JACC: Advances (Jan 2025)

Estimating the Risk of Cardiovascular Events in U.S. Veterans Using the SMART Risk Score

  • Subhash Banerjee, MD,
  • Rick A. Weideman, PharmD,
  • David A. Jacob, PharmD,
  • Helene R. DiGregorio, MD,
  • Kevin C. Kelly, PharmD,
  • Avantika Banerjee, MD,
  • Milan Ravishankar, MD,
  • Patrick T. Strickland, MD,
  • Helayna M. Abraham, MD,
  • Nicole E. Minniefield, MD,
  • Bradley R. Grimsley, MD,
  • Jeffrey M. Schussler, MD,
  • Rohit J. Parmar, MD,
  • Robert C. Stoler, MD,
  • Emmanouil S. Brilakis, MD, PhD,
  • Bertis B. Little, PhD

Journal volume & issue
Vol. 4, no. 1
p. 101459

Abstract

Read online

Background: Estimation of long-term risk for cardiovascular events using the SMART (Secondary Manifestations of Arterial Disease) risk score can be potentially valuable in devising risk mitigation strategies. Objectives: The objective of this study was to apply the SMART risk score to compute the risk for major adverse cardiovascular events (MACE) in the U.S. Veteran patient population. Methods: We used the Veterans Affairs (VA) informatics and computing infrastructure to identify patients referred for an initial outpatient cardiology evaluation between the years 2003 and 2010 to estimate 10-year risk for composite MACE (all-cause death, ischemic stroke, and nonfatal myocardial infarction). Cox regression and survival curves were used to develop and validate the VA SMART score. Results: The study population included 472,702 patients (mean age 60 ± 8.9 years, 96% male) who were allocated into development (n = 94,091) and test cohorts (n = 378,611). The median follow-up time was 7.9 years (IQR: 6.0-9.9). The VA-SMART score allowed accurate estimation of MACE. Patients were stratified in low (<10%), moderate (10% to 20%), high (20% to 30%), and very high (≥30%) risk groups with observed events rates of 6.8%, 17.9%, 28.5%, and 49.5%, respectively, in the test cohort (P < 0.0001 for all intergroup comparisons). Most MACE events were all-cause death, with nonfatal myocardial infarction and stroke also being high, especially in the very high-risk group. The VA SMART score performed similar to other established risk prediction models (C-statistic = 0.67). Conclusions: The VA SMART risk score can estimate the long-term risk of recurrent cardiovascular events in U.S. Veterans and could help implement individualized risk mitigation strategies.

Keywords