Journal of Pain and Symptom Management
Volume 36, Issue 3 , Pages 217-227, September 2008

Characteristics and Quality of Life of Patients Who Choose Home Care at the End of Life

  • Marianne Ahlner-Elmqvist, RN, BEd

      Affiliations

    • Department of Clinical Sciences, Lund University, Lund, Sweden
    • Department of Otorhinolaryngology, Malmö University Hospital, Malmö University, Malmö, Sweden
    • Faculty of Health and Society, Malmö University, Malmö, Sweden
    • Corresponding Author InformationAddress correspondence to: Marianne Ahlner-Elmqvist, RN, BEd, Department of Otorhinolaryngology, Malmö University Hospital, SE-20502 Malmö, Sweden.
    • ,
  • Marit S. Jordhøy, MD, PhD

      Affiliations

    • Pain and Palliation Research Group, Department of Cancer Research and Molecular Medicine, NTNU
    • Cancer Unit, Innlandet Hospital Trust, Gjøvik, Norway
    • ,
  • Kristin Bjordal, MD, PhD

      Affiliations

    • Palliative Care Research Unit and Department of Radiation Oncology, The Norwegian Radium Hospital-Rikshospitalet Medical Centre, Montebello, Oslo, Norway
    • ,
  • Magnus Jannert, MD, PhD

      Affiliations

    • Department of Clinical Sciences, Lund University, Lund, Sweden
    • Department of Otorhinolaryngology, Malmö University Hospital, Malmö University, Malmö, Sweden
    • ,
  • Stein Kaasa, MD, PhD

      Affiliations

    • Unit for Applied Clinical Research, NTNU
    • Palliative Medicine Unit, Department of Radiotherapy and Oncology, University Hospital of Trondheim, Trondheim, Norway

Accepted 31 October 2007. published online 09 April 2008.

Article Outline

Abstract 

Cancer patients with advanced disease and short-survival expectancy were given hospital-based advanced home care (AHC) or conventional care (CC), according to their preference. The two groups were compared at baseline to investigate whether there were differences between the AHC and the CC patients that may help explain their choice of care. The patients were consecutively recruited over 2½ years. Sociodemographic and medical data, and the health-related quality of life (HRQL) of the two groups were compared. HRQL was assessed using a self-reporting questionnaire, including the European Organization for Research and Treatment of Cancer Core Quality of Life Questionnaire (EORTC QLQ-C30), the Impact of Event Scale (IES), five questions about social support, and two items concerning general well-being. The AHC group showed significantly poorer functioning on all the EORTC QLQ-C30 scales and an overall higher symptom burden than the CC patients. Fewer of the AHC patients were receiving cancer treatment. The AHC patients had lived longer with their cancer diagnosis, had a significantly shorter survival after study enrollment, and a significantly poorer performance status. The major differences between the two groups seemed to be related to being at different stages in their disease. The results indicate that patients are reluctant to accept home care until absolutely necessary due to severity of functioning impairments and symptom burden. These findings should be taken into consideration in planning palliative care services.

Key Words: Palliative care, advanced home care, preferences, health-related quality of life, EORTC QLQ-C30

 

Back to Article Outline

Introduction 

Several factors may influence cancer patients' choice of care in their home or in hospital at the end of life.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 These factors may be related to the individual and his or her illness, or to the environment.13 The importance of environmental factors, such as social support, availability of informal care, health care input, and organization of health care services, have been thoroughly investigated and confirmed in several studies. Results regarding the patients' characteristics are less consistent, and disease-related factors have been little explored. Most studies have focused on diagnoses, whereas there are few data on the influence of symptoms and well-being.

One patient-related factor that is consistently reported as an important determinant for place of care until death is the patient's preference. Preferences are, however, found to change with time and the progression of the disease5, 14 and, as pointed out by Thomas et al.,15 seem to be a “social contingent” rather than “an abstractly arrived certainty.” The patients' wishes are influenced by a range of factors, including their perception of the quality and quantity of informal care, experiences of health care services, as well as more existential perspectives.15, 16 The extent to which health-related quality of life (HRQL) may contribute is poorly documented. It seems, however, reasonable to assume that disease-related factors may be of importance, and hence, that groups of patients choosing different alternatives for end-of-life care, may differ in terms of symptom burden and HRQL.

