Key result
Dynamic daily risk predictor beats admission-day models for estimating 24-hour mortality with ~0.90 AUC.
Why the study?
Does a dynamic mortality risk predictor improve 24-hour mortality prediction compared to a static admission-day predictor in pediatric ICU patients?
Population
2,628 pediatric ICU patients in 9 tertiary care centers.
Comparison
Dynamic mortality risk predictor using daily… vs Static admission-day mortality predictor
Design
Cohort
Follow-up
daily during ICU stay
Authors
Loading...
May aid real-time risk stratification in pediatric ICU; hypothesis-generating pending prospective validation and outcome trials.
Observational (n=2,628)
Yes
Does a dynamic mortality risk predictor improve 24-hour mortality prediction compared to a static admission-day predictor in pediatric ICU patients?
Effect estimate: Az 0.904
p-value: p=<0.01
A dynamic mortality risk predictor using daily physiologic data accurately estimates 24-hour mortality risk in pediatric ICU patients, outperforming static admission-day predictors.
Ruttimann et al. (1991) conducted an observational in Pediatric intensive care unit patients (n=2,628). Dynamic mortality risk predictor vs. Admission-day predictor was evaluated on Accuracy of predicting 24-hour mortality risk (Area under the ROC curve, Az) (Az 0.904, p=<0.01). A dynamic mortality risk predictor based on daily physiologic data accurately estimated 24-hour mortality (Az = 0.904) and significantly improved prediction over an admission-day model (p<0.01).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: