Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized 1 90998 This Page is intentionally left blank Cover photo courtesy: Oxfam International, Creative Commons. Republic of Niger Niger Allocative Efficiency and Financial ‎ Sustainability Study Niger’s‎HIV‎response:‎ Targeted investments for a healthy future CISLS Niger authors: Zeinabou Alhousseini and Zakou Harouna World Bank authors: Nicole Fraser, Nejma Cheikh, Markus Haacker, Emiko Masaki, Djibrilla Karamoko and Marelize Görgens University of New South Wales authors: Cliff Kerr, Richard Gray, Andrew Shattock and David Wilson UNAIDS authors: Luisa Frescura and Claude Konan June 2014 HDNHE AFRICA Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The findings, ‎ interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The ‎ World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. ‎The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the ‎ part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries.‎ Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a ‎ violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages ‎ dissemination of its work and will normally grant permission to reproduce portions of the work promptly.‎ For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright ‎ Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, ‎ http://www.copyright.com.‎ All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World ‎Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Niger’s‎HIV‎response: Targeted investments for a healthy future ‎ Findings from the HIV allocative efficiency and financial sustainability study ‎ This page is intentionally left blank Table of Contents Abbreviations .................................................................................................................................. i Acknowledgements....................................................................................................................... iii Executive Summary‎....................................................................................................................... v Section 1. Introduction ................................................................................................................ 1 What is the context of this analysis?‎........................................................................................ 1 Why is this analysis needed?‎................................................................................................... 2 What are the objectives of the analysis?‎.................................................................................. 4 What methods were used?‎....................................................................................................... 5 How are the study findings presented in this report?‎............................................................... 5 Section 2. Development context, health priorities and health system challenges in Niger ................................................................................................................. 7 What is the state of economic and human development?‎....................................................... 7 How much is spent on health and where does the money come from?‎ ................................. 12 What are the health priorities?‎ ................................................................................................ 13 What is the provision and reach of health services in Niger? ‎............................................... 14 Section 3. The state of the HIV epidemic‎ ................................................................................. 17 What has been the historical course and scale of the epidemic?‎.......................................... 17 How is HIV distributed in the population of Niger?‎................................................................. 19 Where do new HIV infections occur?‎..................................................................................... 24 Section 4. HIV financing, HIV services and programs and their efficiency ......................... 27 How is the HIV response financed, and what has been spent on the different programs?‎.................................................................................................. 27 HIV services and ‎ What HIV services and programs have been scaled up?‎...................................................... 29 HIV investments: Impacts from past expenditures and best resource allocations in the ‎ current strategic plan period ......................................................................................... 43 Section 5. HIV investments: Impacts from past expenditures and best resource allocations in the ‎ current Strategic Plan period................................................................ 43 Impact of historical HIV expenditures ‎.................................................................................... 43 Best resource allocation in the current HIV strategic plan period .......................................... 44 Section 6. Financing needs and fiscal space implications of HIV investments .................. 49 NSP 2013–17: Financing needs and gaps ............................................................................. 49 Anticipated impacts of the investments under the NSP 2013–17‎.......................................... 51 Longer-term costs of responding to HIV and future savings .................................................. 52 Annexes A1.1. Optimizing budget allocations for impact........................................................................... 55 A1.2. Population groups and transmission pathways ................................................................. 55 A1.3. HIV programs included in the analysis .............................................................................. 56 A1.4. Overview of the Optima methodology ............................................................................... 56 A1.5. Optima calibration to the HIV epidemic in Niger................................................................ 57 A1.6. Comparison between Optima and Spectrum estimates .................................................... 59 A1.7. Relationships between program costs and effects on behavior and program coverage‎............................................................................................................. 60 A1.8. NASA expenditure data used in Optima............................................................................ 60 Figures Figure 1. Years lost due to disability by cause and age, Niger (2010)‎......................................... vii Figure 2. Optima-estimated trends in HIV incidence by sub-populations, Niger (2000–20)‎ . ...... viii Figure 3. Number of infections averted by historical HIV spending from 2007–12, Niger (2007–35)‎ ..............................................................................................................ix Figure 4. Cumulative infections over 2013–25 that would result from different spending scenarios, ‎Niger..............................................................................................................xi Figure 5. Current versus optimal budget allocations for minimizing incidence (left) and DALYs (right) ‎ over the period 2014–2025, Niger ................................................... xii Figure 6. Health expenditure by source of financing, Niger (1995–2011)‎................................... xxi Figure 7. Shifts in leading causes of DALYs in Niger (1990–2010)‎.............................................. 1 Figure 8. Health expenditure by source of financing, Niger (1995–2011)‎................................... 13 Figure 9. Burden of disease indicators, Niger (1990, 2010; ranking among 46 countries in Sub-‎Saharan Africa)‎................................................................................................. 14 Figure 10. Access to health facility by health district, Niger (2011) ‎ .............................................. 15 Figure 11. Non-users of family planning: contact with providers, Niger (2012)‎ ............................. 16 Figure 12. HIV prevalence by population, Niger (2002--12)‎.......................................................... 17 Figure 13. HIV prevalence by sex and residence, Niger (2002, 2006, 2012)‎............................... 18 Figure 14. New HIV infections, AIDS deaths and ART, Niger (1994–2013)‎................................. 19 Figure 15. HIV prevalence by FSW age and duration of sex work, Niger (2011)‎......................... 20 Figure 16. HIV prevalence by age group, Niger (2002, 2006, 2012)‎ ............................................. 22 Figure 17. Numbers of people living with HIV by region, Niger (2006, 2012)‎............................... 22 Figure 18. HIV prevalence by socio-economic strata, Niger (2006, 2012)‎ ................................... 23 Figure 19. Spectrum-estimated HIV incidence by age and sex, Niger (2012)‎.............................. 24 Figure 20. Optima-estimated trends in HIV incidence by sub-populations, Niger (2000–20)‎....... 24 Figure 21. Concentration of the HIV burden by region, Niger (2012)‎ ............................................ 25 Figure 22. Sex work-related hotspots of transmission, Niger (2013)‎............................................ 26 Figure 23. Summary of efficiency loss along the PMTCT cascade, Niger .................................... 34 Figure 24. Men: Likelihood of recent HIV test by HIV status in socio-demographic strata, Niger ‎(2012)‎.................................................................................................................. 36 Figure 25 Number of ART patients and estimated ART need, Niger (2004–12)‎ Figure 26. Number of infections averted by historical HIV spending from 2007–12, Niger (2007–35)............................................................................................................ 43 Figure 27. Estimated infections by population and time, Niger (2000–35)‎ .................................... 44 Figure 28. Current versus optimal budget allocations for minimizing incidence (left) and DALYs ‎ (right) over the period 2014–25, Niger ...................................................... 45 Figure 29. HIV investment cascade, Niger (2014–17)‎.................................................................. 46 Figure 30. Cumulative infections that would result from different spending ‎ scenarios, Niger (2013–25) ..................................................................................................................... 47 Figure 31. Minimum budgets required to meet NSP coverage targets, Niger ‎ 2013–17 ............... 48 Figure 32. Financial costs of HIV/AIDS, Niger (2010–30, Euro million, 2013 prices) ‎.................. 50 Figure 33. Financial costs of HIV/AIDS, Niger (2010–30, percent of GDP)‎.................................. 51 Figure 34. Modeled HIV incidence and AIDS deaths, Niger (2010–30)‎ ........................................ 52 Figure 35. People living with HIV and ART, Niger (2010–30)‎....................................................... 52 Figure 36. Projected HIV costs caused by new and past HIV infections, Niger (2013–30, GDP)‎........................................................................................................... 53 percent of ‎ Figure 37. Cost of one new infection occurring in 2013, Niger...................................................... 54 Annex Figure 1. Calibration of model to the HIV epidemic in Niger‎............................................. 57 Annex Figure 2. Calibration of the model to the HIV epidemic in Niger. ...................................... 58 Annex Figure 3. Comparison of annual new HIV infections, annual mother-to-child transmissions, and ‎annual AIDS deaths estimated by Optima and Spectrum over 2000–20 ..................................................................................... 59 Tables Table 1. Selected macroeconomic data, Niger (2013) ................................................................... 8 Table 2. HIV/AIDS-related expenditures, Niger (2007–12)‎.......................................................... 28 Table 3. Estimated protected and unprotected sexual acts, Niger (2012)‎ .................................... 31 Table 4. Key gaps and bottlenecks at each step of the PMTCT cascade, and actions identified in the ‎ e-MTCT strategy, Niger ............................................................ 34 Table 5. Comparison of some unit costs of HIV/AIDS services (Euro)‎ ......................................... 39 Table 6. NSP financing acquired, needed, and financing gap (2014–17)‎.................................... 49 Annex Table 1. Estimated new HIV infections by population for the best- fitting simulation (2012–35)‎................................................................................ 61 This page is intentionally left blank Abbreviations ANC Antenatal care ART Antiretroviral therapy ARV Antiretroviral BCC Behavior change communication CD4‎ Cluster of differentiation 4‎ CFA West African currency (franc)‎ CHLD ‎ Children aged 0–14 years (Optima model)‎ CISLS Coordination Intersectorielle de Lutte contre le SIDA (National AIDS Coordination ‎ Authority)‎ CPN Consultation prénatale (antenatal care)‎ DALY Disability adjusted life year DHS Demographic and health survey e-MTCT Elimination of mother-to-child transmission FAD ‎ Females aged 25–49 years (Optima model)‎ FOLD ‎ Females aged 50+ years (Optima model)‎ FP Family planning FSW Female sex worker FYTH ‎ Females aged 15–24 years (Optima model)‎ GAR Gross attendance rate GDP Gross domestic product HCT HIV counseling and testing HDI Human development index IEC Information- education- communication IMF International Monetary Fund KfW Kreditanstalt für Wiederaufbau LRP Low risk population MA Males aged 25–49 years (Optima model)‎ MARP Most at risk populations (also called key populations)‎ MDG Millennium development goal MICS Multiple indicator cluster survey ‎ MIN ‎ Mine workers (Optima model)‎ MIG ‎ Migrants and mobile populations (Optima model)‎ MNIH ‎ Maternal, neonatal and infant health ‎ MOLD ‎ Males aged 50+ years (Optima model)‎ MSM Men who have sex with men i ii Abbreviations MSP Ministère de Sante Publique MTCT Mother-to-child transmission MYTH ‎ Males aged 15–24 years (Optima model)‎ NASA National AIDS spending assessment NDHS Niger Demographic and Health Survey NGO Non-governmental organization NSP National strategic plan (2013–17) ‎ ONPPC Office National des Produits Pharmaceutiques et Chimiques OVC Orphan and vulnerable children PCR Polymerase chain reaction PDES ‎ Economic and Social Development Plan (2012–15)‎ PDS ‎ Health Sector Development Plan (2011–15) also HSS (Health Sector Strategy)‎ PEP Post-exposure prophylaxis PICT Provider initiated counseling and testing PLHIV People living with HIV PMTCT Prevention of mother-to-child transmission PRI ‎ Prisoner (Optima model)‎ Q1-5‎ Wealth quintiles 1 (lowest 20%) to 5 (highest 20%)‎ RBF Results based financing RH Reproductive health RNM Resource needs model SDDCU ‎ Sustainable Development and Inclusive Growth Strategy Niger 2035‎ SIDA-3‎ Regional STI/HIV program in 9 West African countries SM Social marketing SRH Sexual and reproductive health SSG Second generation surveillance STI Sexually transmitted infection SW Sex worker TB Tuberculosis TRU ‎ Truckers (Optima model)‎ UNI ‎ Uniformed security (Optima model)‎ UNSW University of New South Wales VCT Voluntary HIV counseling and testing VPP ‎ Voluntary pooled procurement WB World Bank Acknowledgements The authors express their sincere gratitude to the Government of Niger for entrusting the World ‎ Bank and UNAIDS with this analytical study. We especially acknowledge the leadership of Dr ‎ Zeinabou Alhousseini, Head of the National AIDS Coordination Authority (Coordination ‎ Intersectorielle de lutte contre les IST/VIH/SIDA, CISLS), and the assistance from Mr Zakou ‎ Harouna, M&E expert at the CISLS, in sharing a large collection of epidemiological, behavioral and ‎spending data. We would also like to acknowledge the support from Dr Azara Bamba- Louguet, UNAIDS ‎ Country Director, and Mr Adamou Dambagi, M&E Adviser, UNAIDS Niger. ‎ The authors sincerely acknowledge the active participation of a group of about 45 key stakeholders‎of‎Niger’s‎HIV‎response‎ who met in August 2013 at the introduction and ‎ launch of this study. They convened again in December 2013 to review the preliminary findings. ‎ The authors are thankful to them for sharing their local knowledge and experiences. ‎ Ms Mary Mahy, Senior Epidemiology Adviser UNAIDS Geneva, provided support for the analysis by reviewing ‎ the Spectrum model on which this analysis builds. The financial sustainability model was populated ‎ by Mr Michael Obst, World Bank. The report benefited from review by Ms Emi Masaki, World Bank. Ms Pascale Ledeur-Kraus completed the translation.‎ This study was funded by the World Bank Group with UNAIDS support for two technical advisers. ‎ The views expressed in this report do not necessarily represent the opinion of the World Bank or ‎ UNAIDS. iii This page is intentionally left blank Executive Summary (providing the main conclusions and full policy recommendations)‎ Niger has seen significant progress in economic development and in the reduction of extreme ‎ poverty. The economy grew by over 11% in 2012, thanks to oil, uranium and agricultural production. ‎ The percentage of people living on less than USD 1.25/day has been reduced from 66% in 2005 to 44% in ‎ 2008, scoring above the sub Saharan African average of 49%. Average ‎ income has increased from USD 586 in 2000 to USD 701 in 2012 (expressed as the gross national income [GNI] per capita in purchasing power parity [PPP] ‎ terms, constant 2005 international dollar). The positive changes are mirrored by the improved life ‎ expectancy at birth, growing from 39.5 years in 1980 to 48.3 years in 2000 and 55.1 years in 2012.‎ Despite of these encouraging statistics, living conditions for Nigeriens remain challenging. Food insecurity persists even in years when production yields a surplus. About 60% of households are only able to cover their food requirements for three months, and in periods of food shortages, half of the population suffers from malnutrition. Adults, on average, have a mere 1.4 ‎ years of formal school education, and as a consequence, Niger has ‎ one of the lowest literacy rates in the world in spite of a massive expansion of the education system. ‎ The provision of health services is dominated by the public sector whose services are provided through 829 Integrated Health ‎ Centers linked to 2,000 health posts that are managed by community health workers. Higher ‎ level care is delivered by 33 district hospitals spread across 42 health districts, nine regional ‎ public health departments and six regional hospitals, including several reference hospitals and ‎ maternity centers. Approximately 47% of the population lives within five kilometers of a health ‎ facility, but in some of the rural health districts access to a health facility is much lower. In Niamey, the capital city, nearly 100% ‎ of the population has good geographical access to public sector health provision, which translates ‎ into comparatively better service-use statistics than elsewhere. Overall, health sector ‎ capacity is insufficient, and is compounded by the doubling of the ‎ population every 23 years.‎ In 2011, health spending accounted for 5.1% of gross domestic product (GDP) (Euro 225 million or (United States Dollar [USD] 303 million), and was ‎ dominated by public sector heath spending (57% of total) and private out-of-pocket spending (39% of total). ‎ External financing accounted for almost one-third of all health spending (World Health Organization [WHO], 2014). ‎ Spending on HIV care and treatment accounted for about 1.6% of total health spending, and 2.7% of public health spending in 2011, while HIV/AIDS was responsible for an estimated 1.4% of all Years-of- life-lost in 2010 according to the Burden of Disease Study.‎ As Niger progresses on its development path, it has an opportunity to review the scale and ‎ trajectory of the HIV epidemic, and the investment needed today to appropriately address the ‎ current and future disease burden of HIV/AIDS; and to ensure that HIV investments are high- impact, sustainable, and can best ‎ support‎ Niger’s‎ overall‎ health‎ system‎ and‎ development‎ priorities. v vi Niger’s‎HIV‎response: Targeted investments for a healthy future Equally, it is important to explore‎how‎the‎Niger‎government‎can‎contribute‎its‎“fair‎share”‎to‎the‎ HIV response, in agreement with the 2013 resolutions made by political leaders at the Abuja Summit: to step up the mobilization of domestic resources to strengthen the health system and ensuring diversified and sustainable financing for health. The HIV allocative efficiency and financial sustainability study ‎ The main objective of the study was to provide the Government of Niger with a comprehensive ‎ analysis of HIV and health financing needs, investment opportunities, and health system development in the context of its HIV National ‎ Strategic Plan (NSP) 2013–17. ‎ The analysis provides support for HIV policy decision-making, investment scenarios and ‎ programmatic targeting and prioritization. In addition, the analysis helps Niger build the case for HIV ‎ and health impact investment including delivering estimates of health care savings as a result of these investments. ‎ This analysis was implemented by the World Bank in collaboration with UNAIDS from a request ‎ for analytical support from the Government of Niger. The study involved a desk review of HIV- and ‎health-related evidence, epidemic trends and financial modeling. The Optima model ‎ (formerly Prevtool) was used to estimate optimal resource allocation during the NSP, and the ‎impact and cost-effectiveness of past HIV investments. A financial commitment framework was ‎ used to estimate longer-term costs and savings of the HIV program and the fiscal dimension of HIV ‎ in Niger. ‎ In this executive summary, we summarized the main conclusions, insights and the policy recommendations emanating from this analysis.‎ ‎ 1. Health and development challenges in Niger Conclusion: The health transition in Niger is in its early stage, with communicable diseases and ‎ poverty-related conditions dominating the disease statistics, while malaria and HIV/AIDS are the ‎ leading emerging diseases In Niger, newborn, nutritional and maternal causes such as diarrheal diseases, lower respiratory ‎ infections, and protein-energy malnutrition as well as several communicable diseases including ‎ malaria, meningitis, measles, and tetanus are leading causes of health loss. Non-communicable ‎ diseases are a minor but emerging health issue, responsible for 18% of all disability adjusted life years (DALYs), according to the ‎comprehensive international 2010 Burden of Disease Study. ‎ Rapid‎demographic‎growth‎is‎a‎central‎issue‎in‎Niger’s‎development‎process.‎The‎rate‎of‎growth‎ ‎ has changed very little in the last 20 years and was 3.9% in 2012. On ‎ average, women give birth to 5.6 children in urban areas compared to 8.1 children in rural areas. Women who have secondary education ‎ have an average of 3.1 fewer births than women who do not, highlighting education‎plays‎in‎Niger’s‎demography.‎‎ the profound role ‎ HIV/AIDS has, together with malaria, been the leading cause of emerging disease burden ‎ over the last 20 years. However, compared to other African countries, the burden of HIV/AIDS is relatively ‎ small (see Figure 1, combining HIV/AIDS and TB), and is therefore a more manageable public health ‎ issue. Nevertheless, several top causes of local disease burden—iron deficiency anemia, ‎ malnutrition, TB and diarrhea—interact with the immunocompromised status of HIV- positive ‎ individuals and are likely to accelerate disease progression and morbidity. Executive Summary vii Figure 1. Years lost due to disability by cause and age, Niger (2010)‎ Source: IHME/World Bank; Niger profile: ‎http://www.healthmetricsandevaluation.org/sites/default/files/country- profiles/GBD%20Country%20Report%20-%20Niger.pdf‎ . Note: Years lived with disability (YLD) are estimated by weighing the prevalence of different ‎conditions based on severity. ‎ The size of the colored portion in each bar represents the number of YLD attributable to each ‎ cause. The height of each bar shows which age groups had the most YLDs in 2010. Causes are ‎ aggregated.‎ 2. HIV ‎ epidemic dynamics Conclusion: The low grade HIV epidemic is in decline, but unless strong action is taken to intervene, ‎ there is a possibility that new economic growth-related co-factors such as mining, urbanization, ‎economic development, and migration could result in increases in new infections in future years Empirical data from HIV surveys suggest a contracting HIV epidemic in Niger: HIV prevalence has ‎ decreased rapidly, both in the general population and in key populations. The modest HIV ‎ prevalence decreases in urban females may be linked to a combination of risk behaviors in towns ‎ (leading to new infections), and the relatively good access to antiretroviral therapy (ART) keeping HIV-positive women ‎ alive. The stable HIV prevalence in prisoners suggests a sub-epidemic in correctional facilities with ‎continuing high risk behaviours among prisoners and the possible influence of ART on HIV ‎ prevalence levels. The cultural practice of circumcising males has spared Niger from a generalized ‎ epidemic, and will continue to reduce the probability of HIV transmission from men to women (this ‎ analysis estimated that over 200 million unprotected sexual acts occur every year, or 90% of all sexual ‎acts in Niger). ‎ viii Niger’s‎HIV‎response: Targeted investments for a healthy future Modeled epidemic trajectories―which are developed using empirical HIV prevalence data―‎ suggest that HIV transmission has declined sharply, from 8,470 new HIV infections at its peak in ‎ 2000 to 1,450 in 2013 (Spectrum 2013). The Optima model predicts that the past HIV incidence decline will continue into the future, provided that that current HIV response is maintained, and that there are no changes in exogenous factors that could influence HIV transmission dynamics ‎(Figure 2, by sub-population). ‎ Figure 2. . Optima-estimated trends in HIV incidence by sub-populations, Niger (2000–20)‎ . Source: Optima application Niger, January 2014‎ Note: Graph shows estimated new HIV infections per year.‎ However, the wider context of the HIV epidemic in Niger is changing: ‎ i. Urbanization: Niger’s‎epidemic‎has‎always‎been‎of‎an‎urban‎type,‎‎ and the urban population is growing at 6.2% (significantly higher than the national average of 3.9%). ‎ Urbanization is often linked to risk behaviours outside the controlling village context, to disrupted ‎ family life, poor living conditions and transactional sex for survival or for affording modern ‎ consumer goods, especially among young females.‎ ii. Rapid economic growth: The changing wealth levels, with economic growth at 11%, is associated with higher income levels and the availability of disposable income, which among Niger’s‎men is ‎ linked to buying sex (note that HIV is now concentrated in among the wealthiest 40% ‎ of men).‎ iii. Mining sector development: The development of the petroleum and mining sector has ‎ resulted in sex workers (SWs) relocating from Niamey―where‎there‎are specialist HIV services‎available‎for‎them―to the mining zones and petrol fields. The availability of jobs, infrastructure development and ‎ growing economic opportunities encourage migration and HIV‎(survey‎data‎show‎a‎clear‎link‎between‎men’s‎ mobility, which also create risk contexts for ‎ sexual risk behaviors to mobility and being ‎away from home).‎ iv. Low literacy: Meanwhile, a key factor for protection from HIV risk―school education―has ‎ stagnated in Niger, largely due to the very rapid increase in children of school going age. In 2012, HIV ‎ prevalence was lowest in the most educated population segment (secondary education or above) ‎with only 0.1% of men and 0.2% of women being HIV infected.‎ These evolving contextual factors could put past HIV gains at risk if concerted HIV prevention are not maintained.‎ efforts ‎ Executive Summary ix Recommended policy actions ‎ 1. The government must continue to undertake strategic HIV monitoring activities, especially in key ‎ populations at higher risk—this includes female sex workers (FSWs), sex work clients in the mining areas and young urban ‎ females. Only if HIV prevalence in these populations is well tracked can the risks of mining industries, urban and economic development be assessed and specific interventions be designed. There is a necessity for vigilance and for excellent strategic information on the epidemic dynamics. HIV ‎ prevalence in sex workers should be a key performance indicator of the HIV program.‎ 2. Social sector investments, especially those in the education sector and in family ‎ planning/reproductive health will greatly benefit the HIV program and should be seen part of a broad and powerful response toward mitigating future HIV risks— as integral ‎ improved literacy, better ‎employment opportunities for females, reduced fertility and better closely linked with the future trajectory of the epidemic. ‎ reproductive health are all ‎ 3. Retrospective ‎ impact and cost-effectiveness of past HIV investments Conclusion: Past HIV investments have contributed significantly to the decline in HIV incidence ‎ The impact of historical HIV expenditures on the prevention of new HIV infections was estimated using the Optima model. It was estimated that HIV spending from 2007–12 averted close to 3,900 ‎ ‎ infections in those six years (Figure 3). Furthermore, the downstream impact of these prevented ‎ infections was much larger, since secondary HIV infections and onward transmission were ‎ also averted. By 2035, the spending from 2007–12 will have averted an estimated total of 12,600 ‎ infections, at a cost of Euro 3,500 (USD 4,700) per infection if only HIV program spending is considered, and Euro 4,600 (USD 6,200) if all HIV spending—including management and ‎ coordination costs—is included. This is slightly higher than some cost-effectiveness ratios in other ‎ ‎ countries (e.g., men who have sex with men (MSM) program in Thailand, ‎ USD 1,600–USD 6,400, and Vietnam HIV program USD 890–USD 1,200 per infection averted), suggesting that it might be possible in future to implement the same programs ‎ at a lower cost without compromising their effectiveness. Figure 3. Number of infections averted by historical HIV spending from 2007–12, Niger (2007–35)‎ 14000 12629 11178 HIV infections averted 12000 10000 8528 8000 6000 3893 4000 2000 0 2007-2012 2007-2017 2007-2025 2007-2035 Source: National AIDS spending assessment (NASA) 2007–12, available epidemiological and behavioral data from Niger, and drawing ‎ on the A1.6).‎ relationships between funding, HIV outcomes, and program coverage (Annex ‎ x Niger’s‎HIV‎response: Targeted investments for a healthy future Based on the common understanding of how epidemics spread from core groups, second-order ‎ investment benefits are particularly high when HIV programs target high HIV incidence populations ‎like FSWs. The modeling findings suggest that incidence reductions were indeed greatest in FSWs ‎ (from 3.5% annual HIV incidence in 2000 to 1.1% in 2012), but also had knock- on effects in ‎ other populations like sex-work clients and even in low-risk populations. In the ‎ absence of other major programs being implemented, it can be reasonably assumed that HIV incidence reduction in FSWs after 2000 can in part be attributed to the regional STI/HIV program in 9 West African countries (SIDA-3) and the Multisectoral STI/HIV/AIDS project 2003-2009 supported by the World Bank1. However, no evaluation of SIDA-3’s‎impact‎has been conducted in Niger.‎ Recommended policy action 3. Strengthen capacity for the country to undertake analytical work on the impact and cost-effectiveness ‎ of HIV investments through partnerships and capacity building— Niger disposes a large body of ‎ HIV monitoring data and needs to prioritize evaluation and learning to strive for high-‎ impact, high-return HIV strategies in a resource-constrained health funding environment. Cost-‎ effectiveness and return-on-investment data build powerful economic arguments for HIV ‎ investments, and increase our understanding of the best investment choices in a changing epidemic context.‎ 4. Increasing ‎ the future impact of Niger’s HIV response—“More health for the money”— through more efficient allocation of funds to priority ‎ programs Conclusion:‎Niger’s‎HIV‎response‎can‎achieve‎more‎value‎for‎money;‎with‎the‎resources‎ already ‎ committed until 2017, the country can achieve an estimated additional 12% reduction in HIV ‎ incidence if HIV resources are optimally allocated Better allocation of HIV resources could increase the number of new HIV infections averted: The ‎ modeling analysis on optimal resource allocation during the NSP period of 2013–17 for best HIV ‎ incidence reduction was based on the amount of funding already acquired by end 2013 for the four ‎ remaining years of the NSP (annual average Euro 4.8 million or USD 6.5 million). Although this ‎ amount is lower than the 2012 expenditure of Euro 12.5 million or USD 16.3 million, the optimization ‎ predicts that an additional 12% reduction in new HIV infections could be achieved from ‎ 2014–25, if resources were allocated differently than in 2012 (Figure 4). 1 Implementation completion and results report, WorldBank, December 2009. http://www- wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2010/01/13/000333038_20100113233224/Render ed/PDF/ICR127500P07161C0disclosed011121101.pdf, accessed 27 May, 2014. Executive Summary xi Figure 4. Cumulative infections over 2013–25 that would result from different spending scenarios, ‎ Niger ‎ Sources: Spending data from NASA; Niger epidemic, demographic, behavioral and service data in ‎ the populated Optima model. Note: No spending = a continuation of 2012 spending, and optimized spending at various budget levels ‎ relative to expected funds available, 2014–17; 2012 budget = 2012 expenditure pattern; 100% budget = Euro 4 ‎ .8 m/USD 6.5 m per year, based on acquired funding by end 2013; over time, other costs increase in line with GDP per capita.‎ Drug costs are kept constant ‎ This analysis recommends higher funding allocations for ART, prevention of mother-to-child transmission (PMTCT) and FSW programs; and lower ‎ allocations for untargeted activities for low- management costs forecast‎in‎Niger’s‎resource‎ risk populations (Figure 5). The rapidly escalating ‎ needs model are higher than inflation (currency ‎ inflation at 3% plus population growth at 4% combined). An efficiency of about Euro 1.7 million ‎ (USD 2.3 million) could be gained from 2014– 17 if management costs grew at the inflation rate only. If the cluster of differentiation 4 (CD4) treatment ‎threshold was raised to 500, an ‎ approximate estimate of additional resources needed over the four years is 12%. However, this would require more in-depth analysis using ‎ detailed program data, unit costs and assumptions on delivery models. ‎ An HIV incidence reduction of 50% is only feasible in the longer-term, according to this analysis ‎  Halving HIV incidence during the NSP period would be very costly. It is estimated that achieving this ‎ HIV prevention goal by the end of 2017 would cost Euro 24 million (USD 32.3 million) per year—500% of ‎ the currently acquired budget of an average of Euro 4.8 million (USD 6.5 million) annually from 2014–17- and with optimized program allocations ‎  In contrast, halving HIV incidence by 2025 is more feasible. It is estimated that this would cost Euro ‎5.52 million (USD 7.42 million) per year—115% of the average acquired budget of 2014–17, and ‎ substantially lower than the total 2012 expenditure of Euro 12.5 million (USD 16.3 million)‎ Better program (technical) efficiency could save even more funds: there is a large difference in antiretroviral (ARV) prices between the Global Fund procurement system and the government tender system. ‎ This is currently not increasing costs by much since most ARVs are purchased by the Global Fund ‎ system, and only a small volume of ARVs are purchased via the government system (the small size of ‎ these orders is partly responsible for the higher pricing).‎ xii Niger’s‎HIV‎response: Targeted investments for a healthy future Figure 5. Current versus optimal budget allocations for minimizing incidence (left) and DALYs (right) ‎ over the period 2014–2025, Niger Sources: Spending data from NASA; Niger epidemic, demographic, behavioral and service data in ‎ the populated Optima model. Note: Current = 2012 spending pattern; OVC, blood, PEP, waste, ‎ and management costs were kept fixed. Condoms are part of the HIV intervention packages for ‎ the different populations; FSW = Female sex workers, LRP = low risk populations, a combined group of general population females and males aged 15+ years;‎ MARPs = most at risk populations, a combined group of prisoners, ‎migrants, men who have sex with men, uniformed security/defence personnel, mine workers and ‎ truckers; OVC = orphan and vulnerable children; PEP = post- exposure prophylaxis. Recommended policy actions‎ 4. Use the epidemic and programmatic ‎ intelligence and analytical approaches to identify savings, efficiencies and higher-impact scenarios, to strive towards better value for money in the HIV program—Niger is unlikely to be able to close its NSP resource gap of almost Euro 45 million (USD 60 million) ‎ for the four remaining NSP years. However, this analysis suggests that a greater impact can be ‎ obtained from optimal resource allocation across priority programs, and from data-informed geographical ‎targeting. Resource allocation towards the highest risk population—female sex workers—undoubtedly provides value for money. Management and coordination costs need to be contained and functions integrated with HIV and health administration where possible. 5. Funding needs to be shifted towards ART, PMTCT and FSW programs, and away from untargeted ‎activities for low-risk populations, with budget splits depending on total resource availability. ‎Based on empirical data and impact modeling, SW-related investment increased to approximately 4% of the annual HIV budget—this is likely should be significantly ‎ to prevent an extra 7,000 new HIV ‎ infections over 10 years (with the already acquired NSP resources of Euro 4.8 million/USD 6.5 ‎million annually). 6. Review the government procurement system for ARVs to gain efficiency—although ‎ most ARVs are currently purchased through the Voluntary Pooled Procurement system, there Executive Summary xiii is a ‎ need to ensure that better ARV prices can be acquired through the Government tender system, as ARV ‎needs grow because of an increasing number of people eligible for ART. ‎ 5. Geographical ‎ targeting of priority programs Conclusion:‎Niger’s‎HIV‎response‎can‎ capitalize better by targeting resources to areas of burden and transmission ‎ high HIV ‎ hotspots As HIV epidemics contract, data-informed HIV intervention targeting becomes highly important in order to continue to drive down HIV transmission. Although people on ART are ‎ less infectious overall (lower viral load), they represent a HIV reservoir that is growing ‎annually as more people enter the ART program. Some resources must therefore be targeted ‎ towards maximizing viral suppression in the ART patient population (such as treatment adherence, and ‎ drug resistance monitoring). This analysis estimated that within six years, the number of people ‎ living with HIV (PLHIV) in Niamey will increase from under 9,000 to about 12,000, partly due to ‎ ART improving the life expectancy of PLHIV.‎ About 70% of PLHIV live in one of four regions in the South-West covering only one ‎ quarter of the country surface area. Agadez and Diffa probably had the lowest PLHIV numbers in ‎ 2012 compared to the other regions. Over one half of all PLHIV reside in Niamey, ‎ Dosso or Tahoua with circa 26% of these living in Niamey (according to the 2012 study). In the scale-up of PMTCT, a phased approach may be ‎ recommended, considering the regional distribution of PLHIV and the urban nature of the epidemic. ‎ Data support a PMTCT first phase scale-up in the South-West portion of the country and in towns. ‎ With HIV prevalence at 17% in female sex workers compared to HIV prevalence of below 1% in most ‎populations, targeting areas where sex work takes place is a prerequisite to the success of HIV ‎prevention in Niger (towns, petrol/mining zones, infrastructure construction areas, etc.). There is ‎ an estimated 2.3 million unprotected commercial/paid sexual acts each year in Niger (compared to ‎ over 200 million regular sexual acts). Condom promotion needs to prioritize this programming gap first, given the disproportionate contribution of sex work to ongoing HIV transmission. ‎ Epidemiological data from prisons suggest that there is increased HIV transmission in these settings with HIV ‎prevalence levels at 2.7%, thus targeting HIV and health resources toward prison populations remains important. There is a ‎ lack of data to effectively direct services that target men who have sex with men (MSM), however, it is possible to improve the targeting of male sex workers as a first step. The UNAIDS Investment Framework recognizes that targeting key ‎ populations is done in the larger context of reducing stigma and discrimination2.‎ The ‎ strengthening of an enabling environment is necessary to support the effectiveness and efficiency of program activities which can be affected by barriers to service use due to social ‎ ‎ exclusion, marginalization and criminalization of key populations.‎ Recommended policy actions 7. Shift resource allocations away from untargeted activities for the largely HIV negative population to populations where an impact on HIV can be gained—the most important target populations are all types of FSWs, high-risk MSM, people on ART, people living with 2 UNAIDS, Investing for results, Results for people. UNAIDS/PCB(30)12.CRP.4. 2012. http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2012/JC2359_investing for- results_en.pdf xiv Niger’s‎HIV‎response: Targeted investments for a healthy future HIV (PLHIV), people who are diagnosed but are not yet on ART, prison populations, ‎ and pregnant women in areas of higher HIV prevalence. Consideration should be given to necessary actions that improve access to services such as the reduction of stigma, discrimination and community mobilization3 in these targeted allocations. ‎ 8. HIV testing and case identification must be targeted where yield is likely to be high— such as outreach activities and services for key populations, STI clinics, spouses of known PLHIV, and malnutrition screening for infants. For efficiency, HIV testing activities should be accompanied by strengthening the health professional’s capacity, covering topics such as ‎ non‎ discrimination, informed consent, the obligation to treat all, and confidentiality. It is important that health services are welcoming, supportive and efficient, especially for key ‎ populations. 9. Companies, especially those attracting migrant labor and those in the petrol/mining sector, ‎ should be required, by regulation, to provide a minimum package of HIV/health services to ‎ employees—in some cases, companies should be required to support surrounding communities. ‎ interventions in the ‎ 6. Improving ‎ the implementation efficiency of programs for female sex workers Conclusion: Sex work-related interventions have lacked focus and investment, despite their ‎ epidemiological, cost-prevention and humanitarian importance. Improvements in the implementation of FSW ‎ programs woud increase quality, coverage and reach.‎ An estimated 28,000 FSWs operate in Niger, but there is consensus that the real number may be ‎ higher, with hidden SWs remaining uncounted (mapping in four regions counted a mere‎ ‎ 3,200 FSWs, illustrating the challenge of understanding the true size of this heterogeneous ‎ population). FSWs are 43 times more likely to be HIV-infected than females in the general ‎ population—the previous sections have made it clear that HIV prevention services must focus on ‎ these populations. According to expenditure statistics, less than 1% of HIV expenditure was for sex ‎ work-related interventions. Budget optimization with Optima suggests that if about 4% of the annual budget could be spent on comprehensive, quality‎ services‎ for‎ FSWs,‎ Niger’s‎ future‎ HIV‎ ‎ ‎ incidence,‎ AIDS‎ morbidity‎ and‎ mortality‎ could‎ be‎ minimized‎ and‎ the‎ Government’s‎ HIV‎ policy‎ objectives could be achieved.‎ ‎ In the past, especially within the regional SIDA-3 project, FSW HIV prevention service provision and ‎medical follow-up was provided through specialized, adapted services. One such service site has ‎remained, and is in Niamey, and while integration and mainstreaming of services for key populations is a ‎reasonable policy, it has not worked well in Niger. Only 41% of FSW in Niamey are enrolled in the ‎ “Suivi‎ medical”,‎ and‎ these‎ are‎ predominantly‎ foreigners‎ who‎ enroll for administrative purposes. ‎ The‎variation‎in‎FSW’s‎HIV‎prevalence‎in‎different‎geographical‎areas,‎ ranging from zero to 31%, is ‎ largely unexplained, although proximity to Nigeria seems to be an important risk factor. A history of ‎ violence is associated with highly increased HIV risk (32%), and perpetrators can even include the police who should protect FSWs. Well over half of surveyed 3 UNDP, Understanding and acting on critical enablers and development synergies for strategic investments, 2012. http://www.unaids.org/en/media/unaids/contentassets/documents/unaidspublication/2012/201211_UNAIDS_UNDP_ Enablers_and_Synergies_en.pdf Executive Summary xv FSWs are divorced (others are widowed or separated), pointing to sex work as a livelihood option and family support.‎ for women who are lacking protection ‎ Importantly, sex for money or gifts is not limited to FSWs but may be a frequent transaction among other female key population members, as well as among young females who practice transactional sex. Female defense ‎ and security personnel are many times more likely to be HIV positive than their male colleagues ‎ (7.7% versus 0.1%4),and the same increased HIV risk applies to female teachers, prisoners and ‎ labor migrants. Male circumcision partially protects men from infection, while women ‎ experience much higher biological vulnerability to HIV. Female circumcision (2% of adult females have experienced ‎ genital mutilation, according to the Niger Demographic and Health Survey [NDHS] 2012) can promote the spread of ‎ infections through the use of contaminated objects, and an increased risk of bleeding during sex ‎ can amplify the risk of HIV transmission. The increased prevalence of herpes in women who have ‎ undergone female genital mutilation could also add to the risk of HIV infection.5 FSW data attest to a very rapid acquisition of HIV once females enter the sex industry, with ‎ street prostitution less safe than brothel-based sex work. Male sex work is little documented, and MSM are the least understood key population in Niger. Over half of surveyed MSM get paid for sex. ‎ Some MSM are married and some identify as bisexual. The MSM epidemic—of unknown scale— is not ‎selfcontained and MSM related transmission extends through sexual networks to the general ‎population.‎ This analysis shows that SWs are a population which is currently underserved and concludes that more investment is required for SW-related interventions. An investment toward preventing not only direct ‎ but also downstream infections linked to sex work is expected to have a large impact future course of the HIV epidemic, as well as on the human and financial cost of AIDS. ‎ on the ‎ Recommended policy actions 10. In order to guide programming and targeting of SW interventions, better strategic information on ‎ FSW, female key population members, and high risk (soliciting) MSM should be collected—‎ ongoing FSW mapping and site identification work needs to be further strengthened by taking a ‎ programmatic science approach. More needs to be known about FSW typologies, the ‎ venue profiles where they find clients, hard-to-reach FSW, sexual risk behaviors—including ‎ transactional sex among female defense and security personnel, female teachers, female ‎prisoners and female labor migrants (through oversampling of females in bio-behavioral surveys ‎and qualitative studies), as well as young urban women.‎ 11. The government, in partnership with non-governmental organizations (NGOs) and the private sector, should offer evidence-informed ‎ services that meet the specific needs of sex workers—an important part of improved service ‎ provision is demand creation among SW through strategies such as specially trained peer ‎ educators who can gain access to harder-to- reach high risk SWs (due to their ‎ young age, working conditions as clandestines, and lack of protection). The ongoing scale-up of SW ‎ interventions must work towards ambitious service targets that extend beyond the numbers of SWs ‎ known at the outset, reflecting successful demand creation and outreach to clandestine SWs. ‎ Police and security forces play an important role in the implementation by ensuring the protection of SWs against discrimination and acts of violence, and hence, contributing to the ‎ equitable access by SWs to all health 4 Small sample sizes that warrant more investigation, and female oversampling in future BSS activities. http://www.who.int/reproductivehealth/topics/fgm/health_consequences_fgm/en/‎ 5 xvi Niger’s‎HIV‎response: Targeted investments for a healthy future services. Police and security forces must be sensitized on HIV and on ‎ their role in the response to the epidemic, including the handling of cases of sexual violence. ‎ 7. Improving ‎ the efficiency of implementing Niger’s antiretroviral treatment program through ‎better integration with health services Conclusion: Access to ART is insufficient and AIDS-mortality is high; shifting resources to the ART ‎ program will reduce future AIDS deaths and disease, and new HIV infections By the end of 2012, the estimated ART coverage was 13% in children and 51% in adults, that about 14,000–15,000 HIV infected Nigerians who were eligible were not receiving indicating ‎ ARVs. ART coverage is estimated at only 45% in men and 57% in women suggesting that men are more ‎likely to miss out on treatment. This is also ‎ linked to very low HIV testing rates in men of 2–3%, and some demographics indicate that HIV testing is ‎ particularly low despite elevated HIV prevalence levels in —men in Diffa and Dosso, and in men above 30 ‎ years of age.‎ Low ART access leads to AIDS deaths. The rapid decline of HIV prevalence from 2006 to 2012, ‎ particularly in urban men (1.3% to 0.5%), and rural women and men (0.5% to 0.2%, and 0.6% to ‎ 0.3% respectively) indicates high AIDS mortality. In countries with high ART access and retention, HIV ‎prevalence tends to stabilize or even increase—although decreases in HIV prevalence are ‎ desirable, they should not arise from excessive AIDS mortality. Anecdotal evidence suggests that people delay accessing ART, especially men, who contact health ‎ ‎ services less frequently. There may be challenges to the retention of ART patients (no documentation identified), and ‎ ARV availability at decentralized level has not always been maintained (while there is a surplus of ARVs at central level). Many ART patients will experience food insecurity and malnutrition—the ‎ result of a vicious cycle of no farming, no harvest—which can impair drug adherence and treatment ‎ success. According to Spectrum, an estimated 3,360 PLHIV died in 2012 due to an AIDS-related ‎ cause, and the total cumulative number of AIDS deaths in Niger up to end 2012 is estimated at ‎ 68,730. The insufficient scale-up of the ART program and weaknesses in HIV case finding have to be seen ‎in‎the‎context‎of‎Niger’s‎health‎system.‎Over‎half‎of‎all‎residents‎live‎outside‎the‎ 5 kilometer ‎ catchment area of access to health facilities (however, in Niamey nearly all residents are within a ‎ kilometer-5catchment area). Newborn, nutritional and maternal causes such as diarrheal diseases, lower ‎ respiratory infections, and protein-energy malnutrition as well as several communicable diseases ‎ including malaria, meningitis, measles and tetanus, remain the top causes of morbidity and ‎ mortality. Health sector capacity is insufficient, compounded by the fast population growth which doubles the population every 23 years. However, AIDS (and malaria) are the leading ‎emerging diseases in Niger according to the comprehensive international Burden of Disease Study, ‎and therefore requires attention. ‎ The Optima analysis supports a 50% HIV budget allocation to the ART program (which includes ‎ ARVs, laboratory monitoring, psychological and nutrition support, out-patient and in-patient ‎ services), almost double the 26% allocation in 2012. While the roll-out of the ART program must be ‎ national, a phased scale-up may be justified given the large concentration of PLHIV in four (circa 70% of PLHIV in the four regions covering a quarter of the country surface area).