Inequality in access to pre-exposure prophylaxis (PrEP) for HIV has been a challenge since its implementation in the Brazilian public health system in 2017. This study compares the profile of PrEP users with that of new HIV cases in Brazil from 2018 to 2024, evaluating the “Reach” component of the RE-AIM framework. Descriptive study with secondary data from the PrEP Monitoring Panel, the Notifiable Diseases Information System, and the Laboratory Control System for CD4+/CD8+ and HIV viral load. Reach was analyzed by sociodemographic representativeness (age, race/skin color, education, and population group). All complete records from 2018 to 2024 were included. Simple annual growth rates were calculated in Excel (2018). No qualitative methods or recruitment cost analyses were applied. From 2018 to 2024, there was a reduction in the compound growth rate of new HIV cases (−11.6%), while new PrEP users increased (+51.4%). New HIV cases predominated among men who have sex with men (MSM, 55%), mixed-race individuals (48.1%), people with 8–11 years of education (40.7%), and ages 30–39 years (27.1%). PrEP users were concentrated among MSM (81.8%), White/Asian individuals (55.9%), those with ≥ 12 years of schooling (72.1%), and the same age range (43.8%). Growth among groups previously underserved – such as Black individuals, younger people, and those with lower education – indicates advances in equity. However, there was reduced growth among groups with initially high uptake, such as MSM aged 30–39 years (−13.4% between 2023–2024), suggesting a need for retention strategies. PrEP has contributed to reducing HIV incidence in Brazil, but its initial reach favored groups with lower social vulnerability. The RE-AIM analysis shows positive expansion in sociodemographic representativeness among key populations. Communication strategies, decentralization, and inclusion of qualitative methods may strengthen outreach and broaden representativeness. Limitations include use of secondary data subject to underreporting, absence of qualitative components, and reliance on basic statistical tools.
Rodrigues et al. (Sun,) studied this question.