Setting Limits and their Relation to Well-being in End of Life Care
How to cite
Shapiro, Jeremy; Spitzer, Matt; Brauner, Daniel; Rees, Geoffrey. 2017. "Setting Limits and their Relation to Well-being in End of Life Care." AEA RCT Registry. AEARCTR-0002051. https://doi.org/10.1257/rct.2051-2.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
In 2011, $205 billion was spent on medical care for patients who were in their last year of life (Aldridge and Kelley, 2015). Medicare expenditures alone were responsible for 60% of this figure (Institute of Medicine, 2015). Given the magnitude of this expenditure and its expected increase (Lindgren, 2016) it is important to explore whether patient preferences for their own medical care at the end of life are satisfied. Though death at the end of every life is predictable with certainty, the death of each individual takes its own time, even as new medical technologies have made it possible for physicians to extend this time. Many of these technologies are high cost, and many also are emotionally, if not technically, difficult to stop once started. The result is that patients and their families are often asked to make difficult, sometimes excruciating decisions about limiting and ending intensive medical treatment as the time of death approaches. Added to these difficulties is the fact that the “end of life” is a time that is only known for certain retroactively.
To illuminate some of the difficult issues surrounding decisions concerning medical care at the end of life, our research focuses on two goals: (i) to understand patient preferences in specific medical scenarios and (ii) to understand public attitudes about government policy that could set limits on spending on care at the end of life. A common confound in the study of preferences and attitudes is that respondents may be influenced by social norms, or even the survey itself (e.g., framing and social desirability bias). Also, stated preferences do not always predict behaviors, thus respondents may not know or be able to fully articulate their underlying preferences. Thus, in contrast to most studies on patient attitudes, we augment simple questions pertaining to preferences with a randomization design intended to elicit subtle beliefs and views. Specifically, we present respondents with primes intended to highlight a certain aspect of EOLC (e.g., fear of death, opportunity cost) and observe how shifting the patient’s focus to the given topic affects their stated preferences.
Intervention
We present respondents with primes intended to highlight a certain aspect of EOLC (e.g., fear of death, opportunity cost) and observe how shifting the patient’s focus to the given topic affects their stated preferences.
Experimental design
Google survey with 3 different versions (based on informational primes) varied at the individualistic level.
Design details
Cited by
No citation data available for this registration yet.