Five things physicians and patients should question in hospice and palliative medicine

Five Things Physicians and Patients ShouldQuestion in Hospice and Palliative MedicineDaniel Fischberg, MD, PhD, Janet Bull, MD, David Casarett, MD, MA, MMM,Laura C. Hanson, MD, MPH, Scott M. Klein, MD, MHSA,Joseph Rotella, MD, MBA, Thomas Smith, MD, C. Porter Storey Jr., MD,Joan M. Teno, MD, MS, and Eric Widera, MD, for the AAHPM ChoosingWisely Task ForceDepartment of Geriatric Medicine (D.F.), John A. Burns School of Medicine, University of Hawaii,Honolulu, Hawaii; Four Seasons (J.B.), Flat Rock, North Carolina; University of PennsylvaniaHealth System (D.C.), Philadelphia, Pennsylvania; Division of Geriatric Medicine and University ofNorth Carolina Palliative Care Program (L.C.H.), University of North Carolina-Chapel Hill, ChapelHill, North Carolina; Hospice and Palliative Care (S.M.K.), Visiting Nurse Service of New York, NewYork, New York; Hosparus (J.R.), Louisville, Kentucky; Johns Hopkins Medical Institutions andSidney Kimmel Comprehensive Cancer Center (T.S.), Baltimore, Maryland; American Academy ofHospice and Palliative Medicine (C.P.S.), Glenview, Illinois; Warren Alpert School of Medicine(J.M.T.), Brown University, Providence, Rhode Island; and Division of Geriatrics (E.W.), University ofCalifornia-San Francisco, San Francisco, California, USA AbstractOveruse or misuse of tests and treatments exposes patients to potential harm. TheAmerican Board of Internal Medicine Foundation’s Choosing WiselyÒ campaign isa multiyear effort to encourage physician leadership in reducing harmful orinappropriate resource utilization. Via the campaign, medical societies are asked toidentify five tests or procedures commonly used in their field, the routine use of which inspecific clinical scenarios should be questioned by both physicians and patients based onthe evidence that the test or procedure is ineffective or even harmful. The AmericanAcademy of Hospice and Palliative Medicine (AAHPM) was invited, and it agreed toparticipate in the campaign. The AAHPM Choosing Wisely Task Force, with input fromthe AAHPM membership, developed the following five recommendations: 1) Don’trecommend percutaneous feeding tubes in patients with advanced dementia; instead,offer oral-assisted feeding; 2) Don’t delay palliative care for a patient with serious illnesswho has physical, psychological, social, or spiritual distress because they are pursuingdisease-directed treatment; 3) Don’t leave an implantable cardioverter-defibrillatoractivated when it is inconsistent with the patient/family goals of care; 4) Don’trecommend more than a single fraction of palliative radiation for an uncomplicatedpainful bone metastasis; and 5) Don’t use topical lorazepam (AtivanÒ),diphenhydramine (BenadrylÒ), and haloperidol (HaldolÒ) (ABH) gel for nausea.
These recommendations and their supporting rationale should be considered byphysicians, patients, and their caregivers as they collaborate in choosing those treatments Address correspondence to: Daniel Fischberg, MD, PhD, Pain and Palliative Care Department, The Queen’s Medical Center, 1301 Punchbowl Accepted for publication: December 24, 2012.
Ó 2013 U.S. Cancer Pain Relief Committee.
Published by Elsevier Inc. All rights reserved.
that do the most good and avoid the most harm for those living with seriousillness.
J Pain Symptom Manage 2013;-:-e-. Ó 2013 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Key WordsChoosing WiselyÒ, palliative care, quality of life, dementia, artificial nutrition andhydration, percutaneous endoscopic gastrostomy, PEG, heart failure, implantablecardioverter-defibrillator, ICD, bone metastasis, bone pain, single-fraction radiation, nausea,lorazepam-haloperidol-diphenhydramine gel, ABH gel In response to Brody’s Top Five challenge, the American Board of Internal Medicine Foun- Advances in biomedical science and the de- dation developed a multiyear effort called velopment of novel therapies over the last 50 Choosing WiselyÒ. In 2012, nine medical societies years have been unprecedented. Yet despite developed and, in conjunction with Consumer these advances, Americans experience inferior Reports, publicized an initial series of Top Five quality of care, efficiency, access, and health out- lists. The American Academy of Hospice and comes compared with the citizens of most other Palliative Medicine (AAHPM) was invited to developed nations, all while health care costs participate, along with 15 additional medical so- rise at an unsustainable ratIn the U.S., health cieties, in the next wave of the campaign. Here, we report the five practices the AAHPM Choos- ing Wisely Task Force recommends patients and causes underlying the paradox of spending physicians question in the practice of hospice more while achieving less are complex, overuse and inappropriate use of tests, procedures, and This list is not meant to serve as a rigid tool therapies have been cited as major contributors, and should instead be used as a support for indi- accounting for perhaps 30% of all health care vidualized decision-making born of conversa- tions between physicians and patients. It also Officestimated that $700 billion annually should be understood that these recommenda- goes to health care spending that has not tions are not universally applicable to the situa- been shown to improve health outcomes. Of tions and settings they address. There may be greatest concern, ineffective and nonbeneficial times when they are inappropriate in light of spe- treatments may expose patients to harm from cific additional circumstances facing a patient.
