|
Informing decision-making on male circumcision for HIV prevention in high HIV prevalence settings: insights from modelling
Presented by Cate Hankins (Switzerland).
C. Hankins1, N. Lohse1, T. Hallett2, N. Nagelkerke3, R. White4, L. Abu-Raddad5, H. Weiss4, R. Gray6, J. Stover7, UNAIDS/WHO/SACEMA Expert Group on Modelling the Impact and Cost of Male Circumcision for HIV Prevention
1UNAIDS, Evidence, Monitoring and Policy, Geneva, Switzerland, 2Imperial College, London, United Kingdom, 3United Arab Emirates University, Al Ain, United Arab Emirates, 4London School of Hygiene and Tropical Medicine (University of London), London, United Kingdom, 5Weill Cornell Medical College, Doha, Qatar, 6John Hopkins Bloomberg School of Public Health, Baltimore, United States, 7Futures Institute, Glastonbury, United States
Background: Decision-makers in high HIV prevalence/low male circumcision countries have asked whether introduction or expansion of male circumcion services is likely to be cost-effective for the reduction of HIV incidence over the short-, medium- and long-term. Mathematical models addressing costing and impact questions have helped refine a Decision-Makers' Programme Planning Tool.
Methods: Modellers, epidemiologists, economists, statisticians, policy makers, and male circumcision trial investigators compared assumptions, methodologies, and results of simulation models and costing studies of male circumcision for HIV prevention. Outcomes included cost per HIV infection averted, potential for cost savings, impact on women, effects of risk compensation, influence of early post-operative sex, and potential synergies with other HIV prevention programming.
Results: Not all models addressed all questions but consensus emerged with comparable findings on population-level impact of male circumcision on HIV incidence, despite different methods, baseline assumptions, and input variables. In high HIV prevalence/ low male circumcision populations, one HIV infection is averted for every 5 to 15 circumcisions performed; the cost to avert one HIV infection is $150 to $900, using a 10 year time horizon. Women benefit indirectly from reduced HIV prevalence in male partners. Potentially increased transmission or acquisition due to early post-operative resumption of sex has small population-level effects over 10-20 years. Behavioural risk compensation confined to newly or already-circumcised heterosexual men and their partners has minor effects on projected population-level benefits. Insights from this comparative analysis of models helped refine the Decision-Makers´ Programme Planning Tool subsequently used in Namibia and Botswana for programme planning.
Conclusions: Using mathematical modelling to refine and validate a user-friendly tool that can be deployed locally allows decision-makers to indirectly access main modelling findings to estimate HIV incidence, AIDS deaths, overall costs, and net cost per infection averted as a function of procedures performed, service delivery mode, and rate of scale-up.
|