Detailed knowledge about the characteristics of patients who choose different kinds of palliative care services, and the factors that may influence their preferences, is paramount for the planning and organization of care. Such knowledge is also crucial from a methodological point of view. To evaluate the external validity of a given study, or to compare the results from different groups or settings, the patient samples must be characterized by means of established indicators with which major sociodemographic and medical characteristics are usually well described. Any imbalances between groups can thus be considered in the analysis, or taken into account in the interpretation of the results. However, factors such as attitudes, problems, and needs also may have considerable effects on the study outcome. If these are not adequately described, or if their importance is not known, erroneous conclusions may be drawn.

A study was conducted to compare a hospital-based advanced home care (AHC) program with conventional care (CC).2 A comprehensive assessment of the patients' HRQL was made, and a wide range of sociodemographic and medical factors was recorded. The main question was: Are there any differences in clinical characteristics, symptom burden, or HRQL between patients who prefer AHC and those who prefer CC?

Back to Article Outline

Methods 

Study Design 

The study was a prospective nonrandomized study conducted at the University Hospital of Malmö. Subjects considered for enrollment were patients at the Departments of Oncology, Respiratory Medicine, Hematology, Surgery, Otorhinolaryngology, Urology, and Gynecology at the University Hospital, who lived in the city of Malmö. Inclusion criteria were incurable, malignant disease, life expectancy 2–12 months, age 18 years or older, being informed of the diagnosis and being mentally and physically capable of filling out the HRQL questionnaire. When approached, the patients were given the choice of being referred to the AHC program or continued CC. After the choice, they were asked to participate in the study. Patients who declined to participate in the study still received AHC or CC according to their own preference. The patients completed a self-reporting questionnaire (52 items) at inclusion and thereafter monthly until withdrawal or death. The questionnaire included the European Organization for Research and Treatment of Cancer Core Quality of Life Questionnaire (EORTC QLQ-C30),17 the Impact of Event Scale (IES),18 five questions concerning social support from the MacAdam assessment of suffering in terminal illness scale,19 and two items concerning general well-being derived from a Norwegian health survey (NT).20 A research nurse collected sociodemographic and medical data from the patients' medical records in addition to interviewing the patients. To confirm the representativeness of the study sample, some information (gender, age, performance status, site of cancer disease and later, place of death) was also collected from the nonparticipating patients receiving AHC or CC. The main outcome of the present study is the patients' choice of care. In addition, the sociodemographic and medical characteristics and HRQL profiles of the two groups are explored.

The Regional Ethical Review Board at Lund University approved the study, and return of the first HRQL questionnaire was regarded as written consent.

The AHC Program 

The AHC program was initiated at the beginning of the present study. The program was developed by the Department of Oncology at the University Hospital in Malmö, and has been described in detail elsewhere.2 Briefly, it is a seven-day-a-week service, complementary to existing inpatient and local authority health care, and was initially dimensioned to include 25 patients at home. The staff is multidisciplinary, including experienced nurses, oncologists, physiotherapist, and social worker. AHC provides medical interventions such as blood transfusions and chemotherapy, in addition to symptom treatment, counseling, and emotional and family support. “Back-up” beds are available for the AHC patients, in the Department of Oncology and the hospice, in the University Hospital. CC included inpatient and outpatient services from the University Hospital, local authority health care, and access to inpatient hospice care.

Subjects and Recruitment 

Patients were consecutively recruited over a 2½ year period. To ensure optimal recruitment, information meetings were arranged in all participating departments before the study was started, and the nurses and physicians received written information about the aims, inclusion criteria, and the importance of identifying all eligible patients. During the recruitment period, a full-time research nurse visited all the departments once or twice a week. The physician in charge was consulted before approaching possibly eligible patients, and a nurse on the ward prepared the patients for the arrival of the research nurse. Recruitment rates were continuously monitored, and every month, “trial news” was distributed to the participating departments, containing information on the progress of the study and reinforcement of the aims and inclusion criteria.

