P-026 Effect of Standardized Scoring Conventions on Inter-Rater Reliability in the Endoscopic Evaluation of Crohnʼs Disease
Bibliographic record
Abstract
The Crohn's Disease Endoscopic Index of Severity (CDEIS)1 and the Simple Endoscopic Score for Crohn's Disease (SES-CD)2 are commonly used in clinical trials to evaluate endoscopic disease activity. Scoring of these instruments by expert central readers (ECRs) has become standard practice in trials of new therapies in an attempt to increase trial efficiency by reducing bias and measurement variability.3 Both endoscopic indices (EIs) have demonstrated substantial inter- and intra-rater reliability when scored by ECRs, however some index items have low reliability. Previously identified sources of low reliability include differentiating superficial from deep ulceration, estimation of disease extent, scoring of aphthoid and anastomotic ulcers, defining ulcer location within segments, evaluating anal lesions, scoring ulcers beyond stenotic segments and consistently classifying the location of stenoses.4 We evaluated the application of scoring conventions to improve the reliability of these items. Five gastroenterologists at an academic hospital with no prior experience scoring the CDEIS or SES-CD were trained on the use of these indices. Sixty-five video recordings of colonoscopies were scored blindly by each gastroenterologist prior to and after additional training on index scoring conventions. Changes in intra-class correlation coefficients (ICCs) were used to quantify the effect of application of these conventions on the reliability of the CDEIS, SES-CD and a Global Evaluation of Lesion Severity (GELS) score. Following the application of scoring conventions, inter-rater ICCs (95% confidence interval [CI]) for scoring of the total SES-CD score increased from 0.78 (0.71–0.85) to 0.85 (0.79–0.89). Inter-rater ICCs for the total CDEIS and GELS scores were not affected: the corresponding inter-rater ICCs were 0.74 (0.65–0.81) and 0.49, (0.38–0.61) prior to, and 0.73 (0.65–0.81) and 0.53 (0.42–0.64) following application of scoring conventions. Estimation of ulcer depth, surface area, anatomic location and stenosis were again identified as important sources of variability. Use of scoring conventions developed by ECRs enhanced the reliability of the SES-CD but did not similarly affect the CDEIS or GELS. Our data demonstrate that the SES-CD is more reliable than the CDEIS and can be improved with targeted training. It is therefore the preferred instrument for use in clinical trials.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.233 | 0.432 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".