The n e w e ng l a n d j o u r na l of m e dic i n e

s ounding b oa r d

Early Specialty Palliative Care — Translating Data

in Oncology into Practice
Ravi B. Parikh, A.B., Rebecca A. Kirch, J.D., Thomas J. Smith, M.D., and Jennifer S. Temel, M.D.

Palliative care suffers from an identity problem. with multiple sclerosis4 and congestive heart
Seventy percent of Americans describe them- failure.5,6
selves as “not at all knowledgeable” about pallia- Although there are salient differences between
tive care, and most health care professionals hospice care and palliative care, notably the
believe it is synonymous with end-of-life care.1 limitations on prognosis and use of curative
This perception is not far from current medical therapies with hospice care, most palliative care
practice, because specialty palliative care — ad- is currently provided at the end of life. This per-
ministered by clinicians with expertise in pallia- ceived association between palliative care and
tive medicine — is predominantly offered through end-of-life care has led to a marginalization of
hospice care or inpatient consultation only after palliative care.1 Debates over “death panels,”
life-prolonging treatment has failed. Limiting physician-assisted suicide, and reimbursement
specialty palliative care to those enrolled in hos- for advance care planning have made policymakers
pice or admitted to the hospital ignores the ma- reluctant to devote resources to initiatives per-
jority of patients facing a serious illness, such as ceived to be associated with “death and dying.”
advanced cancer, who have physical and psycho- For example, National Institutes of Health alloca-
logical symptoms throughout their disease. To en- tions for research focused on palliative care re-
sure that patients receive the best care through- main far behind funding for procedure-oriented
out their disease trajectory, we believe that palliative specialties.7 The practice and policy behind palli-
care should be initiated alongside standard med- ative care must be considered independently from
ical care for patients with serious illnesses. end-of-life care. Palliative care should no longer be
For palliative care to be used appropriately, reserved exclusively for those who have exhausted
clinicians, patients, and the general public must options for life-prolonging therapies (Fig. 1).8
understand the fundamental differences between We present three separate cases — clinical,
palliative care and hospice care. The Medicare economic, and political — focused predomi-
hospice benefit provides hospice care exclusively nantly on data in patients with advanced cancer
to patients who are willing to forgo curative to show the value of earlier specialty palliative
treatments and who have a physician-estimated care. We then use these data to propose initial
life expectancy of 6 months or less.2 In contrast, priorities for clinicians and policymakers to
palliative care is not limited by a physician’s es- achieve early integration of palliative care across
timate of life expectancy or a patient’s prefer- all populations with serious illness.
ence for curative medication or procedures. Ac-
cording to a field-tested definition developed by THE CLINIC AL C A SE
the Center to Advance Palliative Care and the
American Cancer Society, “Palliative care is ap- Several randomized studies involving patients
propriate at any age and at any stage in a seri- with advanced cancer show that integrating spe-
ous illness, and can be provided together with cialty palliative care with standard oncology care
curative treatment.”1 Several clinical trials have leads to significant improvements in quality of
shown benefits of early specialty palliative care life and care and possibly survival (Table 1).6,9-12
in patients with advanced cancer.3 The effect of Patients with advanced cancer who receive pal-
early specialty palliative care in other patient liative care consultations early in the course of
populations is less well studied, but there are their disease report better symptom control than
data suggesting a beneficial role in patients those not receiving consultations.11,12 Several

n engl j med 369;24 nejm.org december 12, 2013 2347

The New England Journal of Medicine
Downloaded from nejm.org on December 13, 2013. For personal use only. No other uses without permission.
Copyright © 2013 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

nesses, for whom ensuring the best quality of

Traditional Palliative Care life and care is paramount. Nevertheless, it is
Palliative care
necessary to consider the financial consequences
to manage of serious illness, because 10% of the sickest
Life-prolonging or curative treatment symptoms and
improve quality
Medicare beneficiaries account for nearly 60% of
of life total program spending.18 The growing cost of
Diagnosis Death hospital care is the main driver of the spending
growth observed for seriously ill patients.19 For-
tunately, the quality improvements offered by
Early Palliative Care
early specialty palliative care may also lead to
Life-prolonging or curative treatment lower total spending on inpatient health care.20
Hospitals with specialty palliative care services
have decreased lengths of stay, admissions to
Palliative care to manage symptoms and improve quality of life
the intensive care unit, and pharmacy and labo-
Diagnosis Death
ratory expenses.9,21-23 One study estimated that
Figure 1. Traditional versus Early Palliative Care. inpatient palliative care consultations are associ-
In the traditional care model, palliative care is instituted only after life-­ ated with more than $2,500 in net cost savings
prolonging or curative treatment is no longer administered. In the integrated per patient admission.23
model, both palliative care and life-prolonging care are provided throughout Similarly, outpatient palliative care services
the course of disease. Adapted from the Institute of Medicine.8 have been estimated to reduce overall treatment
costs for seriously ill patients by up to 33% per
patient.6 Early outpatient palliative care achieves
prospective trials have also shown that early pal- these savings by decreasing the need for acute
liative care improves patients’ quality of life.10-12 care services, leading to fewer hospital admis-
For example, patients with metastatic lung can- sions and emergency department visits.11,24 The
cer who receive outpatient palliative care from site of death may be another mediator of sav-
the time of diagnosis and throughout the course ings, because patients receiving early specialized
of their illness report better quality of life and palliative care are more likely to forgo costly in-
lower rates of depression than do controls.11,13 patient care at the end of life than are other pa-
Initiating palliative care upon diagnosis of tients.6 Outpatient palliative care may thus low-
advanced cancer also improves patients’ under- er health care spending by reducing patients’
standing of their prognosis.14 Patients with seri- need for hospital and acute care. The goal of
ous illness often feel that their doctors do not early palliative care, both in and out of the hos-
provide all available information about their ill- pital, is to provide a better quality of life; cost
ness and treatment options.1 These information savings through reduced resource use are an
gaps can lead patients to misunderstand their epiphenomenon of this better care.
treatment goals. For example, recent studies show
that the majority of patients with metastatic THE P OLITIC AL C A SE
cancer incorrectly report that their cancer can
be cured with chemotherapy or radiation.15,16 These data show that earlier specialized pallia-
Palliative care clinicians can remedy this situa- tive care services meet the “triple aim” of better
tion by helping patients develop a more accurate health, improved care, and lower cost.25 Despite
assessment of their prognosis.14 Improved prog- such positive outcomes, legislative efforts to
nostic understanding may explain why patients support the delivery of palliative care have lagged
with advanced cancer who receive early pallia- behind clinical interest. High-profile and contro-
tive care consultations are less likely to receive versial legal cases, such as those of Terry Schiavo
chemotherapy near the end of life than are con- and Dr. Jack Kevorkian, have heightened public
trols.14,17 sensitivity about medical care perceived to has-
ten death. Similarly, the inflammatory language
THE ECONOMIC C A SE surrounding “death panels” that surfaced during
the Affordable Care Act debate left legislators
Cost savings are never the primary intent of pro- wary of addressing policies perceived as promot-
viding palliative care to patients with serious ill- ing end-of-life care.

2348 n engl j med 369;24 nejm.org december 12, 2013

The New England Journal of Medicine

Downloaded from nejm.org on December 13, 2013. For personal use only. No other uses without permission.
Copyright © 2013 Massachusetts Medical Society. All rights reserved.
 sounding board

Table 1. Randomized Trials of Early Specialty Palliative Care Interventions in Patients with Cancer.

Trial Population Intervention Results

6
Brumley et al. 298 homebound patients with a prog- Usual care + in-home multidisci­ Patients assigned to PC had lower rates of
nosis of <1 yr to live and a recent plinary PC (frequency of visits ED visits (P = 0.01) and hospital admis-
hospital or ED visit; included based on individual needs of sions (P<0.001) and lower medical
138 patients with cancer patients) vs. usual care costs (difference in mean cost, $7,552;
P = 0.004) and were more likely to die at
home (P<0.001). There was no signifi-
cant between-group difference in hos-
pice enrollment.
Gade et al.9 517 patients with ≥1 life-limiting Usual care + inpatient multidisci­ Patients receiving PC reported more satis-
­diagnosis and their physician plinary PC consultation vs. faction with care (P<0.001), had fewer
“would not be surprised” if the ­usual care ICU stays on hospital readmission
­patient died ≤1 yr; included (P = 0.04), and had a 6-mo net cost sav-
159 patients with cancer ings of $4,855 per patient (P = 0.001).
There were no significant between-
group differences in hospice use, com-
pletion of advanced directives, symp-
toms and quality of life, or survival.
Bakitas et al.10 322 patients with a life-limiting can- Usual care + phone-based PC ad- Patients assigned to PC reported better
cer and a prognosis of approxi- ministered by advanced-prac- quality of life (P = 0.02) and mood
mately 1 yr to live tice nurse in 4 structured ses- (P = 0.02). There were no significant be-
sions and at least monthly fol- tween-group differences in symptom
low-up vs. usual care burden or intensity of service (hospital
and ICU days or number of ED visits).
Temel et al.11 151 patients within 8 wk after diag­ Usual care + outpatient PC (provid- Patients receiving early PC had better qual­
nosis of metastatic lung cancer ed by physician or advanced- ity of life (P = 0.03), lower rates of depres-
practice nurse) at least monthly sion (P = 0.01), less aggressive end-of-life
and PC consultation if patient care (P = 0.05), and longer median survival
hospitalized vs. usual care (P = 0.02).
Zimmermann 442 patients with metastatic cancer Usual care + early ambulatory PC at Patients receiving early PC reported greater
et al.12 and a physician-provided prog­ least monthly vs. usual care satisfaction with care (P<0.001), better
nosis of 6 mo to 2 yr to live with routine PC quality of life (P = 0.008), and less severe
symptoms (P = 0.05) at 4 mo.

* ED denotes emergency department, ICU intensive care unit, and PC palliative care.

However, policy momentum is now building, that emphasizes it as “an extra layer of support”
bolstered by evidence establishing the quality- that is appropriate at “any stage in a serious ill-
of-life benefit of palliative care for patients with ness.”1 Advocacy groups, practitioners, and re-
advanced cancer. Federal legislative proposals, searchers should use this language consistently
including the Patient Centered Quality Care for to advance this effort to integrate palliative care
Life Act and the Palliative Care and Hospice Ed- earlier in illness.
ucation and Training Act, have built bipartisan
support for federal and state legislation that ad- SOLU TIONS TO MAKE THE TR ANSITION
dresses palliative care research, the palliative care
workforce, and barriers to accessing care. These Although data to date support the use of early
efforts foreshadow more legislative initiatives specialty palliative care for patients with ad-
that prioritize quality of life and survivorship. vanced cancer, the clinical and economic bene-
Although legislation is a key step toward fits are likely to apply to other patient popula-
changing policy regarding palliative care, the tions. Randomized trials of early palliative care
main impediment remains a matter of messag- have shown benefit for patients with chronic ob-
ing. Reframing the policy and professional dis- structive pulmonary disease, congestive heart
cussion around palliative care as a means to im- failure, and multiple sclerosis.4,6,9 Further inves-
prove quality of life without decreasing survival tigation of the role of early specialty palliative
is essential to make this advocacy agenda more care in patients with other serious illnesses is
politically tenable. More than 90% of Americans clearly warranted. In addition, all clinicians car-
react favorably to a definition of palliative care ing for patients with serious illness, not just pal-

n engl j med 369;24 nejm.org december 12, 2013 2349

The New England Journal of Medicine
Downloaded from nejm.org on December 13, 2013. For personal use only. No other uses without permission.
Copyright © 2013 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

