Table 2.
Odds ratios predicting age 21 and age 33 heavy episodic drinking (HED) and alcohol use disorder (AUD) associated with adolescent risk factors
| Variable | Age 21a |
Age 33b |
||||||
| HED |
AUD |
HED |
AUD |
|||||
| 1. Zero-order | 2. w/control | 3. Zero-order | 4. w/control | 5. Zero-order | 6. w/control | 7. Zero-order | 8. w/control | |
| Family | ||||||||
| General | ||||||||
| Family management | 0.92 | 0.76 | 1.42 | 1.19 | 1.47 | 1.36 | 2.15 | 1.88 |
| Family conflict | 1.03 | 0.97 | 1.20 | 1.14 | 1.48 | 1.68 | 1.61 | 1.58 |
| Family involvement | 0.76 | 0.64 | 0.99 | 0.85 | 1.40 | 1.36 | 1.66 | 1.57 |
| Family bonding | 1.00 | 0.86 | 1.49 | 1.33 | 1.50 | 1.39 | 1.72 | 1.54 |
| Alcohol specific | ||||||||
| Parental drinking attitude | 1.57 | 1.46 | 1.35 | 1.24 | 1.57 | 1.28 | 1.33 | 1.20 |
| Respondents’ involvement with parents’ drinking | 2.15 | 2.17 | 1.27 | 1.24 | 1.35 | 1.13 | 1.57 | 1.58 |
| Parental heavy episodic drinking | 1.73 | 1.68 | 1.66 | 1.60 | 1.32 | 1.14 | 1.67 | 1.49 |
| Sibling drinking | 1.67 | 1.59 | 1.64 | 1.51 | 1.25 | 1.08 | 1.31 | 1.05 |
| Peer | ||||||||
| General | ||||||||
| Antisocial peers | 1.65 | 1.14 | 2.74 | 2.06 | 1.43 | 0.88 | 2.20 | 1.38 |
| Antisocial opportunities | 1.30 | 0.95 | 2.42 | 1.91 | 1.51 | 1.17 | 2.60 | 1.81 |
| Alcohol specific | ||||||||
| Drinking peer | 2.55 | 2.55 | 2.59 | 2.50 | 1.16 | 0.81 | 2.26 | 1.82 |
| School | ||||||||
| General | ||||||||
| Bonding | 1.03 | 0.83 | 1.68 | 1.39 | 1.27 | 1.13 | 2.35 | 1.91 |
| Opportunity | 1.18 | 0.94 | 1.57 | 1.27 | 1.63 | 1.39 | 2.41 | 1.99 |
| Involvement | 0.79 | 0.73 | 1.09 | 1.02 | 0.95 | 0.94 | 1.23 | 1.15 |
| Reward | 1.46 | 1.31 | 1.43 | 1.27 | 1.14 | 0.97 | 1.68 | 1.47 |
| Alcohol specific | ||||||||
| Perception of drinking in school | 1.18 | 1.13 | 1.41 | 1.34 | 1.22 | 1.39 | 2.01 | 1.92 |
Notes: Coefficients in bold are significant at p < .05; coefficients in italics are marginally significant at p < .10.
Early alcohol use (ages 10–11), behavioral disinhibition (age 14), and gender were included in the models as covariates;
early alcohol use (ages 10–11), behavioral disinhibition (age 14), gender, and the corresponding age 21 problem alcohol involvement (e.g., for the model predicting HED at age 33, HED at age 21 was added as a covariate) were included in the models as covariates.