The rise in U.S. midlife mortality in recent decades has been substantial, ending the increase in life expectancy
around 2010 and triggering its decline after 2014. The trend has been exacerbated by the COVID-19 pandemic.
One of the most telling features of the rise is its geographic pattern. It has been pronounced in Midwestern and
Southern states and in rural areas and small cities. Explaining these growing geographic disparities is a
necessary step toward identifying the etiologies of rising midlife mortality overall. The overarching objective of
this project is to assess how state policy contexts and county economic contexts collectively predict growing
geographic disparities in 1) all-cause midlife mortality, 2) major trend-driving causes of death for midlife mortality:
suicide, drug overdose, alcohol-induced causes, and cardiometabolic diseases, and 3) psychosocial and health
behavior risk factors for those causes of death The project answers key unresolved questions about the growing
geographic disparities in midlife mortality that have been major obstacles to understanding them. One question
regards the collective influence of state and local contexts. Studies tend to focus on state or local contexts,
providing an incomplete explanation. We advance this work by examining state and local contexts concomitantly,
which is critical because they may affect mortality via independent and synergistic processes. A second question
concerns the influence of states’ policy “contexts”. States have enacted highly correlated, or “bundled”, policies
which necessitates new approaches for understanding their influence on mortality. We advance this work by
using innovative methods to develop annual scores for interpretable policy bundles. A third question concerns
the degree to which state and local contexts collectively predict individual-level psychosocial wellbeing and
health behaviors—i.e., the proximate determinants of the four major causes of death behind rising midlife
mortality. Deindustrialization, declines in good jobs, and concomitant disruptions to families and communities in
some places may have harmed the psychosocial wellbeing of midlife adults, particularly those without a 4-year
college degree, leading to consumption of drugs, alcohol, and unhealthy food. We advance this work by
examining how state and local contexts collectively predict psychosocial and health behavior risk factors for the
four major causes of death. The project will accomplish its objective through three Specific Aims. Aim 1 identifies
how state policy and county economic contexts collectively predict county-level mortality (from all causes and
the four specific causes) from 1990-2025 for all midlife adults and by age, sex, race-ethnicity, education, and
metro status. Aim 2 identifies how state policy and county economic contexts collectively predict individual-level
psychosocial and health behavior risk factors from 2021-2025, and examines hypothesized pathways, among
all midlife adults and by age, sex, race/ethnicity, education, and metro status. Aim 3 merges the 100+ annual
state and county measures into multiple geocoded surveys available via the UM’s Virtual Data Enclave. We will
also disseminate the data, documentation, and code via ICPSR and offer workshops on these resources.