Table 1.

Baseline characteristics of street-involved youth who did and who did not initiate injection drug use and Univariate and Multivariate Cox regression analysis of factors associated with injection drug use initiation among street-involved youth in Vancouver (n = 422).

Characteristic	Baseline characteristics		Univariate and Multivariate Cox regression analysis
	Injection initiation		Unadjusted HR† (95% CI)	p-value	Adjusted HR† (95% CI)	p-value
	Yes (n=77) n (%)	No (n=345) n (%)
Age	21.3 (19.6, 22.5)	21.6 (19.6, 23.3)	0.96 (0.88 – 1.04)	0.297
Per year older
Caucasian
Yes	54 (70.1)	204 (59.1)	1.50 (0.92 – 2.44)	0.104
No	23 (29.9)	141 (40.9)
Male
Yes	54 (70.1)	235 (68.1)	1.02 (0.63 – 1.67)	0.920
No	23 (29.9)	110 (31.9)
Homeless
Yes	57 (74.0)	239 (69.3)	1.80 (1.13 – 2.88)	0.013	1.80 (1.13 – 2.87)	0.014
No	20 (26.0)	106 (30.7)
Non-Injection Crystal Meth Use*
Yes	41 (53.2)	107 (31.0)	1.83 (1.16 – 2.87)	0.009	1.82 (1.16 – 2.85)	0.010
No	36 (46.8)	238 (69.0)
Non-Injection Cocaine Use*
Yes	31 (40.3)	175 (50.7)	0.99 (0.62 – 1.57)	0.476
No	46 (59.7)	170 (49.3)
Crack Cocaine Smoking*
Yes	52 (67.5)	184 (53.3)	1.38 (0.88 – 2.18)	0.164
No	25 (32.5)	161 (46.7)
Non-Injection Heroin Use*
Yes	18 (23.4%)	48 (13.9)	1.83 (1.06 – 3.14)	0.029
No	59 (76.6%)	297 (86.1)
Marijuana Use*
Yes	64 (83.1)	316 (91.6)	0.87 (0.47 – 1.61)	0.653
No	13 (16.9)	29 (8.4)
Daily Alcohol Use*
Yes	14 (18.2)	59 (17.1)	0.99 (0.55 – 1.80)	0.982
No	63 (81.8)	286 (82.9)

^*refers to the activities lagged to the pervious available follow-up

^†HR=Hazard Ratio