Improving Endoscopic Scoring Accuracy in Pediatric Inflammatory Bowel Disease: Impact of a Focused Training...

Background High variability has been reported in assessing endoscopic scoring systems for pediatric inflammatory bowel disease (IBD). This study assessed whether a structured training could improve interobserver agreement when scoring IBD endoscopy findings. Methods Ten ileocolonoscopic videos were scored using the Mayo Endoscopic Score (MES) and Ulcerative Colitis Endoscopic Index of Severity (UCEIS) for ulcerative colitis, and the Simple Endoscopic Score for Crohn's Disease (SES-CD). Ten expert endoscopists and ten trainees evaluated the videos before and after a focused training session. Agreement was measured using intraclass correlation coefficient (ICC) and Fleiss kappa (κ), and perceived benefit was surveyed. Results For MES, κ slightly improved among experts, but ICC declined; trainees showed stable κ and improved ICC. The UCEIS agreement was stable in experts and slightly improved in trainees. The SES-CD scores showed improved agreement for both groups, with trainee accuracy improving from 74% to 86% (Δ = 12%, p = 0.051). Both groups reported subjective improvement, although not statistically significant. Conclusions Interobserver agreement ranged from moderate to good. Structured training produced only modest improvements, especially in SES-CD scoring, with limited overall effect on scoring consistency.