Table 4. Association between genetically predicted alcohol consumption (glass/day) and atrial fibrillation in men.
| Populations | No. | Model | OR (95% CI) | p value |
|---|---|---|---|---|
| Total population | 3,577 | Unadjusted | 3.00 (1.13–8.68) | 0.034 |
| Adjusted | 3.17 (1.18–9.24) | 0.028 | ||
| F statistics = 864.6, adjusted R2 = 0.195, p value ≤0.001* | ||||
| Excluding former drinker | 3,095 | Unadjusted | 2.05 (1.05–4.28) | 0.043 |
| Adjusted | 2.12 (1.08–4.44) | 0.037 | ||
| F statistics = 1,212, adjusted R2 = 0.281, p value ≤0.001* | ||||
Log-transformed alcohol consumption in men was estimated and inversely transformed according to whether the ALDH2 rs671 genotype was GG or GA/AA.
All models were adjusted for age, physical activity, smoking status, body mass index category, hypertension, diabetes, and education status.
ALDH2 = aldehyde dehydrogenase 2; CI = confidence interval; OR = odds ratio.
*F statistics, adjusted R-squared, and p value of regression model which predicted log-transformed alcohol consumption using ALDH2 genotype GG or GA/AA.