Background: To elucidate whether dietary habits affect knee osteoarthritis, we analyzed data from past regional screening programs for arthritis and osteoporosis. Methods: Subjects comprised 493 Japanese individuals (120 men, 373 women; age range, 40-83 years) who participated in a regional screening program for arthritis and osteoporosis in a single town in southwestern Japan. Outcome measure was radiographically determined knee osteoarthritis, categorized according to the criteria of Kellgren and Lawrence as grade ≥ 2, 1 or 0. Associations between dietary habits and knee osteoarthritis were assessed by calculating odds ratios and 95% confidence intervals using a proportional odds model in logistic regression. Results: Odds ratio for knee osteoarthritis was decreased with daily coffee consumption at 40 years of age (odds ratio=0.48, 95% confidence interval=0.28-0.83). Female sex (odds ratio=2.19, 95% confidence interval=1.15-4.15), age (3rd vs. 1st tertile: odds ratio=14.9, 95% confidence interval=8.07-27.5), weight at 40 years of age (3rd vs. 1st tertile: odds ratio=2.50, 95% confidence interval=1.41-4.43), previous knee injury (odds ratio=2.01, 95% confidence interval=1.24-3.24), and physical worker (odds ratio=2.09, 95% confidence interval=1.41-3.11) were significantly associated with knee osteoarthritis. Conclusions: Daily coffee consumption at 40 years of age was inversely associated with severe radiographic knee osteoarthritis.