Population-Level Impact Of A Medical Male Circumcision Program To Prevent HIV Infection: Kenya 2008–2011

Multiple observational studies, confirmed by three randomized control trials, have demonstrated beyond any reasonable doubt that medical male circumcision reduces the risk of female-to-male transmission of HIV. Despite this conclusive evidence and available funding, a lack of information on the longer-term population-level impact of voluntary male medical circumcision (VMMC) programs has proven one barrier to planning and scale-up of circumcision services. To help address this lack of knowledge regarding long-term effectiveness we conducted a series of population-level surveys to describe the impact of a large-scale VMMC program in a community specifically targeted by the world’s most successful national VMMC program in Kisumu, Kenya. The circumcision impact study (CIRCIS) is a series of cross-sectional surveys, each designed to represent the low-to-middle income population of Kisumu municipality at specific time-points during the progression of Kenya’s VMMC program. In this thesis we describe changes in circumcision-related knowledge and beliefs with the identification of associations that may indicate community-wide risk compensation, quantify changes in MC prevalence, describe factors associated with program uptake, and explore the impact of increased circumcision prevalence to the HIV epidemic in this population. We were able to confirm the protective effect of circumcision at the population level and identified a number of important co-factors of HIV infection. VMMC uptake was 24% between 2008 and 2011 resulting in a significant increase in circumcision prevalence from 33.0% to 49.5%. There was significant variation in circumcision uptake across sociodemographic categories with the largest gains observed in young men (<19 years), especially those not yet sexually active, and students. These men are currently at low risk of HIV; however, they are likely to change their risk behaviors as they age. Lastly, we found no evidence of population-level increases in HIV risk behaviors or risk compensation. Our results should help inform VMMC scale-up in Kenya and throughout the region, enabling policy makers to make evidence-based decisions about VMMC programming.