Effect of Affective Temperaments Assessed by the TEMPS-A on the Relationship between Work-Related Stressors and Depressive Symptoms among Workers in Their Twenties to Forties in Japan

Table 5

Multivariate logistic regression model for the GHQ-12 (𝑁=728).

Variables	Model 1^(1,2)			Model 2^(1,3)
	OR	95% CI	𝑃 value	OR	95% CI	𝑃 value
Sex (reference category = female)	1.461	1.022–2.089	0.037	1.366	0.943–1.979	0.099
Age	0.964	0.936–0.994	0.019	0.977	0.947–1.009	0.157
Frequency of working on days off per month (reference category = one day or more)	1.551	0.965–2.492	0.070	1.372	0.837–2.249	0.210
Average working hours per day (reference category = 10 hours or more)	0.796	0.466–1.359	0.403	0.795	0.456–1.385	0.418
Effort	1.085	1.032–1.140	0.001	1.078	1.023–1.135	0.005
Rewards	0.975	0.949–1.001	0.056	0.978	0.951–1.006	0.116
Effort*Rewards	1.004	0.999–1.009	0.087	1.004	0.999–1.009	0.140
Upper tertile of overcommitment (reference category = present)	2.113	1.409–3.169	<0.001	1.589	1.015–2.485	0.043
Depressive ^(4)				0.624	0.164–2.368	0.488
Cyclothymic ^(4)				11.404	2.996–43.409	<0.001
Hyperthymic ^(4)				2.605	0.887–7.647	0.081
Irritable ^(4)				0.337	0.067–1.701	0.188
Anxious ^(4)				6.409	1.484–27.684	0.013

⁽¹⁾OR: odds ratio; Cl: confidence interval.
⁽²⁾Model 1: Hosmer-Lemeshow goodness of fit 𝜒²=3.484, df = 8, 𝑃=0.900.
⁽³⁾Model 2: Hosmer-Lemeshow goodness of fit 𝜒²=6.072, df = 8, 𝑃=0.639.
⁽⁴⁾Affective temperaments assessed by the TEMPS-A.