Aims and background The aim of the present study was to test some hypothetical risk factors for anaplastic thyroid cancer. Study design A case-control study comprised 110 patients with histologically or cytologically verified anaplastic thyroid cancer and 110 hospital controls individually matched with cases by sex, age, place of residence and time of hospitalization. Univariate and multivariate logistic regressions were used for data analysis. Results According to the conditional logistic regression analysis, the following factors were significantly and independently related to anaplastic thyroid cancer: history of goiter or thyroid nodules (odds ratio, OR, 37.55; 95% confidence interval, CI, 4.86–290.11), history of residence in endemic goiter area (OR, 2.56; 95% CI, 1.05–6.22), history of previous malignant non-thyroid tumor (OR, 5.51; 95% CI, 1.04–29.25), diabetes mellitus (OR, 4.06; 95% CI, 1.29–12.81) and low educational level (OR, 2.44; 95% CI, 1.17–5.06). Conclusions The results are in line with the current knowledge on factors related to thyroid cancer, except for the association between thyroid cancer and diabetes mellitus, which as far as we know has not yet been reported.