Objective: An association between needle exchange attendance and higher HIV prevalence rates among injecting drug users (IDU) in Vancouver has been interpreted by some to suggest that needle exchange programmes (NEP) may exacerbate HIV spread. We investigated this observed association to determine whether needle exchange was causally associated with the spread of HIV.
Design and method: Prospective cohort study of 694 IDU recruited in the downtown eastside of Vancouver. Subjects were HIV-negative at the time of recruitment and had injected illicit drugs within the previous month.
Results: Of 694 subjects, the 15-month cumulative HIV incidence was significantly elevated in frequent NEP attendees (11.8+/-1.7 versus 6.2+/-1.5%; log-rank P = 0.012). Frequent attendees (one or more visits per week) were younger and were more likely to report: unstable housing and hotel living, the downtown eastside as their primary injecting site, frequent cocaine injection, sex trade involvement, injecting in 'shooting galleries', and incarceration within the previous 6 months. The Cox regression model predicted 48 seroconversions among frequent attendees; 47 were observed. Although significant proportions of subjects reported obtaining needles, swabs, water and bleach from the NEP, only five (0.7%) reported meeting new friends or people there. When asked where subjects had met their new sharing partners, only one out of 498 respondents cited the needle exchange. Paired analysis of risk variables at baseline and the first follow-up visit did not reveal any increase in risk behaviours among frequent attendees, regardless of whether they had initiated drug injection after establishment of the NEP.
Conclusions: We found no evidence that this NEP is causally associated with HIV transmission. The observed association should not be cited as evidence that NEP may promote the spread of HIV. By attracting higher risk users, NEP may furnish a valuable opportunity to provide additional preventive/support services to these difficult-to-reach individuals.