Next Generation Quality: Assessing the Physician in Clinical History Completeness and Diagnostic Interpretations...

Area of Science:

• Anatomical Pathology
• Health Informatics
• Quality Improvement Science

Background:

• Observational data and funnel plots are standard tools for performance analysis in various fields.
• Their application in pathology can enhance trend understanding and identify areas for improvement.

Purpose of the Study:

• To extract diagnostic rate (DR) data from free-text surgical pathology reports.
• To evaluate the potential for assessing inter-rater variation and clinical history completeness for continuous quality improvement (CQI).

Main Methods:

• Utilized string matching, fuzzy string matching, and hierarchical pruning to categorize 3,854 prostate biopsies.
• Stratified diagnostic rates by submitting physicians and pathologists.
• Employed funnel plots to detect diagnostic bias.

Main Results:

• Achieved a diagnostic classification for all biopsies with a categorization error rate below 1%.
• Observed significant inter-rater variability among 27 pathologists, with older pathologists showing a trend towards diagnosing more WHO/ISUP Grade 1 cancers.
• Demonstrated that normalized deviations plots can highlight individual pathologist over- and under-calls.
• Found substantial variation in clinical history completeness (22% to 100%).

Conclusions:

• Free-text data analysis in anatomical pathology, despite limitations, offers a viable approach for data-driven CQI.
• This methodology has the potential to advance the quality of care in pathology.