Table 4.
Association between alcohol use and head and neck cancer risk by level of education and ALDH2 genotype
| ALDH2*1/*1 (normal function) | ALDH2*1/*2 or *2/*2 (slow or nonfunctional) | |||||
|---|---|---|---|---|---|---|
| Education level ≤ junior high | Education level ≥ high school/ technical school | Level‐specific heterogeneity‐P | Education level ≤ junior high | Education level ≥ high school/technical school | Level‐specific heterogeneity‐P | |
| OR (95% CI) a | OR (95% CI) a | OR (95% CI) a | OR (95% CI) a | |||
| Alcohol drinking | ||||||
| Never + occasional | Reference | Reference | Reference | Reference | ||
| Former regular | 0.68 (0.32‐1.44) | 1.57 (0.77‐3.16) | 0.06 | 1.83 (1.02‐3.27) | 1.98 (1.04‐3.77) | .97 |
| Current regular | 1.21 (0.64‐2.28) | 0.74 (0.45‐1.23) | 0.23 | 4.32 (2.74‐6.80) | 2.80 (1.81‐4.33) | .14 |
| Overall heterogeneity‐P b = .01 | Overall heterogeneity‐P b = .30 | |||||
| Ever (former + current) | 1.02 (0.55‐1.89) | 0.86 (0.53‐1.39) | 0.71 | 3.32 (2.23‐4.96) | 2.56 (1.72‐3.83) | .28 |
| Overall heterogeneity‐P b = .71 | Overall heterogeneity‐P b = .28 | |||||
| Never | Reference | Reference | Reference | Reference | ||
| Monthly | 0.99 (0.30‐3.29) | 1.41 (0.57‐3.47) | 0.57 | 1.03 (0.37‐2.90) | 1.66 (0.70‐3.94) | .71 |
| Weekly | 0.97 (0.43‐2.17) | 0.81 (0.42‐1.57) | 0.75 | 3.15 (1.51‐6.56) | 1.59 (0.84‐3.01) | .08 |
| Daily | 1.01 (0.49‐2.09) | 0.91 (0.52‐1.58) | 0.90 | 3.42 (2.21‐5.32) | 3.22 (2.03‐5.11) | .74 |
| Overall heterogeneity‐P b = .90 | Overall heterogeneity‐P b = .33 | |||||
| Never c | Reference | Reference | Reference | Reference | ||
| Light | 0.82 (0.36‐1.84) | 0.86 (0.48‐1.54) | 0.67 | 2.03 (1.14‐3.62) | 1.71 (0.98‐2.99) | .29 |
| Moderate | 0.96 (0.42‐2.18) | 0.89 (0.45‐1.76) | 0.96 | 2.86 (1.44‐5.66) | 1.61 (0.82‐3.14) | .22 |
| Heavy | 1.26 (0.57‐2.79) | 1.07 (0.56‐2.04) | 0.97 | 4.22 (2.44‐7.30) | 6.63 (3.58‐12.29) | .31 |
| Overall heterogeneity‐P b = .96 | Overall heterogeneity‐P b = .22 | |||||
a
Odds ratio (OR) and 95% confidence interval (CI) were calculated using unconditional logistic regression, adjusted for age, sex, oral hygiene score, consumption of vegetables and fruits, and use of betel quids (pack‐years) and cigarette (pack‐years).
b
Overall heterogeneity‐P evaluated the combined difference by education across the different levels of the lifestyle factors. The heterogeneity‐P was calculated excluding the unknowns.
c
Light, <14 g/d; moderate, 14‐42 g/d; heavy, >42 g/d.