Family's Difficulty Scale in End-of-Life Home Care: A New Measure of the Family's Difficulties in Caring for Patients with Cancer at the End of Life at Home from Bereaved Family's Perspective

Abstract

Background:

The aim of this study was to develop a tool to measure the family's difficulties in caring for cancer patients at the end of life at home: Family's Difficulty Scale in end-of-life home care (FDS).

Methods:

The draft of the FDS was derived from a pilot interview survey and literature reviews. The questionnaires were sent to 395 bereaved family caregivers whose family members were patients with terminal cancer receiving home service.

Results:

We obtained 306 responses (response rate, 81%). Factor analysis resulted in 29 items and 8 factors: Burden of Care, Concerns about Home Care Doctor, Balance of Work and Care, Patient's Pain and Condition, Concerns about Visiting Nurse, Concerns about Home Care Service, Relationship between Family Caregivers and their Families, and Funeral Preparations. The cumulative rate of contribution was 71.8%. Cronbach coefficient α for the FDS was 0.73–0.75; the intraclass correlation coefficient in the test–retest examination was 0.75–0.85. Evidence for construct validity was confirmed by convergent and divergent validity. Concurrent validity was confirmed by significant correlations between identified factors and concurrent measures.

Conclusion:

The validity and reliability of this new instrument were confirmed. This scale should help home care providers to assess and focus on family difficulties and provide individualized care for the family who cares for a patient with terminal cancer at home.

Introduction

Cancer has been increasing and has been the leading cause of death since 1981 in Japan.¹ Consequently, the Japanese government enacted the basic law for the provision of cancer care in 2007. This provision included increasing home care for patients with cancer.² The medical care of patients with cancer has changed from an emphasis on hospitalization to the promotion of treatment at home in Japan.³

In Japan, most people hope for death at home, but a gap between hope and reality exists. In a study concerning end-of-life care in Japan, more than half of the general public hoped for home care at the end of life and death at home.⁴

Approximately 60% of bereaved family members of patients with cancer who had home care also hoped for home care at the end of life and approximately 70% hoped for death at home.⁵ However, only can7.3% of patients with cancer die at home in Japan even though the policy regarding home care is promoted and many Japanese hope for home care at the end of life.⁶ Thus, in recent years, there is a delay of home care for patients with cancer in Japan. So strategies to fulfill the policy of home care are needed in Japan.

In a previous study, patients who hoped to stay at home were unwillingly admitted to a hospital because their symptoms were out of control at home and their families were anxious and the care burden was excessive.^7,8 The continuity of home care for patients with cancer at the end of life is likely to be attributed to the family caregiver's attitude and empowerment.^9–11 Therefore, family caregivers have an important role in home care of patients and are more likely to face difficulties. For this reason, we recommend that a survey of the family's difficulties is needed. Some studies have evaluated the care burden of family caregivers who care for the elderly or people with dementia in Japan.^12–14 However, few the studies have focused on the family of the patient with terminal cancer and there are no studies that have measured the family's difficulties associated with the care for the patient with terminal cancer at home.

In this study, we aimed to develop an instrument that would measure family's difficulties in caring for patients with cancer at the end of life at home and to examine its validity and reliability. We selected bereaved family caregivers as participants, because from an ethical aspect, we believed that participation in a survey for current family caregivers was too much of a burden. Home care providers could use such a tool to assess the family's needs and concerns and use this information to individualize care for the family.

Methods

Pilot study

A pilot study was conducted to develop a questionnaire to evaluate the family's difficulties. We had a semistructured interview to seven bereaved family caregivers and five home care providers (home visiting nurses and care managers). An interview guide was used, which was based on literature reviews relating to home care of terminal patients with cancer. The inclusion criteria of the family caregivers were: (1) patient required care at home for more than 5 days, (2) time elapsed more than 6 months but within 2 ½ years postbereavement, (3) the family caregiver and the patient were over 20 years old, (4) the family caregiver was a primary caregiver, and (5) the family caregivers knew the patient's diagnosis of cancer. The inclusion criteria of the home care providers were 5 or more years of experience of caring for patients with terminal cancer.

