was returned and provided close to an additional US dollar not required to be spent in the future. The cost of averting one Disability Adjusted Life Year was US$917. The costs for averting one HIV-related death and one Disability Adjusted Life Year were US$36,020 and US$1,914 respectively. The life-time cost per Disability Adjusted Life Year averted was estimated to be US$522. According to standard willingness to pay thresholds, these values indicate that the programs are good value for money. If the life-time impacts of NSP are considered, the return-on-investment ratio was 42.82 and the cost required for one Disability Adjusted Life Year aversion was US$270. In comparison to NSP in other international contexts, NSP implemented by DFID/WB in Vietnam were typical. 64 The epidemic benefits derived from condom distribution programs for FSW were unlikely to have been cost-effective during the implementation period. However, when the life-time impacts of the condom distribution program are considered, the return-on-investment ratio was 4.53 and the cost required to avert one Disability Adjusted Life Year was US$425, demonstrating good long-term cost-effectiveness. The STI prevention program also demonstrated moderate cost- effectiveness in the long-term. BCC was deemed not to be cost-effective. The benefit-cost ratio (return on investment ratio) of 42.82 and 4.53 for the IDU and FSW program, respectively, is higher than the estimated 4.08 during appraisal, which is positive. On the negative side, the schooling study is much better powered than the raffle arm. IV. Financial Sustainability The project has played a significant role in building capacity for sustainability. Despite the training of local implementers, there was general consensus among service providers who participated in focus groups that the breadth of activities funded by the DFID/WB project would be impossible to sustain without on-going donor funding. It was reported that the national HIV/AIDS program could only cover a limited number of HIV prevention activities for the most easily accessible IDU and FSW. Sexually transmitted infections (STI) prevention programs for FSW and educational information funded nationally have been influenced by DFID/WB projects. From the perspective of service providers who participated in the qualitative assessment, the design and planning of the project appears to have been systematic and efficient. Staff training provided by the project was perceived as having facilitated effective program design and service providers reported that project-funded training had helped implement provincial HIV prevention interventions. Service providers also reported that training offered by the project had increased organizational capacity and resulted in a more skilled and professional prevention workforce. They reported acquiring new skills through project training activities in planning, project management and coordination, report writing and M&E. The programs were designed in an attempt to be efficient by applying a standard generic format, designed centrally and implemented by each province during the first phase of the DFID project (2004-June/2009). A separate and more flexible program design approach was initiated at the commencement of WB funding (2005). This provided ownership to provinces and allowed them to choose their own plans for implementation. This flexible template is now used in all provinces in Vietnam. During appraisal, it was estimated that the annual cost of sustaining the program after it ended, was (US$1,746,200 per year in 2004 dollars, or US$2,400,000 in 2014 dollar terms). Based on the cost per person reached, and the number of FSW and IDU covered during the project, the annual cost of the program is closer to us$7,089 000, an almost 3-fold higher level of costs needed to sustain the program from what was anticipated. The Government has committed some funds to HIV through the NTP, but by the end of the program, their contribution has been less than US$1,000,000 per year to finance the overall HIV response (not only the harm reduction program). 1.1 There were concerns expressed that if prevention programs established by DFID/WB funding are not sustained into the future then there could be a significant increase in the number of new infections by 2020 (4,698 extra infections), mostly attributable to IDU (4,061), FSW (59) and their clients (327). 65 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending (i.e. persons involved during preparation stage) Maryam Salim Operations Adviser OPSPQ TTL Mai Thi Nguyen Senior Operations Officer EASHH Susan A. Stout Manager OPCSRX Hung Viet Le Sr Financial Management Specialist EASOS Anh Thuy Nguyen Operations Officer EASHH Nga Quynh Nguyen Program Assistant HDNHE Hoang Xuan Nguyen Procurement Specialist EASR2 Toomas Palu Sector Manager EASHH Quynh Xuan Thi Phan Financial Officer GEFOB Ilaria Regondi Junior Professional Associate EASHH Quyen Do Duong Financial Management Specialist EASFM David Wilson Program Director HDNVP Lingzhi Xu Senior Procurement Specialist ECSH1 Robert Oelrichs Senior Health Specialist HDNHE Bekir A. Onursul Senior Environment Specialist EASEN Hoi-Chan Nguyen Senior Counsel LEGEA Hisham A. Abdo Kahin Senior Counsel LEGES Supervision/ICR (i.e. persons involved during implementation Mai Thi Nguyen Senior Operations Officer EASHH TTL since 3/2009 Huong Lan Dao Health Specialist EASHH Mai Thi Phuong Tran Finance Management Specialist EASOS Anh Thuy Nguyen Operations Officer EASHH Nga Quynh Nguyen Program Assistant HDNHE Samuel S.Lieberman Lead Economist EASHD Toomas Palu Sector Manager EASHH Kari Hurt Health Program Cluster Leader EASHH Sang Le Minh Health Environmental Specialist EASHH Ilaria Regondi Junior Professional Associate EASHH Maryam Salim Operations Adviser OPSPQ TTL until 2/2009 David Wilson Program Director HDNVP Lingzhi Xu Senior Operations Officer ECSH1 Xiaolu Bi JPA ECSH1 Marelize Gorgens Senior M&E Specialist HDNHE ICR author Minh Thi Hoang Trinh Team Assistant ECSH1 Nghi Quy Nguyen Social Development Specialist EASVS Hoang Xuan Nguyen Procurement Specialist EASVS 66 (b) Staff Time and Cost WB staff time and travel costs were financed from three sources: WB budget and two Bank- executed Trust Fund grants from DFID as part of the AF. The funds spent from each of these sources are detailed below Staff Weeks Travel Costs 2005 Total BB 14 $ 7,195 2006 Total BB 24 $ 9,637 2007 Total BB 20 $ 8,265 2008 Total BB 33 $ 40,440 2009 Total BB 35 $ 13,874 2010 Total BB 26 $ 5,843 2011 Total BB 17 $ 6,252 2012 Total BB 15 $ 7,269 2013 Total BB 17 $ 24,743 2014 Total BB 5 $ 182 BB SUB TOTAL 207 $ 123,699 2010 Total TF095867 15 $ 7,557 2011 Total TF095867 14 $ 12,531 2012 Total TF095867 14 $ 24,102 2013 Total TF095867 22 $ 14,323 2014 Total TF095867 3 $ - TF095867 SUBTOTAL 67 $ 58,513 2011 Total TF095866 3 $ - 2012 Total TF095866 3 $ - 2013 Total TF095866 4 $ - TF095866 SUBTOTAL 10 $ - Grand Total 285 $ 182,212 67 Annex 5. Project Risks and Project Transition Arrangements Table 5. Project Risks Identified during Appraisal, and the Extent to which they materialized Risks Risk Whether risk materialized Rating Political risks: H  Inadequate focus on vulnerable and marginalized groups No by commune and provincial officials could undermine project effectiveness.  Current approach to IDU and CSW “rehabilitation” may M In some provinces this was be accompanied by denial that there are feasible ways to experienced prevent further transmission of HIV by focusing on these populations.  Bank support to demonstration/piloting of services for H Yes, in 2011, the WB was residents of rehabilitation centers could entail criticized by Human Rights reputational risks if misinterpreted as World Bank Watch for implementing endorsement of lengthy, poor quality treatment in programs in detention centers overextended treatment settings.  Key constraint to the broad implementation of Harm H No, the inverse happened with Reduction strategies remains the national legal the legal environment, environment. partially as a result of the project, changing in a favorable way Technical Risks  BCC for vulnerable groups may not be effective. S Yes, this activity was dropped  There is the possibility while PAPs may support an M Partially overall program they may not link adequately to results.  National policies are aligned with a focus on changing M Yes the behaviors of vulnerable populations, but there is relatively little experience with the design of interventions to reach these groups at the provincial level. Implementation Risks  GDPMAC may have difficulty adapting to a provincial M No approach in the context of a long tradition of very centralized planning.  The project design anticipates a tripling of provincial Yes, in the beginning this was level finances for HIV/AIDS programs, some provinces H an issue may have difficulty absorbing the additional resources, particularly in the first and second years of the project.  The rapid acceleration of donor activity in HIV/AIDS in M Yes, in some provinces Vietnam is creating multiple job opportunities in the foreign NGO community; there is a risk of high turnover among project management staff as demand for skilled administrators from other donor agencies intensifies.  Delays in the establishment of provincial level capacity S Yes, this caused procurement to monitor and evaluate harm reduction and BCC delays in the first four years of campaigns could undermine technical quality of PAPs project implementation and depth of annual performance reviews.  There will be insufficient learning across provinces. M Yes, this was a challenge 68 Table 6. Transition / Sustainability Arrangements for Project Implementation Aspect of Harm Reduction Transition / Sustainability Challenges with transition Program arrangements arrangements FSW and IDU peer education Peer educators would be Reducing number of PEs: and outreach transitioned to LifeGap, NTP and Because of funding constraints, Global Fund not all peer educators could be transitioned. Reducing the number of PE may well results in an adverse effect on program outcomes. Reduced rates of pay for PE. Further diminishing the pool of peer educators, is the reduced value of PE allowances offered under the transition arrangements: PE allowances under the WB/DFID Project are $65/month, plus transport costs. GF will provide between US$55 and $65/month, which includes transport costs. NTP allowances are VND 500,000 (approx. US$24) total per month. Many PE have indicated they will not continue to work if their allowance drops from US$65 to $24/month. Needle-syringe program – Government to take over the Future of commodity distribution of clean needles and provision of clean injecting purchasing syringes paraphernalia (WB project financed 23 million clean injecting equipment in 2012) Condoms and condom social Government to continue with Future of the program marketing condom social marketing of VIP Plus, increasing the prices and thus making it self-sustaining HIV testing and counselling 29 HIV counselling and testing uptake sites would be transitioned to the Global Fund (10 sites) and to the NTP (19 sites) Sexually transmitted infections This would be transitioned to the treatment Government Methadone maintenance This would be transitioned to the treatment Global Fund 69 Annex 6. Stakeholder Workshop Report and Results During the course of the ICR mission, three lessons-learned workshops and discussions were held: one for the southern region provinces, one for the northern region provinces and one for the innovation projects. The purpose was to elicit responses on the project itself and to find out lessons learned from project implementation. The provinces were selected on a non-random basis by VAAC and the World Bank office, and the innovation project recipients were selected randomly, based on this randomization procedure: 2 out of 20 award recipients were selected randomly for the lessons learnt workshop by using random-number function in STATA (Version 12.1). Each project was assigned a numeric ID from 1 to 20. Two integer random numbers were generated from the uniform distribution on [1, 20]. A seed was set to ensure the precise reproducibility of sampling. Projects with the same numeric ID were selected for the interview. The chosen projects were No. 7: ”Be confident, we are always with you” from Ha Tinh Medical College in Ha Tinh City, and No. 10: “Intensifying access to and providing knowledge for HIV(-) spouse and sexual partners of HIV infected persons and injecting drug users” by Bright Futures Group in Thai Binh. Feedback from these lessons learned workshops and the interviews with selected key Innovation Day awardees were as follows: 1. Feedback from the Innovation Project interviews (30 October 2013) Questions asked during the interviews were: (a) What the Innovation Day Award was about? (b) What difference have the Innovation Day activities made to your community? How did innovation day change their lives? (c) What the participants did after the Innovation Day to continue their work? The answers are as follows: (a) The Innovation Day Award in 2007 and 2010 were organized by a team led by a standing committee composed of Vice Minister of MOH and representatives of 5 Ministries (MOH, Ministry of Police, MOLISA, Youth Union, and National Assembly). The event was first advertised through press conference and TV commercials to attract applications. Specialists from MOH were responsible for the selection process. Award recipients received funding to implement the activities and were required to submit completion reports to MOH and WB. (b) Award Recipient #1: “Intensifying access to and providing knowledge for HIV(-) spouse and sexual partners of HIV infected persons and injecting drug users” by Bright Futures Group in Thai Binh. In Thai Binh province, ¾ of AIDS patients live in communes that are 40-80kms from the nearest hospitals. The Bright Futures Group started offering HIV patients outreach and partner education programs in one commune before the Innovation Day in 2007. Upon receiving the Innovation Day funding, the service was expanded to four communes and more HIV patients were reached. An allowance of 50,000VND per member was available to members as needed on a rolling basis. Counselling service for HIV(+) patients and their partners were also provided in the communes. Example: A 12 years old HIV(+) girl whose parents died of HIV was brought to the hospital with the help of members in the commune. 70 Award Recipient #2: ”Be confident, we are always with you” from Ha Tinh Medical College in Ha Tinh City. Ha Tinh Medical College started a students’ volunteering program after learning about the Innovation Day Award through PAC. 120 students volunteered to provide care and education to HIV patients. Prior to receiving the award, the program only covered limited number of communes. The award funding enabled the college to invite specialists from PAC and doctors from the Infectious Disease Department to give students training courses. The service area was also expanded to three target communes and an infectious disease department in the hospital. (c) Currently the award recipients are networking with other NGOs and planning to scale up the activities. The award recipients were advised to mobilize funding from other organizations if they would like to keep the activities going. CPMU was advised to connect with PAC, seeking for sustainability in activities and developing proposals. 2. Feedback and Summary of Lessons Learned from Northern Provinces Workshop Behavior Change Communication (a) The quality and quantity of IEC materials improved. (b) The project contributed to reduce the stigma and discrimination towards the target groups. (c) The project has organized a wide network of peer educators among IOU, FSW and MSM groups. These networks contributed significantly to communication approaching activities, distribution of needles/syringes (injecting equipment), condoms and lubricant for target groups. (d) The project helped in strengthening and diversifying the media. It helped in maintaining the activities of the clubs of infected people, IDU, transferring information about HIV/AIDS on television, radio and electronic media. The project helped in intervention through peer educators of PIOUs, FSW, MSM groups to provide information about HIV/AIDS prevention for high-risk groups. (e) Enhanced local people’s understanding about HIV/AIDS, especially ethnic minority in mountainous districts. (f) Ensured persistent communication to combine diversity of forms to attract people to participate in activities. (g) Propagation and encouragement of high-risk groups to visit the HCT room and established mobile HCT teams for high risk groups for easy access. (h) BCC activities can be implemented through distribution of leaflets, booklets and magazines for prevention HIV/AIDS transmission. (i) BCC activities are implemented through many modes. For example, advocacy through workshops, communication sessions, competition on basic knowledge and the ways to prevent HIV/AIDS transmission for workers, sailors, fishermen, ports, Cat Ba island district and business enterprises. Harm reduction services & activities (a) Many new models of harm reduction intervention have been implemented such as injecting equipment exchange sites, injecting equipment holding boxes at commune health stations, mobilized the many high risk populations in project site to come receive services. (b) Provided suitable compensation for peer educators to take part in HIV/AIDS prevention activities. 