Encounter Appropriateness Score for You Model: Development and Pilot Implementation of a Predictive Model to Identify Visits Appropriate for Telehealth in Rheumatology
Authors' conclusion
The EASY model accurately identifies future visits that are appropriate for telehealth.
Study context
- Population
- Rheumatology encounters at a single academic rheumatology practice, January 1, 2021, to December 31, 2021
- Participants
- 21,679 encounters
- Comparison
- True vs artificial recommendations for change in encounter modality
- Outcome
- Model performance (AUC) in predicting telehealth appropriateness for future rheumatology visits
Methodological detailslogistic regression model, single academic rheumatology practice, multiple validation sets
- Logistic regression model
- Single academic rheumatology practice
- Multiple validation sets
Performance
As reported in the abstract
Model discrimination (AUC) across multiple validation sets
AUC
0.831 to 0.855
Further findings
| Outcome | Cohen's κ | P value |
|---|---|---|
| Pilot: provider agreement with true vs artificial recommendationsProviders more likely to agree with true recommendations | 0.45 | P < 0.001 |
- Pilot: provider agreement with true vs artificial recommendations: 201 encounters; four providers
Pilot: provider agreement with true vs artificial recommendations
- Model increased the number of appropriate telehealth visits
Original publication
Encounter Appropriateness Score for You Model: Development and Pilot Implementation of a Predictive Model to Identify Visits Appropriate for Telehealth in Rheumatology. Arthritis Care Res (Hoboken). Published 13 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →