Skip to main content
. Author manuscript; available in PMC: 2014 Aug 1.
Published in final edited form as: Drug Alcohol Depend. 2013 Jun 17;133(1):100–107. doi: 10.1016/j.drugalcdep.2013.05.025

Table 3.

Adjusted models for CP and patterns of alcohol use from 13 to 15 years

Low alcohol
frequency
(53.9%)
Medium alcohol frequency (30.5%) High alcohol frequency (157%)


Row % Row % Unadjusted OR Adjusted 1 OR Adjusted 2 OR Row % Unadjusted OR Adjusted 1 OR Adjusted 2 OR
Conduct problem trajectory Low (64.8%) 56.1 32.2 1.00 ref 1.00 ref 1.00 ref 11.7 1.00 ref 1.00 ref 1.00 ref
Childhood Limited (14.8%) 58.1 24.8 0.74 [0.43, 1.29] 0.69 [0.35, 1.35] 0.53 [0.21, 1.34] 17.2 1.41 [0.89, 2.26] 1.24 [0.79, 1.97] 1.27 [0.73, 2.22]
Adolescent Onset (11.2%) 38.6 37.3 1.68 [1.01, 2.79] 2.00 [1.09, 3.68] 2.05 [1.09, 3.85] 24.1 2.99 [1.84, 4.86] 3.34 [1.82, 6.12] 3.18 [1.63, 6.21]
Early Onset Persistent (9.2%) 50.2 19.1 0.66 [0.36, 1.22] 0.54 [0.26, 1.13] 0.56 [0.26, 1.21] 30.7 2.93 [1.81, 4.75] 2.44 [1.49, 3.99] 2.66 [1.34, 5.26]

Multinomial odds ratios shown are with reference to the low-alcohol frequency outcome category.

Percentages describing class distributions correspond to the sample of 5659 used for the unadjusted analysis.

Unadjusted model (n = 5659) –main effect of conduct problems trajectory on alcohol frequency pattern.

Adjusted model 1 (n = 5340) –adjusted for gender, housing tenure, over-crowding, parity and maternal education.

Adjusted model 2 (n = 4788) –additionally adjusted for pre-4y maternal alcohol use.