Inpatient Palliative Care for Neurological Disorders: Lessons from a Large Retrospective Series

Abstract

Background:

Patients with neurological disorders are the second most common patient population to receive inpatient palliative care services after cancer; however, there have been few reports describing the characteristics and needs of this population.

Objective:

To understand which patients with neurological diagnoses utilized these services, reasons for consultation, and whether patterns of use changed over time.

Design:

Retrospective case series.

Setting/Subjects:

Neurological patients seen by our academic hospital-based palliative care consult service from January 2010 through December 2014.

Measurement:

Demographics, diagnoses, reasons for consultation, patient characterization, and outcomes were extracted from medical record.

Methods:

Data extracted from the chart was analyzed using primarily descriptive statistics with Chi-square used to compare group proportions and regression to examine temporal trends.

Results:

A total of 3170 patients received inpatient palliative care consultations, of whom 291 (9.2%) had a neurological diagnosis. Among these patients, 85 were excluded because of insufficient data, leaving 206 patients (mean age 70.0 ± 16.0 years; 111 (54%) female). Of the patients, 49% had a chronic diagnosis and 51% received a new diagnosis during their hospitalization, most commonly ischemic stroke. The proportion of patients with chronic diagnoses increased over time, with dementia representing the most common diagnosis in 2014. The most common reason for consultation was eliciting goals of care (82%). Advance directives were present at admission in 61 (30%), 130 (63%) at the time of consultation, and 190 (92%) patients after consultation. The proportion of people with advance directives after consultation significantly increased over the study period (p = 0.01).

Discussion:

Hospital-based palliative care is primarily utilized to clarify goals of care in neurological patients and that advance directives were commonly not completed prior to consultation.

Conclusions:

Outpatient and hospital neurologists should routinely address advance directives, particularly as the proportion of patients with chronic neurological diagnoses increases.

Introduction

Palliative care addresses the suffering of patients with serious illness by providing expert symptom management, psychosocial support, and clarifying goals of care.¹ Palliative care has expanded in several important ways over the past two decades, expanding beyond a focus solely on cancer; engaging with patients and caregivers at any point in the disease course rather than being restricted to the end-of-life; and delivery in inpatient, outpatient, and home settings.^1,2 Availability of inpatient palliative care services has been rapidly expanding with an estimated 80% of U.S. hospitals with 50 or more beds now offering these services.³

To date, there have been few studies on the utilization of palliative care services by patients with neurological conditions, although neurological diagnoses are the second most common diagnostic category, following cancer, of patients accessing inpatient palliative care.^4,5 Chahine et al.⁴ reported the experience of the inpatient palliative care service at the Cleveland Clinic for 177 patients with neurological conditions seen between 2004 and 2007. The palliative care service was most commonly consulted for assistance with goals of care (including transition to hospice), comfort measures, and symptom management. Notable findings included a relatively high rate of patients with chronic neurological diagnoses (56%) and an inability to communicate (64%) with low rates of living wills (14%) and durable power of attorney (22%) at the time of admission. Other smaller studies have looked at the specific needs and outcomes of palliative care consultations for patients with stroke⁶ and dementia.⁷

Given the paucity of data, and particularly lack of recent studies, our primary objective for this study was to replicate and extend Chahine et al.'s⁴ findings in a cohort from a different center and region of the United States. Given trends for increased emphasis on palliative care in medical education,⁸ clinical neurology,¹ neurocritical care,⁹ and society at large,¹⁰ we were also interested to see whether there was any evidence for temporal change in the nature of palliative care consultations and completion of advance care planning documents in this population.

Methods

A retrospective cohort study was performed using the University of Colorado Palliative Care Consult Service clinical database to identify all patients carrying a neurological diagnosis seen by this service between January 2010 and December 2014. This study was approved by the Colorado Multiple Institutional Review Board.

The palliative care consult team collects data at the time of consultation for all patients seen by this service, including goals of care, advance directives, symptoms, prognosis, diagnosis, discharge planning, and reason for consultation. These data were verified and additional data were obtained through direct review of medical records, including reason for admission, length of admission, utilization of critical care services, living situation before admission, presence of advance directives, and discharge status. Patients who had missing values for one or more of the mentioned items were defined as insufficient data and excluded from analyses.

Global symptom burden was documented using a modified version of the Edmonton Symptom Assessment Scale, which measures the severity of several common symptoms.¹¹ Functional status was measured using the Palliative Performance Scale (PPS), which asks clinicians to rate overall patient status on a scale from 0% to 100% based on ambulation, activity levels, self-care capabilities, oral intake, and level of consciousness.¹² Higher scores indicate higher functionality. The PPS was designed to predict mortality rates but has only been evaluated in the hospice setting for patients with advanced dementia or stroke and has not been evaluated for other neurological conditions or for neurology patients in other settings (e.g., outpatient or hospital inpatients).¹³

Statistical analysis

All data outcomes collected were coded and reviewed by an independent researcher (YL) not part of the clinical care team. Analyses presented here are primarily descriptive. Chi-square or Fisher's exact test, as appropriate, was used for analysis of proportion difference among years for advance directives and diagnoses. Logistic and relative probability models were fit for continuous time effects. For testing for differences among groups alpha was set to 0.05. For assessing equivalence, the 95% confidence intervals (95% CIs) provided the boundaries of plausible difference.

Results

Patient characteristics and hospital course

During the 5-year study period, the palliative care consult team saw 3170 new consultations, of whom 291 (9.2%) had a primary neurological diagnosis associated with the consultation. Eighty-five patients were excluded because of insufficient data; 206 patients were included in analyses presented here. Mean patient age was 70.0 ± 16.0 years; 54% were women. Before admission, 36% of patients lived independently, 47% in assisted living, and 17% in a nursing home. Seventy-two percent of patients were unable to communicate at the time of consultation. All patients had caregivers or family present at the consultation visit. The mean PPS for this cohort was 19 ± 14 with 73% being at 20% or less. Patients had an average hospital stay of 15.3 ± 21.5 days; 38 patients (18%) were hospitalized for more than 20 days and 49% received critical care services. The most common disposition was death in the hospital (28%) followed by discharge to inpatient hospice (21%), discharge to home hospice (17%), discharge to hospice in a skilled nursing facility (8%), and home palliative care (7%), with less than 10% going home with or without home health support. Patient characteristics are summarized in Table 1.

Table 1.

Characteristics of Inpatients with Neurological Diseases Seen by Palliative Care Consult, 2010–2014 (N = 206)

Characteristic	Frequency	Percentage
Age (mean ± SD, range)	70 ± 16 (25–96)
Gender (female N, %)	111	54
Marital status
Married	96	47
Single	28	14
Widowed	47	23
Divorced	21	10
Unspecified	14	7
Living situation before admission
Not specified	1	0.5
Assisted living at home	96	47
Independent	74	36
Skilled nursing facility	35	17
ICU admission during hospitalization
No	106	51
Yes	100	49
Communication
Able	45	22
Unable	148	72
Partial	13	6
PPS
Mean (SD)	20 ± 14
10%	93	45
20%	27	13
30%	19	9
40%	14	7
50%	8	4
60%	4	2
>60%	0	0
Not specified	41	20
Discharge disposition
Death in hospital	58	28
Inpatient hospice	44	21
Home hospice	34	17
Hospice in SNF	17	8
Home palliative care	14	7
Palliative care in SNF	11	5
Rehabilitation in SNF	10	5
Home care	8	4
SNF no palliative care	8	4
Non specified	2	1

ICU, intensive care unit; PPS, palliative performance scale; SD, standard deviation; SNF, skilled nursing facility.

The most common neurological diagnosis among the 206 patients was acute stroke (49%; 44 ischemic stroke, 30 intracerebral hemorrhage, 15 subarachnoid hemorrhage, 8 subdural hemorrhage, and 4 ischemic stroke with hemorrhage transformation), followed by dementia in 41 (20%, 8 Alzheimer's disease and 33 unspecified dementia), tumor in 14 (7%), and Parkinson's disease in 13 (6%; Table 2). The proportion of patients with chronic neurological diagnoses increased over the study period (20% in 2010 to 34% in 2014) with a statistically significant evidence for association between dementia and year (p = 0.03), and dementia representing the most common neurological diagnostic category in 2014. The most common reasons for admission were acute hemorrhagic stroke in 52 (25%), acute ischemic stroke in 42 (20%), and pneumonia in 20 (10%). The most frequent symptom-related reasons for admission were altered mental status in 103 (50%), followed by weakness in 59 (29%), speech difficulty in 34 (17%), and pain in 32 (16%).

