Development of a Machine Learning Model to Predict the Use of Surgery in Patients With Rheumatoid Arthritis
Authors' conclusion
Predictive models using clinical data have the potential to facilitate identification of patients who may undergo rheumatoid-related surgery, but not what type of procedure they will need.
Study context
- Population
- Patients with rheumatoid arthritis (RA) identified from electronic health record data278 (5.1%) underwent surgery
- Participants
- 5,481 patients
- Comparison
- Surgery-use model vs Surgery-type model
- Outcome
- Model discrimination (AUC)
Methodological detailsmachine learning, extreme gradient boosting, electronic health record data
- Machine learning
- Extreme gradient boosting
- Electronic health record data
Performance
As reported in the abstract
| Test or model | AUC |
|---|---|
| Surgery-use model | 0.90 ± 0.02 |
| Surgery-type model | 0.58 ± 0.10 |
AUC
Surgery-use model0.90 ± 0.02
Surgery-type model0.58 ± 0.10
Further findings
| Outcome | P value |
|---|---|
| NSAID prescriptions, surgery vs no surgery | P = 0.03 |
- NSAID prescriptions, surgery vs no surgery: NSAID prescriptions more common among patients who had surgery
DMARD or steroid prescriptions, surgery vs no surgery
- No significant difference in frequency of DMARD or steroid prescriptions between patients with and without surgery
Original publication
Development of a Machine Learning Model to Predict the Use of Surgery in Patients With Rheumatoid Arthritis. Arthritis Care Res (Hoboken). Published 19 March 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →