End of Life Care: Estimating the Effectiveness of 20 Years of Health System Reform

End of life care is intensive, expensive, and highly dependent on where one lives. This poses quality, economic, and equity concerns. Intensive care near the end of life (EOL) is a quality concern because most patients do not want intensive interventions at the EOL. It is an economic problem because 25 percent of health care spending is on the six percent of patients who die each year, yet intensive care at the EOL is not associated with better outcomes or greater satisfaction with care. Finally, it is an equity concern because the regional variation is substantial, and not driven by patient preferences.

EOL care for patients with Alzheimer’s disease and related dementias (ADRD) is particularly complex. Three things have to happen for less intense EOL care to be delivered, all of which pose higher hurdles for patients with ADRD. (1) Plan: individuals must state their preferences for care. This is especially important for patients with ADRD, who may lose the ability to consent near the EOL. (2) Communicate: providers must know those preferences in a timely manner. Again, this need is more acute when patients themselves reach a stage when they must rely on systems and caregivers to make their wishes known. (3) Facilitate: regulations and resources must allow those preferences to be honored.

The US health care system has been rapidly changing its financial incentives and system integration in order to improve care delivery for all patients, but to date, the role these system changes has on EOL care has been thus far under studied. Using a mix of claims, surveys, and a newly created database of health care system changes that focus on financial and system integration, this study will estimate the effectiveness of the last 20 years of innovation on the quality of EOL care. In addition to identifying the overall impact, we also will examine the effectiveness of these interventions for individuals with ADRD, because of the complex nature of the systems, the disease, and its predicted increasing prevalence. Finally, we will drill down and identify how EOL care is changing in response to these policy changes, planning vs. communication vs. facilitation, in order to identify where further interventions should be focused.

This project will help explain the geographic variation in EOL care, identify which of our policies have the most influence over patients and providers, and uncover important variation by disease. The impact of this work will be identifying where future efforts for behavior change should focus in order to match patients’ preferences to the care they receive at the EOL.