BACKGROUND Dyslipidemia is a common complication in patients with diabetes and is involved in being prone to cardiovascular disease. The risk of coronary artery disease is known to be lower in light-to-moderate drinkers than in abstainers. The aim of this study was to clarify whether and how alcohol drinking influences the lipid-related indices, good predictors for cardiovascular disease, such as the ratio of LDL cholesterol to HDL cholesterol (LDL-C/HDL-C ratio), the ratio of triglycerides to HDL cholesterol (TG/HDL-C ratio), and the lipid accumulation product (LAP), in patients with diabetes. METHODS The subjects were men with diabetes (n = 1477; mean age, 54.0 years) and they were divided into non-, light (< 22 g ethanol/day), moderate (≥ 22 and < 44 g ethanol/day) and heavy (≥ 44 g ethanol/day) drinkers. The relationships between alcohol intake and the lipid-related indices were investigated by the multivariate analyses with adjustment for age, smoking, regular exercise and drug therapy for diabetes. RESULTS The odds ratio (OR) vs. nondrinkers for high LDL-C/HDL-C ratio tended to be lower with an increase in alcohol intake (OR with 95% confidence interval (CI): 0.80 [0.50-1.29] in light drinkers; 0.24 [0.15-0.38] in moderate drinkers and 0.10 [0.05-0.19] in heavy drinkers). Alcohol intake showed an inverse association with a high TG/HDL-C ratio (OR with 95% CI vs. nondrinkers for high TG/HDL-C ratio: 0.54 [0.36-0.80] in light drinkers; 0.73 [0.56-0.97] in moderate drinkers and 0.72 [0.53-0.98] in heavy drinkers) and a J-shaped relationship with a high LAP (OR with 95% CI vs. nondrinkers for high LAP: 0.66 [0.43-1.02] in light drinkers; 0.82 [0.61-1.10] in moderate drinkers, and 1.29 [0.95-1.77] in heavy drinkers). Similar associations between alcohol intake and the lipid indices were obtained in a covariance analysis. CONCLUSIONS In patients with diabetes, light-to-moderate alcohol consumption is inversely associated with lipid-related indices, and this correlates with previous findings of cardiovascular risk reduction by modest drinking in patients with diabetes.