The HRQL Instruments 

The EORTC QLQ-C30 has been found valid and applicable for use in a palliative care setting.21 The questionnaire includes 30 items and is composed of six scales that evaluate physical (five items), emotional (four items), role, cognitive and social functioning (two items each) as well as global health status (two items). Higher mean scores on these scales represent better functioning. There are also three symptom scales measuring pain, nausea and vomiting (two items each), and fatigue (three items), and six single items assessing financial impact and various physical symptoms. Higher mean scores on these scales/items represent more symptoms.17, 21, 22 Before analysis, all scales and item scores are linearly transformed to a 0–100 scale, and for missing items, the mean values of those completed are imputed if at least half of the items from a scale are completed.

The IES is a 15-item scale that measures psychological distress in response to stressful events such as having incurable cancer.23, 24, 25 Several scoring systems have been used.23 In the present study, each item was scored on a categorical response scale, ranging from “Not at all” to “Often.” The IES item scores are summarized into two subscales: intrusion and avoidance. The seven-item intrusion subscale (range of scores 0–35) describes how thoughts and impressions related to the disease reappear. The eight-item avoidance subscale (range of scores 0–40) assesses behavior characterized by denying the meaning and consequences of the disease. Higher subscale scores indicate more distress. Subscale scores above 20 are defined as pathological values, and patients with such scores are identified as “cases.”24, 26

Social support from family, friends, and other social networks was measured by five questions derived from MacAdam's questionnaire, “Assessment of suffering in terminal illness.”19, 27 Each item was scored on a four-point categorical response scale, ranging from “Not at all” to “Very much.”

To assess general well-being, two items from a Norwegian national health survey20 were included. One item was a seven-point categorical response scale ranging from being “Very strong and healthy” to “Very tired and worn out.” The other item asked whether the respondent felt calm and good about him/herself, and had a four-point categorical response scale from “Almost all the time” to “Never.”

Statistical Analyses 

For the purpose of the present study, only baseline data were analyzed and compared between study groups. Simple descriptive statistics were applied, and to test for statistical significance of differences, Chi-square and Mann-Whitney U tests were used for the sociodemographic and medical variables,28 whereas HRQL scores were tested using analysis of variance (ANOVA). Due to multiple testing, the level of statistical significance was set at P=0.01.

To illustrate the level of functioning impairment and symptom burden of the patients, comparisons were made with normative EORTC QLQ-C30 data from a Swedish sample and with scores for the two items from the Norwegian national health survey for the general well-being items. In this comparison, age and gender corrections were performed.

To detect a difference of 0.5 SD between treatment groups with 85%–90% power and alpha 0.05, a sample of 75–85 patients in each group would be necessary.29, 30 For the EORTC QLQ-C30 scales, 0.5 SD normally approximates 10–15 points. Thus, the present study may be regarded to have the power to detect clinically significant differences in HRQL between treatment groups, that is, a change of ≥10.31, 32

The statistical software SPSS version 13.0 for Windows was used in the statistical analysis.33

Back to Article Outline

Results 

A total of 722 patients were identified, of whom 477 fulfilled the inclusion criteria and were asked to participate in the study. Of these, a total of 181 (38%) patients had chosen AHC, whereas 296 (62%) had chosen to remain within CC. Study participation was refused by 62 (34%) of the AHC patients and 118 (40%) of the CC patients. Thus, overall, 180 of 477 (38%) patients did not take part in the study—83 patients declined upon the first request, and 97 patients did not return the baseline HRQL questionnaire. The study sample comprised 297 patients; 119/297 (40%) AHC patients and 178/297 (60%) CC patients (Fig. 1).

Characteristics of the Participants 

The AHC and CC groups were comparable in terms of major sociodemographic characteristics (Table 1). The median ages were 67 years and 68 years in the AHC and CC groups, respectively. The male/female ratio was similar, as was the level of education (>12 years in school; AHC 18% and CC 15%). More AHC patients were not living alone (75% compared to 68% in the CC group), but the difference was not statistically significant. The number of children at home younger than 18 years was similar, 12%14 in the AHC group and 11%20 in the CC group. However, a larger proportion of the AHC patients received practical help from family and friends than in the CC group (98% compared to 86%, P<0.001), and more AHC patients were supported by the local health authority care (14% compared to 5%, P=0.006).