adverse effects, overtreatment, and delayed de- These recommendations are provided for in- livery of effective and beneficial treatments. At formational purposes only and do not constitute a time of dramatically increased spending, an medical advice. They do not supersede the inde- aging population, and an increasing illness bur- pendent judgment of a medical professional, den, it is absolutely necessary for physicians and and the authors believe that an individual with patients to choose every treatment wisely.
specific medical questions should obtain medi- In light of these challenges and with the cal advice from their health care provider.
goal of maximizing quality of care while mini-mizing the costs, Brody challenged medical so-cieties to each create a ‘‘Top Five’’ list of tests or treatments that are commonly ordered, ex- The president of AAHPM appointed a special pensive, and have been shown not to provide task force to coordinate the development of the any meaningful benefit to at least a major cat- Academy’s list of ‘ Five Things Physicians and Pa- egory of patients for whom they are commonly tients Should Question in Hospice and Palliative Medicine.’ Chaired by a member of the Board the Top Five list as ‘‘a prescription for how, of Directors who previously oversaw AAHPM’s within that specialty, the most money could Education and Training Strategic Coordinating be saved most quickly without depriving any Committee, the task force included representa- patient of meaningful medical benefit.’’ tives of the Academy’s Quality and Practice Five Things Physicians and Patients Should Question in HPM Five Things Physicians and Patients Should Question in Hospice and Palliative Medicine 1. Don’t recommend percutaneous feeding tubes in patients with advanced dementia; instead, offer oral-assisted feeding.
In advanced dementia, studies have found that feeding tubes do not result in improved survival, prevention of aspiration pneumonia, or improved healing of pressure ulcers. Feeding tube use in such patients has actually been associated with pressureulcer development, use of physical and pharmacologic restraints, and patient distress about the tube itself. Assistance with oralfeeding is an evidence-based approach to provide nutrition for patients with advanced dementia and feeding problems; in thefinal phase of this disease, assisted feeding may focus on comfort and human interaction more than the nutritional 2. Don’t delay palliative care for a patient with serious illness who has physical, psychological, social, or spiritual distress because they are pursuing disease-directed treatment.
Numerous studiesdincluding randomized trialsdprovide evidence that palliative care improves pain and symptom control, improves family satisfaction with care, and reduces costs. Palliative care does not accelerate death and may prolong life inselected populations.
3. Don’t leave an ICD activated when it is inconsistent with the patient/family goals of care.
In about a quarter of patients with ICDs, the defibrillator fires within weeks preceding death. For patients with advanced irreversible diseases, defibrillator shocks rarely prevent death, may be painful to patients, and are distressing to caregivers/family members. Currently, there are no formal practice protocols to address deactivation; less than 10% of hospices haveofficial policies. Advance care planning discussions should include the option of deactivating the ICD when it no longersupports the patient’s 4. Don’t recommend more than a single fraction of palliative radiation for an uncomplicated painful bone metastasis.
As stated in the American Society for Radiation Oncology 2011 guideline, single-fraction radiation to a previously unirradiated peripheral bone or vertebral metastasis provides comparable pain relief and morbidity compared with multiple-fractionregimens while optimizing patient and caregiver convenience. Although it results in a higher incidence of later need forretreatment (20% vs. 8% for multiple-fraction regimens), the decreased patient burden usually outweighs any considerationsof long-term effectiveness for those with a limited life expectancy.
5. Don’t use topical lorazepam (Ativan), diphenhydramine (Benadryl), and haloperidol (Haldol) (ABH) gel for nausea.
Topical drugs can be safe and effective, such as topical nonsteroidal anti-inflammatory drugs for local arthritis symptoms.
Although topical gels are commonly prescribed in hospice practice, antinausea gels have not been proven effective in anylarge, well-designed, or placebo-controlled trials. The active ingredients in ABH are not absorbed to systemic levels that couldbe effective. Only diphenhydramine (Benadryl) is absorbed via the skin and then only after several hours and erratically atsubtherapeutic levels. It is, therefore, not appropriate for ‘‘as needed’’ use. The use of agents given via inappropriate routesmay delay or prevent the use of more effective inter ICD ¼ implantable cardioverter-defibrillator.
Standards Task Force, Research Committee, the AAHPM Executive Committee and submit- Ethics Committee, Public Policy Committee, ted to the American Board of Internal Medi- and External Awareness Task Force, as well as at-large appointees who represent distinguishedleaders in the field.
AAHPM aimed for an inclusive list develop- ment process that would afford every member The five recommendations are the following: of the Academy the opportunity to participate 1. Don’t recommend percutaneous feeding in the identification or evaluation of potential tubes in patients with advanced dementia; recommendations. The task force solicited in- instead, offer oral-assisted feeding.
put from AAHPM’s 17 special interest groups,and task force members also offered their own Dementia is the fifth leading cause of death among Americans aged 65 years and older, and Considering the potential impact and evi- recent research suggests that this is an underes- dence to support the proposed recommenda- timateDementias are progressive diseases of tions, the task force settled on seven finalists, cognitive and physical decline. In advanced de- and a rationale and evidence base were further mentia, 86% of patients develop an eating prob- lem that increases risk for malnutrition and were invited to comment on and rank these recurrent infections. Because they may view it seven recommendations. The members’ feed- as a choice between feeding and not feeding, back informed the task force’s final delibera- families are often faced with what they perceive tion, which included narrowing the list to five as a difficult decision on whether to insert a feed- recommendations and refining their verbiage.
ing tube.However, framing the decision in Finally, the list was reviewed and approved by these stark terms ignores the relative risks and benefits of tube feeding compared with contin- untreated pain and other symptoms, high care- ued oral feeding in advanced dementia.
giver burden, poor communication with their A substantial body of research provides evi- health care providers, infrequent discussions dence of the risks and benefits of percutaneous endoscopic gastrostomy (PEG) feeding tube in- and preferences, and high rates of hospitaliza- tion and burdensome treatments at the end of that feeding tubes do not result in improved cus on reducing the suffering of those dealing survival, prevention of aspiration pneumonia, with a serious illness is often delayed until after or improved healing of pressure ulcers.
potentially curative or life-prolonging treat- Teno et conducted an analysis of national ment options have been exhausted, with stud- Medicare claims and the Minimum Data Set, us- ies showing that palliative care consultations ing techniques that accounted for selection occur very late in the disease trajectory bias, and found no survival benefit of feeding There is now convincing evidence that the de- tube insertions in people with advanced cogni- livery of palliative care concurrent with the tive impairment. For nursing home residents disease-directed treatment can improve the with advanced dementia, the one year survival quality of life, symptom control, and family sat- rate after a feeding tube insertion is only isfaction with care, all while reducing costs associated with aggressive end-of-life care. Pal- liative care does not shorten life expectancy tubes are not without risk. Although the mortal- and can improve survival in select populations.
ity rate with the insertion of a PEG feeding tube The evidence that palliative care improves is small, people with advanced dementia who symptom control and leads to greater family sat- had a PEG tube insertion during an acute care isfaction with care has been shown in both obser- hospitalization had more than two times the vational and randomized control trials. Ringdal risk of developing a Stage II or higher pressure et al.conducted a randomized trial of compre- ulcerA small study based on bereaved family hensive palliative care services for patients with member interviews reported that 25.9% of dece- incurable cancer and life expectancy of two to dents with feeding tubes were physically re- nine months. One month after the patients’ strained and 29.2% were pharmacologically deaths, families of the patients who received pal- liative care were more satisfied with most aspects nearly 40% of patients dying with dementia of the care received. The most positive effects were in pain control, speed of symptom treat- Oral-assisted feeding represents a viable ment, communication, quality of family confer- ences, and availability and thoroughness of and caloric intake for patients with dementia.
physicians. Engelhardt et alrandomized 275 High-calorie supplements can support weight patients with chronic obstructive pulmonary dis- stabilization or weight gain for people with de- ease, heart failure, or cancer and with recurrent mentia; assisted feeding programs, modified hospital admissions to usual care or concurrent foods, and appetite stimulants have a potential palliative care case management. Palliative care patients had increased satisfaction with care comfort is an appropriate option in the final and communication and increased use of ad- vance directives (69% vs. 48%, P ¼ 0.006). Addi- and accept the option of oral-assisted feeding, tionally, in a retrospective study, more time rather than tube feeding, when they receive ef- between the initial palliative care consultation and the patient’s death was associated with bet-ter family perceptions of care, most notably for 2. Don’t delay palliative care for a patient with serious illness who has physical, psychologi- Focusing on the relief of suffering and pro- cal, social, or spiritual distress because they moting shared decision-making through concur- are pursuing disease-directed treatment.
rent palliative care has been shown to lower Studies have shown that individuals dealing costs and reduce rates of intensive care use and with a serious illness are at significant risk for hospitalizations. Gade conducted a multicenter Five Things Physicians and Patients Should Question in HPM randomized trial of interdisciplinary hospital- diseases, defibrillator shocks rarely prevent death, based palliative care, enrolling 517 patients with may be painful, and are typically distressing to life-limiting illnesses. Individuals randomized to caregivers and family members. In addition, pa- palliative care reported higher quality of care tients who are at the end of life often experience and better quality of communication. They also electrolyte disturbances, hypoxemia, acidosis, experienced fewer intensive care unit admissions and organ failure, making these devices less effec- and a net cost savings of $4855 per patient, with no Barriers to timely deactivation have been difference in mortalityIn an eight-hospital shown to include both patient and physician fac- study of 4908 palliative care patients and more tors. Patients may be unwilling to discuss deactiva- than 20,000 propensity score-matched controls, tion, yet remain fearful about potential shocks.
palliative care consultations were associated with Often, they believe that physicians should make $1696 direct cost savings per patient for patients the decision regarding deactivation,and re- discharged alive and $4908 direct cost savings search shows that patients may not even realize per patient for patients who died in the h In a controlled study of palliative care for New Although most physicians believe that deacti- York Medicaid patients, palliative care access vation should be discussed with patients, this resulted in similar or greater cost savings.
rarely occurs. Furthermore, physicians’ lack of Finally, there is some evidence that concurrent comfort in discussions with patients has been palliative care may prolong life in select popula- shown to be a major barrier to deactivation.
When discussions do occur, it is often in the randomized 151 outpatients with metastatic last days of a patient’s lifeGiven these short- non-small-cell lung cancer to either standard comings, advance care planning discussions care or concurrent palliative care. Patients who should include the option of deactivating the received concurrent palliative care showed sig- ICD when it no longer supports a patient’s goals, nificant improvements in quality of life and and Do Not Attempt Resuscitation orders should mood. Despite lower use of aggressive end-of- be consistent with deactivation of these devices.
life care, individuals randomized to concurrent Still, fewer than 10% of U.S. hospices have palliative care showed a significant increase in ICD deactivation policies.In a national survey survival compared with standard care (median of hospice organizations, only 42% of hospice survival 11.6 vs. 8.9 months; P ¼ 0.02).
patients with ICDs had their devices deacti- vated, and only 25% of hospices surveyed had directed treatment and palliative care is unwar- a magnet available for emergency deactivation; ranted based on the aforementioned findings.
of those that did, only 64% provided training When patients experience burdensome symp- toms, difficult treatment choices, or emotional ICDs are more likely to have patients with deac- distress related to serious illness, palliative care tivated devices compared with those without should be offered in combination with disease- a policy (73% vs. 38%, P < 0.0In spite of modifying therapies. The resulting treatment their relatively sparse use, such policies could approach can promote physical and emotional be brought to scale in U.S. hospices, and a sam- support, improve shared decision-making, sup- ple policy is available to aid organizations in cre- ating ICD policies and procedureFinally, it is recommended that hospices develop relation-ships with local electrophysiologists or device 3. Don’t leave an implantable cardioverter- manufacturers to assure that reprogramming defibrillator (ICD) activated when it is in- of a device can occur to reduce barriers to deac- The ethical principles around ICD deactiva- As patients approach the end of life, the bene- tion are well established. Informed patients fits and burdens of ICDs need to be readdressed with decisional capacity, or their legally autho- in alignment with the patient and family goals of rized decision-makers, can choose to refuse any care. About a quarter of patients with ICDs expe- and all treatments, including life-sustaining rience a shock from their device within weeks of ones. Furthermore, there is no ethical distinc- tion between withdrawing treatment (e.g., deactivating an ICD) and withholding treatment are frequently consulted to manage their pallia- (e.g., not placing an ICD in the first placeIn tion. Cancer patients referred to HPM specialists the face of a progressive disease, a patient may often have limited performance status, coexist- feel that the benefit of having an ICD prevent ing visceral metastases, and shortened life expec- a fatal arrhythmia is outweighed by the burden tancy. Such patients are likely to be burdened by of treatment. Not allowing deactivation forces short-term side effects, repeated trips to the radi- a patient to suffer potential unwanted continued ation center, and transfers on and off the radia- intervention and violates the ethical principles tion treatment table. Because limited prognosis mitigates concern for a late recurrence of pain requiring retreatment, the primary goal of palli- for a patient’s natural death from disease pro- ation is to restore quality of life as quickly as pos- gression. If for some reason a physician has a con- sible with the least burden to the patient and scientious objection to deactivating a device, the family. Therefore, for most patients considering physician has the obligation to transfer the pa- palliative radiation for painful uncomplicated bone metastasis, SF (8 Gy) EBRT is the bestrecommendation.
4. Don’t recommend more than a single frac- The survival of patients with bone metastasis tion (SF) of palliative radiation for an un- is associated with the origin of the primary can- complicated painful bone metastasis.