71 (c) Helped in continuing and strengthening to implement the needles exchange program and condom use program. It focused on the benefits of using separate needles and condoms for all sexual activities in order to reduce the rate of sharing needles and increase the rate of condom use during sexual activities. (d) Strengthened activities for exchanging experiences in HIVIAIDS activities, especially in the harm reduction interventions activities. (e) In order to ensure sustainability of the Project, besides distribution of needles and condoms for free, it is necessary to transfer gradually to Condoms and Needles Social Marketing Program. M&E (a) The project ensured the data exchange via electronic communication from provincial level to community level. Reports have been done by specialized software nationwide. All work related to reporting, accessing, and synthesizing have become simpler and easier for implementation. (b) The M&E procedures developed have established efficient procedures to support for lower levels; it improved the quality of M&E activities such as development of content and checklist for M&E before field trip. (c) Helped in perfecting the M&E system from the Central to Provincial level. (d) M&E activities were performed regularly and periodically to ensure the project activities were implemented on schedule and with good quality. (e) Organized regular meetings at districts and the cities. (f) Implemented VIP Plus condom social marketing program Capacity building & training (a) Assisted in organizing training courses for improving knowledge and practicing skills on HIV/AIDS prevention for district collaborators and peer educators with at least 1 to 2 training courses every year. These courses provided timely knowledge on HIV/AIDS as well as practicing skills for district collaborators and peer educators and helped them implement their work well at project sites. (b) Before organizing training courses, the organizers should solicit the ideas of trainees for developing plan, content, and training methods accordingly. (c) After attending the training, the trainees should commit to work in the field of training for at least one year. (d) With the Project’s support, training for project staff and PAC staff are held every year. (e) Modern equipment is used in M&E activities at all levels. (f) Organized study tours for sharing experiences for project staff with other provinces. Management & coordination (a) The necessity of setting up the Steering Committee for HIV/AIDS prevention from Provincial to Community level has been recognized. (b) Full time staff should have incentives for maintaining human resources for PPMU. (c) Data collection and synthesizing should be done by specified software to ensure accuracy and easy accessing. (d) Project activities should include integration and use available health system about HIV/AIDS prevention activities in the grass-root levels. What would the province do differently if they were to design and implement the project again? (a) Take action to approach MSM group and develop activities for MSM Group. 72 (b) Allocate suitable human resources of PPMU to ensure coordination of resources with the highest efficiency. (c) Ensure Approved Action Plan in the beginning of the year in order to implement the activities. (d) Implement M&E program on a regular basis. (e) Implement BCC for ethnic minorities (f) Develop standard and full evaluation tool-kit for evaluating the impact of the project when project ends. (g) Use peer educators to continue to maintain implementation of harm reduction program and gradually transfer from distributing for fee to the Social Marketing Program. (h) Strengthen communication through multi-media channels such as radio, television etc. 3. Feedback and Summary of Lessons Learned from Southern Provinces workshop BCC (a) Some people take advantage to carry out illegal behaviors. (b) Need to maintain the PE network to approach the target population to distribute NS and introduce them to the HCT and STls treatment service. (c) Contributed to improved awareness of high risk groups and the general population in health protection, and preventing HIV transmission. (d) Positive group discussion has high effectiveness, assisting them to have more opportunities for approaching health services. (e) Improved awareness of community and target groups on preventing HIV/AIDS transmission, risk behavior reduction. (f) Direct traditional outreach model is no longer suitable because the target groups are no longer concentrated in hotspots; PE with limited knowledge provided limited BCC activities. Harm reduction services & activities (a) The targets groups sometimes still hesitate to approach PEs for injecting equipment exchanging and for other services information. (b) It's necessary to provide more project and funding to implement the HIV harm reduction programs for the HIV transmission prevention. (c) No interruption when not creating habit of buying condom and injecting equipment for target population. (d) Harm reduction activities has help in HIV transmission among MARP (e) More budget for free distribution of injecting equipment, condoms. (f) Provided condoms, needles/syringes (injecting equipment) and lubricant to help target groups to practice safe behaviors in sexual intercourse and drug use. (g) Developed social support fund for post-detoxified drug user to integrate into community, to help them stabilize their lives, and avoid relapse. HlV transmission prevention. M&E (a) Need to strengthen the M&E Staff’s capacity. (b) Need financial support to implement surveys (every 2 years) to have reliable data for harm reduction intervention activities in province. 73 (c) The 100% completion of monthly meetings and monitoring the districts leading to the success and effectiveness of the project. Without the completion of this activity, the project would have been a waste and not reached the results as expected. (d) Important contribution in monitoring and evaluating project activities; helps measure target indicators, efficiency and impact of the project. (e) M&E Skills of some staff are limited. Capacity building & training (a) Need to continue training for PEs and district collaborators. (b) Provided lots of knowledge for health staffs, collaborators and PE in HIV/AIDS prevention field. (c) Need to have a long term plan on training for human resources. Management & coordination (a) Continued monthly planning as well as organized monthly meetings to conduct monitoring at commune/ward level and the cooperation of all provincial district and commune levels. (b) Managed and coordinated project's districts and related agencies during project implementation process, ensuring implementation of activities as expected progress. (c) The procedure of approving procurement plan takes a long time. (d) Need to have a long term plan on training for human resources. (e) Department of Health should guide to implement harmoniously between activities of the project and activities of PAC. What would the province do differently if they were to design and implement the project again? (a) A timely transition plan is needed to avoid issues being outstanding during the transition process. The provinces are very passive when activities of this project end but recipient has not had detail information and guidance yet. (b) Harm reduction program will be continuously covered by national target program such as: maintaining PE and collaborator network by paying allowances, organize group discussion, outreaching high risk groups, distribution of injecting equipment, condoms, lubricant, etc. 74 Annex 6. Comments on Draft ICR from the Borrower The following comments on the draft ICR were received from the Borrower: 1. The HIV/AIDS Prevention Project started right in time of the beginning implementation of the National Strategy on HIV/AIDS for 2004-2010. Hence, the Project activities and budget were developed and allocated based on National Strategy priorities and action programs, which contributed greatly to the success of the Strategy – to control the HIV prevalence under 0.3% among general population as the Strategy objective. 2. The Project has made great contribution to the policy advocacy, especially policies related to implementation and expansion of harm reduction intervention program. 3. Vietnam Government and leaders of provincial People’s Committee have provided important support and guidance during project implementation; and there has been close coordination among Ministry of Health, CPMU and PPMUs to ensure better allocation of project resources in harmonization with resources of the National Target programs and of donors such as PEPFAR, Global Fund and others. 4. The mechanism of decentralization and active, flexible planning, budget allocation has been established. This mechanism has helped PPMUs to integrate and optimize the project resources. Moreover, thanks to this, the PPMUs have been able to develop their intervention models that were suitable for each province/city’s situation. 5. The HIV/AIDS Prevention Project is the 1st organization to focus on the capacity building of the staffs working in HIV/AIDS prevention system which was newly-established at that time. From the beginning to the end of 2013, the Project has supported 56 staffs for oversea master training and 27 staffs for in-country training. These staffs after graduation have contributed greatly to the HIV/AIDS prevention activities. 6. The Project is also the 1st place to provide effective support for developing the national HIV/AIDS M&E system. The information and data of this M&E system has been used for the evidence-based planning and policy advocacy not only for the project but also for the whole national program. 75 Annex 7. Comments of Cofinanciers and Other Partners/Stakeholders DFID noted that overall the project has been well delivered and achieved the original set targets. DFID co-financed with the WB project since early 2010 as a follow-up to the successfully completed project on HIV prevention funded by DFID during 2005-2009. DFID finds the cooperation with WB highly effective, complementary and took full advantage of both agencies’ comparative advantages. The coordination has reduced transaction costs for both the Government and financing agencies. WB team has been flexible in adjusting and adapting to new epidemic trends as well as partner’s priorities; the adjustment in 2010 in project design has been an example. The Task Team Leader has been highly approachable, consultative and effective in dealing with complex situations to lead the project to a successful completion. This project can be showcased as a good example of effective cooperation between development partners. 76 Annex 8. List of Documents Consulted Report Result of project implementation of the first 12 months of Vietnam Ministry November 2012 of Health 2012 Report Vietnam National Monitoring and Evaluation Plan for HIV Ministry of November Prevention and Control Program Health: VAAC 2012 PPT Vietnam National Monitoring and Evaluation Plan for HIV Dr Vo Hai Son, November Prevention and Control Program Vice-Dean, M&E 2012 Department Form Draft of D28 which was commented by Departments under Dr Vo Hai Son, Not dated VAAC, WHO and CDC/LG and project technical teams Vice-Dean, M&E Department Report Evaluation Report EV673: DFID May 2007 Country Program review: Vietnam Report A review of implementation of EMPF in Vietnam World bank March HIV/AIDS prevention project 2009 Government Project Progress Report for 2006 - 2012 CPMU December Report 2012 Government Project Progress Report up to October 2013 CPMU October Report 2013 Implementati ISR No.8, ISR No.9, ISR, No.10, ISR No. 11, ISR No.12, ISR World Bank Dates of on Status No.13 missions Results Reports Government CPMU 2005 Project Operational Manual of CPMU Document Project World Bank 2005 Project Appraisal Document Document PPMU PPMUs 2007 Provincial Action Plans 2007 of Thanh Hoa, Lai Chau, Yen documents Bai, Son La, Cao Bang, Nghe An, Khanh Hoa, An Giang Government VAAC 2006 Ethnic Minority Policy Framework (both English and Document Vietnamese versions) Project Project Midterm review report 10/2005 – 8/2008. Document Project Project Midterm Review Aide Memoire, December 2008 Document CPMU CPMU 2007 “Prevalence of HIV, Syphilis and high risk behavior among Report ethnic minority groups”, Research Report, conducted by CPMU in 2007. Project World Bank 2005 – Aide Memoires Documents 2013 Research data Ministry of Health. Results from the HIV/STI intergrated MOH 2009 biological and behavioral surveillance (IBBS) in Vietnam round II, 2009. Hanoi, Vietnam: Ministry of Health. Retrieved April 12, 2013 from http://aidsdatahub.org/dmdocuments/Vietnam_IBBS_Roun d_II_2009.pdf 2011. Academic García MC, Meyer SB, Ward P. Elevated HIV prevalence BMJ 2012 articles and risk behaviors among men who have sex with men 77 (MSM) in Vietnam: a systematic review. BMJ Open. 2012 January 1, 2012;2(5). Research data Ministry of Health. Results from the HIV/STI integrated MOH 2006 biological and behavioral surveillance (IBBS) in Vietnam, 2005-2006. Ha Noi, Vietnam: Ministry of Health. Retrieved April 10, 2013 from http://www.inthealth.ku.dk/reach/resources/surveillance.pdf /. 2006. World bank HEALTH/NUTRITION/POVERTY (HNP) AND World Bank 2000 report POVERTY SEMINAR REPORT: Improving the Health of Vietnam’s Poor: Analysis and Recommendations, by L. Richard Meyers and Anil Deolalikar, retrieved from http://siteresources.worldbank.org/INTPAH/Resources/Publ ications/Seminars/deolali2.pdf Global Joint United Nations Program on HIV/AIDS (UNAIDS). UNAIDS 2013 Report Global report: UNAIDS report on the global AIDS epidemic. Geneve, Switzerland Retrieved November 26, 2013 from http://wwwunaidsorg/en/media/unaids/contentassets/docum ents/epidemiology/2013/gr2013/UNAIDS_Global_Report_ 2013_enpdf. 2013 Project Mission Note : HIV/AIDS Prevention Project World Bank 2003 Document Reconnaissance Mission, May 26 – June 12, 2003 UNAIDS Viet Nam National AIDS Spending Assessment 2008-2010 UNAIDS 2011 report Global report Procurement Cost Trends for Global Fund Commodities: AIDSpan 2013 Analysis of Trends for Selected Commodities 2005 –2012, by Frank Wafula, Ambrose Agweyu and Kate Macintyre. Aidspan Working Paper 02/2013, April 2013 Global UNODC Fact Sheet about Vietnam: UNODC 2010 Report http://www.unodc.org/docs/treatment/CoPro/Web_Viet_Na m.pdf Global report Young people in Vietnam: From the 2009 Viet Nam UNFPA 2011 Population and Housing Census PPT INTERVENTION RESULTS FOR HIV/AIDS World Bank 2013 PREVENTION FOR ETHNIC MINORITIES 2006 – 2013, PowerPoint Presentation sent by TTL PPT Vietnam impact assessment of harm reduction programs Partnership for 2010 (IA): Provincial stakeholders review of preliminary findings Epidemiological January 21, 2010 Analysis Report Global Burden of Disease Profile for Vietnam, available at Institute for 2010 http://www.healthdata.org/sites/default/files/files/country_p Health Metrics rofiles/GBD/ihme_gbd_country_report_vietnam.pdf and Evaluation Report Evaluation of a decade of DFID and World Bank supported DFID 2013 HIV and AIDS programmes in Vietnam from 2003 to 2012 (by New South Wales University) 78 Annex 9. List of Publications Generated from data that the Project financed List of articles using data source of World Bank project in Southern Vietnam 1. Nguyen Duy Phuc, Khuu van Nghia, Nguyen Quang Bao, Pham Duy Quang, Pham Dang Doan Thuy, Tran Ngoc Huu, Ho Hoang Canh, Tran Nguyen Đuc, Nguyen Vu Thuong. HIV and risk behaviors among male IDU in Dong Nai. Journal of Preventive Medicine (in Vietnamese) , Issue XXI, No. 6 (124): 15-20, 2011 2. Khuu Van Nghia, Nguyen Vu Thuong, Pham Duy Quang, Phan Thu Huong, Nguyen Thanh Long, Bui Hoang Duc, Nguyen Duy Phuc, Nguyen Quang Bao, Tran Van Ut, Le Van Viet, Tran Ngoc Huu. Factors related to HIV transmission among IDU in Vinh Long, 2007. Journal of Preventive Medicine (in Vietnamese), Issue XXI, No. 6 (124): 21-27, 2011 3. Nguyen Quang Bao, Pham Duy Quang, Hoang Quoc Cuong, Khuu Van Nghia, Phan Thu Huong, Nguyen Thanh Long, Nguyen Duy Phuc, Mai Hoang Anh, Tran Tho Anh, Tran Ngoc Huu, Nguyen Vu Thuong. HIV and factors associated with condom use among MSM in An Giang. Journal of Preventive Medicine (in Vietnamese), Issue XXI, No. 6 (124): 15-20: 32-37, 2011 4. Nguyen Vu Thuong, Nguyen Duy Phuc, Khuu Van Nghia, Ho Hoang Canh, Tran Nguyen Duc, Tran Phuc Hau, Tran Ngoc Huu. Factors related to sharing needles/syringes among IDU in Dong Nai. HCMC Health Bulletin, Issue. 