Secular stability and reliability of measurements of the percentage of dense tissue on mammograms

Abstract Elevated mammographic density is associated with increased risk of breast cancer. We conducted a reliability study on mammographic density assessments to determine their potential usefulness for projecting individual breast cancer risk. We used baseline screening mammograms from 7251 women in the Breast Cancer Detection Demonstration Project (BCDDP). Repeated measurements from the same images were used to assess measurement variability by an experienced evaluator. Intraclass correlations of assessments over time usually exceeded 0.9, indicating usefulness for prospective applications. Data also indicated it may be reasonable to include cases identified in the first year of screening together with other cases in developing a risk model. Older ages and increased weight were associated with decreased mammographic density. The density of the right breast slightly exceeded that of the left. Among women who developed breast cancer, the baseline mammographic density of the ipsilateral (diseased) breast was 0.53 (95% confidence interval (CI) 0.20–0.86) percentage units higher than in the contralateral breast.