A population-based case-control study was conducted in two regions of Sweden and Norway to investigate the association between dietary habits and the risk of thyroid cancer. The consumption of selected foods was reported in a self-completed food-frequency questionnaire by 246 cases with histologically confirmed papillary (n = 209) and follicular (n = 37) thyroid carcinoma, and 440 age- and gender-matched controls. Odds ratios (OR) and their 95 percent confidence interval (CI) were calculated as estimates of the relative risk using conditional logistic regression. High consumption of butter (OR = 1.6, CI = 1.1-2.5) and cheese (OR = 1.5, CI = 1.0-2.4) was associated with increased risks. Residence in areas of endemic goiter in Sweden was associated with an elevated risk, especially among women (OR = 2.5, CI = 1.3-4.9). High consumption of cruciferous vegetables was associated with increased risk only in persons who ever lived in such areas. A decreased risk was associated with consumption of iodized salt in northern Norway, and with use of iodized salt during adolescence among women (OR = 0.6, CI = 0.6-1.0). The results of this study suggest a role of diet and environment in the risk of thyroid cancer.