The family caregivers were asked to describe the difficulties that they encountered in caring for a terminally ill family member at home and home care providers were asked to comment about the difficulties that families faced. A content analysis was undertaken to analyze the interviews.¹⁵

The resulting questionnaire about the family's difficulties in caring for their family member was based on the interview analysis and literature reviews. In designing the questionnaire, we solicited the opinions of the homecare physicians and home visiting nurses who were engaged in palliative care and two researchers involved in palliative care. The draft of the Family's Difficulty Scale in end-of-life home care (FDS) included 44 items.

Participants

The facilities included in the study were home visit nursing stations, care facilities that managed home care service, and a clinic associated with home care. The inclusion criteria for the facilities were that they provided 24-hour visits and care for patients with cancer at the end of life. We contacted 27 facilities; 5 refused to participate and 2 did not meet the participant criteria. Therefore, 20 facilities were included in this study. The participants were the bereaved family caregivers of patients with terminal cancer using the home care services of these facilities. The inclusion criteria for participants were the same as the pilot study. Participants who had dementia, mental disorders, or intense grief were excluded from this study.

Procedure

This was a cross-sectional study. We sent a set of questionnaires to the facilities, then the facilities mailed a questionnaire to each participant. We requested that the participants mail the completed questionnaire in an enclosed envelope within a month. We regarded the return completed questionnaire as consent of the participants. Participants who were willing were asked to complete another questionnaire to examine test–retest reliability 2 weeks after they completed the first questionnaire. The ethical and scientific validity of this study was approved by the Institutional Review Board of Graduate School of Medicine, University of Tokyo.

Measurement

Participants were asked to complete the 44-items of the FDS (draft version). Each item was rated on a 6-point scale from 1 (very much) to 6 (not at all) and higher scores indicated a greater degree of difficulty. To examine the instrument's concurrent validity, participants were also asked to complete several other measures, Caregiver Reaction Assessment—Japanese version (CRA-J) and Care Evaluation Score—Short version (CES). Developed by Given et al.¹⁶ the CRA-J is a measure to assess the care burden of the caregiver. The validity and reliability of the Japanese version has been verified.¹⁷ Developed by Morita et al.^18,19 the CES is a measure to assess the quality of care for patients with terminal cancer. The validity and reliability have been established for the bereaved family of patients with terminal cancer. Participants were also asked to fill out basic demographic information.

Item reduction (FDS)

We examined the ceiling and floor effect for 44 items of the FDS. As a result of factor analysis, we deleted some items with low factor loadings (<0.3) and high factor loadings among more than two factors (>0.4). And we also deleted some items that indicated high factor lodings among items.

Data analysis

Scale development

To assess the internal structure of the instrument, a factor analysis using the least-square method without weighting by promax rotation was performed. The selection of factors for rotation was based on the criteria of eigenvalues greater than one.

Scale Validity

Convergent validity

Convergent validity was confirmed by determining the correlation between each item score and the associated domain score.

Divergent validity

Divergent validity was confirmed by determining the correlation between each item score and domain scores that were not associated with the item.

Construct validity

The results of convergent and divergent validity were used to confirm the rate of success of the scale using the Multitrait Scaling analysis. The calculations showed that the rate of convergent correlation was higher than divergent correlation through all correlations of each domain.

Concurrent validity

Pearson's product-moment correlation coefficient demonstrated concurrent validity among some domains of the FDS and some subscales of the CRA-J and CES.

Scale reliability

The internal consistency of the FDS was examined by determining Cronbach coefficient α. We also calculated intraclass correlation coefficients. Test–retest reliability was demonstrated by correlations between the initial FDS questionnaire and the second FDS questionnaire administered 2 weeks later.

All tests were carried out and reported on a p<0.05 level and two-tailed basis by using the statistical package, SAS version 9.1 (StataCorp, College Station, TX).

Results

Demographic data

A total of 395 questionnaires were sent to participants and 18 were returned due to an incorrect address. Of the 377 questionnaires that were actually sent, 306 were returned (81%). A total of 280 (74%) were analyzed; 26 were excluded from analysis because 15 were returned as refusals and 11 were more than 30% incomplete. For the retest, we sent a set of questionnaires to 196 participants who were willing to complete it; 172 patients completed and returned it; 170 were analyzed because two questionnaires were excluded for more than 50% incomplete information.

