Assessing Resident Performance in Screening Mammography: Development of a Quantitative Algorithm

Area of Science:

• Radiology and Medical Imaging
• Medical Education
• Health Outcomes Research

Background:

• Assessing resident performance in screening mammography interpretation is crucial for ensuring quality patient care.
• Existing methods may lack objectivity or fail to provide timely, individualized feedback.
• There is a need for a standardized system to evaluate resident competency in mammography interpretation.

Purpose of the Study:

• To establish an objective performance metric for residents interpreting screening mammograms.
• To provide residents with actionable, case-by-case feedback on examination volumes, recall rates, and concordance with faculty interpretations.
• To longitudinally track resident performance throughout their training.

Main Methods:

• A macro-driven spreadsheet system was developed to record resident and faculty interpretations of screening mammograms.
• Data collected included concordant and discordant interpretations, with outcomes weighted by perceived clinical impact.
• A weighted mammography performance score was generated, alongside rotation-specific goals for recall rate and concordance.

Main Results:

• Over 18,000 mammograms were reviewed by 31 residents across 71 rotations.
• Mean resident recall rate decreased with increasing resident level (R2 to R4).
• Mean resident-faculty discordance rate also significantly decreased from R2 to R4.

Conclusions:

• The developed method offers objective, individualized, and longitudinal performance data for residents interpreting screening mammograms.
• This system provides a valuable tool for assessing resident competency and improving the quality of mammography interpretation training.
• The data indicate a learning curve and improvement in resident performance with experience.