A prospective study of hepatitis C incidence in Australian prisoners.

AIMS: To document the relationships between injecting drug use, imprisonment and hepatitis C virus (HCV) infection.

DESIGN: Prospective cohort study.

SETTING: Multiple prisons in New South Wales, Australia.

PARTICIPANTS: HCV seronegative prisoners with a life-time history of injecting drug use (IDU) were enrolled and followed prospectively (n = 210) by interview and HCV antibody and RNA testing 6-12-monthly for up to 4 years when in prison.

MEASUREMENTS: HCV incidence was calculated using the person-years method. Cox regression was used to identify predictors of incident infection using time-dependent covariates.

RESULTS: Almost half the cohort reported IDU during follow-up (103 subjects; 49.1%) and 65 (31%) also reported sharing of the injecting apparatus. There were 38 HCV incident cases in 269.94 person-years (py) of follow-up with an estimated incidence of 14.08 per 100 py [confidence interval (CI) = 9.96-19.32]. Incident infection was associated independently with Indigenous background, injecting daily or more and injecting heroin. Three subjects were RNA-positive and antibody-negative at the incident time-point, indicating early infection, which provided a second incidence estimate of 9.4%. Analysis of continuously incarcerated subjects (n = 114) followed over 126.73 py, identified 13 new HCV infections (10.26 per 100 py, CI = 5.46-17.54), one of which was an early infection case. Bleach-cleansing of injecting equipment and opioid substitution treatment were not associated with a significant reduction in incidence.

CONCLUSIONS: In New South Wales, Australia, imprisonment is associated with high rates of hepatitis C virus transmission. More effective harm reduction interventions are needed to control HCV in prison settings.