Multivariable models for advanced colorectal neoplasms in screen-eligible individuals at low-to-moderate risk of colorectal cancer: towards improving colonoscopy prioritization
Bibliographic record
Abstract
Abstract Background Advanced colorectal neoplasms (ACNs), including colorectal cancers (CRC) and high-risk adenomas (HRA), are detected in less than 20% of persons aged 50 years or older who undergo colonoscopy. We sought to derive personalized predictive models of risk of harbouring ACNs to improve colonoscopy wait times for high-risk patients and allocation of colonoscopy resources. Methods We characterized colonoscopy indications, neoplasia risk factors and colonoscopy findings through chart review for consecutive individuals aged 50 years or older who underwent outpatient colonoscopy at The Ottawa Hospital (Ottawa, Canada) between April 1, 2008 and March 31, 2012 for non-life threatening indications. We linked patients to population-level health administrative datasets to ascertain additional historical predictor variables and derive multivariable logistic regression models for risk of harboring ACNs at colonoscopy. We assessed model discriminatory capacity and calibration and the ability of the models to improve colonoscopy specificity while maintaining excellent sensitivity for ACN capture. Results We modelled 17 candidate predictors in 11,724 individuals who met eligibility criteria. The final CRC model comprised 8 variables and had a c-statistic value of 0.957 and a goodness-of-fit p-value of 0.527. Application of the models to our cohort permitted 100% sensitivity for identifying persons with CRC and > 90% sensitivity for identifying persons with HRA, while improving colonoscopy specificity for ACNs by 23.8%. Conclusions Our multivariable models show excellent discriminatory capacity for persons with ACNs and could significantly increase colonoscopy specificity without overly sacrificing sensitivity. If validated, these models could allow more efficient allocation of colonoscopy resources, potentially reducing wait times for those at higher risk while deferring unnecessary colonoscopies in low-risk individuals.
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 | 0.006 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".