Using information on ART costs and benefits to mobilise resources - comparing different methods and contexts

Sustaining HIV and AIDS responses depends on a mix of donor, government and private funders. Their decisions about investing in HIV treatment may be informed by various types of economic evaluations, which may be more or less useful for different contexts. This paper benchmarks methods against each other. Epidemiological and demographic impacts of HIV and antiretroviral therapy (ART) from 1996-2015 were quantified using country- specific spectrum files. The study compared societal benefits of ART using the full income (FI) methodology with "conventional" benefit, utility and effectiveness estimates produced with the same data. The FI estimates suggested $3.50 in benefits per dollar spent on ART globally, 2.6 times larger than productivity-related measures of benefits, of $1.33 in benefits per dollar. Higher benefit-cost ratios are mainly because FI reflects value of life beyond what people produce at work and in non-working age groups, and allocates the future stream of benefits in the year that death is avoided. ART costs were 0.78 times per capita GDP per quality-adjusted life-years gained globally. FI benefit-cost ratios are considerably higher in upper- and lower-middle-income countries than in low- or high-income countries. Productivity-based benefits also exceeded costs in all but one region but had smaller ratios and different regional patterns. Per capita GDP per quality-adjusted life-years ratios were below 1.2 for all regions and country income bands, suggesting cost effectiveness. The fact that FI returns of ART are higher than productivity returns, helps to quantify developmental benefits of interventions that directly extend life and its quality, arguably the objective of development. They provide an important argument to increase budget allocations to health sectors for ART scale-up, and not just reallocate existing health resources. Benchmarking FI returns against cost per quality- adjusted life-years may allow comparison to other "cost effective" health interventions. However, caution should be taken in extrapolations between measures, because they produce different rankings across country categories.