Patient and carer preference for, and satisfaction with ... - SAGE Journals

03pm235.qxd 18/5/99 4:19 pm Page 197

Palliative Medicine 1999; 13: 197–216

Patient and carer preference for, and satisfaction with, specialist models of palliative care: a systematic literature review EK Wilkinson Project Researcher, Wessex Institute for Health Research and Development, University of Southampton, C Salisbury Consultant Senior Lecturer, Division of Primary Care, University of Bristol, N Bosanquet Professor of Health Policy, Department of Primary Health Care, Imperial College School of Medicine at St Mary’s, PJ Franks Co-director, The Centre for Research and Implementation of Clinical Practice, Thames Valley University, S Kite Senior Registrar in Palliative Medicine, The Margaret Centre, Whipps Cross Hospital, London, M Lorentzon Honorary Research Fellow, Department of Primary Health Care, Imperial College School of Medicine at St Mary’s and A Naysmith Consultant in Palliative Medicine, Pembridge Palliative Care Centre, St Charles’ Hospital, London Abstract: This systematic literature review examined the impact of specialist models of palliative care on consumer satisfaction, opinion and preference over the past 20 years. From the literature search, 83 relevant papers were identified. Research findings from North America did not reveal any reliable or consistent trends, and this was due primarily to methodological flaws in the research. In the UK, consumers are more satisfied with all types of palliative care, whether provided by inpatient units or in the community, than with palliative care provided by general hospitals. Even though research findings consistently indicate that consumers appreciate the psychosocial climate in hospices, this research was based on small-scale local studies which were mainly focused on a single hospice. The dearth of high quality, comprehensive research was notable on the impact on consumer preference, opinion or satisfaction of hospice home care services, and other forms of palliative care in the community. It was concluded that further research is needed into consumer priorities for patients dying of noncancer diseases, and into the possible impact of patient characteristics on consumer satisfaction, opinion and preference for specialist models of palliative care. Key words: palliative care; terminal care; consumer satisfaction; caregivers; consumer participation; systematic literature review Resumé: Cette revue systématique de la littérature étudie l’impact des modèles de soins palliatifs sur la satisfaction, l’opinion et les préférences des usagers durant ces vingt dernières années. 83 publications pertinentes ont été identifiées. Pour l’Amérique du Nord, on n’a pas mis en évidence de tendance fiable ou concordante, ce qui était avant tout dû à des défauts méthodologiques. Au Royaume-Uni, les utilisateurs ont été davantage satisfaits par tous les types de soins palliatifs pourvu qu’ils soient dispensés dans des unités fixes ou mobiles plutôt que dans des hôpitaux généraux. Même si les résultats de l’étude indiquent de façon concordante que les utilisateurs apprécient la

Address for correspondence: Ms EK Wilkinson, Wessex Institute for Health Research and Development, Health Care Research Unit, Level B, Southampton University, Southampton General Hospital, Southampton SO16 6YD, UK. E-mail: [email protected] © Arnold 1999

0267–6591(99)PM235RA

03pm235.qxd 18/5/99 4:19 pm Page 198

198

EK Wilkinson et al. prise en charge psycho-sociale des centres de soins palliatifs, ils étaient issus d’études locales à petite échelle, généralement centrées sur un établissement de soins palliatifs. La pénurie d’études de qualité bien conduites influence les résultats de l’appréciation des préférences, opinions et satisfactions des usagers concernant les services d’hospitalisation à domicile et les autres formes de soins palliatifs extra-hospitalières. Nous en avons conclu que d’autres études sont nécessaires pour définir les priorités des usagers en terme de soins aux patients non cancéreux en fin de vie et pour connaître l’influence exercée par les caractéristiques du patient sur l’appréciation par les usagers des modèles en soins palliatifs. Mots clés: soins palliatifs; soins terminaux; satisfaction des usagers; soignants; participation des utilisateurs; revue systématique de la littérature.

Introduction Over the past 20 years there has been considerable growth in the variety and complexity of specialist palliative care services provided in the UK. During this time, the relative importance of National Health Service (NHS) funding for these services has increased. Assessment of the effectiveness and value for money offered by palliative care services is now of prime concern to ensure that the best and most efficient ways of caring for patients are developed into the new millennium. This study was carried out within a wide-ranging literature review which aimed to identify the most appropriate and most cost-effective models of service delivery and level of provision of palliative care services. To achieve this aim, the impact of the various models of palliative care was assessed on a number of important outcomes, including quality of care, cost, consumer satisfaction and opinion. This paper describes the research evidence about the impact of different types of palliative care on informal carer and patient satisfaction and opinion. The assessment of these outcomes was considered important for several reasons. Given health care services exist for the benefit of patients and are paid for by them, any meaningful assessment of the impact of these services should include their views. Furthermore, some aspects of care can only be assessed from the unique perspective of the consumer, including, for example, access to care and the quality of doctors’ explanations. Moreover, previous research has shown satisfaction levels to

correlate with a number of other important outcome measures including: compliance with treatment,1 attendance at follow-up,2 improvements in health status,3,4 quality of consultation and effectiveness of patient–doctor communication.5 Clearly, the assessment of consumer satisfaction can help to predict consumer behaviour, and also to assess communication needs. Finally, knowing consumer views on, and satisfaction with, different models of palliative care can help towards the development of new models of care.6

Method Overview The aim of the systematic review was to identify and assess evidence from studies which evaluated patient and informal carer satisfaction with palliative care. The systematic review was carried out with adherence to a strictly predefined protocol to minimize potential bias and to maximize reliability and comprehensive findings. This process was overseen by an expert advisory group which included representatives from palliative medicine, health economics, health policy, primary care and nursing. Scope of the review The scope of the review was defined by a number of predetermined inclusion and exclusion criteria. The literature review, as a whole, included papers which focused on palliative care, regardless of disease type. Further broad inclusion criteria included

03pm235.qxd 18/5/99 4:19 pm Page 199

Patient and carer preference for, and satisfaction with, specialist models of palliative care 199 studies conducted in Europe, North America, Australasia, or Israel and published between 1978 and June 1998 in English, German, French, Italian or Swedish. For this particular study, the focus was on papers which met the above inclusion criteria and which also had patient or carer satisfaction, preference or opinion as outcome measures. The term ‘carer’ was defined as including all informal caregivers such as friends, relatives, spouses, family members or neighbours who cared for patients. Papers focusing on the following issues were excluded from the review: personal commentaries; general or public opinion; ethical, legal or religious issues; education; history of palliative care provision; individual case histories; the development of research instruments or general cancer care services. Data sources and search strategies A variety of data sources and techniques were used to identify documents (see Table 1 below). Data sources included: electronic databases, paper-based databases, and consultation exercises with relevant individuals and organizations. The initial search was conducted on Medline using the search terms ‘palliative care’, ‘terminal care’, ‘hospices’, ‘hospice care’, ‘oncologic nursing’, or ‘cancer care facilities’ combined with terms to include research articles: ‘clinical trials’, ‘program evaluation’, ‘health services research’, ‘outcome assessment health care’ and ‘review literature’. Articles on drug evaluation, radiotherapy, chemotherapy or surgery, and articles falling outside the predefined language and date Table 1

parameters were excluded. To identify papers relevant to patient and carer preference for different types of palliative care a further search was carried out by combining the Index medicus medical subject headings: palliative/hospice/terminal care with other terms such as ‘consumer satisfaction’, ‘consumer participation’, ‘patient acceptance of health care’, ‘caregivers’, ‘knowledge, attitudes, practice’ and ‘patient-centred care’. This strategy formed the prototype search strategy which was used as a basis for further searches on other electronic databases. The paper-based European directory of palliative care research, and the manual index of Palliative Medicine (which has only been indexed in Medline since 1993) were scrutinized. Hand-searches were carried out of recent issues of the five journals which contained the highest number of relevant papers in our initial electronic literature searches, as shown in Table 1. The reference lists of papers were examined as they were retrieved, and further articles were identified in this way. Major funding organizations were also asked to provide details of relevant research which they had recently or were currently funding. After these literature searches, 25 organizations and 30 individual researchers were sent the preliminary bibliography. They were asked about any other research they were aware of, particularly unpublished, unreferenced or current work. Several individuals provided us with personal bibliographies and copies of unpublished work.

Sources of data

Electronic databases

Journals hand-searched

(Searches were conducted in 1997, for articles published since 1978. Searches in Medline, Embase and the Cochrane library were updated to June 1998)

(January 1992–June 1998)

• Medline • Embase • Index of scientific and technical proceedings • Cochrane library • SIGLE (Index of grey literature) • NHS Project Research System

• • • • •

• Health planning and administration • CancerLit • DHSS data

The Hospice Journal Palliative Medicine Journal of Palliative Care Cancer Nursing Oncology Nursing Forum

Paper-based sources • Index of palliative medicine 1987–1993 • European directory of palliative care research

Funding bodies contacted

• NHS Project Register • Cancer Research Council • Imperial Cancer Research Fund • Medical Research Council • Wellcome Trust

03pm235.qxd 18/5/99 4:19 pm Page 200

200

EK Wilkinson et al.

Data management and initial assessment Details of identified articles were entered into a Reference Manager database. The titles and abstracts of the articles were examined by at least two researchers to categorize each paper to the particular area of interest. Copies of these papers were retrieved and photocopied in full wherever possible. Further assessment, data extraction and synthesis The validity of the articles was assessed using separate checklists for studies of different designs. The checklists ensured consideration of the study design, potential biases, the extent to which results are attributable to the intervention, and reliability of the results. Conclusions were based on consideration of the methodological rigour of papers, with more weight given to larger, better quality studies. Reviewers used data extraction forms to record this information and to ensure standard information extraction. These forms were designed to extract the following information: bibliographic details, country of publication, study population, study setting, disease type, study design, intervention, outcome measures, number of subjects, number of conditions, duration of intervention, measurement tools, the results, conclusions and comments on strengths and weaknesses. The data from the extraction sheets was collated in a specially designed Microsoft Access database.

Results We identified 831 documents which met the exclusion and inclusion criteria for the whole review, of which 688 were retrieved and analysed. Papers which were not retrieved were mainly unpublished or from overseas journals not available on interlibrary loan. Eighty-three papers were relevant to the issue of patient and informal carer preference for, or satisfaction with, different types of palliative care. Of these papers, 79 were retrieved and analysed. Of the remaining articles which were not retrieved, abstracts were examined.7–10 Studies from the European mainland were not included in the review as there were too few in number.11–13 No relevant studies were identified from Australasia or Israel.

Direct comparisons were not made between research findings derived from palliative care settings in the UK and those derived from North America due to fundamental differences in the structure of care in these two countries. Differences in the financial arrangements in funding palliative care services has led to the development of unique systems of care in these two countries. Finding evidence that a system of care works well in one country may not indicate that it can work well in another. Before examining the main findings, consideration is given to conceptual issues relating to the term ‘ satisfaction’, the impact of the design of consumer surveys on the data obtained, and the extent to which consumer responses can be considered valid and reliable. Conceptualization of the term ‘satisfaction’ and related assessment issues The term ‘satisfaction’ was rarely defined a priori, with most studies assuming that its meaning was already known. The definition of satisfaction and how it can be effectively measured as an ‘outcome’ has not been supported by a clear consensus either within or beyond palliative care. Furthermore, no well-defined theory of satisfaction exists.14,15 As such, the concept of satisfaction and constructs used to measure it may differ from study to study. Previous research has shown that findings based on questionnaire surveys of consumer satisfaction often lack validity. For example, a large hospital survey found that satisfaction levels for hospital services were very high for most services (over 89%),16 however, after more detailed questioning a large number of specific problems were revealed with regard to these services. The ‘ceiling effect’ of satisfaction surveys, whereby satisfaction levels are high on any given issue and thereby lack variability, is well documented.16–19 It has been estimated that at least 80% of respondents will express satisfaction for any given issue.20 The design and impact of tools used to measure satisfaction/opinions Outcome measures. Outcome measures developed or validated outside the palliative care environment may not be sensitive to the changes that occur

03pm235.qxd 18/5/99 4:19 pm Page 201

Patient and carer preference for, and satisfaction with, specialist models of palliative care 201 within such a setting. Such tools may also lack appropriateness. Instruments that have not been validated or are of unknown reliability may also be poor indicators of effect. Choice of response option. The choice of ‘response option’ used in consumer surveys may impact on the validity of data obtained. Global measures of satisfaction may provide crude, unreliable data that mask preferential differences.17 Specific combination measures are more likely to reveal such differential views or preferences, although composite measures have been criticized for assuming that all items are of equal weighting.21 Impact of wording. There is some evidence that the framing and timing of questions can impact on the quality of data obtained from satisfaction surveys.15,22,23 For example, the same question framed in a negative or a positive way, or posed directly or indirectly can elicit quite different responses.23 Focus. There are many aspects of care which may be examined in consumer surveys, such as overall quality, competence, information, bureaucratic arrangements, physical facilities, provider’s attention to patients’ psychosocial problems, etc. There is some concern that consumer surveys do not reflect the range of issues of importance to the consumer and may in fact reflect the bias of the researcher.18 Other research has indicated that consumer surveys measure aspects of care unevenly.19,24 A meta-analysis of 221 studies24 indicated that cost issues, access to care and bureaucracy issues were least frequently measured. Aspects of care most frequently examined included ‘hotel aspects’ such as catering or environment.19 Validity and reliability of consumer responses Patient responses. Complete and reliable completion of surveys (or other research instruments) may prove problematic due to patients’ lapses of energy or concentration, or due to fluctuations of pain, mood and psychological state.25 The timing of the data collected is therefore crucial. It is important to note such changes and when they occur in relation to any intervention.26 Proxy accounts. Although proxies may be more willing participants and easier to interview, there

is no firm evidence that their views accurately reflect the patients’ perspective. Some research has assumed that proxy accounts can be used as reliable measures.27–30 However, there is some evidence that proxy accounts do not provide reliable accounts of some aspects of a patient’s experience, in particular, of pain and emotional state.31–34 The extent to which proxy views diverge from those of patients is still not fully understood.33 Retrospective accounts. Retrospective assessments made by relatives or friends may be affected by the process of grief. Higginson et al.35 found that in comparison with prospective assessments, bereaved family members’ memories of symptoms and pain were polarized after the death of patients. Socially desirable responses. Both patients and informal carers may feel under great pressure to provide socially desirable answers to sensitive questions33 and the reluctance of recipients of care to criticize providers of care is well documented.36 Methodological issues Attribution of change and patient characteristics. The great challenge of attributing change in outcome measures to the quality of care within a palliative care environment has been well documented.37–39 Indeed, it is difficult to rule out the possibility that other factors such as personal characteristics of a patient or the course of a disease have had an effect on the outcome. For example, The Regional Study for the Care of the Dying studied a random sample of patients across 20 health districts, and found that carers of patients with lymphatic cancer were more satisfied with care than carers of patients with genito-urinary cancer. Although the underlying rationale for this finding was not fully understood, it highlights the need to consider patient characteristics when evaluating models of palliative care. Comparing research studies The above section outlines some of the difficulties encountered in conceptualizing ‘consumer satisfaction’ and assessing consumer opinions in general. Differences in instrumentation design, general focus, and patient characteristics can make meaningful comparisons between studies difficult.

03pm235.qxd 18/5/99 4:19 pm Page 202

202

EK Wilkinson et al.

The impact of specialist models of palliative care on carer/patient opinion, preference and satisfaction The following assessment of the impact of different models of palliative care on consumer opinion and satisfaction was based, for the most part, on a review of comparative research studies. Details of the key papers that were reviewed are summarized in Tables 2 and 3. Satisfaction with and opinions on inpatient hospice care in the UK Comparative studies on opinions of inpatient hospice versus inpatient hospital care. In the late 1970s and early 1980s a small number of studies examined consumer satisfaction with inpatient hospice care compared to inpatient hospital care. Hinton40 examined differences in consumer opinion in four different care environments including: St Christopher’s inpatient and outpatient hospice care, general hospital care and a foundation home. Hospice care was viewed more favourably than other forms of care in terms of ameliorating psychological problems. Opinions on physical care, communication and place of care were generally positive across all environments. Similar findings were found by Parkes.41 In this study, spouses of St Christopher’s Hospice patients reported lower levels of anxiety than their hospital counterparts. Hospice inpatients also reported lower levels of pain compared to their hospital counterparts, although this particular finding has not been replicated in follow-up work or any other study. The follow-up study29 did, however, find that anxiety levels were significantly lower among surviving hospice spouses in comparison with their hospital counterparts. The trend of consumer preference for the psychosocial climate of hospices in comparison to hospitals is prevalent in research findings in the 1990s. A retrospective study by Seamark42 examined lay carers’ views of quality of care in a hospice environment compared to their lay carers’ views on community hospital care. Carers were more critical of the hospital environment, and the quality of communication, lack of nursing staff and bereavement support compared to these aspects of hospice care. Similarly, a study by Seale and Kelly43,44 involving matched pairs of carers whose patients

had either received care from St Christopher’s Hospice or from a nearby hospital, found that carers were more critical of the busy atmosphere and the quality of communication in hospitals. No differences were found in terms of physical care or symptom control. Overall, comparative consumer research on inpatient hospice care and inpatient hospital care from the 1970s to the present day has indicated that consumers consistently favour the psychosocial aspects of inpatient hospice care. However, these studies have mainly been relatively small scale and have involved comparisons of care given by St Christopher’s Hospice with care provided by other nonspecialized care institutions. The extent to which this centre of excellence may have influenced the care provided in other local care institutions is difficult to determine, as is the extent to which findings from this centre can be generalized to other hospices. Comparative studies on satisfaction with inpatient hospice care versus conventional care. Two recent satisfaction surveys have found comparatively higher levels of satisfaction with inpatient hospice care, compared to more conventional, nonspecialized forms of care. However, methodological flaws have undermined this assertion. The existence of medical differences between the two comparative groups in Seale’s45 satisfaction survey somewhat undermines the finding that the quality of hospice care was viewed more favourably than conventional care. Field et al.’s work46 on the satisfaction of 59 lay carers of patients who had experienced both hospice care and conventional care was relatively small scale and may lack external validity. However, such studies have shown that consumers tend to appreciate the nonclinical aspects of inpatient hospice care most, including: humaneness,45 an ability of hospice staff to reduce anxiety,40,41 honest communication41 and supportive nursing care. Satisfaction with and opinions on specialist community services Consumer opinion of hospice home care services. Only one comparative study carried out almost two decades ago examined consumer opinions of hospice home care services compared to other nonspecialized services. This study by Parkes28 based on retrospective relatives’ accounts, indicated

03pm235.qxd 18/5/99 4:19 pm Page 203

Patient and carer preference for, and satisfaction with, specialist models of palliative care 203 that cancer patients who had received home care provided by St Christopher’s Hospice were able to stay at home until a later stage in their illness and spent less time in hospital compared to the control group. However, the hospice group patients also felt more of a burden and showed greater feelings of tension. Given that patients in the St Christopher’s group had greater nursing needs than those in the control group, no firm conclusions can be drawn. A small number of recent, small-scale studies have examined consumer views on hospice home care services. These studies have shown that perceived benefits of home care have included: support and access to 24-h access to professional care and attention to symptom control. Satisfaction with and opinions on community specialist (nonhospice) palliative care outreach teams in the community. A small number of recent, comparative studies on satisfaction with specialist home care services have shown that specialist outreach care compared favourably with other community care services49 and with primary and secondary services.31,50 Indeed, McCann49 found that people with HIV/AIDS rated the community team as significantly more kind and understanding than general practitioners (GPs), and the majority believed that the main role of the community team was to provide advice, which was found to be useful for the most part. A smaller scale study by Butters et al.50 found that although people with HIV/AIDS and their lay carers were more satisfied with the community care team’s services than those provided by primary or secondary care, they were unclear of the role of the community care team. These positive results may not be representative of all the patients’ experiences, as the patients who were interviewed had lower levels of pain, less anxiety and better insight into their condition than those who were not interviewed. Consumer opinions on ‘home’ as a site of death There is a growing body of evidence that ‘home’ as the site of death is generally preferred,48,51–53 although this preference may fluctuate according to the stage of terminal illness.48 It has been suggested that dying at home is preferable as families and their dying relatives can continue with established patterns of interaction, maintain some control over

daily routines and a level of normality in their lives. The nature of some hospice home care services may also contribute to this preference, as patients can have 24-h access to specialized support and care.47 The increasing availability of specialized care at home may have contributed to consumer preference of dying at home. Despite consumer preference for dying at home, some research has shown that consumer preference is not always realized in practice. Indeed, an interview survey of 84 cancer patients found that 58% wished to die at home, but 29% did so, only 20% wished to die in hospital and 54% did so.54 Satisfaction with and opinions on inpatient hospital care and nonspecialist care in the community Consumer views on inpatient hospital versus general community care. The earliest comparative study of hospital care with ordinary home care in the late 1970s, found that although hospital patients dying from cancer experienced less pain than their home care counterparts, they were also less active and conscious than their home care counterparts.55 Although these findings have not been replicated since, studies in the 1980s and 1990s have indicated that consumers have reported a wide range of problems with inpatient hospital care. Wilkes26 examined 262 carers’ retrospective views on hospital and home care received by their relatives. Although care received by almost a third of patients who had received hospital care was praised, almost a quarter of relatives criticized the uncaring attitude of hospital care, poor symptom control and difficulties in extracting information from doctors. This latter complaint has been reported in a number of other studies. Poor communication was reported as the most prominent criticism by cancer patients in a comparative study by Higginson et al.31 on consumer views of community-based support teams, general community and hospital care. Indeed, similar findings have been reported in other studies with staff being perceived as too busy to care and lacking time to discuss medical issues. Another complaint reported in Higginson et al.’s study31 was the lack of adequate co-ordination between primary and secondary services. However, it is notable that a relatively recent randomized controlled trial has found no effect of co-ordinating care for terminally ill cancer

03pm235.qxd 18/5/99 4:19 pm Page 204

204

EK Wilkinson et al.

patients on satisfaction levels.60,61 Overall, the research suggests that consumer criticisms of hospital-based palliative care are not only numerous, but are also wide-ranging. Recent research has also highlighted a range of consumer criticisms regarding nonspecialized palliative care in the community, although they have not been as wide-ranging as criticisms of palliative care in hospitals. These difficulties have centred on difficulties in access to specialist services or equipment,26,31,62 lack of services and out of hours care.51 Consumer concerns reported in small scale, observational studies have included: difficulties with physical care52,62 and a lack of awareness of the full range of options of care.62,63 Satisfaction with inpatient hospital care versus general community care. The variety of case-mix and differences in the interventions of studies which have focused specifically on satisfaction with hospital care compared to other forms of community care make meaningful comparisons difficult to draw across studies. Addington-Hall et al.51 examined the satisfaction of 80 cancer patients and their relatives during the patient’s final week of life with either home care, acute hospital care or inpatient hospice care. They were more dissatisfied with hospital care compared to care in the other settings. Other largescale satisfaction studies have focused mainly on the experiences of terminally ill noncancer patients. A large-scale interview survey of 1683 dementia and cancer lay carers indicated that experiences and satisfaction with hospital and community care varied according to disease type.56 Informal carers’ views of hospital and GP care for patients indicated that dementia patients were slightly more dissatisfied with hospital care and GP care than cancer patients. Their dissatisfaction centred on poor information and the lack of peace and quiet in hospital rooms. Similar complaints were made by cancer care respondents. A secondary analysis of data from the Regional Study of the Care for the Dying examined informal carers’ perspectives on hospital-based palliative care received by 237 stroke patients across 20 health districts, and found that a sizeable minority of them were dissatisfied with their care.59 Their dissatisfaction stemmed from beliefs that the care was too rushed, perceptions of lack of privacy, lack of peace and quiet in the hospital rooms and inadequate

information on the patient’s condition. In a further analysis from the same regional study, focusing on patients dying with heart disease, it was found that almost a third of 600 lay carers of heart disease patients who had received hospital-based palliative care believed that there was inadequate choice regarding the patients’ place of death.64 Despite this criticism and the complaint that patients suffered poor symptom control, over 80% of lay carers rated the care of hospital doctors and nurses as good or excellent. Satisfaction with and opinions on hospice care in North America Satisfaction with inpatient hospice care versus conventional care. The National Hospice Study,66 examined satisfaction of 1754 patients and their lay carers with hospital-based hospice care, hospice home care and conventional care. There were no differences in patient satisfaction across all conditions, although lay carers in the hospice conditions reported greater satisfaction with care before and after death, and with place of death compared to their conventional care counterparts. Concerns regarding the representativeness of sites of care and inconsistencies in the application of patient-selection criteria have undermined the reliability of these results. In contrast to these findings, a randomized controlled trial (RCT)39 found that 137 hospice patients were significantly more satisfied with hospice care compared to 110 conventional care patients. No differences were apparent between the two groups in the alleviation of pain and symptoms or psychological distress. Their families were significantly less anxious compared to their counterparts, although they were not significantly more satisfied with care. However, due to the fact that some hospice patients were cared for in general wards when hospice beds were full, it is hard to determine whether dilution of effect occurred. Overall, comparative examinations of consumer satisfaction with inpatient hospice care with more conventional forms of care have reaped inconsistent results. Drawing any firm conclusions has also proved difficult due to methodological flaws and the variation in focus of these studies. Satisfaction and opinions on different forms of hospice care. The New York Hospice Program focused on retrospective evaluations on the care

Interview survey

64

42

Retrospective interview survey

43

Questionnaire survey

Retrospective interview survey,

Observational study

Interview survey

Study design(s)

• 75% of dementia group and 72% of cancer group thought GP care was excellent or good. • 50% of dementia group compared to 60% for cancer rated the GP as very understanding. • 34% of dementia group described hospital room that the patient had spent most time in as 'not at all peaceful and quiet' compared to 22% of cancer group. • 39% of dementia group were unable to get all information required on patient's condition, compared to 51% of cancer respondents.

Lay carer satisfaction

Random sample of dementia and cancer patients over the age of 65 at death. (Randomly chosen from 20 health districts.) Total: 1683 Two conditions: n = 170 dementia n = 1513 cancer

Cancer patients. Total: 161 lay carers (106 interviewed, 55 given questionnaires) Two conditions: Hospice group: n = 70 Hospitals: n = 91

Investigation of the experience of dying for people with terminal cancer and end-stage dementia during final year of life (from lay carers' perspective)

A comparison of quality of care in community hospitals with care in hospice (from lay carers' perspective, 4–6 weeks after patients' death) Lay carer opinion

Morbidity Provision of care needs

• Lay carers rated communication with hospice professional professionals as significantly 'easier' compared to their community hospital counterparts. • Lay carers in the community hospital group reported more negative comments than those in the hospice group.

• Communication was judged to be better in hospices. • Hospital staff were more likely to be judged as busy. • Hospice respondents were less likely to want improvements and were more likely to judge the institution 'as a family'.

Lay carer opinion

66 patients who died of cancer in 1994. n = 33 hospice n = 33 hospital

Comparison of hospice care versus hospital care

Lay carer opinion

• The best aspects of hospital care were available staff, providing good explanations, spending time with carer/patient. Poor aspects included: understaffing, poor communication, lack of service provision, lack of aftercare, lack of continuity of care. • Giving dietary advice, regular visits, support and easy access to care were seen as the best aspects of community nurses' care.

Lay carer opinion

Results

55 lay carers of cancer patients in Northern Ireland

Outcome measures

Hospital palliative care versus home care during last 6 months of life

Intervention

Subjects: total number, number per condition

Over 90% of lay carers from both conditions assessed terminal care as good or excellent. Hospitals care received more criticisms that included communication difficulties, lack of nursing staff and lack of support in bereavement. Weaknesses: 1) retrospective; 2) limited generalizability – single site [hospice].

Respondents for dementia patients were slightly more dissatisfied with hospital and GP care for their patients, compared to cancer care respondents' views. Dementia patients were less frequently hospitalized, although spent more time in hospital. Lay carers of dementia patients also found caring for patients less rewarding. Weaknesses: 1) carer perspectives not patient; 2) retrospective. Strengths: 1) random sample; 2) relatively large sample.

The psychosocial climate in hospices was rated better than in hospital. The hectic nature of hospitals may not be conducive for effective palliative care. Strengths: matched pairs, random initial recruitment.

Most favourable assessments were made of community nursing. Inpatient hospital care received the most negative comments.

Conclusions and comments

Comparative studies on patient or lay carer opinions or satisfaction with various types of palliative care in the UK

65

Reference

Table 2

03pm235.qxd 18/5/99 4:19 pm Page 205

Patient and carer preference for, and satisfaction with, specialist models of palliative care 205

45

46

50

Reference

Table 2

Random sample of cancer patients age 15 or over.

A comparison of inpatient hospice care/home with conventional care during last year of life. (From the perspective of patients' relatives.)


Hospice group n = 45 Conventional care group n = 126

Total: 171

Hospice patients (details of disease not given) Total: 59 lay carers (interviewed on admission) 37 lay carers (reinterviewed after patient death)

People with AIDS/HIV, majority were homosexual, all had live in carers Of 125 patients, only 19 were interviewed Three groups: 19 lay carers 19 patients 19 ratings by community care team

Comparison of lay carers' views on community care with hospice care (1 week after admission and 3 months after death)

Comparison of views on palliative care given by home care team. (Patients versus lay carers views versus care team views.) After 3–4 weeks of care and later at 4±6 weeks.

Intervention


Interview survey


Interview survey

Study design(s)

Continued

Morbidity

Lay carer opinion

Lay carer opinion

Patient satisfaction

• Hospice nurses were rated as more reassuring and supportive than their conventional care counterparts (92% versus 69% of carers). • Quality of inpatient nursing care was rated more highly by hospice lay carers than their conventional care counterparts (96% versus 57% considered it 'excellent') • 91% of hospice respondents considered doctors' care 'excellent', compared to 40% of conventional care respondents. • No difference between two groups on satisfaction with site of death.

Satisfaction with hospice care was generally higher compared to conventional care. Aspects of hospice care most valued were: supportive nursing care, bereavement support, and attention to symptom control. Weaknesses: 1) significant medical differences between groups: hospice patients had fewer conditions, worse symptoms and were less mobile; 2) proxy views. Strength: 14 hospices involved in study.

Weaknesses: 1) details of disease type not given; 2) small sample.

Lay carers were more satisfied with care from hospice nurses, compared to other professional groups. However, the quality of hospice care was rated as significantly better for patients than lay carers.

• Lay carers were significantly more likely to rank hospice nurses care as 'excellent' than community nurses. • The ratings of the quality of care of GPs and hospice doctors were not significantly different. • Lay carers were significantly less satisfied with their own hospice care compared to the care received by patients. • Lay carers rated 89% of hospice staff as 'willing to give information' although 54'% of carers felt they had to ask for information.



Morbidity

Patients and carers were generally satisfied with Community Care Team and were less content with care of primary and secondary services. Conclusions may be limited as patients with lower levels of pain, less anxiety, better insight and fewer spiritual problems were included in the study. Weaknesses: 1) unable to interview 105/125 patients due to ill health; 2) patients interviewed experienced fewer problems. Strengths: 1) ratings from patients, informal carers and CCT ensured balanced perspective; 2) prospective.


• Most patients (18/19) and carers rated CCT's care as good or excellent, although they were unclear of the role of the CCT. • Problems identified after 3–4 weeks of care included: symptom control (most severe problem), pain control, and family and patient anxiety. • Patients' and carers' negative comments focused on • communication problems with GPs, hospital doctors, and nurses in discussing diagnosis, prognosis, their attitude and difficulties getting through to CCT during out of hours.

Results


Outcome measures

03pm235.qxd 18/5/99 4:19 pm Page 206

Interview survey

Interview survey

Prospective interview survey

51

31

Study design(s)

Continued

49

Reference

Table 2

Terminally ill cancer patients, mean age 66 years, all with a family carer.

Evaluation of community based support teams' services and hospital and community services from family and patient perspective (one main group)

Total: 65 participants

Terminally ill cancer patients, expected to live for 1 year or less Total: 80 family carers of patients Three main conditions: home care n = 31 hospital care n = 41 hospice care n = 8

Males with AIDS. 42% had known about HST for 1 year, 25% less than 6 months Majority still worked full time. Total: 261 Two main groups: HST group n = 119 Control n = 142

A comparison of home care versus hospice care versus hospital care received during last week of life

Comparison of patients views' on Home Support Team (HST) with GP care and outpatient care

Intervention


Patient/lay carer opinion

Patient/lay carer opinion

Morbidity

Carer satisfaction

Utilization of health care


Patient opinion

Outcome measures

• Support teams received most praise by patients (89%) and family (91%) compared to any other profession. • 71% of patients/families rated GP care and district nurses' care as good or excellent. One-third of patients (34%) and over half the family members (54%) rated hospital doctors'/ nurses' care at this level. • Negative comments on care of hospital staff focused on communication, co-ordination of services and being too overworked to care. Complaints re: primary care included communication, attitude and difficulties in getting GPs to do home visits. • Symptom control as most severe problem by both groups.

• Relatives of patients who died at home were significantly more satisfied with site of death, than those whose relatives died in hospital. • 50% of carers of patients who received hospital care were dissatisfied (37/73). • Over a third were dissatisfied with hospital information (39%) • In final week of life, over 50% of patients suffered from pain, dyspnoea, insomnia, and depression; 20% received no effective pain relief.

• Over three-quarters of patients (78%) thought HST staff were kind and understanding; these ratings were slightly higher for social workers (73%), outpatient staff (74%), significantly higher than GP care (49%). • Over half the patients thought the HST provided a lot of reassurance and support (60%). • 58% patients believed that a main role of the HST was to give advice and over half of these found the advice very helpful (64%).

Results

Weakness: no control. Strength: prospective study.

Community-based support teams received the most praise by families and patients. Primary care was rated more favourably than secondary care. Poor communication was identified as most common problem in primary and secondary care.

Although family carers appreciated nursing and medical care for terminally ill cancer relatives, areas of dissatisfaction included: inadequate symptom control, undesirable nature of dying on busy acute wards, difficulties in obtaining information and lack of community services and out of hours support.

People with AIDS viewed the HST care in a positive way, especially in terms of offering support; their views on this type of care compared favourably to other forms of care including social work, outpatient and GP care in particular. Strengths: 1) Prospective study; 2) Patients' own views; 3) Large sample.


03pm235.qxd 18/5/99 4:19 pm Page 207




Prospective interview survey Questionnaire survey Observational study

26

28

40

Study designs(s)

Continued

29

Reference

Table 2

A comparison of care in: general hospital, foundation home, inpatient hospice, outpatient hospice during final 3 months of life.

Views of surviving spouses 13 months after death.

Married cancer patients Total: 80 general hospital n = 20 foundation home n = 20 hospice (inpatient) n = 20 hospice (outpatient) n = 20

Married men/ women, died of cancer under the age of 70. Total 102 surviving spouses Two main conditions: (Matched pairs) St. Christopher's home care group n = 51; Ordinary home care group n = 51

Patients who died of heart disease (38%), cancer (32%), and stroke (12%) Total: Views of relatives of 262 deceased patients

Comparison of carers' views on hospital versus home care.

A comparison of St Christopher's Home Care Team versus ordinary home care.

Patients who had died of cancer In 1967–69 hospice group n = 34 hospital group n = 34 In 1977–79 (matched pairs) hospice n = 30 hospital n = 30

Comparison of hospice care with hospital care – from perspective of surviving spouse Comparison of care in 1967–69 with care in 1977–79

Intervention


• Patients' opinions on physical care, staff, communication with nurses, and place of care were positive across all sites. • Observer ratings showed hospice inpatients were 'less troubled' and less depressed to patients in other conditions. • Hospice outpatients were significantly more anxious, angry and depressed compared to hospice inpatients. Patient opinion Psychological status

Patients gave most praise to outpatient systems of care in spite of experiencing slightly more anxiety. Patients were least distressed in a hospice environment. Weaknesses: 1) ratings made by interviewer, whilst conducting interview; 2) no pre-set criteria for observation; 3) small numbers in each group. Strength: prospective study.

Although hospice patients were able to stay at home longer, their stress levels were higher than the control group, & their need for primary care was not reduced. [However, hospice patients were more seriously ill than the control and had greater nursing needs.] Hospice relatives were very positive about home help, and its spiritual impact. Strength: matched pairs in terms of age, sex, severity of pain before terminal period etc. Weakness: Significant medical differences between groups

• Patients in St Christopher's group were significantly more likely to show slight or great tension and to feel more of a burden. • Hospice group relatives reported significantly greater anxiety whilst patient was in their care than their ordinary care counterparts. • Hospice group patients stayed at home until a later stage in illness. • Surviving relatives expressed positive feelings about help received from home care service and 100% reported improvements in patient's peace of mind and 97% in their own.

Quality of life

Lay carer opinion

Patient and lay carer opinion

Spouses of patients in both groups believed patients suffered less pain and distress during 1977–79 compared to 1967–69. During 1977–79, the hospice group experienced less anxiety. Weaknesses: 1) Statistical results not presented clearly; 2) Closed questionnaire. Strength: matched pairs.


Although all forms of care received some praise, a range of problems were also reported which included: unreliable home care services, difficulties in obtaining information in hospitals and the uncaring attitude of some hospital staff.

• Although anxiety decreased in both conditions over time (1967–79), the hospice group reported significantly less anxiety in 1977–79 than their hospital counterparts. • Spouses in both groups rated patients as suffering from significantly less overall pain and severe distress in 1977–79, compared to 1967–69. • Spouses in both groups experienced significantly less post-bereavement stress and grief in 1977–79 compared to 1967–69.

Results

• Difficulties in home care included: problems in obtaining medical help at night, obtaining equipment, infrequent home GP visits. • Criticisms of hospital care were: difficulties in extracting information, poor symptom control, and an uncaring attitude. • Relatives praised hospital care in 29% of cases, 34% praised hospital nurses, and 24% praised GP skills and support.

Morbidity

Lay carer opinion

Outcome measures

03pm235.qxd 18/5/99 4:19 pm Page 208

41

Reference

Table 2

Cohort study


Study design(s)

Continued

Hospital care versus inpatient hospice care, from perspective of surviving spouse (interviewed 13 months after bereavement)

Intervention

Total: 68 surviving spouses of patients Hospital group n = 34 St Christopher's group n = 34

Patients who died from cancer under age of 65

Subjects: total number, number per condition Lay carer opinion

Outcome measures

• Hospice patients suffered significantly less severe pain/distress than patients in hospital group; and remained more mobile. • More hospice patients (36%) were fully aware of prognosis compared to hospital patients (18%). • Surviving spouses in hospice group were significantly more likely to spend more time with patient and talk more to members of staff. • Nurses, ward sisters and doctors from St. Christopher's were seen as significantly less busy than those at hospitals.

Results

Strengths: 1) Matched pairs; 2) No significant differences between groups.

Inpatient hospice care was reported to be more effective than hospital care by surviving spouses of patients in terms of providing more effective pain control and patients being more mobile and more fully aware of prognosis than hospital patients.


03pm235.qxd 18/5/99 4:19 pm Page 209

03pm235.qxd 18/5/99 4:19 pm Page 210

210

EK Wilkinson et al.

received by 236 deceased patients across 12 hospice programmes.67 Comparisons were made between community-based programmes, two types of hospital-based programmes: scattered beds and autonomous units. Equally high levels of satisfaction were reported by informal carers for services across the three settings, indicating a ‘ceiling’ effect. Smaller-scale observational research studies have shed some light on the perceived benefits of hospice care, including: relaxed visiting hours, home-like atmosphere, attention given to families68 and good nursing care.68,69 Satisfaction with and opinions on models of home care in North America Comparative studies on home care in North America have reaped contrary results in regard to satisfaction with multidisciplinary home care compared to ordinary home care.70,71 Indeed, a RCT examining patient and carer satisfaction with a multidisciplinary, primary home care team, found significantly higher levels for this type of care over a 6-month period.70 Patients in the experimental group had fewer hospitalizations, nursing home visits and outpatient visits. Lay carer satisfaction was associated with physician availability, competence, personal qualities and treatment effects to a lesser extent. In contrast, another comparable RCT of hospital-based home care71 did not find consistently higher levels of satisfaction in the experimental group over a 6-month period compared to the control group.71 Attrition and a substantially smaller sample at follow-up may have influenced this result. Overall, the research has shown that satisfaction with home care services, both specialized and nonspecialized is generally high.70–72 Specific criticisms of palliative care in the home have also been reported. Lay carer concerns have included: inadequate sleep while caring for the patient during his or her final month of life, as well as the need for more overnight help and access to a respite service,73 inadequate pain control and the negative impact of a lack of respite care and distressing symptoms on the family.74 Other research has indicated that carers were not aware of the full range of services offered by community hospice services and these included: financial assistance, legal assistance, education, bereavement support and counselling.72

Discussion Research into the impact of specialist models of palliative care on consumer opinion and satisfaction in the UK has shown that, in general, the care provided by hospitals has been the most subject to criticism compared to any other type of care. Criticisms were wide-ranging, and included complaints regarding the psychosocial aspects of care, communication, staff, the care environment, and co-ordination with other community services. Although inpatient hospice care, outreach teams, and other home care services were generally viewed favourably, the evidence is too patchy to determine whether any particular model of care is preferred over any other. The research over the past 20 years indicates that consumers have consistently appreciated the psychosocial climate and the generally high quality of communication in hospices. However, studies involving hospices have generally been small-scale, and have mainly centred on St Christopher’s Hospice, making generalization problematic. The dearth of good quality, large scale, comparative contemporary studies on hospice home care and ordinary home care services was also apparent. The research into specialist (nonhospice) community outreach teams, although contemporary, is also rather limited. North American studies on the impact of different models of palliative care on consumer opinion and satisfaction did not draw out any consistent trends. Methodological flaws, even in studies based on the most robust design, have been the major reason for the dearth of firm conclusions. It is clear, however, that the findings do highlight the difficulties of carrying out research in a palliative care context. Particular problems include the likelihood of attrition, scarcity of subjects, heterogeneity of case-mix, uncertainties in attributing change to the intervention, and also in quantifying the impact of assessors. Research, particularly from the UK, has highlighted the need for the consideration of the opinions, needs and preferences of terminally ill, noncancer patients. Understanding the similarities and differences in experience, satisfaction and general preferences across different disease groups is paramount in developing effective, models of palliative care that best meet user need. Furthermore, the impact of particular patient characteristics on

Reference

Comparative studies of patient and lay carer opinions or satisfaction with different types of palliative care in North America

Study design(s)

Intervention


Outcome measures

Results


71

Randomized control trial Diary study Interview survey Questionnaire survey

Examined costeffectiveness of hospital-based home care compared to traditional home care. Comparisons were made at referral, one month, 6 months.

Mainly terminal cancer patients with life expectancy of less than 6 months, with an informal caregiver. Total: 171 Two conditions: convention home care n = 85; hospital based home care n = 86

Lay carer satisfaction Patient satisfaction Morale Psychological status

• Patients and carers in the hospital based home care group reported significantly higher levels of satisfaction with care, compared to the control group at 1 month into care programme. • Patients and carers in the hospital based care group reported higher levels of satisfaction with care compared to control group at 6 months but this was not significantly different – possibly due to reduced sample size.

No consistent differences were found between satisfaction levels with hospital-based home care and traditional home care. Strengths: 1) Prospective randomized controlled trial; 2) No significant differences between groups at referral. Weakness: attrition.

72


Families' views on hospice care: at referral, during care, and 3±6 months after bereavement.

Family caregivers of patients who had died of cancer


• Lay carers expected on referral to receive: emotional support whilst caring for patient, respite care and symptom control from hospice care. • Lay carers were 'not sure' or 'did not expect' help with: finances, legal issues, education, bereavement support, counselling & transportation. • Lay carers were 'mostly' to 'very much' satisfied with all hospice services - especially those where they were well informed or held expectations.

Lay carers were generally very satisfied with all aspects of hospice care. They expressed a preference for direct medical and emotional support services as opposed to spiritual and social-environment services. Weaknesses: 1) Retrospective views; 2) Small sample.

A comparison of conventional care with hospice care as part of the National Hospice Study. Comparisons were fortnightly until death of patient.

Cancer patients with a lay carer aged 21 or older Total: 1754 Three conditions: Hospice home care, n = 833 Hospital based hospice, n = 624 Conventional cancer care n = 297

• No significant differences in patient's levels of satisfaction across settings. • Lay carers in the hospice conditions were more satisfied with patient care before and after death compared to their conventional care counterparts. • Lay cares in the hospice conditions were more satisfied with patients' place of death than their conventional care counterparts. (Hospice patients stayed at home longer, and were more likely to die at home)

Terminally ill cancer patients were equally satisfied with hospital based hospice care, home care hospice and conventional care. Families of hospice patients who were more likely to die at home, were more satisfied with the place of patient's death compared to their conventional care counterparts.

66

Interview survey Analysis of routine data

Lay carer opinion

Total: 47 family care givers

Lay carer satisfaction Patient satisfaction Costs Quality of life

03pm235.qxd 18/5/99 4:19 pm Page 211

Table 3

67

74

71

Reference

Interview survey Questionnaire survey Retrospective study

Analysis of routine data

Retrospective study


Interview survey


Interview survey

Randomized controlled trial

Study designs(s)

Table 3 Continued

Evaluation of the New York State Hospice Demonstration Program. Community based hospice versus hospital-based hospice with 'scattered beds', versus hospital-based hospice in an 'autonomous unit'

Male and female terminally ill cancer patients

Home care versus 'institutional care' during final 6 months of life. Interviews were carried out every 2 months until death, and with relatives 3–4 weeks after bereavement.

Hospice care patients Total: 350 lay carers talk about 236 patients. 3 hospice conditions: Community based n = 148 Scattered beds in hospital n = 43, Autonomous unit in hospital n = 159

Total: 122 patients and 96 relatives Two conditions: home care group n = 54 institutional group n = 68

Homebound, chronically or terminally ill elderly, cancer patients, with a prognosis of less than 3 months. Total: 158 at baseline Two conditions: home care group n = 82 control group n = 76

A randomized controlled trial of a multidisciplinary primary home health care team, compared to a control group. Comparisons were at until 6 months, and at 3-monthly intervals.

Intervention


Lay carer/ patient satisfaction Costs Impact of volunteers Utilization of care

Lay carer opinion

Morbidity

Functional status

• Lay carers estimated that patients were very satisfied with a range of services from all types of hospice care (over 90%). • Lay carers held mixed views on patients' emotional and physical quality of life. • Lay carers' estimations of patients' satisfaction with hospital care indicated significant differences across the three models with lower ratings for the community based care group. • Lay carers found bereavement services either very helpful or helpful.

High levels of satisfaction were associated with service provision, emotional support and bereavement support provided to the lay carers. Satisfaction with hospital care in hospital based programmes was higher than for hospital care for community based programmes. Weakness: Details of patients not given.

Patients' relatives were less satisfied with home care services and with availability of care. They reported inadequate pain control in patients and a need for more respite care. In spite of this, relatives preferred home care than any other type of care. Strengths: 1) High response rate (72% of relatives); 2) Validity checks. • Qualities of health care professionals, rather than treatments, were the focus of most examples of satisfaction. • Difficulties reported by 87 home care patients' relatives centred on: physical care (28 relatives), patient's distressing psychological symptoms (27) and the negative impact of caring for patient on home life (18). • Home care relatives were significantly less satisfied with availability of care, their patients experienced more pain than their institutional care counterparts. Home care relatives expressed a preference for home care in spite of above.


Mortality Psychological status Costs

Informal family caregivers expressed significantly higher satisfaction with specialized home care team than control group for their care. There were no significant differences in patients' satisfaction with the different forms of care.


• There were no significant differences between satisfaction levels at baseline, at 3 months and at 6 months for patients in both groups. • Lay carers in the specialized home care team group were significantly more satisfied with care at 3 and 6 months compared to the control group. Their satisfaction correlated with physician availability, competence, personal qualities of physician. • Patients in the specialized home care group had fewer admissions to hospital, nursing homes and outpatient's visits, and used more home services.

Results

Patient and carer satisfaction Utilization

Outcome measures

03pm235.qxd 18/5/99 4:19 pm Page 212

Cohort study

39


Interview survey

Study design(s)

Reference

Table 3 Continued.

A comparison of hospice care with conventional care during last weeks/ months of life.

Intervention Mainly male, terminally ill cancer patients with a prognosis of 2 weeks to 6 months. Total: 247 Two conditions: Hospice group n = 137 Control group n = 110

Subjects: total number, number per condition Results

• Hospice patients were significantly more satisfied with the 'interpersonal care' compared to conventional care patients. • Family caregivers in hospice group were more satisfied & less anxious than those in conventional care group. • No significant differences were apparent between the two groups on measures of pain, activities of daily living, symptoms or affect.

Outcome measures Lay carer/ patient satisfaction Costs Functional status Length of stay Pain control Mortality Symptom control

Although hospice care was not more beneficial than conventional care in pain or symptom relief or alleviation of psychological distress, the higher levels of satisfaction shown by hospice patients and their families indicated an appreciation of other 'qualitative aspects' of hospice care.


03pm235.qxd 18/5/99 4:19 pm Page 213

03pm235.qxd 18/5/99 4:19 pm Page 214

214

EK Wilkinson et al.

satisfaction with models of palliative care has yet to be fully understood. Future research could also address how to improve nonclinical aspects of hospital palliative care, and in particular, how to provide a quiet, peaceful atmosphere where terminally ill patients can die in peace and in dignity, and in the presence of their loved ones if that is their desire. Finally, further research could also assess the impact of self-help groups and volunteers on consumer opinion and satisfaction with palliative care. In conclusion, there are few consistent trends in consumer opinion on and satisfaction with specialist models of palliative care. This may be a reflection of the methodological problems of collecting this type of information within a palliative care environment. Even so, the value of this type of research should not be overlooked by funders or managers of palliative care. Indeed, it can provide invaluable information on consumer perspectives of the quality of communication, access to care, problem areas, and valued aspects of care. This information can be used not only to predict consumer behaviour but also to develop future, more successful models of palliative care. Acknowledgements We acknowledge the work of our late colleague, Debra Weston, who worked on this project until her tragic death in an untimely accident in August 1996. We would like to acknowledge the work of Matthew Wilkinson in designing the Microsoft Access database and also for his advice on data entry. We also thank Alison Lee for her administrative work as research assistant, in particular for locating and photocopying articles. Further thanks to Selina Salisbury for her data entry work. We would like to acknowledge the assistance of the librarians at Imperial College of Medicine at St. Mary’s for their help in devising and carrying out the on line searches. We also thank the librarians at the Halley Stewart Library at St Christopher’s Hospice. We would also like to acknowledge all those researchers who responded to our request for information.

References 1 Kincey J, Bradshaw PJ, Ley P. Patients’ satisfaction and reported acceptance of advice in general practice. J R Coll Gen Pract 1975; 25: 558–66.

2 Roghman K, Hengst A, Zastowny T. Satisfaction with medical care: its measurement and relation to utilisation. Med Care 1998; 17: 461–77. 3 Fitzpatrick R, Hopkin A, Harvard-Watts O. Social dimensions of healing: a longitudinal study of outcomes of medical management of headaches. Soc Sci Med 1983; 17: 501–10. 4 Fitzpatrick R, Bury M, Frank A, Donnelly T. Problems in the assessment outcome in a back pain clinic. Int Rehab Studies 1987; 9: 161–65. 5 Savage R, Armstrong D. Effect of a general practitioner’s consulting style on patients’ satisfaction: a controlled study. Br Med J 1990; 301: 968–70. 6 Bollam M, McCarthy M, Modell M. Patients’ assessment of out of hours care in general practice. Br Med J 1988; 296: 829–32. 7 Barbarin OA, Chesler MA. Relationships with the medical staff and aspects of satisfaction with care expressed by parents of children with cancer. J Comm Health 1984; 9: 302–13. 8 Garland TN, Bass DM, Otto ME. The needs of hospice patients and primary caregivers: comparison of primary caregivers’ and hospice nurses’ perceptions. Am J Hosp Care 1984; 1: 40–45. 9 Grobe ME, Ilstrup DM, Ahmann DL, Miller JC, Gillard M, Haycock H, Jacobsen D. Evaluation of a coordinated community approach to hospice services. Prog Clin Biol Res 1983; 130: 411–15. 10 Mor V, Masterson Allen S. A comparison of hospice vs conventional care of the terminally ill cancer patient. Oncology Huntingt 1990; 4: 85–91. 11 Eriscsson Persson B, Galvan S, Bexell G. Care of senile demented patients in the final stage of the disease as experienced by their close relatives. J Clin Exp Geront 1984; 6: 17–26. 12 Merrouche Y, Freyer G, Saltel P, Rebattu P. Quality of final care for terminal cancer patients in a comprehensive cancer centre from the point of view of patients’ families [see comments]. Support Care Cancer 1996; 4: 163–68. 13 Dawson NJ. Need satisfaction in terminal care settings. Soc Sci Med 1991; 32: 83–87. 14 Wilkinson HJ. Assessment of patient satisfaction and hospice: a review and an investigation. Hosp J 1986; 2: 69–94. 15 Hays RD, Arnold S. Patient and family satisfaction with care for the terminally ill. Hosp J 1986; 2: 129–50. 16 Bruster S, Jarman B, Bosanquet N, Weston D, Erens R, Delbanco TI. National survey of hospital patients. Br Med J 1994; 309: 1542–49. 17 Fitzpatrick R. Surveys of patient satisfaction: I – Important general considerations. Br Med J 1991; 302: 887–89.

