Are venue-based strategies the ticket to the last mile in HIV prevention in Malawi

Background: In 2016, Blantyre District had the highest adult HIV prevalence in Malawi (17% overall; 22% in women) and the lowest viral suppression rate (60%). In response, the MOH expanded prevention and treatment strategies. We hypothesized that social venues patronized by people with high sexual partnerships rates could identify sub-groups currently missed. Methods: We conducted cross-sectional bio-behavioral surveys of representative samples of individuals seeking care in government clinics (n=2313) and social venue patrons (n=1802) Jan-Mar 2022. Clinics were randomly selected from government clinics providing HIV testing. Venues were randomly sampled from urban and rural strata with oversampling of rural venues. Sampling weights were based on 2-stage sampling probabilities. We followed national testing protocols for rapid tests, recency testing and viral load measurements. Acute infections were identified by pooling dried blood spots from persons with an HIV- rapid test. Results: Compared to the clinic population, the venue population was more likely to: be male (68% vs 28%); aged >25 years (61% vs 51%); unmarried (62% vs 40%); drink alcohol daily (43% vs 8%); have more sexual partners in the last year (mean 16 vs 2); report a new sex partner in the past 4 weeks (42% vs 14%); and report transactional sex (52% vs 12%). HIV prevalence (Table 1) was higher among the venue population (19% vs 9%); the proportion HIV+ suppressed was similar (78%). Among women recruited at venues, prevalence increased by age: 0% among age 15-17 to 41% among age 18-21. At venues, factors associated with HIV infection include female sex (39% vs 10%); having a new partner in the past 4 weeks (28% vs 13%) and transactional sex (25% vs 13%). Acute and recent infections were uncommon. Clinic participants who reported visiting venues were less likely to have a suppressed viral load than other PLHIV clinic participants (53% vs 81%). Among both populations, reporting a genital sore in the past 4 weeks was associated with non-suppression (40% vs 20% in clinic; 48% vs 20% in venues). Conclusions: Lower HIV prevalence and greater viral suppression suggests that Blantyre’s HIV epidemic is slowing. Strategies to further reduce transmission should include outreach to venues with higher prevalence of unsuppressed infection and to young women at venues. Testing for acute or recent infection yielded few cases and thus did not provide sufficient value to warrant the cost.