Development and validation of a machine learning ASA-score to identify candidates for comprehensive preoperative screening and risk stratification
Authors' conclusion
We created and validated a machine learning physical status based on preoperatively available data.
Study context
- Population
- Patients who received anesthesia at Beth Israel Deaconess Medical Center (training and internal validation) and Montefiore Medical Center (external validation)training 361,602
- Exposure
- Machine learning-derived ASA-PS vs anesthesiologist ASA-PS
- Outcome
- Agreement between anesthesiologist ASA-PS and ML-PS
Methodological detailsmulticentre, external validation
- Multicentre
- External validation
Main finding
As reported in the abstract
Agreement between anesthesiologist ASA-PS and ML-PS (moderate inter-rater agreement)
57.2%
Observational data: associations, not effects.
Further findings
Net reclassification among patients who died within 30 days
- Reclassified into higher clinical risk category with ML-PS vs anesthesiologist rating, of 3594 deaths
- 1281 (35.6%)
Predictive accuracy in multiple co-morbidity patients (subgroup)
- Anesthesiologist ASA-PS had better predictive accuracy than ML-PS
Original publication
Development and validation of a machine learning ASA-score to identify candidates for comprehensive preoperative screening and risk stratification. J Clin Anesth. Published 8 March 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →