Meeting the Palliative Care Needs of Maintenance Hemodialysis Patients

In this issue of the Clinical Journal of the American Society of Nephrology, two studies highlight the importance of palliative care for patients undergoing maintenance hemodialysis. First, Chettiar et al. (1) examined outcomes of first hospitalizations in 2012–2013 within a United States cohort of patients on hemodialysis propensity matched for receipt of inpatient palliative care versus usual care. Among the 1308 patients who died during the hospitalization, those who received inpatient palliative care had a 21% shorter length of stay and 14% lower hospitalization cost than patients who received only usual care. Although these outcomes were similar among the 5024 patients who survived the hospitalization regardless of whether they received inpatient palliative care, those who did receive inpatient palliative care were eightfold more likely to have enrolled in hospice and 20% less likely to be readmitted in the 30 days after hospitalization. This study included all adults age 18 years old and older (mean age of roughly 66 years old across groups) but excluded patients who died within 90 days of initiating dialysis. Results would likely have been more striking if restricted to an older population and if they included early deaths, thus further underscoring the point: inpatient access to palliative care lessens intensity of care and cost.

Second, Chen et al. (2) examined dialysis withdrawal and palliative care utilization among a cohort of 536 patients on incident hemodialysis across an integrated group of eight dialysis units in the Midwestern United States who died between 2001 and 2015. The mean age of this cohort was 72 years old. Using narrative medical record review, they found that nearly one half withdrew from hemodialysis before death—an estimate more than twice that of estimates of prior studies that used registry data or death notification forms. These studies can have variable consistency and validity as shown in this study, where testing of a small random sampling found that Centers for Medicare and Medicaid Services 2746 (CMS-2746) ESKD Death Notification reporting compared with narrative medical record review had only 73% sensitivity for dialysis withdrawal. Given cultural differences in receptivity to advance care planning and hospice (3), the fact that this cohort was 94% white race likely contributed to this higher estimate. However, availability of outpatient palliative care services in this health care system was a likely contributor as well, because only about 20% of other institutions have outpatient palliative care services available (4). As the authors point out, outpatient hospice services are usually unavailable to patients with ESKD, because patients are required to forego dialysis to be eligible for the Medicare hospice benefit—unless they have a second terminal diagnosis unrelated to ESKD. Although California Senate Bill 1004 marks the first legislation mandating access to outpatient palliative care for patients with certain serious illnesses, ESKD is not one of them (5).

Chen et al. (2) also found that nearly twice as many patients who withdrew from dialysis had a palliative care consultation (34%) within 6 months before death compared with those who did not withdraw from dialysis (19%). Not that dialysis withdrawal is or should necessarily be a goal of end-of-life care planning, but what is necessary is that patients know that it is an option.

Like the study from Chettiar et al. (1), this study lends weight to the message that palliative care consultation is a good thing and ought to be available to all patients with ESKD. However, a looming question remains: how can we meet the palliative care needs of all patients with ESKD?

The question remains, because there simply are not enough palliative care specialists to take care of the palliative care needs of all patients with serious illness. According to the American Academy of Hospice and Palliative Medicine, there is only one palliative care specialist for every 20,000 adults living with a serious illness and <300 graduate from hospice and palliative medicine fellowships each year; therefore, this reality is not soon to change.

It has been suggested that nonpalliative care specialists should be able to attend to all but the most complex of their patients’ palliative care needs. Arguably, given the excess morbidity and mortality suffered by patients on maintenance hemodialysis and the financial incentives for these patients to be seen by a nephrologist and/or a nephrologist surrogate (e.g., physician assistant or nurse practitioner) four times a month, one would think that this would be a priority within the field and that there would be ample opportunities to carry it out. However, there are no guidelines regarding the content of these visits. Drive-by visits will do. In fact, Erickson et al. (6) found that, although a nationally representative database of patients on hemodialysis in the United States indicated at least four visits per month for more than two thirds in 2006 (as opposed to 15%–36% before visits were financially incentivized), patient factors (i.e., sicker patients are seen more often than stable ones) accounted for <1% of the variation in visit frequency. Furthermore, there has historically been little emphasis on palliative care training in nephrology fellowships; this was shown by two findings. It was exemplified by a 2003 survey, in which nephrology fellows reported that they had received little training on end-of-life issues and felt less prepared to take care of patients on dialysis at the end of life compared with other practice skills, and a similar survey of nephrology fellows 10 years later, which found nearly identical results (7,8).

The lack of time is usually stated as the primary reason why palliative care issues are not addressed. However, current financial incentives for initiating and maintaining patients on dialysis only encourage nephrologists to acquire and retain more patients on hemodialysis, thus further limiting time that can be spent with each patient. Although more dialysis patients means more income for nephrologists, patients cared for by nephrologists with large patient panels may suffer worse outcomes, such as mortality, dialysis adequacy, and receipt of kidney transplant, than patients cared for by nephrologists with smaller panels (9).

However, even if nephrologists were adequately trained in palliative care and attending to palliative care needs of patients was required and even financially incentivized, the nephrology workforce is not any better situated than the palliative care workforce to fill the void either. According to the American Society of Nephrology, roughly 40% of nephrology fellowship positions go unfilled, whereas the dialysis population continues to grow each year (10).

What are we nephrologists to do? Should we throw up our hands in defeat and accept that we cannot provide care that attends to our patients’ palliative care needs? I would argue that any nephrologists who find this acceptable should probably consider spending all of their time in a laboratory or perhaps switching over to a career in pathology. The rest of us need to roll up our sleeves and start turning our lemons into lemonade while we push lawmakers to make outpatient palliative care and hospice accessible to our patients and to align payment and quality of care policies with the needs of our patients. For example, the CMS should be pressured to make its Comprehensive ESKD Care Model truly comprehensive by expanding ESKD Seamless Care Organizations (through which dialysis clinics, nephrologists, and other providers together coordinate care for network beneficiaries) to include patients with CKD stage 5 who are not on dialysis. Such an expansion would remove the incentive to initiate and maintain patients on dialysis, and rather, it would lead to alignment with outpatient hospice and palliative care organizations, which in turn, could result in more cost-effective and better quality care.

Making lemonade would mean that, rather than just deferring to our palliative care colleagues whenever there is a tough conversation that needs to be had, we nephrologists instead collaborate with them so that they may teach us, our surrogates, our fellows, and our social workers how to address the palliative care needs of our patients. Furthermore, fellowship directors need to prioritize palliative care training right up there with calculating how much saline to give a patient who is hyponatremic.

Lastly, we need to learn how to teach nephrology concepts in a way that is accessible to medical students, interns, and resident physicians, thus making the field more attractive to those who may be more motivated to attend to the art of nephrology in addition to the math.

Disclosures

V.G. is supported by Cambia Health Foundation Sojourns Scholar Leadership Program and by grant R21DK112100 from the National Institute of Diabetes and Digestive and Kidney Disease (NIDDK, Bethesda, MD).