Findings demonstrated a superior performance of the 10-year atherosclerotic cardiovascular disease (ASCVD) risk prediction models developed in patients with CKD, including novel kidney and cardiac biomarkers, when compared to equations developed for the general population using only traditional risk factors.

• In this study, 10-year ASCVD risk prediction models were built and validated in patients with CKD that included participants without self-reported cardiovascular disease from the Chronic Renal Insufficiency Cohort (CRIC) study.

• A total of 2,604 individuals (mean age 55.8 years; 52.0% male) included, of whom 252 had incident ASCVD within 10 years of baseline.

• Higher discrimination, and an AUC of 0.736, was achieved with a model with coefficients estimated within the CRIC sample, vs the American College of Cardiology/American Heart Association pooled cohort equations (area under the receiver operating characteristic curve [AUC]=0.730).

• An AUC of 0.760 was yielded by the CRIC model developed using clinically available variables.

• An AUC of 0.771 offered by the CRIC biomarker-enriched model was significantly higher than the clinical model.

• Both the clinical and biomarker-enriched models, relative to the pooled cohort equations, were well-calibrated and improved reclassification of nonevents (6.6%; 95% CI, 3.7% to 9.6% and 10.0%; 95% CI, 6.8% to 13.3%, respectively).