Joint effects of substance use disorders and recent substance use on HIV viral non-suppression among people engaged in HIV care in an urban clinic, 2014–2019

Abstract

Aims

To estimate the joint effects of substance use disorder (SUD) and recent substance use on human immunodeficiency virus (HIV) non-suppression.

Design

Retrospective clinical cohort study with repeated observations within individuals.

Setting

Baltimore, Maryland, United States.

Participants

1881 patients contributed 10 794 observations.

Measurements

The primary independent variable was the combination of history of SUD and recent substance use. History of SUD was defined as any prior International Classification of Diseases 9/10 code for cocaine or opioid disorder. Recent substance use was defined as the self-report of cocaine or non-prescribed opioid use on the National Institute of Drug Abuse-modified Alcohol, Smoking and Substance Involvement Screening Test or clinician-documented cocaine or opioid use abstracted from the medical record. The outcome was viral non-suppression, defined as HIV RNA >200 copies/mL on the first viral load measurement within 1 year subsequent to each observation of substance use. We adjusted for birth sex, Black race, age, HIV acquisition risk factors, years in care and CD4 cell count. In secondary analyses, we also adjusted for depressive, anxiety and panic symptoms, cannabis use and cannabis use disorder.

Findings

On their first observation, 31% of patients had a history of an SUD and 18% had recent substance use. Relative to no history of SUD and no recent substance use, the 1-year fully adjusted risk difference (RD) for viral non-suppression associated with cocaine and opioid use disorder and recent substance use was 7.7% (95% CI = 5.3%–10.0%), the RD was 5.5% (95% CI = 1.2%–9.7%) for history of cocaine use disorder without recent substance use, and the RD was 4.6% (95% CI = 2.7%–6.5%) for recent substance use without a SUD.

Conclusions

Substance use and substance use disorders appear to be highly prevalent among, and independently associated with, viral non-suppression among people with HIV.