‎ regions ‎ Executive Summary xvii Recommended policy actions 12. More resources need to be allocated to the ART program to optimize the impact of treatment ‎on HIV incidence and DALYs—allocating half of the budget to the ART program would allow for investment in program infrastructure (like diagnostic equipment); human resources (including ‎capacity building, and task shifting activities); supply chain strengthening (including management ‎ information systems, and ARV prescription decentralization); and patient and adherence monitoring ‎ through community-based health workers. ART program efficiency will be greatly improved through closing these resource gaps and through better data along the ART cascade‎ .‎ 13. With more investment in the ART program, pre-ART services could be innovated and strengthened, and the treatment threshold could be raised to initiate more PLHIV on treatment, ‎ especially men—diagnosing people as HIV positive and then losing them again until their health has ‎ seriously deteriorated is a great inefficiency, and contributes to ongoing HIV transmission. There ‎ are good linkage and retention practices in pre-ART care6 that are documented internationally which can be‎applied‎to‎Niger’s‎context. Gradually switching from the prevention of mother-to-child-transmission (PMTCT) options A and B to option B+ will ‎ further help scale up ART.‎ 14. Institute systematic monitoring of CD4 count at treatment initiation, separately for men and ‎women, to track progress in the implementation of treatment policy (threshold 350 or potentially ‎500)—although CD4 capacity is still being strengthened (and is a priority action), ‎ systematic recording of this key ART access indicator must be started when CD4 counting is ‎done at treatment initiation, and the indicator must be used for performance monitoring (an ‎ increasing CD4 count at initiation shows earlier and better access to treatment). Resolving the technical bottlenecks that are associated with low testing numbers, such as handling the test machine capacity, is an ‎ important efficiency improvement. 15. Provider-initiated counseling and testing must be selectively scaled up for HIV case finding—this ‎primarily concerns services for STI clients, where HIV yields are higher than in the ‎ general population (TB clients are already being tested for co-infection). Once a person has been identified as ‎ HIV positive, post-test counseling should focus on getting a commitment from the client for periodic ‎ monitoring visits so that the client can be initiated on treatment in agreement with ART eligibility ‎ guidelines (a higher CD4 treatment threshold PLHIV earlier).‎ would enable health providers to initiate more ‎ 8. Improving ‎ the implementation efficiency of the prevention of mother-to-child transmission ‎ program through better integration with health services Conclusion: Elimination of vertical transmission is challenged by poor health-seeking behavior of ‎ pregnant women and multiple limitations of the health system Approximately half of all new HIV infections are vertically transmitted. Niger ‎ has committed to eliminating mother-to-child transmission (MTCT), and to keeping mothers alive. This global goal requires a significant scaling up of HIV services, accompanied by intensified sexual ‎ and 6 http://www.aidstar-one.com/focus_areas/care_and_support/resources/report/PreART_Linkage_Retention xviii Niger’s‎HIV‎response: Targeted investments for a healthy future reproductive health interventions, community mobilization and rural service deployment in ‎ Niger and elsewhere. ‎ Achieving the virtual elimination of MTCT in Niger has considerable challenges—the MTCT rate needs ‎to be reduced from the rate of approximately 27% in 2013 to below 5%. Although Niger had 651 ‎PMTCT sites in 2012 and saw positive ‎ growth in the use of antenatal services (CPN1 at 83%) and assisted deliveries (29%), less than half of pregnant mothers in need of ARVs received the drugs in 2013. HIV‎diagnosis of exposed infants is almost nonexistent at 1%. There are many missed family ‎ planning (FP) opportunities that are also missed opportunities to reduce the vertical transmission of HIV. ‎More than one-third of all non-users of FP visited a health facility over the past 12 months, but did not discuss FP during that visit. These missed FP opportunities are especially a concern in women ‎ aged 30–44 years (who have an elevated HIV prevalence compared to younger women), and women living in ‎ Niamey (female HIV prevalence of 1.5%).‎ The success of PMTCT in Niger relies heavily on improving the demand of pregnant women for maternal ‎and neonatal health services. The causes of the high number of those lost-to-follow-up multifaceted and complex: many women‎are‎dependent‎on‎their‎husband’s‎approval‎to‎seek‎ are ‎ and ‎ obtain health services; cultural reasons can prevent ‎ both men and women from seeing pregnancy as a “disease”‎state‎that‎warrants‎the‎‎ intervention of health care providers; the stigma of being HIV positive can prevent women from ‎ getting their HIV test results or obtaining the treatment for themselves or their baby; and the cost of ‎ accessing antenatal care and assisted delivery—costs can be overwhelming even if is just for transport or a lost day ‎ of work, and especially if the perceived benefits are seen as limited. One field study on ‎ maternal health services in Niger also noted that staff attitudes towards patients were often ‎demeaning.7 The Government has reviewed the PMTCT situation in detail and identified inefficiencies, ‎ constraints and priority actions. One of the main delivery solutions was a scheme where female health workers were assigned at the community level (“médiatrices‎communautaires”), such as for example, in the Arlit district. The recent positive Rwanda experience that involved 15,000 ‎ community health workers suggests that investing in linking people to services provides direct ‎ health outcomes. Community based workers and community networks can play a role in information, ‎ education, demand creation, mobilization and in the provision of basic services. The optimization analysis suggests that increased investment in PMTCT will help minimize new HIV ‎infections, AIDS deaths and disease. Spending 18%–20% of the HIV budget on PMTCT instead of 11% as spent in 2012, will lead to best reductions in vertically transmitted infections and avert the maximum amount of‎ ‎ DALYs. It is estimated that an additional 1,400 vertical HIV infections could be averted over a ten year period if ‎ the currently acquired NSP resources were optimally allocated. Using the Option B+ PMTCT ‎ regimen could help simplify the service cascade and improve the uptake of PMTCT by removing ‎ the need for CD4 counting for treatment initiation. By decreasing the number of ‎ necessary interactions with the health system, it theoretically helps to scale-up PMTCT (and would directly ‎ contribute to ART targets). The simplified infant prophylaxis protocol and a unified regimen that ‎ can be used for the entire adult population would also simplify procurement and supply chain ‎ management.8 Of course, switching to Option B+ also entails significant financial and ‎ organizational investments that need to be thoroughly assessed and costed. ‎ 7 Chambers, V., and D. Booth, Delivering maternal health—Why is Rwanda doing better than Malawi, Niger and Uganda? ODI briefing paper 74, May 2012.; Olivier de Sardan, J.-P., Providing public goods: local ‎ ‎ responses to policy incoherence and state failure in Niger. APPP Policy Brief 4. London: APPP,‎2012. 8 UNICEF, Key considerations for options B and B+, July 2012, Chewe Luo. http://www.unicef.org/aids/files/hiv_Key_considerations_options_B.pdf/ Executive Summary xix Recommended policy actions 16. Shift more HIV resources to PMTCT, per the optimization analysis, and use these resources to ‎strengthen systems that are shared between PMTCT, maternal health and family planning—advancement in PMTCT largely relies on progress in the provision of maternal, newborn and child health ‎ (MNCH) and family planning. Unless more women use antenatal care (ANC) and delivery services, PMTCT cannot ‎ be significantly scaled up. Therefore PMTCT resources must contribute to the broader attempt toward ‎ providing better services and building greater demand. A particularly promising shared resource are ‎ community-based health workers (médiatrices) who are tasked with the identification of risk ‎ pregnancies, and who accompany women, children and families enrolled in treatments ‎ (PMTCT regimens, ART, FP, tuberculosis [TB], etc.). The Malawi experience has shown that peer support and ‎ defaulter-tracking are important for minimizing loss-to-follow-up, as are extended counseling and ‎ ongoing support of women, and effective use of the patient 9 monitoring data.‎ 17. Use the knowledge about the barriers, bottlenecks and inefficiencies of the current PMTCT ‎ intervention to make strategic investments, including a phased introduction of the B+ option—more evaluation of demand creation strategies in‎ MNCH and FP need to be done, bearing in mind ‎ that intervention effects will be heterogeneous10‎ (e.g., by literacy level, order of pregnancy, and previous service-user‎ ‎ experience). Regular quality monitoring of services is essential toward retaining women ‎ during antenatal care and throughout their reproductive years. PMTCT option A should be ‎ completely phased out in favor of the more effective option B. Pilot studies of the B+ regimen—a promising ‎ option in settings with scarce CD4 cell count testing capacity and high fertility—should be ‎ conducted together with systematic and ongoing reviews of retention, using medical record data. Option B+ ‎ needs to be fully costed, including community-based activities deemed necessary to make ‎ life-long treatment of HIV-infected mothers work over time.‎ 18. The PMTCT program must expand its integration at all levels—this study found that PMTCT ‎ services would benefit from a higher level of integration with other services and initiatives, ‎ especially at the community level: to improve access to services for women; better facilitate referrals and ‎ follow-up between services; increase efficiency across the PMTCT cascade; and for health supervision services‎ to fully integrate PMTCT into the broader supervision visits. Finally, an ‎ improved and integrated supply chain management for ARVs and for HIV test reagents is required ‎ to ensure ANC clients can be diagnosed for HIV and treated at all PMTCT sites without ‎ interruptions due to shortages. ‎ 9 Shaffer, N., et al., Option B+ for prevention of mother-to-child transmission of HIV in resource-constrained settings: great promise but some early caution. AIDS 28:599–601. 2014.; Tenthani, L., et al., Retention in care under universal antiretroviral therapy for HIV-infected pregnant and breastfeeding women (Option BR) in Malawi. AIDS 28:589–598, 2014. 10 Takasaki, Y., and R. Sato, Why‎don’t‎the‎poor‎receive‎antenatal‎care?‎Evidence‎from‎incentive‎and‎information‎ experiments in rural Nigeria, June 21, 2013. xx Niger’s‎HIV‎response: Targeted investments for a healthy future 9. Financing the future HIV response in Niger Conclusion: There is not sufficient funding for implementing the national HIV strategy, however, the total ‎ funding needed is not as much as anticipated provided that the funds are directed to the most ‎ allocatively-efficient programs, that the programs are implemented in the most technically efficient manner, and that they are integratedwithin‎ ‎ the broader health systems ‎ Even if HIV resources were optimally allocated (as suggested in 2.), a financing gap would still remain. For‎Niger’s‎NSP,‎the‎Government’s‎anticipated‎financing‎shortfall will be Euro 44.9 million (USD 60.4 million) for the remaining 2014–17 period. This implies that by December 2013, only ‎ 35% of the total ‎ NSP-projected funding for the four year period was committed from either internal external sources. But, if funding was optimally allocated, the Government ‎ or ‎ may not need this amount of funding in the future.‎ In the past and from 2007 to 2011, 90% of the financing response to HIV was driven by external funding, with the most important sources in recent years being the Global Fund, UNICEF, the Government of Germany (Kreditanstalt ‎ für Wiederaufbau KfW) and World Bank. In 2012, the share of external funding declined to 67% ‎ and public spending dramatically increased through a World Bank IDA loan, which is ‎counted‎asGovernment of Niger funding. The average costs of the national HIV response (2007–12 data) ‎ was Euro 9.7 million (USD 12.6 million), or 0.24% of GDP, years.‎ bearing in mind that there were large fluctuations between ‎ Annual spending under the provisional NSP budget and the 2012 resource needs model is ‎ projected to increase steeply from Euro 10.5 million (USD 14.1 million) in 2014 to Euro 20.2 million ‎(USD 27.2 million) in 2017. The scaling-up of HIV prevention and treatment envisaged under these ‎ provisional investment plans would result in a steep decline in HIV incidence and mortality, and an ‎ increasing number of people on ART. The decline in HIV incidence would ‎ contribute substantially toward containing the long-term costs of the HIV program. It is estimated that one new HIV ‎ infection in 2013 results in a future cost of about Euro 2,500 (USD 3,400). This level of mortality on ART. HIV incidence declines in 2013–17 alone would cost will vary with the ‎ result in reduced ‎ long-term spending commitments equivalent to a saving of Euro 6.3 million (USD 8.4 million). The ‎ estimates in this analysis imply that—in addition to the obvious health and social gains—the ‎ financial savings achieved by increased HIV investments per the current NSP would offset about ‎ one-third of the additional costs of the scaling-up of prevention and treatment efforts. While the ‎costs of the national HIV response per the NSP are projected to increase in absolute terms, the ‎ costs are expected to peak at 0.29% of GDP in 2017, and decline to 0.22% of GDP in 2030. ‎ Private out-of-pocket health spending is relatively large, considering the low level of wealth in ‎ Niger. In 2011, health spending accounted for 5.1% of GDP (Euro 225 million or USD 303 million) (see Figure 6), which was dominated by the public sector (57% of total) and private out-of pocket spending ‎ (39% of total). Out-of-pocket spending accounted for 95% of household health spending. About ‎one-third of all health spending in Niger is externally financed (which is a lower external dependency than ‎HIV spending). ‎ Executive Summary xxi Figure 6. Health expenditure by source of financing, Niger (1995–2011)‎ In 2011, spending on HIV care and treatment accounted for about 2.7% of public health spending ‎ (1.6% of total health spending), which is a considerable share given the many competing needs of Niger’s‎health‎sector.‎A‎recent‎World‎Bank‎report‎on‎financing‎for‎‎ financing ‎ development post-201511‎ points to four foundational pillars of development financing: domestic ‎ resource finance. ‎ mobilization; better and smarter aid; domestic private finance; and external private ‎ Recommended policy actions‎ 19. Review the NSP budget in the light of this analysis and acquired HIV funding—it is ‎ recommended that budget adjustments be made to ensure financing of high impact ‎ interventions. The Government should plan for budget adjustments now in 2014, identify savings and ‎alternative financing mechanisms, and plan for a gradual decrease in ‎ funding of HIV interventions which have low epidemic impact.‎ 20. Evaluate alternative funding sources especially those that generate permanent resource streams and leverage the private sector—the feasibility and potential revenue from additional resource ‎streams should be investigated, for example: intensify private sector mainstreaming (comprehensive ‎ HIV/AIDS workplace programs initiated by HIV business coalition members); explore growing the ‎ existing social health insurance (SHI) further, where appropriate, to create fiscal capacity by shifting expenditure ‎ from the Government budget (almost 3% of Niamey residents are already members of a SHI); ‎ reinforcing public sector mainstreaming; and other measures. The results based financing (RBF) scheme planned for the ‎health sector represents an important opportunity to ensure financing for ARV-based ‎ interventions, and to stimulate both demand and supply (see Burundi ‎ experience).‎ 21. The burgeoning private sector is creating tax revenue (but also new or increased risk contexts), so the ‎ Government should consider reinvesting a percentage of these fiscal revenues into HIV and health ‎programs—this process has started, and current analysis demonstrates the importance of a ‎ predictable and sustainable resource base for HIV 11 World Bank, Financing for development post-2015, October 2013. http://www.worldbank.org/content/dam/Worldbank/document/Poverty%20documents/WB-PREM%20financing-for- development-pub-10-11-13web.pdf xxii Niger’s‎HIV‎response: Targeted investments for a healthy future services. Niger already instituted an airline levy, ‎ but other innovative financing modalities to increase domestic financing could be investigated to sustain the HIV response.‎ 22. Integrate HIV/AIDS into any future health schemes such as Results Based Financing (with PMTCT and ‎ ART in the essential service package), or any future Universal Health Coverage scheme—while ‎ HIV/AIDS is not a leading cause of burden of disease in Niger, it is nevertheless a leading emerging ‎ cause, and services must be available to manage HIV/AIDS and co-morbidities such as poor ‎nutrition in an integrated manner. A vertical approach to HIV is ineffective in such a ‎ context and therefore HIV and health investment must be built in a concerted and integrated way. ‎ 23. When advocating for HIV and health investment, the Government of Niger should consider the ‎ HIV epidemic potential within the context of unfolding demographic and socio-economic ‎ changes—epidemic drivers and co-factors such as the rapid growth of urban populations, increased labor ‎ migration and high fertility may erode some epidemic be complacent regarding HIV/AIDS. ‎ gains, therefore the country cannot ‎ Section 1. Introduction What is the context of this analysis?‎ Niger’s‎ morbidity‎ and‎ mortality‎ statistics‎ are‎ still‎ dominated‎ by‎ communicable‎ diseases‎ while‎ there ‎ is, until now, a relatively low burden of disease attributable to non-communicable diseases. The ‎macroeconomy is experiencing positive development, however, a large part of ‎ the population are still living in poverty (56% were living on less than USD $1.25/day in 2008). The majority of the population has poor access to health facilities (53% live further than 5 kilometers from a health service). 40% of all health spending is out-of pocket, despite the low income levels of most people. Despite Niger being a low HIV prevalence and comparatively low HIV/AIDS burden country, HIV/AIDS is an emerging and significant cause of DALYs in recent years —see Figure 7 ‎ ‎  The only rapidly growing cause of DALYs between 1990 and 2010, other than HIV/AIDS, was ‎ malaria ‎  TB is still a public health problem. However, latest World Health Organization (WHO) statistics show a falling incidence trend of 104/100,00012. ‎ ‎ Figure 7. Shifts in leading causes of DALYs in Niger (1990–2010)‎ Source: Institute of Health Metrics and Evaluation, Human Development Network, The World Bank; ‎ The global burden of disease: Generating evidence, guiding policy—Sub-Saharan Africa Regional ‎ Edition. Seattle, WA: IHME, 2013.‎ 12 https://extranet.who.int/sree/Reports?op=Replet&name=%2FWHO_HQ_Reports%2FG2%2FPROD%2FEXT%2FTB- CountryProfile&ISO2=NE&LAN=FR&outtype=pdf 1 2 Niger’s‎HIV‎response: Targeted investments for a healthy future In 2011, HIV/AIDS consumed an estimated 2.7% of public health spending (2011 data), and has therefore become a significant cause of health expenditure in Niger.‎ ‎ At the international level, the HIV funding landscape—and indeed the HIV financing paradigm—has changed ‎  After a decade of unprecedented HIV investment, external HIV financing has plateaued leading increasingly to domestic resources being expected to sustain and scale-up national responses13 ‎ ‎‎  Over the last three decades, it has become apparent that structural factors—such as ‎ poverty, limited livelihood options, inequality, and violence—as well as inadequacies in the health system,‎ have compounded and sustained the HIV epidemic, undermining the effectiveness and scale-up of proven ‎HIV interventions14‎  Concurrently, HIV investments have had spill-over effects that have impacted well beyond HIV, having contributed to health sector strengthening and non-HIV outcomes15 ‎  As a consequence, current discussion has shifted from whether an investment is within (or ‎ beyond) the remit of the HIV program, to a broader view on the multiple interactions between health, and development outcomes‎ HIV, ‎  Beyond 2015, future international efforts will increasingly be directed towards achieving maximum ‎ attainable health outcomes equitably and‎ through universal coverage approaches. The thought is that effectively embedding ‎ HIV responses into broader national priorities, will further encourage ‎domestic ownership and sustainability16 ‎  This report adopted some of these new insights by relating Niger’s‎ HIV‎ response‎ and‎ the health sector context and its financing, and included the DALY17‎metric in the financing to ‎ optimal HIV ‎resource allocation modeling ‎ Why is this analysis needed?‎ Although HIV is not the major burden of disease in Niger, it is an important emerging source of disease ‎ burden. Implementing, integrating, and allocating resources more effectively will potentially save time and money, and free up resources to address other key development challenges.‎ 13 Meeting the investment challenge: tipping the dependency balance. Geneva. UNAIDS; 2012; Hecht, R., et al., Financing of HIV/AIDS program scale-up in low-income and middle-income countries, 2009–31. The Lancet, 376:1254–1260. 2010; Lule, E., and M. Haacker, The fiscal dimension of HIV/AIDS in Botswana, South Africa, Swaziland, and Uganda, Washington, DC: The World Bank, 2012. 14 Remme, M., et al., Financing structural interventions: going beyond HIV-only value for money assessments. AIDS, 28:425-434. 2014. 15 Walensky, R. P., and D. R. Kuritzkes, The‎impact‎of‎the‎President’s‎Emergency‎Plan‎for‎AIDS‎Relief‎(PEPfAR)‎ beyond HIV and why it remains essential. Clinical Infectious Diseases, 50:272–275; Samb, B., et al., An assessment of interactions between global health initiatives and country health systems. The Lancet, 373:2137– 2169, 2009. 16 Global Health 2035: a world converging within a generation conference, 16 January 2014. New York. 17 DALY (Disability adjusted life years) is a metric which allows comparison across diseases. One DALY can be thought of as one lost year of "healthy" life. The method uses disability weights, discounting and age weighting. Introduction 3 Niger’s‎HIV‎response‎is‎heavily‎dependent‎on‎external‎funding  Niger has been heavily reliant on international support, with 95% in 2009, 93% in 2010 and 2011 and 67% in 2012 of HIV program funds covered by external finance (the decrease in 2012 is largely due to the World Bank loan)  The primary source of international funding in Niger has been the Global Fund. However, in Niger’s‎HIV‎expenditure‎was‎ supported by the Global Fund compared to 2012, only 18% of ‎ 37%–51% in the three years prior to 2012 Niger may face shortfalls in HIV funding during the NSP period  The African Union Special Summit in Abuja in July 2013 led to a renewed commitment by ‎ Member States to take decisive action toward the elimination of HIV/AIDS by 2030, through increased ‎ access to prevention programs, eliminating mother-to-child transmission of HIV 18 mothers alive, and honoring the‎earlier‎“Abuja‎Commitments”‎ while keeping ‎  In line with its commitments,‎Niger’s‎NSP‎2013–17 envisages enacting a service ‎ scale-up to meet ambitious HIV service coverage targets, however, the funding that has been acquired‎ ‎ falls well short of estimated resource needs  Niger’s‎Global‎Fund‎support‎is‎based‎on‎the‎Transitional‎Funding‎Mechanism‎until‎2015,‎and‎ the Kreditanstalt für Wiederaufbau (KfW) ‎ funding is only assured until 2014. The World Bank support project 2012–16 focuses its ‎investments in specific geographical and service areas (prevention of HIV among sex workers and ‎ their clients, HIV testing and treatment, PMTCT, as well as management and coordination of the ‎ HIV response, and piloting output-based financing of HIV services) Even though HIV incidence has fallen, AIDS care and support costs are considerable in a that has multiple investment priorities ‎ health system ‎  In 2011, health spending accounted for 5.1% of GDP (Euro 225 million/USD 303 million), and about ‎ one-third of all health spending was externally financed (WHO, 2014)  HIV care and treatment spending accounted‎for‎about‎2.7%‎of‎Niger’s‎public‎health‎spending‎ 2011‎ in ‎  There is an incessant flow of new HIV infections and it is estimated that Niger had 46,910 ‎ people living with HIV in 2012 (Spectrum, 2013) and spent Euro 3.38 million (2.22 billion CFA/USD ‎ 4.41 million) on HIV/AIDS treatment in the same year  The global goal of eliminating mother-to-child HIV transmission requires a significant scaling up of ‎ HIV services accompanied by intensified sexual and reproductive health interventions, community ‎mobilization and rural service deployment‎ 18 In April 2001, heads of state of African Union countries met in Abuja and pledged to set a target toward allocating at least 15% of their total annual budget to health. At the same time, they urged donor countries to "fulfil the yet-to-be met target of channeling 0.7% of their Gross National Product as official Development Assistance to developing countries". 4 Niger’s‎HIV‎response: Targeted investments for a healthy future Now is the right time to promote a smart and sustained investment into Niger’s‎ HIV response that strengthens, integrates with, and complements Niger’s‎ broader‎ health‎ priorities  Globally‎ the‎ notion‎ of‎ “shared‎ responsibility”‎ in funding the HIV response has gained momentum. ‎ While the Global Fund has restructured its grant mechanisms towards strategic investment ‎for maximum impact, political leaders have agreed to scale up the mobilization of domestic ‎resources to strengthen the health system, and ensure diversified and sustainable health (July 2013 Abuja Summit) ‎ financing for ‎ The New Funding Model of the Global Fund is designed to produce a greater impact on the three ‎ diseases, to improve funding predictability and reward ambitious vision. Funding is‎ determined‎ by‎ a‎ country’s‎ income‎ level‎and‎ official‎ disease‎burden‎ data.‎ The‎ eligibility ‎ level of allocations ‎ are predetermined based on disease burden, income levels and other criteria such as previous ‎ grant performance, impact, increasing rates of infection, 19 absorptive capacity, and risk.‎ ‎‎  In 2012, Niger has already increased its investment in HIV, spending three times more on in the previous five years combined ‎ HIV than ‎  At a time when the global economic environment is stagnating,‎Niger’s‎economy grew to 11% in 2012, thanks to oil, uranium and agricultural production (International Monetary Fund [IMF] increased in 2012 relative to 2011, ‎As‎part‎of‎Niger’s‎ data), and the country’s‎fiscal‎revenues ‎ current economic strategy, the government intends to ‎ create fiscal space for development spending (IMF, April 2013). The new national HIV strategy ‎ emphasizes the need for diversified revenue streams to finance the growing HIV response, but ‎ much of this may depend on whether HIV/AIDS prevention, treatment and support are deemed to be smart investments ‎ What were the objectives of the analysis?‎ The overarching objective was to provide the Government of Niger with a comprehensive ‎ investment case analysis of HIV financing needs, and investment opportunities in the context of ‎ burden of disease priorities, health financing and health system strengthening, in order to apprise a‎HIV response that is efficient, impactful, cost-effective, and financially sustainable.‎ The specific objectives were: ‎  Understand the broader health priorities and health financing landscape  Within the health priority context, estimate the impacts, cost-effectiveness and returns on ‎ investment of past HIV expenditure  Review evidence of HIV implementation efficiencies and determine where efficiency gains ‎ could be made to save costs, or scope increased without compromising quality‎  Provide an integrated analysis of how, during the ‎current NSP, HIV financing can be best allocated in a way that minimizes HIV incidence, AIDS morbidity and mortality, and reach the 2017 epidemic/health impact targets ‎ ‎ 19 http://www.theglobalfund.org/en/fundingmodel/updates/2014-02-04_Eligibility_List_for_2014_now_available/; http://www.theglobalfund.org/en/fundingmodel/allocationprocess/ Introduction 5  Link the evidence on allocative and programmatic efficiency and intervention effectiveness to an analysis ‎ of‎ thefinancial sustainability of the HIV response, based on domestic fiscal resources and ‎evolving availability of external funding  Determine what the long-term financial consequences and savings of HIV interventions in Niger are, and ‎ potentially what future health care costs can be saved through different policy makers‎ scenarios, to guide policy-‎ Using the Optima model, the analysis (see below) posed the following questions:‎ 1. What‎is‎the‎cost‎of‎meeting‎Niger’s‎2017‎coverage‎targets?‎ 2. How much money is required to reduce HIV ‎ incidence by 50% in the mid-term, if allocated optimally among the basic programs?‎ 3. With the available 2014–17 budget, what is the optimal allocation among the basic HIV programs for best reduction in HIV incidence and DALYs over that period, including follow-on 2025? ‎ effects until ‎ a. Could accelerated ART scale-up contribute to the 2017 goal? ‎ b. Could a change in PMTCT protocol to B+ contribute to the 2017 goal?‎ What methods were used?‎  A Desk review of available data and literature was conducted, comprising: HIV resource tracking data (NASA); ‎ National health account data; macroeconomic data; HIV prevalence and sexual behavior reports; ‎ Spectrum estimations; data from the Resource Needs Modeling; reports on key populations; HIV ‎ policy documents; and HIV program implementation data  Optima (formerly Prevtool), a mathematical model developed by the University of ‎ New South Wales (UNSW) and the World Bank (WB) for estimating HIV incidence in sub-populations was used to calculate the HIV impact and ‎ cost-effectiveness of HIV spending; and optimal resource ‎allocation toward achieving best results in the prevention of new HIV infection and HIV-related morbidity and ‎mortality—see Annex 1 for details ‎  Implementation of a financial commitment analysis that has three building blocks: (1) an ‎ epidemiological component; (2) a costing framework for projecting the costs of the national response ‎ to HIV; and (3) an economic framework to estimate the costs incurred by new infections. The ‎ epidemiological component is harmonized with Spectrum for consistency. The costing framework projects the costs of the national response to HIV/AIDS through calculating the epidemiological estimates and specified coverage rates and the number of people ‎receiving HIV/AIDS services. ‎ Unit costs are informed by past expenditure patterns (as documented in the national AIDS spending assessment [NASA] report), budget ‎ data, and data from the Resource Needs Model—see Annex 2 for details  Monetary amounts were expressed in Euros, using the fixed exchange rate of 1 Euro = and using 503.1 for CFA/USD ‎ 655.957 CFA, ‎ How are the study findings presented in this report?‎  The Executive summary presents the main findings by theme and provides the full conclusions and ‎ recommendations that are derived from this analysis. The recommendations cover eight areas: (1) ‎ Epidemic dynamics; (2) Impact and cost-effectiveness of past HIV 6 Niger’s‎HIV‎response: Targeted investments for a healthy future investments; (3) Impact gains ‎through better allocative efficiencies; (4) Geographical targeting of priority programs; (5) Improving ‎ the efficiency of programs for female sex workers; (6) Improving the efficiency of the antiretroviral ‎ treatment program; (7) Improving the efficiency of the prevention of mother-to-child-transmission ‎ program; and (8) Financing the future HIV response in Niger  Section 1 (Introduction) presents the context and rationale of the analysis, the objectives and the ‎ methodological approach taken  Section 2 (Development context, health priorities and health system challenges in Niger) ‎ summarizes the economic and human development state of affairs, national development policy ‎ framework, health spending, burden of disease data, as well as the provision and services.‎ reach of health ‎  Section 3 (The state of the HIV epidemic) reviews the epidemic situation, course and drivers, and ‎the distribution of HIV across population strata. Section 3 provides a brief characterization of key ‎ populations with population sizes and locations, and provides comments on risk factors and vulnerability. It ‎ presents model estimations on HIV incidence and sources of new infections.‎  Section 4 (HIV financing, HIV services and programs, and their efficiency) reviews the data on HIV ‎ expenditure and funding sources, and what these investments have delivered in terms of HIV ‎ program development and scale-up. It reviews service linkage and integration, unit service utilization. ‎ costs and ‎  Section 5 (Impacts from past HIV expenditures and best resource allocations in the current ‎ strategic plan period) presents modeling results of the HIV impact and cost-effectiveness of six ‎years of HIV investments covering 2007–12. It then presents modeling results for optimized ‎resource allocation until 2017, including scenarios for best HIV incidence reduction and maximum ‎ number of DALYs averted. The analysis makes assumptions that are based on a range of total HIV resources available, while taking savings from possible efficiency gains into account.‎  Section 6 (Financing needs and fiscal space implications of HIV investments) presents results from ‎ the fiscal model that was applied to Niger. It summarizes the financing needs and gaps until 2017, the HIV ‎ prevention and mortality impacts of investments under the current NSP, and the longer-term fiscal ‎effects of HIV investments. ‎ Section 2. Development context, health priorities and health system challenges in Niger What is the state of economic and human development?‎ Macroeconomic situation Niger is one of the poorest countries in the world with a largely agrarian and subsistence- based ‎economy that is frequently disrupted by extended droughts common to the Sahel region of Africa  Niger is landlocked, arid and vast, covering an area of 1,267,000 km2. The closest port is the Cotonou (in Benin) about‎ 1,000‎ km‎ away.‎ Thus,‎ most‎ of‎ Niger’s‎ foreign‎ trade‎ is‎ Port of ‎ transport corridors being Benin,‎Togo‎and‎Côte‎d’Ivoire‎via‎ conducted overland, with the main ‎ Burkina Faso ‎  The economy is dependent on the agro-pastoral sector (40%) and the tertiary service sector (43%)‎  After a year of food shortages, good harvests in 2010 helped the economy recover and grow, but this was interrupted by another ‎ period of drought in late 2011. However, growth was sustained by sizeable ‎investments in the mining and oil sector ‎ Following the trend in the region, Niger has benefitted from positive economic growth and ‎ increased fiscal revenues thanks to new mineral resources  Over the past decade, extractive industries have fuelled economic growth in Africa which, together with ‎continuous demand from China has inflated export prices.20‎ In‎ 2011,‎ Africa’s‎ production accounted for ‎three-quarters‎ of‎ the‎ world’s‎ platinum‎ supply,‎ half‎ of‎ its‎ diamonds of global gold and uranium supplies.21 New exploration has and chromium, and up to one-fifth ‎ identified larger reserves than previously known ‎ which is attracting additional foreign investment ‎  The IMF reported that‎Niger’s‎ economic growth had reached 11% in 2012, thanks to an oil project coming on-stream, higher uranium production and a rebound in agricultural harvest. Over the ‎next 5 years, average real GDP growth is projected to be 6.3% annually (and at 3.1% GDP per ‎ capita)—see Table 1  Fiscal revenues increased in 2012 relative to 2011 but, due to weaknesses in customs and oil revenue, were expected to fall short of program targets (IMF, April 2013)‎ ‎ 20 Africa‎progress‎report‎2013:‎Equity‎in‎Extractives:‎Stewarding‎Africa’s‎natural‎resources‎for‎all.‎ http://africaprogresspanel.org/wpcontent/uploads/2013/08/2013_APR_Equity_in_Extractives_25062013_ENG_HR.p df 21 Mineral Yearbooks, USGS, 2011. http://minerals.usgs.gov/minerals/pubs/country/2011/myb3-sum-2011-africa.pdf 7 8 Niger’s‎HIV‎response: Targeted investments for a healthy future GDP per capita is on an upward trend, and inflation is below 3%‎  In 2013, per capita GDP was at Euro 330 or USD 440‎  Despite fluctuations in farm production, inflation, on average, was under control at 2.1%, below the regional limit of 3% ‎ ‎  IMF and World Bank (2013) suggest that Niger faces a moderate risk-of-debt distress. Public debt ‎has declined since 2006 due to restructuring arrangements obtained from the creditor members of ‎ the Paris Club, as part of the Heavily Indebted Poor Countries Initiative and the Multilateral Debt ‎ Relief Initiative. However, over the last few years, public external debt has increased due to the ‎financing of large resource extraction projects  As‎part‎of‎Niger’s‎current‎economic‎strategy,‎the‎government‎intends‎to‎create‎fiscal‎space‎ for ‎ development spending while tackling revenue weaknesses, implementing multiple measures ‎ to rebuild government deposits, employing structural reforms, improving financial development ‎ and maintaining debt sustainability‎ Table 1. Selected macroeconomic data, Niger (2013)‎ Niger: Selected Macroeconomic Data 2007 2008 2009 2010 2011 2012 2013 Proj. Proj. Real GDP growth 0.6 9.6 -1.0 10.7 2.2 11.2 6.2 Real GDP per capita growth -2.9 5.8 -4.3 7.3 -0.9 7.9 3.0 CPI inflation (annual average) 0.1 10.5 1.1 0.9 2.9 0.5 2.1 GDP per capita (Euro) 226.1 257.7 260.6 284.5 292.0 317.5 333.4 GDP per capita (US$) 310.0 379.3 363.0 377.5 406.3 408.3 440.0 GDP (Euro bn) 3.1 3.7 3.8 4.3 4.6 5.1 5.5 (Percent of GDP) Government revenues 22.4 24.3 18.9 18.2 18.1 22.5 24.3 Government expenditures 23.4 22.8 24.3 20.7 19.6 25.1 28.6 Fiscal balance -1.0 1.5 -5.4 -2.4 -1.5 -2.6 -4.4 Public debt 25.4 21.3 28.1 24.0 27.7 28.8 38.7 ‎ Sources: IMF (2013) Note: bn = billion; CPI = consumer price index; GDP = gross domestic product. The political and administrative landscape is characterized by a multiparty system and ‎ decentralization  There are 10 major political parties, a president who is elected by popular vote and a prime minister who is appointed by the president. The new constitution was adopted in October 2010‎  Niger has established a process of decentralization that has transformed the former ‎ regional administrative structures (Départements, Arrondissements) and created Communes ‎ for better local governance  The country is now subdivided into eight administrative regions, 36 Départements and 266 Communes, of which 52 are urban and 214 are rural ‎ ‎ Development context, health priorities and health system challenges in Niger 9 Human development The Nigerien population, estimated at 17.13 million inhabitants in 2012, continues to grow at a very ‎rapid rate of 3.9% (average intercensus 10-year rate), with a total fertility index of 7.6 children per ‎ woman in 2012‎  The rate of demographic growth has remained virtually unchanged in the last 20 years, and at that rate, the population is expected to double every 23 years  In rural areas, women give birth to 8.1 children on average, and in urban areas, 5.6 children (NDHS ‎ 2012). Zinder and Maradi have exceptionally high indices of 8.5 and 8.4 births per woman ‎ respectively. Women with secondary education have on average 3.1 fewer births than women ‎ without education  ‎ 23% of births are spaced at less than 24 months (NDHS 2012)‎  Women have their first child on average at age 18.6 years, 2.9 years on average after ‎ entering their first union Poverty levels are improving while living conditions remain challenging  The World Bank poverty indicator shows a positive trend in poverty reduction with a poverty ‎ headcount ratio of USD 1.25 a day for 43.6% in 2008, down from 65.9% of the population in 2005, ‎placing Niger above the sub-Saharan Africa average of 49.2% in 2008. However, living conditions ‎remain challenging ‎ Recurring food crises mean that many households, especially in rural areas, are severely affected by ‎under-nutrition  Food insecurity persists even in years when agricultural production yields a surplus. About 60% of households ‎ are only able to cover their food requirements for three months. In periods of food shortages, half ‎ of the population suffers from malnutrition  The situation is often dramatic for children under the age of five with about four out of ten children suffering from ‎chronic under-nutrition and one out of ten children suffering acute under-nutrition. During the 2009–10 food ‎ crisis, some 7.1 million people needed food assistance ‎  The principle challenge is to transition from the cyclical management ‎ of recurring food crises to more structured responses,‎particularly‎through‎the‎“Nigeriens‎ ‎ nourishing‎Nigeriens”‎(3N)‎ development‎ initiative, for promoting sustainable food security and agricultural ‎ Urban centers are growing and standards of housing and living are mostly inadequate ‎  The urbanization rate has risen from 16.3% in 2001 to 17.5% in 2005 and to 20.4% in 2010. The city of ‎Niamey had 1.3 million inhabitants in 2011, almost double the number in 2001 22 (708,000). An ‎ estimated 57% of the country’s‎urban population lives in Niamey‎  The annual growth rate in urban areas (6.2% per year) is about twice the growth rate for the ‎ population as a whole. If the current trend continues, the urban population will represent more ‎ than 40% of the total population by 2030‎  The last ten years were marked by progress made in establishing the institutional and legal ‎ framework for urban development in a context of decentralized management. However, major ‎ challenges remain in the area of urban planning, housing and habitat 22 www.citypopulation.de 10 Niger’s‎HIV‎response: Targeted investments for a healthy future Sustainable access to drinking water has improved, but remains challenging, especially in rural and ‎ periurban areas, while there is considerable under-investment of sanitation facilities in towns  The‎percentage‎of‎Niger’s‎population‎using‎a‎potable‎water‎source‎increased‎from‎22.3%‎in‎ 1992 to 50.1% in 2008 (National Survey on Budget and Household Consumption)  Coverage of drinking water needs increased gradually to 66.5% in 2011, a level still well below the millennium development goal (MDG) target of 80% in 2015‎  Principal constraints in this important sector are low investment levels, insufficient maintenance ‎efforts, and lack of water, or poor quality water, or water at great depth in some areas  More generally, there is an enormous unmet need for the supply of hygiene and sanitation ‎ services including garbage collection, and disposal of rain and wastewater Radio is the only medium able to reach a significant proportion of the population  ‎42% of men and 36% of women listen to the radio at least once a week (61% of urban men and 53% ‎ of urban women) (NDHS 2012)‎  ‎ 3% of men and 59% of women are not using any of the main media (radio, television, print) 5 ‎ regularly (63% of rural men and 66% of rural women)‎ Niger’s‎2012‎human‎development‎index‎(HDI)‎is‎0.304,‎ ranking Niger at 186 in this global ranking comparison (UNDP, 2013)  The index is a composite statistic of education, life expectancy, and ‎ income indices that rank countries‎for‎their‎advancement‎in‎human‎development.‎Niger’s‎HDI‎has‎‎ gradually increased from below 0.2 in 1980 to exceed 0.3 by 2010 HDI/Education: Niger’s‎education‎system‎has‎experienced considerable expansion, but the average duration ‎ of schooling of adults is only 1.4 years which negatively impacts the human development rank23‎  At the basic cycle 1 level of instruction, the number of pupils increased from 210,000 in 1979 to 1,901,115 in 2010. ‎ The gross attendance rate (GAR) has risen to 76% in 2010/11 with a large disparity between boys ‎ (85%) and girls (67%), and between urban (99%) and rural (70%) settings  At the basic cycle 2 level of instruction, the GAR decreases, especially for girls (about 18% in and‎there‎are‎large‎disparities‎between‎Niger’s‎regions‎‎ 2010), ‎  Attendance also remains low in the secondary cycle. The GAR was 4% in 2010, at 5.3% for and 2.5% for girls (the average GAR for sub-region countries was 16% in 2006)‎ boys ‎  In the NDHS 2012, 60% of males and 80% of females had no education, and only 19.5% of males and 8.5% of females had started or completed secondary school Niger’s‎literacy‎levels‎have consistently been very low, due to low school enrolment and ‎ poor teaching quality, with only a minority of teachers being appropriately trained  In 2012, 42% of males and 14% of females were classified as literate (NDHS 2012). In rural areas, ‎ only 32% of males and 7% of females were literate (70% males and 45% females in urban areas). ‎ Niger has one of the lowest literacy rates in the world ‎ 23 http://hdrstats.undp.org/en/countries/profiles/NER.html Development context, health priorities and health system challenges in Niger 11  The limited access to, and retention of girls within the educational system minimizes their economic life and in decision-making‎ opportunities in ‎  There is an urgent need to stimulate demand among vulnerable groups (girls, women, rural ‎ residents, etc.), for the promotion of quality instruction, particularly through policy, human resources ‎management, adaptation of teaching programs, and the gradual establishment of suitable school ‎infrastructure (IMF April 2013) ‎ The life expectancy at birth indicator, the health component of the HDI, shows a positive over the last 30 years‎ trend ‎  In 1980, life expectancy at birth was 39.5 years in Niger. This increased to 48.3 years in 2000 55.1 years in 2012‎ and ‎ The average per capita income (the economic component of the HDI) is moving in a but remains at modest levels ‎ positive direction, ‎  According to the United Nations Development Programme (UNDP) database, the average income has increased from USD 586 in 2000 to USD 701 in 2012 ‎ (expressed as the gross national income [GNI] per capita in purchasing power parity [PPP] terms, constant 2005 international $). The Gini index, that ‎ measures income inequality, was 34.6 in 2008, which is similar to Togo and Algeria‎  Two recent poverty profiles indicated that poverty affects rural areas more than urban areas, and ‎ is more pronounced among the female population. In addition, the phenomenon takes on ‎ endemic dimensions in the Maradi, Dosso, and Tillaberry regions. Although the depth and severity ‎ of poverty, as well as social inequities, declined slightly in recent years, they still levels‎ remain at high ‎ National development policy framework There are three principal strategic documents that are relevant‎to‎Niger’s‎future‎HIV‎investment‎ strategy ‎ and broader social-sector development: the Sustainable Development and Inclusive Growth ‎Strategy‎ (SDDCI NIGER 2035), the Economic and Social Development Plan 2012–15 (PDES), and the Health ‎Sector Development Plan 2011–15 (PDS).‎ The PDES, adopted in September 2012, is the frame of reference for interventions under government’s‎medium-term development agenda and is aligned with the MDGs ‎ the ‎  It capitalizes on goals and progress attained when implementing the Accelerated Development and ‎ Poverty Reduction Strategy, the sectoral strategies, and ministerial action plans‎  The PDES forms the basis for the execution of important structural projects, including the Kandadji ‎ hydroelectric dam, the exploitation and processing of coal from Salkadamna mines, investment in transport infrastructure ‎ and ‎  Improvements in rail and road infrastructure will reduce economic seclusion by better linking Niger to its neighboring countries. This progress‎ can‎ put‎ Niger’s‎ successes‎ in‎ reducing‎ the‎ transmission of HIV epidemic at risk—through creating hotspots where infrastructure is located, and intensifying the ‎ movement of people between low HIV prevalence areas and areas of higher HIV prevalence‎ 12 Niger’s‎HIV‎response: Targeted investments for a healthy future A core objective of the PDES is the improvement in the standard of living by reducing and the growth of the middle class ‎ inequalities ‎  The middle class is regarded as the most active population in terms of available skilled labor ‎ and increased domestic demand, and in particular, household consumption‎  However, epidemic statistics also show that ‎ HIV takes hold in the wealthier segments of the population first. The association is likely because of the availability of disposable income which ‎ can be used for pleasure-seeking purposes, including commercial sex The Health Sector Development Plan 2011–15 (PDS) that was adopted in January 2011, has as its overall ‎objective to contribute toward both the‎ improvement‎ of‎ the‎ population’s‎ health status and the attainment of the health MDGs  The PDS is based on a detailed analysis of the gaps and weaknesses in the public health sector. The ‎ insufficient coverage of quality SRH services is identified as a major shortfall, and urgent ‎ improvements to services of family planning, maternal health, immunization, integrated ‎ management of the sick child, nutrition and PMTCT, feature prominently  HIV prevalence is a key outcome indicator in the PDS results framework (target level <0.7% has been achieved in 2012 and needs to be maintained)‎ which ‎  PMTCT is presented as a high-impact intervention for both maternal/newborn health and child ‎ health, with a 2015 coverage target of 65%  ART coverage targets are 30% for children and 75% for adults with advanced infections, and 30% for ‎ pregnant women identified as seropositive How much is spent on health and where does the money come from?‎  In 2011, health spending accounted for 5.1% of GDP (Euro 225 million or USD 303 million), and was dominated by the public sector (57% of total) and private out-of pocket spending (39% of total)—‎Figure 8‎  According to the NDHS 2012, only 2% of men and women are covered by some medical or social ‎ insurance, however, 10% of women in Agadez and Niamey Regions reported having insurance ‎ medical/social ‎  Out-of pocket spending accounted for 95% of household health spending‎  External financing accounted for about one-third of all health spending (WHO, 2014)‎  Spending on HIV care and treatment accounted for about 1.6% of total health spending, and 2.7% ‎ of public health spending in 2011 Development context, health priorities and health system challenges in Niger 13 Figure 8. Health expenditure by source of financing, Niger (1995–2011)‎ Note: GDP = gross domestic product.  According to the PDS, the government intends to develop alternative health financing mechanisms, ‎ including the promotion of social health insurance (mutuelles de santé), and support for the elderly and destitute ‎ social welfare ‎  The free provision of some services is enshrined in laws: pediatric services for children aged 0–5 prevention‎ services‎ ;years (family planning, antenatal care); services for patients with cancer, TB and ‎ HIV/AIDS; and caesarean section (kit payable). Minimum and maximum are also regulated ‎ amounts for service fees ‎  In three regions, child delivery at health facilities is also free of charge (UNFPA support). In regions, services for sex workers are free (World Bank support) ‎ four ‎ What are the health priorities?‎ Burden of disease Communicable, newborn, nutritional and maternal causes such as diarrheal diseases, lower ‎ respiratory infections, and protein-energy malnutrition remained the top drivers of health loss in ‎ Niger (IHME/WB)  However, in comparison with 46 Sub-Saharan African countries, Niger has progressed burden of disease indicators and regional ranking—see Figure 9 (left and significantly in ‎ best, and rank 46 is worst) ‎ middle; note that rank 1 is ‎  Niger is the country in the whole region with the lowest percentage of DALYs attributable to ‎ non-communicable diseases (only 18%). However, the STEPS study results suggest that there has been a rapid increase ‎in hypertension and diabetes over the last few years (PDS p. 29) ‎  Despite progress, Niger had very poor rankings in the regional comparison for years-of-life- ‎ lost due to meningitis, measles, tetanus, diarrheal diseases and lower respiratory infections, in 2010 ‎ 14 Niger’s‎HIV‎response: Targeted investments for a healthy future  The country had, by regional comparison, relatively better rankings for road injury, ‎ interpersonal violence, diabetes, self-harm, chronic obstructive pulmonary disease, and HIV/AIDS Figure 9. Saharan Africa)‎ Burden of disease indicators, Niger (1990, 2010; ranking among 46 countries in Sub-‎ Age-standardized death Years of life lost (YLLs) due to Years lived with disability rate premature mortality (YLDs): any short- or long- term health loss 2,000 1,765 0 100,000 88,497 0 80,000 10 14,600 14,526 0 1,500 1,268 10 60,000 52,955 14,400 10 20 20 19 1,000 22 14,200 22 20 40,000 30 39 30 30 14,000 500 13,915 30 20,000 46 13,800 40 40 40 - - 13,600 1990 2010 1990 2010 1990 2010 Age-standardised death Age-standardised YLL rate Age-standardised YLD rate rate (per 100,000) (per 100,000) (per 100,000) Rank Rank Rank Source: Institute of Health Metrics and Evaluation, Human Development Network; The World Bank; ‎ The global burden of disease: Generating evidence, guiding policy—Sub-Saharan Africa Regional ‎ Edition. Seattle, WA: IHME, 2013.‎ Note: A rank of 1 is best, a rank of 46 least.‎ What is the provision and reach of health services in Niger? ‎ Niger has a policy approach that is focused on primary health care and on system ‎ the district health ‎ The provision of health services is dominated by the public sector (Health Sector Strategy, with: ‎ HSS), ‎  ‎829 Integrated Health Centers, each managed by a nurse, and linked to a total of 2,160 health posts ‎ managed by community health workers  ‎ 33 district hospitals, spread across 42 health districts‎  ‎ regional public health departments with 6 regional hospitals, 2 regional maternity reference 9 centers and 7 mother-child centers ‎ ‎  ‎ national reference hospitals and a national maternity reference center, of which 3 are in 3 Niamey ‎and one is in Zinder Approximately 47% of the population lives within 5 kilometers of a health facility (Figure 10)‎  In rural and in remote areas, transport costs arising from medical emergencies are a challenge‎ considerable ‎ Service provision by a male nurse is regarded as a barrier for females who use health services‎  However, according to the Ministry of Health (MOH) Human Resources Development Plan (2011–20), an increasing ‎percentage of health workers are female—75% of those working and aged under 35 years, and 85% of those who are still in training‎ Development context, health priorities and health system challenges in Niger 15 Figure 10. Access to health facility by health district, Niger (2011) ‎ 1,000,000 900,000 Population 0-5 km Population >5 km 800,000 700,000 600,000 500,000 400,000 300,000 200,000 100,000 - Tchirozérine Birni N'Konni Niamey III Maine Soroa Bilma Niamey II Boboye Arlit Tanout Aguie Mirriah N'Guigmi Loga Bouza Keita Abalak Magaria Matameye Diffa Gaya Mayahi Kollo Ouallam Say Tessaoua Madaoua Niamey I Doutchi Dakoro Illela Tera Filingué Goure Madarounfa Maradi commune Dosso commune Guidan Roumdji Tchintabaraden Tahoua Commune Zinder Commune Agadez commune Tillaberi commune Agadez Diffa Dosso Maradi Tahoua Tillaberi Zinder Niamey Source: Annuaire MSP, 2012‎ The provision of health services is significantly limited by infrastructure and resource constraints—‎ the following coverage levels were reported in 2011:‎  The geographical coverage of the provision of basic health care was estimated at 71% (HSS)24  The treatment of malnutrition was estimated at 22% (HSS Annex 1)‎  Antenatal care consultation CPN4+ was reported at 26%‎  Anti-parasite treatment of pregnant women was estimated at 21%‎  Detection and treatment of syphilis during pregnancy was reported at 6%‎  Assisted delivery was at 29% in 2012 (e-MTCT plan) ‎  Management of neonatal infection was at 11%‎  Diagnosis and treatment of STIs was reported at 37%‎  Directly observed treatment supervision for TB was at 58%‎ Prevalence of modern contraceptive use has increased by about 1% every 5 years among women living in Niamey, in other urban zones, and in rural areas, but remains low overall (NDHS ‎1998, 2006, 2012)‎  Overall, prevalence of modern contraceptives was calculated to be 11% in 2012, and 12% in women who were in a union (married or cohabiting). In ‎ urban areas, 27% used modern contraceptives compared to 10% in rural areas. Regionally, contraceptive use was highest in Niamey (32%) and ‎lowest in Tahoua (6%)‎  Women obtained modern contraceptives mainly from public sector Integrated Health ‎ Centers and health posts, but also from public and private sector pharmacies and mobile chemists 24 This indicator does not include HIV-specific services like PMTCT, ART and HIV testing. 16 Niger’s‎HIV‎response: Targeted investments for a healthy future ‎ (especially the pill); condoms played a very minor role in contraception with less than 0.1% of women who were ‎ in a union reporting using condoms for contraception (DHS 2012) ‎  The majority of pill users use products that are socially marketed‎ ‎ 0% of all women in a union and aged 15–49 years old expressed a demand for family 3 planning (FP), of which just less than ‎ half (14%) are satisfied (with contraceptive use) and just over half (16%) are not satisfied (unmet need) ‎(NDHS 2012)‎  In 2012, there were 3.1 million women between 15–49 years old who were in union (married or cohabiting), and ‎ an estimated 930,000 of these considered themselves to be in need of FP  An estimated 498,000 women in a union have an unmet FP need (estimations based on NDHS 2012)  ‎ 5% of women and 59% of men have heard a message about FP on the radio within the last 5 month ‎(only 41% of rural women) (NDHS 2012)‎ There are significant missed opportunities in health care contacts with non-users of family (NDHS 2012)‎ planning ‎  ‎ 86% of non-users of FP report not having spoken about FP in the last 12 months, neither with a‎ community health worker nor a clinic staff member  ‎ 6% of all non-users of FP have visited a health facility in the past 12 months, but have not 3 discussed FP during that visit (missed FP opportunity)—see Figure 11 (dotted areas)‎ ‎ These missed FP opportunities are also missed opportunities to reduce vertical transmission of HIV ‎ (PMTCT prong of family planning)—see Figure 11 (dotted orange areas prevalence). Risks may be higher in:‎ show strata with higher HIV ‎  Women aged 30–44 years old (an elevated HIV prevalence compared to younger women, and many ‎ women grand multiparous with five or more births)‎  Women living in the city of Niamey and the other urban areas/towns (female HIV and 0.7%, respectively)‎ prevalence of 1.5% ‎  Women in the regions of Niamey and Diffa (female HIV prevalence of 1.5% and 0.8%, respectively)‎ Figure 11. Non-users of family planning: contact with providers, Niger (2012)‎ 100 90 80 70 60 50 40 30 20 10 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Niamey Rural Agadez Dosso Tahoua Niamey Diffa Tillaberi All urban Maradi Zinder Age Residence Region Home visit, spoken of FP Clinic vist, spoken of FP Clinic visit, not spoken of FP No contact for speaking aboout FP Source: NDHS 2012 Table 5.12 and 15.4‎ Section 3. The state of the HIV epidemic‎ What has been the historical course and scale of the epidemic?‎  Niger’s‎HIV‎epidemic‎is‎contracting. HIV prevalence data from the general population and key ‎ populations at higher risk suggest that the epidemic has become more concentrated over the last ‎ ten years. HIV prevalence levels have been decreasing in almost all surveyed populations (Figure ‎ 12). Sentinel surveillance data from antenatal care services corroborate the contracting nature of ‎the epidemic, with HIV prevalence in pregnant women declining from 1.7% in 2009 to 0.9% in 2012 Figure 12. HIV prevalence by population, Niger (2002--12)‎ 25.6 20.9 AxiHIV prevalence (%) 17.3 2002 2006-2008 2011-2012 2.8 3.8 2.7 0.8 1.9 1.3 2.7 1.6 1.8 0.5 0.83 0.3 1.0 1.1 1.3 1.1 0.4 0.4 0.4 0.1 0.52 0.1 0.57 0.0 FSW Prisoners Military Migrants Mine Truckers Females Females Males Males workers 15-24 25-49 15-24 25-49 Sources: Seroprevalence report Care 2002, SSG 2008, SSG 2011, DHS 2006, DHS 2012.‎  The epidemic trend is decreasing in both urban and rural areas (Figure 13). HIV prevalence has ‎ declined least in urban females who are five times more likely to be HIV positive than females who live in rural areas (a higher HIV prevalence prevails in Niamey- based females). ART may play a ‎ role in elevated female HIV prevalence, especially in Niamey where there are six ART treatment ‎ centers. In 2012, there were nationally 40% more females on ART than males (Rapport annuel ULSS ‎ 2012, T8, page 21) ‎ 17 18 Niger’s‎HIV‎response: Targeted investments for a healthy future Figure 13. HIV prevalence by sex and residence, Niger (2002, 2006, 2012)‎ HIV prevalence (%) 2.6 1.5 1.5 1.3 1.1 0.6 0.7 0.6 0.5 0.5 0.2 0.3 Urban Rural Urban Rural Female Male 2002 2006 2012 Sources: Sero-prevalence report Care 2002, DHS 2006, DHS 2012‎  Niger shares national borders with several countries which report higher HIV prevalence ‎ levels in their populations: Nigeria, Mali, Benin and Burkina Faso. The first three listed countries also report higher ‎prevalence levels in female sex workers (UNAIDS 2013 global report and aidsinfo.com, stat compiler ‎ for DHS data, UNGASS 2012 country reports). Cross-border traffic interactions and mobile ‎populations will impact HIV epidemic dynamics in Niger  Near-universal male circumcision (at 99%) contains the epidemic (male circumcision data from ‎ 2006 DHS). The chief role of male circumcision toward reducing the transmission of HIV from females to ‎ males has been established beyond any doubt. The Niger experience supports the claim that very high ‎levels of male circumcision prevent a generalized epidemic, despite high risk sexual behaviors and ‎ vulnerability of its people. HIV infections are largely contained within females as a result of low transmission ‎ efficiency to their male partners that is attributable to male circumcision  ‎ ccording to model estimations, HIV incidence has decreased over time from the peak A level of ‎8,470 new HIV infections in 2000 to 1,447 in 2013 (Spectrum 2013). This has resulted in reduced HIV ‎ prevalence, that is determined by the incidence of new HIV infections and by AIDS-related ‎mortality‎  Until recently, decreasing HIV incidence has been accompanied by high AIDS-related due to low access to ARV treatment (Figure 14) ‎ mortality ‎ - Since 2004, ART coverage has been scaled up and in 2012 had reached 51% coverage in children below 15 years of age (Spectrum 2013) ‎ in adults and 13% ‎ - In 2012, an estimated 3,360 PLHIV died due to an AIDS-related cause, and the total cumulative ‎number of AIDS deaths in Niger up to the end of 2012 is estimated at 68,730. High AIDS-related mortality ‎ is also suggested by successive demographic and health survey (DHS) data, where a large part of HIV prevalent cases in the 2006 ‎ survey are not present in the respective age cohort of the 2012 survey (see also Figure 16) ‎ - Conditions like iron deficiency anemia and malnutrition and diseases such as diarrhea and TB (see ‎graph in Executive Summary) can be expected to accelerate the progression toward AIDS and death among ‎ untreated people living with HIV ‎ The state of the HIV epidemic 19 Figure 14. New HIV infections, AIDS deaths and ART, Niger (1994–2013)‎ 8000 30000 7000 25000 Incidence and mortality 6000 Numbers on ART 20000 5000 4000 15000 3000 10000 2000 5000 1000 0 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Mort-child Mort-adult Inc-child Source: Spectrum 2013 (revised version)‎ Note: ART = antiretroviral therapy; Mort = mortality; Inc = incidence. How is HIV distributed in the population of Niger?‎  The two key populations that continue to have elevated HIV prevalence are female sex (FSW) and prisoners ‎ workers ‎  Overall HIV infections are concentrated within certain segments of the female population: in female ‎ sex workers, female members within key populations at higher risk (including female prisoners), and ‎specific female demographics within the general population (see below). Epidemiological data suggest that male circumcision is likely to prevent transmission events from HIV-‎ infected females to males (including from female sex workers [FSWs] to their clients) ‎ Female sex workers (FSWs)  Niger has an estimated 28,000 FSWs who are concentrated in urban and industrial zones such as ‎ mining areas and petrol fields (SSG 2011). Despite declining HIV levels in FSWs, the prevalence ‎ differential between FSW and females from the general population is still large (17.3% versus 0.4%, ‎see Figure 12). In 2011/12, HIV infection was 43 times more likely to occur in FSWs than in females from the ‎ general population, indicating the high exposure to unprotected sexual acts of FSWs and their high ‎ susceptibility to HIV infection ‎  Survey data suggests a large regional variation of HIV prevalence in FSWs. In 2011, levels were highest ‎in FSWs surveyed in Tahoua and Maradi (31%), followed by Agadez (23%), Zinder ‎(21%) and Niamey (19%). Sex workers in Dosso and Tillaberi were found to have below-average ‎ HIV prevalence levels (16% and 12%), and there were no HIV infections diagnosed in the 71 FSWs ‎ surveyed in Diffa (also 0% in the 2009 survey) ‎  FSWs get infected rapidly once they join the sex industry. FSWs below 20 years of age were as likely ‎ to be HIV infected as older FSWs (p = 0.98) and FSWs with less than one year of sex work experience ‎ had the highest HIV prevalence of all respondents in the 2011 survey—see Figure 15. A 2010 study ‎ (CISLS, Cartographie sites prostitutionelles et MSM, 2010) indicates that younger FSWs tend to ‎ work on the street, while women up to 50 years of age tend to work in brothels (maison closes). ‎ The number of clients also appears to differ 20 Niger’s‎HIV‎response: Targeted investments for a healthy future according to type of site, with bars and restaurants reporting ‎ the highest proportion of sex workers reporting >20 clients a day‎  The average duration of sex work is 3 years with 41% of SWs working from 1 to 4 years and 28% working for over four years25 ‎ ‎  The majority of women in sex work are divorced (63%, in the 2010 mapping study). Some driven into sex work after becoming widowed, divorced or separated ‎ women are ‎ Figure 15. HIV prevalence by FSW age and duration of sex work, Niger (2011)‎ 21.4 17.6 17.4 17.2 18.3 HIV prevalence 16.4 (percentage) 13.1 15-19 years 20-24 years 25+ years <1 year 2 years 3 years 4+ years Age of FSW Duration of sex work Source: SSG 2011, Table 4.6b and annex 13. Note: FSW‎= female sex worker.  Reported violence is strongly associated with HIV infection. In 2011, FSWs who reported being victims of violence had a ‎ HIV prevalence of 32% whereas those not reporting acts of violence had a HIV prevalence of 14%. Reported ‎ use of drugs and/or alcohol was not associated with HIV status ‎ Prisoners  There are approximately 11,600 prisoners‎ in‎ Niger’s‎ correctional institutions. It is are female.26‎ ‎ reported that 3.2% ‎ The average duration of a prison incarceration is 1–2 years. Prisons represent a risk context for ‎ HIV, TB and other diseases. One official source reports conditions of inadequate ventilation, overcrowding, and incidences of death from AIDS, malaria and TB27‎  HIV prevalence in prisoners has remained stable over the last decade. HIV prevalence has leveled off at between 2.7%–2.8%, but ‎ has not followed the decreasing epidemic trend seen in other key populations. In Niamey, 6% and in Diffa 4.3% of ‎ prisoners surveyed in 2011 were HIV infected, reminding us that localized ‎ epidemics can ignite in risk settings like prisons ‎ 25 Animas-Sutura- SongES – RENIP.‎Résultats‎de‎l’étude‎de‎cartographie‎des‎sites‎Prostitutionnels‎de‎la‎ communauté urbaine de Niamey, 2013. 26 International Center for Prison Studies, World female imprisonment list, second edition. 2012. http://www.prisonstudies.org/sites/prisonstudies.org/files/resources/downloads/wfil_2nd_edition.pdf 27 Bureau of Democracy, Human Rights and Labor, Niger country report, 2012. http://www.state.gov/j/drl/rls/hrrpt/humanrightsreport/index.htm?year=2012&dlid=204151#wrapper The state of the HIV epidemic 21  HIV knowledge has been consistently low in prisoners and may be decreasing (2011 history of STIs is associated with HIV infection‎ SSG). A ‎ Men who have sex with men (MSM) and male sex work  Homosexuality is not socially accepted in Niger and sexual relationships between men are ‎therefore hidden.28 ‎ There are no epidemiological data available on MSM in Niger. Mapping and ‎ estimation activities on MSM have assessed the numbers of MSM in individual regions: Circa 200 ‎ MSM in Tahoua; circa 150–200 MSM in Zinder; and 89 MSM in Niamey, including in bars, brothels and ‎cinemas (Animas 2013). 10 MSM were identified in Maradi but could not be interviewed‎  In one survey, a high proportion of sampled MSM said they were involved in commercial sexual ‎ activities (2010 data, MSM identified through snowball sampling). In 2010, 64% of interviewed ‎MSM said that they receive payment from male sexual partners. There is a proportion of MSM who ‎ identify themselves as bisexual, and some are married, which links any MSM related HIV transmission to the ‎female population‎ Other key populations that include sex work clients  HIV in FSW client populations: Periodic surveys have also been conducted in teachers, truckers, ‎ mine workers, labor migrants and the military (see Figure 6 for HIV prevalence data). In all males ‎ of the key populations combined, HIV prevalence in 2011 was 0.8% (24 infections per 2,943 men with ‎ bio-behavioral data), which can be considered as the HIV prevalence‎of‎“sex‎work‎‎clients”‎‎  HIV in female members of key populations: In all females of the key populations combined ‎ (excluding FSW), HIV prevalence in 2011 was 1.7% (7 infections in 405 females with bio- data) ‎ behavioral ‎  Female members of the key populations have a significantly higher HIV prevalence members. However, female key population samples are small:‎ than male ‎ 7.7% (female – 2/26) versus 0.1% (male – 1/746)‎ - Military: ‎ - Labor migrants: 2.9% (female – 1/35) versus 1% (male – 6/600)‎ - 3.6% (female – 2/55) versus 2.6% (male – 14/548)‎ Prisoners: ‎ - 0.7% (female – 2/289) versus 0% (male – 0/226)‎ Teachers: ‎  In men, sexual risk behaviors and use of FSWs services are linked to being away from disposable income, and mobility (DHS 2012) ‎ home, having ‎  Overall, few surveyed men report that they have paid for sex—1.3% of men aged 15–59 years ‎report ever having paid for sex, and 0.5% say they did so in the last 12 months (NDHS 2012). A history ‎ of paid sex is more frequent in men living in Niamey (4.6% ever paid for sex), divorced or ‎separated men (3.0%), the wealthiest quintile of men (2.9%), men living in Tillaberi region (2.8%) and ‎men with a secondary or higher education, or who were living in an urban area (both 2.7%)‎ 28 CISLS (2010). La cartographie des professionnelles de sexe et des homosexuels. 17 Février, 2010. 22 Niger’s‎HIV‎response: Targeted investments for a healthy future General population  HIV prevalence has declined in almost all of the 5-year age groups as defined by the DHS, and in 2012 ‎remained highest in women in their thirties and early forties (up to 0.8%) and in men in their late ‎thirties and forties (up to 1.0%)—see Figure 16. Some of the prevalent HIV cases in these older age ‎ groups are attributable to ART provided to individuals with advanced HIV infections. Some HIV ‎ positive women will have a history of sex work, given the demographics of FSW, and the turnover ‎ of women in the sex industry  The age and sex pattern of HIV infection reflects the age mixing pattern of sexual relationships in ‎ Niger. Very few young men are infected compared to young women (who tend to have much ‎older partners/spouses) Figure 16. HIV prevalence by age group, Niger (2002, 2006, 2012)‎ 3.1 HIV prevalence (percentage) 2.6 2.6 2.7 2.4 2.2 1.7 1.5 1.4 1.1 1.2 1.1 1.0 1.0 1.0 0.8 0.8 0.9 1.0 0.6 0.5 0.5 0.6 0.6 0.6 0.6 0.3 0.4 0.3 0.4 0.30.3 0.2 0.3 0.2 0.3 0.1 0.1 00 00 0 0.1 15-19 20-24 25-29 30-34 35-39 40-44 45-49 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-59 Female Male 2002 2006 2012 Sources: Care report 2002; NDHS 2006; NDHS 2012 (preliminary).‎  It is estimated that just over a quarter of all PLHIV are in Niamey (an estimated 26% ‎ in 2012), and that over half of all PLHIV reside in either Niamey, Dosso or Tahoua (based on the ‎regional population numbers and the regional HIV prevalence levels 2012—see Figure 17). Two ‎ regions, Niamey and Dosso, have seen a rise in PLHIV numbers since 2006. It is estimated that ‎Tahoua had the largest decrease in PLHIV numbers and that Agadez and Diffa have the lowest ‎PLHIV numbers in 2012 compared to the other regions ‎ Figure 17. Numbers of people living with HIV by region, Niger (2006, 2012)‎ 16,846 Estimated PLHIV 11,974 numbers 8,667 8,682 8,881 6,429 6,873 6,395 5,727 5,210 4,874 4,298 5,031 4,394 2,863 2,135 Niamey Dosso Tahoua Maradi Zinder Tillabéri Diffa Agadez 2006 est. PLHIV 2012 est. PLHIV ‎ 013.‎ Sources: Population estimates based on regional census data, DHS 2006 & 2012, revised Spectrum 2 Note: PLHIV = people living with HIV The state of the HIV epidemic 23  Key demographics with elevated HIV prevalence are (HIV prevalence DHS 2012): ‎ - Female residents of Niamey (1.5%)‎ - Females who are divorced/separated (1.7%), or widowed (3.2%)29‎ - Females away from home at least 1 month per year (1.1%)‎ - Females reporting no sexual partner in the past year (1.7%)‎ - Females with sexual inception at age 20 years or above (1.7%)‎ - Males reporting no sexual partner in past year (1.1%)‎ - Individuals with primary education compared to none or no higher education (female: 1.3%, Male: 1.0%) ‎  The changes in HIV prevalence between 2006 and 2012 in different population strata of education, ‎ wealth and economic activity, suggest a rapidly changing epidemic situation (Figure 18). Even in the ‎ stratum of highest education, HIV prevalence has declined sharply in both females and males, ‎ presumably through a combination of lower HIV incidence and high AIDS-related mortality (the ‎scale of the ART program was small until circa 2009) ‎ Figure 18. HIV prevalence by socio-economic strata, Niger (2006, 2012)‎ 1.5 2006 2012 1.4 1.3 1.2 1.2 HIV prevalence 1.0 1 0.9 0.9 0.9 0.8 0.8 1 0.6 0.6 0.6 0.6 0.6 0.6 0.5 0.5 0.4 0.4 0.4 0.4 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.0 0.0 Secondary/higher Secondary/higher No employment No employment Employment Employment Primary Primary No education Q1 wealth Q2 wealth Q3 wealth Q4 wealth Q5 wealth No education Q1 wealth Q2 wealth Q3 wealth Q4 wealth Q5 wealth Female Male Sources: DHS 2006 and 2012.‎ Note: Q = wealth quintile. 29 Note that in the 2010 FSW survey, 63% of FSW reported being divorced and 4% were widows. 24 Niger’s‎HIV‎response: Targeted investments for a healthy future Where do new HIV infections occur?‎ ‎ Incident infections by age and sex ‎  Niger’s‎epidemic‎is‎fuelled‎by‎both‎vertical‎and‎sexual‎transmission.‎In‎2013, it was estimated that 709 ‎ new HIV infections occurred in children through mother-to-child transmission, and 738 new ‎infections occurred in adults (Spectrum 2013)—see Figure 19 ‎  Estimated HIV incidence is similar for males and females, but young females and older males have ‎ slightly higher incidence due to the age pattern of sexual mixing Figure 19. Spectrum-estimated HIV incidence by age and sex, Niger (2012)‎ 710 738 346 363 213 165 209 151 0-4 15-24 25+ Total Female Male Source: Spectrum 2013 (revised version) Note: MTCT = mother-to-child transmission‎ Incident infections by sub-population  Using Optima and stratifying the population by demographic, epidemiological and behavioral ‎ characteristics into 14 sub-populations, annual new infections in each sub-population were estimated (Figure 20). In 2012, MTCT and new infections in adults aged 25–49, and in ‎ migrants and FSWs, ‎ accounted for an estimated 80% of HIV incidence. The largest single contributor to new infections ‎ was MTCT Figure 20. Optima-estimated trends in HIV incidence by sub-populations, Niger (2000–20)‎ Source: Optima application Niger, January 2014.‎ Note: Graph shows estimated new HIV infections per year.‎ The state of the HIV epidemic 25 Incident infections by location There are no direct estimates of the geographical distribution of HIV incidence available, however, ‎there are clues as to where HIV transmission may be occurring. ‎  The geographical distribution of PLHIV suggests that there may be an important percentage of HIV ‎transmission occurring in the South-West portion of the country—in 2012, an estimated 70% of PLHIV lived in ‎ one of four regions in the South-West covering only a quarter of the country surface area (Niamey, ‎ Dosso, Maradi, and Tahoua)—Figure 21. If untreated, the infectiousness of PLHIV who are at a more advanced ‎stage of HIV/AIDS can perpetuate the onward transmission to sexual contacts Figure 21. Concentration of the HIV burden by region, Niger (2012)‎ ‎ Source: Population estimates based on regional census data, DHS 2012.‎ Note: People living with HIV (PLHIV) density is given per km2: ≥0.3,‎0.1–0.29, 0.01–0.09, et <0.01.  