Table 2.

Crude* and adjusted† OR with 95% CI of prescribed sleep medication use by work time schedule over a 12 years follow-up period among Norwegian nurses (n=2028)

	Random effects model		Fixed effects model
	OR (95% CI)		OR (95% CI)
	Crude	Adjusted	Crude	Adjusted
Work schedule
Shift work with nights	1.00 (ref.)	1.00 (ref.)	1.00 (ref.)	1.00 (ref.)
Shift work without nights	1.18 (0.80 to 1.74)	1.20 (0.76 to 1.89)	0.84 (0.50 to 1.39)	0.66 (0.37 to 1.20)
Only day work	0.54 (0.32 to 0.91)	0.50 (0.27 to 0.93)	0.50 (0.26 to 0.97)	0.32 (0.14 to 0.70)

Models based on longitudinal data. The random effects model links changes in work schedule to changes in sleep medication use both within and between individuals, whereas the fixed effect model only assesses within-individual differences. Total study sample in random effects model, crude: 1893 nurses (8560 observations), and adjusted: 1761 nurses (7269 observations). Total study sample in the fixed effects model, crude: 193 nurses (1121 observations), and adjusted: 168 nurses (933 observations).

*Adjusted for wave.

†In the random effects model: Adjusted for wave, sex, age, age squared, work experience and symptoms of anxiety and/or depression (based on the Hospital Anxiety and Depression Scale) at baseline, marital status, children living at home, percentage of full-time position and type of workplace in each wave. In the fixed effects model: Adjusted for wave, marital status, children living at home, per cent of full position and type of workplace in each wave.