Abstract 13498: Fitness and Reclassification of Risk for Cardiovascular Disease Mortality
Background: Fitness is associated with cardiovascular disease (CVD) mortality. However, the extent to which fitness improves risk classification when added to traditional risk factors (TRF) is unclear.
Methods: We included 66,371 subjects (26% women, mean age 44 years) without prior CVD enrolled in the Cooper Center Longitudinal Study between 1970 and 2006. CVD death ascertained from the National Death Index through 2006 was the primary outcome variable. Fitness was measured by Balke protocol. Logistic regression was used to compare discrimination [assessed by c-index and integrated discrimination index (IDI)] by 2 separate models [(model 1: age, sex, systolic blood pressure, diabetes, total cholesterol and smoking); model 2: model 1 + fitness]. The predicted risk of CVD mortality using model 2 vs model 1 was calculated using Cox proportional hazards models. To assess net reclassification improvement at 10-years (NRI10), the predicted risk was categorized as <1%, 1% to <5%, and ≥5%. Similar analyses were conducted at 25-years (NRI25) with predicted risk categorized as <8%, 8% to 30%, and ≥30%.
Results: During 1048,344 person-years of follow up, 1621 CVD deaths occurred. Adding fitness to TRF resulted in significant improvement in discrimination (c-index: 0.830 vs 0.807 and IDI 0.013, p<0.001 for both). There was a graded decrease in the risk of CVD mortality with higher fitness levels (Table). Adding fitness significantly improved the NRI both at 10- and 25-years (NRI10 = 0.13 and NRI25 = 0.037, p<0.01 for both). The NRI25 was higher in women (0.131) than men (0.041). Most of the NRI10 was due to upward reclassification of those with CVD death (0.108) and a higher NRI was observed for the intermediate risk group (NRI10 = 0.283).
Conclusion: Fitness is not only associated with CVD mortality but also significantly improves measures of discrimination and reclassification when added to the TRF over both short-term (10-year) and long-term (25-year) follow-up.
- © 2010 by American Heart Association, Inc.