The role of FSWs in the ongoing transmission of HIV (Figure 20), the geographical distribution of sex work ‎sites (Figure 22), and the evidence of an urban HIV epidemic (Figure 13), all support ‎the notion that urban and industrial zones in Niger remain areas of elevated HIV transmission. ‎ In 2013, sex work sites were mapped and characterized into four regions to improve the understanding of hotspots of sex work-related HIV transmission. Over 3,200 FSW were counted in 498 identified sites ‎ 26 Niger’s‎HIV‎response: Targeted investments for a healthy future Figure 22. Sex work-related hotspots of transmission, Niger (2013)‎ Source: Cartographie de prostitution, Animas, 2013. Note: #FSW = number of females sex workers ‎ Section 4. HIV financing, HIV services and programs and their efficiency How is the HIV response financed, and what has been programs?‎ spent on the different HIV services and ‎ National spending on HIV and AIDS is well documented in NASA reports covering the years 2007 ‎ and 2008 (CISLS, 2009) and 2009–12 (CISLS, 2013), that is summarized in Table 2.‎ Over six years, the costs of the national response to HIV amounted to Euro 58 million (USD 78 ‎ million), corresponding to an annual average of Euro 9.7 million (USD 12.6 million), or 0.24% of GDP, ‎ over that period, but costs fluctuated dramatically on an annual basis‎  There is no clear trend in HIV/AIDS-related spending—it declined by almost one-half between 2007 ‎ (Euro 11.4 million/USD 14.8 million) and 2010 (Euro 5.9 million/USD 7.7 million), but has increased ‎steeply since, and had attained the highest level observed so far of (Euro 12.5 million/USD 16.3 ‎million) in 2012 The financing of the HIV response has been dominated by external funding, accounting for 90% of HIV spending from 2007 to 2011 ‎ almost ‎  However, much of the increase in HIV/AIDS spending in 2012 was accounted for by public spending ‎(up by Euro 3.4 million/USD 4.5 million) largely due to the new World Bank loan, funding declined by Euro 1.4 million/USD 1.8 million ‎ while external ‎  Consequently, the share of external funding has declined to 67%, while public funding for 30% of total spending in 2012‎ accounted ‎  The most important sources of external support have been the Global Fund (29% of total funding ‎in 2007–12), UNICEF, the Government of Germany (Kreditanstalt für Wiederaufbau from 2012 onwards (2012–16)‎ [KfW]), and the World Bank loan ‎  Looking ahead, there is—as per 1 January 2014—some uncertainty regarding the level of support ‎ Niger can obtain from the Global Fund. The current agreement with KfW is expiring in April 2014, ‎ and at the time of writing only USD 2.6 million in Global Fund commitments HIV/AIDS response remained undisbursed‎ towards the national ‎ At least since 2009, most HIV/AIDS services are delivered through the public sector (about two-‎thirds of all spending), while the private sector accounted for about one-quarter, and NGOs for one-tenth of the total in 2009–12‎ international ‎  The domestic non-public sector as implementer largely reflects prevention service spending that is delivered through Animas-‎Sutura30‎ is‎ and‎ financed by ‎ the Government of Germany (KfW) ‎ 30 Animas-Sutura is an NGO and social marketing organization established with KfW support. It has become an instrument in the implementation of the national population policy and HIV strategy. Its objective is to distribute condoms and other contraceptives in a way that makes them accessible country-wide. 27 28 Niger’s‎HIV‎response: Targeted investments for a healthy future In 2009–12, most HIV prevention spending was devoted to condom social marketing and ‎ distribution (37%), prevention of mother-to-child transmission (19%), voluntary counseling and ‎testing (15%), and measures targeting young people (13%) ‎  Only 7% of prevention spending was targeted toward “accessible‎risk‎populations”, and only 1% was allocated to ‎ sex workers  Expenditure for sex work programs has been low and has fluctuated, with the majority of funding ‎ coming from international sources (1.9% in 2009 and 0.9% in 2012 of total prevention spending). ‎ This raises questions with respect to allocative efficiency, sustainability and ownership since HIV prevalence among FSW has ‎ historically been and remains very high, ‎ and there has been a substantial decline of SW investment (65% between 2007 and 2008) bilateral cooperation project SIDA-3 came to an end31‎‎ when the regional ‎ Table 2. HIV/AIDS-related expenditures, Niger (2007–12)‎ Niger: HIV/AIDS-Related Expenditures, 2007-2012 2007 2008 2009 2010 2011 2012 (CFA franc million, unless stated otherwise) Total 7495.3 6420.7 5377.9 3862.3 6897.5 8216.8 Prevention 2507.3 1473.1 2089.8 1796.8 3719.5 3791.8 Treatment and Care 2420.1 1770.2 1440.6 687.0 1920.8 2226.7 Social Mitigation 296.3 327.6 402.9 292.8 130.1 130.9 Management, Coordination, Support 2271.7 2849.7 1444.6 1085.7 1127.1 2067.5 By source of financing Public domestic sources 286.9 253.5 37.2 53.0 200.5 2443.9 External sources 6247.9 5352.1 5134.7 3583.3 6434.3 5493.7 Multilateral and Bilateral 5689.3 4693.2 4469.1 2833.1 5777.8 4918.9 o/w: GFATM 1500.3 1428.3 2757.1 1216.9 2538.2 1494.0 o/w: UNICEF 105.8 99.9 142.2 133.9 1479.6 953.9 International NGOs 556.1 658.8 665.6 750.1 656.5 574.8 Private sources 960.5 815.1 206.0 226.0 262.7 279.2 By implementing agency/sector Public sector 2553.1 2693.7 3498.0 1904.8 4624.6 5600.7 Private sector 3841.1 2657.2 1157.6 1312.8 1775.0 2109.0 International NGOs and Other 1101.1 1069.7 722.3 644.7 497.9 507.2 Total spending (percent of GDP) 0.37 0.27 0.21 0.14 0.23 0.25 o/w: Public domestic sources 0.01 0.01 0.00 0.00 0.01 0.07 Total spending (Euro million) 11.4 9.8 8.2 5.9 10.5 12.5 Sources: CISLS (2009, 2013) for HIV/AIDS spending, IMF (2013) for GDP. 31 Batona, G., et al., Challenges and barriers for conducting STI/HIV. Prevention projects targeting female sex workers within national programs in Benin and Niger. Sexually Transmitted Infections, 87: A354. 2011. HIV financing, HIV services and programs and their efficiency 29 What HIV services and programs have been scaled up?‎ Niger’s‎HIV‎prevention‎services‎have‎ focused on reducing unprotected heterosexual acts the risk of vertical transmission during pregnancy and birth ‎ and ‎  Homosexual transmission has had less focus. Although there is evidence that sexual transmission ‎between male partners happens, there are no adapted services for MSM of any coverage and ‎ scale  HIV transmission through injecting drugs has not been regarded as a public health issue to date, and ‎there are no specific services developed. Drug trafficking is occurring and the first ‎ consultations have been held (e.g., in Agadez) to address the emerging threat of illicit drug use in Niger‎ Services for sex workers and their clients Since the end of SIDA-3 in 2007, a regional bilateral cooperation project, medical follow-up of sex ‎ workers has been the responsibility of national STI reference centre (Center National de ‎ Référence des IST). Service delivery essentially consists of community-based ‎ activities (peer education, and condom distribution), linkage to health facilities (STI, HCT), medical ‎ follow-up of SWs, and communication activities with SW clients. Actual implementation recently, been patchy and lacking any scale:‎ has, until ‎  It is not known what program coverage has been achieved—in 2008, an estimated 75% of SWs were ‎ reached with some prevention activities; in 2011, an estimated 54% were reached with activities ‎ (second generation surveillance [SSG] 2011). It is not known what ART coverage has been achieved among HIV-positive SWs‎  There is no comprehensive, defined standard package with quality-assured, standard operating procedures that ‎can be applied in all parts of the country (except peer education among SWs, ‎ which is well defined in terms of content and delivery modalities32‎ )‎  Services planning has historically not been sufficiently based on strategic information migration, site characteristics and the dynamic nature of hotspots) ‎ (such as SW ‎  Services have been irregular and some adapted services have abandoned their specific ‎ focus on SWs (the only adapted service that remained was the Center National de which ensures the medical follow-up of SWs)‎ Référence des IST, ‎  SWs have to pay fees for their medical card and for each consultation even at the adapted service ‎ center. The consultation fee increased from FCFA 200 to FCFA 1,500, however, this fee was overturned ‎ in the four World Bank-supported regions where consultation is now free for SWs Only 4 of 10 Niamey sex workers are linked into the medical follow-up program, consisting screening and treatment services at a health center33 ‎‎ ‎ of ‎  The monthly medical monitoring service is only provided for SWs in Niamey. The majority of workers accessing the services are foreigners i.e., non-Nigerien‎ sex ‎ 32 CISLS‎(2011).‎Directives‎sur‎l’identification‎et‎la‎formation‎des‎paires‎educatrices.‎Draft‎Fevrier,‎2011. 33 Animas-Sutura- SongES—RENIP.‎Resultats‎de‎l’etude‎de‎cartographie‎des‎sites‎prostitutionnels‎de‎la communaute urbaine de Niamey, 2013. 30 Niger’s‎HIV‎response: Targeted investments for a healthy future  Barriers to using services cited by respondents of the mapping study are the non-proximity of the ‎ health center, poor reception received, slowness of services, and lack of privacy. An additional barrier cited ‎ anecdotally in the study related to the need to re-purchase a medical log book after a missed ‎ appointment‎  Key informants emphasize the challenge that the vast majority of FSW are clandestine— known to the authorities and therefore are difficult to serve and protect ‎ not ‎ The country has started to systematically define and characterize SW typologies and sites, which ‎ represents indispensable strategic information toward tailoring services to the locality and needs ‎  SW venues and sites are34 bars,‎“buvettes”,‎tourist‎camps,‎night‎clubs,‎restaurants,‎markets,‎ hotels,‎“maison‎closes”‎and‎the‎street‎‎ ‎  SWs below‎30‎years‎of‎age‎tend‎to‎work‎in‎the‎street,‎while‎“maison‎closes”‎tend‎to‎include‎ up to 50 years of age ‎ women ‎  The number of clients differs according to type of site, with bars and restaurants having the highest ‎ proportions of SWs serving >20 clients a day Sex‎workers’‎condom‎use‎has‎multiple‎determinants‎and‎barriers  Client refusal is the chief reason for unprotected commercial sex—client refusal was responsible ‎for 85% of all unprotected sex acts in a 2010 SW study in Niamey35. This level of client refusal of ‎ condom use emphasizes the need to augment risk perception among higher- risk men (SWs in Maradi ‎ report that 65% of clients refuse condoms; 72% of SWs permit unprotected sex with clients who refuse to use condoms36) ‎  SW characteristics such as age, length of time involved in sex work, education and affiliation to venues, ‎ determine the motivation to use condoms, and their self-efficacy in defending their wish to use ‎condoms  Condoms are often, but not always available—86% of SW in a recent mapping study say condoms ‎ were always available  The type of sex partner determines condom use—condom use is highest when with commercial partners ‎ and lowest with boyfriends (about 23% of SW report using a condom at every sexual act37), and ‎ ‎ ‎ this differential indicates a need for risk counseling of SWs SW interventions appear to offer significant scope for strengthening scale, ‎ coverage and impact. ‎  Few Nigerien SWs benefit from the full combination package of services ‎ 34 CISLS. Cartographie sites prostitutionelles et MSM, 2010. 35 Barakatou Adamou Yahaya and Gado Moussa, Seroprévalence de la syphilis et du VIH chez les patients IST et les professionnelles du sexe du Center national de référence des IST et du Center national de Dermato Lèpre de Niamey,‎du‎24‎mai‎au‎5‎juillet‎2010,‎‎Mémoire‎de‎fin‎d’étude,‎IPSP,‎2010. 36 Souleymane Hamidou Garito, et al., Évaluation de la prévention des IST/VIH/sida en milieu des professionnelles du sexe‎dans‎la‎commune‎urbaine‎de‎Maradi,‎2010,‎mémoire‎de‎fin‎d’étude,‎IPSP,‎2010. 37 CISLS‎et‎cabinet‎national‎d’expertise‎en‎sciences‎sociales‎(CNESS‐Bozari), Enquête nationale de surveillance de seconde génération sur les IST/VIH/sida au Niger, rapport d'indicateurs, 2008. HIV financing, HIV services and programs and their efficiency 31  With higher HIV prevalence observed in SW sites bordering Nigeria—due to the associated migratory activities and presence of SW client populations—targeting resources to these SW ‎ seems especially promising ‎ sites ‎  Strengthening program monitoring is required to inform management, foster learning ‎ and contribute to evaluation activities ‎ Condom promotion and distribution services ‎ Prevention of HIV and STIs through condom use and promotion has been a focus for many years with free distribution, and socially-marketed and commercial condoms being made available. ‎ Summarized results are as follows:‎  Condom use is only significant in commercial (circa 87% use across several surveys) and casual acts ‎(circa 40% use across several surveys), they are virtually excluded from use in and play a very minor role in FP ‎ regular partnerships (circa 3% use), ‎  There are an estimated 2.32 million unprotected commercial sex acts per year in Niger, which ‎ represents the top priority for condom targeting (see Table 3). It is also estimated that there are ‎over 3 million unprotected casual acts, which is a second priority for the targeting of condom promotion activities ‎ for HIV and STI prevention  Young men increasingly report condom use in casual sex—38% in 2006 (DHS) and 66% in 2011 (SSG) ‎ (very few young females report casual sex)  Condom use data from key populations suggest that more could be done to make sex safer: among ‎ those reporting two or more partners, 40% of prisoners, 45% of migrants, 51% of miners, and 64% of truckers, reported condom use at last sex (SSG 2011)‎ military, 43% of ‎  In recent years 85%–95% of sex workers reported using condoms with paying clients, however,‎it‎was‎only‎64%‎with‎“petits‎amis”‎(cartographie‎PS/HSH),‎who‎may‎often‎be‎men‎in‎ ‎ ‎ regular partnerships with women of low risk behavior, and therefore link these inadvertently to ‎ higher risk sexual networks Table 3. Estimated protected and unprotected sexual acts, Niger (2012)‎ Est. total acts Est. % ‎ protected Est. protected acts Est. unprotected acts Regular acts ‎ 210.46 m ‎ 3%‎ ‎ 6.31 m ‎ 204.14 m Casual (non-‎regular) ‎ 5.47 m ‎ 40%‎ ‎ 2.19 m ‎ 3.28 m acts ‎ paid) Commercial ( ‎ 17.91 m ‎ 87%‎ ‎ 15.58 m ‎ 2.33 m acts‎ Total ‎ 233.84 m ‎ 10%‎ ‎ 24.08 m ‎ 209.75 m Sources: Estimated values based on reported condom use data (2012 DHS, 2011 SSG), the size of ‎ sub-populations (census data) and estimated sexual frequencies by type of partnership.‎ Note: Est. = estimated. Social marketing of condoms was launched in 2003 and has led to:‎  The‎‘Association‎Nigerienne‎de‎Marketing‎Social:‎Animas-Sutura’‎is‎the‎main‎role player, with medium ‎term financing from KfW (phase 1 2003–2008, phase 2, 2008–12, phase 3, 2013– 17)‎ 32 Niger’s‎HIV‎response: Targeted investments for a healthy future  At least 44 wholesalers who provide condom outlets with social marketing condom units  “ ‎ Foula”‎ condoms‎ dominate‎ the‎ market and are sold in packets of 3 for FCFA 75‎ . Animas- Sutura‎have‎distributed‎30‎million‎“Foula”‎‎ condoms  Targeted condom marketing and distribution located along main transport routes via kiosks and mobile ‎ vendors  Integrated promotion and sales together with socially-marketed malaria and reproductive oral‎contraceptive‎“Sutura”‎since‎2010)‎ health (RH) commodities (distribution of ‎ There are certain limitations to the scale-up of condom program activities:‎  Many adults disapprove promotion of condom use for youth aged 12–14 years—only 32% of females and 48% of adult males are supportive (about half in urban areas)‎ adult ‎  In pro-natalist societies, there are often cultural and perception barriers to condom use  In health facilities, condoms are often stocked as FP commodities and since the demand for ‎ condoms as a contraception method is very low, these condoms can expire at the clinic level instead of being made available for HIV/STI preventionDistribution, especially to remote areas, through the public health system (free condoms) or through the ‎ commercial channels (socially marketed condoms) remains a challenge in a vast country with overall weak push- and-pull factors for condom distribution ‎ ‎ Services for the Prevention of Mother-to-Child Transmission In 2011, Niger committed itself to accelerate progress in PMTCT with the goal of virtually eliminating new pediatric HIV infections by 2015. The e-MTCT strategy for 2013–15 ‎ presents a ‎ clear analysis of the bottlenecks and details a series of actions to reduce the MTCT rate of 27% in ‎ 2013 (2013 revised Spectrum data) to less than 5% in 2015. Key results to date are: ‎  Moving from 7 PMTCT pilot sites in 2003 to 651 PMTCT sites38‎in 2012‎  Increased assisted deliveries—a SRH objective—to 29.3% in 2012 (up from 18% in 2006), which ‎ can help women accessing PMTCT offered in maternity care  There was an increased attendance for the first prenatal visit (CPN1 = 83% up from 46% in 2006), however, a CPN4 of below 26% in 2012 severely limits PMTCT uptake  In 2012, 47% of pregnant women in need received ARVs through the PMTCT program Spectrum 2013; of 2,847 mothers in need, 1,334 received ARVs—895 for option A (revised ‎ B)‎ and 439 for option ‎  Diagnosis of exposed infants is almost nonexistent at 1% ‎  The‎“husband‎school/écoles‎des‎maris”‎tested‎with‎United Nations Population Fund (UNFPA) assistance for SRH purposes—this program is expected to help address some of the PMTCT scale-up since expectant mothers’ access to assisted delivery is largely challenges in ‎ dependent‎on‎their‎husband’s‎approval‎‎ 38 A prevention of mother-to-child transmission (PMTCT) site means PMTCT service provision in an Integrated Health Center, a hospital, maternity or any other health facility. HIV financing, HIV services and programs and their efficiency 33 The roll-out of PMTCT services is, in part, in alignment with epidemic evidence  Niamey—where DHS-female HIV prevalence is the highest—had reached the highest rollout PMTCT in 2012 (89% of eligible health facilities) (e-MTCT strategy 2013)‎ of ‎  Diffa—with second highest DHS-female HIV prevalence and third-highest ANC HIV prevalence—had achieved a 82% rollout to eligible health facilities  Tillaberi—with low DHS and ANC female HIV prevalence—had the slowest PMTCT roll-out of 44%; ‎Zinder also had relatively slow roll-out at 71% and somewhat lower DHS and ANC female HIV ‎ prevalence,‎ and‎ Agadez’‎ roll-out was at 76%, commensurate with its average female HIV ‎prevalence level  However, Dosso—with highest ANC HIV prevalence of 1.3%—had only reached a 70% roll- out  Tahoua—with second highest ANC HIV prevalence of 1.1%—had only reached 71%‎  Maradi—with 0% DHS-HIV prevalence and <1% ANC HIV prevalence—had reached a high of 85%‎ roll-out ‎ The efficiency of existing PMTCT sites is constrained by numerous organizational, technical and ‎financial challenges  Even though 176 new PMTCT sites were created in 2011 within existing integrated health centers, ‎ the number of pregnant women tested for HIV decreased that year due to a 6-month shortage of ‎ testing kits and reagents  Scale-up of PMTCT services suffers from a limited and irregular availability of human resources‎ or/and‎ trained professionals, and insufficient number of CD4 and PCR diagnostic tools  The high rate of utilization of CPN1 (83%) is explained by the additional benefits of the first visit—registration of the birth in the administrative system, and receipt of a mosquito net for malaria ‎ prevention. Subsequent attendance for pre-natal care is progressively lower  Renewed efforts to provide free delivery assistance has raised utilization, but it is still too low, and ‎ also affects the follow-up of HIV-exposed infants (in 2012, only 1%, or 669 children out of received the ARV treatment they needed, revised Spectrum data) ‎ 5,117 ‎  The‎ “PMTCT‎ cascade”‎ is‎ illustrated‎ in‎ Figure‎ 23,‎ showing‎ the‎ massive‎ effect‎ that the poor facility-based antenatal and delivery services has on PMTCT coverage ‎ utilization of ‎ 34 Niger’s‎HIV‎response: Targeted investments for a healthy future Figure 23. Summary of efficiency loss along the PMTCT cascade, Niger Sources: Data from NDHS 2012, PDS, Sentinel surveillance/Spectrum 2013; illustration by authors Note: ANC = antenatal care.‎ Program inefficiencies and bottlenecks have been diagnosed (presented in the e-MTCT strategy) ‎ and a series of actions has been identified to address them (Table 4 provides a summary). The two main ‎ avenues to scale up PMTCT are:  Improve access to treatment through the integration and decentralization of services, the delegation of ‎ART prescriptions to primary care providers and improvement of the technical CD4 count) ‎ platform (for PCR and ‎  Achieve community mobilization with a strong focus on the participation of men (potentially generalization‎ of‎ the‎ “écoles‎ des‎ maris”)‎ to‎ facilitate‎ the higher uptake of services by via ‎ women, improve the continuum of care, and reduce the number of lost-to-follow-up ‎ Table 4. Key gaps and bottlenecks at each step of the PMTCT cascade, and actions identified in the ‎ e-MTCT strategy, Niger PMTCT Current status Bottlenecks Actions identified cascade identified Improve ANC CPN1 >85%, CPN Cultural barriers, low  Improve support from community networks uptake 1–4 <45%. quality of services including men (école des maris) provided  Improve services via decentralization of tasks and better supply-chain management Improve HIV Less than 40% Lack of trained  Improve technical platform testing coverage personal, weak  Improve patient services through increased Improve CD4 Less than 25% of coverage of technical capacity of health care providers and testing and HIV positive platform for CD4 support from PLHIV networks process to women are under count, limited number  Accelerate decentralization and task launch ARV treatment of prescription and delegation at district level of prescription treatment follow up centers sites as well as CD4 counts  Improve supply chain management for treatments and testing materials. HIV financing, HIV services and programs and their efficiency 35 PMTCT Current status Bottlenecks Actions identified cascade identified Improve Less than 30% of Cultural barriers  Scale-up experience of Arlit district where uptake of pregnant women community mediators facilitated access to assisted benefit from PMTCT delivery delivery assisted by  Promote the involvement of men in SRH/HIV trained professional services, in particular via extension of the école des maris Improve follow CPoN = 24 % Lack of communication  Implement follow-up process of mother and up of exposed between the different child from screening to post delivery children role players (PMTCT-  Involve PLWHIV and communities (women, maternity-prescription traditional and religious opinion leaders) site) Improve early Almost nonexistent Lack of materials (only  Improve technical platform (e.g., PCR diagnostic for one PCR machine equipment) children currently available and  Accelerate decentralization and task shifting Improve follow Very low diagnosis located in Niamey) at district level for ARV prescription up at 18 for exposed infants Lack of providers  Integration of services with maternal and months (<10% of children qualified for newborn infant health services to ensure early are under ART) follow-up diagnosis Lack of integrated database to ensure follow-up and transfer between structures/sites Source: e-MTCT strategy 2013, MOH‎ . Note: ART = antiretroviral therapy; ARV = antiretroviral; CD4 = cluster of differentiation 4; CPN = consultation prénatale; e-MTCT = elimination of mother-to-child transmission; PCR = polymerase chain reaction; PLHIV = people living with HIV; PMTCT = prevention of mother-to-child transmission. HIV counseling and testing services The health sector offers HIV counseling and testing (HCT) in integrated health services, at PMTCT ‎ sites, blood transfusion centers and sites for TB testing and treatment. In 2008, there were 172 HCT ‎sites in the country. The following results are summarized:‎ Annual HIV testing is 4–5 percentage points higher in females than males  HIV testing is reported by 7–8% of females and by 2–3% of males (NDHS 2012)—the higher testing ‎rates in females are due to PMTCT-related HCT. Figure 24 illustrates the strata of men with high HIV ‎ prevalence and low HCT rates. Such analyses can inform better geographic targeting and ‎ promotion of HCT services (e.g., men who reside in Dosso and Diffa), and demographics for demand ‎ creation (e.g., older men)  Key populations have higher testing rates overall, with 76% of sampled FSWs, 61% of the military, ‎ 46% of miners, 41% of prisoners, 39% of truckers and 28% of migrants, reporting testing (2011 SSG)‎  In 2012, the Government allocated dedicated resources for provider-initiated counseling and ‎ testing (PICT) demonstrating the intention to target HCT at individuals who are: (a) already in contact ‎ with the health sector; and (b) may provide high yields of HIV-positive individuals, especially when ‎applying provider initiated counseling and testing (PICT) to sexually transmitted infection (STIs) and SRH services for adults, and child malnutrition screening‎ 36 Niger’s‎HIV‎response: Targeted investments for a healthy future (2012)‎ Figure 24. Men: Likelihood of recent HIV test by HIV status in socio-demographic strata, Niger ‎ Higher HIV, very low testing Source: NDHS 2012. Note: % = percentage. ‎ A proportion of women and men who have accepted HIV testing fail to obtain the test result, ‎ especially those living in poverty  Overall,‎ 11%‎ of‎ females‎ don’t‎ receive‎ the‎ test‎ result,‎ but‎ this‎ is‎ elevated‎ in‎ women‎ in‎ the‎ lowest‎wealth‎quintiles‎(>15%‎don’t‎receive‎the‎result)‎and‎also‎in‎those‎without‎school‎ three ‎ and living in rural areas (NDHS 2012)‎ education ‎  Although men undergo HIV testing less often, they are more likely to obtain the test results ‎ with only 5% not receiving the result. Failure rates are highest in Dosso and Maradi (>10%) the lowest wealth quintile (25%, small sample size)‎ and in ‎ There are reports of stock-out of HIV test reagents at service levels  According to key informants, stocks at the central level are sufficient, but there are reports of stock-‎ outs at the decentralized level (e.g., in the Maradi region)  The system of ordering commodities at the decentralized level is not working well, and the has had some challenges ‎ cold chain ‎  There are efforts to strengthen the supply chain logistics for HIV test