Keywords
cancer - caregiver burden - hope - oncology - quality of life - shame
Dr Mehdi Reza Sarafraz
Introduction
Cancer is a chronic disease that results in considerable, permanent, and variable
needs and problems for patients. Because the unit of care in cancer care is the patient
and family,[1] investigation of the mutual effects of patient and caregiver relationship is of
great significance in the process of sickness, therapy, and recovery.
Care is something that is neither predictable nor selected by people. When someone
is confronted with cancer diagnosis and therapy, his/her family members feel a high
responsibility toward caregiving.[2] Moreover, they are committed to offer infinite care and support.[3] However, family members accept this responsibility with low or no training and limited
resources.[4] Caregiver burden is defined as the cognitive evaluation of the multidimensional
response to demands and their consequences within the context of the evolving caregiving
experience.[5] Regarding cancer, the following variables are shown to be effective in the responsibility
of caregiver: caregiver age and sex, relation to care receiver, and duration of care
preparation.[6] Caregiver burden includes physical burden,[7] psychological burden,[8] social burden,[8] and financial burden.[9] Physical burden such as sleep disturbance, fatigue, and pain are often experienced
by caregivers.[7] Providing emotional support to patients and helping cancer patients to deal with
their feelings about cancer are considered as psychological commitments.[8] Lack of employment due to caregiving activities and taking care of others besides
cancer patients are known as social responsibilities.[10] Financial burden might include high medical expenses, loss of income, and savings.[8] In an extensive review of literature, Schulz et al reviewed 41 studies and reported
high physical and psychological distemper of caregivers compared with normal populations
or control group.[11] In fact, most of the psychological distress levels reported by caregivers are comparable
or even more than those experienced by care receivers.[12]
This study tries to identify whether the caregiver’s distress affect the patient?
if yes, how? It is assumed that caregiver burden affects patients’ hope and QoL. Furthermore,
it is predicted that patients can keep and raise their hope to control their environment,
be less dependent on the caregiver, and recover faster. Recent studies have also shown
that quality of life (QOL) can offer distinct prognostic information as a predictor
for the recovery duration of different types of cancers.[13]
As a dynamic and multidimensional structure, hope is affected by various factors,
and it is defined as the likelihood of better future than a hard and uncertain present.[14] Identifying factors that influence the hope of newly diagnosed cancer patients is
of great importance as hopelessness is a risk factor that may lead to suicide,[15] depression, and tendency to early death[16] in cancer patients. Energy and pain levels are among the physical features effective
in hope. Energy is considered one of the features of hope, and it is discovered that
low energy has a significant relation to hope.[17] The involvement of pain in the daily life of cancer patients has an inverse relation
to their level of hope.[18] It is also found that psychological factors such as depression and anxiety are related
to hope.[19] An integrative literature review conducted by Lin and Bauer-Wu[20] identified that living with hope and goal is the main element of psychospiritual
well-being in terminal patients. Moreover, with this enhanced sense of psychospiritual
well-being, patients can face a terminal disease more effectively.
The life quality of cancer survivors shows their therapy pattern and psychophysical
functions that may be used to discover patients’ subgroups in need of more surveillance
and directive perspectives for patient-based interventions after the completion of
cancer therapy.[21] Although most cancer survivors cannot retain their physical, psychological, and
social functions as before cancer therapy, some specific subgroups of them are at
risk of decreased QOL even after 5 years of early diagnosis.[22]
It also seems that caregivers’ negative evaluation, whether real or as perceived by
patients, may lead to the feeling of shame in the very same patients. Hence, in individuals
suffering from cancer who have low ability to control life events, behavioral compensation
is not possible, negative evaluation of caregiver is predictable, and feeling shame
is inevitable. Thus, studying the effect of the shame felt by the patient on hope
and QOL elements is another goal of this research.
The feeling of shame has been studied in psychological theories and has been implicated
in lots of psychopathological conditions. Shame is associated with a general punitive
judgment of the self, which results in an intense emotional response and a tendency
to withdraw and hide.[23] Shame is also defined as a self-conscious negative emotion about self or personal
self-blame.[24]
Conclusion
This study is a novel and important step to exploring the role of shame and hope as
cardinal features in the cancer patients’ experiences and presenting a composite picture
of their quality of life affected by caregiver burden.
Methods
This research was conducted with patient–caregiver dyads to identify the different
views of cancer patients and their caregivers. The patients and caregivers were allowed
to complete the research form in multiple settings as needed and to ask assistance
from a friend or an intimate person to answer the questionnaire. The patients’ form
and caregivers’ form needed 15 minute and 5 minutes to be completed, respectively.
