From collaborator to colleague: a community-based program science approach for engaging Kenyan communities of gay, bisexual and other men who have sex with men in HIV research

ABSTRACTSince the 1990s, researchers have used community-based participatory approaches to achieve outcomes relevant to local communities, to build collaborative and sustainable research infrastructures, and to address disparities in knowledge production. Notwithstanding these strengths, communities and researchers have questioned its success in addressing power imbalances inherent in collaborative research encounters. In this methodological paper, we describe a novel community-based program science approach to guide an interdisciplinary research project on HIV self-testing among men who have sex with men in three Kenyan counties. Drawing on ethnographic field notes, we detail how community researchers and their academic and programmatic partners collaborated through all phases of the research process, including research design and data collection. Importantly, community researchers also played an integral role in data analysis and dissemination, going well beyond the conventional role of ‘community engagement’ in global health research. We also present findings from qualitative interviews conducted by community researchers with their peers to inform the rollout of HIV self-testing kits in their respective county-contexts. Our approach highlights that engaging community directly in evidence production allows research findings – owned and generated by communities on their own behalf – to be fed more swiftly and effectively into community-led program delivery.KEYWORDS: Community-based participatory researchmen who have sex with menHIV testing Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1. Article 162 of the Kenyan Penal Code, a relic of British colonial-era laws criminalizing ‘carnal knowledge against the order of nature’, was upheld by the Kenyan High Court in 2019.Additional informationFundingThe work was supported by the Bill and Melinda Gates Foundation [OPP-11191068].