cer and presence of visceral metastases or Bone is the most common site of cancer me- skeletal-related events (SRE). Breast and pros- tastasis, and although bone metastasis is most tate cancer patients with metastasis only to prevalent in breast and prostate cancers (found bone have median life expectancies measured in w70% at autopsy), it also is common in mul- in years. However, the median survival for pa- tiple myeloma and cancers of thyroid, kidney, tients with lung cancer metastatic to bone is and bronchial origin. Because bone metastasis is the leading cause of cancer-related pai (pathological fracture, cord compression, hy- HPM specialists are frequently consulted to percalcemia, or any condition requiring bone treat symptoms and address suffering associ- surgery or radiation) increases mortality. A study of breast cancer patients in the Danish Cancer Registry reported a five year survival evidence-based practice guideline published by rate of 75.8% for patients without bone metas- the American Society for Radiation Oncology tasis, 8.3% for those with bone metastasis, and (ASTRO) for palliation of bone metastases The ASTRO guideline reports the findings of who receives radiation for painful bone metas- a systematic review of the literature concerning tasis (by definition an SRE) has a shortened the comparative effectiveness of SF vs. multiple- life expectancy, perhaps best expressed as fraction (MF) regimens of external beam radio- therapy (EBRT) for palliation of uncomplicated EBRT provides relief of associated pain in painful bone metastases (those not associated 50%e85% of patients with bone metastasis, de- with spinal cord compression or an unstable pending on what methods of pain assessment and definition of relief are applied. Up to Compared with an MF regimen, SF palliative a third of patients experience complete relief EBRT provides equivalent short-term symptom of pain at the treated site. SF treatment of relief, fewer side effects, and less inconve- 8 Gy to a previously unirradiated bone metasta- nience for patients. There is a higher incidence sis provides equivalent pain relief to various of symptom recurrence for SF compared with schedules of MF treatment (30 Gy in 10 frac- MF regimens (20% vs. 8%), but recurrences tions, 24 Gy in six fractions, or 20 Gy in five usually can be irradiated a second time. These fractions). There is no significant difference findings are the same for both peripheral bone between SF and MF regimens in the risk of de- pathological fracture. The only therapeutic a common source of pain and morbidity in pa- tients with advanced cancer, and HPM specialists is the incidence of recurrent pain requiring Five Things Physicians and Patients Should Question in HPM retreatment at the site (20% for SF compared drugs. Topical drugs can be safe and effective, be more willing to retreat a site of recurrent such as topical nonsteroidal anti-inflammatory pain if the previous treatment was SF rather drugs for local arthritis symptoms.Topical Acute radiation reactions are generally worse in hospice practice, with one large hospice and more prolonged with MF than with SF treat- pharmacy reporting two-thirds of patients get- ment. The incidence of a temporary postradia- tion flare in pain may be higher with SF but can However, antinausea gels have not been pro- ven effective in any well-designed or placebo- MF is more expensive than SF treatment.
controlled trials, and the available evidence is Given the equivalent short-term efficacy, SF is a more cost-effective option for most patients.
The active ingredients in one commonly pre- Several authors have noted significant interna- scribed antinausea gel, ABH, are not absorbed tional variation in the use of SF treatment to to systemic levels that could be effective by any palliate bone metastasis.Specifically, physi- cians in the U.S. use MF for painful bone me- healthy volunteers apply the standard 1.0 mL tastasis more often than their counterparts in dose (2 mg of lorazepam, 25 mg of diphenhy- other countries, despite the evidence of thera- dramine, and 2 mg of haloperidol in a pluronic peutic equivalence. The creators of the AS- lecithin organogel), rubbed on the volar sur- face of the wrists as is done in practice. Blood would drive a change in the patterns of care.
samples were obtained at 0, 30, 60, 90, 120, Even if the costs of SF and MF regimens were 180, and 240 minutes. No lorazepam (A) or hal- equal, however, the decreased patient burden operidol (H) was detected in any sample from alone would be sufficient reason to recom- any of the 10 patients, down to a level of mend SF treatment for most patients seen in 0.05 ng/mL. Most volunteers had undetectable levels of diphenhydramine at most time points, Physicians should tailor their recommenda- tions to the individual patient’s condition, a single volunteer of 0.30 ng/mL at 240 min- prognosis, and goals of care. MF treatment utes. The therapeutic level of diphenhydramine may be a reasonable option for a patient likely fore, none of the lorazepam (A), haloperidol have difficulty accessing retreatment if pain (H), or diphenhydramine (B) in ABH gel is ab- were to recur. Some patients with favorable tu- sorbed in sufficient quantities to be effective in mor type, excellent performance status, and the treatment of nausea and vomiting.
aggressive care goals may prefer MF treatment.
This is an important issue for quality of care, Patients with complications such as cord com- safety, and cost. The advantage of ABH and pression or instability in weight-bearing bones other gels is the easy patient-controlled appli- require a multidisciplinary approach including cation and the low cost. But the use of agents orthopedic surgery, neurosurgery, radiation given via ineffective routes may delay or pre- oncology, and palliative medicine working in vent the use of more effective interventions, concert. In addition to EBRT, practitioners causing suffering and even more expense by should consider other treatment modalities precipitating hospital admission. Therefore, for painful bone metastasis, including anti- the use of ABH and similar gels is not recom- inflammatory drugs, other analgesic drugs, bisphosphonates, radiopharmaceuticals, and 5. Don’t use topical lorazepam (AtivanÒ), di- Over the past 10 years, the field of palliative peridol (HaldolÒ) (ABH) gel for nausea.
care has grown and evolved with a remarkable, unprecedented rapidity. Palliative care now palliative care consultations at some cancer subspecialties with defined domains of knowl- concerns that come with seriously ill patient edge and skills and an expanding evidence populations create challenges that can be sub- base. Perhaps equally important, palliative stantialAlso, unlike many fields, there is not care is gaining widespread recognition as a spe- yet a clear consensus about all the outcomes cialty that helps patients, families, and provi- that should define ‘‘effectiveness’’ in palliative ders to achieve an improved quality of life.
