Men’s preferences for HIV self-test distribution strategies in Uganda: A discrete choice experiment

INTRODUCTION: HIV-self testing (HIVST) can increase men's HIV testing in Eastern and Southern Africa (ESA). While secondary HIVST distribution from pregnant women to male partners is standard of care in ESA, little is known about men's preferences for HIVST distribution strategies. METHODS: We administered a Discrete Choice Experiment (DCE) to men from Kampala and Wakiso, Uganda. DCE Attributes were: HIVST distributor (counselor, community health worker, partner, pharmacy staff), distributor sex, HIVST type (oral, blood-based), counseling/support (toll-free hotline, in-person, SMS) and HIVST distribution location (facility, home, mobile clinic, workplace, pharmacy, booth). We used conditional logit regression and latent class analysis. RESULTS: Among respondents (N=396, mean: 30 years old, 90% in a relationship), HIVST distribution was highly acceptable (a large negative opt-out preference weight); participants preferred oral HIVST with in-person or phone counseling/support and had a negative preference for receiving an HIVST at home from female partners. We identified three subgroups: 1) "Prefer HIVST distribution" (71.5% of sample) whose preferences aligned with the main analysis aside from only preferring in-person counseling and support, 2)"Prefer oral HIVST with real-time support" (21.2% of sample), whose preferences also aligned with the main analysis except for positive opt-out (indicating they would only uptake the intervention if it aligned with their preferences), and 3) "Prefer blood-based HIVST in pharmacy with real-time support (7.3%)", which preferred blood-based HIVST from pharmacies with in-person support. CONCLUSIONS: HIVST distribution was highly acceptable; overall men preferred oral HIVST with in-person support and did not prefer receiving HIVST at home from female partners.