Differences in colonoscopy performance among four endoscopy centers in Western Norway: Influence of case-mix

Unmodifiable patient factors such as age, sex, and indication (case-mix) may influence colonoscopy performance. In this study, we explored how case-mix affected polyp detection, cecal intubation, and pain on a center level.

A cross-sectional study was performed on data from four centers in Western Norway registered in the national endoscopy quality registry, Gastronet, in 2020 and 2021. We extracted demographics, indication, and the performance measures cecal intubation rate (CIR), proportion of at least one polyp ≥ 5 mm in size per colonoscopy (PDR-5), and pain. We also analyzed the explanatory variables bowel preparation, withdrawal time, and sedation/analgesia.

First colonoscopies in 14,765 patients were included. Median age was 60 years (interquartile range 46-71) and 54% were women. Case-mix differed between centers and significantly influenced performance measures. Increased PDR-5 was associated with higher age and male sex (odds ratio [OR] 1.27, 95% confidence interval [CI] 1.18-1.37). The indication surveillance had the highest PDR-5 (44.9%, 95% CI 42.6-47.1) and inflammatory bowel disease the lowest (14.6%, 95% CI 12.3-16.8). CIR decreased with increasing age. Men had less pain (OR 0.33, 95% CI 0.27-0.39). Among indications, surveillance and IBD had higher CIRs and less pain. Performance measures differed among centers, even after adjustment for case-mix and other known explanatory variables such as sedation/analgesia and bowel preparation.

Case-mix influenced performance measures. Although we showed center differences in performance, other factors, such as individual endoscopist skills, probably influence performance measures. Our study demonstrates the importance of considering case-mix when assessing colonoscopy performance.