Abstract 161: Trends In Time-Incremental Mortality After Hospitalization For Incident Acute Myocardial Infarction Among Medicare Beneficiaries, 2008-2017

Abstract

Background: New medical technologies and advancements in care quality have transformed care of patients hospitalized with incident acute myocardial infarction (AMI). While some advances may impact short-term mortality (e.g., mechanical circulatory support), others may take much longer to impact mortality (e.g., cardiac rehabilitation). Little is known about trends in time-incremental AMI mortality rates. Methods: We identified all patients hospitalized with incident AMI using a 5% sample of Medicare fee-for-service beneficiaries from 2008 to 2017. We used a three-year look back to exclude patients with a prior AMI hospitalization. We calculated unadjusted in-hospital, 30-day, 1-year, 2-year, and 3-year mortality rates by dividing the total number of all-cause deaths during each period by the corresponding number of patients who were hospitalized. We used logistic regression models to adjust mortality rates for patient demographics (age, sex, race/ethnicity) and a variety of comorbid conditions. Using 2008 as the baseline, we calculated annual incremental mortality rates for each time period. We performed subgroup analyses among AMI patients with ST-segment and non-ST-segment elevation. Results: From 2008 to 2017, we identified 42,567 patients hospitalized with incident AMI (mean age 81 years, 44% male, 86% white). The largest incremental declines in mortality rates were observed for the in-hospital (2.9% [CI 2.7 to 3.1%]), 30-day (2.1% [CI 1.9 to 2.3%], and 1-year (2.9% [CI 2.6 to 3.2%] time periods. Mortality declines were smaller but persisted for the 2-year (1.5% [CI 1.2 to 1.8%]), and 3-year (0.5% [CI 0.2 to 0.8%]; Figure) time periods. Similar patterns were observed among both ST-segment and non-ST-segment elevation AMI patients. Conclusions: For patients hospitalized with incident AMI, advances in care have led to incremental declines in mortality across time periods spanning the hospital stay through 3 years after discharge.