The parent R01 focuses on developing reliable and interpretable statistical methods for the assessment of simultaneous health effects of multiple chemicals. This is challenging due to the statistical curse of dimensionality, to moderate to high correlation in levels of exposure to different chemicals, and to missing data and limit of detection issues. Current statistical methods for nonparametric regression fail to adequately address these challenges, and can produce uninterpretable dose response surfaces and high error rates in detecting interactions. The parent R01 is developing transformative methods that incorporate mechanistic constraints on response surfaces, allow for the complications inherent in epidemiology studies of mixtures, produce interpretable results including for interactions, and borrow information across different data sources. This R01 has already produced new statistical tools that clearly improve upon the state-of-the-art, and that can be implemented routinely by epidemiologists using publicly- available software packages (e.g., Ferrari and Dunson, 2020a,b). This proposal describes a competitive revision of the parent R01 to provide a transformative statistical toolbox for epidemiologists studying risk factors for COVID-19 infection, morbidity and mortality. This toolbox builds on the Bayesian modeling frameworks developed by the parent R01, while crucially accounting for the types of large spatially and temporally structured datasets that are now being collected as part of the COVID-19 monitoring effort. A new class of computational algorithms is proposed for rapid analysis of massive and complex spatial-temporal data, these algorithms are used to develop statistical tools for epidemiologists studying COVID-19 including an R package, and the approach is applied to study interactions between environmental exposures, age, and other comorbidities with COVID-19 mortality.