care. However, none of these challenges is insur- Much of this growth is attributable to the mountable.In fact, there is a growing body of way that palliative care promotes open and evidence that evaluates palliative care interven- honest communication with patients and their tions using both prospective randomized con- a chief contribution of palliative care to the na- tional dialogue about end-of-life care has been More such studies are needed in two areas.
the recognition that such communication can First, a productive line of research would rigor- help patients to avoid treatment that they do ously evaluate novel treatments. The ability to not wantMore generally, this focus on com- generate new treatments and advance the sci- munication, decision-making, and patient- ence of comfort is perhaps the most visible ev- centered outcomes has raised questions about idence of palliative care’s success as a field.
the risks and potential benefits of interven- However, a second parallel effort also is needed. Just as it is essential to develop and This article highlights two such interven- test novel interventions, it will be equally im- tionsdfeeding tubes for those with advanced portant to critically examine the risks and po- dementia and ICDs near the end of lifedthe tential benefits of the existing palliative benefits of which are highly questionable.
treatments that are widely used but unproven.
Research along both these pathways will help that their use should be the focus of much to ensure that the palliative care evidence more careful decision-making by patients and base continues to grow and that the actual health care providers. More broadly, this arti- practice of palliative care is consistent with cle suggests opportunities for palliative care providers to find ways to shape practices thatare more consistent with the existing evidence.
questions about its own practice. For instance, This article was written and reviewed by the the use of MF radiation therapy and ABH gel in palliative care settings offers a valuable cau- Wisely Task Force. The task force and develop- tionary lesson. Just as providers in other fields ment of this article were provided institutional support by AAHPM. The authors and the task hope of prolonging life, palliative care pro- force members declare no relevant conflicts of viders also may rely on unproven interventions out of a desire to enhance the quality of life.
The authors would like to express their grat- This lesson highlights the fact that as the itude to AAHPM staff members Jacqueline Ko- field of palliative care continues to develop, cinski, MPP, and Patrick Hermes, MS, and all there is an urgent need to ensure its evidence base keeps pace with those of other fields. In Task Force, whose hard work, careful consider- particular, palliative care will need to carefully ation, and expert contributions made this arti- examine its own treatments, making a substan- tial investment in comparative effectiveness re-search. More generally, palliative care needs to aspire to an evidence base in which all palliativeinterventionsdfrom opioids to family meet- 1. Public Policy Committee of the American Col- lege of Physicians, Ginsburg JA, Doherty RB, et al.
Achieving a high-performance health care system Of course, there are challenges to conducting with universal access: what the United States can high-quality comparative effectiveness research learn from other countries. Ann Intern Med 2008; in HPM settings. For instance, the ethical Five Things Physicians and Patients Should Question in HPM 2. Truffer CJ, Keehan S, Smith S, et al. Health in a comprehensive cancer center. J Palliat Med spending projections through 2019: the recession’s impact continues. Health Aff (Millwood) 2010;29: 18. Elsayem A, Swint K, Fisch MJ, et al. Palliative care inpatient services in a comprehensive cancer 3. Fisher ES, Bynum JP, Skinner JS. Slowing the center: clinical and financial outcomes. J Clin Oncol growth of health care costsdlessons from regional variation. N Engl J Med 2009;360:849e852.
19. Gelfman LP, Meier D, Morrison RS. Does palli- 4. Congressional Budget Office. Increasing the ative care improve quality? A survey of bereaved value of federal spending on health care. 2008.
family members. J Pain Symptom Manage 2008;36: 20. Higginson IJ, Finlay IG, Goodwin DM, et al. Is there evidence that palliative care teams alter end- 5. Brody H. Medicine’s ethical responsibility for of-life experiences of patients and their caregivers? health care reform: the Top Five list. N Engl J J Pain Symptom Manage 2003;25:150e168.
21. Jordhoy MS, Fayers P, Saltnes T, et al.
6. Finucane TE, Christmas C, Travis K. Tube feed- A palliative care intervention and death at home: ing in patients with advanced dementia: a review of a cluster randomized trial. Lancet 2000;356: the evidence. JAMA 1999;282:1365e1370.
