Agreement of Administrative Pharmacy Dispensing With Self‐Reported Use of Oral Prednisone in US Veterans With Rheumatoid Arthritis
Authors' conclusion
Algorithms derived from VA claims detecting prednisone prescriptions have high validity compared to patient self-report. The modest sensitivity of these algorithms may reflect stockpiling and non-VA prescriptions.
Study context
- Population
- Participants with rheumatoid arthritis (RA) enrolled at seven Veterans Affairs Rheumatoid Arthritis (VARA) Registry sites
- Participants
- 284 participants
- Comparison
- Veterans Affairs (VA) prescription claims-based algorithms vs self-reported prednisone use
- Outcome
- Performance characteristics and agreement of claims-based algorithms for glucocorticoid use vs self-report
Methodological detailsmulticentre, registry-based, benchmarked on self-report
- Multicentre
- Registry-based
- Benchmarked on self-report
Performance
As reported in the abstract
Performance of VA claims-based algorithms vs self-report (benchmark)
| Test or model | Sensitivity | Specificity | PPV | NPV | Cohen's κ |
|---|---|---|---|---|---|
| Current use | 0.70 | 0.98 | 0.84 | 0.96 | 0.68 |
| 90-day use | 0.71 | 0.92 | 0.72 | 0.92 | 0.63 |
Sensitivity
Current use0.70
90-day use0.71
Specificity
Current use0.98
90-day use0.92
PPV
Current use0.84
90-day use0.72
NPV
Current use0.96
90-day use0.92
Cohen's κ
Current use0.68
90-day use0.63
Further findings
Self-reported prednisone use
- Reported current prednisone use
- 13%
- Reported 90-day use
- 20%
Agreement for dose among participants reporting ≤5 mg/day
- Weighted κ
- 0.67
Stockpiling and non-VA prednisone
- Reported a stockpile
- One in four
- Reported receiving prednisone from a non-VA provider
- one in four
Original publication
Agreement of Administrative Pharmacy Dispensing With Self‐Reported Use of Oral Prednisone in US Veterans With Rheumatoid Arthritis. Arthritis Care Res (Hoboken). Published 8 August 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →