Performance of a Statistical Model to Predict Stroke Outcome in the Context of a Large, Simple, Randomized, Controlled Trial of Feeding
Bibliographic record
Abstract
BACKGROUND AND PURPOSE: Statistical models to predict the outcome of stroke patients have several uses. Their utility depends on their predictive accuracy in patients other than those on whom they were developed (ie, external validity). We sought to test the external validity of some recently described models in patients enrolled in the FOOD (Feed Or Ordinary Diet) trial: a large randomized trial evaluating feeding policies in patients with stroke. METHODS: The predictive variables were collected during a telephone call to randomize the patient a median of 5 days after stroke onset. Patients were followed up 6 months later to establish their survival, functional status, and residence. Charts were plotted to demonstrate the discrimination and calibration of the models. RESULTS: The models performed well in the first 2955 patients enrolled and followed up in the FOOD trial. The area under the receiver operating characteristic curves varied between 0.78 and 0.81 (with 0.5 indicating no discrimination and 1.0 indicating perfect discrimination). The discrimination was marginally better for patients enrolled within the first day of stroke than later. The models tended to provide rather pessimistic predictions in all groups except those predicted to have a high likelihood of surviving free of dependency. CONCLUSIONS: As one might predict, the discriminatory power in the selected cohort of trial patients was marginally less good than in previously studied unselected cohorts used to test their external validity. These models provide a well-tested tool for stratification in trials, comparing outcomes in different cohorts and examining the additional predictive power of novel factors.
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 | 0.167 | 0.208 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.004 | 0.005 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| 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".