Performance of the Systematic COronary Risk Evaluation 2 (SCORE2) in the EPIC-Norfolk prospective population study

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): EPIC-Norfolk is supported by programme grants from the Medical Research Council UK (MRC G0401527, MRC G0701863, MRC G1000143) and Cancer Research UK (CRUK 8257). Introduction The latest European guidelines recommend the use of the Systematic COronary Risk Evaluation 2 (SCORE2) to assess 10-year risk of fatal and non-fatal cardiovascular disease (CVD) in individuals aged 40-70 years. Therefore, validation of the SCORE2 algorithm in a large population is needed. Purpose To assess the predictive performance of SCORE2 in individuals without prior myocardial infarction, stroke or diabetes mellitus. Methods Individuals aged 40-70 without prior myocardial infarction, stroke or diabetes mellitus were included from the European Prospective Investigation of Cancer Norfolk (EPIC-Norfolk) prospective population cohort. We assessed discriminative power using the C-statistic; calibration was assessed by plotting calculated SCORE2 (low-risk algorithm) 10-year risks against observed CVD event rates. We defined the composite cardiovascular outcome as death due to ischemic heart disease, cardiac failure, cerebrovascular disease or peripheral-artery disease, or non-fatal myocardial infarction or non-fatal stroke. Observed probabilities were adjusted for competing mortality events. Results We included 20,318 individuals (56% women) with a median (IQR) age of 56 (50-63) years, with baseline measures between 1993-1997. Mean predicted 10-year risk was 3.7% (95% confidence interval (CI) 3.7-3.8) versus adjusted observed 4.2% (95% CI 3.9-4.5) 10-year CVD incidence, yielding a ratio of 1.1 (95% CI 1.1-1.2).(Figure) Observed events consisted of 1.5% (95% CI 1.3-1.7) fatal CVD events, 1.8% (95% CI 1.6-2.0) non-fatal myocardial infarctions, 1.5% (95% CI 1.3-1.6) non-fatal strokes and 23.4% (95% CI 22.1-23.2) competing mortality events. Overall C-statistic was 0.75 (95% CI 0.73-0.77). Conclusion SCORE2 demonstrates reasonable discriminative ability and accurate risk estimations in low-risk individuals in a large, prospective population cohort. However, as actual CVD incidence is expected to be lower given the baseline between 1993-1997, these findings do not indicate that SCORE2 is suitable for current use for risk assessment in individuals from the United Kingdom. Calibration plot: predicted vs. observed.