Coronary disease & ACSDiagnostic or prognostic study
Meta-prediction of coronary artery disease risk
- Prognostic model development and validation study
- external validation
~2,000 candidate predictive features; prevalent coronary artery disease (CAD) cohort trained baseline models used as meta-features for the incident CAD model
Study context
- Population
- UK Biobank (development); independent test cohort from the All of Us research program
- Comparison
- Meta-prediction model vs Standard clinical scores
- Outcome
- 10-year incident CAD risk prediction (AUC)
- Follow-up
- 10 years
Performance
As reported in the abstract
10-year incident CAD
| Test or model | AUC (UK Biobank) | AUC (All of Us test cohort) |
|---|---|---|
| Meta-prediction model | 0.84 | 0.81 |
Further findings
Individualized risk reduction profiles
- Genetic risk influences the extent to which standard clinical interventions reduce overall CAD risk
Original publication
Meta-prediction of coronary artery disease risk. Nat Med. Published 16 April 2025. doi:10.1038/s41591-025-03648-0
Summarised from the published abstract; every number is checked against it