Seroprevalence of and risk factors for HIV-1 infection in injection drug users in Montreal and Toronto: a collaborative study.

OBJECTIVE

To determine the prevalence of antibodies to HIV-1 and risk factors for HIV-1 infection among injection drug users.

DESIGN

Questionnaire survey. A venous blood sample was taken for HIV-1 antibody testing.

SETTING

Montreal and Toronto.

PARTICIPANTS

A total of 810 subjects who had used injection drugs in the previous 6 months recruited mainly from treatment centres and from the street in Montreal (425 subjects) and from treatment centres in Toronto (385 subjects) between September 1988 and September 1990. The overall participation rate was 82%.

OUTCOME MEASURES: HIV-1 seropositivity, sociodemographic and behavioural risk factors for HIV-1 infection.

RESULTS

The overall seroprevalence rate of HIV-1 infection was 4.8% (95% confidence limits [CL] 3.5 and 6.5). In Montreal the rate was 8.2% (95% CL 6.0 and 11.2), and in Toronto 1.0% (95% CL 0.4 and 2.6) (p < 0.001). Seropositive subjects were significantly older (p = 0.041) and were more likely to have a history of imprisonment (p = 0.006) than seronegative subjects. In univariate analysis seropositivity was associated with the following behaviours: more frequent cocaine use (p < 0.001), injecting drugs in "shooting galleries" (p = 0.002), sharing equipment with a person known to be HIV-1 seropositive (p = 0.006), "booting" fresh blood (p = 0.004), homosexual or bisexual orientation (p = 0.006), engaging in prostitution (p < 0.001) and, for men, number of male sexual partners in the previous 6 months (p = 0.007). In multivariate analysis the determinants of HIV-1 seropositivity were Montreal as the city of recruitment (odds ratio [OR] 6.7, 95% CL 2.32 and 19.42), engaging in prostitution (OR 2.13, 95% CL 1.01 and 4.75), a history of imprisonment (OR 3.51, 95% CL 1.33 and 9.29) and sharing equipment with a person known to be HIV-1 seropositive (OR 4.43, 95% CL 1.43 and 13.74).

CONCLUSIONS

Our findings show that HIV-1 is circulating among injection drug users in Montreal and Toronto and that both drug use and sexual behaviours are implicated in the transmission of infection in the populations studied. Adapted preventive programs should be developed to prevent further spread of HIV-1 infection in this population.