Skip to main content
. Author manuscript; available in PMC: 2015 Jun 3.
Published in final edited form as: Am J Drug Alcohol Abuse. 2014 Jan 9;40(2):151–156. doi: 10.3109/00952990.2013.860983

Table 3.

Multivariate cox regressions for factors associated with initiation of injection drug use among female and male street-involved youth in Vancouver, Canada, September 2005–November 2011 (N = 422).

Characteristic Adjusted HRa (95% CIb) p Value
Model I: Male gender
    Homelessd (yes vs. no) 1.73 (0.98–3.06) 0.058
    Sex workd (yes vs. no) 4.74 (1.45–15.5) 0.010
    Downtown Eastside residencyd (yes vs. no) 1.95 (1.12–3.41) 0.018
Model II: Female gender
    Homelessd (yes vs. no) 1.79 (0.76–4.21) 0.185
    Non-injection crystal meth usec,d (yes vs. no) 4.63 (1.89–11.35) 0.001
    Crack cocaine smokingc,d (yes vs. no) 1.79 (0.73–4.41) 0.202
    Non-injection heroine usec,d (yes vs. no) 1.66 (0.62–4.44) 0.309
a

HR, Hazard Ratio

b

CI, Confidence Interval

c

Lagged to previous follow-up

d

Denotes behavior in the preceding 6 months