PROJECT ABSTRACT Sarcoidosis, a multiorgan disease with an uncertain etiology and variable clinical outcomes, is potentially influenced by environmental exposures, such as organic fuel burning or inorganic dust. The higher incidence and prevalence of sarcoidosis among veterans than in civilians provides a rationale to examine the role of service-related toxic exposures in the natural history of the disease. Such studies are consistent with the goals of the recently approved PACT Act, which recognizes sarcoidosis as a service-related illness and calls for studies of service-related exposures as risk factors for lung disease. The Veteran's Health Administration (VHA), the largest integrated healthcare system in the U.S., obtains comprehensive longitudinal data on millions of veterans, including their unique history of service-related toxic exposures. This wealth of data enables the exploration of these relationships with high precision. My central hypothesis is that service-related toxic exposures affect the incidence of sarcoidosis and all-cause mortality. I propose two specific aims: (1) to determine in a nested case- control study whether service-related exposures increase sarcoidosis incidence risk among veterans and (2) to identify predictors of all-cause mortality among veterans with sarcoidosis using a time-dependent Cox proportional hazards model and two other machine learning models; in two subaims, I will determine whether individual clinical features predict all-cause mortality and whether incorporating the service-related exposures into the models will improve this prediction. This study will reveal the effects of toxic exposures on sarcoidosis incidence, develop predictive models for all-cause mortality, and stratify veterans into risk subgroups—which will improve longitudinal care and guide efforts to identify new sarcoidosis phenotypes. The findings will lay the foundation for a Career Development Award proposal focusing on real-world, evidence-based management strategies for sarcoidosis.