Syringe exchange and risk of infection with hepatitis B and C viruses

FRANÇAIS :
L'étude porte sur une cohorte d'usagers de drogues par voie injectable de Seattle participant à une étude plus large (RAVEN Study - Risk Activity Variables, Epidemiology, and Network Study), et concerne 187 personnes séronégatives au virus de l'hépatite C (VHC) et 460 séronégatives au virus de l'hépatite B (VHB), recrutées de juin 1994 à janvier 1996 et suivies une année plus tard. 39 infections au VHC et 46 au VHB ont été enregistrées. Les résultats montrent aucun effet protecteur de l'échange de seringues sur l'infection à VHC ou à VHB.
ENGLISH :
The authors utilized a cohort study among Seattle injection drug users (IDUs) to assess whether participation in a syringe exchange program was associated with incidence of hepatitis B virus (HBV) and hepatitis C virus (HCV) infection. Susceptible I DU subjects (187 seronegative for antibody to HCV, and 460 seronegative for core antibody to HBV) were identified in drug treatment, corrections, and social service agencies from June 1994 to January 1996, and followed for seroconversion one year later. The subjects included in the analysis were Seattle-King County (Washington State) area IDUs enrolled in a larger multipurpose cohort study, the Risk Activity Variables, Epidemiology, and Network Study (RAVEN Study). There were 39 HCV infections (20.9/100/year) and 46 HBV infections (10.0/100/year). There was no apparent protective effect of syringe exchange against HBV (former exchange users, relative risk (RR) = 0.68, 95% confidence interval (Cl) 0.2-2.5; sporadic exchange users, RR = 2.4, 95% Cl 0.9-6.5; regular users, RR = 1.81, 95% Cl 0.7-4.8; vs. RR = 1.0 for nonusers of the exchange; adjusted for daily drug injection). Neither did the exchange protect against HCV infection (sporadic users, RR = 2.6, 95% CI 0.8-8.5; regular users, RR = 1.3, 95% Cl 0.8-2.2; vs. RR = 1.0 for nonusers; adjusted for recent onset of injection and syringe sharing prior to enrollment). While it is possible that uncontrolled confounding or other bias obscured a true beneficial impact of exchange use, these data suggest that no such benefit occurred during the period of the study. (Editor' s abstract)