Related Experiment Video
Updated: Apr 27, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
42.6K
Identification of abnormal screening mammogram interpretation using Medicare claims data.
Rebecca A Hubbard1, Weiwei Zhu, Steven Balch
1Group Health Research Institute and Department of Biostatistics, University of Washington, 1730 Minor Ave, Suite 1600, Seattle, WA, 98101.
Health Services Research
|July 1, 2014
Summary
Medicare claims can identify abnormal screening mammograms with high accuracy. This approach using claims for follow-up imaging or biopsy is feasible for research and quality improvement.
Area of Science:
- Medical Informatics
- Radiology
- Public Health
Background:
- Accurate identification of abnormal screening mammography interpretations is crucial for early breast cancer detection.
- Existing methods for identifying abnormal results may have limitations in large-scale research or quality assessment.
Purpose of the Study:
- To develop and validate Medicare claims-based methods for identifying abnormal screening mammography interpretations.
- To assess the feasibility of using claims data for research and quality improvement in mammography.
Main Methods:
- Utilized linked mammography data and Medicare claims from 387,709 screening mammograms (1999-2005) within the Breast Cancer Surveillance Consortium (BCSC).
- Employed split-sample validation of algorithms based on claims for subsequent breast imaging or biopsy.
- Pooled Medicare claims and BCSC mammography data at a central Statistical Coordinating Center.
Main Results:
- Algorithms using claims for subsequent imaging or biopsy demonstrated a sensitivity of 74.9% and specificity of 99.4%.
- A classification and regression tree model improved sensitivity to 82.5% while maintaining high specificity (96.6%).
Conclusions:
- Medicare claims represent a feasible data source for identifying abnormal screening mammography interpretations.
- These claims-based approaches can support research and quality improvement initiatives focused on high rates of abnormal mammograms.
