Alcohol flushing after light drinking is triggered mainly by severe acetaldehydemia in individuals possessing inactive aldehyde dehydrogenase (ALDH)-2. Inactive ALDH2 encoded by ALDH2*1/2*2 and the low-activity form of alcohol dehydrogenase (ADH)-2 encoded by ADH2*1/2*1 enhance the risk for esophageal cancer in Japanese light to heavy drinkers, a significant association that emphasizes the importance of screening tests for inactive ALDH2 based on alcohol flushing. The objectives of the present report were (a). to evaluate the reliability of a simple questionnaire that asks about both current and past flushing for detecting inactive ALDH2 and (b). to predict cancer risk based on flushing in a case-control manner. The study subjects consisted of 233 Japanese men with esophageal squamous cell carcinoma and 610 cancer-free Japanese men. When current or former flushing individuals were considered to have inactive ALDH2, the sensitivity and specificity of the test were 84.8% and 82.3%, respectively, for the cases and 90.1% and 88.0%, respectively, for the controls. To clarify the characteristics of men who had genetically inactive ALDH2 but did not report alcohol flushing, we analyzed individuals possessing the ALDH2*1/2*2 genotype and found that those who also had ADH2*1/2*1 (both cases and controls) tended not to report current flushing, and those who did not report current flushing (controls only) tended to be heavier drinkers. As compared with overall never or rare drinking, the cancer risks for light (1-8.9 units/week; 1 unit = 22 g of ethanol), moderate (9-17.9 units/week), and heavy (18+ units/week) drinkers with current or former flushing (odds ratio = 6.69, 42.66, and 72.86, respectively) significantly exceeded the risks for those who had never flushed (odds ratio = 1.27, 10.12, and 15.61, respectively), even after adjustment for age, smoking, and diet. The flushing questionnaire may be used in large-scale epidemiological studies as a surrogate marker of ALDH2 genotype to predict individual cancer risk.