Initiation to drug injection among street youth: a gender-based analysis.

OBJECTIVE(S): To estimate the incidence rate of initiation into drug injection and to identify predictors of initiation into drug injection separately among street girls and boys.
DESIGN: Data from two consecutive prospective street youth cohort studies (1995-2001 and 2001-2005) were used to conduct these analyses, stratified by gender.
METHODS: Data were collected using semi-annual interviewer-administered questionnaires. Variables from the following domains were considered in Cox regression models: socio-demographic characteristics, early and current substance abuse, marginalization, childhood traumatic sexual events and injection exposure.
RESULTS: Of the 946 youth who had never injected drugs at study entry, 86.4% completed at least two questionnaires representing 243 girls and 574 boys. Incidence rates of injection of 7.0 and 5.9 per 100 person-years were observed among these girls and boys respectively. Among girls, cocaine or crack use (adjusted hazard ratio (AHR)=1.97), heroin use (AHR=2.86), homelessness (AHR=2.49) and hanging out regularly with people who inject (AHR=4.46) all independently increased risk of first injection. Among boys, age decreased risk of initiating injection (AHR=0.90/year), while cocaine or crack use (AHR=2.14), heroin use (AHR=3.56), homelessness before age 16 (AHR=1.68), incest or rape before age 14 (AHR=1.98) and hanging out regularly with people who inject (AHR=1.66) all independently increased this risk.
CONCLUSIONS: Our findings suggest similar rates of initiation among girls and boys; however, factors associated with initiation vary by gender. This might lead to the design of more effective programs to prevent initiation into drug injection.