3 Suppli.16, 2012 5. Quang Duy Pham, Thuong Vu Nguyen, Cuong Quoc Hoang, Van Cao, Nghia Van Khuu, Huong Thu Thi Phan, Anh Hoang Mai, Huu Ngoc Tran, David Peter Wilson, Lei Zhang. Prevalence of HIV/STIs and Associated Factors among Men Who Have Sex with Men in An Giang, Vietnam. Sexually Transmited Diseases. 2012 Oct;39(10):799-806. 2012 6. Nguyen Vu Thuong, Nguyen Duy Phuc, Tran Ngoc Huu, Ho Hoang Canh, Tran Tho Anh, Vu Dinh Tuyen, Khuu Van Nghia. Factors related to HIV testing among pregnant Khmer-Vietnamese and Chinese-Vietnamese females in Southern Vietnam. Paper submitted to the fifth National AIDS Conference in Hanoi, 12/2013 7. Nguyen Vu Thuong et al. 2013. Factors related to positive attitudes towards people living with HIV / AIDS among ethnic minority individuals in Southern Vietnam. Paper submitted to the fifth National AIDS Conference in Hanoi, 12/2013 8. Khuu Van Nghia et al. 2013. Factors related to HIV testing among Khmer-Vietnamese and Chinese- Vietnamese males in Southern Vietnam. Paper submitted to the fifth National AIDS Conference in Hanoi, 12/2013 9. Nguyen Duy Phuc et al. 2013. Factors related to HIV testing among Khmer-Vietnamese and Chinese- Vietnamese females in Southern Vietnam. Paper submitted to the 2013 Public Health Conference in Ho Chi Minh City, Dec. 2013 10. Nguyen Vu Thuong et al. 2013. Factors related to STI/HIV among female sex workers in Ben Tre, 2012. Paper submitted to the 2013 Public Health Conference in Ho Chi Minh City, Dec. 2013 11. Nguyen Vu Thuong et al. 2013. Prevalence of HIV, syphilis and associated risk behaviors among Khmer-Vietnamese and Chinese-Vietnamese in Southern Vietnam. Paper submitted to the 2013 Public Health Conference in Ho Chi Minh City, Dec. 2013 12. Khuu Van Nghia et al. 2013. Factors related to unlidding used needles/syringes prior to disposal among people who inject drugs in 4 southern provinces, 2010 . Paper submitted to the 2013 Public Health Conference in Ho Chi Minh City, Dec. 2013. 79 Annex 10. Report on review of implementation of project’s transition arrangements Introduction This review was undertaken by the WHO during the last six months of the Project. A consultant was hired to support transition arrangements and help MOH and VAAC plan for continued implementation of the harm reduction program after the project had ended. This review was undertaken by a team of two consultants – one local and one international – who visited several provinces and held key informant interviews with stakeholders at national level, regional level, in the districts, and with beneficiaries. They also did a desk review of documentation from VAAC, CPMU, and the PPMUs. A. Transition process: observations, issues and recommendations i. PE and outreach Observations and Issues: The PE program is facing major cuts across the 32 Project provinces. PE are the ‘eyes and ears’ of the HIV response – they reach high-risk individuals, identify new hotspots, provide prevention commodities, and facilitate referrals; they provide the health department with information and data on high-risk populations, and observe and report changing trends in behaviors among people at risk for HIV. Without an effective outreach arm, health services will lose their link to high-risk individuals. While some Project provinces have elected to use remaining WB/DFID funds to continue support to the PE program through to December 2013, a large proportion of peer educators in numerous provinces/districts stopped working after 30 June 2013. Some provinces are waiting for funding to be mobilized from the Global Fund to Fight AIDS, Tuberculosis and Malaria (GF), while others without GF support are hoping there may be some financing from their provincial/district governments or through the national target program (NTP). Here is the most recent overview of numbers of current WB/DFID PE transferred to other programs: WB/DFID PE transferred: 64% of PE for people who inject drugs (IDU) 71% of PE for female sex workers (FSW)14 Here is the breakdown of which projects/funding sources received the PE: To GF: 45% of PE for IDU 48% of PE for FSW To NTP: 19% of PE for IDU 22% of PE for FSW 14 Data from the WB/DFID CPMU, 2013 80 To LifeGap: 0.3% of PE for IDU 1.2% of PE for FSW The following provinces will be left with none or scarce IDU PE: Cao Bang, Lang Son, Quang Ninh, Thai Binh, Hau Giang, Soc Trang, An Giang, Kien Giang. The Project provinces in which PE for MSM are supported will be continued by the GF program. PE allowances under the WB/DFID Project are US$65/month, plus transport costs. GF will provide between US$55 and $65/month, which includes transport costs. NTP allowances are VND 500,000 (approximately US$24) total per month. Many PE have indicated they will not continue to work if their allowance drops from US$65 to US$24/month. Thankfully, a significant number of PEs across the 12 provinces that the consultants have visited to date have said they would continue to work at this reduced allowance rate. RECOMMENDATIONS: 1. CPMU should develop and print a certificate of appreciation for PE: “Thank You for Your Excellent Service to Addressing HIV in Vietnam. Without you, Vietnam could not reach the people most at risk.” These certificates should be for all PE, in all districts, under the WB/DFID Project. Names could be hand written at the district level. 2. Provinces and districts with little prospect of future funding through GF or NTP, but still with a low to medium IDU burden, should focus any local PE funding they have on IDU. This target group still drives the HIV epidemic in Vietnam, they are generally less able to form their own networks, they have a high incidence of psycho-social & physical co-morbidities, and their overall HIV prevalence and risk level (for both transmission and acquisition) is far greater than either MSM or FSW. 3. FSW PE can and should be encouraged to generate additional income through condom social marketing – i.e. selling condoms to guest houses, hotels, entertainment establishments (EE) and to FSW in all provinces. 4. Districts and provinces should retain only the best PE for IDU & FSW, and work out how to pay them a larger allowance – this will require an amendment to cost norms. These strong PE can then recruit volunteer deputies (người i di n or “IDU leaders”) to help undertake IEC, distribute commodities and refer clients to services – thereby exponentially expanding the network of reach. In addition, provinces/districts should explore new and resource-efficient outreach modalities – see Section B. New outreach models and their application to harm reduction. 5. PE for MSM should be encouraged to establish their own networks, adopt outreach models using existing gay internet sites and chat rooms. District health services could provide allowances to only a minimum number of MSM PE, in order to continue overall coordination and data collection. 6. PACs and district health authorities should explore existing commune-level capacity to assist in maintaining outreach (including NSP): use motivated PEs, community volunteers, commune/village health workers, the Department of Labor, Invalids and Social Affairs’s network of social workers. 