Background: The relationships among breast density, age, and use of hormone replacement therapy (HRT) in breast cancer detection have not been fully evaluated.

Objective: To determine how breast density, age, and use of HRT individually and in combination affect the accuracy of screening mammography.

Design: Prospective cohort study.

Setting: 7 population-based mammography registries in North Carolina; New Mexico; New Hampshire; Vermont; Colorado; Seattle, Washington; and San Francisco, California.

Participants: 329 495 women 40 to 89 years of age who had 463 372 screening mammograms from 1996 to 1998; 2223 women received a diagnosis of breast cancer.

Measurements: Breast density, age, HRT use, rate of breast cancer occurrence, and sensitivity and specificity of screening mammography.

Results: Adjusted sensitivity ranged from 62.9% in women with extremely dense breasts to 87.0% in women with almost entirely fatty breasts; adjusted sensitivity increased with age from 68.6% in women 40 to 44 years of age to 83.3% in women 80 to 89 years of age. Adjusted specificity increased from 89.1% in women with extremely dense breasts to 96.9% in women with almost entirely fatty breasts. In women who did not use HRT, adjusted specificity increased from 91.4% in women 40 to 44 years of age to 94.4% in women 80 to 89 years of age. In women who used HRT, adjusted specificity was about 91.7% for all ages.

Conclusions: Mammographic breast density and age are important predictors of the accuracy of screening mammography. Although HRT use is not an independent predictor of accuracy, it probably affects accuracy by increasing breast density.