LAUSR

Background Standardizing the Fried criteria (S‐FC) using cutoffs specific to the patient population improves adverse outcome prediction. However, there is limited evidence to determine if a S‐FC assessment can improve discrimination of cardiovascular disease (CVD) risk in middle‐aged and older women. Design The objective of this cross‐sectional analysis was to compare the ability of the Fried frailty phenotype criteria (FC) to discriminate between individuals at higher risk for CVD according to the Framingham Risk Score and Rasmussen Disease Score in comparison to the S‐FC. Setting Asper Clinical Research Institute, St. Boniface Hospital Research Centre. Participants 985 women 55 years of age or older with no previous history of CVD. Measurements Discrimination of individuals with high CVD risk according to the Framingham and Rasmussen Disease scores was assessed using receiver operating characteristic (ROC) curves, integrated discrimination index (IDI) and net reclassification index (NRI). Results The S‐FC showed superior ability to discriminate CVD risk as assessed by area under the ROC curve (AUROC) based on the Framingham (0.728 vs 0.634, p < 0.001), but not for the Rasmussen (0.594 vs 0.552, p = 0.079) risk score. Net reclassification index identified improved discrimination for both the Framingham (67.9%, p < 0.001) and Rasmussen Disease scores (26.0%, p = 0.003). Integrated discrimination index also identified improved CVD risk discrimination with the Framingham (3.0%, p < 0.001) and Rasmussen Disease scores (1.5%, p < 0.001). Conclusion In this study, the Fried frailty phenotype better discriminated cardiovascular disease risk when standardized to the study population. HighlightsUsing the same Fried frailty cut offs across all populations may lead to misclassification of frailty statusStandardizing Fried frailty criteria may improve cardiovascular disease risk discriminationFrailty assessment may be a means of improving cardiovascular disease management