Prevalence of HIV infection and risk behaviours among Montreal street youth

É. Roy MD MSc, N. Haley MD, P. Leclerc MSc, N. Lemire MA, J-F. Boivin Md ScD, J-Y. Frappier MD and C. Cleaessens MSc

Montreal Regional Public Health Department,, Department of Epidemiology and Biostaatistics, McGill University, Department of Pediatrics, University of Montreal and Laboratoire de Santé Publique du Québec, Canada.

One of the constants found among all street youth is their precarious living conditions, which include poverty, residential instability, and emotional and psychological vulnerability. These conditions often lead to behaviours that expose the youth to sexually transmitted diseases (STDs) and blood-borne viruses such as HIV. Data on HIV infection prevalence among this group are still rare and are available mostly from Latin America and the United States.

In the United States, estimates varied according to recruitment site and city. In a New York City study of homeless youth aged 15-20 years conducted in 1987-90, the prevalence rate was 5.3%. Drug injection, homosexual activity among boys, prostitution, and a history of STD were associated with the infection. At a later period, between 1990 and 1992, an American sentinel network composed of medical clinics showed that HIV infection prevalence was 1.1% among youth recruited in 5 homeless centres; prevalence varied among the 4 participating cities from 0% (Dallas) to 4.1% (San Francisco). The highest rates were among males reporting sex with men.

Due to the limited information available on HIV among street youth in Canada, a seroprevalence study among Montreal street youth was initiated in 1995. In addition to prevalence, the study was designed to determine risk factors associated with infection.


The 20 agencies offering free services and seeing a significant number of youth per year were invited to participate. Selection criteria were chosen to capture as much as possible the whole spectrum of street youth. Youth have different experiences with varying degrees of street involvement and residential instability. Inclusion criteria were: being 12-25 years of age; French or English speaking; being able to provide informed consent and to complete the questionnaire; and being "street active". Youth were considered street-active if they had either used the services of Montreal street youth agencies or been without a place to sleep for 3 days or more in the last 6 months. The 3-days period was chosen to eliminate youth having left home for too short a period to have experienced any kind of real street involvement.

The interview included a 30 min face-to-face questionnaire and the collection of 2 saliva samples (more precisely gingival exudate) for HIV testing using the Orasure device. The questionnaire included 70 sociodemographic and behavioural questions.


We included 909 participants: 37.2% (338) were recruited in emergency shelters, 35.5% (323) in outreach vans, 14.0% (127) through outreach agencies, and 13.3% (121) in drop-in centres. Overall, 71.1% were male and the mean age was 19.4 years (range: 13-25 years). Most youth (93.8%) were born in Canada, as were their parents, with 84.8% having a father and 89.6% a mother born in Canada.

Most participants (94.1%) had had at least one episode of homelessness in their life, meaning that they had needed to look for a place to sleep (such as a shelter) or had had to sleep outside in a park or an abandoned house, in a bus or train station, or with friends or relatives because they had nowhere to go, did not want to return home, or did not have a home. The mean age at first episode of homelessness was 15.3 years old.

Most street youth reported more than one source of income during the last 6 months. In all, 30.2% had an illegal main source of income (e.g. selling drugs, panhandling, prostitution or theft) and 69.8% a legal main source of income.

Approximately half (47.0%) of the participants reported having at least one tattoo (1 to 38; mean 3.3) while 73.9% had at least one part of their body pierced (29.3% excluding ears).

Overall, 41.1% of girls and 22.2% of boys reported having had at least one STD. Almost half of the girls (47.1%) had been pregnant at least once and 35.6% of the boys reported having had intercourse with a girl which resulted in pregnancy. Two-thirds of the girls (66.9%) and one-quarter (27.1%) of boys reported having been sexually abused; 64.7% of girls and 33.9% of boys were abused by more than one person.

Youth reported using alcohol and drug frequently. Among those reporting ever drinking, 48.2% drank at least once a week in the previous month; this figure includes 8.9% who drank daily. A total of 63.6% reported ever binging on alcohol, with 57.3% doing so in the last month. The mean age at first alcohol binge was 13.4 years.

Roughly 97% of subjects had used at least one drug in their lifetime and 76.2% had used 4 different drugs or more. A total of 56.8% of youth reported using drugs regularly (more than twice a week).

In all, 36.4% of youth reported having injected drugs (332), with similar proportions among girls (38.8%) and boys (35.4%). Age at first injection ranged from 8-25 years with a mean of 17.0 years (girls: 16.2; boys: 17.3). One-third of these youth had injected drugs 10 times or less, and 43% more than 100 times. A total of 58.2% of injectors had borrowed a used needle at least once and 67.5% had borrowed other injection materials (cotton, spoons, water, or something else). Overall, 26 youth (7.9% of injectors) reported borrowing a used needle from an HIV-infected person; 12 of them were aware of the HIV status of their partner at the time of sharing. Two-thirds of youth injectors were current injectors (previous 6 months).

Serological status

Among the 909 participants, 17 (14 boys and 3 girls) tested positive for HIV, for a prevalence rate of 1.87% (95% CI: 1.09-2.98). Prevalence varied with age, with a rate of 0.49% (95% CI: 0.11-1.4) among youth aged 13-20 years (all were above 17 y.o.) and 4.7% (95% CI: 2.6-7.7) among youth aged 21-25 years. HIV prevalence did not vary significantly with sex: 1.1% among girls and 2.2% among boys (P=0.42).

Factors significantly associated with HIV prevalence in the multivariate logistic regression analysis are: being over 20 years of age, having ever injected drugs, having been engaged in prostitution, and being born outside Canada.

Since we suspected possible interactions between sex and some of the selected variables, such as prostitution, and since 14 out of the 17 HIV-infected subjects were boys, we conducted a multivariate logistic regression analysis among boys only. Three variables were independently associated with infection among boys: being over 20 years of age (AOR= 6.7; 95% CI: 1.4-31.0), having ever injected drugs (AOR= 7.1; 95% CI: 1.5-32.9), and having been engaged in prostitution (AOR= 5.9; 95% CI: 1.8-19.6).

Published in: International Journal of STD & AIDS 2000:1;241-247