Table 2.

Diagnoses, Reasons for Admission, and Presenting Symptoms

	Frequency	Percentage
Neurological diagnosis
Stroke	101	49
Ischemic stroke	44	21
ICH	30	15
SAH	15	7
Ischemia stroke with hemorrhagic stroke Transformation	4	2
SDH	8	4
Dementia	41	20
Alzheimer's disease	8	4
Unspecified dementia	33	16
Nervous system tumor	14	7
Parkinson's disease	13	6
Nervous system injury or trauma	11	5
MS	7	3
Central nervous system infection	5	2
ALS	4	2
MD	2	1
Other	8	4
Reasons for admission
Stroke (bleeding)	52	25
Stroke (ischemia)	42	20
Pneumonia	20	10
Nervous system tumor	15	7
Urinary tract infection	14	7
Advanced dementia	12	6
Hypernatremia/hyponatremia	11	5
Sepsis	9	4
Respiratory failure	9	4
Nervous system infection	8	4
Presenting symptoms (top 6)
AMS	103	50
Weakness	59	29
Speech difficulty	34	17
Pain	32	16
Falling	27	13
Shortness of breath	24	12
Poor oral intake	11	5

ALS, amyotrophic lateral sclerosis; AMS, altered mental status; ICH, intracerebral hemorrhage; MD, myotonic dystrophy; MS, multiple sclerosis; SAH, subarachnoid hemorrhage; SDH, subdural hemorrhage.

Table 3 shows the global symptom burden in the subset of patients who were able to respond and for whom data were available (N = 36–84 depending on item). The most prevalent symptoms were tiredness (83%), drowsiness (58%), anorexia (57%), and delirium (50%), followed by pain (36%), shortness of breath (26%), secretions (26%), anxiety (24%), constipation (13%), nausea (10%), and depression (8%).

Table 3.

Modified Edmonton Symptom Assessment Scale in Inpatients with Neurological Diagnoses Seen by Palliative Care Consult (Able to Respond, 2010–2014)

	ESAS
Symptoms	None, n (%)	Mild, n (%)	Moderate n, (%)	Severe, n (%)	Total, n (%)	Prevalence (%)
Pain	35 (64)	11 (20)	9 (16)	0 (0)	55 (100)	36
Tiredness	8 (17)	20 (43)	17 (36)	2 (4)	47 (100)	83
Drowsiness	24 (42)	12 (21)	14 (25)	7 (12)	57 (100)	58
Depression	33 (92)	2 (5)	1 (3)	0 (0)	36 (100)	8
Anxiety	31 (76)	7 (17)	2 (5)	1 (2)	41 (100)	24
Nausea	35 (90)	4 (10)	0 (0)	0 (0)	39 (100)	10
Anorexia	19 (43)	11 (25)	9 (20)	5 (12)	44 (100)	57
Shortness of breath	34 (74)	9 (20)	3 (6)	0 (0)	46 (100)	26
Secretions^a	62 (74)	22 (26)			84 (100)	26
Constipation^a	62 (87)	9 (13)			71 (100)	13
Delirium^a	31 (50)	31 (50)			62 (100)	50

These items were only assessed as present or absent on modified form.

ESAS, Edmonton Symptom Assessment Scale.

Reasons for consultation and care goals

Table 4 summarizes reasons for palliative care consultation and care goals. The top three reasons for consultation were eliciting goals of care (167, 81%), psychosocial/spiritual support (82, 42%), and end-of-life transition (68, 33%). Comfort (145, 70%) was the most common care goal identified during the consultation, followed by spending quality time with family (82, 40%), and improving quality of life (42, 20%).

Table 4.

