Supportive Care in Cancer

, Volume 17, Issue 10, pp 1247–1254

Administrators’ perspectives on end-of-life care for cancer patients in Japanese long-term care facilities

Authors

    • Department of System Management in Nursing, Graduate School of Health Care SciencesTokyo Medical and Dental University
  • Mitsunori Miyashita
    • Department of Adult Nursing/Palliative Care Nursing, School of Health Sciences and Nursing, Graduate School of MedicineThe University of Tokyo
  • Tatsuya Morita
    • Department of Palliative and Supportive Care, Palliative Care Team, and Seirei HospiceSeirei Mikatahara General Hospital
  • Takayuki Ichikawa
    • Department of Adult Nursing/Palliative Care Nursing, School of Health Sciences and Nursing, Graduate School of MedicineThe University of Tokyo
  • Nobuya Akizuki
    • Psycho-Oncology Division, Research Center for Innovative OncologyNational Cancer Center Hospital East
  • Miki Akiyama
    • Faculty of Policy ManagementKeio University
  • Yutaka Shirahige
    • Shirahige Clinic
  • Kenji Eguchi
    • Department of Internal Medicine and Medical OncologyTeikyo University School of Medicine
Supportive Care International

DOI: 10.1007/s00520-009-0665-8

Cite this article as:
Fukahori, H., Miyashita, M., Morita, T. et al. Support Care Cancer (2009) 17: 1247. doi:10.1007/s00520-009-0665-8

Abstract

Purpose

The purpose of this study was to clarify administrators’ perspectives on availability of recommended strategies for end-of-life (EOL) care for cancer patients at long-term care (LTC) facilities in Japan.

Methods

A cross-sectional survey was conducted with administrators at Japanese LTC facilities. Participants were surveyed about their facilities, reasons for hospitalization of cancer patients, and their perspectives on availability of and strategies for EOL care.

Results

The 97 responses were divided into medical facility (n = 24) and non-medical facility (n = 73) groups according to physician availability. The most frequent reasons for hospitalization were a sudden change in patient’s condition (49.4%), lack of around-the-clock care (43.0%), and inability to palliate symptoms (41.0%). About 50% of administrators believed their facilities could provide EOL care if supported by palliative care experts. There was no significant difference between facility types (P = 0.635). Most administrators (81.2%) regarded unstable cancer patients as difficult to care for. However, many (68.4%) regarded opioids given orally as easy to administer, but regarded continuous subcutaneous infusion/central venous nutrition as difficult. Almost all administrators believed the most useful strategy was transferring patients to hospitals at the request of patients or family members (96.9%), followed by consultation with palliative care experts (88.5%).

Conclusion

Although LTC facilities in Japan currently do not provide adequate EOL care for cancer patients, improvement might be possible with support by palliative care teams. Appropriate models are necessary for achieving a good death for cancer patients. Interventions based on these models are necessary for EOL care for cancer patients in LTC facilities.

Keywords

Palliative careAdministratorsLong-term care facilitiesCancer care

Introduction

With a rapidly aging society, long-term care facilities and residential homes for older people are becoming more common in Japan as well as in western countries [20, 21]. In the USA, over 1.75 million people reside in nursing homes, and 43% of individuals who reach age 65 will spend time living in a nursing home [5, 6]. In Japan, 1.4% of those over 65 years old live in nursing homes [20] and approximately 0.8 million people used institutional services financed through the long-term care insurance system in 2006 [23]. Some researchers predicted that the number of residents would increase because implementation of long-term care insurance promoted willingness to use institutional care for older adults [10, 11, 23]. Many Japanese elders are likely to enter long-term care facilities in the near future.

Some residents and their family members might prefer nursing homes as a place for dying, though only about 2% of Japanese people died in nursing homes in 2003 [20]. In western countries, patients tend to prefer nursing homes and assisted living facilities to acute care hospitals as a site of death [10, 14, 16, 18]. To promote end-of-life (EOL) nursing home care, in 2006, the Japanese government amended the insurance law so that nursing homes could receive additional reimbursement for providing EOL care. The number of deaths in the facilities or residential homes is expected to increase in the near future in Japan.

Currently, more than 80% of patients with cancer die in general hospital wards in Japan [13]. However, more patients with cancer may die in long-term care settings in coming years because there are few hospices or palliative care units. In the near future, long-term care facilities in Japan are likely to have to care for more terminal cancer patients than at present.

