BACKGROUND: Several studies have suggested that smoking may cause insulin resistance. However, the association between smoking and insulin resistance is still controversial. The purpose of this study was to investigate the association between smoking status and insulin resistance in Korean nondiabetic male population. METHODS: A total of 5,969 men, aged > 20 years were recruited from those who visited the Health Promotion Center, Samsung Medical Center between 2005 and 2006. All subjects were divided into three categories: on-smokers (n = 2,594), ex-smokers (n = 1,580), and current-smokers (n = 1,795). Fasting values for glucose and insulin were used to estimate insulin resistance by HOMA (homeostasis model assessment). An independent association between smoking status and HOMA-IR (homeostasis model assessment of insulin resistance) was assessed after adjustment for factors influencing insulin sensitivity such as age, exercise, alcohol, body mass index, abdominal circumference, and blood pressure. RESULTS: HOMA-IR was signifi cantly higher in ex-smokers and current-smokers than in non-smokers (2.09 +/- 0.94 vs. 2.04 +/- 0.90, 1.96 +/- 0.86, P < 0.01, respectively). After adjustment for factors potentially affecting insulin sensitivity (age, alcohol intake, exercise, BMI, body fat, abdominal circumference, and blood pressure), the association persisted for ex-smokers (P = 0.008). And HOMA-IR was signifi cantly higher in > or = 40 pack-years smokers than in non-smokers CONCLUSION: Based on HOMA-IR, previous-smoking and chronic smoking were significantly associated with insulin resistance in apparently healthy Korean nondiabetic men.