Performance of Clinical Risk Prediction Models for Post-ERCP Pancreatitis: A Systematic Review
Bibliographic record
Abstract
OBJECTIVES: Pancreatitis is common following endoscopic retrograde cholangiopancreatography (ERCP). Despite increased vigilance of post-ERCP pancreatitis (PEP), both its incidence and associated mortality are rising. Risk prediction models may provide more accurate stratification of patient risk and proactive mitigation of PEP incidence and/or severe associated outcomes. METHODS: We conducted an electronic search of MEDLINE, PubMEd, Cochrane, and CINAHL from inception through April 9, 2024 for studies evaluating the details and performances of available PEP prediction models. Studies were eligible if they used statistical measures to quantify their model's predictive ability. Risk of bias was determined using the PROBAST tool. RESULTS: Nineteen studies met eligibility criteria and were included. Logistic regression models were used in 15 studies, with machine learning models representing the second most commonly used approach. Ten studies reported the performance of their risk prediction models using derivation data, with areas under the receiver operating curve (AUC) ranging from 0.68 to 0.86. Fifteen studies reported the performance of their risk prediction models on internally validated data, with AUCs ranging from 0.66 to 0.97. Eight studies reported on the performance of their risk prediction models on external validation data, with AUCs ranging from 0.67 to 0.98. DISCUSSION: Numerous PEP clinical prediction models exist with variable performances. The use of PEP prediction tools can support the management of patients following ERCP. Implementation studies assessing the optimal usability of these tools, followed by prospective evaluations, are needed to evaluate their potential impacts on reducing PEP in real-world 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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.008 | 0.002 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".