Inter-observer agreement for Crohn's disease sub-phenotypes using the Montreal Classification: How good are we? A multi-centre Australasian study
Bibliographic record
Abstract
BACKGROUND: Crohn's disease (CD) exhibits significant clinical heterogeneity. Classification systems attempt to describe this; however, their utility and reliability depends on inter-observer agreement (IOA). We therefore sought to evaluate IOA using the Montreal Classification (MC). METHODS: De-identified clinical records of 35 CD patients from 6 Australian IBD centres were presented to 13 expert practitioners from 8 Australia and New Zealand Inflammatory Bowel Disease Consortium (ANZIBDC) centres. Practitioners classified the cases using MC and forwarded data for central blinded analysis. IOA on smoking and medications was also tested. Kappa statistics, with pre-specified outcomes of κ>0.8 excellent; 0.61-0.8 good; 0.41-0.6 moderate and ≤0.4 poor, were used. RESULTS: 97% of study cases had colonoscopy reports, however, only 31% had undergone a complete set of diagnostic investigations (colonoscopy, histology, SB imaging). At diagnosis, IOA was excellent for age, κ=0.84; good for disease location, κ=0.73; only moderate for upper GI disease (κ=0.57) and disease behaviour, κ=0.54; and good for the presence of perianal disease, κ=0.6. At last follow-up, IOA was good for location, κ=0.68; only moderate for upper GI disease (κ=0.43) and disease behaviour, κ=0.46; but excellent for the presence/absence of perianal disease, κ=0.88. IOA for immunosuppressant use ever and presence of stricture were both good (κ=0.79 and 0.64 respectively). CONCLUSION: IOA using MC is generally good; however some areas are less consistent than others. Omissions and inaccuracies reduce the value of clinical data when comparing cohorts across different centres, and may impair the ability to translate genetic discoveries into clinical practice.
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 | 0.039 | 0.066 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".