Importance: Little is known about the factors affecting viral suppression among people living with HIV returning to the community after incarceration (ie, reentrants) in Africa.
Objective: To describe changes in viral suppression among reentrants living with HIV in Zambia and estimate the association between postrelease alcohol use and viral suppression.
Design, setting, and participants: This cohort study enrolled incarcerated people living with HIV from 5 prisons and surrounding communities in Zambia and conducted study visits before and after prison release to assess HIV-associated health behaviors and changes in clinical outcomes between March 2017 and November 2019. Eligible participants were incarcerated people aged at least 18 years, living with HIV, with release planned fewer than 30 days from screening, enrolled in the national HIV program, planning on living in Lusaka postrelease, and providing voluntary informed consent. Data were analyzed from April 20, 2024, to October 2, 2025.
Exposure: Postrelease unhealthy alcohol use (UAU), per the World Health Organization Alcohol Use Disorders Identification Test.
Main outcomes and measures: The primary outcome was HIV-1 viral load (VL) suppression (<1000 copies/mL) before and after release. Mixed-effects Poisson regression was used to estimate the association between postrelease UAU and loss of viral suppression, using a directed acyclic graph to identify confounders.
Results: Of 396 individuals screened, a total of 295 incarcerated people living with HIV (74.7%; median [IQR] age, 34 [29-41] years; 237 [80.3%] male) were included, with median (IQR) baseline CD4+ of 378 (244.5-506.5) cells/mL. At release, participants had been incarcerated a median (IQR) of 7.4 (3.5-15.4) months and receiving antiretroviral therapy for 13.2 (4.3-62.5) months. Documented viral suppression decreased from 237 individuals (80.3%) before release to 211 individuals (71.5%) after release. A total of 251 participants (85.1%) completed follow-up at a median (IQR) of 7.8 (5.4-11.6) months after release. UAU (prevalence ratio [PR], 3.35; 95% CI, 1.82-6.15) and unhealthy drug use (PR, 2.82; 95% CI, 1.39-5.71) were associated with postrelease unsuppressed VL. Among 205 participants (69.5%) with prerelease viral suppression, postrelease UAU was significantly associated with loss of viral suppression (adjusted risk ratio, 4.07; 95% CI, 1.97-8.42).
Conclusions and relevance: This cohort study of reentrants living with HIV in Zambia found that postrelease UAU was significantly associated with loss of viral suppression after community reentry. These findings suggest that new care models are needed to support HIV care continuity and address comorbid substance use in this population.