July 12, 2026Scientific Reports0 citationsOpen Access
Scalable risk stratification of undiagnosed heart failure using routine health data and its association with imaging phenotypes and outcomes
Key Result
The FIND-HF electronic health record model predicted incident heart failure within one year across international cohorts (AUC 0.72-0.85), with high risk associated with worse clinical outcomes.
Key Points
- This research aims to create a tool to identify individuals at high risk for undiagnosed heart failure using routine health data.
- Developed and validated the FIND-HF logistic regression model using UK primary care electronic health records.
- Conducted internal validation with a sample size of 3,520,186 and external validation across multiple regions.
- Analyzed associations between FIND-HF risk levels and cardiac outcomes using imaging data.
- FIND-HF model achieved an AUC of 0.79 for incident heart failure in internal validation.
- The model showed good predictive performance in external cohorts with AUCs ranging from 0.72 to 0.85.
- High FIND-HF risk was linked to a 21.4% positive predictive value for heart failure hospitalization or cardiovascular death.
Study Design
Type
Observational (n=18,792,646)
Multicenter
Yes
Structured PICO
Does the FIND-HF model accurately identify patients at high risk of undiagnosed heart failure using routine electronic health records?
P
Population
18,792,646 patients from primary care electronic health records across the UK, Japan, USA, and Taiwan evaluated for incident heart failure within one year.
E
Exposure
FIND-HF model (a logistic regression tool using routine electronic health records to identify high risk of undiagnosed heart failure)
C
Comparator
Low FIND-HF risk (in the cardiac magnetic resonance imaging cohort)
O
Outcome
Incident heart failure diagnosis within one yearsurrogate
The FIND-HF model is a scalable, EHR-based tool that demonstrates good predictive performance for identifying patients at high risk of undiagnosed heart failure across diverse international healthcare systems.
Main Result
Effect estimate: AUC 0.79
Abstract
Abstract Late diagnosis of Heart failure (HF) is associated with worse outcomes. We aimed to develop a scalable tool to identify those at high risk of undiagnosed HF using routine electronic health records (EHR). We developed and internally validated a logistic regression (FIND-HF) model for incident HF diagnosis within one year in United Kingdom primary care EHRs (CPRD-Aurum, n=3 520 186), with good prediction performance (area under the receiver operating characteristic curve (AUC) 0.79), equal to more complex modelling techniques. We externally validated FIND-HF in United Kingdom (CPRD-GOLD, n=570 850, AUC 0.72), Japan (JMDC, n=6 820 694, AUC 0.73), United States of America (Epic Cosmos, n=7 710 398, AUC 0.78), and Taiwan (NTUH, n=170 518, AUC 0.85). In a cohort who had undergone HF diagnostics an optimised FIND-HF threshold had a positive predictive value of 21.4% and a negative predictive value of 96.9%. Amongst patients with HF who had undergone cardiac magnetic resonance imaging, high FIND-HF risk compared with low FIND-HF risk as reference, was associated with increased risk of a primary composite outcome of heart failure hospitalisation or cardiovascular death and more advanced adverse remodelling including lower left ventricular ejection fraction. FIND-HF is a scalable EHR-based model which has the potential to help rule out undiagnosed HF in low risk cases, whilst high risk cases are associated with more advanced cardiac dysfunction and worse prognosis.