Reasons for Palliative Care Consultation and Care Goals

	Frequency	Percentage
Reasons for palliative care consultation
Eliciting goals of care	167	81
Psychosocial spiritual support	87	42
End-of-life transition	68	33
Nonpain symptoms	19	9
Advance care planning	7	3
Pain	7	3
Care goals
Comfort care	145	70
Spend quality time with family	82	40
Improve quality of life	42	20
Control symptoms	34	17
Die at home	26	13
Not to be a burden of family	15	12
Maintain dignity	21	10
Communication and education	16	8
Address spiritual concerns	11	5
Maximize length of life	8	4

Use of advance directives

At the time of admission, 61 out of 206 patients (30%) had advance directives (14 patients [7%] living will, 8 [4%] medical durable power of attorney [MDPOA], and 39 [19%] both living will and MDPOA). At the time of initial palliative care consultation, 130 out of 206 patients (63%) had advance directives and 190 (92%) had advance directives documented after consultation.

No statistically significant change was found among the proportions of people with living wills and/or MDPOA at the time of admission or time of initial consultation over the study period (p > 0.05). However, there was a statistically significant increase in proportion completed after palliative care consultations comparing years 2010 and 2014 (69 vs. 97%; p = 0.01). Similarly, a logistic regression model, with a linear year effect, found the odds of having a living will and/or MDPOA after consultation increased by an estimated 75.8% each year (odds ratio = 1.758, 95% CI [1.133–2.726], p = 0.008) and a relative risks model found an estimated increase of 6.11% in the proportion per year (relative risk = 1.061, 95% CI [1.012–1.112], p = 0.01).

Discussion

Similar to prior articles⁴ we found that patients with neurological diagnoses receiving inpatient palliative care consultations were severely ill and disabled with more than 70% being unable to communicate, more than 60% completely bedbound, and less than 10% returning to independent function at time of discharge. Although approximately half of these patients suffered an acute neurological event, mainly stroke, the other half suffered from chronic conditions and the proportion of patients suffering from chronic neurodegenerative illnesses increased over the study period and is projected to continue to increase over time.^14,15 These data suggest several opportunities to improve care for neurological patients both in and outside of the hospital setting and support recent efforts to routinely assess and address palliative care needs for these patients.¹⁶

Palliative care is best viewed as an approach to care for patients and families affected by serious illness rather than as a specific service or specialty.¹⁷ In this model, a patient's primary care or specialty physicians (e.g., neurologists) should be able to provide basic palliative care to patients and their families when appropriate and recognize situations in which specialist palliative care would be beneficial.¹⁸ For hospital-based neurologists, this would include providing patients and families with diagnostic and prognostic information in a compassionate manner, symptom management (e.g., confusion and dysphagia), and addressing advance directives.¹⁹ Our results suggest that neurologists are primarily consulting the palliative care service for establishing goals of care for patients nearing the end of life. Future study is needed to determine whether this reflects neurologists' comfort with other aspects of palliative care (e.g., symptom management) or lack of knowledge for what other services may be provided by the palliative care team. Further education may be needed for neurologists to provide palliative care services and appropriately refer patients for specialist palliative care.²⁰ Although palliative care education is required by the American College of Graduate Medical Education for neurology residency, many programs do not provide organized or didactic activities to meet these goals.²¹ Also notable is the finding that all of our patients had involvement of family at the time of the consultancy. Although complex family dynamics may be a reason for consulting palliative care, our data suggest that this was not a common reason for consultation. Rather it may be that patients with involved families, particularly patients with communication issues, are more likely to have palliative care support. This may be because of advocacy from families and has been documented in prior studies.²² Consultations should be collaborative with the neurology service staying involved as there are likely to be aspects of the neurological diagnosis and disease course that are less familiar to the palliative medicine specialists.²³

Our data document a trend toward increasing use of palliative care consultations for patients with chronic neurological illnesses. Although we cannot say with our data whether this reflects an overall increase in hospitalizations for this population, epidemiological studies would support this being a likely driving force.²⁴ As the rates of patients with chronic neurodegenerative illness continue to increase, there is a correspondingly heightened responsibility for outpatient neurological care to also address palliative care issues such as providing information regarding the prognosis and progression of neurodegenerative illnesses to allow patients and families an opportunity to plan and prepare for the future.²⁵ Addressing advance directives as a routine part of outpatient care is also critical as waiting for the right time, or signs of decline often mean that advance directives are not completed or are completed in an emergent situation (e.g., during hospitalization). Our data highlight the importance of having these conversations early in neurological populations, in which loss of cognitive and communication abilities is common and precludes meaningful input later in the disease course. Providers should be reassured that studies of patients and caregivers support early rather than late completion of advance directives.²⁶

