Self-reported alcohol abuse in HIV-HCV co-infected patients: a better predictor of HIV virological rebound than physician's perceptions (HEPAVIH ARNS CO13 cohort).

Aims: Studying alcohol abuse impact, as measured by physicians' perceptions and patients' self-reports, on HIV virological rebound among patients chronically co-infected with HIV and hepatitis C virus (HCV).

Design: Cohort study.

Setting: Seventeen French hospitals.

Participants: Five hundred and twelve patients receiving antiretroviral therapy (ART) with an undetectable initial HIV viral load and at least two viral load measures during follow-up.

Measurements: Medical records and self-administered questionnaires. HIV virological rebound defined as HIV viral load above the limit of detection of the given hospital's laboratory test. Alcohol abuse defined as reporting to have drunk regularly at least 4 (for men) or 3 (for women) alcohol units per day during the previous 6 months. Correlates of time to HIV virological rebound identified using Cox proportional hazards models.

Findings: At enrolment, 9% of patients reported alcohol abuse. Physicians considered 14.8% of all participants as alcohol abusers. Self-reported alcohol abuse was associated independently with HIV virological rebound [hazard ratio (95% confidence interval): 2.04 (1.13-3.67); P = 0.02], after adjustment for CD4 count, time since ART initiation and hospital HIV caseload. No significant relationship was observed between physician-reported alcohol abuse and virological rebound (P = 0.87).

Conclusions: In France, the assessment of alcohol abuse in patients co-infected with HIV and hepatitis C virus should be based on patients' self-reports, rather than physicians' perceptions. Baseline screening of self-reported alcohol abuse may help identify co-infected patients at risk of subsequent HIV virological rebound.