Support Care Cancer

DOI 10.1007/s00520-013-2077-z

ORIGINAL ARTICLE

Modifiable factors associated with caregiver burden

among family caregivers of terminally ill Korean
cancer patients
Seok-Joon Yoon & Jong-Sung Kim & Jin-Gyu Jung &
Sung-Soo Kim & Samyong Kim

Received: 1 July 2013 / Accepted: 25 November 2013

# Springer-Verlag Berlin Heidelberg 2014

Abstract
Purpose Higher caregiver burden is associated with poor
quality of life among family caregivers. However, in Korea,
very few studies have examined factors associated with care-
giver burden. The present study investigated factors associat-
ed with caregiver burden among family caregivers of termi-
nally ill Korean cancer patients, particularly modifiable factors
as a potential target of intervention strategies.
Methods A cross-sectional study using self-administered
questionnaires was performed. Sixty-four family caregivers
of terminally ill cancer patients who were admitted to the
hospice-palliative care unit of a university hospital in South
Korea were included. To identify caregiver burden, the
Caregiver Reaction Assessment scale (CRA) was used in this
study. Time spent in providing care per day, number of visits
per week from other family members, family functioning, and
a positive subscale, self-esteem, of the CRA were deemed as
modifiable factors. Other sociodemographic, caregiving char-
acteristics of the subjects were non-modifiable factors.
S.<J. Yoon : J.<S. Kim : J.<G. Jung : S.<S. Kim
Department of Family Medicine, Research Institute for Medical
Sciences, Chungnam National University School of Medicine,
Daejeon, South Korea
S. Kim
Department of Internal Medicine, Cancer Research Institute,
Chungnam National University School of Medicine, Daejeon, South
Korea
S.<J. Yoon : J.<S. Kim : J.<G. Jung : S.<S. Kim
Department of Family Medicine, Chungnam National University
Hospital, 282 Munhwa-roJung-gu Daejeon, South Korea
S. Kim (*)
Division of Hemato-Oncology, Department of Internal Medicine,
Chungnam National University Hospital, 282 Munhwa-roJung-gu
Daejeon, South Korea
e-mail: frkim@cnu.ac.kr
Results Longer time spent providing care per day, fewer
weekly visits from other family members, poor family func-
tioning, and low self-esteem were considered as modifiable
factors associated with caregiver burden. Low monthly in-
come and the spouse being the family caregiver were non-
modifiable factors.
Conclusions Our study has practical significance in that it
identifies modifiable factors that can be used to devise inter-
vention strategies. Developing and applying such intervention
strategies for alleviating the factors associated with high care-
giver burden could be important for improving the quality of
life of both patients and their families.
Keywords Caregivers . Quality of life . Cancer . Palliative
care
Introduction
The World Health Organization (WHO) has defined the pur-
pose of palliative care as improving the quality of life (QOL)
of both patients and their families [1]. The WHO suggests that
in order to improve the QOL of terminally ill patients, im-
proving the QOL of patients' families is crucial.
Taking care of a terminally ill family member places sig-
nificant physical (e.g., coronary heart diseases) and mental
(e.g., depression) burdens on the caregiver(s), which, in turn,
leads to decreased QOL within the family [25]. Financial
burden is one of the most important predictors of depression
and has the strongest relationship with poor QOL [6, 7]. Many
studies have examined factors related to caregiver burden
among family members of cancer patients, and the following
factors related to high caregiver burden were found: young
age of the caregiver [8, 9], limitations imposed on daily
activities [8], caregiver depression [10, 11], lack of other
family members to share caregiving [9], poor relationships

with other family members [12], lack of communication be-
tween caregivers and patients [13], unmet psychosocial needs
among patients [11, 14], poor financial status [9, 15], and
being a female caregiver [9].
In Korea, like China or Japan, taking care of ill family
members is a duty prescribed by filial piety from Confucian
culture. A literature review noted that Korean caregivers re-
ported filial obligation, while Caucasian caregivers reported
filial affection as a major caregiving motivation [16]. Koreans
usually sought help from other family members, while
Caucasian Americans used formal support from government
or other social services [16]. Thus, the factors related to
caregiver burden among Korean caregivers may be different
from those of Western caregivers. However, few studies have
examined these factors among family caregivers of terminally
ill Korean cancer patients [9, 1719]. We argue that the
previous studies have some limitations. The previous studies
did not include detailed data associated with caregiving such
as duration and type of care or directly measure caregiver
burden.
Developing and applying intervention strategies to reduce
caregiver burden could be important as well as finding the
factors related to caregiver burden. A meta-analysis conducted
to determine the effect of interventions offered to family
caregivers of cancer patients showed that these interventions
significantly reduced caregiver burden and increased their
self-efficacy [20]. But a high number of sessions and duration
of interventions could limit their practical applicability in a
palliative care setting, especially in Korea because of late
referral to palliative care. It is reported that family meeting
as a single session therapy could be an alternative and was
useful and effective in the palliative care setting [21, 22].
Modifiable factors associated with caregiver burden could be
important topics for discussion in the family meetings to
reduce caregiver burden in palliative care settings in Korea.
The objective of the present study was to identify factors
associated with high caregiver burden among family care-
givers of terminally ill Korean cancer patients. We placed a
special emphasis on modifiable factors related to caregiver
burden as a potential target of intervention strategies.
Subjects
Subjects were family caregivers of terminally ill cancer pa-
tients who were admitted to the hospice-palliative care unit of
Chungnam National University Hospital in South Korea from
April 2012 through September 2012. Family caregivers who
were providing informal care to patients were included in the
study if they were aged 18 years or older, assessed to be
mentally and physically capable of filling out questionnaires
as determined by subjective observation of medical staff
members, and agreed to fill out the questionnaires. A total of
Support Care Cancer
68 family caregivers agreed to participate, with 64 family
caregivers forming our final sample (4 caregivers were ex-
cluded because of missing data).
Research methods
Family caregivers filled out self-administered questionnaires
on the day the patients were admitted to the aforementioned
hospice-palliative care unit. Age, sex, educational status, mar-
ital status, religion, and monthly income were assessed as
sociodemographic variables. Relationship with the patient,
job loss due to caregiving, diagnosis of any diseases after
the onset of caregiving, self-reported overall health status,
time spent in providing care per day, care duration, number
of visits per week from other family members, number of
activities of daily living (ADLs) in which patients need help,
and family functioning were assessed as variables associated
with caregiving. These variables were adopted from other
studies [814, 1719]. Self-reported overall health status
was examined using one item from the Korean version of
the Short Form-36 [23]. The number of ADLs in which the
patients need help was determined using seven items from the
Korean Activities of Daily Living scale [24]. Family function-
ing was assessed using the family adaptation, partnership,
growth, affection, and resolve (APGAR) score [25].
Data on age, sex, performance status, and primary sites of
the patients' cancer were obtained by reviewing medical re-
cords. Patients' performance status was determined using the
Eastern Cooperative Oncology Group performance status
(ECOG PS) [26]. An ECOG PS score of 1 or 2 was defined
as good performance status, while scores of 3 or 4 were
defined as poor performance status.
To identify caregiver burden, the Caregiver Reaction
Assessment (CRA) was used in the present study. The CRA
is a self-administered questionnaire that consists of 24 items,
developed by Given et al. in 1992 [27]. It contains five
subscales: four subscales assess negative aspects of caregiv-
ing, and one subscale assesses the positive aspects of caregiv-
ing. The advantage of the CRA is that it enables the investi-
gation of various aspects of caregiving by using a single
questionnaire [28]. Seventeen items comprise the four nega-
tive subscalesdisrupted schedules, lack of family support,
health problems, and financial problems. The positive sub-
scale comprises seven items assessing care-derived self-
esteem. Each item was rated on a 5-point Likert scale (from
strongly disagree to strongly agree); thus, scores on each
subscale ranged from 0.0 to 5.0 points. A total score is not
recommended to identify caregiver burden; we used scores for
each subscale instead [27]. Higher scores on the four negative
subscales and lower scores on the positive subscale indicate
higher burden. In the present study, we used the Korean
version of the CRA (CRA-K); the CRA was translated in a
Support Care Cancer

