Résumé : Background: The short- and long-term effects of anti-hepatitis C treatment on mortality in the HIV-HCV-coinfected population have not been evaluated in observational cohorts. Such evaluations must use methods that allow for time-varying prognostic factors that both predict treatment and are affected by prior treatment. We aimed to study immunological changes in HIV-HCV-coinfected individuals during HCV treatment and to estimate the effect of HCV-treatment on mortality. Methods: Patients were included if they were aged ≥16 years, were HIV-HCV-coinfected and were enrolled in the COHERE cohort. Data were pooled within COHERE in December 2009 in EuroCoord. Random-effects models were used to model immunological changes during HCV treatment. Marginal structural models were used to estimate the effect of HCV treatment on mortality, allowing for time-dependent confounders affected by prior treatment. Results: In total, 780/6,433 (12%) HIV-HCV-coinfected patients initiated HCV treatment (interferon [IFN] and ribavirin n=692, IFN alone n=88). CD4+ T-cell counts decreased during the first 12 weeks of treatment (P<0.0001) and stabilized from week 24 onwards. The estimated mortality hazard ratio for comparing HCV-treated with -untreated individuals was 0.72 (95% CI 0.43, 1.21). The estimated hazard ratio for liver-related death was 0.57 (95% CI 0.21, 1.55). Conclusions: Despite its effect in reducing CD4+ T-cell counts, the effect of HCV treatment on mortality was in the direction of benefit and our results excluded a substantial increase in mortality. Such benefit may be related to a lower risk of liver-related death. New HCV treatment strategies might contribute to a further reduction in mortality. ©2012 International Medical Press.