Table 3.

Logistic Regression and Instrumental Variable Estimated Associations Between Physical Abuse, Household Alcohol Abuse and Household Mental Illness Among Individuals With a Sibling (n=5923), National Longitudinal Survey of Youth 1979, 1979-2010.

	Physical Abuse		Household Alcohol		Household Mental
Logistic regression	OR	95% CI	OR	95% CI	OR	95% CI
Obesity age 25	1.43	0.91, 2.26	1.02	0.71, 1.45	0.92	0.44, 1.95
Obesity age 40	1.08	0.76, 1.53	0.96	0.72, 1.27	0.72	0.46, 1.14
Obesity II age 40	1.13	0.72, 1.79	1.09	0.77, 1.56	0.89	0.49, 1.60
Smoking, Ever	1.65	1.20, 2.26	1.51	1.14, 1.99	1.41	0.83, 2.38
Smoking, Current	1.84	1.30, 2.60	1.41	1.05, 1.89	1.67	1.01, 2.75
Instrumental Variable	OR	95% CI	OR	95% CI	OR	95% CI
Obesity age 25	0.76	0.10, 5.50	0.69	0.35, 1.36	0.34	0.09, 1.21
Obesity age 40	1.39	0.69, 2.79	0.77	0.56, 1.07	0.45	0.13, 1.53
Obesity II age 40	1.59	0.46, 5.50	0.83	0.52, 1.33	ID
Smoking, Ever	1.63	0.95, 2.80	1.36	1.08, 1.71	2.90	0.31, 2.73
Smoking, Current	1.84	0.31, 10.90	2.01	1.01, 3.98	ID

Table notes: ID indicates insufficient data to estimate the coefficient. Instrumental variable 95% confidence intervals did not account for the primary sampling unit clustering of the sample.