Table 3.

Multivariate adjusted hazard ratios (HR) and 95% confidence intervals (CI 95%) for suicide attempt in people sickness absent due to stress-related mental disorders (N=36,304)

Characteristics	HR*	CI 95%
Socio-demographic characteristics
Age (years)
17–25	1.98	1.31–2.98
26–35	1.18	0.86–1.64
36–45	1
46–55	1.13	0.80–1.60
56–65	0.44	0.24–0.82
Education (years)
0 to 9	2.43	1.63–3.63
10 to 12	2.00	1.45–2.78
Above 12	1
Family situation
Married/cohabiting with children a	1
Married/cohabiting without children a	0.95	0.54–1.66
Single people without children a	1.23	0.89–1.7
Single people with children a	1.40	0.99–1.96
Sickness absence characteristics b
Duration (days) c
1–14	1
15–90	1.19	0.85–1.68
91–180	1.85	1.20–2.84
181–365	1.86	1.14–3.05
> 365	2.20	1.44–3.35
Number of spells (any disorders) d
1	1
2	1.44	1.07–1.94
> 2	2.02	1.27–3.20
Health care characteristics
Hospital stay due to mental diagnoses (days) e
No hospital stay	1
1 to 8	3.37	2.25–5.03
> 8	4.15	2.84–6.06
Hospital stay due to somatic diagnoses (days) e
No hospital stay	1
1 to 4	1.51	1.10–2.06
> 4	1.86	1.36–2.53
Outpatient care visits due to mental diagnoses (days) e
No visits	1
1	1.85	1.25–2.74
>1	3.08	2.20–4.31
Outpatient care visits due to somatic diagnoses (days) e
No visits	1
1–4	1.36	0.85–2.18
> 4	2.41	1.51–3.85

^* Analyses are adjusted for sex, age, education, type of family, duration of first sickness absence spell due to SRMD, number of spells due to SRMD, and all health care related predictors, ^a Children living at home; ^b Sickness absence due to stress-related mental disorders; ^c Duration of the first sickness absence spell due to stress-related mental disorders initiated in 2005; ^d Number of new sick leave spells initiated in 2005 regardless of diagnosis; ^e Cut-offs are based on the diagnosis-specific median.