Vasculitis & polymyalgia rheumatica
Computable phenotype for real-world, data-driven retrospective identification of relapse in ANCA-associated vasculitis
Authors' conclusion
This CP accurately quantifies the individualised probability of relapse in AAV retrospectively, using objective, readily accessible registry data.
Study context
- Population
- Patients with ANCA-associated vasculitis (AAV) and >6 months follow-up recruited to the Rare Kidney Disease registry (national longitudinal, multicentre cohort study)1,209 development and 377 validation encounters
- Participants
- 536 patients
- Comparison
- Computable phenotype vs ground truth from independent encounter adjudication of primary medical records
- Outcome
- Accuracy (F1-score, sensitivity, specificity) of the CP for retrospective identification of relapse
Methodological detailscomputable phenotype development and internal validation, multilevel regression modelling, multicentre registry cohort, retrospective, Web application developed for computable phenotype (CP) implementation, outputting an individualised probability of relapse
- Computable phenotype development and internal validation
- Multilevel regression modelling
- Multicentre registry cohort
- Retrospective
- Web application developed for CP implementation, outputting an individualised probability of relapse
Performance
As reported in the abstract
Relapse identification, validation
| Test or model | F1-score | Sensitivity | Specificity |
|---|---|---|---|
| CP | 0.85 (95% CI 0.77 to 0.92) | 0.89 (95% CI 0.80 to 0.99) | 0.96 (95% CI 0.93 to 0.99) |
CP
F1-score0.85 (95% CI 0.77 to 0.92)
Sensitivity0.89 (95% CI 0.80 to 0.99)
Specificity0.96 (95% CI 0.93 to 0.99)
Further findings
Models handling missingness
- Where DE5 was missing, DE2 plus either DE1/DE3 were required to match the accuracy of BVAS
Original publication
Computable phenotype for real-world, data-driven retrospective identification of relapse in ANCA-associated vasculitis. RMD Open. Published 30 April 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →