Investigation of the social context and physical environment on cardiovascular disease disparities in the All of Us Research Program

Abstract Persistent prevalence of cardiovascular disease (CVD) and its high morbidity and mortality suggest that there are significant gaps in our understanding of the mechanisms that impair CVD health and better strategies are needed. It is well known in the scientific community that the impact of environmental justice on CVD extends

beyond that of air pollution and encompasses neighborhood deprivation and social determinants. The literature also consistently demonstrates that low income, vulnerable, and underrepresented individuals are at the greatest risk of excessive exposure and lower resources to cope with and respond to the problems posed by these factors,

including CVD. However, integrated, and comprehensive investigation of the multiple layers of environmental influence with individual-level factors is limited, especially in studies powered to investigate racial/ethnic disparities. We recently estimated that median All of Us participant PM2.5 level is 10 μg/m3, slightly higher than

the US average of 7 μg/m3 and that people in the highest category have a 1.5 times increased risk of stroke. These preliminary results suggest that effects are certainly being detected, but the story is complex and requires analytic methods to tease apart correlated effects, weight influence, consider the effect of bias and offer

insight into the factors commonly thought to be protective among diverse populations and communities. This project leverages the unprecedented opportunity to utilize existing data in the All of Us Researcher Workbench, our prior linkage of environmental factors into the dataset, and novel analytic methods to investigate how the

social context and physical environment jointly contributes to disparities in CVD risk, and to further evaluate the impact of individual-level factors (e.g., established genes, individual demographics) on the modification and/or mitigation of these impacts. Our prior work in this area highlights how novel methods such as weighted quantile

sum regression can offer interpretable effects and clear messaging on the percent contribution of different factors to chronic disease risk and disparities. As such, in addition to leveraging the unprecedented diversity in this national cohort, the approach is responsive to what our IPMC communities and community partners have

requested in terms of areas of research interest and understandable answers.