7. Gillick MR. Rethinking the role of tube feeding 22. London MR, McSkimming S, Drew N, Quinn C, in patients with advanced dementia. N Engl J Med Carney B. Evaluation of a Comprehensive, Adapt- able, Life-Affirming, Longitudinal (CALL) palliative 8. Hanson LC, Ersek M, Gilliam R, Carey TS. Oral care project. J Palliat Med 2005;8:1214e1225.
feeding options for people with dementia: a system- 23. Temel JS, Greer JA, Muzikansky A, et al. Early atic review. J Am Geriatr Soc 2011;59:463e472.
palliative care for patients with metastatic non- 9. Kuo S, Rhodes RL, Mitchell SL, Mor V, Teno JM.
small-cell lung cancer. N Engl J Med 2010;363: Natural history of feeding-tube use in nursing home residents with advanced dementia. J Am Med Dir As- 24. Berger JT. The ethics of deactivating implanted cardioverter defibrillators. Ann Intern Med 2005; 10. Palecek EJ, Teno JM, Casarett DJ, et al. Comfort feeding only: a proposal to bring clarity to decision- 25. Goldstein NE, Lampert R, Bradley EH, Lynn J, making regarding difficulty with eating for persons Krumholz HM. Management of implantable cardi- with advanced dementia. J Am Geriatr Soc 2010; overter defibrillators in end-of-life care. Ann Intern 11. Sampson EL, Candy B, Jones L. Enteral tube 26. Goldstein NE, Carlson M, Livote E, Kutner JS.
feeding for older people with advanced dementia.
Management of implantable defibrillators in hos- Cochrane Database Syst Rev 2009;2:CD007209.
pice: a nationwide survey. Ann Intern Med 2010; 12. Stratton RJ, Ek AC, Engfer M, et al. Enteral nu- tritional support in prevention and treatment of 27. Russo JE. Deactivation of ICDs at the end of life: pressure ulcers: a systematic review and meta-analy- a systematic review of clinical practices and provider and patient attitudes. Am J Nurs 2011;111:26e35.
13. Teno JM, Gozalo P, Mitchell SL, et al. Feeding 28. Lutz S, Berk L, Chang E, et al. Palliative radio- tubes and the prevention or healing of pressure ul- therapy for bone metastases: an ASTRO evidence- cers. Arch Intern Med 2012;172:697e701.
based guideline. Int J Radiat Oncol Biol Phys 14. Teno JM, Mitchell SL, Gozalo PL, et al. Hospital characteristics associated with feeding tube place- 29. Smith TJ, Ritter JK, Poklis JL, et al. ABH gel is ment in nursing home residents with advanced cog- not absorbed from the skin of normal volunteers.
nitive impairment. JAMA 2010;303:544e550.
J Pain Symptom Manage 2012;43:961e966.
15. Teno JM, Mitchell SL, Kuo SK, et al. Decision- 30. Weschules DJ. Tolerability of the compound making and outcomes of feeding tube insertion: ABHR in hospice patients. J Palliat Med 2005;8: a five-state study. J Am Geriatr Soc 2011;59:881e886.
16. Delgado-Guay MO, Parson HA, Li Z, Palmer LJ, 31. Wachterman M, Kiely DK, Mitchell SL. Report- Bruera E. Symptom distress, intervention and out- ing dementia on the death certificates of nursing comes of intensive care unit cancer patients re- home residents dying with end-stage dementia.
ferred to a palliative care consult team. Cancer 32. Teno JM, Gozalo PL, Mitchell SL, et al. Does 17. Elsayem A, Smith ML, Palmer JL, et al. Impact feeding tube insertion and its timing improve sur- of a palliative care service on in-hospital mortality vival? J Am Geriatr Soc 2012;60:1918e1921.
33. Simmons SF, Keeler E, Zhuo X, et al. Prevention 47. Morrison RS, Dietrich J, Ladwig S, et al. Pallia- of unintentional weight loss in nursing home resi- tive care consultation teams cut hospital costs for dents: a controlled trial of feeding assistance. J Am Medicaid beneficiaries. Health Aff (Millwood) 34. Monteleoni C, Clark E. Using rapid-cycle quality 48. Sulmasy D. Within you/without you: biotech- improvement methodology to reduce feeding tubes nology, ontology, and ethics. J Gen Intern Med in patients with advanced dementia: before and af- 49. Goldstein NE, Mehta D, Siddiqui S, et al. That’s 35. Hanson LC, Carey TS, Caprio AJ, et al. Improv- like an act of suicide: patients’ attitudes toward de- ing decision-making for feeding options in ad- activation of implantable defibrillators. J Gen Intern vanced dementia: a randomized, controlled trial.
36. Teno JM, Clarridge BR, Casey V, et al. Family Bradley EH, Morrison RS. ‘‘It’s like crossing perspectives on end-of-life care at the last place of a bridge’’: complexities preventing physicians from discussing deactivation of implantable defibrillatorsat the end of life. J Gen Intern Med 2008;23:2e6.
37. Desbiens NA, Mueller-Rizner N, Connors AF Jr,Wenger NS, Lynn J. The symptom burden of seri- 51. Beauchamp TL, Childress J. Principles of bio- ously ill hospitalized patients. SUPPORT Investiga- medical ethics, 5th ed. New York: Oxford University Preferences for Outcome and Risks of Treatment.
52. Zellner RA, Aulisio MP, Lewis WR. Should im- J Pain Symptom Manage 1999;17:248e255.
plantable cardioverter-defibrillators and permanent 38. Walke LM, Gallo WT, Tinetti ME, Fried TR. The pacemakers in patients with terminal illness be de- burden of symptoms among community-dwelling activated? Deactivating permanent pacemaker in pa- older persons with advanced chronic disease. Arch tients with terminal illness. Patient autonomy is paramount. Circ Arrhythm Electrophysiol 2009;2:340e344.
39. A controlled trial to improve care for seriouslyill hospitalized patients. The study to understand 53. Coleman R. Clinical features of metastatic bone prognoses and preferences for outcomes and risks disease and risk of skeletal morbidity. Clin Cancer of treatments (SUPPORT). The SUPPORT Princi- pal Investigators. JAMA 1995;274:1591e1598.
54. Sugiura H, Yamada K, Sugiura T, Hida T,Mitsudomi T. Predictors of survival in patients with 40. Osta BE, Palmer JL, Paraskevopoulos T, et al. In- bone metastasis of lung cancer. Clin Orthop Relat terval between first palliative care consult and death in patients diagnosed with advanced cancer ata comprehensive cancer center. J Palliat Med 2008; 55. Yong M, Jensen A, Jacobsen J, et al. Survival in breast cancer patients with bone metastases andskeletal-related events: a population-based cohort 41. Hui D, Elsayem A, De la Cruz M, et al. Availabil- study in Denmark (1999-2007). Breast Cancer Res ity and integration of palliative care at US cancer 56. Chow E, Harris K, Fan G, Tsao M, Sze WM. Pal- 42. Ringdal GI, Jordhoy MS, Kaasa S. Family satis- liative radiotherapy trials for bone metastases: a sys- faction with end-of-life care for cancer patients in tematic review. J Clin Oncol 2007;25:1423e1436.
a cluster randomized trial. J Pain Symptom Manage2002;24:53e63.
57. Steenland E, Leer J, van Houwelingen H, et al.
The effect of a single fraction compared to multiple 43. Engelhardt JB, McClive-Reed KP, Toseland RW, fractions on painful bone metastases: a global anal- et al. Effects of a program for coordinated care of ysis of the Dutch Bone Metastasis Study. Radiother advanced illness on patients, surrogates, and health- care costs: a randomized trial. Am J Manag Care2006;12:93e100.
58. Hartsell W, Scott C, Brunner DW, et al. Ran-domized trial of short- versus long-course radiother- 44. Casarett D, Pickard A, Bailey FA, et al. Do palli- apy for palliation of painful bone metastases. J Natl ative consultations improve patient outcomes? J Am 59. Hird A, Chow E, Zhang L, et al. Determining 45. Gade G, Venohr I, Conner D, et al. Impact of an the incidence of pain flare following palliative radio- inpatient palliative care team: a randomized control therapy for symptomatic bone metastases: results trial. J Palliat Med 2008;11:180e190.
from three Canadian cancer centers. Int J Radiat 46. Morrison RS, Penrod JD, Cassel JB, et al. Cost savings associated with US hospital palliative care 60. Hird A, Zhang L, Holt T, et al. Dexamethasone consultation programs. Arch Intern Med 2008;168: for the prophylaxis of radiation-induced pain flare after palliative radiotherapy for symptomatic bone Five Things Physicians and Patients Should Question in HPM metastases: a phase II study. Clin Oncol 2009;21: 66. Quill TE. Initiating end-of-life discussions with seriously ill patients: addressing the ‘‘elephant in 61. Fairchild A, Barnes E, Ghosh S, et al. Interna- the room’’. JAMA 2000;284:2502e2507.
tional patterns of practice in palliative radiotherapy 67. Hanson LC, Tulsky JA, Danis M. Can clinical in- for painful bone metastases: evidence-based prac- terventions change care at the end of life? Ann tice? Int J Radiat Oncol Biol Phys 2009;75: 1501e1510.
62. Dhillon N, Kopetz S, Pei BL, et al. Clinical find- 68. Casarett DJ, Karlawish JHT. Are special ethical ings of a palliative care consultation team at a guidelines needed for palliative care research? comprehensive cancer center. J Palliat Med 2008; J Pain Symptom Manage 2000;20:130e139.
69. Phipps EJ. What’s end of life got to do with it? 63. Derry S, Moore RA, Rabbie R. Topical NSAIDs Research ethics with populations at life’s end. Ger- for chronic musculoskeletal pain in adults. Co- chrane Database Syst Rev 2012;9:CD007400.
64. Bleicher J, Bhaskara A, Huyck T, et al.
70. Kutner JS, Smith MC, Corbin L, et al. Massage Lorazepam, diphenhydramine, and haloperidol therapy versus simple touch to improve pain and transdermal gel for rescue from chemotherapy- mood in patients with advanced cancer. Ann Intern induced nausea/vomiting: results of two pilot trials.
71. Casarett D, Johnson M, Smith D, Richardson D.
65. Winek CL, Wahaba WW, Winek CL Jr, Balzer TW.
The optimal delivery of palliative care: a national Drug & chemical blood-level data. 2001. Available comparison of the outcomes of consultation teams vs inpatient units. Arch Intern Med 2010;171:

Source: http://www.lifemattersmedia.org/wp-content/uploads/2013/06/Fischberg-JPSM-2013.pdf