2012年-世界发展银行全球_Assessment_of_the_Allocation_of_HIV_Funding_in_Indonesia_73页_4mb

Summary of the Assessment of the Allocation of HIV Funding in Indonesia

Core Content

This document presents an analysis of the cost-effectiveness and return on investment (ROI) of HIV programs in Indonesia from 2000 to 2010, with a focus on informing the optimal allocation of resources for the 2015-2019 national HIV budget. The study was conducted in the context of decreasing international donor funding and aims to provide evidence-based insights to improve the efficiency and impact of HIV prevention efforts.

Key Findings

Overall Impact of HIV Funding

• Total HIV/AIDS funding in Indonesia increased from approximately US$25 million in 2003 to US$70 million in 2010, with a total of US$363 million spent over the 2000-2010 period.
• This investment averted 130,000-240,000 HIV infections, corresponding to a 53-61% reduction in population incidence.
• It also averted 11,000-13,000 HIV/AIDS-related deaths.
• The majority of the benefits were observed in non-Papua regions.

Cost-Effectiveness of Prevention Programs

• Prevention programs for MSM/waria were the most cost-effective, with a cost of US$61-114 per QALY gained, and were estimated to avert 19,000-52,000 infections.
• Needle-syringe programs (NSPs) targeting injecting drug users (IDUs) were the next most cost-effective, with a cost of US$105-321 per QALY gained, and were estimated to avert 57,000-102,000 infections.
• NSPs were found to be twice as cost-effective as methadone maintenance therapy (MMT) due to lower unit costs and better individual-level efficacy.
• Prevention programs for female sex workers (FSWs) averted 4,200-9,200 infections but had a high cost per QALY gained (US$3,073-6,688).
• General population prevention programs were not found to be cost-effective.

Economic Benefits

• The total investment of US$363 million is estimated to be fully recovered in healthcare cost savings by 2050, with a return of US$1.15-1.32 for every US$1 invested.
• Indirect costs, which were not included in the initial cost-effectiveness analysis, would reduce the ROI to ~US$0.55-0.63 per US$1 invested.

Key Recommendations

• Prioritize MARPs: Resources should be reallocated from general prevention programs to most at-risk populations (MARPs), particularly injecting drug users (IDUs), MSM/waria, and FSWs.
• Scale-up harm reduction for IDUs: Increase funding for NSPs and MMT to maintain control over IDU-driven transmission.
• Focus on MSM/waria programs: These programs should be scaled up further as more resources become available.
• Improve efficiency of FSW programs: FSWs act as a bridge for HIV transmission from MARPs to the general population; thus, targeting these groups is essential, but programs must be made more efficient.
• Consider sexual partners of MARPs: These groups should be included in prevention efforts, especially as they are at higher risk of transmission.
• Optimize ART use: While ART for prevention (initiating therapy for people with CD4 counts >350) is not a priority with current resources, scaling up ART for treatment is essential to reduce morbidity and mortality among people living with HIV.

Methodology Overview

Mathematical Epidemic Model

• A modified version of the HIV in Indonesia Model (HIM) was used to simulate the impact of HIV programs on the epidemic.
• The model includes 10 distinct population groups, including IDUs, FSWs, MSM, and low-risk populations.
• It incorporates realistic biological transmission processes, detailed infection progression, and sexual mixing patterns.

Model Calibration

• The model was calibrated using BSS or IBBS data from previous studies, endorsed by the Ministry of Health.
• The model was split into non-Papua and Papua regions due to distinct epidemics in these areas.
• Data from 2009-2010 was used to calculate proportional allocations for prior years.

Costing and Spending Breakdown

• Funding allocation was based on National AIDS Spending Assessments (NASA) and other cost studies.
• Indirect costs were assumed to include categories such as program management, human resources, and social services.
• MARPs received the majority of prevention funding, with IDUs getting 67%, FSWs 30%, and MSM only 3%.

Economic Analysis

• Cost-effectiveness was assessed using Quality-Adjusted Life Years (QALYs) as the outcome of interest.
• Counterfactual scenarios were used to compare observed outcomes with those under reduced or no funding conditions.
• Sigmoid/logistic curves were fitted to data to estimate the relationship between program spending and changes in risk behaviors.

Key Assumptions and Data Gaps

• Proportional spending was assumed to be the same as in 2009-2010 for all years.
• Unknown allocations were reallocated to MARPs.
• Non-targeted programs were assumed to affect low-risk populations.
• Behavioral parameters were modeled using sigmoid/logistic curves based on empirical data.
• Fixed ratios were assumed between prevention and indirect costs.
• No movement between population groups was assumed over the analysis period.
• Health utilities were based on a meta-analysis of HIV/AIDS utility estimates.

Conclusion

The study emphasizes the need for more efficient allocation of HIV funding in Indonesia, particularly targeting most at-risk populations (MARPs) to maximize epidemiological impact and health outcomes. It also highlights the importance of data-driven modeling and cost-effectiveness analysis in guiding future HIV funding decisions. With reduced international donor funding, the Indonesian government is urged to reallocate resources to more effective prevention programs and to increase technical efficiency and reduce overhead costs.