. Author manuscript; available in PMC: 2015 Jul 30.

Published in final edited form as: J Clin Psychiatry. 2014 Oct;75(10):1047–1053. doi: 10.4088/JCP.13m08967

Table 3.

Cox regression modelling with risk factors for suicide among al individuals releases from prison in S weden 2005-2009 (n=38,955).

	Unadjusted Hazard ratio (95% CI)	Adjusted Hazard ratio (95% CI)
Female gender	1.24 (0.68-2.24)	1.05 (0.57-1.94)
Age 17-24^§	0.97 (0.61-1.54)	reference
Age 25-44^§	1.11 (0.78-1.57)	0.86 (0.53-1.40)
Age 45-64^§	0.86 (0.58-1.28)	0.67 (0.39-1.16)
Age 65-^§	1.54 (0.58-4.83)	1.53 (0.46-5.14)
Born in Sweden vs. abroad	2.70^*** (1.60-4.56)	2.09^** (1.23-3.56)
Psychotic disorder	2.35^*** (1.47-3.76)	1.59 (0.97-2.61)
Affective disorder	1.53 (0.71-3.28)	1.00 (0.46-2.18)
Substance use disorder	3.12^*** (2.12-4.59)	2.13^** (1.39-3.26)
Personality disorder	1.38 (0.76-2.51)	0.66 (0.35-1.24)
Any psychiatric disorder (including substance use disorder)	3.49^*** (2.28-5.34)	not used due to predictor collinearity
Any psychiatric disorder (excluding substance use disorder)	1.06 (0.54-2.08)	not used due to predictor collinearity
Any previous suicide attempt	3.58^*** (2.51-5.09)	2.58^*** (1.76-3.78)
Any violent crime	1.50 (0.99-2.26)	1.20 (0.79-1.88)
Released once^#	0.68 (0.47-1.00)	reference
Released twice^#	1.54^* (1.01-2.34)	1.33 (0.86-2.04)
Released more than twice^#	1.12 (0.60-2.09)	0.94 (0.49-1.78)

Notes:

p<0.05,

^**

p<0.01,

^***

p<0.001.

^§

Used as a dichotomous variable in the unadjusted analysis, and together with the three other age strata as a categorical variable for the adjusted, multivariate analysis.

Used as a dichotomous variable in the unadjusted analysis, and together with the two other “Released…” variables as a categorical variable for the adjusted, multivariate analysis.