Table 1. Characteristics at Inclusion of Participants and Non-Participants
ParticipantsNon-participants
AHCCCP-valueaTotalTotalP-valueb
119178297180
%%%%
Gender
Males52470.35749500.803
Females4853 5150

Age
Median67680.67167700.010
(Min-max)38-8828–85 28–8824–88

KPS
≥7055720.0026540<0.001
<704528 3560

Primary cancer diagnosis
Breast8110.0111080.201
Lung1321 1827
Gastrointestinal3745 4136
Prostate83 54
Urological136 97
Othersc2114 1618

Weeks from diagnosis to inclusion
Median55230.01632270.327
(Min-max)1–5651–564 1–5651–829

Weeks from inclusion to death
Median1423<0.001195<0.001
(Min-max)1–1112–152 1–1520–129

Place of death
Home45100.00125250.771
Hospital2263 4647
Hospice3327 2928

KPS=Karnofsky performance status.

aMann-Whitney U test between participants in AHC and CC.

bMann-Whitney U test between participants and non-participants.

cHematological, gynecological, skin/melanoma, head and neck, and brain/eye.

A number of differences were revealed regarding disease-related characteristics. Lung and gastrointestinal cancers were more frequent in the CC group, whereas cancers classified as “others” were more frequent among the AHC patients (Table 1). A significantly smaller proportion of the AHC patients were receiving tumor treatment (chemotherapy), 36 (30%) compared to 104 (59%) in the CC group (P<0.001). The AHC patients had poorer Karnofsky performance status (KPS) (Table 1).

The median time from diagnosis to trial entry was 55 weeks in the AHC group compared to 23 weeks in the CC group, and median survival time after inclusion was 14 weeks in the AHC group and 23 weeks in the CC group.

Thus, the time from diagnosis to death differed by 23 weeks between the AHC and CC groups (P<0.001) (Fig. 2). In addition, the place of death varied between groups, as discussed elsewhere.2 The occurrence of distant metastases was, however, comparable between the groups (80%), metastases in the lymph nodes and skeleton being most common, followed by metastases in the liver. Comorbidity was frequent in both groups; approximately 40%–50% had one or more chronic diseases and/or impaired hearing or vision.

Comparison Between Patients Who Participated and Those Who Did Not 

Overall, the patients who did not participate in the study were older than those who gave their consent, their KPS was significantly lower, and time from inclusion to death was shorter. When comparing between participants and non-participants in the AHC and CC group, respectively, the same differences were revealed (Table 1).

Health-Related Quality of Life 

Compliance 

As completion of the baseline HRQL was mandatory for trial inclusion, no questionnaires were missing. The frequencies of missing items were low and comparable between groups. For the EORTC QLQ-C30 and the NT items, a mean of 1% (range 0%–4%) was missing. For the IES and social support items, the corresponding values were 1.5% (1%–3%) and 3% (2%–4%), respectively.

EORTC QLQ-C30 

At trial inclusion, the AHC patients reported statistically and clinically significantly poorer functioning on all EORTC QLQ-C30 functioning scales than the CC patients, that is, poorer physical, emotional, role, cognitive and social functioning, as well as poorer global health. There were differences of similar magnitude in symptom scales and items regarding fatigue, pain, dyspnea, and appetite loss. These findings are further underlined by marked differences between the groups in the percentage of patients with functioning scores below 30, and symptom scores (fatigue, pain, dyspnea, and appetite loss) above 70 (Table 2). No differences were found in sleep disturbance, constipation, diarrhea, or financial impact (Table 2). In comparison with normative data from a Swedish population, both groups had substantially impaired functioning and a high symptom burden (Table 2).

Table 2. EORTC QLQ-C30 at Inclusion
AHCCCΔaP-valuebAHCCCNormdatac
n=119n=178AHC vs. CC%%n=2602
Functioning scalesdMeanMean % level function<30Mean
Physical295122<0.001552585
Role173720<0.001714485
Emotional506313<0.001211583
Cognitive587315<0.00112788
Social4558130.001271890
Global health354490.003372976

Symptom scalese % level symptoms >70
Fatigue775621<0.001623122
Nausea/vomiting322480.0161073
Pain5341120.002292021

Single items
Dyspnea564016<0.001321519
Sleep disturbances363330.4549719
Appetite loss5341120.00926214
Constipation403190.03016137
Diarrhea181440.173625
Financial impact252410.6521087

IES mean
Intrusion1613 0.008
Avoidance1715 0.047

Mean scale score and frequencies (%) of patients with low functioning (<30) or high level of symptoms (>70).

aClinically significant Δ≥10.

bANOVA.

cNormative data Swedish population sample of EORTC QLQ-C30.

dHigh score represent high/healthy level of functioning.

eHigh score represent high level of symptoms/problems.

Psychological Distress—IES 

Psychological distress, assessed by the IES, was relatively high in both groups, but overall more pronounced among the AHC patients. Mean scores on the intrusion scale for the AHC and CC groups were 16 and 13, respectively. With regard to avoidance, the mean scores were comparable: 17 for the AHC group and 15 for the CC group (Table 2). More patients in the AHC group (49%) were above the cut-off point for classification as a “case” (>20) on at least one of the two subscales compared to the CC group (37%) (P=0.036) (data not shown).

Social Support 

Almost all patients in both groups felt supported by family and friends, and around 60% felt needed by their family. The AHC patients worried more about the future of their family, and had less contact outside the family (Table 3).

Table 3. Fraction (%) of Patients Who Responded “Quite a Lot” or “Very Much” on the Items in the MacAdam Support Questionnaire
At InclusionAHC (n=78) (%)CC (n=125) (%)P-valuea
Worried about families future54340.001
Support from family and friends95930.104
Families need of you59650.256
Share feelings with others77710.009
Contact outside the family60660.078

aANOVA.

General Well-Being—Comparison with Norwegian Normative Data 

Although most patients reported feeling tired and worn out at study entry, the AHC patients reported more fatigue (being “Tired and worn out” or “Very tired and worn out”) (69%) than the CC patients (48%) (P<0.001). Regarding the other question about “Feeling calm and good,” 58% of the AHC patients and 49% (P=0.2) of the CC patients answered that they felt calm and good about themselves, “Now and again.” In the AHC group, 8% of the patients never “Felt calm and good” compared to 6% of the CC patients (Fig. 3). Overall, both groups of patients reported reduced physical and psychological well-being compared with data from a Norwegian population study.20

  • View full-size image.
  • Fig. 3 

    Comparison between a normative population and the study patients at inclusion on “About feeling strong and healthy or tired and worn out” and “About feeling calm and good about themselves.”

Back to Article Outline

Discussion 

In the present study, patients were given AHC or CC according to their preferences. Our findings show that this procedure resulted in substantial differences between the two groups at the time of study enrollment, mainly in terms of HRQL scores and medical characteristics, but also regarding some sociodemographic factors.

Living alone was somewhat less frequent among those patients who chose AHC, a large proportion of whom died at home. More AHC patients also received help from family and friends and/or were familiar with the local health care services at study entry. Thus, the AHC group seemed to have better access to formal and informal care than the CC group, although the majority of patients from both groups reported feeling supported by their loved ones. Because the intervention in this study consists of around-the-clock care for the patients at home, one may have expected that the amount of social support of family members would not have had a major influence on the choice of type of care. These findings strongly support findings from other studies on the importance of the patients' social network as a determinant for where cancer patients are cared for and die.5, 7, 8, 13, 15, 16, 34, 35, 36, 37, 38, 39, 40, 41 Therefore, social network variables should be regarded as highly important in the planning of place of care. It may be prudent to plan for care in institutions of those with little social support.

Other major differences between the groups were concerned with disease-related factors. The AHC patients had lived longer with their disease, were closer to death, had poorer performance status, poorer functioning and more symptoms, and more patients were no longer on chemotherapy. With regard to symptoms, the most pronounced difference between the groups was found for fatigue, a finding that was confirmed by the scores from the NT question “Tired and worn out.” The differing factors as stated are closely interrelated. A significant association between poorer functioning on all EORTC QLQ-C30 scales and poorer KPS has been reported previously, as has also an increase in pain, appetite loss, and fatigue as performance status deteriorates.42 The KPS is an important predictor for survival, with poorer scores indicating shorter survival, and patients' HRQL scores are closely related to the time from assessment to death, that is, the closer to death, the poorer the functioning and the higher the symptom burden.42, 43, 44, 45 Thus, most of the differences may be related to the fact that the two groups of patients were included at different points in the course of their disease, although differences in the distributions of diagnoses also may have contributed.

The extent to which each of the factors that differed between the study groups may have influenced the patients' choice of AHC, cannot be determined from our results. However, it is highly likely that poor performance status, poor functioning, and multiplicity of symptoms are important determinants. According to our clinical experience, patients and relatives are often reluctant to accept a palliative care program instead of a “traditional oncology program,” especially if the support is not required immediately. This is consistent with studies reporting that there are mixed views about the presence of professionals in the home, as some patients and relatives may regard this as intrusive.16 Furthermore, late enrollment into palliative care is a frequently cited problem. Other studies have shown levels of physical impairment and fatigue among patients admitted to palliative care services comparable to those in the present AHC group.44, 46 Patients entering similar programs have been found to have short, intensive periods of care prior to the last month of life,40 also indicating that the severity of problems might be decisive in the patients' choice. Because the AHC group consists of patients with more symptoms and poorer function, it is a challenge to offer these patients optimal symptom control at home, that is, systematic monitoring of symptoms, intensive treatment, nutrition and fluid support, etc. These findings underline the need for physicians and nurses who are highly trained in palliative care in such programs.

Our findings have some limitations. First of all, after having chosen their preferred place of care, a large proportion of the patients declined to participate in the study, the refusal rate being higher among those preferring CC (40%) than AHC (34%). Consequently, the question arises of whether the participating AHC and CC patients were representative of all the patients choosing AHC and CC. When comparing the characteristics of the participants to non-participants, we found that non-participation were a few years older, had a lower KPS, and were closer to death in both groups. This is consistent with findings from other studies,46, 47 and indicates that participation had occurred according to the same pattern in the two groups. We, therefore, believe that the differences in characteristics represent the true differences between those choosing AHC and CC.

Another factor that warrants discussion is that CC was chosen more often than AHC (62% vs. 38%). One reason might be that the desire for home care was less in a Swedish cancer population living in a densely populated area than reported in studies from other countries.13, 41, 48, 49, 50 Another explanation may be that AHC was a new concept, which most patients and relatives were unfamiliar with, and, therefore, reluctant to try. The limited number of 25 AHC patients also may have reduced the enrollment to this group, that is, health professionals may, consciously or unconsciously, have advocated this option less positively if no AHC “bed” were available at the time an eligible patient was identified. Thus, culture, knowledge about the service, and its availability may have had an effect on the patients' preferences and decisions.

Third, the distribution of diagnoses in our study sample did not correspond to the incidence in the population. In particular, breast and prostate cancers are underrepresented due to the lack of formal cooperation with the departments treating these patients during the study period. Thus, to firmly establish the external validity of our findings, similar analyses of patients with other diagnoses should be undertaken.

Although a systematic review of patients' wishes regarding place of death concluded that home care is the most common preference, our findings clearly indicate that it might be difficult to convince patients to accept AHC early in the palliative phase, before the symptom burden is severe.14 Patients' preferences and needs change with disease progression, and introducing the right service at the right time is a challenge. However, this is necessary not only to fulfill the wishes of more patients, but also to optimize the utilization of services. Early involvement of a palliative counseling team could help identify when supportive care should be replaced by more AHC. However, as new cancer therapies are introduced, the use of active tumor treatment in later palliative stages may hamper the timely introduction of palliative care counseling. The palliative care community, in close cooperation with oncologists, must discuss this.

Back to Article Outline

Conclusions 

Preferences for AHC are likely to be influenced by several factors. Patients who choose AHC are in the later stages of their disease and have poorer physical functioning, more symptoms, and more social needs than those who do not choose such care. The introduction of AHC at the appropriate point in time requires close follow-up and adequate assistance during the entire palliative phase. Discussions with the patients and their families may be called for, as there seems to be a tendency to postpone participation in programs involving strangers in the home. Further investigations on patient- and disease-related factors that may influence the patient's preferences for AHC are needed.

Back to Article Outline

Acknowledgments 

The authors would like to thank all the patients who participated in the trial and Eva Almqvist Wedeberg for her invaluable work as a research assistant.

Back to Article Outline

References 

  1. Jordhoy MS, Fayers P, Saltnes T, et al. A palliative-care intervention and death at home: a cluster randomised trial. Lancet. 2000;356(9233):888–893
  2. Ahlner Elmqvist M, Jordhoy MS, Jannert M, Fayers P, Kaasa S. Place of death: hospital-based advanced home care versus conventional care. A prospective study in palliative cancer care. Palliat Med. 2004;18(7):585–593
  3. Axelsson B, Christensen SB. Place of death correlated to sociodemographic factors. A study of 203 patients dying of cancer in a rural Swedish county in 1990. Palliat Med. 1996;10(4):329–335
  4. Townsend J, Frank AO, Fermont D, et al. Terminal cancer care and patients' preference for place of death: a prospective study. BMJ. 1990;301(6749):415–417
  5. Hinton J. Which patients with terminal cancer are admitted from home care?. Palliat Med. 1994;8(3):197–210
  6. De Conno FCA, Groff L, Brunelli C, et al. Effect of home care on the place of death of advanced cancer patients. Eur J Cancer. 1996;32A(7):1142–1147
  7. Cantwell P, Turco S, Brenneis C, et al. Predictors of home death in palliative care cancer patients. J Palliat Care. 2000;16(1):23–28
  8. Brazil K, Bedard M, Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care. 2002;1(1):2
  9. Costantini M, Higginson IJ, Boni L, et al. Effect of a palliative home care team on hospital admissions among patients with advanced cancer. Palliat Med. 2003;17(4):315–321
  10. Hearn J, Higginson IJ. Do specialist palliative care teams improve outcomes for cancer patients? A systematic literature review. Palliat Med. 1998;12(5):317–332
  11. Gyllenhammar E, Thoren-Todoulos E, Strang P, et al. Predictive factors for home deaths among cancer patients in Swedish palliative home care. Support Care Cancer. 2003;11(9):560–567
  12. Chvetzoff G, Garnier M, Perol D, et al. Factors predicting home death for terminally ill cancer patients receiving hospital-based home care: the Lyon comprehensive cancer center experience. J Pain Symptom Manage. 2005;30(6):528–535
  13. Gomes B, Higginson IJ. Factors influencing death at home in terminally ill patients with cancer: systematic review. BMJ. 2006;332(7540):515–521
  14. Higginson IJ, Sen-Gupta GJ. Place of care in advanced cancer: a qualitative systematic literature review of patient preferences. J Palliat Med. 2000;3(3):287–300
  15. Thomas C, Morris SM, Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med. 2004;58(12):2431–2444
  16. Gott M, Seymour J, Bellamy G, Clark D, Ahmedzai S. Older people's views about home as a place of care at the end of life. Palliat Med. 2004;18(5):460–467
  17. Aaronson N, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993;85(5):365–376
  18. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective stress. Psychosom Med. 1979;41(3):209–218
  19. MacAdam D. An initial assessment of suffering in terminal illness. Palliat Med. 1987;1:37–47
  20. Holmen J, Midthjell K, Bajartveit K. The North-Troendelag Health Survey 1984-1986. Report No. 4. Oslo and Verdal, Norge: National Institute of Public Health; 1990;
  21. Kaasa S, Bjordal K, Aaronson N, et al. The EORTC core quality of life questionnaire (QLQ-C30): validity and reliability when analysed with patients treated with palliative radiotherapy. Eur J Cancer. 1995;31A(13–14):2260–2263
  22. Fayers P, Aaronson NK, Bjordal K, Sullivan M. EORTC QLQ-C30 scoring manual. Brussels: EORTC Data Center; 1995;
  23. Briere J, Elliott DM. Clinical utility of the impact of event scale: psychometrics in the general population. Assessment. 1998;5(2):171–180
  24. Mahon SM, Cella DF, Donovan MI. Psychosocial adjustment to recurrent cancer. Oncol Nurs Forum. 1990;17(3 Suppl):47–52discussion 53–54
  25. Sundin EC, Horowitz MJ. Horowitz's Impact of Event Scale: evaluation of 20 years of use. Psychosom Med. 2003;65(5):870–876
  26. Kaasa S, Malt U, Hagen S, et al. Psychological distress in cancer patients with advanced disease. Radiother Oncol. 1993;27(3):193–197
  27. Fowlie M, Berkeley J, Dingwall-Fordyce I. Quality of life in advanced cancer: the benefits of asking the patients. Palliat Med. 1989;3:55–59
