The relative impact of smoking, alcohol use and drug use on general sickness absence among Norwegian employees

BMC Public Health

Table 3 Factors associated with having had three or more sickness absences in the past 12 months

	AOR	P	95% CI for AOR
Smoking
Daily	1.65	0.019	1.09–2.50
Occasional	1.19	0.432	0.77–1.84
Former	1.08	0.617	0.79–1.49
Daily snus use	1.18	0.492	0.73–1.90
Weekly binge drinking	1.76	0.147	0.82–3.78
Illegal drug use	0.77	0.737	0.16–3.65
Medical drug use	2.36	0.007	1.27–4.37
Poly- and drug use
Smoking and drug use	0.88	0.807	0.33–2.40
Smoking and weekly binge drinking	1.26	0.702	0.39–4.05
Drug use and weekly binge drinking	0.39	0.329	0.06–2.58
Male	0.70	0.024	0.52–0.96
Age (ref = younger than 30 yrs)
30–39 yrs	1.10	0.625	0.76–1.59
40–49 yrs	0.54	0.004	0.36–0.83
50–59 yrs	0.35	0.000	0.22–0.56
60+ yrs	0.36	0.002	0.19–0.69
Education (ref = middle school)
High school	1.84	0.225	0.69–4.94
University	1.51	0.409	0.57–4.03
Constant	0.14	0.000	0.05–0.42

Multilevel logit regression (mixed effect logistic regression) with sickness absence (dummy variable) as the dependent variable, controlled for random effects of branch and workplace. AOR = Adjusted Odds Ratio
N = 1811, 22 workplaces (var = 0.12, SE = 0.13) within 5 branches (var = 0.14, SE = 0.08), LR-test for sign. Effect of random stratification: p < 0.001

ISSN: 1471-2458