Table 4.

Random-effects model of the relationships between shift schedule characteristics and sleep quality, stratified by morning and evening chronotype (scale 0–10; higher score indicating poorer sleep quality). Boldface indicates statistical significance (P< 0.05). [CI=confidence interval.]

	Morning chronotype	Intermediate chronotype	Evening chronotype
	B (95% CI)a	B (95% CI) a	B (95% CI) a
Type of shift b
Morning	0.27 (-0.62–1.17)	0.31 (-0.39–1.01)	0.26 (-0.38–0.90)
Day	Ref.	Ref.	Ref.
Evening	0.35 (-0.16–0.86)	0.06 (-0.40–0.52)	0.01 (-0.33–0.35)
Night	1.14 (0.74–1.55)	0.66 (0.37–0.96)	0.48 (0.25–0.71)
Day off following workday	0.43 (0.03–0.83)	-0.10 (-0.41–0.20)	0.07 (-0.16–0.31)
≥2 days off following a working day	-0.09 (-0.46–0.29)	-0.05 (-0.34–0.24)	-0.06 (-0.29–0.17)
Time between shifts (hours rest)
>16	Ref.	Ref.	Ref.
11–16	-0.15 (-0.60–0.29)	0.42 (0.11–0.77)	-0.15 (-0.37–0.06)
<11 (quick return)	-0.07 (-1.14–1.00)	-0.20 (-0.73–0.48)	0.41 (-0.12–0.95)
Rotation direction
Forward	Ref.	Ref.	Ref.
Stable	-0.32 (-0.80–0.16)	0.21 (-0.13–0.55)	-0.08 (-0.33–0.18)
Backward	-0.12 (-0.73–0.49)	0.40 (-0.04–0.85)	0.06 (-0.29–0.40)
Successive night shifts
0	Ref.	Ref.	Ref.
1	1.22 (0.90–1.55)	0.56 (0.31–0.80)	0.59 (0.41–0.77)
2	0.66 (0.14–1.18)	0.89 (0.55–1.23)	0.36 (0.11–0.61)
3	..c	1.15 (0.47–1.83)	0.31 (-0.06–0.68)
≥4	..	..	0.07 (-0.40–0.54)

^aAnalyses model per roster characteristic adjusted for age, gender, education, children living at home under the age of 12, general health, and type of shift. Successive night shifts was not adjusted for type of shift.

^bMorning shift (starts after 06:00 and before 11:00 and mostly at 07:00 hours); day shift (starts after 11:00 and before 14:00 hours); evening shift (starts after 14:00 and ends before 01:00 hours); night shift (starts after 14:00 and ends after 01:00 hours).

^cCensored due to number of observations is too low (<20) for reliable estimations.