Previous studies have demonstrated an association between current hormone replacement therapy (HRT) use and increased mammographic breast density. Many of these studies have also shown that only 20-35% of women initiating HRT respond in this manner. This subgroup of HRT responders may be at an increased risk of breast cancer. We performed a case-control study to investigate how women who experience increased density in response to HRT (cases) differ from women who do not experience an increase in density with HRT use (controls) with regard to breast cancer risk factors, type of HRT, weight change, and baseline breast density. Participants were female residents of Olmsted County, Minnesota who received routine screening mammograms at the Mayo Clinic. Cases included 172 women identified between the years 1998 and 1999 by Mayo radiologists as having a HRT response. Controls were women who did not experience an increase in mammographic density with HRT use and were matched to cases on age (+/-3 years), menopausal status, duration of HRT, month of initiation of HRT, and months between baseline and follow-up mammograms. Mammograms were obtained from cases and controls before and during HRT therapy. Breast density was read as a four-category Bi-Rads density grade measure and as a quantitative percentage estimate, using a computer-assisted method. Risk factor information was obtained from both chart review and a mammography database of patient-provided information. There was no association between HRT response and first-degree family history of breast cancer [odds ratio (OR), 0.8; 95% confidence interval (CI), 0.4-1.5], parity (OR, 0.8; 95% CI, 0.4-1.7), later age at first birth (OR, 0.8 for age >25 years versus nulliparous women; 95% CI, 0.4-1.8), or history of biopsy (OR, 0.9; 95% CI, 0.6-1.5). There was also no association with baseline weight or change in weight between a woman's baseline and follow-up mammograms. However, there was evidence of an association between HRT response and type of HRT used; women who experienced a mammographic increase in density with HRT had 2.3 greater odds (95% CI, 1.4-3.7) of having taken estrogen-progestin combined therapy than estrogen alone, compared with controls. This association was stronger among women with a baseline weight below the median (OR, 5.2; 95% CI, 1.6-17.6). Also, there was an inverse association between HRT response and baseline density. Because all risk factors examined accounted for only 26% of the variation in the HRT response, genes or other unmeasured factors are thought to be involved.