Prediction of incident heart failure in individuals without prior cardiovascular disease: the SCORE2-HF risk model

Background and aims: Heart failure (HF) presents a significant and growing public health challenge. The aim of this study was to develop and validate SCORE2-HF, a model for HF risk estimation in European adults aged over 40 years without previous cardiovascular disease. Methods: Using data from 25 prospective cohorts (14 countries, 611 778 individuals, 21 818 incident HF events), the sex-specific, competing risk-adjusted SCORE2-HF models were derived, including age, smoking status, systolic blood pressure, antihypertensive treatment, body mass index (BMI), estimated glomerular filtration rate, and type 2 diabetes mellitus (including age at diagnosis and glycated haemoglobin). Using Europe-wide statistics from the World Health Organization and linked health records from five countries (>36 million individuals, 515 466 incident HF events), models were recalibrated to contemporary 10-year and 30-year HF incidence in four European risk regions. SCORE2-HF was validated using data from three further cohorts (three countries; 1 336 824 participants; 36 841 incident HF events). Results: In the three external validation cohorts, C-indices (95% confidence interval) were .827 (.824-.829), .839 (.827-.850), and .874 (.863-.884). SCORE2-HF risks varied importantly by individual's risk factors and risk region. For example, in the low-risk region, the average 10-year risk for 70-year-old individuals with zero vs four adverse risk factors (smoking, type 2 diabetes mellitus, hypertension, and BMI ≥ 30 kg/m2), was 8% vs 24% in men and 6% vs 20% in women. By contrast, in the very high-risk region, the average SCORE2-HF risk with four adverse risk factors was 59% in 70-year-old men or women. Conclusions: SCORE2-HF-a model derived, recalibrated, and validated to estimate 10-year and 30-year risk of incident HF across European countries-may enhance the identification of individuals at higher risk of developing HF.