How should HIV-specific charitable interventions like PEPFAR be evaluated?

Fauci & Eisinger (F&E), in a short piece for New England Journal of Medicine, celebrate 15 years of the President’s Emergency Plan for AIDS Relief (PEPFAR) and count the lives saved. PEPFAR was established by President George W. Bush in 2003 in recognition of the obligation owed by a resource-rich country such as the US to the 90% of HIV/AIDS infected people living in low-income countries, given the existence of effective interventions that were unavailable to them at the time. Over the 15 years the number of recipient countries has grown from the initial 14 to whole over 50 worldwide, included 22 in sub-Saharan Africa.

Now for the counting. As of September 2017, PEPFAR funded programmes have: provided ART for 13.3 million men, women and children; performed 15.2 million voluntary male circumcisions; averted 2.2 million perinatal HIV infections; supported 6.4 million orphans and vulnerable children. Recent data suggest that, thanks to PEPFAR, five African countries are on track to achieve UNAIDS 90-90-90 targets by 2020: Lesotho, Malawi, Swaziland, Zambia and Zimbabwe.

Quantifying the benefits of PEPFAR funding is difficult, however. First, because, as Heaton & Way (STI) point out, there may be complex synergies between different interventions F&E themselves mention the training, through PEPFAR funding, of 220,000 health-care workers whose future role will presumably not be restricted to the delivery of PEPFAR HIV interventions.  Reynolds & Cates (STI) discuss the case, not specifically mentioned by F&E, of contraception interventions undertaken as a more cost-effective method than perinatal ART to prevent HIV positive births. Quantification is also difficult, however, because the benefits specifically attributable to PEPFAR may be hard to separate out from those deriving from other interventions. One can begin to see the logic of the recent move from a vertical (i.e. disease specific) mode of evaluating interventions to a more horizontal (non-disease specific) mode.

At the same time, this celebratory piece by F&E is itself a demonstration of the need for quantitative evaluations of interventions in order to justify their continued financial support. Baernighausen & Humair (STI) (B&H) argue that, ultimately, these two imperatives, arising simultaneously but independently from different quarters within international health policy, are potentially in conflict with each other. Evaluations of a programme dedicated to a specific disease cannot be adequately conducted on a vertical basis; but moving to a more horizontal mode of evaluation runs counter to the very aims of programmes like PEPFAR, and, as B&H argue, risks undermining the basis on which they justify their continuation.