Interobserver agreement for the assessment of erosive reflux esophagitis: a <i>post hoc</i> analysis of clinical trial data
Bibliographic record
Abstract
Interobserver agreement for the Los Angeles (LA) classification of erosive reflux esophagitis was good in validation studies, but limited agreement data exists from clinical trials (CTs). We conducted a post hoc evaluation of interobserver agreement between CT endoscopists and independent expert adjudicators in a multi-center, randomized controlled trial of a new acid suppression therapy. Trial endoscopists captured endoscopic images/videos and documented esophagitis severity using the LA classification. Adjudicators reviewed images/videos on a web-based platform. If the first two adjudicators disagreed and the third adjudicator did not produce a majority verdict, all three conferred to reach consensus. Cohen's kappa (κ) evaluated interobserver agreement. Cohen's weighted kappa (κw) evaluated agreement corrected for disagreement extent. Of 388 images/videos with adequate quality, trial endoscopists and adjudicators agreed on esophagitis severity in 168 (43.3%) cases, and assigned more severe grades than adjudicators for 185 (47.7%) cases. Agreement was fair between trial endoscopists and adjudicators (κ: 0.27; κw: 0.40), moderate between individual adjudicators (κ: 0.43 to 0.47), and good between adjudicators and final diagnosis (κ: 0.75 to 0.78). After adjusting for disagreement extent, agreement was good between individual adjudicators (κw: 0.63 to 0.66), and very good between adjudicators and final diagnosis (κw: 0.84 to 0.87). Interobserver agreement on esophagitis severity between CT endoscopists and adjudicators was fair. Initial agreement between adjudicators was moderate, but agreement between adjudicators and consensus diagnosis was very good. Accurate esophagitis grading for CTs requires further training on LA classification and a robust central reading protocol.
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 | 0.281 | 0.305 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.009 | 0.014 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.002 | 0.005 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.005 | 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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".