A case-control study was conducted to assess whether certain "high-risk" mammographic parenchymal patterns are associated with the increased occurrence of breast cancer in postmenopausal women. Patients in the case group included 105 women with histologically confirmed breast cancer; subjects in the control group included 104 women with fibrocystic breast disease and 103 women with clinically normal breasts. All mammographic results were evaluated "blindly" by a radiologist who classified the breast parenchyma into "high-risk" and "low-risk" categories according to the criteria proposed by Wolfe. These data showed a similar proportion of allegedly "high-risk" breast parenchymal patterns among patients in all three groups. The finding that breast parenchymal patterns are not associated with an increased risk of breast cancer in women 50 years of age or older is consistent with the results of earlier studies in which the association is present only in younger, premenopausal women, and is absent in older, predominantly postmenopausal women. Consequently, it is concluded that these parenchymal patterns should not be used to select postmenopausal women for breast cancer screening programs, or to guide the evaluation of postmenopausal women with breast lumps or symptoms.