Predictors of Intense Treatment in the Emergency Department Among Older Adults With Serious Life-Limiting Illnesses: A Five-Year Cross-sectional Analysis of Medicare Claims Data

BackgroundTreatment intensity of end-of-life care is the degree of aggressiveness of medical care aimed at providing life-prolonging medical care to patients with serious life-limiting illnesses. This study aims to assess the demographic and health characteristics associated with older adults with serious life-limiting illnesses who received highly intense end-of-life care. MethodsFor this cross-sectional analysis, we pooled the 2015 to 2019 Medicare claims data of adults 65 years and older who visited at least one of the 29 emergency departments (EDs) enrolled in the Primary Palliative Care for Emergency Medicine. We identified those with serious life-limiting illnesses using a Gagne score of seven or higher. Our outcome measure was treatment intensity, defined using acute care and intensive care unit (ICU) admissions. Acute care admission was measured as a binary variable and ICU admission was measured as a three-point nominal variable. The predictor variables were age, sex, race/ethnicity, and illness severity (Gagne score). To assess the odds of acute care and ICU admissions, we used a generalized estimating equation model and a multinomial regression model, respectively. We performed the same analyses among the population without serious life-limiting illnesses to observe differences in effect sizes of intense treatment. ResultsOf the 301,083 older adults that visited one of the 29 EDs, 13% had serious life-limiting illnesses. Age was associated with 9% and 7% increased odds of acute care (95% CI: 1.04 - 1.14) and ICU (95% CI: 1.02 - 1.12) admissions. We reported significant associations by sex, race/ethnicity, and illness severity (Gagne score). The effect sizes of the observed association between measures of treatment intensity and the demographic and health characteristics were smaller among those with serious life-limiting illnesses compared to those without serious life-limiting illnesses. ConclusionOlder adults with serious life-limiting illnesses who present to the ED experience intense treatment. Identifying demographic and health characteristics associated with treatment intensity may inform the need for serious illness conversations in the ED.