Table 5.

Predictors of a prolonged duration of sickness absence during 12-month follow-up (results of univariable and multivariable Cox regression, N = 657)

Independent variables	Univariable Cox regression			Multivariable Cox regression
	HRa	95% CI	p-value	HR	95% CI	p-value	PIb	95% CI
Age
<50 yrs	1(ref)
≥50 yrs	0.99	0.78–1.26	0.9
Gender
Female	1(ref)
Male	0.79	0.64–0.99	0.04*	0.77	0.62–0.97	0.03*	0.56	0.51–0.62
Employment status
Temporary agency	1(ref)
Loss of employment contractc Unemployed	0.56 0.52	0.31–1.02 0.30–0.91	0.06 0.02*	0.48 0.47	0.26–0.90 0.27–0.84	0.02* 0.01*	0.68 0.68	0.53–0.79 0.54–0.79
Psychological symptoms d
Mild-to-moderate	1(ref)
Severe	0.63	0.46–0.85	0.003*	0.64	0.46–0.87	0.005*	0.61	0.53–0.68
RTW self-efficacy e
High	1
Low	0.80	0.60–1.07	0.1	0.86	0.63–1.17	0.3	0.54	0.46–0.61
Study
Co-Work	1 (ref)
Brainwork Cohort	0.95 0.96	0.73–1.23 0.71–1.31	0.7 0.8
Study allocation
Intervention	1 (ref)
Control Cohort	0.87 0.93	0.68–1.12 0.70–1.24	0.3 0.6

^a An HR of < 1 indicates a prolonged duration of sickness absence compared to the reference group

^b The PI indicates the probability of a prolonged duration of sickness absence compared to the reference group. A PI > 0.5 indicates a higher probability

^c Loss of employment contract during sickness absence period

^d Cut-off point based on the medians of the GHQ12 (Brainwork participants) and 4DSQ (Cohort and Co-Work participants)

^e Cut-off point based on the medians of the RTW-SE scale (Brainwork participants) and ASE-SE subscale (Cohort and Co-Work participants)

*p < 0.05