A population-based, case-control study was conducted to examine whether infrequent breast self-examination (BSE) is associated with the detection of breast cancer at a later stage of disease. A group of 435 women with newly diagnosed, regional or distant breast carcinoma from Connecticut were interviewed and compared with 887 age-frequency-matched women without breast cancer. A case-control method eliminates lead time bias that may have been present in prior studies that evaluated BSE efficacy. Cases were found to have practiced monthly BSE more frequently than controls (27 vs. 21%). No differences in frequency of BSE was found between cases and controls after controlling for the potentially confounding effects of breast cancer risk factors and mammography (OR = 1.26; 95% CI, 0.77 to 2.07).