Disentangling the Roles of Disease Incidence and Case Fatality in the US Cardiovascular Mortality Stagnation

Marcus Ebeling , Max Planck Institute for Demographic Research
Neil Mehta, University of Texas Medical Branch
Leah Abrams, Tufts University
Mikko Myrskylä, Max Planck Institute for Demographic Research

Improvements in US mortality have slowed markedly since around 2010, with stagnation in cardiovascular disease (CVD) mortality emerging as one of the key factors. Most evidence to date has relied on cause-of-death data, which capture overall mortality trends but cannot disentangle the mechanisms driving changes in cause-specific death rates. Such rates can shift because more people develop disease (changing incidence) and/or because more people with disease die from it (changing case fatality). Each mechanism implies distinct underlying processes and points to different public health interventions. To clarify the drivers of CVD mortality stagnation, we examine how changes in incidence and case fatality have shaped mortality from myocardial infarction (MI) and stroke between 1988 and 2021. Using multiple cause-of-death data, the National Inpatient Sample - a nationally representative database of hospital admissions - and the Human Mortality Database, we estimate age-, sex-, and race/ethnicity-specific incidence, case fatality, and death rates. Counterfactual analyses will quantify the contributions of incidence and case fatality to overall mortality change and assess prospects for future improvements. The analysis will identify the periods and population groups in which each mechanism has dominated. Cross-national comparisons using Swedish registry data will provide additional context and validation. Beyond helping to disentangle the drivers of US mortality stagnation, this analysis will also provide methodological insights into the caveats and dynamics underlying changes in cause-of-death-specific mortality more generally. Understanding these mechanisms is crucial for developing effective prevention policies and for returning to mortality improvements in the future.

See extended abstract

 Presented in Session 27. Mortality and Longevity