Impact of the coronavirus disease 2019 (
COVID
‐19) pandemic and stringency index on the dispensation of human immunodeficiency virus post‐exposure prophylaxis in Brazil: A retrospective study
Matheus Negri Boschiero, Nathália Mariana Santos Sansone, Fernando Augusto Lima Marson Abstract
Introduction
The coronavirus disease 2019 (COVID‐19) pandemic substantially affected healthcare systems worldwide and may have compromised access to human immunodeficiency virus (HIV) prevention strategies, including post‐exposure prophylaxis (HIV‐PEP). However, evidence regarding temporal changes in HIV‐PEP dispensation and their association with restrictive governmental measures in Brazil remains limited. Therefore, this study aimed to evaluate the dynamics of HIV‐PEP dispensation in Brazil before, during and after the COVID‐19 pandemic and to investigate its temporal association with the COVID‐19 Stringency Index.
Methods
This retrospective epidemiological study used secondary data from the Brazilian Ministry of Health HIV‐PEP Monitoring Panel between 2018 and 2024. The study period was categorized into pre‐pandemic (until February 2020), pandemic (March 2020 to May 2023) and post‐Public Health Emergency of International Concern (post‐PHEIC; from June 2023 onward) periods. An interrupted time‐series analysis (ITSA) using negative binomial regression was performed to evaluate temporal changes in HIV‐PEP dispensation, including immediate‐level changes and changes in trends associated with the pandemic and post‐PHEIC periods. Additional generalized linear mixed‐effects models were used to evaluate factors associated with the volume of recorded HIV‐PEP dispensation events, with a random intercept for Brazilian state. Data on the COVID‐19 Stringency Index were obtained from the Oxford COVID‐19 Government Response Tracker. Exploratory analyses using negative binomial regression and Spearman correlation were performed to evaluate the temporal association between the Stringency Index and HIV‐PEP dispensation.
Results
From 2018 to 2024, 863 201 HIV‐PEP dispensation events were recorded in Brazil, most frequently in the Southeast region (54%). Consensual sexual exposure accounted for most dispensation events (69%), followed by accidents involving biological material (26%). Before the pandemic, HIV‐PEP dispensation showed an increasing temporal trend (rate ratio [RR] = 1.019; 95% confidence interval [95% CI]: 1.015, 1.024). At the onset of the COVID‐19 pandemic, an immediate reduction in the number of recorded HIV‐PEP dispensation events was observed (RR = 0.657; 95% CI: 0.604, 0.714). The volume of dispensations subsequently resumed an increasing trajectory during both the pandemic (RR = 1.016; 95% CI: 1.014, 1.018) and post‐PHEIC periods (RR = 1.013; 95% CI: 1.006, 1.020), although the observed trajectory remained below the pre‐pandemic trajectory. In the adjusted mixed‐effects model, the expected number of recorded HIV‐PEP dispensation events was lower during both the pandemic (RR = 0.74; 95% CI: 0.71, 0.76) and post‐PHEIC periods (RR = 0.75; 95% CI: 0.71, 0.79). In the primary temporal model, the Stringency Index was inversely associated with HIV‐PEP dispensation (RR = 0.968; 95% CI: 0.961, 0.974). However, this association was no longer statistically significant after the pandemic onset was incorporated into the segmented sensitivity model (RR = 0.990; 95% CI: 0.980, 1.001), indicating that the association was not robust across all model specifications.
Conclusion
HIV‐PEP dispensation in Brazil showed an increasing trend before the COVID‐19 pandemic, followed by an immediate reduction at pandemic onset and a subsequent recovery that remained below the model‐based counterfactual trajectory. Although the Stringency Index was inversely associated with HIV‐PEP dispensation in the primary exploratory model, this association was not statistically significant after adjustment for pandemic onset. These findings should therefore be interpreted as temporal associations rather than evidence of independent or causal effects.