Adherence to Antiretroviral Therapy, Polypharmacy, and Virologic Failure in Women with HIV in the United States: Retrospective Cohort Analysis

Abstract

Background: Women with HIV (WHIV) in the United States face many challenges with adherence to antiretroviral drugs and suboptimal adherence often leads to virologic failure. Objectives: The main aim of this study was to identify the association between trajectories of adherence to antiretroviral drugs and time to virologic failure among WHIV. Methods: We conducted a retrospective analysis of longitudinal data among WHIV enrolled in the Women's Interagency HIV Study (WIHS) in the United States from 4/1/2014 to 9/30/2019 (10 consecutive visits). The study included women aged >18 years on antiretroviral therapy (ART) who had at least two consecutive measurements of HIV RNA. Women with less than three measurements of adherence were excluded. Group-based trajectory modeling (GBTM) was used to identify adherence trajectories and Cox proportional hazard regression models were fitted to identify differences in time to virologic failure among adherence trajectories, adjusted for factors potentially influencing the adherence-viral suppression relationship (age, smoking, presence of depression symptoms, alcohol intake at the start of this analysis, drug regimen at the last visit, and frequency of experiencing viremia during the whole study period). Results: A total of 1,437 women were eligible. The median age was 49 (interquartile range 42 – 54) years. The results of GBTM analysis revealed four latent adherence trajectories, namely ‘consistently high’ (N=378; 26.3%), ‘moderate increasing’ (N=137; 9.5%), ‘moderate decreasing’ (N=440; 30.6%), and ‘consistently low’ (N=482; 33.5%). Virologic failure occurred among 173 (12%) women. Compared to the ‘consistently high’ trajectory, the adjusted hazard ratio (aHR) for time to virologic failure was higher among patients classified in the ‘consistently low’ adherence trajectory aHR 2.8 (1.6-4.9; P 0.001) and decreased proportionally as adherence level increased among other trajectories as follows: ‘moderate decreasing’ aHR 1.8 (1.02-3.2; P=0.04) and ‘moderate increasing’ aHR 1.0 (0.4-2.5; P.001). Other predictors of virologic failure were current smoking (aAHR 1.5, 1.05-2.2; P=0.02), being on a protease inhibitor-based regimen at the last visit (aHR 2.7,1.8-4.0; P.001), being on no therapy at the last visit (aHR 8.8, 5.7-13.3; P.001), and frequent presentation with detectable viremia (aHR 2.8, 2.0-4.0; P.001). Conclusion: Time to virologic failure was inversely proportional to adherence trajectories. Multilevel targeted behavioral interventions for women with suboptimal adherence are urgently needed. In addition, it is imperative to optimize the drug regimen by selecting potent ART with longer half-lives together with developing anti-smoking campaigns to reduce the risk of inflammation.