Hospitalizations for patients with neurodegenerative illnesses may portend a worsening of functional status and poor prognosis and should trigger consideration of advance care planning discussions. Recurrent hospitalizations, particularly for bladder infections and pneumonia, can in fact be used to support referral to hospice per Medicare guidelines and are associated with increased mortality rates in dementia and Parkinson's disease.^27,28 Although this is a retrospective study and highly selective cohort, the high mortality rates seen among our patients with chronic neurological illnesses are consistent with these data. Although not well studied, there is some evidence that proactive use of advance directives may prevent hospital and intensive care unit stays and could thus both reduce health utilization and prevent unnecessary hospitalizations and hospital deaths, both of which may be traumatic to patients and caregivers.²⁹

Regarding advance directives, overall rates were low at the time of admission (30%) and fair at the time of consultation (63%). Although these rates did not statistically change over the course of the study, they were higher than seen in older studies and suggest the possibility that proactive use of advance care plans is improving.^4,30 In contrast, rates of advance directive documentation after palliative care consultation improved significantly and approached 100% by the end of the study. This change occurred after a quality improvement (QI) initiative of the inpatient palliative care service and suggests that similar QI initiatives could be successful if initiated by neurology services.

As a retrospective case series from a single tertiary care center, this work has several notable limitations, including lack of a comparison group, limitations of data to what was documented on the hospital chart, omission of brain cancer patients (who were more often identified as cancer and not neurology in the palliative care database), and potential biases regarding which patients were referred for palliative medicine consultations. Importantly, data were not available to compare patients receiving a palliative care consultation with other neurology patients not receiving such a consultation. There is likely to be referral biases toward more advanced and/or complex patients receiving palliative care consultations, limiting the generalizability of these data. Although it is reassuring that our results are similar to a prior study from the Cleveland Clinic, it is possible that these results would be different if performed in a community hospital or setting outside of the United States.⁴

This study supports the need for palliative care approaches for patients with acute and chronic neurological conditions in both hospital and outpatient-based settings. Further studies are needed to better define the needs of specific populations and to test the effectiveness of therapeutic strategies for meeting the palliative care needs of these populations. Studies are also needed to determine the efficacy of educational and QI initiatives to change practice patterns and improve patient and caregiver outcomes.

Footnotes

Author Disclosure Statement

No competing financial interests exist. S.S. conducted statistical analysis.

References

Boersma

, Miyasaki

, Kutner

, et al.: Palliative care and neurology: Time for a paradigm shift. Neurology, 2014; 83:561–567.

Strand

, Kamdar

, Carey

: Top 10 things palliative care clinicians wished everyone knew about palliative care. Mayo Clin Proc, 2013; 88:859–865.

Morgan

: Palliative Care Growth Trend Continues, According to Latest Center to Advance Palliative Care Analysis. 2014; www.capc.org/about/press-media/press-releases/2014-9-2/palliative-care-growth-trend-continues-according-latest-center-advance-palliative-care-analysis. (Last accessed May 12, 2015).

Chahine

, Malik

, Davis

: Palliative care needs of patients with neurologic or neurosurgical conditions. Eur J Neurol, 2008; 15:1265–1272.

Gott

, Gardiner

, Ryan

, et al.: Prevalence and predictors of transition to a palliative care approach among hospital inpatients in England. J Palliat Care, 2013; 29:147–153.

Mazzocato

, Michel-Nemitz

, Anwar

, et al.: The last days of dying stroke patients referred to a palliative care consult team in an acute hospital. Eur J Neurol, 2010; 17:73–77.

Campbell

, Guzman

: A proactive approach to improve end-of-life care in a medical intensive care unit for patients with terminal dementia. Crit Care Med, 2004; 32:1839–1843.

