Attitudes of medical oncologists toward palliative care for patients with advanced and incurable cancer

Report on a survey by the European Society of Medical Oncology taskforce on palliative and supportive care

Authors

  • Nathan I. Cherny M.B.B.S.,

    Corresponding author
    1. Medical Oncology and Palliative Medicine Service, Department of Oncology, Shaare Zedek Medical Center, Jerusalem, Israel
    • Medical Oncology and Palliative Medicine Service, Department of Oncology, Shaare Zedek Medical Center, Jerusalem, Israel,
    Search for more papers by this author
    • Fax: (011) 972 26666731

    • Both authors are members of the European Society of Medical Oncology Taskforce on Palliative and Supportive Care.

  • Raphael Catane M.D.

    1. Sheba Medical Center, Tel Aviv University Medical School, Tel Aviv, Israel
    Search for more papers by this author
    • Both authors are members of the European Society of Medical Oncology Taskforce on Palliative and Supportive Care.


  • The following members of the European Society of Medical Oncology Taskforce on Palliative and Supportive Care participated in this study: Maurizio Tonato (Policlinico Divisione Oncologia Medica, Perugia, Italy); K. Winand Lange (Johanniter–Krankenhaus Rheinhausen, Duisburg, Germany); Herbert Kappauf (Institut für Medizinische Onkologie, Nürnberg, Germany); Alexandru–Calin Grigorescu (Oncologic Institute of Bucharest, Bucarest, Romania); Sergei Tjulandin (Cancer Research Center, Moscow, Russia); Piotr Siedlecki (Marie Skledowska–Curie Institute Cancer Center, Warszawa, Poland); Marco Maltoni (Department of Medical Oncology, Pierantoni Hospital, Forlfi, Italy); Marianne Kloke (Universitätsklinikum Essen, Essen, Germany); Kaija Holli (Department of Oncology, University Hospital, Tampere, Finland); Vladimir Bryuzgin (Cancer Research Center, Moscow, Russia); Dirk Schrijvers (Middelheim Hospital, Antwerp, Belgium); Özgür Özyilkan (Baskent University Faculty of Medicine, Ankara, Turkey); Martin Chasen (Richard Eek Sandton Oncology Centre, Pretoria, South Africa); Maria Wagnerova (University Teaching Hospital, Kosice, Slovak Republic); Janos Szanto (Debrecen Medical University, Debrecen, Hungary); and Joao Oliveira (Istituto Portugues de Oncologia, Losboa Codex, Portugal).

Abstract

BACKGROUND AND METHODS

In part of a quality improvement program, the European Society of Medical Oncology (ESMO) surveyed its membership regarding their involvement in and attitudes toward the palliative care (PC) of patients with advanced cancer.

RESULTS

Of 895 members who responded, 82.5% were European and 12.1% were American. Sixty-nine percent of respondents reported that patients with advanced cancer constituted a major proportion of their practice; for 22% of respondents, patients with advanced cancer constituted most of their practice. Only a minority of respondents collaborated often with a PC care specialist (35%), a palliative home care service (38%), an in-patient hospice (26%), or a psychologist (33%). In response to questions regarding specific involvement in PC clinical tasks, respondents were involved more commonly in treating physical symptoms, such as pain (93%), fatigue (84%), and nausea/emesis (84%), than in managing psychological symptoms and end-of-life care issues, such as depression/anxiety (65%), existential distress (29%), or delirium (12%). Forty-three percent of respondents reported that they directly administered end-of-life care often, and 74% reported that they derived satisfaction from their involvement in end-of-life care. Overall, 88.4% of respondents endorsed the belief that medical oncologists should coordinate the end-of-life care for their patients, but a substantial minority (42%) felt that they were trained inadequately for this task. Positive attitudes toward PC were correlated highly with the degree of direct involvement in PC practice. Practitioners in private practice or teaching hospitals had substantially more positive attitudes regarding PC compared with physicians based in comprehensive cancer centers (P < 0.05). Although most of the responding medical oncologists expressed positive views regarding their involvement in the PC of patients with advanced cancer and dying patients, 15% of respondents had pervasively negative views.

CONCLUSIONS

Most ESMO oncologists recognize the importance of PC and supportive care for patients with advanced cancer. Despite this, many are prepared inadequately for these tasks, and actual participation levels commonly are suboptimal. Cancer 2003. © 2003 American Cancer Society.

Despite major progress in cancer care, cancer-related mortality remains high, and more than 50% of patients with cancer ultimately develop advanced and incurable disease, leading to death.1, 2 The care of these patients constitutes a major part of the case load of oncologists in Europe and world wide. Indeed, in the developing world, where presentation with advanced cancer is the norm, this constitutes the overwhelming bulk of cancer care.

Patients with incurable, advanced cancer present a major challenge. They commonly have numerous physical symptoms.3–8 Psychologic distress, including anxiety, depression, and suicidal ideation, are commonplace.9, 10 By virtue of increasing incapacitation, these patients have complex care needs that put stress on their families and on healthcare resources. Indeed, their families very often are distressed.11–14 Commonly, these patients also suffer from severe existential distress related to issues of their past (guilt, remorse, unfulfilled expectations), their altered self image (dependency, altered family and professional roles), and their future (dashed hopes, meaning in the time remaining, fear of death).15, 16

The optimal care of patients with advanced cancer is complex, and patients require an integration of approaches to best control the underlying illness with approaches to optimize physical and emotional symptom control and social support. The European Society of Medical Oncology (ESMO) has developed a task force as part of a quality-improvement program to improve the quality of supportive and palliative care delivered by oncologists in Europe and around the world. The ESMO policy regarding the role of the oncologist in the delivery of supportive and palliative care has been developed by this taskforce. Other educational initiatives are in the process of development.

Until now, there have been no studies describing the degree to which European oncologists are involved in the care of patients with advanced cancer, their participation in supportive and palliative care, or their attitudes toward this dimension of cancer care. The EMSO, as part of a quality-improvement initiative, chartered the Supportive and Palliative Care Taskforce to survey the ESMO membership.

Objectives

The objectives of this survey were to obtain valid and accurate descriptive data regarding the practice and attitudes of ESMO oncologists in relation to the supportive and palliative management of patients with advanced and incurable cancer and oncologist-related barriers to the provision of optimal supportive and palliative care.

Methodology

Based on a focus group of oncologists participating in the Palliative Care Working Group of the ESMO, a survey tool was developed. The tool was reviewed by oncologists with the working group and the executive of ESMO for content validity. The final version of the survey tool (Table 1) addressed demographics (7 items), collaboration with supportive/palliative care clinicians (7 items), direct involvement in palliative and supportive care (16 items), and attitudes (24 items). The survey was sent to all ESMO members.

Table 1. European Society of Medical Oncology Supportive/Palliative Care Survey
  1. ESMO: European Society of Medical Oncology.

  2. a

    The definition of “advanced cancer” is any incurable cancer, including most metastatic or locally advanced solid tumors.

inline image

Data from returned responses were entered into an Access data base for subsequent analysis. A frequency analysis was performed on data relating to oncologist demographics, level of collaboration with supportive/palliative care clinicians, oncologist practice in supportive/palliative care, and attitudes toward supportive/palliative care. Summary indices for attitude to palliative and supportive care, collaboration with palliative and supportive care clinicians, and direct involvement in palliative and supportive care were calculated. For collaboration with and direct involvement in palliative care, related items were calculated as an average score on a 0–2 scale. Because attitudes can be negative, the 5-point attitude items in the survey were scores on a scale of − 2, − 1, 0, + 1, and + 2. The attitude index was calculated as an average score for the items evaluating oncologist attitudes toward palliative care. Stepwise regression analyses were performed to evaluate the factors that contributed to attitudes toward palliative care and sensitivity to burn out from working with patients with advanced cancer.

