Table 3.
Multinomial logistic regression: associations between current alcohol use and alcohol delivery policies
| Current light drinking vs. no alcohol use in 30 days | Current heavy drinking vs. no alcohol use in 30 days | |||
|---|---|---|---|---|
| Predictors | aOR | 95% CI | aOR | 95% CI |
| (Intercept) | 3.34 | 1.56, 7.14 | 1.22 | 0.4, 3.67 |
| Policies | ||||
| On-premises delivery | 0.60 | 0.34, 1.06 | 0.51 | 0.22, 1.17 |
| Off-premises delivery | 1.93 | 1.24, 3.01 | 1.96 | 0.97, 3.95 |
| Demographics | ||||
| Age, years | ||||
| 22–24 | Ref | |||
| 14–17 | 0.10 | 0.05, 0.19 | 0.03 | 0.01, 0.15 |
| 18–21 | 0.47 | 0.32, 0.69 | 0.19 | 0.11, 0.35 |
| Race/ethnicity | ||||
| White | Ref | |||
| American Indian or Alaska Native | 0.76 | 0.36, 1.60 | – | – |
| Asian | 0.94 | 0.44, 2.03 | 1.37 | 0.62, 3.02 |
| Black | 1.10 | 0.61, 1.97 | 1.34 | 0.55, 3.25 |
| Hispanic/Latinx | 1.09 | 0.67, 1.75 | 1.67 | 0.58, 4.80 |
| Multiracial | 0.90 | 0.46, 1.76 | 3.45 | 0.60, 19.92 |
| Sexual identity | ||||
| Straight | Ref | |||
| Asexual/ace spectrum | 0.25 | 0.09, 0.69 | – | – |
| Bisexual/pansexual | 1.31 | 0.78, 2.19 | 1.37 | 0.62, 3.02 |
| Gay/lesbian | 1.56 | 0.88, 2.76 | 1.34 | 0.55, 3.25 |
| Queer | 1.95 | 0.98, 3.87 | 1.67 | 0.58, 4.80 |
| Questioning | 1.45 | 0.38, 5.46 | 3.45 | 0.60, 19.92 |
| Gender identity | ||||
| Woman/girl | Ref | |||
| Agender | 2.51 | 0.31, 20.20 | – | – |
| Gender queer | 1.15 | 0.40, 3.30 | 2.53 | 0.44, 14.61 |
| Man/boy | 0.80 | 0.50, 1.28 | 1.46 | 0.72, 2.94 |
| Non-binary | 1.38 | 0.66, 2.88 | 2.10 | 0.58, 7.61 |
| Questioning | 4.66 | 0.35, 62.72 | 1.84 | 0.09, 36.71 |
| Gender modality | ||||
| Cisgender | Ref | |||
| Trans and gender diverse | 0.51 | 0.28, 0.94 | 0.36 | 0.12, 1.06 |
| Not sure | 0.11 | 0.01, 1.24 | 0.39 | 0.03, 5.74 |
“–” Unstable estimates due to small sample cells. Log-likelihood: − 606.75; McFadden R2: 0.10997; likelihood ratio test: χ2 = 149.93 (p < 0.0001). aOR is adjusted odds ratio; 95% CI is 95% confidence intervals. Boldface indicates statistical significance at p < 0.05