Accounting for individual differences and timing of events: estimating the effect of treatment on criminal convictions in heroin users

Table 1 Cox proportional hazards regression models

	Univariate regression models		Multiple regression model 1
Variable	Hazard ratio (95% CI)	p-value	Hazard ratio (95% CI)	p-value
OMT	0.40 (0.35,0.45)	<0.001	0.79 (0.72,0.87)	<0.001
Female gender	0.48 (0.40,0.57)	<0.001	0.75 (0.66,0.85)	0.001
Age [10 years]	0.60 (0.54,0.67)	<0.001	0.79 (0.73,0.85)	<0.001
>27 criminal days prior to OMT application	4.83 (4.19,5.55)	<0.001	1.50 (1.35,1.66)	<0.001
Criminal conviction while on waiting list	6.65 (5.70,7.76)	<0.001	2.84 (2.47,3.25)	<0.001
>1 OMT period	1.74 (1.33, 2.28)	<0.001	1.43 (1.20,1.70)	<0.001
Criminal conviction last 30 days	93.9 (87.1, 101.2)	<0.001	45.2 (40.4, 50.5)	<0.001

Cox proportional hazards regression models with day of criminal conviction as a recurrent event outcome and opoid maintenance treatment (OMT) a time-dependent explanatory variable. Results should be interpreted with care as the multiple model failed the assumption of proportional hazards.

ISSN: 1471-2288