EOL care for cancer patients, however, might be especially difficult for facilities in Japan due to lack of trained staff, including doctors, nurses, and nursing assistants, inappropriate pain/symptom management, and negative attitudes toward EOL care [19, 22, 24, 25]. Earlier research indicated that the quality of EOL care might also be low in long-term care facilities in western countries [2, 3, 12]. For example, pain and other symptoms may be managed inappropriately [14, 19]. Hanson et al. have reported that surviving family members expressed greater dissatisfaction with nursing homes than with any other component of terminal care [4, 7].

To improve palliative care for cancer patients in the community, including long-term care facilities, the Ministry of Health, Labor, and Welfare in Japan launched a nationwide intervention research project, the Outreach Palliative Care Trial of Integrated Regional Model Study (OPTIM-Study), which is a 5-year project from 2006 to 2011 with a yearly budget of 2.5 million US dollars. The details of the OPTIM-Study design are published elsewhere [26]. The present survey was conducted as a part of the OPTIM-Study.

As decision makers in medical facilities, administrators influence the quality of care that is provided by those facilities [7, 17]. Little is known, however, about administrators’ perspectives on EOL care for cancer patients in long-term care facilities and residential homes in Japan. The purpose of the present study is to clarify administrators’ perspectives on the current availability of and recommended future strategy for EOL care for cancer patients at long-term care facilities in Japan in preparation for the OPTIM-Study. This study provides necessary information for considering and creating future effective interventions and systems of EOL care for cancer patients in the community, including long-term care facilities, in Japan.

Methods

Design

This cross-sectional study was conducted in Chiba, Shizuoka, Nagasaki, and Yamagata Prefectures in Japan. These areas were openly recruited for the OPTIM-Study and selected because of the specific features of EOL care resources in each community. Specifically, Chiba Prefecture has a national cancer center that plays a central role in the EOL care system, Shizuoka Prefecture has a general hospital that is renowned for palliative care and plays a central role in EOL care, and Nagasaki Prefecture has an efficient support system managed by an association of regional primary care physicians. In contrast, Yamagata Prefecture was selected because it did not have any noteworthy resources at the time.

Participants and procedures

We mailed a structured survey to 275 administrators in all of the long-term care facilities in four areas. Researchers had developed the survey questionnaire from earlier research and their own clinical experience. Administrators received the questionnaire and were informed of the survey protocol. The 103 administrators who consented to participate in the study replied anonymously. We excluded six responses because of missing data regarding facility type. The responses with missing data on other items were included in the study in order to maximize the sample size for analysis. During statistical analysis, items with missing responses were excluded.

Questionnaire

Characteristics of facilities

The questionnaire included items on types of targeted facilities. For analysis, we divided the study sample into two groups according to the availability of physicians: medical facility and non-medical facility. We chose this simple classification due to the small sample size, though there is considerable variety among long-term care facilities in Japan [9]. The survey also asked about other characteristics of the facilities. Size (number of beds) was divided into three groups. The number of nurses working at night was summarized dichotomously. We also included questions on the number of cancer patients in the previous year who had died, either in the facility or acute-care hospitals after transfer, and the number of cancer patients who had died in the facility without being transferred in the previous year.

Reasons for transfer to acute-care hospitals

Respondents rated the frequency of ten reasons for transfer of cancer patients to acute-care hospitals. The reasons included items such as “unexpected change in patient’s condition.” Each reason was rated on a five-point Likert-type scale (1 = none, 2 = rarely, 3 = sometimes, 4 = often, and 5 = always) and responses were divided dichotomously.

Administrators’ perspectives on availability of EOL care

We asked the administrators about their opinion on the feasibility of EOL care in their facilities under a variety of situations. First, we inquired whether it was possible to provide cancer patients with EOL care if palliative care experts supported them. Second, we asked whether they could accept respite stay and admissions of four types of cancer patients into their facilities if palliative care experts supported them. The types of patients illustrated included ones such as “respite stay of stable cancer patient without intravenous line.”

Administrators’ perspectives on availability of palliative therapies

Administrators were asked about the availability of six types of therapy if palliative care experts were available to patients, such as opioids given orally and continuous subcutaneous infusion of morphine. Participants rated the questions using three categories (1 = impossible, 2 = possible with expert support, and 3 = possible) and responses were divided dichotomously.

Administrators’ perspectives on strategies to provide EOL care

Finally, participants answered questions about the usefulness of coordinating interdisciplinary teams for six coordinated actions, which included items such as “regular visits by palliative care experts,” “contact with practitioners in specialized home care support clinics whenever needed,” and “transfer to hospitals whenever patients or family desire admission.” Specialized home care support clinics were uniquely defined in 2006 in Japan as facilities that provide home care for a wide range of patients in the community and include long-term care facilities [26]. Administrators rated the responses using a five-point Likert-type scale (1 = useless, 2 = somewhat useful, 3 = useful, 4 = very useful, and 5 = indispensable) and responses were divided dichotomously.

Analyses

Descriptive comparisons between facility types were conducted utilizing cross-tabular breakdowns with Fisher’s exact test for dichotomous variables and Kruskal–Wallis test for ordinal variables. Significance was defined as P < 0.05. All analyses were conducted using SAS 9.1. The study method was reviewed and approved by the review board of the OPTIM-Study organized by the Ministry of Health, Labor, and Welfare.

Results

Characteristics of study sample

Of the 97 administrators whose data were included in the analysis, 24 were from medical facilities (MF) and 73 were from non-medical facilities (NMF). MF included ten geriatric hospitals and 14 geriatric health service facilities; NMF included 36 nursing homes, 21 public facilities similar to assisted living facilities in the USA, and 16 private for-profit nursing homes.

Table 1 shows the characteristics of the participants. About half of the facilities had 41 to 80 beds. Sixty percent of them did not have night-shift nurses. This was especially true of most of the NMF. There were no cancer patient deaths in 30% of the facilities, including death after being transferred to acute-care hospitals. Also, 60% of the facilities had no cancer patients.
Table 1

Characteristics of facilities

 

Total

Medical facility

Non-medical facility

P value

n (%)

n (%)

n (%)

Number of beds (n = 97)a

    

 1–40

13 (13.4)

3 (12.5)

10 (13.7)

0.002

 41–80

55 (56.7)

6 (25.0)

49 (67.1)

 81–436

29 (29.9)

15 (62.5)

14 (19.2)

Nurses working at night (n = 95)a

    

 Yes

36 (37.9)

23 (100.0)

13 (81.9)

<0.001

 No

59 (62.1)

0 (0.0)

59 (18.1)

Death of cancer patients in hospital or facility after a transfer (n = 93)a

    

 0

30 (32.3)

4 (19.0)

26 (36.1)

0.012

 1–5

57 (61.3)

12 (57.1)

45 (62.5)

 6–10

6 (6.5)

5 (23.8)

1 (1.4)

Death of cancer patients in the facility without being transferred (n = 93)a

    

 0

56 (60.2)

12 (54.5)

44 (62.0)

0.368

 1–5

35 (37.6)

8 (36.4)

27 (38.0)

 6–10

2 (2.2)

2 (9.1)

0 (0.0)

aSample size for items varied because of missing values

Reasons for transfer to acute-care hospital

Table 2 shows possible reasons for hospitalization of cancer patients. The most frequent reasons were unexpected change in patient’s condition (49.4%), followed by lack of around-the-clock care during nights and weekends (43.0%), and the inability to palliate pain and other symptoms (41.0%). Of the ten reasons, significant differences between facility types were observed for the lack of doctor (P = 0.025) and caregiving burden felt by family members (P = 0.018).
Table 2

Reasons for transfer to acute-care hospital

Reasons for hospitalization

na

Total (%)b

Medical facility (%)b

Non-medical facility (%)b

P value

Unexpected change in patient’s condition

83

49.4

50.0

49.2

1.000

Lack of around-the-clock care during nights and weekends

79

43.0

37.5

45.5

0.623

Inability to palliate pain or other symptoms

83

41.0

33.3

44.1

0.463

Lack of nurses

79

34.2

30.4

35.7

0.796

Lack of careworkers

80

27.5

16.7

32.1

0.183

Anxiety of patient or family member

55

27.3

25.0

28.2

1.000

Lack of doctor

80

26.3

8.3

33.9

0.025

Inability to palliate delirium

79

19.7

29.2

15.4

0.216

Caregiving burden felt by family member

70

18.6

36.4

10.4

0.018

Lack of collaboration between doctors and nurses/family members

76

13.2

4.3

17.0

0.266

aSample size for items varied because of missing values

bPercentage for responses of often/always among five options including none, rarely, sometimes, often, and always

Administrators’ perspectives on availability of EOL care

The first line in Table 3 presents administrators’ perspective on the ability of their facilities to care for cancer patients with support from palliative care experts. Overall, slightly more than 50% of administrators speculated that they could provide EOL care for cancer patients. Fisher’s exact test did not show a difference between facility types (P = 0.635).
Table 3

Administrators’ perspectives on availability of end-of-life care

Questions

Total (n = 96)

Medical facility (n = 24)

Non-medical facility (n = 72)

P value

Is it possible to provide cancer patients with end-of-life care to the end if palliative care experts support them? (%)a

56.3

62.5

54.2

0.635

Is it possible to accept respite stay of stable cancer patients without intravenous lines if palliative care experts support them? (%)b

69.8

83.3

65.3

0.126

Is it possible to accept respite stay of stable cancer patients without intravenous lines but taking opioids orally if palliative care experts support them? (%)b

60.4

70.8

56.9

0.335

Is it possible to accept admissions for end-of-life care for stable cancer patients expected to live for some time with a chronic illness like post-stroke if palliative care experts support them? (%)b

48.4

66.7

42.3

0.058

Is it possible to accept admissions for end-of-life care for unstable patients using continuous subcutaneous infusion of morphine to control symptoms if palliative care experts support them? (%)b

18.8

37.5

12.5

0.013

aPercentage for response of possible between two options including possible and impossible

bPercentage for response of possible among three options including impossible, difficult, and possible

Administrators believed that they could accept respite stay and admission of stable cancer patients with palliative care expert support. However, if a patient was unstable and used continuous subcutaneous infusion of morphine, admission was regarded as difficult. Administrators of NMF considered it more difficult than administrators of MF (P = 0.013).

Administrators’ perspectives on availability of palliative therapies

Table 4 presents the availability of six palliative therapies with support from palliative care experts. Administrators believed that using opioids given orally was easiest, and continuous subcutaneous infusion and central venous nutrition were the most difficult routes of administration. MF had greater availability than NMF for all therapies, especially central venous nutrition (P = 0.002).
Table 4

Administrators’ perspectives on availability of palliative therapies

Types of therapies

na

Total (%)b

Medical facility (%)b

Non-medical facility (%)b

P value

Opioids given orally

95

68.4

73.9

66.7

0.612

Intravenous hydration

94

52.1

65.2

47.9

0.160

Subcutaneous hydration

93

41.9

60.9

35.7

0.051

Continuous subcutaneous infusion of sedative

93

36.6

52.2

31.4

0.085

Continuous subcutaneous infusion of morphine

93

35.5

52.2

30.0

0.078

Central venous nutrition

93

33.3

60.9

24.3

0.002

aSample size for items varied because of missing values

bPercentage for response of possible/possible with expert support among three options including possible, possible with expert support, and impossible

Administrators’ perspectives on strategies to provide EOL care

Table 5 shows administrators’ perspectives on coordination to care for cancer patients. Administrators thought that the most useful coordination was transport to hospitals whenever patients or families desired admission (96.9%), followed by consultation with palliative care experts (88.5%), and contact with physicians in specialized home care support clinics (85.3%). Rental of medical devices was regarded as the least useful (60.4%). Differences between facilities were shown only for contact with physicians in specialized home care support clinics (P = 0.004).
Table 5

Administrators’ perspectives on strategies to provide end-of-life care

Items for coordination

na

Total (%)b

Medical facility (%)b

Non-medical facility (%)b

P value

Transfer to hospital whenever patients or family desire admission

97

96.9

95.8

97.3

1.000

Consultation with palliative care experts about palliation whenever needed

96

88.5

79.2

91.7

0.136

Contact with physicians in specialized home care support clinic whenever needed

95

85.3

65.2

91.7

0.004

Regular visit of palliative care experts

95

80.0

69.6

83.3

0.229

Contact with visiting nurses whenever needed

92

72.8

60.9

76.8

0.177

Rental of rarely used medical devices from regional hospitals

91

60.4

54.5

62.3

0.618

aSample size for items varied because of missing values

bPercentage for response of useful, very useful, and indispensable among five options including useless, somewhat useless, useful, very useful, and indispensable

Discussion

This article has described reasons for hospitalization of cancer patients who have been admitted to Japanese long-term care facilities and administrators’ opinions on the feasibility of EOL care, availability of palliative therapies, and effective therapeutic coordination strategy for EOL care for cancer patients. The results also indicated some differences in administrators’ opinions according to the type of long-term care facility (medical vs. non-medical).

The results showed that the most frequent reason for hospitalization was the unexpected change in a patient’s condition. This result differs from Buchanan’s research, which indicated that non-clinical factors such as patient preferences were more important in the decision to hospitalize a resident. Lack of an advance directive system in Japan might also contribute to this difference. To prevent the unexpected deterioration of a patient’s condition or symptoms, educational outreach by palliative care specialists, which is included in the OPTIM-Study [26], might be effective. Additionally, this approach would also be useful in improving the ability to palliate pain and other symptoms, a reason that was cited for hospitalization. Lack of around-the-clock care, which was noted as another reason, might arise partly because NMF were not required to employ a night-shift physician or nurses. Previous research also indicated that inadequate staffing was an obstacle for the provision of EOL care [1, 8]. Coordination with physicians in specialized home care support clinics is likely to be effective for supporting such institutions in providing EOL care. Addressing these causes of hospitalization may help to reduce unnecessary hospitalization of cancer patients. This, in turn, may improve patients’ quality of life and decrease medical costs.

It should be noted that half of the administrators believed that their facilities could provide EOL care with proper support by palliative care experts. Moreover, there was no significant difference according to type of facility. These results were similar to Sloane’s work, which suggested that EOL care in assisted living facilities appeared to be similar in process and outcomes to that provided in nursing homes [18]. Our results suggest that Japanese long-term care facilities might be able to provide EOL care for cancer patients with proper support, despite gaps between facility types. Palliative care team consultation, such as what occurs in hospices in American nursing homes [12, 15], might facilitate EOL care in Japanese long-term care facilities and improve the quality of care. Therefore, regional palliative care centers included in the OPTIM-Study, which provide coordination and information services, may be effective. The administrators also believed that their facilities could accept respite stay or admission of stable cancer patients, though they would not admit unstable cancer patients. As a whole, although patients with unstable symptoms or pain should be cared for in hospice or hospitals with a palliative care team, EOL care for cancer patients in long-term care facilities might be feasible with a proper intervention system in Japan.

As for specific palliative therapies, central venous nutrition and continuous subcutaneous infusion for morphine and sedative agents were regarded as difficult to provide in these facilities. These difficulties may be due to the paucity of physicians and nurses in long-term care facilities. However, the administration of opioids given orally and intravenous hydration could be feasible with proper support and existing staff. Zimmerman et al. reported that staff of long-term care facilities emphasized the need for more education in EOL care [27]. Education for less invasive palliative procedures might be effective in promoting palliative care in such institutions. Standardized tools and guidelines derived from the OPTIM-Study will be useful for this type of education [26].

Administrators said that the need for coordination of services was greatest in the transport of patients to other hospitals when patients or family desired admission elsewhere. This request is noteworthy for EOL care for cancer patients in institutions in Japan, as people generally do not regard a long-term care facility as a site for dying. Therefore, patients and family members might prefer hospitalization when cancer is detected or causes symptoms. Moreover, resources such as hospice or outreach programs for EOL care in Japanese long-term care facilities do not exist at this time. Accordingly, administrators are not confident that their facilities can care for cancer patients until death. Administrators’ concerns about hospital support are reasonable because the preferences of patients and family members should be valued at the EOL. Coordination between hospitals and facilities providing EOL care is indispensable.

Consultations and regular visits by palliative care experts were also highly desired. These results suggest that interventions of the OPTIM-Study, establishing community palliative care teams and educational outreach programs, would be needed for long-term care facilities. Although contact with practitioners in specialized home care support clinics was also required, administrators of MF regarded this as less important because MF had at least one full-time physician. In addition, MF administrators may also have assumed that clinic physicians do not have specialized knowledge in EOL cancer care and that physicians who have EOL care competency might be required. In NMF, physicians in specialized home care support clinics are important because many of their facilities have no regular doctor.

The results of this study should be interpreted in light of its limitations. This study has limited applicability because of the small sample size and low response rate. However, this study is noteworthy for being the first published article on Japanese EOL care for cancer patients in institutions for elderly people.

In spite of the study limitations, the results of this study have implications for health care policy and future research. Although EOL care for cancer patients in Japanese institutions is currently insufficient, improvement might be possible with appropriate interventions, such as staff education or support by efficient interdisciplinary expert teams. Future research should create appropriate models for coordination that lead to a good death for cancer patients in Japanese institutions. Previous research conducted abroad has indicated that intervention studies were effective in improving the quality of care [3, 6]. The results of studies in various countries indicate that interventions using OPTIM-Study processes could open new frontiers for EOL care for cancer patients in long-term care facilities in Japan.

Financial disclosure

This study was supported by a grant-in-aid for the Third Term Comprehensive Strategy for Cancer Control in Japan.

Authors’ contributions

Morita, Miyashita, Akizuki, Akiyama, Shirahige, and Eguchi were responsible for the study concept and design; Miyashita, Ichikawa, Akizuki, Akiyama, and Shirahige for the acquisition of subjects and data; Fukahori and Miyashita for the analysis and interpretation of data; and Fukahori, Morita, Miyashita, Ichikawa, Akizuki, Akiyama, Shirahige, and Eguchi for the preparation of the manuscript.

Sponsor’s role

None.

Copyright information

© Springer-Verlag 2009