Social Network Approach for Increasing Testing Coverage Among Men Who Have Sex With Men

HIV testing is essential in shortening the time to identify a new infection, the first 90 of the UNAIDS 90-90-90 targets. However, over one-third of the men who have sex with men (MSM) had never been tested for HIV; even if they did, one-fifth had their tests done more than a year ago. Assortative mixing pattern observed in the HIV-positive MSM group shaped the transmission dynamics and could be leveraged for intervention. Barriers to access HIV testing services could, on the other hand, be hurdled by self-tests. A network approach for intervention could therefore be promising in delivering effective HIV self-tests. To experiment with such an approach, a 2-phase study was conceptualised incorporating actual network-based referred HIV self-tests and an agent-based simulation evaluating its impact. Sixty-four MSM would be recruited as seeds for promoting HIV self-tests within their network and those being referred could refer their friends for the same after passing online training. To facilitate the process, an online platform would be developed offering information, collecting informed consent, requesting HIV self-test kits, returning results, performing online training, and referring peers. Participants could opt to receive self-tests by delivery or to conduct it on-site with staff assistance. A hotline with video conferencing support would be maintained to assist those who self-test at home. They could also choose between blood and oral fluid tests. Two user interfaces, namely gamification and neumorphism, would be randomly assigned. Primary outcomes to measure are number and proportion of MSM who had never or not tested within 12 months and the associating factors, and usability of the two user interfaces. Data collected in the empirical study would be used for parameterising the agent-based simulation to evaluate the impact of the approach in increasing testing coverage and shortening time to diagnosis. Its economic assessment would also be performed to cost each new infection to be identified. The approach could be feasible and effective to be adopted for future broader implementation for peer-led HIV self-test kit or HIV prevention message distribution.