Radiologist errors by modality, anatomic region, and pathology for 1.6 million exams: what we have learned.

Area of Science:

• Radiology
• Medical Quality Improvement
• Pathology

Background:

• Radiologist error characterization is crucial for quality assurance.
• Current schemes often lack detailed pathology information.
• This limits targeted peer review and learning.

Purpose of the Study:

• To assess the feasibility of incorporating pathology into radiologist error analysis.
• To determine if pathology data can refine peer review and learning.
• To enhance the specificity of quality improvement initiatives.

Main Methods:

• Collected quality assurance data from 349 radiologists over one year.
• Categorized errors as major, minor, or administrative.
• Classified interpretive errors by modality, region, and 64 pathology types.

Main Results:

• A discrepancy rate of 0.5% (8181/1,634,201 studies) was observed.
• Major interpretive errors were most common in CT (75.5%), particularly abdomen/pelvis (40.4%).
• Mass/nodule/adenopathy was the most frequent discrepancy overall; fracture/dislocation and bleed were most common in major errors.

Conclusions:

• Integrating error-related pathology into peer review is feasible and practical.
• This approach provides a detailed guide for targeted quality improvement.
• Further research is needed to confirm measurable improvements post-intervention.