  28. Altman DG. Practical statistics for medical research. 8th ed.. London: Chapman & Hall; 1999;
  29. Fayers P, Machin D. Quality of life: Assessment, analysis and interpretation of patient-reported outcomes. 2nd ed.. Chichester: Wiley; 2007;
  30. Machin D, Campbell MJ, Fayers PM, Pinol APY. Sample size tables for clinical studies. Oxford: Blackwell Scientific; 1997;
  31. Osoba D, Rodrigues G, Myles J, Zee B, Pater J. Interpreting the significance of changes in health-related quality-of-life scores. J Clin Oncol. 1998;16(1):139–144
  32. King MT. The interpretation of scores from the EORTC quality of life questionnaire QLQ-30. Qual Life Res. 1996;5:555–567
  33. SPSS Inc. . SPSS (Statistical Package for the Social Sciences), Release 13.0 for Windows. Chicago: SPSS; 2003–2004;
  34. Bruera E, Kuehn N, Emery B, Macmillan K, Hanson J. Social and demographic characteristics of patients admitted to a palliative care unit. J Palliat Care. 1990;6(4):16–20
  35. Bradshaw PJ. Characteristics of clients referred to home, hospice and hospital palliative care services in Western Australia. Palliat Med. 1993;7(2):101–107
  36. Costantini M, Camoirano E, Madeddu L, et al. Palliative home care and place of death among cancer patients: a population-based study. Palliat Med. 1993;7(4):323–331
  37. Addington-Hall J, Altmann D, McCarthy M. Which terminally ill cancer patients receive hospice in-patient care?. Soc Sci Med. 1998;46(8):1011–1016
  38. Fried TR, Pollack DM, Drickamer MA, Tinetti ME. Who dies at home? Determinants of site of death for community-based long-term care patients. J Am Geriatr Soc. 1999;47(1):25–29
  39. Grande GE, Todd CJ. Why are trials in palliative care so difficult?. Palliat Med. 2000;14(1):69–74
  40. Grande GE, McKerral A, Todd CJ. Which cancer patients are referred to hospital at home for palliative care?. Palliat Med. 2002;16(2):115–123
  41. Tang ST. When death is imminent: where terminally ill patients with cancer prefer to die and why. Cancer Nurs. 2003;26(3):245–251
  42. Jordhoy MS, Fayers P, Loge JH, et al. Quality of life in advanced cancer patients: the impact of sociodemographic and medical characteristics. Br J Cancer. 2001;85(10):1478–1485
  43. Jordhoy MS, Fayers P, Loge JH, Ahlner-Elmqvist M, Kaasa S. Quality of life in palliative cancer care: results from a cluster randomized trial. J Clin Oncol. 2001;19(18):3884–3894
  44. Lundh Hagelin C, Seiger A, Furst CJ. Quality of life in terminal care—with special reference to age, gender and marital status. Support Care Cancer. 2006;14(4):320–328
  45. Strömgren AS, Sjogren P, Goldschmidt D, et al. A longitudinal study of palliative care: patient-evaluated outcome and impact of attrition. Cancer. 2005;103(8):1747–1755
  46. Strömgren AS, Goldschmidt D, Groenvold M, et al. Self-assessment in cancer patients referred to palliative care: a study of feasibility and symptom epidemiology. Cancer. 2002;94(2):512–520
  47. Kopp I, Lorenz W, Rothmund M, Koller M. Relation between severe illness and non-completion of quality-of-life questionnaires by patients with rectal cancer. J R Soc Med. 2003;96(9):442–448
  48. Tang ST, McCorkle R. Determinants of congruence between the preferred and actual place of death for terminally ill cancer patients. J Palliat Care. 2003;19(4):230–237
  49. Beccaro M, Costantini M, Rossi PG, et al. Actual and preferred place of death of cancer patients. Results from the Italian survey of the dying of cancer (ISDOC). J Epidemiol Community Health. 2006;60(5):412–416
  50. Tiernan E, O'Connor M, O'Siorain L, Kearney M. A prospective study of preferred versus actual place of death among patients referred to a palliative care home-care service. Ir Med J. 2002;95(8):232–235

 The study was supported by grants from The Swedish Cancer Society (grant no. 3650-B95-01XAC), The Vardal Foundation for Health Care Sciences and Allergy Research (grant no. V98262), and The Association for Cancer and Traffic Victims (grant no. C24405).

PII: S0885-3924(08)00116-4

doi:10.1016/j.jpainsymman.2007.10.010

Journal of Pain and Symptom Management
Volume 36, Issue 3 , Pages 217-227, September 2008

Article Tools

* Image Usage & Resolution