81 7. Associated with outreach, Methadone maintenance therapy satellites could contribute greatly to reaching hidden IDU and encourage them to enter treatment. The consultants witnessed this phenomenon in Chau Doc/An Giang Methadone maintenance therapy clinic, where eight clients had never had any contact with a PE or any other harm reduction service prior to accessing Methadone maintenance therapy services. 8. There is basically no (or extremely limited) civil society to support populations at risk in any of the provinces visited. Provinces need to encourage the growth of groups, networks, clubs, CBOs to help carry out outreach and link people at risk for HIV to prevention and treatment services. ii. NSP Observations and Issues: If NSP and PE for IDU are cut, there could be major consequences for both HIV incidence and community attitudes towards harm reduction services. As IDU in An Giang stated, NS cleaning, re-use and eventually sharing would likely increase in the absence of free NS distribution and/or friendly pharmacies. And if PE are cut then there will be extremely limited staff numbers to collect dirty/used NS – leading to a likely community backlash against IDU. Low dead space NS, proven to minimize the risk for HIV transmission in circumstances when NS are immediately re-used by a second user, are not broadly available in Vietnam. IDU in Ho Chi Minh City (HCMC) prefer a 1mL insulin NS (the so called “red tip” NS), which is also a low dead space NS. Population Services International (PSI) is piloting a NS social marketing intervention in three provinces, promoting and supplying these 1mL insulin NS for sale at various non-traditional and some traditional outlets. See section B. iv for more information. Here is the most recent overview of Project NS data and information: - 18 million NS remain from WB-DFID, and distribution will continue though GF and NTP.15 - Coverage in the 32 Project provinces will decided by PACs. - With an estimated 77,041 IDU in the 32 provinces of the WB-DFID project, and with 18 million NS left, each IDU could receive 1 NS per day for 233 days from Oct 15th, 2013 – i.e. up to June 15th, 2014. - WB-DFID CPMU has plans to purchase more NS (for free distribution) with remaining central level funds, which would extend this coverage timeline. RECOMMENDATIONS: 1. Provinces need to ensure a steady supply of NS for free distribution for the immediate to medium term. NS social marketing may prove to be effective in future, but to reach scale would take years and in the meantime infections need to be averted through the distribution of free NS to IDU most in need. 15 WB-DFID CPMU data from Oct 15th 2013 82 2. The Project should endeavor to promote and supply low dead space NS, where appropriate, using any remaining Project funds in the final months of the Project life. 3. The presence of Methadone maintenance therapy in a community should not lead to the diminishing of NSP, for the following reasons: (a) current Methadone maintenance therapy coverage is far too low to make a real impact on the prevention of HIV transmission; (b) clients on Methadone maintenance therapy often slip and inject drugs; and (c) there will always be IDU in the community who are not ready for opioid substitution treatment and who may continue to adhere to risky injection behaviors. 4. There is an urgent need to install fixed NS boxes (or other type of discreet 24 hour NS access) in a number of provinces. This would need to occur in parallel with, or after, Police Advocacy efforts. In the face of reduced outreach/PE, fixed boxes remain an effective and financially efficient way to ensure IDU still have access to new NS at all times of day. 5. Fixed boxes can manifest in a variety of forms – from metal boxes attached to trees, to plastic bags full of NS hanging behind a gate, wall or sign outside a commune health station. 6. As in Recommendation 6 above, PACs and district health should explore existing commune- level capacity and systems to assist in maintaining NSP: engage motivated PEs, volunteers, commune/village health workers, the Department of Labor, Invalids and Social Affairs’s network of social workers. iii. Condoms and condom social marketing (CSM) Observations and Issues: The WB/DFID Project has supported the development, purchase and distribution of a new high quality, low cost condom brand – VIP Plus. Since April 2012, this brand has been distributed by companies and PE in over 30 Project provinces. In several provinces we now see: (a) FSW PE distributing other brands of condoms in times when the VIP Plus supply has run out, which appears to make little difference to the client, EE or FSW – as they still use and/or stock condom brands which they believe are of acceptable quality; (b) There is a wide variety of condoms available on the market, at all quality and price points; (c) FSW PE continue to distribute free condoms (purchased by the Health Department or PAC; sometimes even VIP Plus) to all FSW, MSM and IDU. Note: Most respondents in provinces have said FSW will buy and/or obtain condoms regardless of whether VIP Plus are available. Condom distribution companies working with VIP Plus have said that they would be happy to purchase VIP Plus themselves in future, without government subsidies, and that the condoms could be sold for more than three (3) times the current retail price. Obviously companies have taken advantage of the kick-start funding provided by the government (through WB/DFID), which supported initial brand development, promotion, marketing and condom cost subsidy. While it is encouraging news that companies say they would purchase VIP Plus directly from manufacturers in future, (a) the Government currently holds the trademark and would need to either sell it to a company or at least allow companies to use it, and (b) there are other well- known brands at a similar price and quality points to VIP Plus, which were also previously socially marketed by donors –e.g. OK condoms through DKT– and VIP Plus would have to compete with these brands at a similar retail price. 83 RECOMMENDATIONS: 1. Further develop policies, legislation and locally enforced regulations to ensure all guest houses, hotels and EE have condoms available at all times. 2. Owners/managers of guest houses, hotels and EE should be continuously encouraged to participate in condom programs. 3. FSW PE could form cooperatives to purchase condoms (any brand) at a discount or wholesale price, and then sell these to guest houses/hotels, EE and FSW at a modest profit. This will increase personal income for PE and ensure overall sustainability of the social marketing approach. Individual PEs are already doing this in areas of Ben Tre and Dong Nai provinces. 4. Seek clarity from the Vietnam Administration for AIDS Control (VAAC)/MOH on whether they intend to support VIP Plus condom social marketing in future. If there are insufficient government funds to continue purchasing and subsidizing VIP Plus, then there seems little point in the Government retaining the VIP Plus trademark. 5. STOP GIVING OUT FREE CONDOMS (never VIP Plus for free), unless a high-risk sub group has been identified and it is considered critical to provide them with condoms – e.g. street-based sex workers. 6. District health and peoples’ committees could encourage all condom distribution companies to give out free condoms at events, certain hotspot locations, etc., which will also help the companies’ marketing efforts. iv. HTC uptake and ARV treatment referral Observations and Issues: All IDU, MSM and FSW interviewed by the consultants have said they would really like to access discrete, truly confidential and fast-result testing for HIV. Getting an HIV test is already a harrowing and stressful event; people want to know their results quickly, not to wait for a week or more. They would also like to avoid any chance of being stigmatized or discriminated against. They often do not trust the established HCT clinics to be confidential or to maintain client anonymity. Current PE referrals to testing are generally low. Some PE can encourage 1 to 3 clients per month at most to be tested, and in some provinces (Khanh Hoa, Ben Tre) many PE had not managed to physically accompany a client to be tested for over a year. Most high-risk respondents said they get tested only once per year at most – whether IDU, MSM or FSW. Overall, high risk individuals are not accessing HIV testing at a frequency high enough to enable an effective HIV response. Some relevant data from HSS+ 2012: - 31% of IDU accessed HIV testing and know their results in the last 12 months - 26% of HIV positive IDU received ARV treatment - 40% of FSW accessed HIV testing and know their results in the last 12 months - 10% of HIV positive FSW received ARV treatment 84 RECOMMENDATIONS: 1. Roll out rapid testing at the commune level. Respondents said that commune level rapid testing would be much preferred to the current HCT system, but even if available they would probably go and get tested at a commune far away from their own residence in order to ensure anonymity – as they fear the lack of confidentiality. 