A recent study reported that the aortic–brachial arterial stiffness gradient, defined as carotid–radial/carotid–femoral pulse wave velocity (PWV ratio), predicts all-cause mortality better than carotid–femoral pulse wave velocity (CFPWV) alone in dialysis patients. However, the prognostic significance of PWV ratio for cardiovascular disease (CVD) in the community remains unclear. Accordingly, we assessed the correlates and prognostic value of the PWV ratio in 2114 Framingham Heart Study participants (60±10 years; 56% women) free of overt CVD. Mean PWV ratio decreased from 1.36±0.19 in participants aged <40 years to 0.73±0.21 in those aged ≥80 years. In multivariable linear regression, older age, male sex, higher body mass index, diabetes mellitus, lower high-density lipoprotein cholesterol, higher mean arterial pressure, and higher heart rate were associated with lower PWV ratio (P<0.001 for all). During a median follow-up of 12.6 years, 248 first CVD events occurred. In Cox regression models adjusted for standard CVD risk factors, 1-SD changes in CFPWV (hazard ratio, 1.33; 95% confidence interval, 1.10–1.61) and PWV ratio (hazard ratio, 1.32; 95% confidence interval, 1.09–1.59) were associated with similar CVD risks. Models that included conventional CVD risk factors plus CFPWV or PWV ratio gave the same C statistics (C=0.783). Although PWV ratio has been reported to provide incremental predictive value over CFPWV in dialysis patients, we could not replicate these findings in our community-based sample. Our findings suggest that the prognostic significance of PWV ratio may vary based on baseline CVD risk, and CFPWV should remain the criterion standard for assessing vascular stiffness in the community.