Background: To improve colonoscopy quality, reports must include key quality indicators that can be monitored.
Objective: To determine the quality of colonoscopy reports in diverse practice settings.
Setting: The consortium of the Clinical Outcomes Research Initiative, which includes 73 U.S. gastroenterology practice sites that use a structured computerized endoscopy report generator, which includes fields for specific quality indicators.
Design: Prospective data collection from 2004 to 2006.
Main outcomes measurements: Reports were queried to determine if specific quality indicators were recorded. Specific end points, including quality of bowel preparation, cecal intubation rate, and detection of polyp(s) >9 mm in screening examinations were compared for 53 practices with more than 100 colonoscopy procedures per year.
Results: Of the 438,521 reports received during the study period, 13.9% did not include bowel-preparation quality and 10.1% did not include comorbidity classification. The overall cecal intubation rate was 96.3%, but cecal landmarks were not recorded in 14% of the reports. Missing polyp descriptors included polyp size (4.9%) and morphology (14.7%). Reporting interventions for adverse events during the procedure varied from 0% to 6.5%. Among average-risk patients who received screening examinations, the detection rate of polyps >9 mm, adjusted for age, sex, and race, was between 4% and 10% in 81% of practices.
Limitation: Bias toward high rates of reporting because of the standard use of a computerized report generator.
Conclusions: There is significant variation in the quality of colonoscopy reports across diverse practices, despite the use of a computerized report generator. Measurement of quality indicators in clinical practice can identify areas for quality improvement.