BACKGROUND: Cohort studies estimating rates of incidence are scarce despite they are the most appropriate for following the epidemics of HIV-1 and evaluating the effectiveness of preventive interventions. The rate of incidence of infection by HIV-1 in intravenous drug users (IVDU) and the determining factors for such infection are estimated in the present study. METHODS: 3,693 IVDU attended in two centers for information and prevention of AIDS from 1987-1992, were included in the study. Out of 1,868 seronegative subjects, 579 attended a second visit in which a new serologic test was performed against HIV-1. The rates of incidence and 95% confidence intervals were calculated with a Poisson regression model being adjusted in order to control the confounding factors. RESULTS: The estimated global incidence was 11.7/100 people/year (CI 95%: 9.6; 14.46). Those highest rates were observed in IVDU who exchanged syringes (23.5/100 people/year), in those who had more than one sexual partner in the previous month (21.2/100 people/year), in those referring a history of prostitution (22.6/100 people/year), in carriers of hepatitis B antigens (25.6/100 people/year) and in the IVDU younger than 20 years of age (16.5/100 people/year). The variables which independently predicted the rates of incidence were: the exchange of syringes (RR = 6.85; CI 95%: 3.99; 11.75), having more than one different sexual partner in the previous month (RR = 3.51; CI 95%: 1.70; 7.20) and age, with lower risk in those under the age of 20 (RR = 0.49; CI 95%: 0.25; 0.95). CONCLUSIONS: The incidence of infection by type-1 human immunodeficiency virus observed in intravenous drug users is the highest described in Western countries. Preventive measures in general must be intensified particularly in subgroups of intravenous drug users with the greatest risk of seroconversion.
BACKGROUND: Cohort studies estimating rates of incidence are scarce despite they are the most appropriate for following the epidemics of HIV-1 and evaluating the effectiveness of preventive interventions. The rate of incidence of infection by HIV-1 in intravenous drug users (IVDU) and the determining factors for such infection are estimated in the present study. METHODS: 3,693 IVDU attended in two centers for information and prevention of AIDS from 1987-1992, were included in the study. Out of 1,868 seronegative subjects, 579 attended a second visit in which a new serologic test was performed against HIV-1. The rates of incidence and 95% confidence intervals were calculated with a Poisson regression model being adjusted in order to control the confounding factors. RESULTS: The estimated global incidence was 11.7/100 people/year (CI 95%: 9.6; 14.46). Those highest rates were observed in IVDU who exchanged syringes (23.5/100 people/year), in those who had more than one sexual partner in the previous month (21.2/100 people/year), in those referring a history of prostitution (22.6/100 people/year), in carriers of hepatitis B antigens (25.6/100 people/year) and in the IVDU younger than 20 years of age (16.5/100 people/year). The variables which independently predicted the rates of incidence were: the exchange of syringes (RR = 6.85; CI 95%: 3.99; 11.75), having more than one different sexual partner in the previous month (RR = 3.51; CI 95%: 1.70; 7.20) and age, with lower risk in those under the age of 20 (RR = 0.49; CI 95%: 0.25; 0.95). CONCLUSIONS: The incidence of infection by type-1 human immunodeficiency virus observed in intravenous drug users is the highest described in Western countries. Preventive measures in general must be intensified particularly in subgroups of intravenous drug users with the greatest risk of seroconversion.