What Is the Evidence That Palliative CareTeams Improve Outcomes for Cancer Patients and Their Families?

What Is the Evidence That Palliative CareTeams Improve Outcomes for Cancer Patients and Their Families?

Patients with advanced cancer experience a complex web of problems, all of which interact. Specialist palliative care services have developed to meet these needs, but their effectiveness should be considered. We sought to determine whether specialist palliative care teams achieve their aims and improve outcomes for patients with advanced cancer and their caregivers, in terms of improving symptoms and quality of life and/or reducing the emotional concerns of family caregivers. We conducted a systematic review, searching standard databases augmented by reference lists of earlier reviews. The review focused on specialist (ie, with trained and dedicated professionals) palliative care in the home, hospital, or designated inpatient settings for patients with cancer. Outcomes were pain, symptoms, quality of life, use of hospital services, and anxiety. Studies were excluded if they did not test specialist palliative careservices. We identified 8 randomized controlled trials and 32 observational or quasi-experimental studies. Overall, the evidence demonstrated that home, hospital, and inpatient specialist palliative caresignificantly improved patient outcomes in the domains of pain and symptom control, anxiety, and reduced hospital admissions. The results suggest that specialist palliative care should be part of care for cancer patients. Although the appraisal of evidence found improvements across domains, there is a need to understand better the effects of different models of palliative care and to use standardized outcome measurement.


Patients with advanced cancer experience a complex web of problems, all of which interact. These include profound symptoms, which, unless alleviated, result in great suffering for the person with cancer and his/her family,1 and emotional, social, and spiritual consequences associated with cancer, disability, and facing the end of life, for patients, their families, and those close to them.2,3 Palliative care-holistic multiprofessional care with skilled assessment and management of symptoms and problems-seeks to alleviate these problems and to enable patients to live well for as long as possible, to die with comfort and dignity, and to support the family.

There are many reasons why palliative care is and will continue to be increasingly important in cancer. First, unfortunately death from cancer continues to increase. By 2020, the World Health Organization estimates that globally more than 15 million people will experience cancer, and 10 million will die of it each year,4 a 60% increase in deaths. By 2020, in North America, the number of people who die annually of cancer will increase by 51% to just under 1 million. An 83% increase is predicted for South America, a 31% increase for Western Europe, and a 66% increase for sub-Saharan Africa5 (Table 1).


Deaths From Cancer in Europe and Selected Other Parts of the World: 2000 and Predicted for 2020

Second, despite extensive efforts to prevent and cure cancer, overall 5-year survival from cancer is only between 50% and 60% even in high-income countries. Optimal treatment combined with early detection leads to a high rate of cure or remission for some cancers (eg, of the cervix, breast, oral cavity, and colon and lymphomas).5 Yet, survival remains poor for other cancers (eg, of the esophagus, pancreas, and lung). Many of the new treatments bring lengthened survival-and although this is much desired for most patients, it brings with it a longer period of advanced cancer, where symptom relief and support are needed. Third, the general aging of the population means that people who have cancer increasingly are older. Other chronic and progressive diseases are more common at old ages, meaning that more and more cancer patients have comorbid conditions, which bring other symptoms and emotional, social, and spiritual problems. The effects on the family can continue long into bereavement, affecting their subsequent health and well-being.2-4

In response to all these needs that numerous studies showed were not well met among cancer patients, and the increased demands of patients and families for effective and coordinated support at this phase of the illness, more than 100 countries of the world have developed specialist palliative care and hospiceservices. As Dame Cicely Saunders, the founder of the modern palliative caremovement, explained: „Palliative care begins from the understanding that every patient has his/her own story, relationships, and culture and is worthy of respect as a unique individual. This respect includes giving the best available medical care and making the advances of recent decades fully available, so that all have the best chance of using their time well.6,7“ The World Health Organization defined palliative care in 2002 as „…an approach that improves the quality of life of patients and their families facing the problem associated with life-threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial, and spiritual.8“


Frequently, palliative care is categorized as9:

  • (1) Generalist palliative care (which is the requirement of all clinicians, as most meet patients and families at some stage who are reaching the end of life): Generalists comprise practitioners not exclusively concerned with specialist palliative care,10 for example, general practitioners, district nurses, care home staff, and oncologists, who look after a range of patients and so cannot be completely up to date in all the complex treatments.
  • (2) Specialist palliative care (which is offered by professionals specifically trained in palliative and hospice care, including symptom management, emotional support, communication, and spiritual and social support). In the United Kingdom, Australia, and some other countries, palliative medicine is a recognized medical specialty, which has a 4-year training program just as other medical specialties. Specialist palliative care can be provided in dedicated inpatient units, home care programs, hospital support teams, and day care and outpatient clinics.11 This review focuses on the first 3 services, as these are the most common.

There is a difference in the way that some terms are defined and understood between the United States and most other countries (eg, United Kingdom, Australia, Canada, Europe, Africa). In Europe/Australia and so on, the terms palliative care and hospice are often used interchangeably and mean the same things.11 The term palliative care is used more often in modern times in Europe and Canada, because in French „hospice“ means hospital. Sometimes, the word hospice is used only to refer to an inpatient unit. In the United States, the term hospice refers to a program that is primarily home care, and palliative care tends to refer to hospital-based services and teams.11 For the purpose of this review and simplicity, the European/Australasian definitions will be used, but where the situation is different for the United States, this will be noted.

How might specialist palliative care services work? What outcomes should be measured?

Specialist palliative care teams provide a detailed holistic assessment, review, and management of the patient and family. Dame Cicely Saunders defined the patient and family as the unit of care.6,7 Palliative care takes a problem- and goal-orientated approach to care, assessing the main problems for the patient and family.8,11-13 Teams address symptoms and physical, emotional, social, and spiritual problems, prioritize these, and attempt to address them, also attempting to coordinate care. They often advise generalist services, as well as directly provide care.

The difficulties to undertake evaluations on palliative care are well rehearsed in the literature. There are major challenges to undertaking evaluation research on palliative care both because of the ethical and practical difficulties of engaging very ill people and their families and the range of interventions and population groups viewed as in the domain of palliative care.14 Evaluations incorporate varied designs, and many are observational.15 This reflects the difficulties in undertaking randomized trials in this area, notably attrition and ethical concerns when involving people at the end of life in research studies. Observational studies can provide valuable understanding on the effectiveness of palliative care services, particularly in the community.16,17 The purpose of this review was not to reiterate these difficulties, but to cast the search net wide enough to consider the evidence on the effectiveness of specialist palliative care teams (SPCTs) for people with advanced cancer. This encompasses both randomized and observational or nonrandomized studies (NRSs).


We have undertaken 2 previous reviews on the effectiveness of palliative care in general for people with cancer18 and specifically SPCTs for cancer and noncancer populations.15 The former review drew on the latter. The reviews demonstrated the positive effect of palliative care teams on patient outcomes and small benefit for patients in managing pain. Both reviews demonstrated support for different inpatient palliative care and home care teams, but identified the need for more work to compare and identify which team model is most effective in achieving patient outcomes and what are the active components in such teams. There have been several other significant reviews that also considered palliative and end-of-life care services.19,20 However, many of the reviews included a range of services, some of which were not specialist palliative care services, but were focused on hospice, at home, or single specialty services. In addition, a special issue of the Journal of Pain and Symptom Management in 2009 was dedicated to health economic issues of palliative care services and included appraisals of several country-specific studies.20-25 This review updates these previous reviews but focuses specifically on the effectiveness of SPCTs for cancer patients.

Review Aims

In this review, we sought to address the following: Do SPCTs achieve their aims and improve outcomes for patients with advanced cancer and their caregivers, in terms of improving symptoms and quality of life and/or reducing the emotional concerns of family caregivers? We also sought to comment on the quality of the evidence and gaps and whether there was support for some models more than others.


Data Sources

The systematic review protocol26 drew on, and updates, our previous effectivenessreviews on palliative care teams and end-of-life care15 and palliative care for people with cancer.18 The search protocol intended to identify randomized studies and NRSs evaluating SPCTs for people with cancer and searched MEDLINE, EMBASE, CINAHL, CancerLit, and PsychInfo, from 2000, the end of our earlier reviews, to end of 2009. We used the following key words: palliative care or hospices, terminal care, terminally ill, palliat$, hospice$, dying$, end of life; and neoplasms or cancer$; and service or team, primary health care, community health services, hospitals; and effective or evaluate, random, methods, economics, statistics, trends. We searched the reference lists of previous and new reviews.

Study Selection

Studies were included that compared a specialist palliative care team (SPCT) with usual care (present or historical) and incorporated either a randomized study or NRS design. A SPCT comprised 2 or more health workers; at least one had specialist training in palliative care, who provided specialist palliative carethrough education and support to generalist health care workers and/or direct palliative care to patients with cancer. Trials of SPCTs were included when they aimed to provide specialist palliative care in a community, hospital (inpatient/outpatient), and/or hospice setting that aimed to help those with advanced cancer, and their caregivers, to live as well as possible until death. Caregivers were defined as family, friends, or significant others. Usual care comprised conventional community and general hospital/oncology services. Outcomes included pain and symptom management, quality of life and death, and patient and carer satisfaction/morbidity before and after bereavement. We included trials of palliative care teams that had recruited exclusively patients with cancer, or trials of mixed patient populations where randomization was stratified by diagnosis and data for the cancer patient subgroup could be obtained. We included only English-language articles.

Studies were rejected at full article review if they used an un-stratified population sample of noncancer and cancer groups defined as terminally ill or requiring palliative care. Although all included patients with cancer, the analyses compared control and intervention groups with no analysis by disease group.27-32 Other studies were excluded on full article review as they did not evaluate a SPCT, including hospice care,33,34 palliative day care,35 Macmillan Services,36 home-cancer-health-care programs,37 and palliative care services by region,38individuals working as nurse-led service that was not a SPCT,39 or a coordinating service.40

Validity Assessment and Data Extraction

There is a lack of general guidance on quality appraisal applicable for differing research methods that include randomized and NRS. We limited the inclusion of nonrandom studies by study design or methodological quality and examined in the data analysis the degree of susceptibility of the included NRS designs to bias.26 Nonrandom studies were excluded when they did not assess effectivenessby comparing between the intervention and control, or at defined time points, or had no defined outcome measures. The validity assessment of the included studies used 2 hierarchy of evidence: study design and risk of bias.26 The methods of grading and data extraction procedures were piloted and cross-checked, before full data extraction. The assessment of study design used Cochrane Grade, which is applicable for randomized studies and NRSs and uses 4 quality levels: high, moderate, low, and very low (Table 2). Nonrandom studies are, however, generally assessed as low, but can be judged higher if indicated by a large magnitude of effect or lack of concern about confounding.26 A risk of bias was assessed for each included study and documentation for the judgment recorded (eg, study design).26


Grade of Evidence

In assessing design, we followed these main principles:

  • In palliative care, the question of interest cannot be always answered by randomized trials, and review authors may be justified in including NRSs.
  • Potential biases are likely to be greater for NRSs compared with randomized trials, so results should always be interpreted with caution when they are included in reviews. Particular concerns arise with respect to differences between people in different intervention groups (selection bias) and studies that do not explicitly report having had a protocol (reporting bias).
  • Eligibility criteria, data collection, and critical assessment of included studies place an emphasis on specific features of study design (eg, which parts of the study were prospectively designed) rather than „labels“ for study designs (such as case-control vs cohort).
  • Risk of bias in NRSs can be assessed in a similar manner to that used for randomized trials, although more attention must be paid to the possibility of selection bias.


In this review, we have focused on meta-synthesis, that is, combining the results in tabular format without attempting meta-analysis. We contrast the strength of the evidence on the effectiveness of SPCTs for cancer patients and specifically which team model, and its components, is most effective.

Details of the study designs were extracted into Review Manager software from the Cochrane Collaboration to examine their strengths and limitations. Data were also extracted into standardized tables to examine potential solutions, to contrast methods, study settings, types of intervention, and main outcomes.


From the searches, we identified 40 trials that considered different SPCTs (Table 3). These included 8 randomized controlled trials (RCTs)-the remainder of studies were observational, usually with a control group, although a few without. There were few very low-grade studies, but many of the designs graded very low were excluded by our methods.


Studies Included in the Review-Authors, Grade, Setting, Type of Intervention, and Main Outcomes

One of the randomized trials failed to deliver results and so was graded low; the remainder were graded moderate to strong, depending on the ways they accounted for biases, sample size calculations, and other features of design. Five of the RCTs were from the United States; there was one each from the United Kingdom, Canada, and Norway. Two recent trials (2009 and 2010) suggested positive results, in terms of improved symptoms/mood and/or quality of life; the earlier trials were more equivocal, except for the early RCT of a home care team by Zimmer et al.51,52 All trials suffered from problems of sample size and were often underpowered or suffered from attrition. No trial reported negative effects of a palliative care service. The trials ranged across all types of service-inpatient, home care, and hospital support-and in some instances combinations of these.

The remaining 32 studies were generally graded moderate or low, except in a few instances where patients in the control group were matched to those in the intervention group (Table 3). The evidence (moderate and low) supports SPCTs working in home, hospitals, and inpatient units as a means to improve outcomes for cancer patients, such as pain, symptom control, and satisfaction, and in improving care more widely, including reducing hospital admissions. The benefit is demonstrated quantitatively. Studies indicated either benefit in favor of a palliative care team or no difference. Some studies suggested lower costs. Quality of life, when measured, less often was different between groups and tended to deteriorate over time. However, there were biases in all studies, even when there were control groups, because of selection bias of patients into the palliative careservice or control.

Given the variety of interventions within each team and the wide range of services tested, more work is needed to test the specific components of palliative careteam activity (eg, to compare different types of hospital team or inpatient unit or to test specific ways of working within their practice) and to discover if a different skill mix or interventions performed by the team are more effective than each other. More recent trials considered more integrated palliative care and oncologyservices, with joint activity and earlier referral to palliative care, and this seemed to provide benefit.

Studies tended to be from large teaching hospitals in urban areas, and only a few were multicenter, and included rural areas, although there was no difference in the nature of the results in these areas.


In this systematic review, we identified 8 RCTs and 32 observational or quasi-experimental studies offering a mixture of high-, moderate-, and low-grade evidence. Taken together, the results indicate positive benefits for specialist palliative care services of hospital teams, home care teams, and inpatient services. The main outcomes that showed improvement were pain and symptom control, hospital admissions or stays, and, in some studies, caregiver outcomes. Studies also found improved satisfaction for patients and caregivers and, in some instances, lessened anxiety or depression. Quality of life, when measured, showed fewer differences.

Contrary to earlier reviews, where most evidence was available for home care and inpatient services,15,18 there is now evidence for home, hospital, and inpatient services, with perhaps least evidence now for inpatient services. This is partly because most of the randomized trials are from the United States, where home care services predominate. A review found that palliative care teams have benefit in the management of patients with HIV and AIDS,100 although the evidence in this field was weaker. Separate reviews have considered the factors that enable patients to remain at home,101 and our review lends support to their finding that home support is an important factor in patients remaining at home, as it was primarily the home palliative care services that increased home care.

Further work is needed to investigate all types of team, to compare models of service, and to consider the impact of the different mix of individuals within teams. One way to achieve this is by better routine data collection within palliative care services. Currow et al102 and Eagar et al103 have developed the Palliative Care Outcomes Collaboration, an Australian initiative that allows specialist palliative care services to collect nationally agreed-upon measures to better understand quality, safety, and outcomes of care. Data include phase of illness, place of discharge, and, at each transition in place of care, the person’s functional status, dependency, and symptom scores. This allows benchmarking of services and comparison between different initiatives, which would now be valuable in further developing palliative care services.

Palliative and end-of-life care has traditionally emphasized supporting people with cancer, but since the 1980s104 and continuing today,105 U.K. policy has advocated extension to all population groups at the end of life. Many of the evaluations on palliative care reflected this with mixed populations of patients with either cancer or noncancer diagnoses. Publication of subgroup analysis by condition type could develop the evidence base on the effectiveness of palliative care for different population types, for example, the very old.

Strengths and Limitations

We were able to identify and include a wide range of robust literature, focusing more closely on specialist palliative care services and overcoming some of the weaknesses of earlier reviews that included specialist and nonspecialist services. However, our review was still weakened by the wide range of outcomes measured. There is a need to standardize outcome measurement in palliative care. One European-funded collaborative, PRISMA, is seeking to harmonize outcome measurement, which may help in the future.106 This is building on measures such as the Palliative Outcome Scale and the Support Team Assessment Schedule, which are now widely used to make available resources to enable their integration into research and practice.


The review shows a benefit of SPCTs in cancer, in terms of pain and symptom management and psychosocial and health care outcomes. This suggests that specialist palliative care should be part of oncology practice, and if the more recent studies hold true, be introduced earlier into practice, as this may prevent subsequent symptoms and inappropriate treatments.


  1. Solano JP, Gomes B, Higginson IJ. A comparison of symptom prevalence in far advanced cancer, AIDS, heart disease, chronic obstructive pulmonary disease and renal disease. J PainSymptom Manage. 2006;31:58-69.
  1. Davies E, Higginson IJ. Better Palliative Care for Older People. Denmark: World Health Organization; 2004. Available at: www.euro.who.int/document/E82933.pdf.
  1. Davies E, Higginson IJ. Palliative Care: The Solid Facts. Denmark: World Health Organization; 2004. Available at: www.eho.int/document/E82931.pdf.

View full references list

Palliative care; hospice; oncology; effectiveness; systematic review; pain

© 2010 Lippincott Williams & Wilkins, Inc.