Simplified Machine Learning Models Can Accurately Identify High-Need High-Cost Patients With Inflammatory Bowel Disease
Bibliographic record
Abstract
INTRODUCTION: Hospitalization is the primary driver of inflammatory bowel disease (IBD)-related healthcare costs and morbidity. Traditional prediction models have poor performance at identifying patients at highest risk of unplanned healthcare utilization. Identification of patients who are high-need and high-cost (HNHC) could reduce unplanned healthcare utilization and healthcare costs. METHODS: We conducted a retrospective cohort study in adult patients hospitalized with IBD using the Nationwide Readmissions Database (model derivation in the 2013 Nationwide Readmission Database and validation in the 2017 Nationwide Readmission Database). We built 2 tree-based algorithms (decision tree classifier and decision tree using gradient boosting framework [XGBoost]) and compared traditional logistic regression to identify patients at risk for becoming HNHC (patients in the highest decile of total days spent in hospital in a calendar year). RESULTS: Of 47,402 adult patients hospitalized with IBD, we identified 4,717 HNHC patients. The decision tree classifier model (length of stay, Charlson Comorbidity Index, procedure, Frailty Risk Score, and age) had a mean area under the receiver operating characteristic curve (AUC) of 0.78 ± 0.01 in the derivation data set and 0.78 ± 0.02 in the validation data set. XGBoost (length of stay, procedure, chronic pain, drug abuse, and diabetic complication) had a mean AUC of 0.79 ± 0.01 and 0.75 ± 0.02 in the derivation and validation data sets, respectively, compared with AUC 0.55 ± 0.01 and 0.56 ± 0.01 with traditional logistic regression (peptic ulcer disease, paresthesia, admission for osteomyelitis, renal failure, and lymphoma) in derivation and validation data sets, respectively. DISCUSSION: In hospitalized patients with IBD, simplified tree-based machine learning algorithms using administrative claims data can accurately predict patients at risk of progressing to HNHC.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| 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 | 0.000 |
| 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".