previous study and the reliability and validity of the translated status. The patients' most prevalent primary cancer site was
version were confirmed [29]. Psychometric properties (QOL, the stomach (21.9 %; Table 1).
anxiety, depression, social support) were confirmed in the
previous studies [29, 30]. Cronbach's alphas for each subscale
of the CRA-K, disrupted schedules, lack of family support, Table 1 General characteristics of family caregivers and patients (N =64)
health problems, financial problems, and self-esteem were
Variables MeanSD or N (%)
0.75, 0.80, 0.73, 0.71, and 0.52, respectively [29]. In our
study, Cronbach's alphas were 0.695, 0.853, 0.817, 0.824, Age (years) 47.9113.43
and 0.625, respectively. Sex (male) 21 (32.8)
Time spent in providing care per day, number of visits per Relationship with patientsa (spouses) 23 (35.9)
week from other family members, family functioning, and a Education (12 years) 35 (54.7)
positive subscale, self-esteem, of the CRA were deemed as Marital status (married) 49 (76.6)
modifiable factors in the opinion of the clinicians. Religion (yes) 37 (57.8)
Monthly income (KRW) (2,000,000) 32 (50.0)
Job loss due to caregiving (yes) 24 (37.5)
Statistical analysis Diagnosis of disease after the onset of 8 (12.5)
caregiving (yes)
Self-reported overall health status
Relationships between the sociodemographic variables of
Very poor 5 (7.8)
family caregivers, variables associated with caregiving, pa-
Poor 9 (14.1)
tients' variables, and caregiver burden were analyzed using
As usual 25 (39.1)
independent samples t tests, one-way ANOVAs, and
Good 21 (32.8)
Pearson's correlations at the univariate level. After that, we
Very good 4 (6.3)
extracted variables associated with caregiver burden (four
Care time (h/day)b 14.939.18
negative subscales of the CRA). We conducted four multiple
Care duration (months)c 13.8019.90
regression models to determine the factors associated with
Number of visits from other family members 3.802.52
caregiver burden. Significant variables (p <0.05) obtained
(number/week)
from the univariate analyses were used in each regression Number of ADLs in which patients need help 4.172.61
model. We conducted a stepwise method of multiple regres- Family functioningd (good) 38 (59.4)
sion analysis to remove unrelated factors. All statistical anal- Age of patients (years) 64.6312.02
yses were performed using SPSS version 13.0 for Windows Sex of patients (male) 31 (48.4)
(SPSS, Inc., Chicago, IL, USA). We set the statistical signif- Performance status of patientse (poor) 45 (70.3)
icance level of univariate analyses at p <0.05. We used Patients' primary cancer sites
Bonferroni correction to the four multiple regression analyses Stomach 14 (21.9)
and set the statistical significance level at p < 0.0125. Colon/rectum 12 (18.8)
Descriptive statistics of our variables were expressed as Lung 7 (10.9)
meanstandard deviation.
Hepatobiliary 7 (10.9)
Pancreas 4 (6.3)
Breast 3 (4.7)
Results Genitourinary 9 (14.1)
Brain 3 (4.7)
General characteristics of family caregivers and patients Others 5 (7.8)

