Breast imaging reporting and data system: Inter- and intraobserver variability in feature analysis and final assessment

OBJECTIVE, We sought to evaluate the use of the Breast Imaging Reporting an d Data System (BI-RADS) standardized mammography lexicon among and within o bservers and to distinguish variability in feature analysis from variabilit y in lesion management. MATERIALS AND METHODS. Five experienced mammographers, not specifically tra ined in BI-RADS, used the lexicon to describe and assess 103 screening mamm ograms, including 30 (29%) showing cancer, and a subset of 86 mammograms wi th diagnostic evaluation, including 23 (27%) showing cancer. A subset of 13 screening mammograms (two with malignant findings, 11 with diagnostic eval uation) were rereviewed by each observer 2 months later. Kappa statistics w ere calculated as measures of agreement beyond chance. RESULTS, After diagnostic evaluation, the interobserver kappa values for de scribing features were as follows: boast density, 0.43; lesion type, 0.75; mass borders, 0.40; special cases, 0.56; mass density, 0.40; mass shape, 0. 28; microcalcification morphology, 0.36; and microcalcification distributio n, 0.47. Lesion management was highly variable, with a kappa value for fina l assessment of 0.37, When we grouped assessments recommending immediate ad ditional evaluation and biopsy (BI-RADS categories 0, 4, and 5 combined) ve rsus follow-up (categories 1, 2, and 3 combined), five observers agreed on management for only 47 (55%) of 86 lesions. Intraobserver agreement on mana gement (additional evaluation or biopsy versus follow-up) was seen in 47 (8 5%) of 55 interpretations, with a kappa value of 0.35-1.0 (mean, 0.60) for final assessment. CONCLUSION. Inter- and intraobserver variability in mammographic interpreta tion is substantial for both feature analysis and management. Continued dev elopment of methods to improve standardization in mammographic interpretati on is needed.