Table 2. Hazard ratios for the incidence of diabetes associated with long working hours by shift work schedules among all participants (non-shift workers n = 2371, shift workers n = 824).

Working Hours			Model 1a		Model 2b		Model 3c
	Person-years	Cased	HR	(95% CI)	HR	(95% CI)	HR	(95% CI)
Non-shift workers
35–44 h/week	6022	93	1.00		1.00		1.00
≥45 h/week	4295	45	0.83	(0.56–1.20)	0.79	(0.53–1.15)	0.84	(0.57–1.24)
Per 10-hour increasee			0.74	(0.51–1.01)	0.71	(0.49–0.99)	0.71	(0.49–0.99)
Shift workers
35–44 h/week	1412	17	1.00		1.00		1.00
≥45 h/week	2245	29	2.36	(1.22–4.79)	2.26	(1.17–4.57)	2.43	(1.21–5.10)
Per 10-hour increasee			1.20	(0.91–1.55)	1.14	(0.85–1.49)	1.19	(0.88–1.59)

CI, confidence interval; HR, hazard ratio.

^aAdjusted for age, fasting plasma glucose concentration at baseline, family history of diabetes (yes, no), and month in which participants underwent their annual health checkup at baseline.

^bModel 1+ additionally adjusted for educational level (≤12 years, >12 years), marital status (married, others, unknown), and job category (clerical workers, professional workers, manual workers, fire fighters).

^cModel 2+ additionally adjusted for body mass index, smoking status (current smoker or not), drinking habits (every day, sometimes, never), exercise habits (≥2 times per week and ≥30 minutes per session) (yes, others, unknown), and hypertension at baseline (systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg and/or under treatment) (yes, no).

^dCase refers to diabetes.

^eHRs for the association between increasing working hours by 10 hours per week and developing diabetes.