Concerns about deteriorating mental health, substance use disorders, early development of chronic conditions, exposure to economic shocks, and to violence---causes that may all lead to premature mortality among adolescents and young adults in the United States, have been rising in recent public discourse. Despite these concerns, and the importance of this age group for the society’s future well-being, research on the trends, variation, and above all distal causes of mortality in this age group has been limited when compared to evidence on the causes and consequences of mortality among infants and older adults. This dearth of evidence likely stems from the absence of data sets with wide population coverage for this age group that links mortality—a very rare outcome—with anything beyond the basic demographic information about the individuals. In this project, we propose to take advantage of extensive infrastructure for linking US administrative data that has been developed by the U.S. Census Bureau, to construct a new database that links administrative birth and death dates for 10 to 25 year olds, with data on their demographics (race, ethnicity, sex, age, disability status), education, familial circumstances, parental income, access to public health insurance, healthcare providers, and local geographies, covering the time period from 2005 to the 2025. Using this new database, we will document how youth mortality evolved over time and space in the Unites States. We will also describe mortality inequalities across different subpopulations based on a rich set of demographic and socio-economic characteristics of adolescents, young adults, their families, and communities. Finally, we aim to use natural experiments and statistical techniques of causal inference to improve our understanding of the underlying distal causes of observed mortality patterns.