03pm235.qxd 18/5/99 4:19 pm Page 215

Patient and carer preference for, and satisfaction with, specialist models of palliative care 215 18 Department of Health. Consumers and research in the NHS: consumer issues within the NHS. London: Department of Health, 1994. 19 Glickman, M. Making palliative care better: quality improvement, multi-professional audit and standards. London. National Council for Hospice & Specialist Palliative Care Services. 1997; occasional paper number 12. 20 Fitzpatrick R. Surveys of patient satisfaction: II – Designing a questionnaire and conducting a survey. Br Med J 1991; 302: 1129–32. 21 Locker D, Dunt D. Theoretical and methodological issues in sociological studies of consumer satisfaction with medical care. Soc Sci Med 1978; 12: 283–92. 22 Delbanco TI. Quality of care through the patient’s eyes. Br Med J 1996; 313: 832–33. 23 Cohen G, Forbes J, Garraway M. Can different patient satisfaction survey methods yield consistent results? Comparison of three surveys. Br Med J 1996; 313: 841–44. 24 Hall JA, Dornan MC. What patients like about their medical care and how often they are asked: a metaanalysis of the satisfaction literature. Soc Sci Med 1988; 27: 935–39. 25 Wales J, Kane R, Robbins S, Bernstein L, Krasnow R. UCLA hospice evaluation study. Methodology and instrumentation. Med Care 1983; 21: 734–44. 26 Wilkes E. Dying now. Lancet 1984; i: 950–52. 27 Field D, Douglas C, Jagger C, Dand P. Terminal illness: views of patients and their lay carers. Palliat Med 1995; 9: 45–54. 28 Parkes CM. Terminal care: evaluation of an advisory domiciliary service at St Christopher’s Hospice. Postgrad Med J 1980; 56: 685–89. 29 Parkes CM, Parkes J. ‘Hospice’ versus ‘hospital’ care – re-evaluation after 10 years as seen by surviving spouses. Postgrad Med J 1984; 60: 120–24. 30 Lobchuk M, Kristjanson L, Degner LF, Blood P, Sloan JA. Perceptions of symptom distress in lung cancer patients: I. Congruence between patients and primary family caregivers. J Pain Symptom Manage 1997; 14: 136–47. 31 Higginson IJ, Wade A, McCarthy M. Palliative care: views of patients and their families. Br Med J 1990; 301: 277–81. 32 Spiller JA. Domiciliary care: a comparison of the views of terminally ill patients and their family caregivers. Palliat Med 1993; 7: 109–15. 33 Fakhoury WKH, McCarthy M, Addington-Hall JM. The effects of clinical characteristics of dying cancer patients on informal caregivers’ satisfaction with palliative care. Palliat Med 1997; 11: 107–15. 34 Hinton J. How reliable are relatives’ retrospective reports of terminal illness? Patients’ and relatives’ accounts compared. Soc Sci Med 1996; 43: 1229–36.

35 Higginson IJ, Priest P, McCarthy M. Are bereaved family members a valid proxy for a patient’s assessment of dying? Soc Sci Med 1994; 38: 553–57. 36 Kelson, M. Consumer involvement initiatives in clinical audit and outcomes. London: College of Health, 1995. 37 Rinck G, Van den Bos TG, Kleijnen J, de Haes HJ, Schade E, Veenhof CH. Methodologic issues in effectiveness research on palliative cancer care: a systematic review. J Clin Oncol 1997; 15: 1697–707. 38 McWhinney IR, Bass MJ, Donner A. Evaluation of a palliative care service: problems and pitfalls [see comments]. Br Med J 1994; 309: 1340–42. 39 Kane RL, Wales J, Bernstein L, Leibowitz A, Kaplan S. A randomised controlled trial of hospice care. Lancet 1984; i: 890–94. 40 Hinton J. Comparison of places and policies for terminal care. Lancet 1979; i: 29–32. 41 Parkes CM. Terminal care evaluation of in-patient service at St Christopher’s Hospice: Part 1 and 2. Postgrad Med J 1979; 55: 517–25. 42 Seamark D. Dying from cancer in community hospitals or a hospice: closest lay carer’s perception. Reporting back the results of R & D conference, Bristol, 1996. 43 Seale C, Kelly M. A comparison of hospice and hospital care for people who die: views of surviving spouses. Palliat Med 1997; 11: 93–100. 44 Seale C, Kelly M. A comparison of hospice and hospital care for the spouses of people who die. Palliat Med 1998; 11: 101–106. 45 Seale C. A comparison of hospice and conventional care. Soc Sci Med 1991; 32: 147–52. 46 Field D, Dand P, Ahmedzai S, Biswas B. Care and information received by lay carers of terminally ill patients at the Leicestershire Hospice. Palliat Med 1992; 6: 237–45. 47 Hinton J. Services given and help perceived during home care for terminal cancer. Palliat Med 1996; 10: 125–34. 48 Hinton J. Can home care maintain an acceptable quality of life for patients with terminal cancer and their relatives? Palliat Med 1994; 8: 183–96. 49 McCann K. The work of a specialist AIDS home support team: the views and experiences of patients using the service. J Adv Nurs 1991; 16: 832–36. 50 Butters E, Higginson I, George R, McCarthy M. Palliative care for people with HIV/AIDS: views of patients, carers and providers. AIDS Care 1993; 5: 105–16. 51 Addington-Hall JM, MacDonald LD, Anderson HR, Freeling P. Dying from cancer: the views of bereaved family and friends about the experiences of terminally ill patients. Palliat Med 1991; 5: 207–14.

03pm235.qxd 18/5/99 4:19 pm Page 216

216

EK Wilkinson et al.

52 Sykes NP, Pearson SE, Chell S. Quality of care of the terminally ill: the carer’s perspective. Palliat Med 1992; 6: 227–36. 53 Brown P, Davies B, Martens N. Families in supportive care – Part II: Palliative care at home: a viable care setting. J Palliat Care 1990; 6(3): 21–27. 54 Townsend J, Frank AO, Fermont D, Dyer S, Karran O, Walgrove A, Piper M. Terminal cancer care and patients’ preference for place of death: a prospective study [see comments]. Br Med J 1990; 301: 415–17. 55 Parkes CM. Home or hospital? Terminal care as seen by surviving spouses. J R Coll Gen Pract 1978; 28: 19–30. 56 McCarthy M, Addington-Hall JM, Altmann D. The experience of dying with dementia: a retrospective study. Int J Geriatric Psychiat 1997; 12: 404–409. 57 Hockley JM, Dunlop R, Davies RJ. Survey of distressing symptoms in dying patients and their families in hospital and the response to a symptom control team. Br Med J Clin Res Ed 1988; 296: 1715–17. 58 Hull MM. Hospice nurses. Caring support for caregiving families. Cancer Nurs 1991; 14: 63–70. 59 Addington-Hall JM, Lay M, Altmann D, McCarthy M. Symptom control, communication with health professionals, and hospital care of stroke patients in the last year of life as reported by surviving family, friends, and officials. Stroke 1995; 26: 2242–48. 60 Addington-Hall JM, MacDonald LD, Anderson HR et al. Randomised controlled trial of effects of coordinating care for terminally ill cancer patients. Br Med J 1992; 305: 1317–22. 61 MacDonald LD, Addington-Hall JM, Anderson HR. Acceptability and perceived effectiveness of a district co-ordinating service for terminal care: implications for quality assurance. J Adv Nurs 1994; 20: 337–43. 62 Jones RVH, Hansford J, Fiske J. Death from cancer at home: the carer’s perspective. Br Med J 1993; 306: 249–51.

63 Cowley S. Supporting dying people. Nurs Times 1993; 89: 52–55. 64 McCarthy M, Lay M, Addington-Hall JM. Dying from heart disease: symptoms and hospital care in the last year of life reported by informal carers. J R Coll Physicians 1996; 30: 325–28. 65 Field D, McGaughey J. An evaluation of palliative care services for cancer patients in the Southern Health and Social Services Board of Northern Ireland. Palliat Med 1998; 12: 83–97. 66 Greer DS, Mor V, Morris JN, Sherwood S, Kidder D, Birnbaum H. An alternative in terminal care: results of the National Hospice Study. J Chron Dis 1986; 39: 9–26. 67 Hannan EL, O’Donnell JF. An evaluation of hospices in the New York State Hospice Demonstration Program. Inquiry 1984; 21: 338–48. 68 Kellar N, Martinez J, Finis N, Bolger A, von Gunten CF. Characterization of an acute inpatient hospice palliative care unit in a U.S. teaching hospital. J Nurs Admin 1996; 26: 16–20. 69 Harrison E. Nurse caring and the new health care paradigm. J Nurs Care Qual 1995; 9: 14–23. 70 Zimmer JG, GrothJuncker A, McCusker J. A randomized controlled study of a home health care team. Am J Public Health 1985; 75: 134–41. 71 Hughes SL, Cummings J, Weaver F, Manheim L, Braun B, Conrad K. A randomized trial of the cost effectiveness of VA hospital-based home care for the terminally ill. Health Serv Res 1992; 26: 801–17. 72 Foster LW. Hospice care: art and science evaluating primary caregiver perceptions. Hosp J 1987; 3: 31–45. 73 Bramwell L, MacKenzie J, Laschinger H, Cameron N. Need for overnight respite for primary caregivers of hospice clients. Cancer Nurs 1995; 18: 337–43. 74 McCusker J. The use of home care in terminal cancer. Am J Prev Med 1985; 1: 42–52.