This retrospective cohort compared the National Early Warning Score (NEWS) with simplified NEWS, a demographic score and a gradient-boosting model. It covered 2·08 million adult encounters at 15 Danish hospitals, 2018–2023, with 24 h mortality as the outcome. Treatment within 24 h, such as ICU admission or an operation, can avert predicted deaths and bias validation. This intervention bias was handled by setting treated encounters aside and weighting the rest by propensity scores. Overall, the model added at most 1·9 correct identifications per 10 000 over NEWS, and NEWS 1·5 over simplified NEWS. The design is retrospective, and unmeasured factors behind treatment decisions may bias the weighted estimates.
Why it is interesting: Most validations report discrimination, but this one counts extra deaths correctly flagged. The gain over NEWS was under 2 per 10 000 overall, but 78·1 when initial NEWS was 7 or more.
Cite this item
BibTeX
Free sign-up. A digest of the new papers twice a week, grouped by domain.
Other drivers in Data, models and systems
-
Forecasts and their record
How the forecasts are made, how well they explained the past, and how well mortality can be forecast — and by whom.
3 this week -
What linked data reveals
What linked registries make it possible to know about whole populations
8 this week -
AI in medicine
Whether machine learning predicts, detects and decides better than the tools we have, and whether that has yet changed an outcome for a population.
4 this week -
Health systems and access
What the health system can provide and who gets it.
3 this week -
Value in health
What a treatment, health programme or policy is worth against what it costs, and how far the models that decide it can be trusted.
3 this week -
Revisions and new sources
What the figures are built from, what is newly available, and how much the numbers move when they are revised.
2 this week
