Abstract
BACKGROUND: Social determinants of health (SDH) have been found to contribute to cardiovascular risk and the development of cardiovascular disease (CVD). However, few studies have examined early-life exposure to SDH and the combined effect of multiple SDH measures on CVD. There is an urgent need to develop an SDH index for use in practice and clinical settings. METHODS: A total of 3189 participants from the CARDIA Study who had chest CT scans at the year-25 (Y25) visit were included in this study. Personal and parental SDH measures, including education, occupation, income, financial strain, and childhood family environment, were obtained through interviews. The participants' coronary artery calcification (CAC) was measured using chest CT scans, and left-ventricular mass (LVM) was measured using M-mode echocardiography. The values of the individual social determinants of health (iSDH) index were determined based on individual-level measures and CAC using a supervised learning method-the Boosted Regression Tree (BRT) model. This index's association with the LVM index (LVMI) was evaluated as an external validation using linear regression models adjusting for race, sex, BMI, smoking status, alcohol intake, and physical activity. RESULTS: Each one-standard-deviation (SD) increase in the iSDH index was associated with an increase in LVMI ranging from 0.376 (95% CI -0.016, 0.767) at year 0 to 0.468 (95% CI 0.115, 0.821) at year 20. The association between the iSDH index and the LVMI was more pronounced as the participants aged. Also, the iSDH indices were more strongly associated with LVMI among Black participants (β = 0.969, 95% CI = 0.081, 1.858) than White participants (β = 0.202, 95% CI = -0.228, 0.633) at year 5. CONCLUSIONS: Higher iSDH indices in early adulthood were associated with increased LVMI values in midlife. The association between the iSDH index and CVD was stronger among Black adults than with White adults.