Purpose: Several studies have reported an inverse association between moderate alcohol consumption and prevalence of fatty liver (FL) in men. We aimed to clarify this association in women. Methods: We collected health checkup data from 4,921 Japanese women without concurrent liver disease (mean age 46.4 years) and performed a cross-sectional study to evaluate the influence of alcohol drinking patterns (frequency and amount) on the prevalence of FL as assessed by ultrasonography. Results: Alcohol consumption was reported in 30.8 % of participants, and FL was observed in 13.8 % (15.5 % nondrinkers, 10.1 % drinkers). Alcohol consumption was inversely associated with FL prevalence [adjusted odds ratio (AOR) 0.79, 95 % confidence interval (CI) 0.63-0.98]. In analyses stratified by drinking frequency and/or amount of alcohol consumed, the risk of FL decreased for the following categories: 0.1-19.9 g/drinking day (AOR 0.61, 95 % CI 0.44-0.83) and 0.1-69.9 g/week (AOR 0.74, 95 % CI 0.55-0.98). The amount of alcohol consumed directly correlated with the prevalence of FL in daily drinkers (p &lt
0.05), whereas there was no correlation between the frequency of alcohol consumption and FL prevalence. Alanine aminotransferase levels were significantly lower for the following categories: 0.1-19.9 g/drinking day for 1-3 days a week (p = 0.016) and 0.1-69.9 g within 1-3 drinking days a week (p = 0.004). Conclusions: Minimal alcohol consumption appears to have protective effects against nonalcoholic FL disease in women, although an increase in the amount of alcohol consumed appears to nullify the protective effect. © 2013 Asian Pacific Association for the Study of the Liver.