Dickinson

: Thirty-five years of end-of-life issues in US medical schools. Am J Hosp Palliat Care, 2011; 28:412–417.

Tran

, Back

, Creutzfeldt

: Palliative care consultations in the neuro-ICU: A qualitative study. Neurocrit Care, 2016; 25:266–272.

10.

Schwartz

: Is this palliative care's moment? Science of Caring. 2012; http://scienceofcaring.ucsf.edu/acute-and-transitional-care/palliative-cares-moment. (Last accessed May 12, 2015).

11.

Chang

, Hwang

, Feuerman

: Validation of the Edmonton Symptom Assessment Scale. Cancer, 2000; 88:2164–2171.

12.

Anderson

, Downing

, Hill

, et al.: Palliative Performance Scale (PPS): A new tool. J Palliat Care, 1996; 12:5–11.

13.

Harrold

, Rickerson

, Carroll

, et al.: Is the palliative performance scale a useful predictor of mortality in a heterogeneous hospice population?. J Palliat Med, 2005; 8:503–509.

14.

Mayeux

: Epidemiology of neurodegeneration. Annu Rev Neurosci, 2003; 26:81–104.

15.

Reitz

, Brayne

, Mayeux

: Epidemiology of Alzheimer disease. Nat Rev Neurol, 2011; 7:137–152.

16.

Casaubon

, Boulanger

, Glasser

, et al.: Canadian stroke best practice recommendations: Acute inpatient stroke care guidelines, update 2015. Int J Stroke, 2016; 11:239–252.

17.

Phillips

: A service versus an approach: The importance of building primary palliative care. Int J Palliat Nurs, 2014; 20:471.

18.

Creutzfeldt

, Robinson

, Holloway

: Neurologists as primary palliative care providers: Communication and practice approaches. Neurol Clin Pract, 2016; 6:40–48.

19.

Frontera

, Curtis

, Nelson

, et al.: Integrating palliative care into the care of neurocritically ill patients: A report from the improving palliative care in the ICU project advisory board and the center to advance palliative care. Crit Care Med, 2015; 43:1964–1977.

20.

Schuh

, Biondo

, An

, et al.: Neurology resident learning in an end-of-life/palliative care course. J Palliat Med, 2007; 10:178–181.

21.

Weissman

, Block

: ACGME requirements for end-of-life training in selected residency and fellowship programs: A status report. Acad Med, 2002; 77:299–304.

22.

Sudore

, Casarett

, Smith

, et al.: Family involvement at the end-of-life and receipt of quality care. J Pain Symptom Manage, 2014; 48:1108–1116.

23.

Manu

, Marks

, Berkman

, et al.: Self-perceived competence among medical residents in skills needed to care for patients with advanced dementia versus metastatic cancer. J Cancer Educ, 2012; 27:515–520.

24.

Low

, Ben-Shlomo

, Coward

, et al.: Measuring the burden and mortality of hospitalisation in Parkinson's disease: A cross-sectional analysis of the English Hospital Episodes Statistics database 2009–2013. Parkinsonism Relat Disord, 2015; 21:449–454.

25.

Holloway

, Gramling

, Kelly

: Estimating and communicating prognosis in advanced neurologic disease. Neurology, 2013; 80:764–772.

26.

Bernacki

, Block

, American College of Physicians High Value Care Task Force: Communication about serious illness care goals: A review and synthesis of best practices. JAMA Intern Med, 2014; 174:1994–2003.

27.

Fall

, Saleh

, Fredrickson

, et al.: Survival time, mortality, and cause of death in elderly patients with Parkinson's disease: A 9-year follow-up. Mov Disord, 2003; 18:1312–1316.

28.

Hsiao

, Peng

, Wen

, et al.: Care needs and clinical outcomes of older people with dementia: A population-based propensity score-matched cohort study. PLoS One, 2015; 10:e0124973.

29.

Tuck

, Zive

, Schmidt

, et al.: Life-sustaining treatment orders, location of death and co-morbid conditions in decedents with Parkinson's disease. Parkinsonism Relat Disord, 2015; 21:1205–1209.

30.

Jaren

, Selwa

: Causes of mortality on a university hospital neurology service. Neurologist, 2006; 12:245–248.