liative care specialists, must be capable of prac- end of life. Instead, we believe that health pro-
ticing “primary palliative care,” which includes fessional schools should establish content areas
managing illness- and treatment-related symptoms in palliative care during the preclinical and clin-
to improve quality of life and assessing treatment ical years and train students in managing symp-
preferences and prognostic understanding.26 toms, providing psychosocial support, and dis-
cussing prognosis and treatment preferences for
Incentive Changes all seriously ill patients. Furthermore, lawmakers
To reinforce the practice of early palliative care should adjust the current cap on training posi-
for all serious illnesses, hospitals, insurance pro- tions in graduate medical education and increase
viders, and the government would need to pro- funding for fellowship programs in palliative
vide practice and payment incentives for clini- care to expand the palliative care workforce.
cians. Medicare reimbursement for clinicians to
counsel patients about their goals and options Expanding Hospital-Based Palliative Care Teams
for care throughout their illness is necessary to Integrated palliative care requires patients to
encourage and reinforce early palliative care. Un- have access to palliative care services in the in-
fortunately, congressional efforts to reimburse patient and outpatient settings, across both the
for this service have been unsuccessful. Hospital acute and chronic phases of disease. Although
administrators have also identified several barri- hospital-based palliative care teams improve
ers to implementing consultative specialty palli- quality of care while reducing inpatient costs,
ative care teams, including limited institutional their prevalence varies considerably according to
budgets, poor reimbursement, and few trained geographic region and is quite low in some loca-
staff.27 Although public awareness of the clinical tions. Among adult-care hospitals with 50 or
benefits of palliative care may itself drive hospi- more beds, the statewide prevalence of inpatient
tal-level integration, increased reimbursement palliative care teams ranges from 20 to 100%
would most strongly convince hospitals and phy- across the United States.32 Small, for-profit, and
sicians to integrate primary and specialty pallia- public hospitals are far less likely to have pallia-
tive care into routine practice. tive care teams than large and nonprofit institu-
More broadly, reimbursement structures should tions.22 Hospital leaders should ensure that all
encourage coordinated medical care that aligns hospitals have access to integrated palliative care
treatments with patients’ goals. Health care sys- services within the next decade. Data suggest
tems that provide structured palliative care ser- that this trend has already begun: more than
vices in coordination with disease-centered treat- half of administrators at major cancer centers
ment have enjoyed tremendous success. The Aetna plan to increase palliative care professional re-
Compassionate Care Program of early nurse- cruitment in the short term.27 The American
managed palliative care and advanced care plan- Hospital Association and Center to Advance Pal-
ning alongside usual care has decreased hospi- liative Care have released guidelines advocating
tal lengths of stay and admissions while the use of specialist palliative care services for
decreasing costs at the end of life by 22%.28,29 the management of complex conditions in inpa-
The success of such initiatives should convince tient settings.33 These efforts, coupled with
Medicare and commercial insurers to reimburse strong external incentives such as Medicare
for palliative care services regardless of progno- Conditions of Participation and Joint Commis-
sis and treatment goals. sion accreditation requirements, will reinforce
hospital penetration of palliative care.
Educational Reform
Dedicated clinical exposure to seriously ill pa- CONCLUSIONS
tients, in combination with structured didactic
teaching, improves medical students’ attitudes Early provision of specialty palliative care im-
toward palliative care.30 A study based on survey proves quality of life, lowers spending, and helps
data from 1998 through 2006 from the Associa- clarify treatment preferences and goals of care
tion of American Medical Colleges showed for patients with advanced cancer. However,
greater student exposure to palliative care train- widespread integration of palliative care with
ing over the past decade.31 However, current cur- standard medical treatment remains unrealized,
ricula generally focus exclusively on care at the and more evidence is needed to show the poten-

2350 n engl j med 369;24 nejm.org december 12, 2013

The New England Journal of Medicine

Downloaded from nejm.org on December 13, 2013. For personal use only. No other uses without permission.
Copyright © 2013 Massachusetts Medical Society. All rights reserved.
 sounding board

tial gains of early palliative care in other popula- 13. Pirl WF, Greer JA, Traeger L, et al. Depression and survival
in metastatic non-small-cell lung cancer: effects of early pallia-
tions. This will require improved public and pro- tive care. J Clin Oncol 2012;30:1310-5.
fessional awareness of the benefits of palliative 14. Temel JS, Greer JA, Admane S, et al. Longitudinal percep-
care and coordinated action from advocacy tions of prognosis and goals of therapy in patients with meta-
static non-small-cell lung cancer: results of a randomized study
groups, health professionals, educators, and of early palliative care. J Clin Oncol 2011;29:2319-26.
policymakers. Patients who access earlier spe- 15. Weeks JC, Catalano PJ, Cronin A, et al. Patients’ expectations
cialty palliative care have better clinical out- about effects of chemotherapy for advanced cancer. N Engl J
Med 2012;367:1616-25.
comes at potentially lower costs — a compelling 16. Chen AB, Cronin A, Weeks JC, et al. Expectations about the
message for providers, policymakers, and the effectiveness of radiation therapy among patients with incurable
general public. lung cancer. J Clin Oncol 2013;31:2730-5.
17. Greer JA, Pirl WF, Jackson VA, et al. Effect of early palliative
Disclosure forms provided by the authors are available with care on chemotherapy use and end-of-life care in patients with
the full text of this article at NEJM.org. metastatic non-small-cell lung cancer. J Clin Oncol 2012;30:394-
400.
From Harvard Medical School (R.B.P.) and Massachusetts 18. Medicare spending and financing fact sheet. Menlo Park, CA:
General Hospital (J.S.T.) — both in Boston; the American Can- Kaiser Family Foundation, 2012 (http://www.kff.org/medicare/
cer Society, Washington, DC (R.A.K.); and Johns Hopkins Uni- upload/7305-07.pdf).
versity, Baltimore (T.J.S.). Address reprint requests to Dr. Temel 19. Meier DE, Isaacs SL, Hughes R, eds. Palliative care: trans-
at the Division of Hematology/Oncology, Massachusetts Gen- forming the care of serious illness. San Francisco: Jossey-Bass,
eral Hospital, 55 Fruit St., Yawkey 7B, Boston, MA 02114, or at 2010.
[email protected]. 20. Meier DE. Increased access to palliative care and hospice
services: opportunities to improve value in health care. Milbank
1. 2011 Public opinion research on palliative care: a report Q 2011;89:343-80.
based on research by public opinion strategies. New York: Center 21. Ciemins EL, Blum L, Nunley M, Lasher A, Newman JM. The
to Advance Palliative Care, 2011 (http://www.capc.org/tools-for economic and clinical impact of an inpatient palliative care con-
-palliative-care-programs/marketing/public-opinion-research/ sultation service: a multifaceted approach. J Palliat Med 2007;
2011-public-opinion-research-on-palliative-care.pdf). 10:1347-55.
2. Medicare hospice benefits. Baltimore: Centers for Medi- 22. Goldsmith B, Dietrich J, Du Q, Morrison RS. Variability in
care and Medicaid Services, 2013 (http:// www.medicare.gov/ access to hospital palliative care in the United States. J Palliat
publications/pubs/pdf/02154.pdf). Med 2008;11:1094-102.
3. Greer JA, Jackson VA, Meier DE, Temel JS. Early integration 23. Morrison RS, Penrod JD, Cassel JB, et al. Cost savings asso-
of palliative care services with standard oncology care for pa- ciated with US hospital palliative care consultation programs.
tients with advanced cancer. CA Cancer J Clin 2013 July 15 (Epub Arch Intern Med 2008;168:1783-90.
ahead of print). 24. Brumley RD, Enguidanos S, Cherin DA. Effectiveness of a
4. Higginson IJ, Costantini M, Silber E, Burman R, Edmonds P. home-based palliative care program for end-of-life. J Palliat Med
Evaluation of a new model of short-term palliative care for peo- 2003;6:715-24.
ple severely affected with multiple sclerosis: a randomised fast- 25. The IHI Triple Aim initiative. Cambridge, MA: Institute for
track trial to test timing of referral and how long the effect is Healthcare Improvement, 2013 (http://www.ihi.org/offerings/
maintained. Postgrad Med J 2011;87:769-75. Initiatives/TripleAim/Pages/default.aspx).
5. Hauptman PJ, Havranek EP. Integrating palliative care into 26. Quill TE, Abernethy AP. Generalist plus specialist palliative
heart failure care. Arch Intern Med 2005;165:374-8. care — creating a more sustainable model. N Engl J Med 2013;
6. Brumley R, Enguidanos S, Jamison P, et al. Increased satis- 368:1173-5.
faction with care and lower costs: results of a randomized trial 27. Hui D, Elsayem A, De la Cruz M, et al. Availability and inte-
of in-home palliative care. J Am Geriatr Soc 2007;55:993-1000. gration of palliative care at US cancer centers. JAMA 2010;303:
7. Overall appropriations. Bethesda, MD: National Institutes of 1054-61.
Health, Office of Budget, 2013 (http://officeof budget.od.nih 28. Spettell CM, Rawlins WS, Krakauer R, et al. A comprehen-
.gov/pdfs/FY13/Vol%201%20Tab%202%20-%20Overall%20 sive case management program to improve palliative care. J Pal-
Appropriations.pdf). liat Med 2009;12:827-32.
8. Improving palliative care for cancer: summary and recom- 29. Krakauer R, Spettell CM, Reisman L, Wade MJ. Opportuni-
mendations. Washington, DC: Institute of Medicine, 2003 (http:// ties to improve the quality of care for advanced illness. Health
iom.edu/~/media/Files/Report%20Files/2003/Improving-Palliative Aff (Millwood) 2009;28:1357-9.
-Care-for-Cancer-Summary-and-Recommendations/PallativeCare 30. Morrison LJ, Thompson BM, Gill AC. A required third-year
8pager.pdf). medical student palliative care curriculum impacts knowledge
9. Gade G, Venohr I, Conner D, et al. Impact of an inpatient and attitudes. J Palliat Med 2012;15:784-9.
palliative care team: a randomized control trial. J Palliat Med 31. Sulmasy DP, Cimino JE, He MK, Frishman WHUS. U.S. med-
2008;11:180-90. ical students’ perceptions of the adequacy of their schools’ cur-
10. Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative ricular attention to care at the end of life: 1998-2006. J Palliat
care intervention on clinical outcomes in patients with advanced Med 2008;11:707-16.
cancer: the Project ENABLE II randomized controlled trial. 32. A state-by-state report card on access to palliative care in our
JAMA 2009;302:741-9. nation’s hospitals. New York: Center to Advance Palliative Care,
11. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care 2011 (http://www.capc.org/reportcard).
for patients with metastatic non-small-cell lung cancer. N Engl 33. Palliative care services: solutions for better patient care and
J Med 2010;363:733-42. today’s health care delivery challenges. Chicago: American Hos-
12. Zimmermann C, Swami N, Rodin G, et al. Cluster-random- pital Association, 2012 (http://www.hpoe.org).
ized trial of early palliative care for patients with metastatic
cancer. Presented at the 2012 American Society for Clinical On- DOI: 10.1056/NEJMsb1305469
cology Annual Meeting, Chicago, June 1–5, 2012. abstract. Copyright © 2013 Massachusetts Medical Society.

n engl j med 369;24 nejm.org december 12, 2013 2351

The New England Journal of Medicine
Downloaded from nejm.org on December 13, 2013. For personal use only. No other uses without permission.
Copyright © 2013 Massachusetts Medical Society. All rights reserved.