RESULTS

Demographics

Of 2600 surveys that were mailed to ESMO members, 895 completed questionnaires were received. A total of 64 countries were represented. In all, 738 responses (82.5%) were received from 35 European countries (Germany, 147 responses; Italy, 87 responses; France, 57 responses; Spain, 48 responses; United Kingdom, 42 responses; and The Netherlands, 36 responses), and other responses were received from ESMO members in America (108 responses; 12.1%), Asia (23 responses; 2.6%), Australia (20 responses; 2.2%), and Africa (6 responses; 0.7%).

Most of the responding oncologists were male (n = 701 respondents; 78.3%). The median respondent age was 45–49 years, and the median experience in oncology was 15–19 years.

The largest proportion of respondents worked in teaching hospitals (n = 334 respondents; 37.3%). Other respondents worked in comprehensive cancer centers (n = 185 respondents; 20.7%), community hospitals (n = 176 respondents; 19.7%), or private oncology practice (n = 167 respondents; 18.7%).

Overall, most respondents were very involved in the care of patients with advanced cancer: Six hundred fifteen respondents (68.8%) reported that the management of patients with advanced and incurable cancer represented a substantial proportion of their practice, and another 197 respondents (22.0%) reported that such patients constituted most of their practice. Few respondents reported that patients with advanced and incurable cancer represented a small proportion of their practice (8.7%), and very few oncologists had no involvement in the management of patients with advanced cancer (0.4%).

Attitudes

Greater than 75% of responding oncologists agreed (agree or strongly agree) to each of the following statements: All advanced cancer patients should receive concurrent palliative care, even if they are receiving antitumor therapies (92%); I derive satisfaction from managing the physical symptoms of my patients (89.3%); all cancer centers should have a palliative care service (89.1%); medical oncologists should coordinate the care of cancer patients at all stages of disease, including end-of-life care (88.4%); medical oncologists should be expert in the management of the physical and psychological symptoms of advanced cancer (88.1%); I am usually successful in managing my patients' pain (86.9%); I read journals and papers related to the palliative care of advanced cancer (83.6%); and I own a textbook of palliative care (76.0%).

A lesser but significant consensus (60–75% agree or strongly agree) was expressed for the following attitudes: The medical oncologist is the best person to coordinate the palliative care of patients with advanced cancer (74.5%); I derive satisfaction from my work managing patients with advanced cancer and dying patients (74.0%); I am interested in participating in research in palliative treatments of advanced cancer (73.3%); I have a close working relationship with the palliative care (or hospice) services in my region (70.3%); and I am expert in the management of the physical and psychological symptoms of advanced cancer (60.4%).

Greater than 65% of respondents disagreed (disagree or strongly disagree) with the following attitudes: Palliative care begins where medical oncology ends (84.3%); dying patients do not belong in the oncology ward (73.5%); I deal with palliation in the nondying patients (symptoms management), but not with the palliation of the dying patient (end-of-life care) (70.4%); managing patients with advanced cancer and dying patients depresses me (68.2%); palliative care specialists steal patients who would otherwise benefit from medical oncology (68.0%); and I would rather have someone else look after my dying patients (66.0%).

Opinions were polarized regarding five items (Table 2): I received good training in palliative care during my oncology fellowship (residency); I feel emotionally burned out by having to deal with too many deaths; most medical oncologists I know are expert in the management of the physical and psychological symptoms of advanced cancer; a palliative care specialist is the best person to coordinate the palliative care of patients with advanced cancer; and palliative care (or hospice) physicians do not have enough understanding of oncology to counsel patients with advanced cancer regarding their treatment options.

Table 2. Attitudes in which Respondents Had Polarized Opinions
AttitudeAgree/strongly agree (%)Disagree/strongly disagree (%)
I received good training in palliative care during my oncology fellowship (residency).52.842.0
I feel emotionally burned out by having to deal with too many deaths.33.855.6
Most medical oncologists I know are expert in the management of the physical and psychological symptoms of advanced cancer.37.541.8
A palliative care specialist is the best person to coordinate the palliative care of patients with advanced cancer.36.339.4
Palliative care (or hospice) physicians do not have enough understanding of oncology to counsel patients with advanced cancer regarding their treatment options.35.239.2

Collaboration with Supportive and Palliative Care Professionals

Responses to the question, “in your dealings with patients with advanced cancer, how often do you collaborate with…” are summarized in Table 3.

Table 3. Frequency (Percent) of Collaboration with Other Professionals when Managing Patients with Advanced Cancer
ProfessionFrequency of collaboration (%)
NeverOccasionallyOften
Social worker13.938.247.9
Home hospice (palliative care) team19.143.137.8
Palliative care medical specialist27.337.635.1
Psychologist17.449.333.3
Palliative care nurse specialist33.235.131.7
Inpatient hospice team29.044.726.4
Psychiatrist22.462.714.9

Direct Participation in Supportive and Palliative Care of Patients with Advanced Cancer

Responses to the question, “in your dealings with patients with advanced cancer, how often are you directly involved with…” are summarized in Table 4.

Table 4. Frequency (Percent) of Direct Administration of Supportive and Palliative Care for Patients with Advanced Cancer
Treatment categoryFrequency of administration (%)
NeverOccasionallyOften
Treating cancer pain0.46.393.3
Treating dyspnea0.427.572.0
Managing fatigue0.815.483.8
Managing nausea and emesis0.39.190.6
Managing the complications of chemotherapy0.78.391.1
Managing the psychological consequences of advanced cancer (depression and anxiety)1.733.464.9
Managing delirium22.665.611.8
Managing constipation or diarrhea0.717.282.1
Discussing end of life care preferences with patients6.437.056.6
Directly administering end of life care to dying cancer patients14.042.843.2
Coordinating meetings with the family of dying patients14.446.139.4
Recommending an inpatient hospice20.052.327.6
Managing intestinal, biliary, or ureteric obstruction7.554.238.3
Managing spinal cord compression6.963.929.2
Managing existential or spiritual distress12.956.627.5
Patients who request elective death (euthanasia or assisted suicide)61.836.02.1

Multivariate Analyses

Stepwise regression analyses were performed to evaluate the factors that contributed to a positive attitude toward the oncologist's role on supportive/palliative care (Table 5) and factors that contributed to a predilection to burn out from managing patients with advanced cancer and dying patients (Table 6). In the first analysis, the variables evaluated included age, gender, years of experience, degree of involvement in the management of advanced cancer, place of work, average collaboration index, and the average direct participation in palliative care index. The analysis of factors that contributed to burn out included all of those used in the first analysis and also the average attitude toward the palliative care index. The statistical description of summary scores of the variables included in the multivariate analyses is presented in Table 7.

Table 5. Analysis of Variance: Multivariate Analysis of the Dependant Variable “Positive Attitude toward the Role of the Oncologist in Supportive/Palliative Care”
Contributing variableBSignificancea
  • PC: palliative care.

  • a

    Correlation coefficient (R2) = 0.223.

Working in a comprehensive cancer center−0.1310.000
Average direct involvement in PC administration index0.6140.000
Working in a community hospital−0.0810.013
Female gender−0.0770.013
Average collaboration with PC index0.0670.038
Table 6. Stepwise Regression Analysis for the Dependant Variable “I Feel Emotionally Burned Out by Dealing with Too Many Deaths”
Contributing variableBSignificancea
  • a

    Correlation coefficient (R2) = 0.150.

Average attitude toward palliative care index−0.6840.000
Female gender−0.2290.007
Age0.0430.030
Table 7. Descriptive Statistics of the Summary Scores of the Variables Included in the Multivariate Analyses
CharacteristicMeanSD
  1. SD: standard deviation.

Age4.821.756
Gender (1 male, 2 females)1.210.411
Years of experience3.951.516
Involvement in management of patients with advanced cancer3.120.558
Private practice0.190.390
Practice in a community hospital0.200.399
Practice in a teaching hospital0.370.484
Practice in a comprehensive cancer center0.200.404
Average attitude to palliative care index0.7560.422
Average palliative care collaboration index1.090.426
Average direct involvement index1.420.280

DISCUSSION

The management of patients with advanced cancer is central to the role of the medical oncologist, and it constitutes a major part of the clinical workload of most oncologists. Overall, this survey of the ESMO membership demonstrates a high level of consensus regarding the importance of the oncologist's role in the provision of supportive and palliative care for patients with advanced cancer and for dying patients. Most medical oncologists believe that oncologists should coordinate the care of patients with advanced cancer and dying patients, that supportive and palliative care should be initiated during the active treatment phase, that medical oncologists should be expert in the management of the physical and psychological symptoms of advanced cancer, and that cancer centers should provide supportive and palliative care as part of their basic array of services.

The survey results demonstrated the following discrepancies relating to practice and skill levels of the respondents: 1) Although 88.4% of respondents agreed that medical oncologists should coordinate the care of cancer patients at all stages of disease, including end-of-life care, it was found that actual practice was much less thorough: Only 43% of respondents commonly coordinate the care of cancer patients at all stages of disease, including end-of-life care; < 50% of respondents collaborate commonly with any supportive/palliative care clinician (including social workers); only 39% of respondents commonly coordinate meetings with the family of dying patients; and only 11.8% of respondents commonly manage delirium (despite the high prevalence of this problem among patients with far advanced cancer).

2) Whereas 60.4% of respondents reported that they were expert in the management of the physical and psychological symptoms of advanced cancer, only 37.0% of respondents said most oncologists they knew were expert in the management of the physical and psychological symptoms of advanced cancer. This indicates either a self-selection bias regarding the respondents favoring those with greater expertise, or a tendency to either inflate self-evaluation and/or deprecate colleagues.

3) Whereas 70.4% responded that they have a close working relationship with the palliative care (or hospice) services in their region, reported levels of collaboration with palliative care and hospice services were relatively low: Only 37.8% of respondents often collaborated with a home hospice (palliative care) team, and only 35.1% or 33.3% of respondents often collaborated with a palliative care medical specialist or nursing specialist, respectively.

Analysis of the summary scores for attitudes regarding supportive and palliative care identified that 15–20% of the responding medical oncologists have pervasively negative views regarding the oncologist's role in the supportive/palliative care of patients with advanced cancer and end-of-life care. This subgroup reported low levels of perceived self-efficacy in the management of physical and psychological symptoms, low levels of satisfaction from managing patients with advanced cancer, sensitivity to burn out from involvement with dying patients, and a predilection toward depression from their involvement with dying patients. They were more likely to indicate that it is not part of the oncologist's role and to endorse the idea that palliative care specialists should manage these patients. Despite that, these clinicians demonstrated low levels of collaboration with supportive/palliative care physicians. Negative attitudes toward the oncologist's role in supportive/palliative care were much more prevalent among oncologists working in comprehensive cancer centers and community hospitals than among oncologists in private practice or in teaching hospitals.

The following issues toward which there was a polarization of attitudes are worthy of specific attention: 1). A substantial 42% of respondents reported that they had not received adequate training in palliative care during their residency training. This finding is consistent with previously published data regarding the training of medical oncologists in the management of cancer pain,17–19 communication skills,20, 21 and palliative care.19, 22

2) In response to the statement, “most medical oncologists I know are expert in the management of the physical and psychological symptoms of advanced cancer,” more respondents disagreed (41.8%) than agreed (37.5%). It seems, therefore, that most oncologists have a low assessment of their colleagues' readiness to manage the physical and psychological symptoms of advanced cancer. This is consistent with the findings regarding oncologist training in these aspects of patient care.