KRW 2,000,000 is approximately equivalent to USD 1,900

On average, family caregivers were 47.9113.43 years old,
ADL activities of daily living
and patients were 64.6312.02 years old. Twenty-one of the a
family caregivers (32.8 %) and 31 of the patients (48.4 %) Non-spousal caregivers were children, parents, and daughters-in-law of
the patients
were male. Twenty-three of the family caregivers were b
Care time means time spent providing care per day
spouses (35.9 %). Thirty-two of the family caregivers c
Care duration means duration providing care since diagnosis
(50.0 %) had a monthly income of KRW 2,000,000 (approx- d
Evaluated by family APGAR (adaptation, partnership, growth, affec-
imately equivalent to USD 1,900) or less. Average time spent tion, resolve) score
providing care per day was 14.939.18 h. The number of e
Eastern Cooperative Oncology Group performance status score of 1 or 2
visits per week from other family members was 3.802.52. was defined as good performance status, while scores of 3 or 4 were
Forty-five of the patients (70.3 %) had a poor performance defined as poor performance status
 Support Care Cancer

Univariate analyses of relationships
between sociodemographic, caregiving characteristics
of family caregivers and caregiver burden
Factors related with higher burden of disrupted schedules
were being an unmarried caregiver, low monthly income, job
loss due to caregiving, longer time spent in providing care per
day, and a greater number of ADLs in which patients need
help. Factors associated with higher burden of lack of family
support were low monthly income, longer care duration,
fewer visits from other family members, moderate or poor
family functioning, and low self-esteem. Factors related to
more health problems were old age, the spouse being the
caregiver, low monthly income, longer time spent providing

Table 2 Univariate analyses of

relationships between Variables MeanSD or correlation coefficients
sociodemographic, caregiving
characteristics of family care- DS LF HP FP
givers and caregiver burden
Age 0.002 0.044 0.286 0.159
Sex
Male 3.420.80 2.261.06 2.640.74 2.920.89
Female 3.620.74 2.100.85 3.030.87 2.991.04
Relationship with patients
Others 3.460.83 1.990.91 2.680.77 2.700.90
Spouses 3.730.61 2.430.87 3.290.85 3.450.97
Education
12 years 3.690.67 2.310.74 3.080.79 3.300.96
>12 years 3.390.85 1.951.07 2.690.87 2.570.90
Marital status
Married 3.420.72 2.090.75 2.910.86 2.960.99
Unmarried 3.990.78 2.361.33 2.880.84 3.001.02
Religion
Yes 3.650.74 2.140.89 3.060.95 3.120.99
No 3.430.79 2.160.97 2.690.63 2.770.97
Monthly income (KRW)
2,000,000 3.840.62 2.380.81 3.270.85 3.510.94
>2,000,000 3.280.80 1.930.97 2.540.68 2.430.71
Job loss due to caregiving
Yes 4.030.65 2.400.99 3.090.93 3.220.87
No 3.280.69 2.000.85 2.790.79 2.821.03
Diagnosis of disease after the onset of caregiving
Yes 4.000.88 2.531.04 3.661.22 3.131.10
No 3.490.73 2.100.89 2.790.73 2.950.98
KRW 2,000,000 is approximately Self-reported overall health status
equivalent to USD 1,900 Very poor 3.440.93 2.120.86 2.751.51 2.531.07
DS disrupted schedule, LF lack Poor 3.780.87 2.600.95 3.441.24 3.260.86
of family support, HP health As usual 3.620.58 2.220.71 3.050.63 3.280.94b
problem, FP financial problem,
SE self-esteem, ADL activities of Good 3.330.85 1.881.04 2.600.54 2.790.95
daily living Very good 3.950.93 2.151.33 2.560.94 1.830.79b

p <0.05 obtained by indepen- Care time 0.525 0.169 0.389 0.443

dent samples t tests or Pearson's Care duration 0.150 0.291 0.237 0.258
correlation analyses
Number of visits from other family members 0.186 0.455 0.081 0.231
a
Evaluated by family APGAR Number of ADLs in which patients need help 0.282 0.061 0.207 0.002
(adaptation, partnership, growth,
affection, resolve) score Family functioninga
b
There was a statistical signifi- Poor, moderate 3.780.76 2.620.96 3.080.83 3.420.88
cant difference between two cate- Good 3.410.74 1.830.74 2.780.85 2.660.95
gories by post-hoc test of one- Self-esteem 0.224 0.451 0.362 0.268
way ANOVA
Support Care Cancer

care per day, and low self-esteem. The spouse being the
caregiver, low educational status, low monthly income, usu-
al self-reported overall health status, longer time spent pro-
viding care per day, longer care duration, moderate or poor
family functioning, and low self-esteem were factors associ-
ated with more financial problems (Table 2).
Univariate analyses of relationships between characteristics
of patients and caregiver burden
Among patients' characteristics, poor performance status was
related with higher burden of disrupted schedule, and male
patients were associated with more health problems of care-
giving (Table 3).
family members (R 2 change = 0.207), low self-esteem
(R 2 change=0.167), and poor family functioning (R 2
change=0.075) were factors related with higher burden of
lack of family support. Factors associated with more health
problems were low monthly income (R 2 change=0.187), low
self-esteem (R 2 change=0.086), and longer time spent in
providing care per day (R 2 change=0.082). Low monthly
income (R 2 change=0.304), poor family functioning (R 2
change=0.063), and the spouse being the caregiver (R 2
change=0.043) were related with more financial problems
(Table 4).
Discussion

Multivariate analyses of relationships between multiple The present study identified modifiable factors associated
variables and caregiver burden with caregiver burden among family caregivers of terminally
ill Korean cancer patients; the findings have practical signif-
Factors associated with higher burden of disrupted icance as they may be helpful in tailoring intervention strate-
schedule were longer time spent in providing care per gies to suit the needs of the patients and their caregivers. Our
day (R 2 change=0.276) and poor performance status of results suggest the followings as possible intervention strate-
the patients (R 2 change=0.103). Fewer visits from other gies for lessening caregiver burden: decreasing time spent in
providing care per day, encouraging other family members to
visit, enhancing family functioning (e.g., resolving conflicts
Table 3 Univariate analyses of relationships between characteristics of and encouraging communication among family members),
patients and caregiver burden and increasing self-esteem.
Variables MeanSD or correlation coefficients The aforementioned family meeting would be an interven-
tion to decrease caregiver burden. A study investigated the effect
DS LF HP FP of the family meeting in the palliative care setting and noted that
symptom management, discharge planning, and update on the
Age 0.148 0.054 0.013 0.115
patient's medical condition were key topics discussed [22]. Most
Sex
(90 %) family caregivers reported that the meeting was benefi-
Male 3.580.74 2.120.79 3.120.81 3.091.12
cial because of its informative perspective, while 40 % noted
Female 3.530.80 2.181.03 2.700.84 2.860.85
that the family-focused (e.g., talking about feelings/concerns)
Performance statusa aspect was beneficial [22]. According to our results, discussing
Poor 3.720.69 2.250.97 3.020.79 2.960.95 the factors related to caregiver burden and solutions to them
Good 3.160.81 1.920.74 2.630.93 3.001.11 (e.g., sharing time spent in providing care per day and more
Primary cancer sites visits from other family members) in the family meeting would
Stomach 3.360.90 2.210.73 3.000.69 3.001.05 be helpful in lessening caregiver burden. Further studies on
Colon/rectum 3.670.64 2.250.97 3.100.88 3.701.13 family meeting are needed for the palliative care settings in
Lung 3.690.81 2.000.95 2.711.08 2.810.66 Korea in order to identify the efficacy of such intervention.
Hepatobiliary 3.310.88 2.061.39 2.541.02 2.760.81 In the present study, longer time spent in providing care per
Pancreas 3.350.75 2.600.59 3.250.79 2.751.26 day was the most important factor associated with higher
Breast 3.270.57 2.471.29 3.000.90 2.700.92 burden of disrupted schedules and was associated with more
Genitourinary 3.930.75 2.110.97 2.911.01 3.110.51 health problems. Taking care of patients all day long hinders
Brain 3.470.70 1.800.72 2.920.52 3.110.51 the caregivers from accessing their personal resources avail-
Others 3.720.78 1.800.91 2.550.69 2.330.62 able for coping with stress (e.g., hobbies, relationships with
friends). This finding is consistent with that of a previous
DS disrupted schedule, LF lack of family support, HP health problem,
FP financial problem, SE self-esteem study [8] where limitations in daily activities were related to
a
Eastern Cooperative Oncology Group performance status score of 1 or 2
caregiver burden. Other studies [18, 19] showing that longer
was defined as good performance status, while scores of 3 or 4 were time spent in providing care per day is associated with depres-
defined as poor performance status sion and lower QOL are also consistent with our results. When

p <0.05 obtained by independent samples t tests the main caregiver's burden is deemed heavy as judged by
 Support Care Cancer

Table 4 Multivariate analyses of relationships between multiple variables and caregiver burden

Variables DS LF HP FP

R 2 change R 2 change R 2 change R 2 change

Caregivers' characteristics
Age 0 0
Spousea 0 0 0.483 0.043
More educated 0 0
Unmarried 0 0
Higher monthly income 0 0 0 0 0.269 0.187 0.733 0.304
Job loss 0 0
Self-reported health status 0 0
Care time 0.043 0.276 0.034 0.082 0 0
Care duration 0 0 0 0
More visits of other families 0.131 0.207
More needed ADLs by patients 0 0
Good family function 0.528 0.075 0.617 0.063
Self-esteem 0.566 0.167 0.574 0.086 0 0
Patients' characteristics
Male 0 0
Poor performance status 0.534 0.103
Total R 2 0.379 0.449 0.354 0.410

A dash means that this variable was not included in the statistical model because of non-significance on univariate analyses. A 0 means that this
variable was included in the statistical model but was not significant on multivariate analysis. All R 2 means adjusted R 2
DS disrupted schedule, LF lack of family support, HP health problem, FP financial problem, ADL activities of daily living
a
Relationship among the patients, children, parents, and daughters-in-law was compared together with spouses

