Investigator-observed alcohol-induced flushing but not self-report of flushing is a valid predictor of ALDH2 genotype.

OBJECTIVE: This article presents data on the validity of using self-report of alcohol-induced flushing and actual investigator-observed flushing following alcohol challenge to predict ALDH2 genotype in Asian-American men.
METHOD: Men between the ages of 21 and 25 years who were of Chinese, Japanese or Korean descent completed questionnaires about their drinking history and their alcohol-induced flushing history and associated symptoms. Fifty men selected for participation in the study were genotyped for alleles of ALDH2 and individually tested on two separate occasions following oral administration of placebo and 0.75 ml/kg (0.56 g/kg) alcohol. Facial flushing was assessed at baseline and at intervals over a 150-minute period after drinking using observational ratings.
RESULTS: By comparing the results of ALDH2 genotype with investigator-observed flushing and with previous self-report of facial flushing, it was found that investigator-observed flushing is both a sensitive (100%) and specific (96%) predictor of ALDH2 genotype, whereas self-report of facial flushing is a sensitive (100%) but not a specific (68%) predictor of ALDH2 genotype.
CONCLUSIONS: The results suggest that investigator-observed flushing provides a valid estimate of ALDH2 genotype, but that self-report of facial flushing is biased, giving a substantial overestimate. Due to an increase in error variance, studies that rely solely on self-report of flushing will more often lead to the conclusion that no association exists. This study supports the importance of using ALDH2 genotype, rather than self-report of flushing or ALDH2 phenotype, when examining factors associated with differences in drinking behavior, response to alcohol and risk for alcoholism or alcohol-related disease among Asians.