Heart failure & cardiomyopathies
Predictors of 1-Year Mortality Among Patients With Heart Failure With Preserved Ejection Fraction
Authors' conclusion
Our transparent risk tool using routinely available admission data appears feasible, allowing for patient-level, precision health risk assessment.
Study context
- Population
- Adults with ejection fraction (EF) ≥50% admitted for a first-time heart failure (HF) exacerbationadmissions analysed
- Participants
- 7,840 admissionsEvents: 31.5%
- Predictors
- Extreme Gradient Boosting, Light Gradient Boosting, Cox proportional hazards, random survival forest (RSF), and gradient boosting survival (GBS); variables spanning demographics, comorbidities, laboratory tests, echocardiographic variables, medications, and outcomes
- Predicted outcome
- 1-year all-cause mortality
Methodological detailsdevelopment and internal validation of prediction models, electronic medical records from a large academic health system
- Development and internal validation of prediction models
- Electronic medical records from a large academic health system
Model performance
As reported in the abstract
Test-set AUROC
Extreme Gradient Boosting
0.751(95% CI: 0.727-0.775)
Light Gradient Boosting
0.749(95% CI: 0.721-0.776)
- Acceptable calibration
Further findings
Concordance index
GBS
0.718(95% CI: 0.696-0.740)
RSF
0.711(95% CI: 0.690-0.734)
Cox
0.704(95% CI: 0.680-0.728)
Time-dependent AUROC
GBS
0.759(95% CI: 0.716-0.799)
RSF
0.750(95% CI 0.708-0.789)
Cox
0.735(95% CI 0.692-0.777)
Cohort characteristics
- Mean age
- 78 years
- Women
- 55.6%
Most consistent risk signals
- Lower albumin, older age, higher N-terminal pro-B-type natriuretic peptide, renal dysfunction, and lower hemoglobin
Original publication
Predictors of 1-Year Mortality Among Patients With Heart Failure With Preserved Ejection Fraction. JACC Adv. Published June 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →