PROJECT SUMMARY The 2019 novel coronavirus disease (COVID-19) is now a global pandemic with severe consequences. Individuals with Alzheimer’s disease and related dementias (ADRD) are at a higher risk of contracting COVID- 19 and experiencing severe COVID-19. However, risk factors for COVID-19 severity among individuals with ADRD beyond older age and comorbidities remain largely undefined, and the ability to predict COVID-19 severity among ADRD patients is limited. There are large overlaps between the currently known risk factors of severe COVID-19 and the health conditions that are affected by environmental exposures, and emerging evidence suggested that long-term environmental exposures may be important determinants of COVID-19 severity. Our current R21 (1R21ES032762) was set to identify novel environmental exposures associated with COVID-19 severity considering the totality of the external environment (or the external exposome). Compared with the general population, individuals with ADRD are substantially different in their individual characteristics and environmental exposure profiles. Therefore, a more focused study on the relationships between environmental exposures and COVID-19 severity among ADRD individuals is needed. The rise of real-world data (RWD) creates the perfect opportunity to create a synthetic COVID-19 cohort of individuals with ADRD at the scale needed to understand the full-spectrum of COVID-19 outcomes among individuals with ADRD. Nevertheless, there are two key barriers: (1) the lack of a consensus on methods for examining the external exposome-health associations, and (2) the lack of a pipeline to accurately identify ADRD patients from RWD and extract important risk factors that are only available in unstructured clinical notes. Expanding our parent award on the external exposome and COVID-19 severity to focus on individuals with ADRD, we aim to fill these important gaps to: (1) create a synthetic longitudinal COVID-19 cohort of individuals with ADRD, (2) systematically evaluate statistical methods for external exposome-wide association study (ExWAS), specifically for COVID-19 outcomes among ADRD patients, and (3) through an external ExWAS, identify external exposome factors associated with COVID-19 severity and then develop predictive models of COVID-19 severity among individuals with ADRD. This study builds upon our continuous work on the external exposome and OneFlorida – a repository of RWD with linked electronic health records, claims, and vital statistics data that covers > 60% of Floridians. This study will generate: (1) a large longitudinal COVID-19 cohort of individuals with ADRD, (2) standardized state-of-the-art statistical methods to conduct external ExWAS, and (3) novel external exposome factors associated with COVID-19 severity and predictive models of ADRD patients at high- risk of severe COVID-19. Our approach can be scaled up through PCORnet to create a national cohort and expanded to examine COVID-19’s l...