Arrhythmias & electrophysiologyDiagnostic or prognostic study
Development and validation of the GLORIA-AF Stroke Weighted Risk Score for patients with atrial fibrillation
- Development and validation of the GLORIA-AF Stroke Weighted Risk Score
- LASSO-penalized Cox regression with stability selection
- external validation in EORP-AF and APHRS-AF registries
- time-to-event framework
Authors' conclusion
The GLORIA-AF Stroke Weighted Risk Score provides risk refinement beyond CHA2DS2-VA while retaining discrimination consistent with more complex models. External validation supports its transportability and potential adjunctive role in guideline-directed thromboembolic risk assessment.
Study context
- Population
- Patients with atrial fibrillation (AF)derivation cohort, external validation 8309 patients · derivation cohort (GLORIA-AF Phase II/III) 20,517 · external validation (EORP-AF and APHRS-AF) 8,309
- Participants
- 20,517 patients
- Comparison
- Derived weighted score vs CHA2DS2-VA, CHA2DS2-VASc, the unweighted score, and full Cox or machine learning models
Performance
As reported in the abstract
mean (SD) age, years
69.9 (10.3)
women
9196 (44.8%)
had stroke
487 (2.4%)
receiving anticoagulation at baseline
17 397 (84.8%)
Further findings
C-index
| Outcome | GLORIA-AF Stroke Weighted Risk Score | CHA2DS2-VA | CHA2DS2-VASc | unweighted score | P value |
|---|---|---|---|---|---|
| Discrimination (C-index), derivation cohort — GLORIA-AF Stroke Weighted Risk Score vs CHA2DS2-VA · higher with derived score | 0.661 (95% CI 0.636–0.685) | 0.626 | 0.615 | Not reported in the abstract | P < 0.001 |
| Discrimination (C-index), derivation cohort — GLORIA-AF Stroke Weighted Risk Score vs CHA2DS2-VASc · higher with derived score | P < 0.001 | ||||
| Discrimination (C-index), derivation cohort — GLORIA-AF Stroke Weighted Risk Score vs unweighted score · higher with derived score | P = 0.016 | ||||
| Discrimination (C-index), external validation · higher with derived score | 0.652 (95% CI 0.614–0.690) | 0.616 (95% CI 0.598–0.634) | Not reported in the abstract | Not reported in the abstract | P < 0.001 |
| Discrimination among patients without baseline anticoagulation · higher with derived score · subgroup | P < 0.001 |
- Discrimination (C-index), external validation — strokes in external validation
- 147
- Discrimination (C-index), external validation: 8309 patients
Discrimination (C-index), derivation cohort
- no significant difference from the full Cox or machine learning models
Integrated discrimination improvement (IDI)/net reclassification improvement (NRI), calibration, and decision curve analyses
- supported improved risk differentiation, close calibration, and generally greater net benefit than CHA2DS2-VA
Original publication
Summarised from the published abstract; every number is checked against it