Effectiveness of Hospital Palliative Care Teams for Cancer Inpatients: A Systematic Review

Abstract

Background:

Although the benefits of palliative care in the outpatient setting are well established, there has been little to support the employing of hospital palliative care services for inpatients with cancer.

Objective:

We conducted a systematic literature review to evaluate the effectiveness of palliative care for cancer patients in the acute inpatient hospital setting.

Methods:

Two electronic databases—PubMed and CINAHL Plus—were searched for articles published between 1 January 2005 and 28 May 2015. The search was augmented by hand-searches of specific journals and by examining the reference lists of short-listed articles. Studies were included if they evaluated a hospital palliative care service for cancer patients. Data extracted included study design, patient population, study setting, composition of the team, nature of the intervention, outcomes measured, and main findings.

Results:

No randomized controlled trials were found. There were 14 pre-post studies that evaluated patient outcomes, of which only 2 had a control group. We also reviewed a further seven studies that evaluated other aspects of the palliative care intervention. The studies were not robust enough to confirm the efficacy of hospital palliative care services for cancer patient outcomes. Nonetheless, published studies provide a glimpse into the wider benefits of palliative care interventions.

Conclusions:

Data to support the benefit of palliative care interventions in the inpatient acute hospital setting are still lacking. Future studies should employ innovative strategies to further this field of research.

Background

The European Society for Medical Oncology (ESMO) and the American Society for Clinical Oncology (ASCO) advocate the role of palliative care in comprehensive cancer care.^1,2 Data showing a positive impact on patient outcomes in numerous clinical settings have served to cement this position and fuel the expansion of palliative care services.^3–6

In the outpatient setting, landmark randomized controlled trials (RCTs) have shown that palliative care can improve patient outcomes among cancer patients.^7–10 Temel et al. showed that palliative care involvement in newly diagnosed metastatic non-small-cell lung cancer patients resulted in better quality of life, less depressive symptoms, less aggressive end-of-life care, and even longer median survival.⁹ This has prompted ASCO to issue a provisional clinical opinion, recommending that combined standard oncology care and palliative care be considered early in the course of illness for any cancer patient with high symptom burden or with metastatic cancer.¹¹

The evidence for improvement in patient outcomes among cancer patients in the inpatient acute hospital setting, however, is less clear.^12,13 Although evaluations of hospital palliative care teams do show benefit in a range of patient outcomes such as improved symptom severity scores, reduced cost of hospital stay, and greater patient satisfaction, heterogeneity of palliative care teams and target patient populations, validity of measured outcomes, and methodological weaknesses limit the generalizable conclusions that can be drawn from these studies.^12,14–23

The aim of this systematic review is to describe recent evidence for the effectiveness of hospital palliative care teams for cancer patients. The composition of palliative care teams, descriptions of patient care interventions, and patient outcomes measured are also reviewed.

Methods

We performed a systematic literature review. The results were qualitatively synthesized based on the nature of interventions and results. We did not combine the data into one variable due to the heterogeneity of the studies.

Literature search

The systematic review protocol was based on published guidelines.^24,25 Two electronic databases—PubMed and CINAHL Plus—were searched for articles published between 1 January 2005 and 28 May 2015. The following search terms were used: palliative AND (team OR service) AND (effect* OR evaluat* OR benefi*) AND hospital AND (cancer OR neoplas*). This was augmented by hand-searching of four specific journals from 1 January 2005 to 28 May 2015: Palliative Medicine, Journal of Pain and Symptom Management, Journal of Palliative Medicine, and Journal of Clinical Oncology. The reference lists of short-listed articles were also examined. Two investigators (S.L. and G.Y.) conducted independent searches, and the final list of articles to be included for review was discussed and agreed on.

Inclusion/exclusion criteria

Palliative care is a complex intervention involving multiple components.²⁶ For the purposes of this review, we considered studies examining the effects of a hospital-based palliative care team on adult cancer patient outcomes. The terms “palliative care involvement” and “palliative care intervention” are used in relation to what the palliative care team did. An inpatient hospital palliative care consultation team refers to a team that includes at least 1 physician and that operates in the hospital inpatient setting. Papers with a broad range of study designs, outcomes, and process measures were included. Studies of adults with cancers at any stage were included. Only English-language articles were included in the review.

We excluded studies involving pediatric or adolescent patients. As this review focuses on cancer patients, studies with a mixture of cancer and noncancer patients were excluded if less than 50% of the sample population were cancer patients. Case studies, conference abstracts, letters to the editor, and editorials were excluded.

Data extraction

The following data were extracted from the short-listed studies that met the inclusion/exclusion criteria: study design, patient population, study setting, composition of the team, nature of the palliative care team's input, outcomes measured, and main findings.

Data analysis

We performed a qualitative meta-synthesis based on data extracted into tables to compare features and results of the studies. The Downs and Black checklist, with a maximum score of 32, was used to evaluate the quality of nonrandomized studies.²⁷ The Drummond checklist was used for the assessment of the quality of the economic evaluation study.²⁸ The 32-item consolidated criteria for reporting qualitative research (COREQ) checklist was used to evaluate the quality of the qualitative study.²⁹ Two authors (G.Y. and S.N.) scored each paper independently, and the average score is presented in Table 1. Where scores differed by more than 3, G.Y. and S.N. discussed and agreed on the final score.

Table 1.

Description of Study Characteristics

First author, publication year	Setting and period of study	Participant and group characteristics	PCT composition	Description of intervention	Main outcomes measures	Effect on measured outcomes	Score for quality of study
Study type: pre-post study with control group
Jack, 2006¹⁷	National Health Service University hospital, England, 1300 beds	100 patients: n = 50 (specialist hospital PCT), n = 50 (traditional care)	4 CNSs (all degree holders with additional specialist palliative and cancer care qualifications)A PC consultantA specialist registrar	Individualized assessment, psychological support, symptom control, and evaluation (e.g., appropriate analgesia)	Pain scores using self-reported Palliative Care Assessment Tool scores	All admitted cancer patients had an improvement in pain controlIntervention group demonstrated a statistically significant greater improvement than control group	18 out of 32
Kao, 2014³⁵	National Taiwan University Hospital, Taipei, January–July 2006	60 patients: n = 30 (usual care only), n = 30 (usual care plus visits from the hospital-based PCT)	Physicians trained in PCNursesSocial workersChaplains	Regular visits from palliative team providing advice on medications, education to patient and families on symptom relief and end-of-life preparation, and emotional support	Symptom Distress ScaleHospital Anxiety and Depression ScaleSpiritual Well-Being ScaleSocial Support Scale	Both groups had improvement in pain, dyspnea, and dysphagia.Intervention group had greater improvement in edema, fatigue, dry mouth, and abdominal distension.Neither group had improvement in anxiety, depression, and social support.	21 out of 32
Study type: pre-post study without control group
Bell, 2011³⁶	Queen's Medical Centre, Hawaii, 542 beds, 1 January 2005–31 December 2009	1362 patients	PC-trained physiciansNursesSocial workers	Discussions were held with patient, family, and medical team to clarify goals of care, disposition, or appropriate use of medical interventions.	Hospice referralsChanges in code status to “Do Not Attempt Resuscitation”	454 (33%) hospice referrals were generated.234 (17%) code status changes were generated.Intensive plan-of-care consultations were strongly associated with hospice referral and code status changes.	20 out of 32
Braiteh, 2007¹⁸	University of Texas M.D. Anderson Cancer Centre, USA, 480-bed unit, February 2004 and May 2004	61 patients	A PC-trained physicianA fellow	Patient review with subsequent recommendation for imaging, counseling, or medication changes depending on assessment of patient's clinical condition	Symptom identification using the ESAS, CAGE questionnaire, and Mini-Mental Score ExaminationSymptom improvement post–palliative team review	Increased symptom identification (n = 449) after palliative team review, compared with (n = 86) at point of referral38% of patients had symptom improvement within 72 hours.	12 out of 32
Centeno, 2009³⁷	Clínica Universidad de Navarra, Spain, 300-bed hospital, October 1, 2001–September 30, 2007. PCT was implemented in 2004.	All patients referred to the PCT were studied.Cancer patients were compared with a baseline group of noncancer patients.	2 doctors2 nursesA psychologistA researcher	Advisory role to other healthcare professionals within the hospital	Impact of PCT on opioid prescription trends using: Prescription of different opioids during the period both before and after implementation of the PCT by calculating yearly the DDD adjusted to 1000HS Rate of effectiveness (proportion of patients treated using opioids compared with the total estimated in need of treatment) Rate of improvement (proportion of patients potentially requiring opioids but not treated who were incorporated into the treatment group)	Total opioid prescription for cancer patients increased from 240 to 558 DDD/1000HS.Rate of effectiveness improved from 64% to 87%.Rate of improvement increased from 43% to 64%.Noncancer patients presented a steadily low opioid consumption.	10 out of 32
Chang, 2010³⁸	Taipei Veterans General Hospital, Taiwan, March 2005–January 2006	313 terminal patients	2 oncologists with more than 5 years of PC experienceA PC-qualified nurse2 social workers for socioeconomic supportA spiritual consultant	Evaluation and advice regarding patient's symptom control. Discussions regarding care disposition were also held.	Improvement in symptom control, which was assessed using a modified Chinese questionnaire similar to the Memorial Symptom Assessment Scale	Significant improvement was noted in symptoms such as pain, dyspnea, and insomnia for 162 patients.	15 out of 32
Chou, 2013³⁹	Chang Gung Memorial Hospital, Northern Taiwan, January 2006–December 2010	2887 terminally ill patients	Qualified physiciansNurse specialistsSocial workersPsychologistsA Buddhist	Symptom control, promotion of awareness of disease, and end-of-life planning	Disease awareness of the patients and their families at the beginning and end of the PCCS	Disease awareness improved from 69% pre- PCCS to 85% post-PCCS review.Longer duration of PCCS review and longer interval between date of hospital admission and referral to PCCS was associated with higher disease awareness.	17 out of 32
Delgado-Guay, 2009⁴⁰	University of Texas M.D. Anderson Cancer Centre, USA, 52-bed ICU, 1 June 2006–31 October 2007	88 ICU patients	A PC physician (certified by the American Board of Hospice and Palliative Medicine)PC and oncology fellowsAdvance practice nursesA social workerA case managerA child-life specialistA psychiatric nurse counselor	Symptom control, participation in code status and withdrawal of therapy discussions, and psycho-emotional-spiritual support to patients and families	Symptom distress and subsequent improvement, evaluated using the ESAS, the Memorial Delirium Assessment Scale or by physician assessmentOutcomes of patients in ICU, evaluated using the presence of advance directives, living wills, out-of-hospital DNR orders, medical power of attorney, and the code status of resuscitation from the admissionNumber of diagnostic evaluations and specific suggestions provided by the PCT	Conversions to DNR orders occurred in 62 patients.Withdrawal of therapies such as mechanical ventilation, bilevel positive airway pressure, total parenteral nutrition, and dialysis occurred in varying degrees.Symptom improvement was also observed in varying degrees.Mean number of pharmacological and nonpharmacological interventions per patient was 9.Coordination of transfer between ICU and the PCU occurred.	13 out of 32
Hanson, 2008²¹a	North Carolina, USA, 668-bed hospital, 1 July 2002–30 June 2005	304 seriously ill hospitalized patients referred to the PCT, 61% of whom had cancer.	PhysicianAdvance practice nurse(Both with certification in PC)Unit-based social workersChaplains	Recommendation regarding management of physical and psychological symptoms, assistance with end-of-life decision making, facilitation of communication between patient, family, and treating physicians, and education of referring physician and unit nurses about how and why to implement changes	Impact on symptoms (evaluated with the McCorkle Scale, Detroit Medical Centre nonverbal pain scale, or an original proxy rating scale derived from the National Cancer Institute Common Toxicity Criteria)Impact on end-of-life treatment (evaluated using frequency of DNR/do-not-intubate orders, new hospice referrals, and new comprehensive comfort care protocol)Effect of PC consultation on hospital costs	Consult recommendations were implemented in 88% of patients.New “DNR/do not intubate” orders occurred for 26% of patients; new comprehensive PC protocol orders occurred for 34% of patients.Symptom scores for pain, dyspnea, and nausea improved within 48 hours.27% were referred for hospice.PC cases had lower variable cost per day as compared with non-PC controls.	14 out of 32
Iwase, 2007⁴¹	University of Tokyo Hospital, Japan, at least 1000 beds, June 2004–December 2005	217 patients	A PC physicianA cancer care nurseA psycho-oncologist	Recommendations on control of physical symptoms	Symptom improvement evaluated using a modified version of the STAS	Improvement in pain, nausea, and vomiting was noted, though certain symptoms (such as anorexia, fatigue) remained hard to alleviate.	13 out of 32
Lefkowits, 2015³¹	Magee-Womens Hospital of the University of Pittsburgh Medical Centre, USA, 1 March 2012–28 February 2013	95 patients with a gynecologic malignancy	Full-time members: A physician A nurse practitioner A social workerPart-time services: Behavioral health Massage therapy Pet therapy Music therapy	Consultation for goals of care and symptom management	Magnitude in symptom burden was evaluated using a modified Edmonton Symptom Assessment System scale.Prevalence of moderate-to-severe symptom intensity was compared between the day of PC consultation, the day after PC consultation, and the last recorded symptoms before discharge.	Majority of improvement in symptom control was noted to occur within one day of consultation.	13 out of 32
Loke, 2011⁴²	A tertiary medical centre, Southern Taiwan, January–December 2009	354 terminal patients	PC physiciansPC-qualified nursesSocial workersA Chaplain	Recommendations for symptom control, psychosocial and family support, and facilitation of discussion about goals of care between PCT, primary cancer physician, patient, and their family members. Dispositional planning was also done.	Frequency of sign rate of DNR ordersPercentage of patients who had disease awarenessPercentage of hospice utilization	Increase in DNR sign rate from 42% to 71%Increase in awareness of disease prognosis from 46% to 58%64% of patients who remained in the acute hospital received hospice care.	14 out of 32
Morita, 2005¹⁴	Seirei Mikatabara Hospital, Japan, 750-bed acute care general hospital, April 2003–March 2004	111 patients were referred to the PCT. These were compared with a baseline group of 100 patients admitted to the PCU.	Full-time members: A PC physician with 12 years of clinical experience as an attending physician in the PCU An officially certified PC nurse who had 19 years of clinical experiencePart-time members: A psychiatrist 2 pharmacists A liaison nurse	Recommendations and prescription of medications for symptom control, discussion of patients during multidisciplinary team rounds	Improvement in symptom severity, evaluated using the Japanese version of the STAS	PCT patients had significant improvements in control of certain symptoms such as pain, appetite loss, and nausea.Primary role of the PCT was for symptom palliation, whereas primary role of the PCU was for end-of-life care and transfer of care.	17 out of 32
Sasahara, 2010⁴³	See Process Evaluation section for details
Study type: economic evaluation
Morrison, 2011⁴⁴	4 New York State hospitals (a community hospital, 2 academic medical centers, and a safety-net hospital), USA, 2004–2007	Patient sample included all adult patients with Medicaid as their primary and sole insurer, who had hospital stays from 6 to 44 days.Included patients either had advanced disease (cancer or noncancer) or had been in the ICU for more than 5 days (regardless of diagnosis).Patients were stratified into those who were discharged alive vs. those who died in hospital.Discharged alive, usual-care patients (n = 1427) were matched with discharged alive, PC patients (n = 296).Hospital death, usual-care patients (n = 149) were matched with hospital death, PC patients (n = 189).The subset of patients who died in hospital did not have a sufficient percentage of cancer patients and were, thus, excluded from this review.	Hospital A: 1 physician 1 nurse practitionerHospital B: 1.6 physicians 2 nurse practitioners 0.2 social worker 0.2 psychologistHospital C: 1 physician 1 nurse practitioner 1 social workerHospital D: 2 physicians 2 nurse practitioners 2 social workersPC consultation teams had been in existence for more than 5 years and had participated in training sessions conducted by the Centre to AdvancePalliative Care.	Care processes recommended by the National Quality Forum's framework were used.Consultation teams assisted patients, families, and referring teams in clarifying diagnoses, treatment options, goals of care, and hospital discharge options.	Total costs for each subject for each hospital day and for the entire admission periodAverage daily total costsICU LOS in daysLikelihood of discharge to hospice or death in the ICU	PC was associated with an average reduction in total cost of $4,098 per admission ($36,741 for usual care patients vs. $32,643 for patients who received PC).Compared with usual-care patients, PC patients were significantly more likely to receive hospice care in an appropriate setting (home, nursing home, or inpatient hospice) after hospital discharge (30% vs. 1%).No statistically significant differences in average hospital LOS or average ICU LOS were noted between usual-care patients and PC patients.	5 out of 10
Study type: process evaluation
Birks, 2011³³	St. George's University of London, 7 June 2008–28 March 2009	336 Saturday assessments of 252 patients (81.7% with cancer diagnoses)Comparison sample of 93 assessments of 78 patients (78.2% with cancer diagnoses) on 9 randomly selected weekdays	An on-site CNS with telephone support from: A training grade doctor A consultant	Consultation process involved and/or face-to-face visits with patients, family meetings, discussions with healthcare professionals, and telephone calls. Recommendations included change of medications, starting (or stopping) the Liverpool Care Pathway, and care disposition planning.	All patient assessments that resulted in a management change were considered clinically appropriate.Assessments of patients who subsequently died before Monday afternoon were considered clinically appropriate and to represent strong evidence of need.	Most of the Saturday assessments resulted in a significant change in management (57%) or were undertaken on patients close to death (10%).There were 39/336 (12%) new referrals assessed on Saturdays compared with weekdays (22%).There were few differences between the nature of the Saturday and the weekday service and no evidence of “inappropriate” l referrals.	11 out of 32
Corbett, 2013³²	The Peter MacCallum Cancer Centre, Australia, 1 January 2007–31 December 2010	Hematological malignancies	DPPC doctorsNurse consultants	Not mentioned	Proportion of hematological referrals as a percentage of all referrals received by the palliative team each yearEffect of PC involvement on the outcome of death in the ICU	Percentage of hematological referrals was noted to increase every year.Percentage of hematology inpatient deaths referred was also noted to increase every year.Patients dying in hospital who had been referred to PC were less likely to have died in ICU than those who had not been referred.	13 out of 32
Hall, 2013³⁴	Royal Surrey County Hospital, UK, 527 beds, 1 September 2011–31 August 2012	651 face-to-face weekend consultations with patients (80% diagnosed with cancer)Patients who had malignant spinal cord compression, on the Liverpool Care Pathway, who were dying or identified as needing symptom control were included for weekend reviews.	An on-site CNS with telephone support from:A consultant in palliative medicine	Symptom control and assessment using the Memorial Symptom Assessment Scale-Short Form or a specific assessment for patients on the Liverpool Care Pathway	Number and proportion of weekend consultationsProportion of patients who died over the weekend or in the early Monday morning hours	25% of total consultations and 18% of new patient consultations were undertaken on the weekends.23% of new patients died over the weekend or in the early hours of the Monday morning.CNSs identified new problems in 60% of patients during weekend reviews.	9 out of 32
Kao, 2014⁴⁵	Chang Gung Memorial Hospital, Taiwan, January 2006–December 2010	2020 terminally ill cancer patients who had no DNR documentation before being seen by the PC consultative services	PhysiciansSpecialist nursesSocial workersA Buddhist spiritual consultantVolunteers	Distress management, promotion of disease awareness, and awareness about DNR designation	Retrospective analysis of the relationship of baseline demographic variables, performance status, physical distress level, and interval from PC referral to death with the outcome of DNR designation	Of the 2020 patients who were not designated DNR at the beginning, 1301 (64.4%) of them had a DNR designation at the end of PCCS care.A longer duration of PC was associated with DNR designation.	15 out of 32
Sasahara, 2010⁴³	Showa University Hospital, Japan, 1100 bed, February 2004– March 2005	53 patientsViews of 98 nurses in the referring wards	Full-time members: A PC physician trained as a pain clinician, 2.5 years of service in the PC team A certified nurse specialist in oncology with 10 years of clinical experiencePart-time members: A psychiatristWeekly multidisciplinary case conferences also involve a social worker, link nurses, and pharmacists.	Advice on symptom management, emotional support for patient and family, and care transitions	Number of categories of support provided by the PC teamImprovement in symptom control as evaluated by the STAS–Japanese version (STAS-J) and the EORTC QLQ-C30Referring staff's views	The PC team provided a median of 3 types of support.No patient's quality of life measure improved after 1 week of referral, except for insomnia.Referring staff highly evaluated the activities of the PC team.	16 out of 32
Yang, 2012⁴⁶	An acute hospital trust, England, 1100 beds, 17 Nov 2006–16 February 2007	12 patients who had received inpatient care from the SPCS for symptom control or psychological support were discharged from the SPCS, either to general ward care or to home. (91.7% of whom had a diagnosis of cancer)10 of their caregivers	Consultants in palliative medicine: 1 on study leave, 1 full-time, 1 part-time1 part-time associate specialist in palliative medicine2 part-time specialist registrars1 full-time, 4 part-time PC CNSs2 full-time administrative staff	Symptom control, patient and caregiver support, and financial and social support. There is 24-hour access to advice from a consultant in an emergency, and patients are seen on the day of referral where requested	The interviews covered: How the patient/caregiver came to be referred to the SPCS and their previous contacts with PC services Their views of using the service Its impact on the patient/caregiver, particularly in relation to symptom control and psychological problems	Patients and caregivers saw PC's role as 3-fold: physical symptom control, psychological support, and a reliable liaison.The PCT made time for patients, giving them a sense of value and worth.	21 out of 32
Yennurajali-ngam, 2007²⁰	University of Texas M.D. Anderson Cancer Centre, USA, November 2004–January 2005	100 consecutive cancer patients who were referred to the PC mobile team	PC specialistsPC and oncology fellowsAdvance practice nursesA social workerA case managerA child-life specialistA psychiatry nurse counselor	Symptom control, PC, and end-of-life issues	Type of interventions instituted by the mobile team	The main interventions by the palliative team were changes in medication types and medication doses.Almost universal counseling services were also provided to patients and families.	13 out of 32

The study also had a component on economic evaluation of the PCT.

CAGE, acronym of four questions used in the screening questionnaire for alcohol problems. Available at http://pubs.niaaa.nih.gov/publications/inscage.htm (Last accessed June 25, 2016.); CNS, clinical nurse specialist; DDD, defined daily dose; DNR, do-not-resuscitate; DPPC, Department of Pain and Palliative Care; ESAS, Edmonton Symptoms Assessment Scale; HS, hospital stays; ICU, intensive care unit; LOS, lengths-of-stay; PC, palliative care; PCCS, Palliative Care Consultation Service; PCT, palliative care team; PCU, palliative care unit; SPCS, specialist palliative care services; STAS, Support Team Assessment Schedule.

Results

The initial search by S.L. yielded 467 unique publications, and the independent search by G.Y. identified an additional 4 publications. The titles and abstracts of these articles were screened, after which 63 potentially relevant publications were identified. The reference lists of these articles were screened, and five additional publications were identified. Hence, 68 publications were evaluated against the inclusion/exclusion criteria.

Of these, 47 were excluded for the following reasons: Less than 50% of the sample were cancer patients (n = 26), the study was not in an inpatient hospital setting (n = 10), the main aim was not to evaluate the effectiveness of the palliative care team (n = 5), team composition did not meet our definition (n = 4), the study was on pediatric and adolescent patients (n = 1), and the publication was a case study (n = 1). Hence, a total of 21 studies were included in the final review. (Table 1)

Studies were categorized according to Portela et al.³⁰ Designs included pre-post studies (with or without a control group), process evaluations, and economic evaluations. No RCTs were identified. The quality of included studies was generally poor. Of the 19 nonrandomized studies that were evaluated with the Downs and Black checklist, only 5 scored more than 16 out of 32 points. The economic evaluation study scored 5 out of 10 points in the Drummond checklist, and the qualitative study scored 21 out of 32 points in the COREQ checklist.

Palliative consultation team composition

All teams included at least a physician and a nurse, except for one study that comprised only physicians.¹⁸ The makeup of the allied healthcare staff varied. Background qualifications, specialties, and experience levels of the physicians and nurses also differed from team to team.

Patient populations

The majority of studies involved a heterogeneous group of cancer patients, except for two studies, which focused on patients with gynecological³¹ and hematological malignancies.³²

Input of the palliative care team

What the palliative care team actually did was varied and was not always described in the papers reviewed. Common interventions described by the palliative care team included symptom assessment and control, change in prescription of the number and types of medications, and discussions regarding resuscitations, withdrawal of therapies, and care disposition. The intervening palliative care teams often provided emotional support to patients and their families and also acted as facilitators of communication between patient and families and/or between patients, families, and referring teams. However, descriptions of interventions were not detailed, and it was not possible for the palliative care team to evaluate the quality of these interventions.

Among the seven process evaluation studies, two studies analyzed the results of novel weekend palliative care services introduced to two different hospitals in the United Kingdom.^33,34 Both studies consisted of a system where an on-site clinical nurse specialist would attend to patients in the hospital with telephone support from a palliative care physician.

Main outcomes of interest

Outcomes used were varied and were categorized into direct patient outcomes (such as improvement in symptom control and quality of life), management outcomes (such as care disposition and increase in comfort care orders), and economic outcomes (decrease in healthcare costs). Often, the scales used for assessment of symptom control and quality of life differed between studies, as did the measures used to evaluate management and economic outcomes.

Key results

There were two pre-post studies with control groups.^17,35 Both showed that there were additional benefits of palliative care versus usual care in terms of symptom control. The 12 pre-post studies without control groups also generally demonstrated that palliative care consultation was associated with improved outcomes, such as improved symptom control, code status changes, transition to palliative care, increased use of opioids, as well as improved disease and prognosis awareness of patients.^{14,18,21,31,36–43}

Both the economic evaluation studies noted that palliative care consultations were associated with decreased hospital costs.^21,44 Lastly, there were seven process evaluation studies. Outcomes were again generally positive.^{20,32–34,43,45,46} For the two studies that evaluated novel weekend palliative care services, both concluded that it was appropriate to have a weekend palliative care service, as this provided symptomatic patients with increased accessibility to the palliative care team.^33,34 Two studies also sought views of referring teams or patients and their caregivers, and both concluded that the interventions of the palliative care teams were often viewed very positively.^43,46

Discussion

The studies reviewed suggest that hospital palliative care teams improved pain and symptom control, increased awareness of disease and resuscitation status, and reduced the cost of hospital stays. However, the quality of included studies was generally poor, with only 6 out of 21 studies scoring more than half of the available points in the respective checklists. Therefore, the evidence from these papers is not robust enough to conclude that hospital palliative care teams are effective in improving patient outcomes.

Limitations of included quantitative studies

The first limitation of the included studies is that descriptions of what the palliative care teams actually did were not detailed. The degree of involvement of palliative care teams in patient management is also not clear. If the palliative care teams resulted in improved outcomes, it will be impossible to discern which aspect of the teams contributed to the improvement. It could be the interdisciplinary nature of the teams, the holistic approach, or the focus on symptom control. If palliative care did not result in improvement, it is also unclear whether this is due to the suboptimal quality of care provided or the insufficient exposure of the palliative care team.

Second, studies that included a control group were limited by selection bias, as nonequivalent control groups were used.^17,35 In one study, groups were determined according to whether patients were referred to palliative care or not.¹⁷ In the other study, groups were determined according to whether the patients and their families chose to receive usual care or regular visits from the hospital-based palliative care team.³⁵ Another source of selection bias is the low recruitment rate: In Kao et al., nearly 46% of patients were not eligible, because they were either too ill or likely to die within 24 hours; of the remaining 54% who were eligible, only 55% consented to participate.³⁵

Third, greater improvement seen in the intervention group compared with the control group could be due to regression to the mean rather than the effect of the palliative care team. For example, in the study by Kao et al., severity scores for 15 physical symptoms were evaluated, and there was greater improvement in 4 of the symptoms—edema, fatigue, dry mouth, and abdominal distension. However, the baseline severity scores for all four symptoms were worse in the intervention group. For instance, abdominal distension improved from 2.22 to 1.78 in the control group, and from 3.24 to 1.81 in the intervention group.³⁵

The fourth limitation is that in the studies without a control group, the benefits seen may have come about due to factors other than the palliative care team. We can infer this from the two studies that had a control group: Both reported significant improvements for patients in the control arm who received “usual care” and no involvement from the palliative care team.^17,35 Therefore, in pre- and post-studies without a control group, these improvements may be the result of effective treatment of the symptoms by hospital teams providing standard care. Hence, the benefits seen in studies that do not have a control group may have come about due to factors other than palliative care interventions.

Furthermore, the included studies spanned a long time frame of 10 years. Professional policies and guidelines published during this time may have had a confounding effect.^3,47–49 For example, the European Association for Palliative Care guidelines on the use of opioids for the treatment of cancer pain may have resulted in improved pain control, independent of any hospital palliative care team involvement.⁴⁸

Notwithstanding the limitations of quantitative studies, the process evaluations and qualitative studies provide evidence of the benefits of palliative care teams in the acute hospital setting.⁵⁰ Qualitative studies that report the views of referring teams or patients and their caregivers show that palliative care teams do seem to improve patient care.^43,46,51 Further studies using qualitative methodology may be valuable in elucidating the mechanisms through which palliative care teams bring about benefit.

Choice of outcomes to evaluate a palliative care intervention

Evaluation of the effectiveness of any intervention requires the measurement of appropriate outcomes. A range of outcome measures is available for palliative care research, and several tools may need to be employed so that the domains of interest will be included.^52,53 Most of these tools are patient-reported outcome measures, and concerns about patient burden and fatigue in completing these instruments limit their use. Consequently, the choice of primary outcome may not reflect the domain mostly likely to be affected by the intervention. For example, in one of the reviewed studies, disease awareness was the primary outcome, even though the palliative care team was more widely involved in symptom control.³⁹

The most representative outcome may be difficult to measure. According to the World Health Organization's definition, palliative care seeks to improve the quality of life and patients and their families.⁵⁴ By that account, the most appropriate outcome for a palliative care intervention may be patient-reported quality of life. Many instruments that can measure quality of life in palliative care exist; however, most have not yet been adequately validated.⁵⁵ The short follow-up time in the inpatient acute hospital setting may also limit the studied outcomes to short-term issues such as physical symptom burden. Interventions to address psychosocial and emotional needs may require more time to take effect, and this may not be captured in an actual change in quality of life measures within a short time interval.⁵⁶ Response shift, where health changes lead to shifts in internal standards, values, and conceptualization of quality of life, may also distort the interpretation of change in quality of life measures over time.⁵⁷

Innovative approaches are needed to circumvent the challenges of conducting palliative care research in the inpatient acute hospital setting. Novel patient-reported outcome measures specifically for use in the palliative care setting are being developed. For example, the St Christopher's Index of Patient Priorities (SKIPP) includes both current and retrospective components, thus addressing response shift and allowing for detection of change with only one administration.⁵⁸ Also, the Palliative Care Outcome Scale (POS) is a brief instrument that measures aspects of physical, psychological, and spiritual domains, which are pertinent to palliative care.⁵⁹ Proxy reporting by staff is also being explored in the POS group of questionnaires. If found to be reliable, this would be one way of addressing the problem of low data completion rate due to patient severity of illness or fatigue. The streamlining and standardization of patient-reported outcome measures used in future studies would facilitate comparison across studies.

Future research

Palliative care is a complex intervention, making it a challenge to evaluate. The patient-centered ethos also means that the domain of benefit depends on the domain of problems, which differs from one individual to another. This makes quantitative evaluations of palliative care challenging. This review has highlighted the dearth of high-quality studies that evaluate the effectiveness of palliative care teams in the acute inpatient hospital setting. Future research on the development and evaluation of palliative care interventions should endeavor to have higher methodological rigor so as to build better evidence for the effectiveness and cost effectiveness of palliative care teams in patient care.^26,50,60 This may be achieved with better research training in the field of palliative medicine or collaborations with scientists in the field of health services research.

Conclusion

Heterogeneity of interventions, patient populations, and outcome measures in the reviewed studies make comparison across studies impossible. Despite methodological limitations, published studies provide a valuable glimpse into the wider benefits of palliative care interventions. Future studies should employ innovative strategies to further this field of research.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Cancer care during the last phase of life. J Clin Oncol, 1998; 16:1986–1996.

Cherny

, Catane

, Kosmidis

: ESMO takes a stand on supportive and palliative care. Ann Oncol, 2003; 14:1335–1337.

Ferris

, Bruera

, Cherny

, et al.: Palliative cancer care a decade later: Accomplishments, the need, next steps—from the American Society of Clinical Oncology. J Clin Oncol, 2009; 27:3052–3058.

Hui

, Bruera

: Models of integration of oncology and palliative care. Ann Palliat Med, 2015; 4:89–98.

Hui

, Bruera

: Integrating palliative care into the trajectory of cancer care. Nat Rev Clin Oncol, 2015; 13:159–171.

Davis

, Strasser

, Cherny

: How well is palliative care integrated into cancer care? A MASCC, ESMO, and EAPC Project. Support Care Cancer, 2015; 23:2677–2685.

Bakitas

, Tosteson

, Li

, et al.: Early versus delayed initiation of concurrent palliative oncology care: Patient outcomes in the ENABLE III randomized controlled trial. J Clin Oncol, 2015; 33:1438–1445.

Dionne-Odom

, Azuero

, Lyons

, et al.: Benefits of early versus delayed palliative care to informal family caregivers of patients with advanced cancer: Outcomes from the ENABLE III randomized controlled trial. J Clin Oncol, 2015; 33:1446–1452.

Temel

, Greer

, Muzikansky

, et al.: Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med, 2010; 363:733–742.

10.

Zimmermann

, Swami

, Krzyzanowska

, et al.: Early palliative care for patients with advanced cancer: A cluster-randomised controlled trial. Lancet, 2014; 383:1721–1730.

11.

Smith

, Temin

, Alesi

, et al.: American Society of Clinical Oncology provisional clinical opinion: The integration of palliative care into standard oncology care. J Clin Oncol, 2012; 30:880–887.

12.

Higginson

, Finlay

, Goodwin

, et al.: Do hospital-based palliative teams improve care for patients or families at the end of life?. J Pain Symptom Manage, 2002; 23:96–106.

13.

Higginson

, Evans

: What is the evidence that palliative care teams improve outcomes for cancer patients and their families?. Cancer J, 2010; 16:423–435.

14.

Morita

, Fujimoto

, Tei

: Palliative care team: The first year audit in Japan. J Pain Symptom Manage, 2005; 29:458–465.

15.

O'Mahony

, Blank

, Zallman

, Selwyn

: The benefits of a hospital-based inpatient palliative care consultation service: Preliminary outcome data. J Palliat Med, 2005; 8:1033–1039.

16.

Penrod

, Deb

, Luhrs

, et al.: Cost and utilization outcomes of patients receiving hospital-based palliative care consultation. J Palliat Med, 2006; 9:855–860.

17.

Jack

, Hillier

, Williams

, Oldham

: Improving cancer patients' pain: The impact of the hospital specialist palliative care team. Eur J Cancer Care (Engl), 2006; 15:476–480.

18.

Braiteh

, El Osta

, Palmer

, et al.: Characteristics, findings, and outcomes of palliative care inpatient consultations at a comprehensive cancer center. J Palliat Med, 2007; 10:948–955.

19.

Ciemins

, Blum

, Nunley

, et al.: The economic and clinical impact of an inpatient palliative care consultation service: A multifaceted approach. J Palliat Med, 2007; 10:1347–1355.

20.

Yennurajalingam

, Zhang

, Bruera

: The impact of the palliative care mobile team on symptom assessment and medication profiles in patients admitted to a comprehensive cancer center. Support Care Cancer, 2007; 15:471–475.

21.

Hanson

, Usher

, Spragens

, Bernard

: Clinical and economic impact of palliative care consultation. J Pain Symptom Manage, 2008; 35:340–346.

22.

Gade

, Venohr

, Conner

, et al.: Impact of an inpatient palliative care team: A randomized control trial. J Palliat Med, 2008; 11:180–190.

23.

Smith

, Brick

, O'Hara

, Normand

: Evidence on the cost and cost-effectiveness of palliative care: A literature review. Palliat Med, 2014; 28:130–150.

24.

Higgins

JPT

, Green

(eds.): The Cochrane collaboration. 2011. In: Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. www.cochrane-handbook.org.

25.

Stewart

, Clarke

, Rovers

, et al.: Preferred Reporting Items for Systematic Review and Meta-Analyses of individual participant data: The PRISMA-IPD Statement. JAMA, 2015; 313:1657–1665.

26.

Craig

, Dieppe

, Macintyre

, et al.: Developing and evaluating complex interventions: The new Medical Research Council guidance. BMJ, 2008; 337:a1655.

27.

Downs

, Black

: The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health, 1998; 52:377–384.

28.

Drummond

, Sculpher

, Torrance

, et al.: Methods for Economic Evaluation of Health Care Programmes. 3rd edition. Oxford University Press. 2005, p. 396.

29.

Tong

, Sainsbury

, Craig

: Consolidated criteria for reporting qualitative research (COREQ): A 32-item checklist for interviews and focus groups. Int J Qual Health Care, 2007; 19:349–357.

30.

Portela

, Pronovost

, Woodcock

, et al.: How to study improvement interventions: A brief overview of possible study types. BMJ Qual Saf, 2015; 24:325–336.

31.

Lefkowits

, Teuteberg

, Courtney-Brooks

, et al.: Improvement in symptom burden within one day after palliative care consultation in a cohort of gynecologic oncology inpatients. Gynecol Oncol, 2015; 136:424–428.

32.

Corbett

, Johnstone

, Trauer

, Spruyt

: Palliative care and hematological malignancies: Increased referrals at a comprehensive cancer centre. J Palliat Med, 2013; 16:537–541.

33.

