BACKGROUND:
Oral pre-exposure prophylaxis (PrEP) is a key HIV prevention strategy. We evaluated the impact of PrEP implementation on municipal HIV incidence rates in Brazil.

METHODS:
We conducted a quasi-experimental ecological study using a longitudinal panel of all 5,570 Brazilian municipalities(2014–2023). Municipalities were stratified by population size(small:≤100,000; large:>100,000 inhabitants), with data integrated from national health information systems and administrative databases. The primary outcome was the municipal HIV incidence rate, age- and sex-standardized to the municipalities’ populations. The exposure was defined as a binary indicator of municipal PrEP availability(no vs. yes). After comparing pre-intervention(2014–2017) and post-intervention(2018–2023) trends via mixed-effect linear regression, we estimated the causal impact of PrEP availability using a staggered difference-in-differences(DiD) approach. This model exploited temporal variation in PrEP rollout across municipalities while adjusting for time-varying socioeconomic and healthcare covariates.

RESULTS:
Between the 2014–2017 and 2018–2023 periods, median HIV incidence rates rose from 0.0 (IQR:16.6-0.0) to 5.3 (20.4-0.0) in small municipalities(p<0.0001), with a slight increase in large municipalities: from 27.5 (IQR:18.7–40.7) to 29.2(IQR:41.9-19.2)(p>0.05). By 2023, PrEP was implemented in only 6.2%(333/5,334) of small municipalities(median of PrEP coverage-21.5 users per 100,000 inhabitants), compared to 85.9% (220/256) of large municipalities(median of PrEP coverage - 36.7 users per 100,000 inhabitants). While no significant association was found in small municipalities, PrEP implementation in large municipalities was associated with an average reduction of 73.3% in HIV incidence compared to the pre-intervention period. This impact intensified over time: significant reductions were sustained at three, four, and five years of post-implementation, with average treatment effects on the treated of 77.85% (−24.20; 95% CI −46.25 to −2.15), 116.87% (−36.33; −66.21 to −6.44), and 189.01% (−58.76; −114.82 to −2.69), respectively(Figure 1).

CONCLUSIONS:
PrEP implementation was associated with reduced HIV incidence in large municipalities after sustained rollout, but no impact was observed in smaller municipalities, highlighting persistent geographic inequities.