[Skip to Content]
[Skip to Content Landing]
Invited Commentary
October 11, 2019

Stopping Dialysis, Good Deaths, and Social Justice

Author Affiliations
  • 1Adult and Child Consortium for Outcomes Research and Delivery Science, University of Colorado School of Medicine, Aurora
  • 2Division of Geriatric Medicine, Department of Medicine, University of Colorado School of Medicine, Aurora
  • 3VA Eastern Colorado Geriatric Research Education and Clinical Center, Denver
JAMA Netw Open. 2019;2(10):e1913110. doi:10.1001/jamanetworkopen.2019.13110

One of the greatest challenges in modern medicine is that there are so many things that can be done to help people live longer, such as dialysis. This amazing therapy effectively replaces the work of the kidney and allows people who would otherwise have died of kidney failure a chance to continue living. A major challenge, one not unique to dialysis, is knowing when it is the right time to stop pursuing these aggressive, organ-replacement therapies and instead pursue an aggressive approach focused on high-quality palliative and end-of-life care. In the study by Richards et al,1 the authors conducted an elegant analysis of multiple data sets to explore the association of stopping dialysis and receipt of hospice services with end-of-life care measures, such as quality of care. Overall, they found that only 28% of patients stopped dialysis prior to death. Of those, 58% received hospice services. Of the remaining 72% of patients who continued dialysis, only 18% received hospice services prior to death. Compared with patients who continued dialysis, families of patients who chose to stop dialysis were more likely to rate the quality of end-of-life care as excellent (46% vs 56%). Of patients who continued dialysis, choosing to receive hospice services was associated with a markedly higher proportion of family members rating the quality of care as excellent at the end of life compared with those who did not receive hospice services (61% vs 40%). Put another way, patients who continued dialysis and did not receive hospice services had lower family member ratings of quality of care at the end of life.

Readers should understand that linking all the disparate data sources used in the study by Richards et al1 was no small task. By doing this difficult work, the authors have given us a gift of highly credible evidence. However, 2 cautions should be considered in interpreting these results. First, the US Department of Veterans Affairs (VA) is quite different from non-VA health systems in terms of options at the end of life. Patients in the VA have the option of pursuing treatments, such as chemotherapy and dialysis, while they pursue hospice services. In contrast, patients in the Medicare fee-for-service system often must forgo expensive, life-prolonging therapies at the time they begin receiving hospice services simply because of the payment model. Second, and more importantly, the study by Richards et al1 provides evidence of an association, not causation. One should not conclude that stopping dialysis leads to better quality of care. Indeed, the fact that those who continued dialysis but received hospice services had the highest ratings of care at the end of life suggests there is a lack of a causal link. An alternative interpretation would be that people who stopped dialysis or received hospice services were very different than the larger group of people who continued dialysis and did not receive hospice services. This article highlights several important questions about death. Death is inevitable, but what is a good death?

A good death is an ideal that suffers from a lack of clarity and constantly shifting definitions. In the 1960s and 1970s, the term good death was synonymous with physician-assisted death or euthanasia.2 The dominant attitude was that decisions of life and death were solely the concern of the physician, and the patient had little or no agency. This paradigm began to shift in the 1980s when the definition of a good death began to evolve to include ideas of symptom management and patient preferences for treatment decisions.3 Though still underused, this shift directly coincided with the increasing use and popularity of hospice services in the United States. Yet, while we have embraced patient preferences in some respects, today’s unspoken paradigm of medicine instructs students to view death as a failure rather than a natural process that all of us are fated to experience. To add to the convolution, modern conceptions often oversimplify a good death as one that occurs at home with hospice services and a bad death as one that occurs in a health care facility.4 All permutations of a good death are heavily criticized, and rightly so.4

The finding in the study by Richards et al1 of higher quality-of-care rating at the end of life among families of patients who stopped dialysis or received hospice services does not tell us if these deaths were good. We will never know, because we cannot know what services those patients were offered and how those discussions occurred. This highlights the underlying problem with all definitions of a good death, which is that they fail to consider the heterogeneity of patients’ individual needs and desires. However, there is concerning evidence that the US health care system fails to navigate this challenge. First, use of hospice services is relatively low with only 50% of decedents receiving hospice services prior to death, and one-third of those patients receive hospice services for less than 7 days.5 Second, the study by Richards et al1 reported important differences between groups in that patients who stopped dialysis were less likely to be black or live in the South Atlantic region. Third, this same research group published another 2019 study6 in which they reviewed the health records of patients who chose not to start dialysis and found significant documentation that clinicians felt the patients were violating some kind of social norm, questioned their competency, and used repeated strategies to convince the patients to start dialysis.