Birks

, Krikos

, McGowan

, Stone

: Is there a need for weekend face-to-face inpatient assessments by hospital specialist palliative care services? Evaluation of an out-of-hours service. Palliat Med, 2011; 25:278–283.

34.

Hall

, Thompson

, Davies

: An evaluation of the activity of a 7-day, nurse-led specialist palliative care service in an acute district general hospital. Int J Palliat Nurs, 2013; 19:148–151.

35.

Kao

, Hu

, Chiu

, Chen

: Effects of the hospital-based palliative care team on the care for cancer patients: An evaluation study. Int J Nurs Stud, 2014; 51:226–235.

36.

Bell

, Kuriya

, Fischberg

: Hospice referrals and code status: Outcomes of inpatient palliative care consultations among Asian Americans and Pacific Islanders with cancer. J Pain Symptom Manage, 2011; 42:557–564.

37.

Centeno

, Portela

, Noguera

, et al.: Impact of a new palliative care consultation team on opioid prescription in a University Hospital. BMC Palliative Care, 2009; 8:7.

38.

Chang

, Liu

, Chao

, et al.: A new hospice consulting system for terminal cancer patients in transferring to post-acute care options in Taiwan. Eur J Cancer Care (Engl), 2010; 19:267–272.

39.

Chou

, Hung

, Kao

, et al.: Impact of palliative care consultative service on disease awareness for patients with terminal cancer. Support Care Cancer, 2013; 21:1973–1981.

40.

Delgado-Guay

, Parsons

, Li

, et al.: Symptom distress, interventions, and outcomes of intensive care unit cancer patients referred to a palliative care consult team. Cancer, 2009; 115:437–445.

41.

Iwase

, Murakami

, Saito

, Nakagawa

: Preliminary statistical assessment of intervention by a palliative care team working in a Japanese general inpatient unit. Am J Hosp Palliat Care, 2007; 24:29–35.

42.

Loke

, Rau

, Huang

: Impact of combined hospice care on terminal cancer patients. J Palliat Med, 2011; 14:683–687.

43.

Sasahara

, Miyashita

, Umeda

, et al.: Multiple evaluation of a hospital-based palliative care consultation team in a university hospital: Activities, patient outcome, and referring staff's view. Palliat Support Care, 2010; 8:49–57.

44.

Morrison

, Dietrich

, Ladwig

, et al.: Palliative care consultation teams cut hospital costs for Medicaid beneficiaries. Health Aff (Millwood), 2011; 30:454–463.

45.

Kao

, Wang

, Tang

, et al.: Predictive factors for do-not-resuscitate designation among terminally ill cancer patients receiving care from a palliative care consultation service. J Pain Symptom Manage, 2014; 47:271–282.

46.

Yang

, Ewing

, Booth

: What is the role of specialist palliative care in an acute hospital setting? A qualitative study exploring views of patients and carers. Palliat Med, 2012; 26:1011–1017.

47.

Borras

, Albreht

, Audisio

, et al.: Policy statement on multidisciplinary cancer care. Eur J Cancer, 2014; 50:475–480.

48.

Caraceni

, Hanks

, Kaasa

, et al.: Use of opioid analgesics in the treatment of cancer pain: Evidence-based recommendations from the EAPC. Lancet Oncol, 2012; 13:e58–e68.

49.

Kuehn

: ASCO: Patients with advanced cancer need advice on end-of-life care options. JAMA, 2011; 305:1185–1186.

50.

Evans

, Harding

, Higginson

: ‘Best practice’ in developing and evaluating palliative and end-of-life care services: A meta-synthesis of research methods for the MORECare project. Palliat Med, 2013; 27:885–898.

51.

Ewing

, Farquhar

, Booth

: Delivering palliative care in an acute hospital setting: Views of referrers and specialist providers. J Pain Symptom Manage, 2009; 38:327–340.

52.

Agar

, Luckett

: Outcome measures for palliative care research. Curr Opin Support Palliat Care, 2012; 6:500–507.

53.

Bausewein

, Daveson

, Currow

, et al.: EAPC White Paper on outcome measurement in palliative care: Improving practice, attaining outcomes and delivering quality services—Recommendations from the European Association for Palliative Care (EAPC) Task Force on Outcome Measurement. Palliat Med, 2016; 30:6–22.

54.

WHO: WHO definition of Palliative Care. www.who.int/cancer/palliative/definition/en (Last accessed June 25, 2016.)

55.

Albers

, Echteld

, de Vet

, et al.: Evaluation of quality-of-life measures for use in palliative care: A systematic review. Palliat Med, 2010; 24:17–37.

56.

Sprangers

, Moinpour

, Moynihan

, et al.: Assessing meaningful change in quality of life over time: A users' guide for clinicians. Mayo Clin Proc, 2002; 77:561–571.

57.

Hamidou

, Dabakuyo

, Bonnetain

: Impact of response shift on longitudinal quality-of-life assessment in cancer clinical trials. Expert Rev Pharmacoecon Outcomes Res, 2011; 11:549–559.

58.

Addington-Hall

, Hunt

, Rowsell

, et al.: Development and initial validation of a new outcome measure for hospice and palliative care: The St Christopher's Index of Patient Priorities (SKIPP). BMJ Support Palliat Care September 16, 2013.

59.

Hearn

, Higginson

: Development and validation of a core outcome measure for palliative care: The palliative care outcome scale. Palliative Care Core Audit Project Advisory Group. Qual Health Care, 1999; 8:219–227.

60.

Simon

, Higginson

: Evaluation of hospital palliative care teams: Strengths and weaknesses of the before-after study design and strategies to improve it. Palliat Med, 2009; 23:23–28.