3) Just over one-third of respondents reported that they felt emotionally burned out by having to deal with too many deaths. In this survey, this predilection to burn out was associated closely with negative attitudes toward involvement in supportive and palliative care (univariate and multivariate analyses) and low levels of actual involvement or referral to specialist colleagues (univariate analysis only). Burn-out, therefore, was not associated as much with over-exposure; rather, it was related more to poor attitudinal preparedness and a limited role definition that did not include personal involvement in the organization of or the provision for palliative and supportive care. Conversely, physicians with positive attitudes toward end-of-life care and their role in it expressed a much lower tendency to burn out, despite the fact that they were involved much more personally in the active care of their dying patients.

The phenomenon of burn-out among medical oncologists has been described well.23–28 It is associated with depression and depersonalization,23, 24, 28 it impacts adversely on patient care, and it is associated with low levels of preparedness for emotionally taxing professional activities, such as communication with patients and their families and end-of-life care. Physicians who manifest burn-out and depersonalization are more likely to endorse euthanasia or assisted suicide as appropriate end-of-life care options.29

4) Respondents were divided regarding who should coordinate the palliative care of patients with advanced cancer. Although 74.5% of respondents agreed that the medical oncologist is the best person to coordinate the palliative care of patients with advanced cancer, when asked about the role of palliative care specialists, 36.3% of respondents agreed that they are the best professionals to manage the palliative care of the same patient population. This ambivalence is evidenced by the relatively low proportion of oncologists who frequently consult with a palliative medicine physician and by the relatively high prevalence of respondents (35.2%) who felt that palliative care (or hospice) physicians do not have enough understanding of oncology to counsel patients with advanced cancer regarding their treatment options.

The limited data available from a survey of American oncologists suggest that many of the issues highlighted in this survey also may be relevant to oncologic practice in the U.S. The 1998 survey by the American Society of Clinical Oncology, presented in 1998,30 found that 20–25% of the responding oncologists reported that end-of-life care was the worst part of being an oncologist and that 10% experienced a sense of failure when their patients became terminally ill. The overwhelming majority of responding oncologists perceived themselves as having a lot of expertise in managing pain, constipation, nausea/emesis, fever, and neutropenia but much less ability in treating depression especially and loss of appetite. Among the barriers to end-of-life care reported in that survey were unrealistic patient/family expectations, family conflicts, lack of reimbursement for end-of-life care, and a lack of availability of palliative care specialist consultations.

The conclusions drawn from this survey are tempered by the selection bias of survey responders versus nonresponders. Data from the survey suggest that, overall; the responders may represent disproportionately those clinicians with interest in supportive and palliative care. Although 60.4% of respondents reported that they were expert in the management of physical and psychological symptoms among patients with advanced cancer, only 37.5% reported that most oncologists that they know shared similar expertise. These data suggest either a selection bias, a tendency to elevated self-appraisal of responders, or a deprecation of colleagues.

Irrespective of this potential bias, we believe that the findings of the study are highly relevant for the practice of medical oncology in Europe. Indeed, the findings underscore the need for initiative to address common shortcomings in oncologist training and practice regarding the palliative and supportive care of cancer patients with advanced and incurable disease. Clinicians who have a positive attitude toward the role of the oncologist in providing palliative care for their patients are much more likely to deliver that care. Education, therefore, must address not only the technical side of care but also the issues pertaining to role definition. It is important to highlight the care of incurably ill cancer patients with regard to the moral imperative of care and nonabandonment, the skills needed to deliver care to the patient and the family, and strategies to minimize the risk of burning out from the frequent exposure to this high-stress situation.

The gulf between widely expressed, positive attitudes and much more limited implementation suggests the need for infrastructural changes to facilitate better actualization of palliative care in clinical practice. Indeed, the findings of this survey suggest that comprehensive cancer centers and community hospitals may have particular need for this sort of infrastructural intervention to make palliative care a normal part of service delivery for patients with advanced cancer.

Ancillary