TABLE 2.

Hazard Ratios From a Cox Proportional Hazard Model With Time-Dependent Covariates in Men Examining the Effect of Marriage on Risk for Onset of Alcohol Use Disorder

	Model 1		Model 2		Model 3aa		Model 3ba		Model 4
Variable	Hazard Ratio	95% CI	Hazard Ratio	95% CI	Hazard Ratio	95% CI	Hazard Ratio	95% CI	Hazard Ratio	95% CI
Marriage	0.41	0.40, 0.42	0.40	0.39, 0.41					0.41	0.39, 0.42
Marriage to spouse with     no history of registration     for alcohol use disorder					0.38	0.37, 0.39	0.38	0.37, 0.39
Marriage to spouse with     a history of registration     for alcohol use disorder							1.29	1.16, 1.43
Birth year (by year)			1.01	1.01, 1.02	1.01	1.01, 1.02	1.01	1.01, 1.02	1.01	1.01, 1.02
Parental education
Middle versus low level			0.90	0.88, 0.92	0.90	0.89, 0.92	0.90	0.89, 0.92	0.90	0.89, 0.92
High versus low level			0.64	0.63, 0.66	0.64	0.63, 0.66	0.64	0.63, 0.66	0.64	0.63, 0.66
Early-onset deviant behavior			5.33	5.23, 5.43	5.31	5.21, 5.41	5.31	5.21, 5.41	5.31	5.21, 5.40
Family history of alcohol   use disorder			2.26	2.22, 2.30	2.26	2.22, 2.30	2.26	2.22, 2.30	2.29	2.25, 2.33
Spouse with a history   of registration for alcohol   use disorder					3.37	3.02, 3.75			3.18	2.84, 3.56
Interaction of family history   of alcohol use disorder   and marital status									0.89	0.85, 0.95

^aModels 3a and 3b are reparameterizations of one another with an identical model fit and explanatory power. Model 3a demonstrates the impact of marriage to a spouse with no history of registration for alcohol use disorder and then shows the additional risk involved if that spouse has a history of alcohol use disorder (increasing risk ~3.4-fold). Model 3b reparameterizes this to allow us to compare directly the risk association with marriage to aspouse without versus aspouse with a history of registration for alcohol use disorder (hazard ratios of 0.38 versus 1.29).