External Validation of the 2023 Australian Cardiovascular Risk Calculator

BACKGROUND: The Australian Cardiovascular (CVD) Disease Risk Calculator is used to estimate the individual risk of developing cardiovascular disease in the next 5 years. A new version was recently published (July 2023), with the aim of improving on the predictive performance of its predecessor (released in 2012). We present the findings of an external validation study comparing the predictive performance of the 2023 and 2012 Australian CVD risk calculators using data prospectively collected in the Hunter Community Study (HCS; NSW Australia), a longitudinal community-based cohort of people aged 55-85 years. METHODS: We compared the risk predicted by the two calculators to the observed 5-year events in the HCS, in terms of discrimination (using area under the receiver operator characteristic curve, AUROC), calibration (using observed vs expected, O/E, ratio), sensitivity, and specificity. RESULTS: Discrimination was very similar for the 2023 and 2012 calculators, with AUROC measured to be 0.71 95% confidence interval (CI; 0.66, 0.75) and 0.71 95% CI (0.67, 0.75), respectively. With the updated calculator, sensitivity was better in males, while specificity was better in females; there were also modest improvements in positive likelihood ratios for both males and females. The 2023 calculator was found to overpredict risk for males (O/E ratio 0.57, p<0.001), but was better calibrated for females (O/E ratio 1.02, p=0.46). CONCLUSIONS: We conclude that the 2023 calculator provides some improvements in the prediction of CVD, specifically the positive likelihood ratios. However, there are also benefits in observing the old 2012 calculator for some purposes and specific population groups. We find that there is a need for a larger, nationwide cohort to allow further external validation of the 2023 Australian CVD Risk Calculator.