While good death as currently envisioned may not be the best construct, rather than abandon the term, we should reimagine it as a social justice issue. John Rawls7 conceptualized social justice as each person having an equal right to the most extensive basic liberty compatible with a similar liberty for others. Considering a good death from a Rawlsian perspective would mean that all individuals are entitled to define what a good death means to them and possess the information, agency, and opportunity to choose what helps them achieve their definition of a good death. This is consistent with the fundamental American ideals of liberty, equality, and equity. To achieve this, clinicians must inform patients of their options and respect their decisions, and we must ensure that those options are available to them. For example, a black patient receiving dialysis in the southern United States may indeed choose to continue dialysis and avoid hospice services. However, if that choice was made because no one ever introduced the options of stopping dialysis or receiving hospice services, then that patient had no option to have a good death.

Back to top
Article Information

Published: October 11, 2019. doi:10.1001/jamanetworkopen.2019.13110

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Tate CE et al. JAMA Network Open.

Corresponding Author: Daniel D. Matlock, MD, MPH, Adult and Child Consortium for Outcomes Research and Delivery Science, University of Colorado School of Medicine, 13199 E Montview Blvd, Ste 300, Aurora, CO 80045 (daniel.matlock@cuanschutz.edu)

Conflict of Interest Disclosures: None reported.

Richards  CA, Hebert  PL, Liu  C-F,  et al.  Association of family ratings of quality of end-of-life care with stopping dialysis treatment and receipt of hospice services.  JAMA Netw Open. 2019;2(10):e1913115. doi:10.1001/jamanetworkopen.2019.13115Google Scholar
Kehl  KA.  Moving toward peace: an analysis of the concept of a good death.  Am J Hosp Palliat Care. 2006;23(4):277-286. doi:10.1177/1049909106290380PubMedGoogle ScholarCrossref
O’Neil  R.  Defining a good death.  Appl Philos. 1983;1(4):9-17.PubMedGoogle Scholar
Smith  AK, Periyakoil  VS.  Should we bury “the good death”?  J Am Geriatr Soc. 2018;66(5):856-858. doi:10.1111/jgs.15321PubMedGoogle ScholarCrossref
Medicare Payment Advisory Commission.  Report to the Congress: Medicare payment policy. http://medpac.gov/docs/default-source/reports/mar19_medpac_entirereport_sec.pdf?sfvrsn=0. Accessed September 3, 2019.
Wong  SPY, McFarland  LV, Liu  C-F, Laundry  RJ, Hebert  PL, O’Hare  AM.  Care practices for patients with advanced kidney disease who forgo maintenance dialysis.  JAMA Intern Med. 2019;179(3):305-313. doi:10.1001/jamainternmed.2018.6197PubMedGoogle ScholarCrossref
Rawls  J.  A Theory of Justice. Cambridge, MA: Belknap Press; 1971.
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    Who Do We Serve? Patients? Or Survivors?
    Andrew Kahr, Ph.D. | Harvard University
    There is a fundamental and for me fatal flaw here that cannot be palliated.

    We're drawing inferences based on what survivors say. Physicians might optimize ratings of survivors by "letting him die." To physicians, the fatally ill are patients with at least some of the rights accorded other patients. To prospective survivors, they can be a nuisance, an expense, an inconvenience--or worse.

    But physicians often spend more time talking to survivors than talking to patients! They are drawn to serving the survivors. "I understand this is a lot of expense." The dead
    won't bother you, but survivors can.

    Let's shine a spotlight on this. The interests and preferences of survivors don't necessarily coincide with those of patients. Often they are contrary to what patients want. Yet, for our own comfort--our own benefit--we habitually rely on what survivors want as if they wanted what patients want. They often don't.

    If you want your wishes carried out, you might try appointing someone other than a prospective survivor to make your healthcare decisions for you.