consumption, or observer rated); such differences likely result in different endorsement rates of symptoms that make it difficult to directly compare across studies. It is also important to note that retrospective self-reports may be biased in ways that differ across individuals with varying ALDH2 genotypes, drinking histories, or family history of alcohol problems. The consistency of the findings for low doses of alcohol and for initial drinking episodes, however, as well as the trends in the data across ADH1B-ALDH2 genotype combinations, suggest these findings may be robust. Regardless of whether the genetic findings are due to actual differences in low-dose and initial responses to alcohol or to biased reports, we can conclude that individuals with ALDH2*2 and ADH1B*2 remember experiencing stronger responses to alcohol in these situations, which may affect their alcohol expectancies and in turn alter their alcohol consumption behaviors. The retrospective self-reports used in the present study, however, preclude testing of this possibility and of the causal pathways proposed by the mechanistic model, which would require prospective data.
