Table 3.

Univariate and Multivariable Models of Reasons for Drinking at Baseline Predicting New Onset Alcohol Dependence at 10 Years in 423 New Jersey Residents Without a Diagnosis of Alcohol Dependence at Baseline Conditioning Reasons for Drinking on Having a Positive Family History of Alcohol Problems^a

	IOR (95% CI)
Model 1: drinking for social facilitation
Unadjusted model:
No family history of alcoholism	reference
Positive family history and < median	1.08 (0.51, 2.29)
Positive family history and ≥ median	3.52 (1.89, 6.57)
Adjusted model:
No family history of alcoholism	reference
Positive family history and < median	1.12 (0.51, 2.47)
Positive family history and ≥ median	4.54 (2.31, 8.93)
Male	0.75 (0.42, 1.34)
White	0.87 (0.41, 1.87)
Age	0.26 (0.10, 0.66)
Binge drinking	1.29 (1.13, 1.46)
Model 2: drinking to reduce negative affect
Unadjusted model:
No family history of alcoholism	reference
Positive family history and < median	0.95 (0.42, 2.15)
Positive family history and ≥ median	3.39 (1.86, 6.20)
Adjusted model:
No family history of alcoholism	reference
Positive family history and < median	1.10 (0.47, 2.58)
Positive family history and ≥ median	3.79 (2.00, 7.18)
Male	0.82 (0.46, 1.46)
White	0.98 (0.45, 2.10)
Age	0.27 (0.11, 0.69)
Binge drinking	1.26 (1.11, 1.43)

^aHaving a positive family history and being below the median on the Reasons for Drinking Scale (RDS) was coded as 1 and 0 otherwise. A positive family history and being at or above the median on the RDS was coded 1 and 0 otherwise.