2. VAAC should explore options for do-it-yourself, take-home test kits to be made available in Vietnam, through commune health stations and the private sector (e.g. pharmacies), and through existing NS and condom distribution channels (PE, social marketing, non-traditional outlets). v. STI treatment Observations and Issues: When supported by the Project up to 2012, mobile STI screening clinics were very popular and uptake was excellent among FSW. There is also a well-functioning collaboration with private STI clinics to provide data to the public health system. RECOMMENDATIONS: 1. Provinces should attempt to continue mobile STI screening services on a periodic basis, and if funding is scarce, consider implementing a user co-pays system, whereby the FSW and/or the EE owner could contribute to costs – for both screening and treatment. MSM could also be targeted by this service. 2. VAAC should endeavor to continue the link with private STI clinics, for ongoing data provision and for possible future collaboration. vi. Methadone maintenance treatment (Methadone maintenance therapy) Observations and Issues: Many districts/provinces cannot afford to establish a ‘traditional’ Methadone maintenance therapy clinic under the current algorithm. As such, both co-payment models and local level Methadone maintenance therapy satellites represent affordable and effective alternatives. Additionally, commune level satellites could help to bring hidden IDU out and into treatment. The consultants witnessed this phenomenon in Chau Doc/An Giang Methadone maintenance therapy clinic, where eight (8) Methadone maintenance therapy clients had never had any contact with a PE or any other harm reduction service prior to accessing the Methadone maintenance therapy service. They had learnt of and been referred to Methadone maintenance therapy by family and local community members. FHI 360 is currently working to commence a pilot Methadone maintenance therapy satellite program in Dien Bien province, aiming to deliver Methadone maintenance therapy at the commune level. See section B. iii for more details. RECOMMENDATIONS: 1. Overall, new patient eligibility criteria for Methadone maintenance therapy need to be enforced. Many provinces (Da Nang, Son La, An Giang) are not yet following in full the new 85 eligibility criteria, and hence client numbers are very low. This is not acceptable, particularly given the high financial resources presently allocated to establishing and operating a standard Methadone maintenance therapy clinic under the current algorithm. 2. VAAC should push to commence, evaluate and roll out both co-payment and Methadone maintenance therapy satellite pilots as soon as possible. Many provinces need these (likely) more financially efficient and effective models for delivering Methadone maintenance therapy. Without these models many provinces and districts will not be able to afford Methadone maintenance therapy. vii. Management and human resources Observations and Issues: In many provinces with GF funding, there are often communication gaps between WB/DFID provincial PPMU, GF PPMUs and PACs. In addition, significant resources overlap (e.g. human resources for related services) and services are not integrated and co-located for ease of client/patient access to services. Several provinces are concerned that they have not received clear guidance from GF about what exactly GF will support and when funds will reach the province. At the district level, health departments say they are waiting for guidance from the provincial level. Some Methadone maintenance therapy and PE staff have not received salaries for several months as they wait for GF funding to arrive (e.g. Son La). Many districts fall just below the GF selection criteria limits (a combination of minimum sizes for risk groups and HIV prevalence among risk groups), meaning they will not receive support from GF. Finally, provinces visited to date have expressed a need to have renewed guidance from VAAC/MOH, MPS and MOLISA on how to deal with major funding cuts to their harm reduction and Methadone maintenance therapy programs, and what efficiencies they should be seeking. RECOMMENDATIONS: 1. Directors of PACs should ensure communication and handover of responsibilities for WB/DFID funded services is prioritized and transparent, with clear handover plans, timelines and resource implications listed, and strategies developed. 2. VAAC, PACs and district health departments need to pursue “true” one stop shops, where various related services are co-located, and in which these services share staff and other resources as appropriate. For example: (a) Train the Methadone maintenance therapy counsellor/s to do both HTC and ART adherence counselling, and also provide addiction counselling to current IDU who access services; (b) One medical doctor could work across Methadone maintenance therapy, HTC and ART; (c) Methadone maintenance therapy could operate from 07:00am to 10:30am only, freeing up Methadone maintenance therapy staff to undertake other duties with different services; (d) Nurses could work across HTC, Methadone maintenance therapy and ART – even STI if present; (e) Administrative and cleaning/maintenance services could be shared; 86 (f) Rent and other utility costs would be shared and therefore reduced for each service budget; 3. Districts that will not receive significant GF or NTP funding will need some sort of “in -reach” from neighboring districts with greater resources. This in-reach could include: (a) Methadone maintenance therapy satellite dispensaries (in future); (b) Periodic visits from trained/experienced PE of neighboring districts/provinces for IDU, MSM an FSW (as appropriate to the local epidemiology); (c) Rapid HIV testing at the commune level; (d) Periodic assessment of NS fixed boxes (or other 24 hour free NS access) by experienced PE supervisors; (e) and more. 4. In terms of renewed VAAC guidance to inform provinces on how to seek more efficient yet still effective interventions, here are some of the recommended ideas emerging from discussions with provinces and central-level partners: (a) Outreach/PE: Combining outreach responsibilities for prevention, treatment & care (this would require limited training); Combining multiple risk groups into the TOR of a single PE (e.g. one MSM IDU could reach both MSM and IDU); Guidance and support on other innovative outreach models (see Section B.) (b) IEC/BCC: Minimize funding; fund only basic IEC and BCC. Ensure any continuing BCC focuses on IDU/IDU, FSW and MSM – not on the community (c) NSP: More fixed boxes and other forms of “24 hour NS access” (e.g. plastic bags of NS discreetly placed behind the gate of the commune health station), mobile boxes, friendly pharmacies, PE “deputies” (d) Condoms: Social marketing only; Provincial & district monitoring, supervision and TA; Encourage individual and/or cooperative SM, for profit, through PE & other channels that provinces identify (e) Methadone maintenance therapy: Seek permission to relax the strict methadone drug control & handling laws; Dispense methadone from community health stations to stabilized patients; Implement the user co-pay system (f) HCT/HTC: Roll out rapid testing at the commune level; Recommend only 1 test per 6- 12 months (perhaps every 3 months for extremely high risk behavior), thereby drastically reducing costs associated with the current testing algorithm. Explore and develop guidance for take-home, do-it-yourself test kits to be supplied and sold through various outlets. (g) UIC, Capacity Building: Manual UIC data collection approach and tools to continue; Capacity building > seek in-country