Table 6 Predicting the propensity to being susceptible versus immune to any sickness absence (zero‐inflation part) and the duration of sickness absence, if susceptible (negative binomial part).

	Zero‐inflated part (ZI)		Negative binomial part (NB)
	Coefficient	OR (95% CI)	Coefficient	MR (95% CI)
Baseline (odds for ZI, mean for NB)	−0.54	0.58 (0.37 to 0.91)	1.73	5.67 (4.12 to 7.80)
Male blue‐collar	2.02	7.53 (4.76 to 11.90)	0.34	1.40 (1.04 to 1.88)
Male white‐collar (reference)	0	1	0	1
Female white‐collar	1.44	4.24 (2.24 to 8.00)	−0.26	0.77 (0.52 to 1.15)
Age (years)
18–39 (reference)	0	1	0	1
40–44	−0.37	0.69 (0.40 to 1.21)	−0.11	0.90 (0.68 to 1.17)
45–61	−0.72	0.49 (0.31 to 0.75)	−0.12	0.89 (0.72 to 1.10)
Musculoskeletal impairment due to work (per 1 unit; scale 0–10)	0.13	1.13 (1.02 to 1.26)	0.13	1.14 (1.09 to 1.19)
Insufficient sleep (per hour)	0.32	1.38 (1.14 to 1.66)	−0.09	0.92 (0.86 to 0.98)
Predicted work ability
“Able”	0	1	0	1
“Uncertain”	0.11	1.12 (0.63 to 1.98)	0.69	2.00 (1.53 to 2.62)
“Not able”	0.63	1.87 (0.50 to 6.99)	1.13	3.09 (1.89 to 5.05)

Estimated model coefficients, odds ratios (OR) and mean ratios (MR) with 95% CI from fitting a zero‐inflated negative binomial regression Model 3 including age, occupational grade, gender, musculoskeletal impairment, insufficient sleep and self‐rated future working ability as covariates.

The estimate of the dispersion parameter was ϕ = 0.62.

“Zero‐inflated part” refers to the model component for predicting membership to the subpopulation A with high propensity to zero absence, and “negative binomial part” to the component predicting the days on sick leave among the susceptible subpopulation B. To facilitate interpretation, for the zero‐inflation part we have provided the ORs associated with the complementary propensity to having any sickness absence—that is, inclusion in subpopulation B.