They included valid survey tools and demographic information. The medical reports
of patients were not evaluated. In fact, information reported by the individuals themselves
was used. The inclusion criteria for patients were (1) older than 18 years, (2) being
diagnosed with cancer, (3) being under therapy or follow-up at present, and (4) being
in an appropriate physical and mental state to fill the questionnaire. A caregiver
was defined as a family member or a close relative who had the highest responsibility
to take care of the cancer patient and spend most of his or her time with the patient.
Caregivers younger than 18 years or those with severe cognitive disorder were disregarded
from the research.
The following four valid questionnaires were used in this study.
Zarit Burden Interview
Zarit[25] was the first author who offered an operational definition of caregiver burden.
He designed a tool called Zarit Burden Interview (ZBI) for evaluating perceived caregiver
burden. It consists of 22 questions that are rated on a scale from 0 to 4 based on
the presence or severity of positive response. It measures caregiver health, psychological
well-being, social life, financial status, and patient–caregiver relationship. The
ZBI is translated into many languages, and its internal consistency is reported between
0.85 and 0.94.[26]
Guilt and Shame Proneness Scale
The Guilt and Shame Proneness Scale[27] is a 16-item, 7-point scale (1 = very unlikely and 7 = very likely), which measures
individuals’ variance in the tendency to experience guilt and shame. It consists of
four subscales of guilt-negative behavior evaluation, guilt–repair, shame–negative
self-evaluation (NSE), and shame–withdraw. The α coefficients tend to show lower reliability
in scenario-based measures because each item consists of a unique variance for the
scenario[28]; thus, the reliability was in the range of 0.61 to 0.69.[27] In this study, shame subscales were only used and guilt subscales were removed as
they were not related to the study purpose.
Quality of Life-C 30
QOL[29] is a questionnaire designed by the European Organization for Research and Treatment
of Cancer to measure the physical, mental, and social functions of cancer patients.
It consists of five functional domains, three sign scales, a general health domain,
and six individual items. The internal reliability of the questionnaire obtained by
Cronbach’s α is in the range of 0.56 to 0.85 and 0.84 in Cankurtaran et al’s research[30] and Tan et al’s research,[31] respectively.
Herth Hope Index
The Herth Hope Index[32] is a 12-item, 4-point Likert scale (1 = completely disagree and 4 = completely agree)
that was designed to evaluate the rate of hope in adults based on clinical cases.
It suggests three factors of hope namely (a) temporality and future, (b) positive
readiness and expectancy, and (c) interconnectedness. The total rate is in the range
of 12 to 84. Higher rates imply higher levels of hope. The α coefficient was 0.97
with a 2-week test–retest reliability of 0.91.
Results
Patients and caregivers are considered as a dyad when the two parties are willing
to participate in the study and only after they have completed the consent form consciously
and separately. Patients and their caregivers were evaluated by a trained interviewer
in the hospital. The goal and research process were explained by the interviewer.
Of 176 dyads requested to participate, 141 dyads accepted (participation rate: 80.11%).
If there was any nonreplied item in the questionnaire, those dyads were omitted from
the analysis (n = 41). As a result, 100 dyads were included in the final analysis. For this study,
patients were selected from Namazi and Shahid Mohammadi Hospital in Shiraz and Bandar
Abbas, respectively.
The patient characteristics are summarized in [Table 1]. The current study was conducted on 46 men and 53 women (1 missing), with the diagnosis
of cancer. In the “severity” section, much data were not completed that may be due
to the low education level of the patients and their lack of awareness of their status.
In addition, the percentage of cancer type among participants is summarized in [Table 1]. According to [Table 1], leukemia and liver cancer had the most and the least plentitude among the study
individuals, respectively.
Table 1
Patient characteristics (n = 100)
|
Characteristic
|
Value
|
|
Abbreviation: SD, standard deviation.
