Development and validation of a multi-institutional electronic juvenile idiopathic arthritis phenotype
Authors' conclusion
We developed and reproduced a computable and portable phenotype algorithm to identify JIA patients in structured, retrospective data, validated this across multiple institutions, and demonstrated its consistent performance across sites.
Study context
- Population
- Subjects with ICD-9 and ICD-10-CM codes for juvenile idiopathic arthritis (JIA) and/or psoriatic juvenile arthropathy, age ≤20 at first eligible encounter, excluding multiple ICD codes for SLE, DM and PM; three sites of the Maternal and Pediatric Precision in Therapeutics Hub253 (3 %) manually reviewed
- Participants
- 8,093 subjects
- Comparison
- Computable JIA phenotype algorithms with increasing JIA code counts, validated against manual chart review
- ≥4 code countsn = 5,377
- Outcome
- PPV of the computable JIA phenotype vs manual chart review
Methodological detailsmulti-institutional, multi-site validation and replication, manual chart review, Each site manually reviewed a subset of charts to validate diagnosis
- Multi-institutional
- Multi-site validation and replication
- Manual chart review
- Each site manually reviewed a subset of charts to validate diagnosis
Performance
As reported in the abstract
PPV
| Test or model | PPV |
|---|---|
| 1 code count | 0.74 |
| ≥4 code counts | 0.95 |
PPV
1 code count0.74
≥4 code counts0.95
Original publication
Development and validation of a multi-institutional electronic juvenile idiopathic arthritis phenotype. Semin Arthritis Rheum. Published 16 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →