Review and meta-analysis of the association between self-reported sharing of needles/syringes and hepatitis C virus prevalence and incidence among people who inject drugs in Europe.

BACKGROUND: Although sharing needles/syringes (N/S) is a recognised risk factor for the hepatitis C virus (HCV), epidemiological studies have shown inconsistent associations between self-reported N/S sharing and biological markers of HCV infection. This review aims to summarise, and explore factors that may explain the variation in, the measure of association between self-reported sharing of N/S and HCV prevalence/incidence among people who inject drugs (PWID).

METHODS: Studies undertaken in Europe during 1990-2011 were identified through an electronic literature search. Eligible studies reported HCV prevalence (or incidence) among those who reported ever/never (or recent/non-recent) sharing of N/S. Meta-analysis was undertaken to generate a pooled estimate of the association and heterogeneity was explored using stratified analyses.

RESULTS: Sixteen cross-sectional studies and four longitudinal studies were included. Pooled prevalence and incidence of HCV was 59% and 11% among PWID who reported never and not recently sharing N/S, respectively. Random effects meta-analysis generated a pooled odds ratio (OR) of 3.3 (95% CI 2.4-4.6), comparing HCV infection among those who ever (or recently) shared N/S relative to those who reported never (or not recently) sharing. There was substantial heterogeneity between the study effect sizes (I(2)=72.8%). Differences in pooled ORs were found when studies were stratified by recruitment setting (prison vs. drug treatment sites), recruitment method (outreach vs. non-outreach), sample HCV prevalence and sample mean/median time since onset of injecting.

CONCLUSION: We found high incidence/prevalence rates among those who did not report sharing N/S during the risk period, which may be due to a combination of unmeasured risk factors and reporting bias. Study design and population are likely to be important modifiers of the size and strength of association between HCV and N/S sharing.