The backgrounds of the patients and their bereaved family caregivers are summarized in Table 1. Of the participants, 79% were women and 61% were more than 60 years old. More than half were spouses of the patient, and 69% of the patients died at home.

Table 1.

Background of Patients and the Bereaved Family Caregivers (n=280)

	n	%
Bereaved Families
Gender
Male	58	20.7
Female	222	79.3
Age in years
20–29	1	0.4
30–39	13	4.6
40–49	32	11.4
50–59	64	22.9
60–69	86	30.7
Over 70	84	30.0
Relationship
Spouse	158	56.4
Child	80	28.6
Child-in-law	23	8.2
Parent	7	2.5
Sibling	7	2.5
Other	5	1.8
Patients
Gender
Male	162	57.9
Female	118	42.1
Age in years (mean±SD)		72.8±11.9
Marital status
Married	199	71.1
Unmarried, estranged, bereavement	81	28.9
Place of death
Home	193	68.9
Hospital	86	30.7
Unknown	1	0.4
Period from diagnosis to death
Less than 6 months	65	23.2
Less than 1 year	53	18.9
Less than 3 years	87	31.1
Less than 5 years	27	9.6
More than 5 years	47	16.8
No response	1	0.4

SD, standard deviation.

FDS Scale

We confirmed the structure of eight factors as a result of factor analysis (Table 2). We distributed 29 items into 8 domains to form the scale. These factors were labeled as follows: (1) Burden of Care, (2) Concerns about Home Care Doctor, (3) Balance of Work and Care, (4) Patient's Pain and Condition, (5) Concerns about Visiting Nurse, (6) Concerns about Home Care Service, (7) Relationship between Family Caregivers and their Families, and (8) Funeral Preparations. Cronbach α ranged from 0.73 to 0.75 (Table 3).

Table 2.

Factor Validity of Family's Difficulty Scale in End-of-Life Home Care

	Standardized regression coefficients
	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5	Factor 6	Factor 7	Factor 8	Communality
Burden of Care mean±SD=86.7±23.0
Family caregivers struggle to provide daily care for the patient.	0.95	0.09	−0.21	−0.01	0.01	0.02	−0.01	0.04	0.81
Family caregivers sacrifice for patient-centered life.	0.80	0.03	0.08	−0.05	−0.09	0.03	0.04	−0.02	0.70
Family caregivers experience hectic times everyday.	0.79	0.02	−0.04	0.10	0.03	−0.06	−0.02	0.01	0.66
Family caregivers become shut-in because of patient care.	0.70	−0.02	0.19	−0.03	−0.03	0.03	0.10	−0.06	0.67
Family caregivers experience physical stress during patient's care.	0.61	−0.12	0.14	0.00	0.09	−0.01	0.05	0.03	0.50
Family caregivers get insufficient sleep due to anxiety about the patient.	0.58	−0.06	0.08	0.22	0.01	0.00	−0.04	0.05	0.56
Concerns about Home Care Doctor mean±SD=39.2±20.4
No explanation from home care doctor about dealing with patient's possible symptoms.	−0.01	0.90	0.01	0.02	0.01	−0.06	0.02	0.00	0.80
Insufficient control of patient's symptoms by home care doctor.	−0.02	0.85	0.03	0.05	0.01	0.00	−0.02	−0.04	0.74
Family caregivers cannot express their ideas to home care doctor even though they may have different ideas.	0.04	0.82	−0.02	−0.03	0.05	0.05	−0.04	0.01	0.74
Family caregivers have insufficient consultation time with home care doctor.	−0.09	0.79	−0.06	0.07	−0.07	0.10	0.10	0.04	0.69
Home care doctors do not visit the patients when they need a home visit.	0.07	0.75	0.07	−0.08	0.08	−0.06	−0.03	0.01	0.60
Balance of Work and Care mean±SD=68.7±32.7
Family caregivers have to give up work to continue the patient's care.	−0.01	−0.02	0.94	0.00	0.01	0.01	0.01	0.01	0.88
Difficulties in managing work (or hobby) and the patient's care because care requires excessive time.	0.06	0.01	0.92	−0.01	−0.04	−0.02	0.00	0.03	0.91
Family caregivers have to consider whether they will quit their work (or hobby) because of caregiving.	0.11	0.04	0.79	0.02	−0.01	0.00	0.01	0.06	0.77
Patient's Pain and Condition mean±SD=91.0±23.5
Family caregivers cannot do anything to relieve patient's suffering.	−0.05	0.02	−0.02	0.85	−0.02	−0.01	0.05	0.04	0.71
Family caregivers cannot deal with symptoms of the patient.	0.08	0.01	−0.16	0.84	−0.01	−0.04	0.06	0.12	0.82
Bitterness due to patient's suffering.	0.06	−0.02	0.10	0.80	0.01	0.01	−0.03	−0.10	0.69
Bitterness due to patient's weakness.	0.22	0.01	0.14	0.54	0.03	0.07	−0.14	−0.12	0.50
Concerns about Visiting Nurse mean±SD=34.7±18.0
Visiting Nurses provide insufficient care for the patient.	0.03	−0.02	−0.01	−0.01	0.99	0.01	−0.03	0.00	0.97
Poor relationship between visiting nurses and patient.	0.04	0.09	−0.01	−0.03	0.85	0.01	0.03	0.00	0.83
Visiting nurses have insufficient knowledge to care for the patient.	−0.06	0.10	−0.01	0.04	0.85	0.00	0.02	0.00	0.84
Concerns about Home Care Service mean±SD=56.0±23.3
Insufficient information on how to correctly use home care services.	0.03	0.06	−0.02	−0.06	−0.02	0.96	0.01	−0.04	0.91
Family caregivers do not know how to use home care services.	−0.06	−0.04	−0.03	0.06	0.04	0.85	0.03	0.08	0.78
High cost of home medical care and services.	0.21	0.02	0.13	0.03	0.02	0.41	−0.08	0.00	0.32
Relationship between Family Caregivers and their Families mean±SD=40.7±21.4
Relatives are divided over the place of care and death of the patient.	−0.06	−0.07	0.08	0.04	0.04	0.02	0.93	−0.01	0.90
Family caregivers struggle with the burden of many opinions from relatives.	0.16	0.02	−0.11	−0.03	−0.04	0.00	0.76	−0.01	0.59
Family caregivers cannot consult with relatives about patient's recuperation.	−0.04	0.16	0.25	−0.02	0.04	−0.03	0.45	−0.03	0.38
Funeral Preparations mean±SD=50.0±28.0
Family caregivers do not know how to arrange the funeral after the patient's death.	−0.06	0.05	0.04	−0.04	0.03	0.00	−0.04	0.91	0.82
Family caregivers have trouble making funeral arrangements because preparation time is limited.	0.09	−0.04	0.05	0.03	−0.02	0.03	0.02	0.80	0.73

The lest-square method, promax rotation

The cumulative rate of contribution: 71.8%

SD, standard deviation.

Table 3.

Reliability of Family's Difficulty Scale in End-of-Life Home Care

	Cronbach α coefficients	Test–Retest ICC
Burden of Care	0.73	0.85
Concerns about Home Care Doctor	0.74	0.83
Balance of Work and Care	0.74	0.81
Patient's Pain and Condition	0.75	0.85
Concerns about Visiting Nurse	0.75	0.82
Concerns about Home Care Service	0.73	0.75
Relationship between Family
Caregivers and their Families	0.75	0.78
Preparation of Funeral	0.74	0.80

ICC, intraclass correlation coefficients.

The results of the convergent and divergent validity are shown in Table 4. The convergent validity and the divergent validity for each domain ranged from 0.42 to 0.91 and from 0.11 to 0.53, respectively. We had a 100% success rate in the development of the scale for each domain by Multitrait Scaling analysis (Table 4). The concurrent validity of five factors is shown in Table 5. There was a significant positive correlation between three domains of the FDS. Values ranged from 0.50 to 0.65. There was a significant negative correlation between the other two domains of the FDS. Values ranged −0.60 and −0.51, respectively.

Table 4.

