CT colonography: interpretative performance in a non-academic environment.

Area of Science:

• Radiology
• Gastroenterology
• Medical Imaging

Background:

• Computed tomography (CT) colonography is a widely used colorectal cancer screening tool.
• Assessing the interpretative accuracy and reporting time of radiologists in routine clinical practice is crucial for quality assurance.

Purpose of the Study:

• To evaluate the interpretative accuracy and reporting time of radiologists performing CT colonography in non-academic clinical settings.
• To compare performance metrics against experienced and recently trained radiologists in an academic environment.

Main Methods:

• Thirteen radiologists from seven non-academic centers interpreted 15 colonoscopically validated CT colonography cases.
• Performance (accuracy and interpretation time) was compared to aggregated data from experienced and recently trained academic radiologists.
• Spearman's rank correlation was used to assess the effect of experience on accuracy.

Main Results:

• Individual accuracy ranged from 53% to 93%, with a mean accuracy of 75%, significantly lower than experienced radiologists (88%) but not different from recently trained ones.
• Interpretation time was comparable to experienced radiologists (12.4 min vs. 11.7 min) but shorter than recently trained radiologists.
• A significant positive correlation was found between prior experience and accuracy, with no performance plateau observed.

Conclusions:

• On average, non-academic radiologists' CT colonography accuracy is similar to those recently trained but shows wide individual variability.
• Performance often falls below published meta-analysis averages, indicating a need for standardized competency assessment.
• While experience enhances accuracy, it is not solely sufficient to guarantee competence in CT colonography interpretation.