Compared to other ethnic groups Asians are more likely to be sensitive to alcohol, due to polymorphisms of alcohol-metabolizing enzymes. Although previous studies have found positive association between regular alcohol use and high-sensitivity C-reactive protein (HsCRP), whether this association is modified by alcohol sensitivity has not been clarified. We therefore sought to examined this potential effect modification in a cross-sectional community sample with high prevalence of alcohol sensitivity, using data from 2903 men aged >= 50years recruited during phase 1 of the Guangzhou Biobank Cohort Study. Information on alcohol consumption and sensitivity (facial flushing, palpitation or dizziness after drinking) was obtained by questionnaire and HsCRP was measured by an immunoturbidometric assay. Elevated HsCRP was defined as HsCRP level equal to or higher than 2.81 mg/L(median). Excessive alcohol use was defined as use of >= 210 g ethanol per week. After adjustment for age, educational level, occupation, smoking status, physical activity and history of cardiovascular disease, alcohol use was associated with HsCRP in a dose-response pattern. The risks of elevated HsCRP were higher in those who drank daily (odds ratio (OR) = 1.38 (1.10, 1.72)) or drank excessively (1.57 (1.22, 2.02)), and were even higher in alcohol users with alcohol sensitivity (1.82 (1.24, 2.65) for daily users and 2.34 (1.48, 3.71) for excessive users). Results of this study have showed an important role of alcohol sensitivity in modifying the association between alcohol use and HsCRP level. Reduction of alcohol use should be an important public heath target, particularly among populations with high prevalence of alcohol sensitivity. (C) 2016 Elsevier Inc. All rights reserved.