|
|
Age (y), mean ± SD (range)
|
45.1 ± 16.3 (21–80)
|
|
Sex, n (%)
|
|
|
Male
|
46 (46)
|
|
Female
|
53 (53)
|
|
Marital status, n (%)
|
|
|
Married/living with partner
|
66 (66)
|
|
Single/separated/divorced/widowed
|
34 (34)
|
|
Education level, n (%)
|
|
|
Primary education
|
34 (34)
|
|
High school
|
31 (31)
|
|
University
|
24 (24)
|
|
Other/unknown/uneducated
|
11 (11)
|
|
Tumor classification, n (%)
|
|
|
Leukemia
|
47 (47)
|
|
Prostate
|
5 (5)
|
|
Breast
|
19 (19)
|
|
Liver
|
1 (1)
|
|
All others
|
28 (28)
|
|
Severity, n (%)
|
|
|
Low
|
35 (35)
|
|
Mediate
|
26 (26)
|
|
High
|
32 (32)
|
|
Unknown
|
7 (7)
|
|
Time since diagnosis, years (range)
|
1.1 (0.08–11)
|
|
Signed informed consent for participation, n (%)
|
|
|
Yes
|
89 (89)
|
|
No
|
11 (11)
|
The mean, standard deviation, and correlation of research variables are summarized
in [Table 2]. According to [Table 2], there is a significant correlation among exogenous, endogenous, and mediator variables,
which made analysis completion possible. The direct and indirect effects are summarized
in [Table 3]. [Fig. 1] shows the relationships between the variables and the coefficients of the fitted
model. The fitted indices obtained by route analysis employing maximum likelihood
method showed the appropriate fitness of the pattern with data (Goodness-of-Fit Index
(GFI) = 0.99, Comparative Fit Index (CFI) = 0.99, Adjusted Goodness-of-Fit Index (AGFI)
= 0.94, Incremental Fit Index (IFI) = 0.99, Root Mean Squared Error of Approximation
(RMSEA) = 0.03, and Normed Chi-Square (CMIN/DF) = 1.07).
Table 2
Mean, standard deviation, and correlation matrix
|
Study Variables
|
Mean ± SD
|
1
|
2
|
3
|
4
|
5
|
|
Abbreviations: QOL, quality of life; shame–NSE, shame–negative self-evaluation; SD,
standard deviation.
a
p < 0.05.
b
p < 0.01.
|
|
1. Shame–NSE
|
20.89±5.54
|
1
|
|
|
|
|
|
2. Shame–withdraw
|
14.64 ± 5.17
|
0.255a
|
1
|
|
|
|
|
3. Hope
|
37.72 ± 5.77
|
0.223a
|
–0.201a
|
1
|
|
|
|
4. Caregiver burden
|
33.72 ± 9.60
|
–0.112
|
0.177
|
–0.287b
|
1
|
|
|
5. QOL
|
–45.07 ± 16.13
|
0.205a
|
–0.100
|
0.349b
|
–0.188
|
1
|
Table 3
Direct and indirect effects of variables
|
Path
|
Direct effect
|
Indirect effect
|
Total effect
|
|
Abbreviations: QOL, quality of life; shame–NSE, shame–negative self-evaluation.
p<0.05.a
p<0.01.b
|
|
Caregiver burden
|
|
|
|
|
Shame–NSE
|
–0.11
|
–
|
–0.11
|
|
Shame–withdraw
|
0.21
|
–0.3
|
0.18
|
|
Hope
|
–0.22a
|
–0.07a
|
0.29b–
|
|
QOL
|
–
|
–0.10b
|
0.10b–
|
|
NSE
|
|
|
|
|
Shame–withdraw
|
0.28b
|
–
|
0.28b
|
|
Hope
|
0.26a
|
–0.06a
|
0.20
|
|
QOL
|
–
|
0.07
|
0.07
|
|
Shame–withdraw
|
|
|
|
|
Hope
|
–0.23a
|
–
|
–0.23a
|
|
QOL
|
–
|
–0.08a
|
–0.08a
|
|
Hope
|
|
|
|
|
QOL
|
0.35b
|
–
|
0.35b
|
Fig. 1 Structural equation model to assess the predictive manner of quality of life based
on caregiver burden with the mediatory role of hope and shame. The one-headed arrow
indicates a directional relationship between two variables. The number next to the
arrow represents the relationship (β). NSE, negative self-evaluation; QoL, quality
of life.
Caregiver burden negatively correlated with hope (r =–0.28, p < 0.01) ([Table 2]). Moreover, in the model, hope (β =–0.22) and QOL (β =–0.10) both formed strong
relations with caregiver burden ([Table 3]
[Fig. 1]).
Hope and shame–NSE subscale positively correlated with each other (r = 0.22, p < 0.05) ([Table 2]). Furthermore, hope negatively correlated with shame–withdraw subscale (r =–0.20, p < 0.05) ([Table 2]). QOL was found to be statistically correlated (r = 0.34, p < 0.01) ([Table 2]) and strongly related with hope (β = 0.35) ([Table 3]
[Fig. 1]).
Factors that positively correlated with shame–NSE subscale were QOL (r = 0.20, p < 0.05) and shame–withdraw subscale (r = 0.25, p < 0.05) ([Table 2]). The model suggests a positive regression between shame–NSE and shame–withdraw
subscales (β = 0.28) ([Table 3]
[Fig. 1]).