reagents (and ARVs) with committees at the central‎ and‎ regional‎ level‎ (the‎ “comités‎ d’approvisionnement”), and specific funding ‎ to improve the cold chain ART services Public sector ART provision started in 2004 and was gradually scaled up. ARVs are provided free-of-‎ charge in the public health sector. The change in national ART treatment policy in 2010 meant that ‎ the‎ number‎ of‎ eligible‎ adults‎ “in‎ need”‎ doubled‎ and‎ relative‎ coverage‎ decreased (change from ‎ CD4 200 to CD4 350 treatment eligibility). The following results are highlighted:‎  In 2012 there were 15 sites providing ART services with 11,182 adult and pediatric patients register—see Figure 25 ‎ on the ‎ HIV financing, HIV services and programs and their efficiency 37  ART coverage in children was 13% and 51% in adults in 2012 (57% in adult women and 45% in adult ‎ men), based on Spectrum estimates of ART need, and statistics on ART provision in the program Figure 25. Number of ART patients and estimated ART need, Niger (2004–12)‎ Paediatric ART Adult ART 6000 20000 5000 4000 15000 3000 10,513 10000 8,924 2000 6,961 6,187 5000 1000 496 669 2,624 1,6282,232 2 15 94 139 123 258 335 100 680 0 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2004 2005 2006 2007 2008 2009 2010 2011 2012 Receiving ART in need Receiving ART in need Sources: MSP report 2012 on ART; Spectrum estimates on ART need (revised 2013)‎ Note: ART = antiretroviral therapy. ‎ RT retention is 70% at 12 months, 65% at 24 months and 59% at 60 months (2013 A program ‎ statistics). This is lower than the WHO-reported low and middle income country (LMIC) average of 86%, 82% and 72%, respectively ‎ (WHO 2013 progress report) ‎  Retention is similar among male and female ART recipients, and between adult and child recipients ‎ (2013 program statistics)  Confirmed deaths among ART patients are 3.9% at 12 months, 5.4% at 24 months and 7.7% months (loss-to-follow-up 26%, 30% and 33%, respectively) ‎ at 60 ‎ Estimation of drug needs suffer from weak routine data and are often based on program instead of actual service data‎ targets ‎  Supervision documents suggest that at several sites, statistics are not updated, or deaths of ART ‎ patients are not reflected in the statistics  Needs for ARVs, other medicaments and inputs are estimated annually by the APPR group which is ‎ technically supported by Solthis. Incomplete data from the ART sites make this exercise difficult. ‎However, attempts are being made to understand the number of new initiations, those lost-to-follow-up, ‎ mortality, and actual consumption from each site, with the largest volume of data coming from the ‎ Niamey-based ART sites  The resulting estimates are weak and are not supporting correct drug procurements, with ‎ overestimations leading to wastage, and underestimations causing shortages ‎  Data collection and sharing for informing procurement needs to be urgently strengthened with ‎ better data on ARV consumption and the treatment cohorts from each site Human resource constraints negatively affect the performance of the ART program  Only 3 out of 8 regions have a pharmacist for the management of the inputs into the ART (ARVs, other drugs and consumables) ‎ program ‎ 38 Niger’s‎HIV‎response: Targeted investments for a healthy future  Personnel in the pharmacies at the decentralized level are insufficiently trained or qualified and this ‎ affects the daily management of the supply chain, as well as limiting a full of the treatment cohort‎ understanding of the drug needs ‎ The procurement of ARVs is centrally managed, which allows for a harmonized “drug- procurement–to-match-identified-needs” approach. However, the purchase of ARVs, other ‎ and inputs is not harmonized‎ drugs, ‎  The ONPPC (Office National des Produits Pharmaceutiques et Chimiques) coordinates all ‎ procurements on behalf of the Global Fund, other financial partners and the government‎  The process of buying ARVs is not unified: Global Fund orders are processed through the Voluntary pooled ‎ procurement (VPP) service while government-financed orders use a tender process  Proforma invoices reveal a large differential in drug prices between these two purchasing ‎ mechanisms, and between international prices as per the Clinton Initiative ARV drugs purchased through the government system cost twice as much as the prices endorsed ‎by the Global Fund, the Clinton Initiative and WHO  For instance, the cost of one pack of 60 tablets of 3TC/AZT/NVP (150/300/200 mg) costs Euro 17.15 (11 260 ‎ FCFA) as per ONPPC proforma invoice, compared with Euro 8.59 as per Clinton Initiative. ‎ The median cost reported by the Global Fund for West and Central Africa (period 2005–12) for ‎ the same pack of tablets is Euro 8.74 (interquartile range, Euro 7.36— 10.58) or 5 735 FCFA (4 830–‎ 6 943)39 ‎  The cost of 60 3TC/AZT (150/300 mg) has a proforma invoice price of Euro 13.01 compared 6.75 as per the Clinton Initiative.‎ to Euro ‎  Other comparisons show the same pattern: 60 NVP 200 mg: Euro 4.40 versus Euro 2.30; 120 Euro 59.17 versus Euro 24.16 (ONPPC compared to Clinton Initiative) ‎ LVP/r ‎ The ARV distribution systems underperform and sometimes lead to storage problems and shortages at the sites ‎ ARV ‎  ONPPC is responsible for storage and distribution of ARVs, other drugs and reagents (and has a ‎ specific contract with CISLS to ensure the delivery of these commodities to the regions)  There are, at times, challenges with cold storage of stocks at the central level. Three vehicles‎are‎being‎financed‎by‎the‎Global‎Fund‎to‎reduce‎ONPPC’s‎reliance‎on‎ refrigerated ‎ the cold chain ‎ partners  At decentralized level, there are issues with the number of depots (only two, in Tahoua and ‎ Zinder), which means that all other ART sites need to be serviced directly by the central level every ‎ 3 months. Due to delays in the reporting of treatment data and drug requirements, ARV stock-outs ‎ have been experienced at the treatment sites. This can occur at times when the central depot is ‎ over-stocked  Storage conditions at the decentralized level can be sub-optimal, which can affect the quality products‎ of the ‎ 39 In USD, this is 11.40 per pack (interquartile range USD 9.60–13.80),‎as‎per‎the‎Aidspan‎report‎“Procurement‎cost‎ trends for Global Fund commodities: Analysis of trends for selected commodities 2005–12”,‎Aidspan‎working‎paper, April 2013. HIV financing, HIV services and programs and their efficiency 39  The‎“Comites‎d’approvisionnement”‎ are put in place at the central and decentralized level to remedy ‎ the supply chain problems, and the information system, by the Global Fund for HIV, TB and malaria ‎ supported by Solthis, will further support the adequate stocking and reagents for ART provision ‎ availability of ARVs, and ‎ Comparison of selected unit costs of HIV services The bottom-up costing exercise of the Resource Needs Model in 2012 provides an insight into unit ‎ costs, which can be compared to other countries using the same methodology (Table 5)  For some services, the unit costs are higher than for the comparison countries Togo and the Democratic Republic of Congo (RDC): The ‎ service package for SWs and MSM is costed at a higher level in Niger  Costs for HCT provision, STI treatment and examination of a woman in PMTCT services are lower in ‎ Niger compared to the other countries  First line ARV costs are higher in Niger than in Togo, but lower than in the RDC (PMTCT and ART services), ‎however, second line ARVs are significantly more expensive in Niger‎ Table 5. Comparison of some unit costs of HIV/AIDS services (Euro)‎ Services RNM Niger RNM Togo RNM RDC Prevention Cost per SW service package 59.8 22.2 22.2 Cost per MSM service package 59.8 22.2 54.5 Cost per STI episode treated 7.7 11.5 9.2 Cost per HCT client 3.8 10.7 32.2 Cost per women examined in PMTCT 3.8 6.1 - Cost per pregnant women treated with ARV 46.8 34.5 76.7 ART Annual cost first line 166.4 109.7 189.4 Annual cost second line 913.5 628.2 289.2 Note: 1 Euro = 655.957 CFA (fixed), 1 USD = 503.1 CFA; RNM = Resource Needs Model.‎ IEC/BCC services Information- education- communication/behavior change communication (IEC/BCC) activities are designed to increase risk perception and improve the use of HIV prevention ‎ methods and services such as condoms, HCT, PMTCT and STI care. ‎ IEC and BCC have been targeted at young people in primary and secondary schools and at youth in ‎ out-of-school settings for HIV and STI prevention purposes. IEC/BCC activities have resulted in the following levels of knowledge, norms and behaviors relating to ‎sexual behaviors:‎  Among youth aged 15–24, 65% of males and 46% of females cite condom use as a means to HIV infection (NDHS 2012) ‎ prevent ‎ 40 Niger’s‎HIV‎response: Targeted investments for a healthy future  ‎ 0% of young males and 65% of young females know that limiting sexual contacts to one HIV 8 ‎ negative partner reduces the risk of HIV  ‎ 4% of young females think that it is justified for a woman to refuse sex with her spouse if she 5 ‎ knows he has sex with other women, and 66% think it is justified for a women to ask for condom ‎ use if she knows her spouse has an STI  However, most women marry early, have much older partners and little or no school education—it is therefore questionable whether they are in a position to negotiate safe marital sex ‎ ‎  Among 15–24 year olds, 1.1% of males and 0.2% of females report having had 2 or more sex ‎ partners in the last 12 months, and 1.3% of young men report having paid for sex  ‎ he median age at first sexual encounter is 16 years for females and 23 years for males (self- T reported NDHS data ‎ 2012) To what extent are HIV services integrated or linked? ‎ The authorities are aware that linkage and integration are important, and strategy documents state ‎ the need for integrated supervision activities and improved integrated monitoring and reporting. At ‎ the same time, linkage and referral systems are being discussed and proposals for improvements ‎ have been presented. However, this is not a comprehensive review of integration, which encompasses: (a) ‎ intra-provider linkages (i.e., integration of services by one provider during one patient consultation); ‎ (b) intra-facility integration, which encompasses linkages between services by different providers, ‎ at a facility; and (c) inter-facility integration, which is about referrals between facilities. At each ‎ health facility (and standalone HIV-service-providing facility), the real challenge is ‎ how intra-provider, intra-facility and inter-facility integration can be optimized. In summary, the ‎ following observations are made:‎ General integration of, and linkages with HIV and health programs:‎  Social marketing of condoms is integrated with malaria and RH commodities through joint ‎ marketing and common sales channels, under the direction of Animas-Sutura  Free new-born delivery services have raised the assisted delivery rate and have indirectly helped raise PMTCT coverage  The roll-out and promotion of provider initiated counseling and testing (PICT) helps target HCT (which is very important in a low grade HIV ‎ epidemic), and provides opportunities for linking clients to prevention, treatment and care services ‎ (functional linkages between HCT earlier)‎ and pre-ART/ART is a priority for getting PLHIV into treatment ‎  The offer of selected free-of-charge health services by the public sector (especially family planning and ‎antenatal care) holds great potential for increasing the use of HIV and SRH services  Integration of HCT into child malnutrition screening works (3–4% of children were found to be HIV+, suggesting ‎that such targeted HCT is an important way to improve HIV case finding among children and ‎ eventually increase pediatric ART coverage)‎  Community-based‎agents‎(such‎as‎the‎“médiatrices‎communautaires”)‎and‎community-based ‎ organizations can help link people to health services and hold great promise in a country where ‎ 53% of people live outside the 5 kilometer (km) catchment area of health facilities. Priority tasks for such ‎ community workers should include the identification of risk pregnancies and the accompaniment ‎ of women, children and families enrolled in treatments (PMTCT HIV financing, HIV services and programs and their efficiency 41 regimens, ART, FP, TB, etc.). In ‎ Malawi, peer support and defaulter tracking were important for minimizing loss-to-follow-up of ‎ women on ART, as were extended counseling and ongoing support of women, and effective use ‎ of the patient monitoring data.40 ‎ The Rwanda experience with male and female health workers at ‎ community level demonstrates the importance of mobilization and demand creation at grassroots ‎ level  Full‎ integration‎ of‎ the‎ regional‎ HIV‎ coordination‎ into‎ the‎ health‎ sector’s‎ coordination‎ structures will ‎support the integration agenda and increase sustainability of both HIV coordination and HIV ‎ interventions at the decentralized level (delays in resource flow have disrupted HIV coordination in the ‎ past) ‎  Linkage and joint monitoring between public and the private service providers is poor despite the ‎country’s‎ multisectoral approach to HIV/AIDS strategy. For instance, 8% of men and women in ‎ Agadez and 3% of men and 6% of women in Niamey have medical insurance from an employer, ‎ but private sector HIV service provision is not being reported in the national statistics Linkages between a package of services for sex workers: ‎ Although it is more difficult to link a package of HIV services for sex workers to general health service delivery, the different components of FSW programs—the‎ bedrock‎ of‎ Niger’s‎ HIV‎ ‎ response—can and need to be better integrated in the following ways: ‎ ‎  Accessible STI services as a bedrock for ensuring an entry point to service provision for FSWs  The link between community-based activities for SWs and the adapted services at the facility level is ‎ weak, with insufficient collaboration between peer educators and clinic staff. This also link between VCT services and care/treatment ‎ applies to the ‎  Structural interventions need to be linked to a comprehensive service package for SWs to address ‎stigmatization and discrimination which currently limits service use (this also applies to male SWs)‎  Integrated HIV workplace interventions, especially in mining and oil companies, are seen as ‎ important complementary activities toward making sex work activity safer by providing targeted HIV prevention ‎ services to employees (who may be involved in commercial sex due situations) ‎ to their income and living ‎  The integrated services for migrants (OIM-Niger) are considered a useful model for delivering HIV ‎ and STI services in an integrated way to key populations by stakeholders in the national response.‎ HIV ‎ PMTCT: ‎  A strategy of rolling out PMTCT into all of the 857 Integrated Health Centers is being pursued (e-MTCT ‎ plan), which is underpinned by the global and regional plans for the virtual elimination of vertical ‎ transmission (despite low HIV prevalence in the general population, and correspondingly high costs ‎ per vertically transmitted HIV infection averted) ‎ 40 Shaffer, N., et al., Option B+ for prevention of mother-to-child transmission of HIV in resource-constrained settings: great promise but some early caution. AIDS 28:599–601. 2014.; Tenthani, L. et al., Retention in care under universal antiretroviral therapy for HIV-infected‎pregnant‎and‎breastfeeding‎women‎(‘Option‎BR’)‎in‎Malawi.‎AIDS‎ 28:589–598. 2014. 42 Niger’s‎HIV‎response: Targeted investments for a healthy future  Government documents emphasize the weak integration of maternal, neonatal and infant health ‎services (MNIH), SRH, FP, ART and PMTCT services (this includes supervision, i.e., visits by district ‎ cadres for supervision purposes miss including PMTCT services adequately)  This weakness leads to missed opportunities to offer the complete service package, including ‎ ARV-based interventions, to HIV+ women and children at the same time and in the same place ‎  The linkage of community based activities targeting men in order to increase utilization of ‎ antenatal care—and hence PMTCT—services‎has‎been‎piloted‎‎(“école‎des‎maris”)‎and‎this‎ regarded as a promising intervention, particularly for FP and PMTCT ‎ is locally ‎  The screening of HIV-exposed infants at 6 weeks is infrequent due to lack of integration of this ‎ activity into the treatment of the HIV-positive mother (and the underlying problem of very low ‎assisted delivery rates) ‎  The importance of linking FP, PMTCT and child survival efforts in Niger cannot be overstated. International‎experience‎has‎shown‎that‎“wanted‎pregnancies”‎benefit‎from‎better‎follow-up in ‎ ‎ terms of antenatal care, with direct implications on utilization of PMTCT (pers comm. Karusa, ‎ UNAIDS) Section 5. HIV investments: Impacts from past expenditures and best resource allocations in the ‎ current Strategic Plan period Impact of historical HIV expenditures ‎ Using Optima, the impact of past HIV investments was estimated by comparing the expected ‎ number of new HIV infections according to actual conditions with the estimated numbers under‎ the‎ counterfactual scenario of no funding for specific HIV programs. Local data were used to formulate ‎evidence-based assumptions on the relationship between funding, HIV outcomes, and program ‎ coverage (Annex A1.6). Optima was used to project epidemic trajectories according to ‎ the counterfactual scenarios, and compare these trajectories with the calibrated epidemic ‎ trajectory according to actual conditions, resulting in an estimation of the impact of the HIV program. ‎ Historical spending over the period 2007–12 helped reduce HIV incidence and therefore averted new HIV infections (Figure 26) ‎ ‎  Spending from 2007–12 averted close to 3,900 infections just during that period  The downstream effects of these prevented infections mean that by 2035, the spending from 2007–12 will have averted a total of 12,600 infections at a cost of Euro 3,500 (USD 4,700) ‎ per ‎ infection averted if only HIV program spending is considered (see Annex A1.3 for programs ‎ included), and Euro 4,600 (USD 6,200) if all HIV spending, management and included ‎ coordination is ‎  These second-order, future benefits of the six years of historical spending are therefore much higher than the HIV infections averted while spending was happening from 2007 to 2012 Figure 26. Number of infections averted by historical HIV spending from 2007–12, Niger (2007–35) 14000 12629 11178 HIV infections averted 12000 10000 8528 8000 6000 3893 4000 2000 0 2007-2012 2007-2017 2007-2025 2007-2035 Source: NASA 2007–12, available epidemiological and behavioral data from Niger, and drawing ‎ on the relationships between funding, HIV outcomes, and program coverage (Annex A1.6).‎ 43 44 Niger’s‎HIV‎response: Targeted investments for a healthy future The HIV incidence decline occurred across sup-populations and was most dramatic in those with ‎ highest HIV incidence in the past (Figure 27)  The largest HIV risk reduction occurred in FSWs, from an estimated 3.5% HIV incidence in 2000 which dropped to approximately 1.1% in 2012 ‎  Using the best available local and regional data on MSM, the estimated HIV incidence was at 0.9% ‎ in 2012 (reduced from 1.4% in 2000). This figure is uncertain given the scarcity of MSM data in Niger. However, the Optima-derived estimations point to this key demographic as a programming and strategic information strengthening ‎ priority population for ‎  The HIV incidence decline is expected to continue into the future as long as HIV program remains at current levels ‎ coverage ‎ Figure 27. Estimated infections by population and time, Niger (2000–35)‎ 14.00 12.00 New infections per thousand 10.00 8.00 6.00 4.00 2.00 0.00 CHLD FYTH MYTH FAD MAD FELD MELD FSW PRI MIG MSM UNI MIN TRU 2000 0.30 0.62 0.08 0.96 1.06 0.22 0.77 35.31 3.72 1.61 13.56 1.15 0.87 1.17 2012 0.10 0.12 0.02 0.24 0.35 0.07 0.28 11.26 1.42 0.58 8.60 0.44 0.30 0.42 2017 0.08 0.06 0.01 0.13 0.20 0.04 0.16 6.35 0.81 0.33 5.87 0.25 0.17 0.24 2025 0.06 0.03 0.00 0.06 0.10 0.02 0.08 3.32 0.43 0.17 3.97 0.13 0.09 0.12 2035 0.04 0.01 0.00 0.03 0.05 0.01 0.04 1.70 0.22 0.09 2.72 0.07 0.04 0.06 Sources: Niger epidemic, demographic, behavioral and service data in the populated Optima ‎model, Annex 1 Section A1.2.. Note: Estimations for MSM are based on assumed HIV prevalence data using information from the sub-‎ region and the resulting HIV incidence estimates for MSM are therefore uncertain; The Y axis is truncated, HIV incidence in FW in 2000 are not fully displayed; CHLD = children aged 0‎ –14; FYTH = females aged 15–24; MYTH = males 15–24 years; FAD = females aged 25–49; MAD = ‎ males aged 25–49; FELD = females aged 50+; MELD = males aged 50+; FSW = female sex workers; ‎ PRI = prisoners; MIG = migrants; MSM = men who have sex with men; UNI = uniformed security/defence personnel; MIN = mine workers; TRU = truckers.‎ Best resource allocation in the current HIV strategic plan period One of the key interests of this analytical study was to use the Optima model to determine the ‎ allocation of HIV resources or spending required that best meet the objectives of HIV incidence or ‎ DALYs reduction (see detailed objectives in Section 1). This optimization was done by combining the ‎information on the relationship between costs and outcomes (Annex A1.6) with a simple ‎ optimization algorithm that found the best allocation of resources to meet the objective at hand; in ‎ this case, minimization of either HIV incidence or DALYs until 2025 in order to capture longer- term ‎benefits of investments.‎ ‎ urrent strategic plan period HIV investments: Impacts from past expenditures and best resource allocations in the c 45 There are specific shifts in budget allocation that are required to optimize program impacts on HIV ‎ incidence and morbidity/mortality (DALYs)—Figure 28. Note that condoms are included in the ‎ intervention packages for the different populations; some community mobilization activities are ‎part of the PMTCT package; HCT is included in ART, SW and PMTCT service packages; and SW ‎ interventions include mobilization activities‎  Compared to the 2012 spending pattern, spending to minimize either incidence or DALYs would ‎prioritize ART, PMTCT and FSW spending, while reducing spending on the low-risk populations ‎ (primarily the general population) ‎  In‎Niger’s‎epidemic‎context,‎the optimal spending allocations look virtually the same for best ‎ reductions in HIV incidence or DALYs, i.e.: ‎ - Increased spending on ART from 31% currently to circa 50%‎ - Increased spending on PMTCT from 11% currently to circa 19%‎ - Increased spending on FSW interventions from 1% currently to circa 4%‎  These shifts in spending reflect better resource targeting to populations where the ‎ majority of new HIV infections occur (especially true for FSWs and their sexual contacts); in addition, ‎ they emphasize the relatively high effectiveness of ARV treatment as prevention, since PLHIV on ‎treatment have significantly reduced infectiousness.41 Figure 28. Current versus optimal budget allocations for minimizing incidence (left) and DALYs ‎ (right) over the period 2014–25, Niger Sources: Spending data from NASA; Niger epidemic, demographic, behavioral and service data in ‎the populated Optima model. Note: Current = 2012 spending pattern. OVC, blood, PEP, waste, ‎ and management costs were kept fixed; Condoms are part of the HIV intervention packages for ‎ the different populations. FSW = Female sex workers, MARPs = most at risk populations, a combined group of migrants, men having sex with men, uniformed security/defence personnel, mine workers and ‎ prisoners, ‎ truckers; LRP = low risk populations, a combined group of general population females and males aged 15+ years; OVC = orphan and vulnerable children; PEP = post-exposure prophylaxis. ‎ 41 Optima and Spectrum used the factor 0.80 for reduction in HIV transmission when on ART (modified from the Spectrum default value of 0.92), based on programmatic effectiveness of ART on viral load reductions in several African ART interventions. 46 Niger’s‎HIV‎response: Targeted investments for a healthy future Optimization of the 2014–17 budget (annual average of Euro 4.8 million or USD 6.5 million) for incidence reduction averts an estimated 8,900 additional HIV infections from 2014–25 ‎ HIV ‎  Most of the benefit is obtained by increasing funding for FSWs (7,000 extra infections are averted, ‎across all population groups, counting direct and indirect/secondary transmission events), and by ‎increasing funding for PMTCT (which reduces MTCT by 50%, or 1,400 extra ART (3,400 extra infections are averted) ‎ infections are averted) and ‎  These increases require reducing funding for low-risk populations, leading to 3,000 additional ‎ infections, but the net benefit is still very positive (7000 + 1,400 + 3,500 - 3,000 = 8,900 infections averted) If HIV funds are either very limited or more abundant, allocations can be guided by an investment ‎ cascade which helps prioritize expenditures by HIV impact—Figure 29  Given limited funds (below that of what has actually been acquired for the NSP period‎ 42‎ ), PMTCT and ‎ ART should be given first priority, followed by FSW spending—this explicit rationing is not desirable ‎ since easily preventable HIV infections will occur within such a tight funding context‎  With larger budgets (notably larger than the average budget over the period 2014–17), programs ‎ for the low-risk population should be given significant funding with spending patterns similar to the ‎ 2012 distribution, except with a far greater proportion being allocated to PMTCT Figure 29. HIV investment cascade, Niger (2014–17)‎ 20000 18000 16000 Spend (€ '000s) 14000 12000 10000 8000 6000 4000 2000 0 100% 150% 200% 2012 400% 20% budget 50% budget budget budget budget expenditure budget PMTCT 629 1356 1265 2567 3716 1515 7045 ART 0 701 3193 4206 4462 3588 6509 LRP 0 0 0 0 871 4165 4560 MARPs 27 28 25 81 150 407 482 FSW 295 293 274 281 313 55 431 Sources: Spending data from NASA; Niger epidemic, demographic, behavioral and service data in ‎ the populated Optima model.‎ Note: 100% budget = Euro 4.8 m/USD 6.5 m per year, based on acquired funding by the end of 2013; FSW = ‎ female sex workers; MARPs = most at risk populations, a combined group of prisoners, migrants, men having sex with men, ‎ uniformed security/defence personnel, mine general population females and males aged 15+ years. ‎ workers and truckers; LRP= low risk populations, a combined group of ‎ 42 The yearly average of Euro 4.8 million (USD 6.5 million) is skewed toward the earlier NSP years. Very low funding had been acquired for the years 2016–17 at the time of analysis, so the assumptions of 50% or 20% of the yearly average in the investment cascade are relevant. In turn, 2014 had a higher than average amount of acquired funding, and the investment cascade can provide guidance on prioritization in years with a strong availability of funding. ‎ urrent strategic plan period HIV investments: Impacts from past expenditures and best resource allocations in the c 47 The reallocation of available funds results in a 12% additional reduction in new HIV infections over ‎the period until 2025 compared to a continuation of the 2012 expenditure pattern, despite the fact that available funds are only roughly half the 2012 budget—Figure 30.‎ scenarios, Niger (2013–25) Figure 30. Cumulative infections that would result from different spending ‎ ‎ Sources: Spending data from NASA; Niger epidemic, demographic, behavioral and service data in ‎ the populated Optima model. Note: No spending = a continuation of 2012 spending, and optimized spending at various budget levels ‎ relative to expected funds available 2014–17; ‎ 2012 budget = 2012 expenditure pattern; 100% budget = Euro 4 ‎ .8 m/USD 6.5 m per year, based on acquired funding by end 2013; Drug costs are kept constant ‎ over time, other costs increase in line with GDP per capita; CHLD = children aged 0 ‎ –14; FYTH = females aged 15–24; MYTH = males 15–24 years; FAD = females aged 25–49; MAD = ‎ males aged 25–49; FELD = females aged 50+; MELD = males aged 50+; FSW = female sex workers; ‎ PRI = prisoners; MIG = migrants; MSM = men who have sex with men; UNI = uniformed security/defence personnel; MIN = mine workers; TRU = truckers.‎ Optima modeling also provided some broad estimates of additional costs as a result of raising the ‎ ART eligibility threshold to CD4 500, and of the savings gained with more modest management cost ‎increases (Figure 31). Compared to the default expenditure for meeting NSP coverage targets, ‎ increasing the treatment CD4 threshold to 500 would increase ART coverage cost by an estimated ‎ 12% over the four NSP years. By 2017, ART coverage cost would be an extra Euro 2.7 million (USD ‎ 3.6 million) annually. Management costs are currently expected to grow far faster than inflation ‎ (4% population growth, +3% currency inflation). If these costs grew at the inflation rate instead, Euro 1.7 ‎ million (USD 2.3 million) could be saved over the four NSP years, as shown in Figure 31. ‎ 48 Niger’s‎HIV‎response: Targeted investments for a healthy future 2013–17 Figure 31. Minimum budgets required to meet NSP coverage targets, Niger ‎ Spending (€m) 25 20 15 10 5 0 2014 2015 2016 2017 Default (€m) 10.5 13.1 16.6 20.2 Higher CD4 threshold 11.7 14.7 18.6 22.9 Lower management costs 10.5 13.0 16.0 19.2 Sources: National Strategic Plan 2013–17, relationships between funding, HIV outcomes, and ‎ program coverage (Annex 1 section A1.6). Note: Bar groups show different scenarios, Default (current conditions), CD4 threshold for treatment of 5 ‎ 00 instead of 350, and lower management costs; Spending: = Euro, million; Drug costs are kept constant over time, other ‎ costs increase in line with GDP per capita.‎ Section 6. Financing needs and fiscal space implications of HIV investments NSP 2013–17: Financing needs and gaps The NSP 2013–17 envisages a reduction in the number of new HIV infections by 50%, and a further ‎ expansion in access to ARV treatment to at least 80% of eligible people. Annual spending ‎ under the NSP is projected to increase steeply from Euro 8.8 million (USD 11.7 million) in 2013 to ‎ Euro 18.0 million (USD 23.9 million) in 2017. Table 6 presents the acquired NSP funding, estimated ‎needs and funding gaps ‎  For the remaining four years of the NSP, only 35% of the total NSP-projected spending of Euro 69.2 ‎ million/USD 93.1 million had been secured by December 2013 (leaving a funding million/USD 60.4)‎ gap of Euro 44.9 ‎  While NSP-projected spending was fully acquired for 2014, only 31%, 6% and 1.5% had been acquired for the years 2015, 2016 and 2017 respectively ‎ ‎ Table 6. NSP financing acquired, needed, and financing gap (2014–17)‎ (2013) 2014 2015 2016 2017 2014–17 Acquired resources (6.3) 11.0 4.0 2.8 0.3 24.4 (Euro mn) Required (Euro mn) (8.8) 10.5 13.1 16.6 20.2 69.2 Financing gap (Euro mn) (2.5) -0.5 9.1 13.8 19.9 44.9 Sources: Cartographie des fiancements, 2013; PSN 2013–17‎ Note: mn = million Concerning the longer-term financial consequences of the national response to HIV/AIDS, the cost projections under the NSP do not provide the long term picture.  On the expenditure side, the expansion of access to treatment has consequences that extend well beyond 2017. Individuals who newly access treatment in 2013–17 will require ‎ treatment well into the future—the expansion in access to treatment therefore commits financial ‎ resources beyond the NSP period  From the savings side, the increased investments in HIV prevention and treatment reduce future ‎ spending needs that would be associated with future incidence. A lower number of HIV infections reduce the ‎ demand for HIV/AIDS-related services and resulting costs in future years. These future savings need to be taken into ‎ account when evaluating the consequences‎‎ of the NSP absorbing fiscal space The projected costs of the national response to HIV/AIDS that are based on the costing of the NSP ‎estimates provided by authorities through 2017, but are extended through 2030, are Figures 32 and 33‎ summarized in ‎ 49 50 Niger’s‎HIV‎response: Targeted investments for a healthy future  Where the NSP costing assumes an annual inflation rate of 3%, the longer-term estimates have ‎been‎specified‎in‎“real”‎terms,‎i.e.,‎without‎inflation.‎This‎change‎facilitates‎comparisons‎ over long ‎ periods  Cost assumptions for prevention spending are aligned with the NSP assumptions through ‎ 2017. Thereafter, prevention spending is assumed to increase in proportion to the size of the relevant population (e.g., all adults, or—for PMTCT—HIV-positive women), keeping coverage ‎ of the relevant interventions constant‎ rates ‎  For treatment, unit costs were set at set at Euro 310 (USD 400) for first-line ART and Euro 1,110 ‎(USD 1,430) for second-line ART, based on NASA expenditure data for the ART program and the ‎ number of patients treated in 2012. This is higher than the assumptions in the NSP/Resource ‎ Needs Model costing—which covered ARV drug costs only—in order to include additional costs of a ‎ comprehensive ART service such as laboratory monitoring costs, psychosocial counseling, ‎ nutritional support and clinical out-patient services‎  Social mitigation spending is taken from the NSP costing through 2017, and is assumed to grow in ‎ proportion to the relevant population (PLHIV and AIDS orphans, respectively) thereafter‎  Overhead expenses are set at 13% of direct costs (or 12% of total costs), in line with the assumptions underlying the NSP.43 ‎ ‎ Figure 32. Financial costs of HIV/AIDS, Niger (2010–30, Euro million, 2013 prices) ‎ 35 30 Euro (million, 2013 prices) 25 Overhead 20 Mitigation 15 Treatment 10 and care Prevention 5 0 2013 2018 2023 2028 Source: Fiscal‎framework‎model‎for‎Niger‎and‎authors’‎estimates.‎ and Overhead represents overhead costs. ‎ Note: The area between Mitigation ‎ 43 The overhead expenses mark-up that is assumed in the NSP costing is slightly higher (14% of direct costs), but has been adjusted to reflect that the current analysis captures the full cost of care and treatment. Financing needs and fiscal space implications of HIV investments 51 Figure 33. Financial costs of HIV/AIDS, Niger (2010–30, percent of GDP)‎ 0.35 0.30 Overhead 0.25 Cost as % GDP Mitigation 0.20 0.15 Treatment and care 0.10 Prevention 0.05 0.00 2013 2018 2023 2028 Source:‎Fiscal‎framework‎model‎for‎Niger‎and‎authors’‎estimates.‎ Note: The area between Mitigation ‎ and Overhead represents overhead costs.; GDP = gross domestic product. The projected costs of the national response to HIV/AIDS increase steeply during the NSP and beyond‎ period ‎  The largest driver of costs is increasing HIV prevention spending. This reflects increasing ‎ coverage rates of HIV prevention interventions during the NSP, but also high population growth ‎(according to the World Population Profile 2012, the adult population (15+ years) of grow by 3.6% annually and 88% overall between 2013 and 2030)‎ Niger is projected to ‎  The costs of HIV/AIDS care and treatment increase steeply, as treatment coverage ‎ is assumed to increase to 80% during the NSP. Thereafter, the number of people on treatment projected to keep increasing—a consequence of people who newly need and access is ‎ treatment, ‎ coupled with low mortality rates when on treatment (see in following section and footnote)‎  While the costs of the national response to HIV/AIDS are projected to increase in absolute terms, ‎ the costs peak at 0.29% of GDP in 2017, and decline to 0.22% of GDP in 2030 (assuming that GDP ‎ growth rates above 5%, as projected in the IMF World Economic Outlook, until 2018, can be maintained over this period) Anticipated impacts of the investments under the NSP 2013–17‎ The scaling-up of HIV prevention and treatment envisaged under the NSP results in a steep decline ‎ in HIV incidence and mortality (Figure 34). While the number of PLHIV declines overall, the number ‎of people receiving ARV treatment increases44 (Figure 35).‎ 44 The emphasis of the analysis is the long-term financial consequences of the HIV/AIDS program and especially of new HIV infections. While access to second-line treatment plays a small role in Niger at present, the importance may increase over the next year, and (because of higher unit costs) contribute significantly to long-term cost. The current projections therefore make a small allowance for transition to second-line treatment (1% of people on first- line treatment annually). 52 Niger’s‎HIV‎response: Targeted investments for a healthy future Figure 34. Modeled HIV incidence and AIDS deaths, Niger (2010–30)‎ 3,000 Estimated new HIV infections/AIDS deaths 2,500 HIV infections 2,000 AIDS deaths 1,500 1,000 500 0 2010 2015 2020 2025 2030 Source:‎Fiscal‎framework‎model‎for‎Niger‎and‎authors’‎estimates Figure 35. People living with HIV and ART, Niger (2010–30)‎ Source:‎Fiscal‎framework‎model‎for‎Niger‎and‎authors’‎estimates Longer-term costs of responding to HIV and future savings Until 2030, the costs of the national response to HIV/AIDS will be dominated by prevention spending ‎and the demand from people already living with HIV as of 2013 (Figure 36), for three reasons:‎  Low mortality on HIV treatment means that people already receiving treatment in 2013 tend remain on treatment over a long period ‎ to ‎ Financing needs and fiscal space implications of HIV investments 53  HIV infections translate into a demand for HIV/AIDS-related services and spending slowly‎  Because the NSP is preceded by a period in which HIV incidence was already in decline, the number of new HIV infections—relative to the number of people who were already HIV ‎ positive in 2013—is ‎ low at the outset of the NSP GDP)‎ Figure 36. Projected HIV costs caused by new and past HIV infections, Niger (2013–30, percent of ‎ 0.35 Total projected costs 0.30 Projected costs as % of GDP Proj. costs excluding 0.25 post-2013 HIV infections Population-based 0.20 spending 0.15 Costs of new infections from 0.10 2013 are represented by vertical difference between total 0.05 projected costs and the projected costs excluding post- 0.00 2013 2018 2023 2028 Source:‎Fiscal‎framework‎model‎for‎Niger‎and‎authors’‎estimates. Note: Population-based spending denotes spending that cannot be attributed to people living with HIV (i.e., most ‎ prevention spending), and a proportional share of the overhead expenses; GDP = gross domestic product; Proj. = projected. Nevertheless, new HIV infections result in substantial costs, which extend over a long period ‎ (Figure 37). Note that this projection covers a lengthy time horizon, and projections become much less ‎ precise as they extend further into the future. The projection does however illustrate the long- term‎costs‎and‎“treatment‎liability”‎caused‎by‎new‎infections.‎ ‎ One new HIV infection in 2013 results in a cost of about Euro 2,500 (USD 3,300) for both women, applying a discount rate of 5%‎ men and ‎  The estimates are sensitive to the assumptions made for mortality on treatment. The Spectrum ‎ files for Niger assume very low AIDS-related mortality rates for people receiving ARV treatment, ‎and the estimates may overstate the long-term costs45 ‎  Even if mortality rates on ART were higher than those assumed in Spectrum, the long-term costs ‎resulting from one additional HIV infection would be substantial. An additional 1% annually in AIDS-‎ related mortality reduces the estimated cost resulting from each new HIV infection by about 15%.‎  The projected costs of one new infection differs between sexes principally due to PMTCT (leading ‎ to higher projected cost for females in the short-term), transition to treatment rates (for technical ‎ reasons regarding assumptions on treatment coverage, the path of HIV incidence, and certain ‎features of Spectrum, fewer women transition to treatment, therefore the female curve is lower in the ‎ mid-term), and survival (it is assumed that survival for female PLHIV is higher, so the cost curve ‎declines more slowly in the long-term for females) 45 E.g., for individuals initiating treatment at a CD4 count of 100–199, the Spectrum default parameters for West Africa assume an annual HIV-related mortality of 0.1%–1.1% for women, and 0.6%–1.6% for men. 54 Niger’s‎HIV‎response: Targeted investments for a healthy future Figure 37. Cost of one new infection occurring in 2013, Niger 200 Projected cost (Euro) 150 100 Male Female 50 0 2013 2018 2023 2028 2033 2038 2043 2048 2053 2058 2063 ‎‎ Source:‎Fiscal‎framework‎model‎for‎Niger‎and‎authors’‎estimates The decline in HIV incidence achieved during the course of the NSP (and over the longer period ‎ until 2030) contributes substantially toward containing the long-term costs of the HIV program  The decline in HIV incidence between 2013 and 2017 results in reduced long-‎ term spending commitments, a savings equivalent to Euro 6.3 million (USD 8.4 million)‎  The increase in HIV/AIDS-related spending over the same period totals Euro 16.5 million (USD 22.0 million)‎ ‎  Assuming that the number of new HIV infections would have remained constant without the scaling-up of prevention spending, these estimates imply that—in addition to the obvious ‎ gains—the financial savings achieved by the increases in investment in HIV prevention health ‎ one-third of the additional scaling-up of long-term prevention and treatment costs‎ offset about ‎ Niger’s‎HIV‎response: Targeted investments for a healthy future | Annex A1.1–A1.7 55 Annex 1. Optima modeling: Analytical Framework and Methods A1.1. Optimizing budget allocations for impact The analysis is a result of estimating costs based on cost-coverage curves, and optimizing budget ‎ allocations to meet a given objective. Both minimizing new HIV infections and minimizing DALYs ‎were included (see main report for research questions). ‎ In the optimization function, only intervention program costs are reallocated (optimized), while ‎ other costs like management/coordination expenditures are fixed and are not included in the ‎ optimization of allocations. Drug costs are kept constant over time, while other costs increase in GDP per capita.‎ line with ‎ A1.2. Population groups and transmission pathways The following populations were defined for the Optima analysis (maximum of 14 are possible):‎ 1. Children aged 0–14 years (CHLD)‎ 2. Females aged 15–24 years (FYTH)‎ 3. Males aged 15–24 years (MYTH)‎ 4. Females aged 25–49 years (FAD)‎ 5. Males aged 25–49 years (MAD)‎ 6. Females aged 50+ years (FOLD)‎ 7. Males aged 50+ years (MOLD)‎ 8. Female sex workers (FSW)‎ 9. Prisoners (PRI)‎ 10. Migrants and mobile populations (MIG)‎ 11. Men who have sex with men (MSM)‎ 12. Uniformed security (UNI)‎ 13. Mine workers (MIN)‎ 14. Truckers (TRU)‎ We assume HIV transmission between these populations only occurs through sexual partnerships. ‎ All male populations were assumed to partner with all female populations, except that males did ‎not pair with females older than themselves, and males paired with females, at most, one age ‎ stratification younger than themselves (e.g. males 50+ did not form partnerships with females <24). ‎MSM partnered only with each other.‎ 56 Annex A1.1–A1.7 | Niger’s‎HIV‎response: Targeted investments for a healthy future A1.3. HIV programs included in the analysis Eight basic HIV prevention and treatment programs were distinguished:‎ 1. Targeted FSW intervention package 2. Prevention services targeted at MARPs like prisoners, military personnel, truckers, migrants, MSM (IEC, BCC, ‎ condom promotion, HIV counseling and testing)‎ 3. Prevention services for non-MARPs 4. Public sector condom distribution ‎ 5. Social marketing of condoms 6. PMTCT 7. Antiretroviral treatment services 8. OVC support A1.4. Overview of the Optima methodology The Optimization and Prioritization Tool Optima (formerly Prevtool) is an extremely flexible ‎ population-based HIV model. Optima can assess HIV epidemic trends, resource needs, the cost- ‎ effectiveness of past programs, and the impact of potential future programs using a detailed ‎ mathematical model of HIV transmission and disease progression. Optima can perform detailed ‎ country-specific HIV program evaluations, and can build on the results from other models (like Goals) ‎to assess allocative efficiency and the resources required to achieve complex objectives varying periods.‎ over ‎ Up to 14 arbitrarily defined population groups are available within Optima to describe the specific ‎ characteristics of an HIV epidemic. Optima incorporates a model of HIV transmission and ‎ progression using a coupled system of ordinary differential equations to track the movement of ‎ people between health states and population groups. Individuals are assigned to a given ‎ population based on their dominant risk. The model incorporates the effects of different sexual ‎ partnership types, condom use, infection stage, CD4 count categories, homosexual and ‎ heterosexual partnerships, the extent of male circumcision, diagnosis, first and second line anti- retroviral treatment, and treatment failure, on the transmission of HIV and infection progression.‎ ‎ Optima requires detailed demographic, epidemiological, behavioral, clinical, and health economic ‎ data which is entered using an Excel spreadsheet. Data entry is flexible, allowing for everything from a ‎separate data point for every population for each year, or a single data point for all populations ‎over the entire simulation period (Optima assumes that parameters with no data are zero). Using ‎available data, we calibrated Optima to accurately represent past and current trends in Niger’s‎ HIV‎ ‎ epidemic. The model uses healthcare costs and disability-weights specified by CD4 count to ‎ perform cost-effectiveness and return-on-investment calculations. ‎ Given available coverage, spending and outcome data for implemented HIV prevention and ‎ treatment programs, Optima can determine the optimal allocation of resources or the amount of ‎ spending required to meet specified objectives (e.g., minimizing infections, minimizing AIDS ‎ deaths) or a combination of objectives. The model also enables additional programmatic scenarios ‎ representing the potential introduction of new programs or changes to current program implementation as required. ‎ ‎ Niger’s‎HIV‎response: Targeted investments for a healthy future | Annex A1.1–A1.7 57 A1.5. Optima calibration to the HIV epidemic in Niger We used all available demographic, epidemiological, behavioral, and clinical data to calibrate the ‎ Optima model to the HIV epidemic in Niger (shown in Figure 38 and Figure 39).‎ ‎n general, prevalence in Niger is declining rapidly in most population groups, due to both I ‎ reductions in incidence and deaths of people currently living with HIV. Exceptions include MSM, ‎ migrants, and prisoners, for whom there are not sufficient data to confidently determine epidemic trends; however, current indications are that prevalence is relatively stable among these groups.‎ ‎ Annex Figure 1. Calibration of model to the HIV epidemic in Niger‎ 58 Annex A1.1–A1.7 | Niger’s‎HIV‎response: Targeted investments for a healthy future Annex Figure 1. (continued) Note: Black discs represent available data for ‎ HIV prevalence. Lines attached to these discs represent uncertainty bounds, where available. The ‎ solid curve is the best fitting simulation and the light band represents the 95% confidence interval ‎ for the model outputs. MSM prevalence was assumed based on data from the West Africa region, in ‎ the absence of local data. Annex Figure 2. Calibration of the model to the HIV epidemic in Niger Note: Black discs represent available data for ‎ the number of people on first and second line anti-retroviral treatment. Lines attached to these ‎ discs represent uncertainty bounds, where available. The solid curve is the best fitting simulation ‎ and the light band represents the 95% confidence interval for the model outputs. Niger’s‎HIV‎response: Targeted investments for a healthy future | Annex A1.1–A1.7 59 A1.6. Comparison between Optima and Spectrum estimates Part of the calibration process involved aligning the future epidemic trends projected by Optima ‎ with estimates from the Spectrum software package. Figure 40 shows how the projections from ‎ Optima compare to Spectrum estimates for overall prevalence, overall incidence, mother-to-child ‎ transmission and AIDS deaths (revised Niger Spectrum version of 2013).‎ Annex Figure 3. annual Comparison of annual new HIV infections, annual mother-to-child transmissions, and ‎ AIDS deaths estimated by Optima and Spectrum over 2000–20 Note: Prevalence from Optima is ‎the prevalence in the overall population while the HIV prevalence from Spectrum is for population 15 years old and over. ‎ Optima captures the same epidemiological trends as that obtained from the Spectrum software ‎ package (revised Niger Spectrum version, 2013). There are differences, however, in the projections ‎ from each model, particularly for overall HIV prevalence and annual HIV incidence. The reason for ‎these differences between the models—despite using the same demographic, behavioral and clinical data inputs—is due to differing model structures and epidemiological, ‎ modeling ‎assumptions. ‎ Both Spectrum and Optima are deterministic population-based models using similar modeling ‎ methods to describe HIV epidemics and assess the effectiveness of HIV prevention programs. Each ‎model describes the transmission of HIV and the resulting change in CD4 count within those ‎ infected. Spectrum uses a greater number of CD4 count categories but Optima includes 60 Annex A1.1–A1.7 | Niger’s‎HIV‎response: Targeted investments for a healthy future diagnoses ‎ and the impact of testing, first and second line ART and treatment failure, and the impact of ‎ treatment on CD4 cot. Mortality of HIV-positive people is dependent on the CD4 count with Spectrum using lower death rates than Optima. ‎ in both models ‎ The major difference between Spectrum and Optima is the underlying structure of the models. ‎ Spectrum uses a generic population structure dividing the overall population into low, medium, ‎ and high-risk groups and separate most-at-risk populations with associated partnerships. In this ‎ analysis Optima is set up to directly match the characteristics of the Nigerian HIV epidemic and ‎ include all populations driving the HIV epidemic (many of which cannot be included in Spectrum)—with transmission able to match the partnership mixing in Niger, rather than using partnerships. ‎ pre-specified ‎ A1.7. Relationships between program costs and effects on behavior and program coverage‎ A central component of our analyses is the relationships between the cost of HIV prevention ‎ programs and the resulting population coverage and behavioral outcomes. Such relationships are ‎ required to understand how incremental changes in spending ultimately affect HIV epidemics and ‎ determine the optimal funding allocation. For each population, we derived a set of relationships ‎ directly linking estimated funding to coverage and associated behavioral data. To produce these relationships‎we‎used‎an‎ecological‎“top-down”‎approach‎to‎relate‎program‎cost‎and‎outcomes.‎ ‎ ‎ Indirect costs have no direct impact on HIV transmission parameters, but changes to HIV programs ‎ may affect these costs, for example in the supply of additional condoms. The proportion of funding ‎ allocated to indirect costs is assumed to remain the same as the overall HIV program budget ‎ changes. A limitation of our approach is the assumption that all changes in behavior are assumed to ‎ be due to changes in program funding.‎ A1.8. NASA expenditure data used in Optima All NASA data and worksheets came from the national assessments covering the years 2007 to 2012, provided to the authors by the CISLS. ‎ ‎ Niger’s‎HIV‎response: Targeted investments for a healthy future | Annex A2.1 61 Annex 2. Optima modeling: Additional results on incidence estimations up to 2035‎ Annex Table 1. Estimated new HIV infections by population for the best-fitting simulation (2012–35)‎ Population Group 2012 2015 2017 2025‎ 2035 Children 0–14 years‎ ‎ 806‎ 714‎ ‎ ‎ 659‎ ‎ 480‎ 338‎ ‎ Females 15–24 years‎ ‎ 184‎ 137‎ ‎ ‎ 117‎ ‎ 75‎ 54‎ ‎ Males 15–24 years‎ ‎ 17‎ 13‎ ‎ ‎ 11‎ ‎ 8‎ 6‎ ‎ Females 25–49 years‎ ‎ 499‎ 379‎ ‎ ‎ 327‎ ‎ 220‎ 161‎ ‎ Males 25–49 years‎ ‎ 543‎ 425‎ ‎ ‎ 376‎ ‎ 269‎ 202‎ ‎ Females 50+ years 56‎ ‎ 43‎ ‎ 38‎ ‎ 27‎ ‎ ‎ 20‎ Males 50+ years 237‎ ‎ 184‎ ‎ 164‎ ‎ 118‎ ‎ ‎ 89‎ Female sex workers 350‎ ‎ ‎ 273‎ ‎ 241‎ 173‎ ‎ 132‎ ‎ Prisoners 17‎ ‎ ‎ 14‎ ‎ 12‎ 9‎ ‎ 7‎ ‎ Migrants/mobile populations 464‎ ‎ ‎ 365‎ ‎ 324‎ 233‎ ‎ 176‎ ‎ Men having sex with men 96‎ ‎ ‎ 84‎ ‎ 80‎ 74‎ ‎ 75‎ ‎ Uniformed security 26‎ ‎ ‎ 21‎ ‎ 18‎ 13‎ ‎ 10‎ ‎ Mine workers ‎ 30‎ ‎ 24‎ 21‎ ‎ ‎ 15‎ 11‎ ‎ Truckers ‎ 27‎ ‎ 21‎ 19‎ ‎ ‎ 13‎ 10‎ ‎ Total estimated HIV incidence ‎ 3,352‎ ‎ 2,697‎ 2,407‎ ‎ ‎ 1,727‎ 1,291‎ ‎