Perioperative medicine & outcomes
Development and Validation of an Electronic Health Record-based Score for Triage to Perioperative Medicine
Authors' conclusion
A parsimonious electronic health record-based predictive model demonstrates improved performance for identifying pre-surgical patients who are at risk than previously-used assessments for POM triage.
Study context
- Population
- Elective surgical cases within a health system, 2014–2019
- Participants
- 470,727 elective surgical cases
- Comparison
- Comorbidity Assessment for Surgical Triage (CAST) score vs surgeon's complexity designation and ASA class for predicting post-operative morbidity
- Outcome
- Discrimination (AUC) for 30-day post-operative morbidity (composite of mortality and major NSQIP complications)
- Follow-up
- 30 days
Methodological detailssplit validation, electronic health record-based, single health system
- Split validation
- Electronic health record-based
- Single health system
Performance
As reported in the abstract
Discrimination for post-operative morbidity (AUC)
| Test or model | AUC |
|---|---|
| CAST | 0.75 |
| Surgeon's complexity designation | 0.63 |
| ASA | 0.65 |
AUC
CAST0.75
Surgeon's complexity designation0.63
ASA0.65
- CAST vs Surgeon's complexity designation: P < 0.001 — significantly higher discrimination with CAST
- CAST vs ASA: P < 0.001 — significantly higher discrimination with CAST
Further findings
| Outcome | Net reclassification improvement index | P value |
|---|---|---|
| Reclassification, CAST score groups vs complexity designation | 10.4% | P < 0.001 |
CAST model incorporating complexity designation (AUC)
- Discrimination not significantly altered (AUC 0.75; P = 0.098)
Original publication
Development and Validation of an Electronic Health Record-based Score for Triage to Perioperative Medicine. Ann Surg. Published 9 November 2021. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →