Background: Sub-Saharan African (SSA) countries, including Kenya, are considering early infant male circumcision (EIMC) for HIV prevention to complement and perhaps replace adult male circumcision (AMMC) programmes. Evaluation of the costs and effectiveness of EIMC in comparison to AMMC are required for budgeting and resource allocation decisions. Such analyses should incorporate the effects of complementary interventions, especially treatment as prevention. Comprehensive evaluations of this type are lacking.
Methods: Using a narrow health systems perspective and actual male circumcision programme data, we conducted a cost-impact evaluation comparing AMMC to an AMMC plus EIMC (AMMCplus) programme in western Kenya. Incremental cost and effectiveness ratios (ICERs) were calculated. Inputs included: personnel time and remuneration, consumables, non-consumables, management/supervision, mobilization, training, and monitoring/evaluation. Impact estimates were adjusted for under-15 mortality and vertical transmission of HIV. Extreme case scenario analyses were conducted adjusting for discount rates, lifetime HIV treatment costs, priority population targets, and efficacy and scope of complementary HIV prevention programmes. 2014 is the reference year. We used a modified version of the Decision Makers Program Planning Tool for analyses.
Results: Under the reference case, one AMMC costs $40.11 (range $36.08, $46.14). One EIMC costs $58.23 (range $29.37, $114.15). Between 2008 and 2030, AMMC is expected to avert 81,346 (40,409, 116,835) incident infections and 33,933 (25,550, 50,258) AIDS deaths with 8 (6, 10) AMMCs per HIV infection averted (HIA) at a cost of $435 ($323, $676) per HIA and net savings of $4,315 ($3,952, $13,900) per HIA. AMMC dominates AMMCplus when fewer than 80% of infants are circumcised. Above 80%, AMMCplus results in an ICER of $384 (315, 813) per HIA. Our results are sensitive to the scale-up of complementary HIV programmes and are robust to discount rate.
Conclusions: Using WHO thresholds, MMC, at any age is highly cost-effective in western Kenya. AMMC dominates AMMC plus EIMC unless EIMC coverage is greater than 80%. Governments considering introducing EIMC services should consider complementary programmes and the long horizon between EIMC introduction and HIV impact. It is unlikely EIMC programmes could achieve 80% prevalence and therefore be cost-effective relative to AMMC programmes in Kenya.