Abstract MP18: Association Of Longitudinal Social Determinants Through Young Adulthood With Mid-life Cardiovascular Health: The Coronary Artery Risk Development In Young Adults (CARDIA) Study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Introduction: Deeper understanding of the longitudinal associations of social determinants of health (SDOH) at younger ages with cardiovascular health (CVH) in middle age is needed to address persistent health disparities. Methods: We used repeated measures of individual and neighborhood-level SDOH data from CARDIA participants (ppts) to understand how SDOH change from ages 18-44 years. There were 48 diverse SDOH exposures measured. Using a novel, machine learning, sequential pattern mining method (Subgraph Augmented Non-negative Matrix Factorization), we identified clusters of time-dependent SDOH variables occurring in 5% or more of ppts. Clusters were then used as independent variables in logistic regression models to assess associations with CVH at age ≥45 years. CVH status was defined using the AHA’s Simple 7 metrics on diet, physical activity, smoking, BMI, cholesterol, blood pressure, and blood glucose, and summarized by a 14-point score (poor 0-7, moderate 8-11, high 12-14 points). We tested associations of clusters with the outcome of poor CVH at age ≥45 years, and identified the characteristics of significant SDOH clusters. Results: There were 3,522 ppts included (46.3% Black, 55.9% women; mean age 25.3 years at baseline). Mean CVH score (out of 14) was 10.4 ± 1.8 points at baseline and 8.9 ± 2.3 points at age ≥45 years; 483 ppts (13.7%) had high, 2026 (57.5%) had moderate, and 1013 (28.8%) had poor CVH in middle age. The best performing logistic model included 4 SDOH trend clusters (Table). SDOH Clusters 1, 2, and 3 were each associated with significant 50-80% lower odds of having poor CVH in middle age, whereas Cluster 4 was associated with 2.5-fold higher odds of poor CVH. The most informative variables over time comprising each SDOH cluster are also shown in the Table. Conclusions: Specific SDOH clusters through young adulthood are significantly associated with reaching mid-life with poor CVH status. Components of these SDOH clusters may be potential targets for social interventions and policies to improve CVH.