p <0.0125 was obtained by stepwise multiple regression analyses

staff members in weekly team meetings, paid caregivers are
sometimes provided to patients of hospice-palliative care units
in Korea free of charge funded by voluntary donations.
Volunteers' physical contacts with the patients are usually
allowed in Korea. So volunteers' support could be used to
decrease caregiver burden. These could be an intervention to
decrease time spent to providing care. Further studies regard-
ing such interventions would be helpful in Korea.
Fewer visits from other family members and poor family
functioning were associated with higher burden of lack of
family support. Poor family functioning was also related to
more financial problems. These results are consistent with
those of previous studies [12, 19] showing that having poor
relationships with other family members is associated with
greater caregiver burden and that the prevalence of depression
increases as visits from other family members decrease. Some
studies argue that less visits from other family members have a
negative effect on caregivers' ability to vent about their hard-
ships and are related to less financial support from other
family members. We suggest that providing educational pro-
grams giving information about the difficulties and possible
burnout of main caregivers would be helpful in alleviating the
main caregivers' burden.
In the present study, low self-esteem was associated with
higher burden of lack of family support and more health
problems. This result is consistent with results obtained from
a previous study [29]. This result is also consistent with a
previous finding [9] that positive aspects of caregiving are
negatively associated with caregiver burden. Items assessing
positive aspects of caregiving (e.g., Care-giving makes me
feel good) were included in the positive subscale of the CRA.
We assume that family caregivers gladly take care of their ill
family member(s), even though caregiving is very taxing. Part
of this could be due to an ethical desire to fulfill duties related
to filial piety in the Korean culture [16]. We suggest that some
kind of educational program emphasizing meanings and pos-
itive aspects of caregiving would be helpful in lessening
caregiver burden.
Using our results, preventative interventions to reduce
caregiver burden could be possible at the early days of admis-
sion to the hospice-palliative care unit. We suggest that these
programs must include difficulties facing primary caregivers,
the importance of assistance from other family members,
positive aspects of caregiving, loneliness and meaninglessness
of patients, and techniques to care for patients. Further studies
regarding these interventions will be needed.
Support Care Cancer
The present study has some limitations that need to be
addressed in the future. First, it is difficult to generalize our
results given our small caregiver sample. Additionally, we
employed a cross-sectional design. It is thus difficult to deter-
mine causal relationships between multiple variables and
caregiver burden without examining these relationships in
longitudinal research (as well as how modifiable factors can
be improved upon through interventions). Regarding job loss
due to caregiving, we did not consider shortening of work
hours or switching from a full-time job to a part-time job.
These could also cause financial distress due to a curtailment
of income. Cronbach's alpha for the self-esteem scale was low
in a previous study [29]. Although the value of our study was
higher than that of the previous study, it could be a possible
error related to the low reliability of the scale. Finally, psy-
chological variables, such as depression and anxiety, were not
evaluated in the present study. In our results, adjusted R 2
of each multivariate analysis were not particularly high
(0.350.45). Thus, there may be other factors related to care-
giver burden. Future studies should include such variables in
order to have a better understanding of the full range of factors
related to caregiver burden.
Despite these limitations, the current study identified that
time spent providing care per day, number of visits from other
family members, family functioning, and self-esteem were
modifiable factors associated with caregiver burden among
family caregivers of terminally ill Korean cancer patients.
Monthly income and the spouse being the primary caregiver
were non-modifiable factors. Developing and applying inter-
vention strategies to help address the modifiable factors relat-
ed to caregiver burden could be important for improving the
QOL of both patients and their families.
Acknowledgments The nursing staff, social worker, spiritual care
team, volunteers, and the greater number of cancer patients are
acknowledged.
Conflict of interest The authors declare that there is no conflict of
interest.
References
1. World Health Organization (2011) WHO definition of palliative care.
