Perioperative medicine & outcomes
Multicenter Validation of a Machine Learning Model for Surgical Transfusion Risk at 45 US Hospitals
Main finding · Cases with type and screen recommendedResult — go to the resultMedian difference 17.9 absolute percentage pointsAuthors' conclusion
In this cohort study of 45 hospitals, a personalized surgical transfusion risk prediction algorithm demonstrated external validity and discrimination.
Study context
- Population
- All surgical cases in 2020 or 2021 at 45 US hospitals of the Multicenter Perioperative Outcomes Group; obstetric and nonoperative cases excludedsurgical cases, 45 hospitals
- Participants
- 3,275,956 cases
- Exposure
- S-PATH machine learning model vs standard-of-care maximum surgical blood ordering schedule (MSBOS) approach
- Outcome
- Difference in % of patients with type and screen order recommendations between S-PATH and MSBOS at each hospital (thresholds set for 96% sensitivity)
Methodological detailsmulticentre, external validation
- Multicentre
- External validation
Main finding
Cases with type and screen recommended
Median difference
17.9 absolute percentage points
By group
As reported in the abstract
Observational data: associations, not effects.
Secondary outcomes
Median
| Outcome | S-PATH | MSBOS |
|---|---|---|
| AUROC | 0.929 (IQR 0.915–0.946) | 0.857 (IQR 0.822–0.884) |
Original publication
Multicenter Validation of a Machine Learning Model for Surgical Transfusion Risk at 45 US Hospitals. JAMA Netw Open. Published 2 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →