Perioperative medicine & outcomes
Decision Curve Analysis of In-Hospital Mortality Prediction Models: The Relative Value of Pre- and Intraoperative Data For Decision-Making
Authors' conclusion
When it comes to predicting in-hospital mortality and subsequent decision-making, preoperative demographics, comorbidities, and surgery-related data provide the largest benefit for clinicians with risk-averse preferences, whereas preoperative laboratory values provide the largest benefit for decision-makers with more moderate risk preferences.
Study context
- Population
- Surgical patients in the VitalDB database subcohort141 predictors
- Participants
- 6,043 patients
- Exposure
- Preoperative vs Intraoperative vs All data
- Outcome
- Net benefit in decision curve analysis (DCA) of in-hospital mortality prediction models (treatment threshold probabilities 0%–20%)
Methodological detailsretrospective database (VitalDB), multiple machine-learning methods, stacked learner
- Retrospective database (VitalDB)
- Multiple machine-learning methods
- Stacked learner
Main finding
As reported in the abstract
Net benefit (DCA)
- Preoperative data provide the largest overall benefit for decision-making; intraoperative values provide only limited benefit compared to preoperative data
- Preoperative demographics, comorbidities, and surgery-related data provide the largest benefit for low treatment thresholds up to 5% to 10%; preoperative laboratory measurements dominant for higher thresholds
Observational data: associations, not effects.
Further findings
Predictive performance (AUROC, and under the precision-recall curve (AUPRC))
- AUROC range
- 0.53 to 0.78
- AUPRC range
- 0.02 to 0.25
- Incidence of in-hospital mortality
- 0.09
- High AUPRC values from models based on preoperative laboratory values
Original publication
Decision Curve Analysis of In-Hospital Mortality Prediction Models: The Relative Value of Pre- and Intraoperative Data For Decision-Making. Anesth Analg. Published 5 February 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →