In the U.S., approximately 840,000 Americans die from cardiovascular disease (CVD) each year. The prevalence of CVD is on the rise and widespread disparities in CVD exist across economic, racial, and ethnic groups. In order to address the rising prevalence of CVD and persistent disparities, there has been a shift in focus to public health strategies addressing cardiovascular health (CVH). CVH is a broader and more positive construct beyond the absence of CVD. Despite this recent focus on improving CVH, widespread disparities still exist. Social determinants of health (SDOH) may be important contributors to these continued disparities. The World Health Organization (WHO) defines SDOH as the ?structural determinants and conditions in which people are born, grow, live, work, and age.? There has been limited work in studying how a diverse set of SDOH change over time and perform in the prediction of CVH. To address this need, we will identify patterns of SDOH exposure over time and determine if the addition of SDOH variables allows for better prediction of an individual's CVH status. The primary hypothesis is that a diverse set of SDOH will be associated with and improve the prediction of CVH, independent of baseline CVH and other covariates. To examine this hypothesis, we will pursue the following Specific Aims: 1) identify patterns of SDOH exposure up to age 50 and define exposure subgroups and 2) determine whether overall and domain-specific patterns of SDOH exposures from young adulthood to middle age, identified in Aim 1, are associated with and improve the prediction of CVH and its component metrics. We will utilize the Coronary Artery Risk Development in Young Adults (CARDIA) study, a prospective cohort study with detailed information on cardiovascular risk factors and disease in a geographically and racially diverse sample of young adults.
In Aim 1, we will use a novel sequential pattern mining method to identify the associations among SDOH and determine the SDOH exposure patterns from baseline to age 50.
The Aim 1 SDOH exposure definitions will be included as predictors of CVH at age 50 and beyond using supervised machine learning techniques. By including SDOH in predictive models, health services professionals and clinicians may have an improved understanding of patients at high-risk for low CVH and may better tailor social and clinical interventions to each patient's needs. If awarded, this fellowship will allow me to contribute novel research to the SDOH and cardiovascular health fields, gain competency in new research skills, and improve my writing and presentation skills. The completion of this project and training will prepare me for my long-term career goal of becoming an independent research scientist in an academic setting, studying SDOH and using data science and informatics tools to improve public health.
The incorporation of social determinants of health (SDOH) in predictive models for cardiovascular health (CVH) may help health services professionals and clinicians in identifying patients at high-risk for low CVH. This has important public health implications because social and clinical interventions may be tailored to each patient's needs. This research may also help to identify which SDOH are predictive of CVH from young adulthood to middle age, allowing public health professionals to target specific SDOH through directed social, behavioral, and policy-related interventions.
