Predictive factors of HIV-infection in injecting drug users upon incarceration

Objective: To identify predictors of HIV-infection in injecting drug users upon incarceration. Patients and methods: We studied 639 IDU or ex-IDU prisoners admitted to a provincial prison of Northwestern Spain between 1 January 1991 and 31 December 1995. Each was interviewed by health personnel and tested for HIV-infection (ELISA followed by immunoblot confirmation in positive cases). Statistical analysis was based on logistic regression. Results: The prevalence of HIV-infection was 46.9% (95% CI: 43.1%-50.8%). No decreasing tendency in annual prevalence of HIV-infection was observed (p = 0.88); however, for those incarcerated for the first time prevalence fell from 38% in 1991 to 19% in 1995 (p = 0.20). Gypsies (OR: 0.43; 95% CI: 0.23- 0.80) and prisoners who were older upon first incarceration (OR: 0.94; 95% CI: 0.90-0.99) were associated with lesser risk of HIV-infection. Women (OR: 2.17; 95% CI: 1.29-3.65), older prisoners (OR: 1.06; 95% CI: 1.02-1.11), those with multiple incarceration histories (OR: 1.06; 95% CI: 1.01-1.11) and long-term prisoners (OR: 1.01; 95% CI: 1.00-1.02) were associated with higher risk. Conclusions: The high prevalence of HIV-infection, especially in women, younger prisoners, repeat offenders and long-term prisoners, suggests that prevention measures directed toward the most marginal IDU have not been very effective. Harm-reduction programs must be made to reach the IDU population, both in and outside prison walls.