http://www.who.int/cancer/palliative/en. Accessed 9 Oct 2012
2. Lee S, Colditz GA, Berkman LF et al (2003) Caregiving and risk of
coronary heart disease in U.S. women: a prospective study. Am J
Prev Med 24(2):113119
3. Braun M, Mikulincer M, Rydall A et al (2007) Hidden morbidity in
cancer: spouse caregivers. J Clin Oncol 25(30):48294834
4. Song JI, Shin DW, Choi JY et al (2011) Quality of life and mental
health in family caregivers of patients with terminal cancer. Support
Care Cancer 19(1):15191526
5. Dumont S, Turgeon J, Allard P et al (2006) Caring for a loved one
with advanced cancer: determinants of psychological distress in
family caregivers. J Palliat Med 9(4):912921
6. Rhee YS, Yun YH, Park S et al (2008) Depression in family care-
givers of cancer patients: the feeling of burden as a predictor of
depression. J Clin Oncol 26(36):58905895
7. Yun YH, Rhee YS, Kang IO et al (2005) Economic burdens and
quality of life of family caregivers of cancer patients. Oncology 68(2
3):107114
8. Goldstein NE, Concato J, Fried TR et al (2004) Factors associated
with caregiver burden among caregivers of terminally ill patients with
cancer. J Palliat Care 20(1):3843
9. Park CH, Shin DW, Choi JY et al (2012) Determinants of the burden
and positivity of family caregivers of terminally ill cancer patients in
Korea. Psychooncology 21(3):282290
10. Grov EK, Fossa SD, Sorebo O et al (2006) Primary caregivers of
cancer patients in the palliative phase: a path analysis of variables
influencing their burden. Soc Sci Med 63(9):24292439
11. Hwang SS, Chang VT, Alejandro Y et al (2003) Caregiver unmet
needs, burden, and satisfaction in symptomatic advanced cancer
patients at a Veterans Affairs (VA) medical center. Palliat Support
Care 1(4):319329
12. Francis LE, Worthington J, Kypriotakis G et al (2010) Relationship
quality and burden among caregivers for late-stage cancer patients.
Support Care Cancer 18(11):14291436
13. Fried TR, Bradley EH, O'Leary JR et al (2005) Unmet desire for
caregiverpatient communication and increased caregiver burden. J
Am Geriatr Soc 53(1):5965
14. Sharpe L, Butow P, Smith C et al (2005) The relationship
between available support, unmet needs and caregiver burden
in patients with advanced cancer and their carers. Psychooncology
14(2):102114
15. Yusuf AJ, Adamu A, Nuhu FT (2011) Caregiver burden among poor
caregivers of patients with cancer in an urban African setting.
Psychooncology 20(8):902905
16. Kong EH (2007) The influence of culture on the experiences of
Korean, Korean American, and Caucasian American family care-
givers of frail older adults: a literature review. J Korean Acad Nurs
37(2):213220
17. Yun YH, Rhee YS, Nam SY et al (2005) The socioeconomic impact
of terminal cancer on patients' families and its associated factors. J
Korean Acad Fam Med 26(1):3139
18. Jung JG, Kim JS, Kim SS et al (2006) Quality of life among family
caregivers of terminal cancer patient. Korean J Hosp Palliat Care
9(1):110
19. Han KH, Jung JG, Oh SK et al (2005) Depression level among family
caregivers of terminal cancer patients. J Korean Acad Fam Med
26(12):752758
20. Northouse LL, Katapodi MC, Song L et al (2010) Interventions with
family caregivers of cancer patients: meta-analysis of randomized
trials. CA Cancer J Clin 60(5):317339
21. Hudson P, Quinn K, O'Hanlon B et al (2008) Family meetings in
palliative care: multidisciplinary clinical practice guidelines. BMC
Palliat Care 7:12
22. Hudson P, Thomas T, Quinn K et al (2009) Family meetings in
palliative care: are they effective? Palliat Med 23(2):150157
23. Koh SB, Sj C, Kang M et al (1997) Reliability and validity on
measurement instrument for health status assessment in occupational
workers. J Prev Med Public Health 30(2):251266
24. Won CW, Yang KY, Rho YG et al (2002) The development of
Korean activities of daily living (K-ADL) and Korean instrumental
activities of daily living (K-IADL) scale. J Korean Geriatr Soc 6(2):
107120
25. Smilkstrein G (1978) The family APGAR: a proposal for a
family function test and its use by physician. J Fam Pract 6(6):
12311239
26. Oken MM, Creech RH, Tormey DC et al (1982) Toxicity and
response criteria of the Eastern Cooperative Oncology Group. Am J
Clin Oncol 5(6):649655

27. Given CW, Given B, Stommel M et al (1992) The caregiver
reaction assessment (CRA) for caregivers to persons with
chronic physical and mental impairments. Res Nurs Health 15(4):
271283
28. Hudson PL, Trauer T, Graham S et al (2010) A systematic review of
instruments related to family caregivers of palliative care patients.
Palliat Med 24(7):656668
Support Care Cancer
29. Rhee YS, Han IY, Lee IJ et al (2009) Cross cultural validation of the
caregiver reaction assessment (CRA) for cancer patient's caregiver.
Korean J Health Promot Dis Prev 9(3):189198
30. Grov EK, Fossa SD, Tonnessen A et al (2006) The caregiver reaction
assessment: psychometrics, and temporal stability in primary care-
givers of Norwegian cancer patients in late palliative phase.
Psychooncology 15(6):517527