Convergent and Divergent Validity of Family's Difficulty Scale in End-of-Life Home Care

Subscale	Number of items	Convergent validity ^a (range of correlation)	Divergent validity^b (range of correlation)	Success of scale (%)
Burden of Care	6	0.66–0.78	0.44–0.53	42/42 (100%)
Concerns about Home Care Doctor	5	0.73–0.83	0.10–0.52	35/35 (100%)
Balance of Work and Care	3	0.84–0.91	0.13–0.51	21/21 (100%)
Patient's Pain and Condition	4	0.58–0.80	0.11–0.51	28/28 (100%)
Concerns about Visiting Nurse	3	0.81–0.90	0.09–0.52	21/21 (100%)
Concerns about Home Care Service	3	0.42–0.77	0.13–0.44	21/21 (100%)
Relationship between Family Caregivers and their Far	3	0.47–0.70	0.12–0.38	21/21 (100%)
Preparation of Funeral	2	0.71	0.21–0.34	14/14 (100%)

The correlation between each item score and the associated domain score.

The correlation between each item score and domain scores not associated with the item.

Number of items for which convergent correlation is stronger than divergent correlation/the total number of correlations.

Table 5.

Concurrent Validity of Family's Difficulty Scale in End-of-Life Home Care

	CRA-J			CES
FDS-HC	Impact on schedule	Impact on health	Lack of family support	Physical care by Physician	Physical care by nurse
Burden of Care	0.65	0.60	0.31	—	—
Balance of Work and Care	0.62	0.53	0.34	—	—
Relationship between Family Caregivers and their far	0.39	0.40	0.50	—	—
Concerns about Home Care Doctor	—	—	—	−0.60	−0.29
Concerns about Visiting Nurse	—	—	—	−0.26	−0.51

Pearson's product-moment correlation coefficient.

CRA-J, Caregiver Reaction Assessment—Japanese version; CES, Care Evaluation Score—Short version.

As a result of test-retest reliability, the intraclass correlation coefficients for each domain ranged from 0.75 to 0.85 (Table 3).

Discussion

We developed a new measure to assess the family's difficulties, which we believe is a useful tool to clarify the family's difficulties in caring for patients with cancer at the end of life at home. The strength of this scale is that family difficulties can be comprehensively assessed throughout the entire course of caregiving at home and through to the patient's death. In addition, validity and reliability were verified, and the success of scale development was confirmed. This scale also emphasized that some items of the FDS concern qualitative issues such as family caregivers sacrifice and difficulties in communicating with home care doctor. So we believe that this tool will be strengthened by follow-up study of such topics using a qualitative methodology.

With regards to the quality of end-of-life care, the important things are sufficient explanations to families, consideration of the caregiver burden, and the relationship with the doctor and nurse.¹⁹ Our scale also includes items containing this content. During the end of life, patients with cancer often have distressing symptoms such as fatigue, pain, and dyspnea and these symptoms are often uncontrolled, therefore patients suffer.²⁰ Thus, the Patient's Pain and Condition is a specific domain that is particularly relevant for the family of the patient with terminal cancer.

Factorial validity was confirmed by factor analysis. We confirmed eight domains and our assumed domains have an interrelationship with each other.

This study also confirmed the concurrent validity. Correlation between five factors and concurrent measures, CRA-J and CES, were examined. We also confirmed the success of scale development by convergent and the divergent validity. The reliability of the FDS was identified by good intraclass correlation coefficients.

This study had several limitations. First, the facilities were limited to one area in Japan. Therefore, the results might not have been representative of all the potential samples to be investigated. Second, the participants of this study were bereaved family caregivers and there was some extensive variation in length of time from 6 months to 2 ½ years postbereavement, so the responses might have recall bias. Therefore the results should be interpreted with caution. On the other hand, the FDS has the advantage of evaluation of the total process of home care including the patient's death. Further study is needed to assess this scale for current family caregivers and revisions of the FDS may be needed so that we can assess the current family caregiver's difficulties. We also need to investigate methods and interventions that provide relief for the family and help them to cope with the difficulties.

Conclusion

We developed the FDS to measure the families' difficulties in caring for patients with cancer at the end of life at home. The FDS consists of eight factors and the validity and reliability were confirmed. Home care providers can use this tool to measure the family's needs and difficulties as they care for their loved ones at home.

Footnotes

Acknowledgments

We thank all participants and staffs of the participating facilities. This study was supported by the foundation of Yuumi Memorial Fundation for Home Health Care.

Author Disclosure Statement

No competing financial interests exist.

References

Minister's Secretariat, Ministry of Health, Labour and Welfare. Report on politics regarding cancer [in Japanese] www.mhlw.go.jp/seisaku/24.html. 2011 July 6.

Minister's Secretariat, Ministry of Health, Labour and Welfare. Report on politics regarding cancer [in Japanese] www.mhlw.go.jp/shingi/2007/06/s0615-1.html. 2011 July 6.

Minister's Secretariat, Ministry of Health, Labour and Welfare. Report on concepts of hospitals specialized cancer [in Japanese] www.mhlw.go/shingi/2005/07/s0713-6.html. 2011 July 6.

Sanjo

, Miyashita

, Morita

, Hirai

, Kawa

, Akechi

, Uchitomi

. Preferences regarding end-of-life cancer care and associations with good-death concepts: A population-based survey in Japan. Ann Oncol, 2007; 18:1539–1547.

Choi

, Miyashita

, Hirai

, Sato

, Morita

, Tsuneto

, Shima

. Preference of place for end-of-life cancer care and death among bereaved Japanese families who experienced home hospice care and death of a loved one. Support Care Cancer, 2010; 18:1445–1453.

Minister's Secretariat, Ministry of Health, Labour and Welfare. Vital statistics. 2009. www.e-stat.go.jp/SG1/estat/List.do?lid=000001032157. 2011 July 6.

Brink

, Smith

. Determinants of home death in palliative home care: Using the interRAI palliative care to assess end-of-life care. Am J Hosp Palliat Med, 2008; 25:263–270.

Bakitas

, Ahles

, Skalla

, Brokaw

, Byock

, Hanscom

, Lyons

, Hegel

. Proxy perspectives regarding end-of-life care for person with cancer. Cancer, 2008; 112:1854–1861.

Gomes

, Higginson

. Factors influencing death at home in terminally ill patients with cancer: Systematic review. BMJ, 2006; 332:515–521.

10.

McLaughlin

, Sullivan

, Hasson

. Hospice at home service: The carer's perspective. Support Care Cancer, 2007; 15:163–170.

11.

Grov

, Fossa

, Tonnessen

, Dahl

. The caregiver reaction assessment: psychometrics, and temporal stability in primary caregivers of Norwegian cancer patients in late palliative phase. Psychooncology, 2006; 15:517–527.

12.

Miyashita

, Sakai

, Iizuka

, Machida

, Nakamura

, Yokoi

, Araya

, Shiigai

, Tomura

. Burdens to family members in home care and related QOL factors [in Japanese] J Jpn Assoc Rural Med, 2006; 54:767–773.

13.

Hiramatu

, Kondo

, Umehara

, Kuze

, Higuchi

. Burdens of family members and related factors in home care (The first report) J Health Welfare Stat, 2006; 53:19–24.

14.

Hiramatu

, Kondo

, Umehara

, Kuze

, Higuchi

. Burdens of family members and related factors in home care (The second report) J Health Welfare Stat, 2006; 53:8–13.

15.

Krippendorff

, Mikami

, Shiino

, Hashimoto

. Content analysis: An introduction to its methodology [in Japanese] Keisou Shobou, 2006.

16.

Given

, Given

, Stommel

, Collins

, King

, Franklin

. The caregiver reaction assessment (CRA) for caregivers to persons with chronic physical and mental impairments. Res Nurs Health, 1992; 15:271–283.

17.

Misawa

, Miyashita

, Kawa

, Abe

, Nakayama

, Given

. Validity and reliability of the Japanese version of the Caregiver Reaction Assessment Scale (CRA-J) for community-dwelling cancer patients. Am J Hosp Palliat Care, 2009; 26:334–340.

18.

Morita

, Chihara

, Kashiwagi

. A scale to measure satisfaction of bereaved family receiving inpatient palliative care. Palliat Med, 2002; 16:141–150.

19.

Morita

, Hirai

, Sakaguchi

, Maeyama

, Tsuneto

, Shima

. Measuring the quality of structure and process in end-of-life care from the bereaves family perspective. J Pain Symptom Manage, 2004; 27:492–4501.

20.

Ardith

, Charles

, Barbara

, Verbitsky

. Symptom experience in the last year of life among individuals with cancer. J Pain Symptom Manage, 2006; 32:403–412.