The bottom portion of Table 2 shows the means and standard deviations of total symptoms and SRE_5 across ADH1B genotype stratified by ALDH2 genotype. We performed ANOVA analyses to examine mean differences across ADH1B genotypes in the ALDH2*1/*1 and ALDH2*1/*2 groups and t-tests to assess mean differences between ADH1B*1/*2 and ADH1B*2/*2 in the ALDH2*2/*2 group. ADH1B genotype was not significantly associated with total symptoms or SRE_5 in the ALDH2*1/*1 or ALDH2*2/*2 groups, but was significantly associated in the ALDH2*1/*2 group (F = 3.89, 2/262 df, p = .022 for total symptoms; F = 5.55, 2/254 df, p = .004 for SRE_5). In the ALDH2*1/*2 group, individuals with two ADH1B*2 alleles had higher total symptom scores than individuals with only one ADH1B*2 allele (2.24 vs. 1.80, Tukey HSD p = .041), and had significantly lower SRE_5 scores than individuals with no ADH1B*2 alleles (2.05 vs. 3.15, Tukey HSD p = .004); the difference between those with one versus no ADH1B*2 alleles approached but did not reach significance for SRE_5 (2.32 vs. 5.15, Tukey HSD p =.050). Thus, only a few of
