The influence of hepatitis C and alcohol on liver-related morbidity and mortality in Glasgow's injecting drug user population.

Infection with the hepatitis C virus (HCV) is associated with the development of severe liver disease, but cofactors--namely alcohol abuse--in Scotland's HCV-positive population complicate estimation of the unique contribution of HCV. We compared the risk of hospital admission/death for a liver-related cause in a large cohort of Glasgow's injecting drug users (IDUs) testing HCV-positive with IDUs testing HCV negative. Data for 6566 current/former IDUs who had been tested for anti-HCV and/or HCV RNA by polymerase chain reaction in Greater Glasgow health board between 1993 and 2007 were linked to the national hospitalization database and deaths registry to identify all admissions and deaths from a liver-related condition. Relative risks were estimated using Cox proportional hazards regression for recurrent events. Time at risk was censored at 2 years following an HCV test to address bias owing to unobserved seroconversion. The risk of hospitalization/death from a liver-related or an alcoholic liver-related condition following HCV testing was greater for those IDUs with no prior alcohol-related hospitalization who tested positive [adjusted hazard ratio (HR) = 3.2, 95% CI: 1.5-6.7; 4.9, 95% CI: 1.8-13.1, respectively], compared with those who tested anti-HCV negative, but not for those IDUs with a prior alcohol admission (HR = 0.8, 95% CI: 0.4-1.5; 0.8, 95% CI: 0.4-1.6). There was little evidence for an increased risk of hospitalization/death for an exclusively nonalcoholic liver condition for those testing positive (HR = 1.5, 95% CI: 0.8-2.7), after adjustment for previous alcohol-related admission. Within Glasgow's IDU population, HCV positivity is associated with an increased risk of a liver-related outcome, but this is not observed for those IDUs whose problem alcohol use already increases their risk.