opportunities (best practice model sites, NIHE, Pasteur, etc.) (h) Management structures, M&E: Revise TORs/SOWs of staff in health departments to ensure that new responsibilities in harm reduction are covered, managed and services effectively implemented B. New outreach & service delivery models and their application to harm reduction i. MSM Online The HCMC PAC (supported by the US Centers for Disease Control – or CDC) presented preliminary results for this innovative outreach model designed to reach high risk MSM more effectively, with lower financial and human resources. “MSM Online” in HCMC is pro ving to be 87 effective and adaptive in reaching MSM and encouraging them to (a) adopt safer sex behaviors and (b) get tested for HIV. Preliminary results after 1 year of implementation: The model used less than half the number of PEs employed in the traditional MSM outreach model; it reached far greater numbers of hidden MSM and provided information on HIV, AIDS, HTC, STI and ARV services and referred them to testing; the pilot demonstrated good primary cost effectiveness - the annual budget for MSM Online outreach was about VND 200 million, whereas the traditional was about VND 3.2 billion. PE shared experiences chatting with MSM online: they work 3 to 4 hours/day and can chat with 5 to 6 people at one time. The messages they give to MSM include encouraging safe sex and referral to services such as STI and HCT/HTC. ii. Peer driven intervention (PDI) and the indigenous leader outreach model (ILOM) HCMC PAC (supported by CDC) and PSI presented preliminary results for these innovative outreach models designed to reach high risk IDU and FSW in HCMC more efficiently and effectively 1. The PDI model: piloted on IDU and FSW in HCMC. This pilot applies a modified version of the Respondent Driven Sampling methodology normally used in research, in order to identify potential strong seeds, then incentivize these seeds to introduce IDU and FSW to HTC services. Results after Year 1: IDU: Selected 7 seeds who successfully introduced 169 IDU to HTC. One seed was selected until the seed ceased to yield referrals, then another seed was chosen. FSW: Selected 6 seeds who successfully introduced 131 FSW to HTC. Similar to IDU, one seed was chosen until referrals ceased, then another seed was selected. Comparison of PDI to the traditional outreach model: (a) Traditional model employs 16 PE for IDU and FSW, while the new model employs only 2 PE; (b) Under PDI, 87-90% of IDU and FSW that presented at HTC were 'first time' testers; and (c) 98% clients returned to receive their results, whereas an average of only 70-80% return in the traditional model. NOTE irregularity: Positive results for HIV among IDU and FSW tested in PDI were 3.4% & 2.1% respectively, whereas HIV prevalence among IDU from sentinel surveillance in 2012 in HCMC was 30% (FSW results were similar between HSS and PDI generated data). This discrepancy in IDU prevalence is currently being analyzed. PSI also shared its PDI model in HCMC, using the Song Dep group supported by PSI and established in 2010. Its purpose is to reach hidden and high risk FSW who inject drugs. PEs provide information on HIV/AIDS, new needle-syringes and HTC services. This group covers district 6, 7 and four parks in HCMC. Limited findings to date but HTC attendance and the rate of return for results is encouraging. 2. The ILOM: Reach and identify natural leaders of IDU groups, train him/her as a pseudo-PE and use them as information providers and case managers. These leaders will meet other IDU and provide HIV/AIDS information, clean needle-syringes, introduce to HTC services, speak with 88 other IDU to understand their needs and how to work with them, and share the contact information with PAC staff. Limited findings to date but several 'leaders' have been identified and are providing information/data to PAC. Overall, these three innovative pilot models in (i) and (ii) above target high risk groups, and are being implemented alongside traditional models. Activities have been implemented only 3 to 12 months and need more time to produce significant results. iii. Methadone maintenance therapy satellite model pilot Supported by FHI 360 in Dien Bien province, the purpose of this pilot is to demonstrate that commune-level dispensing of methadone and management of stabilized clients is feasible and more cost effective than the current Methadone maintenance therapy algorithm. In addition, clients would be better linked to other commune-level services. The Pilot is yet to commence, due to provincial/district government licensing and salary issues, but it is eagerly awaited. iv. Needle-syringe social marketing (Ha Noi, HCMC, Thai Nguyen - PSI) Emerging evidence suggests the use of low dead space syringes can significantly reduce the risk of HIV and Hepatitis C transmission among IDU. While all syringes retain some fluid in the “dead space” when the plunger is fully depressed — some syringe designs retain more fluid than others. “High dead-space” syringes can retain more than 40 times the amount of blood compared to syringes. Based on market and IDU insights, the USAID-funded needle and syringe social marketing project designed a pilot to use social marketing approaches to improve convenient access to and motivate use of LDSS among IDU in Hanoi and HCMC. Pilot activities were extended to Thai Nguyen province with support from Merck Sharp & Dohme (Asia) for HCV Prevention. Now PSI is proposing a study to assess coverage and results achieved during the first year of LDSS social marketing activities in select provinces, and to inform future scale up. The specific objectives of the study are to assess: (a) Changes in levels of awareness about LDSS product attributes and benefits among male IDU in pilot areas (b) Knowledge of HCV risk, transmission and prevention methods (c) Changes in % of IDU who practice safer injecting behaviors including LDSS use & non- sharing (d) Barriers to motivating IDU to switch from HDS to LDS in Vietnam & assessing opportunity, ability & motivational indicators and population characteristics associated with LDS/HDS use and sharing behaviors (e) Exposure to social marketing activities including targeted LDSS distribution through non-traditional outlets accessible to IDU, trade marketing to motivate pharmacies to stock/display/promote LDSS and behavior change communication to introduce and promote the benefits of LDSS—as well as related, broader harm reduction programming (f) Correlation between exposure to harm reduction & LDSS social marketing activities and safer injecting practices including non-sharing & LDSS use While this study is yet to commence, preliminary pilot implementation data is showing that IDU in provinces that do not normally use LDS NS (such as the red tip 1mL insulin NS) are embracing the benefits of LDS and purchasing the socially marketed NS at non-traditional and some traditional outlets. 89 v. Treatment 2.0 (Can Tho, Dien Bien - WHO) The purpose of the Treatment 2.0 pilot is innovation, simplification and improved community access to HIV treatment & care at the commune level. It aims to integrate HIV testing, treatment and care into commune health systems (which it should be noted are already very busy, having to manage and deliver around 27 different state health care programs/priorities). Treatment 2.0 intends to simplify service delivery and improve ART adherence. The pilot asks PEs to identify and refer more clients to ART at the commune level. There is also rapid testing at the commune level, providing quicker results, which should also lead to a rise in the uptake of testing. As an adjunct to this model, mobile testing is also being explored to reach & test high risk people in more remote and hard to access locations. 90 Annex 11. Summary of Return on Investment Calculations (from UNSW report) Source: 2013 UNSW Evaluation Report 91 Source: 2013 UNSW Evaluation Report 92 Annex 12: Details of correlation between project provinces achieving their IDU and FSW coverage and concomitant reductions in HIV prevalence. Source: PEMA analysis, 2010 93 Summary of HIV prevalence trends in provinces Source: UNSW report, 2013 94 Annex 13. Project provinces, additional Project provinces after DFID AF and the overlap with PEPFAR provinces in Vietnam 95