An electronic health record (EHR)-based risk calculator can predict fractures comparably to FRAX: a proof-of-concept study
Authors' conclusion
Routinely collected data in EHR systems can generate adequate fracture predictions without the need to verbally ascertain fracture risk factors.
Study context
- Population
- Subjects presenting to primary care, 2010 to 2018training 18,141
- Participants
- 24,189 subjects
- Comparison
- EHR-based fracture risk model vs FRAX
- Outcome
- Incident major osteoporotic fractures (MOFs), captured by physician diagnosis codes; discrimination and calibration vs FRAX in test set
Methodological detailsproof-of-concept, training and test sets, Cox regression, backward stepwise
- Proof-of-concept
- Training and test sets
- Cox regression
- Backward stepwise
Performance
As reported in the abstract
Discrimination for MOF in the test set
| Test or model | AUC |
|---|---|
| EHR model | 0.73 |
| FRAX | 0.70 |
AUC
EHR model0.73
FRAX0.70
- EHR model vs FRAX: P = 0.08
Original publication
An electronic health record (EHR)-based risk calculator can predict fractures comparably to FRAX: a proof-of-concept study. Osteoporos Int. Published 15 August 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →