Abstract P332: Framingham Risk Score to Predict Subsequent Coronary Artery Calcification in a Young Adult Population

Introduction: Coronary artery calcification (CAC) has been identified as an independent predictor of coronary events, suggesting the potential use of CAC for risk assessment in asymptomatic individuals. However, ionizing radiation exposure associated with CAC CT scans remains a concern. If elevated risk for developing subclinical cardiovascular disease (CVD) could be identified based on less invasive risk assessment, such as the Framingham risk score, intensified prevention and screening services could be provided to this targeted population. Hypothesis: This study aims to assess the association between the Framingham risk score in early adulthood and subsequent subclinical CVD measured by CAC. Additional risk factors including demographics, socioeconomic status and health behaviors were tested in terms of their capabilities to enhance prediction of subclinical CVD beyond the Framingham risk score. Methods: This study used the Coronary Artery Risk Development in Young Adults (CARDIA) data, with a total of 5,115 Caucasian and African American males and females. Information collected at examination year 10 was used to calculate the Framingham risk score. CAC was measured ten years later (examination year 20). Participants’ demographics, health behaviors (alcohol consumption, BMI, and exercise), socioeconomic status and medical needs at year 10 were identified as potential risk factors associated with the subsequent presence of CAC beyond the Framingham risk score. Multiple logistic regression was used to examine the adjusted association between CAC, Framingham risk score and proposed risk factors. Model comparison was estimated using the area under the receiver operating characteristic curve (AUC) and Akaike information criterion (AIC). Results: By year 20, CAC was present in 19% of the CARDIA population. The Framingham risk score in young adulthood was strongly associated with the subsequent presence of CAC ten years later, regardless of race and gender. Overall, 42% of the CARDIA participants with elevated Framingham risk scores at year 10 had CAC at year 20, compared to 16% of participants with normal scores. The Framingham risk score may underestimate the risk of CAC for males compared to females (Negative Predictive Value: 75% vs. 91%). Beyond the Framingham risk score, the subsequent presence of CAC was associated with being overweight or obese in all populations, at-risk alcohol consumption in African American males, and having high school level or lower education and financial hardship in African American females. Conclusions: Our findings support the potential use of the Framingham risk score as a screening tool for subsequent subclinical atherosclerosis in a young adult population. However, other gender-specific risk factors beyond the Framingham risk score such as obesity also may be important to better predict subclinical CVD risk, especially in male populations.