Objective: Recent studies suggest all-cause mortality in HIV mono-infected patients approaches that of the general population. We aimed to compare participants in the Canadian Co-infection Cohort to the general population to determine if co-infected patients have had similar improvements in mortality.
Design: Prospective multicentre cohort study.
Methods: Between 2003 and 2013, deaths were captured using specific case reports and through linkage to provincial vital statistics for participants lost to follow-up. Standardized mortality ratios (SMRs) were calculated using age, sex and province-specific mortality rates from the Canadian Human Mortality Database, 2009, and compared across behavioural and clinical characteristics of participants at their most recent visit.
Results: Among the 1150 patients, we observed 133 deaths over 3351 person-years (4.0 per 100 person-years, 95% confidence interval 3.3, 4.6). SMRs (95% confidence interval) were: 12.1(10.1, 14.2) overall; 9.3 (7.5, 11.1) for men and 19.4 (12.7, 26.2) for women. CD4 cell counts below 200 cells/μl [25.5 (17.7, 33.3)], active injection drug use [19.9 (13.9, 25.9)] and smoking [14.9 (12.1, 17.7)] were strongly associated with excess mortality. Lowest SMRs were seen for those who had spontaneous [4.5 (-0.6, 9.5)] or treatment-induced clearance of hepatitis C virus (HCV) infection [5.1 (1.3, 8.8)]. Conversely, high SMRs were seen with advanced liver disease [17.0 (11.7, 22.3)]. In no category did SMRs approach mortality seen in the general Canadian population.
Conclusions: HIV-HCV co-infected persons remain at markedly increased risk for death despite antiretroviral therapy. Interventions targeting modifiable risk factors such as substance use, smoking, adherence to antiretrovirals and timely provision of HCV therapy could substantially reduce death rates.