Descriptive statistics were implemented to assess demographic information and clinical
characteristics. Data analysis was performed by using IBM Corp. Released 2015. IBM
SPSS Statistics for Windows, Version 23.0. Armonk, NY, USA: IBM Corp. SPSS Statistics
23.0 for Windows and Amos Graphics Arbuckle, J. L. (2014). Amos (Version 23.0) [Computer
program]. Chicago, USA: IBM Corp. 23.0 for Windows. To evaluate the validity of the
research tool, Cronbach’s α was used that was estimated to be over 0.70 for all questionnaires.
Before parametric statistical analysis, the data were analyzed with regard to consistency
to substructural premises of this type of analysis. These premises are as follows:
analyzing missed data, evaluating the normality of variables’ data distribution, estimating
outlier values’ data, and the linearity of variables’ relation.[33] For measuring missed values and their probable pattern, missing value analysis algorithm
was used. No missing values were found in any of the measures.
For the evaluation of variables’ data normality, Kolmogorov–Smirnov and Shapiro–Wilk
tests were used besides histogram diagram. According to the results, only the data
of QOL and shame–withdraw variables followed normal distribution. For the variables
that did not follow a normal distribution, data were transformed by Rankit formula[34] and logarithm.[33] After that, the distribution of hope and caregiver burden grades was normalized
by using Box plot.[33] The outlier data recommended by Munro et al[35] were corrected regarding the closest data to the outliers. The last step in data
evaluation was investigating the premises of linear relation among the research variables.
The presence of a linear relation among the variables was approved by a Scatter plot.
Discussion
The current study was performed to investigate the predictive model of QOL and hope
based on caregiver burden and shame perceived by the patients. The suggested model
was approved, and the results showed a negative and direct relationship between caregiver
burden and hope. They also showed a positive and direct relationship between hope
and QOL. The research hypothesis implied that caregiver burden predicts hope in patients
with cancer and hope predicts their QOL. Shame was also significantly related to hope
and QOL. Caregivers of cancer patients report that they spend more time in taking
care of the patients and pay more efforts during a shorter period of time, besides
they have to tolerate a higher financial burden compared with other caregivers. As
a result, the more the patient percepts this burden, the less he/she will recover
and the patient’s hope will diminish faster. Caregiver characteristics that may lead
to increase in burden include lower age,[36] female gender,[36] lower level of education,[37] a mixed relation,[38] and higher levels of anxiety and depression.[39] Consequently, it is predicted that patients who are served by such caregivers have
lower hope.
Caregiver burden has a negative and indirect effect on QOL with the mediatory role
of hope. QOL includes physical, mental, and social elements of patients. Caregiver
burden reduces the general health of patients. Recent studies have shown that QOL
can bring a distinctive prognostic information to predict the recovery period of various
cancers. Hence, as the caregiver burden increases, recovery is delayed.
It seems that cancer patients’ hope depends on some factors, both physical (e.g.,
pain and energy level) and mental (anxiety and depression). Cancer patients suffer
from physical and mental disturbances. They feel loss of goal, pride, and self-glory.
In fact, the hope in the case of cancer care is connected to the hope of treatment.
As expected, the high level of hope results in better life quality. Our findings support
the outcome of van der Biessen et al.[40]
Shame is known as a disturbing feeling toward personal faults. From another perspective,
shame is a self-conscious emotion that is aroused through self-assessment and leads
to self-regulation. Shame–NSE subscale items describe bad feelings about oneself.
Patients with lower degree in NSE are more likely to have personal disturbance and
lower self-respect and self-compassion. They are also more likely to ruminate when
they are sad. Consequently, they begin to show depressive symptoms such as disappointment
and lower mental and physical performance. Shame–withdraw subscale items describe
action tendencies focused on hiding or withdrawing from public. This factor has a
direct negative effect on hope and an indirect negative effect on QOL. It is likely
that loosing social support due to isolation leads to these negative impacts.
This survey has some limitations. First, convenient sampling restricts the generalization
of findings. Second, even though caregiver- and cancer patient-related elements were
discovered and implemented in the analyses of the relationship between caregiver burden,
hope, and QOL, other dimensions of caregiving such as caregiver challenges, caregiver
readiness, family function, and social support that might have had a role in the burden
were not studied. Seven caregivers of this study were cancer patients. They were included
to provide a general overview on caregiver’s burden, but it is probable that their
burden and QOL are different from that of other caregivers in this study. It is likely
that cancer patients are sometimes their own caregivers, so their unique needs require
more scrutiny. It is more helpful to perform studies on a larger population and in
different centers to get better insights that support our findings.
Our study showed that cancer patients’ hope has a significant relation to their QOL.
In addition, we found a negative relationship between caregiver burden, hope, and
QOL. This finding should be included in the interventions to offer support and notification.
In addition, based on the results, interventions must also be caregiver based so as
to help a faster recovery by considering caregiver’s needs.
