Abstract
Background:
Multi-systemic symptoms of varying severity in myotonic dystrophy type 1 (DM1) may pose difficulties in caregiving. However, the factors which affect their care burden are yet to be sufficiently understood.
Objective:
We investigated care burden and its correlates among caregivers of patients with DM1.
Methods:
General demographic information was obtained from patients with DM1, as well as Barthel index (ADL), body mass index, and genetic information. Patients completed SF-36v2 (health-related quality of life), CES-D (depressive symptoms), and ESS (daytime sleepiness) questionnaires. Caregivers reported their perception of patient’s status through these questionnaires, and completed Zarit Caregiver Burden Interview (ZBI). Correlation analysis of these variables were performed, and regression analysis was utilized to assess the relationship between caregiver burden and other variables.
Results:
Forty-three patient-caregiver dyads participated. Mean ZBI score was 20.7±17.4, and 32.6% reported a significant care burden. ZBI correlated with caregiver-reported CES-D, but not with patient-reported CES-D. Both patient-reported and caregiver-reported physical QoL of patients correlated with patient ADL. Multiple regression analysis revealed that the combination of caregiver-reported CES-D, caregiver-reported mental QoL, and genetic characteristics predicted caregiver burden.
Conclusions:
Caregiver burden was felt although patients were relatively well-functioning. Patients’ and caregivers’ assessment of patients’ physical condition were similar. However, they did not agree on the evaluation of the patients’ psychological state. Cognitive characteristic of the patients and the caregivers’ perception of the patients’ state may have affected the results. Future DM1 care strategies need to work on improvement of patient-caregiver communication and provide support for the caregiver’s psychological health.
Keywords
ABBREVIATIONS
Activities of daily living
Barthel index
Body mass index
The Center for Epidemiologic Studies depression scale
Cytosine-thymine-guanine
Myotonic dystrophy type 1
Epworth sleepiness scale
Health-related quality of life
Quality of life
Mental component summary
National Center of Neurology and Psychiatry
Physical component summary
Short-Form Health Survey, version 2
Zarit Caregiver Burden Interview
INTRODUCTION
Myotonic dystrophy type 1 (DM1) is a multi-systemic disease characterized not only by muscular symptoms, such as myotonia and muscle atrophy, but also by endocrine and metabolic disorders, cardiac disorder, cataract, cognitive dysfunction, and respiratory problems [1]. It is caused by the mutation in myotonic dystrophy protein kinase gene on chromosome 19, where an abnormal increase in the number of cytosine-thymine-guanine (CTG) repeats is observed in the 3’ untranslated region [2]. It is one of the most common muscular diseases in adults, with a prevalence of approximately 5.5 per 100,000 people [3], with no fundamental cure to date. Due to this symptomatic complexity, comprehensive care that involves medical, as well as social and psychological care, becomes important for maintaining and improving a patient’s quality of life (QoL), and the matter of how care is best provided for each patient is similarly critical. Presence of multiple symptoms means that the challenges of living with patients with DM1 vary between individuals, with some needing long hospital stays while others live at home with less severe conditions. Naturally, what is needed to care for these patients differs between individuals accordingly, so guidance for carers about symptom management should also be tailored to each patient’s needs. Paying attention to a carer’s experienced burden from caregiving is therefore as crucial as examining the problems faced by the patient, in order for medical staff to provide relevant disease management guidance. For this reason, recognizing the nature and associated factors of the DM1 caregiver burden is essential, yet only few reports on this issue currently exist [4–8].
The concept of caregiver burden was first defined by Zarit et al. as the degree of damage the caregiver suffers in terms of emotional, physical health, social life, and economic conditions as a result of caring for relatives [9, 10]. One of the first pieces of research on this topic was by Grad et al., which reported that care burdens affected the family members of patients with mental illness in the United Kingdom both psychologically and physically [11]. Since the 1980 s, the acceleration of aging and the accompanying increase in the size of the elderly population in need of long-term care in Europe and the United States have drawn more attention to caregiver burden. From then, a number of studies have been published, mainly in the fields of psychology, gerontology, psychiatry, and nursing. Diseases that are researched most frequently in such studies are cerebrovascular disorder and dementia [12–17], although studies of the caregiving burden of other illnesses, such as spinal cord injury, orthopedic disease, and cerebral palsy, can also be found in this pool of research. As of today, there is some research on carer burden and its correlates in neuromuscular disease in general that includes participants with DM1, together with those with other illnesses such as Duchenne muscular dystrophy and Becker muscular dystrophy [4, 7]. Furthermore, though not directly measuring the concept of caregiver burden, there are also studies which focus on the psychological and relational impact of DM1 on partners and next of kin who live with DM1 patient [5, 8]. These studies draw attention to the patients’ lack of initiative and its negative impact on partners’ physical health, mental health and marital satisfaction [5, 8]. Moreover, relatives of patients too, tend to have high anxiety, distress and lower psychological QoL compared to physical QoL [6].
While proximal muscles are mainly affected in general muscular disorders, in DM1, symptoms present more often in facial, sternomastoid, and distal limb muscles [1]. Muscular stiffness, or myotonia, is another distinct characteristic of this disease, as well as the presence of dysfunction in multiple organ systems, including cognitive and psychological symptoms [1, 18]. Moreover, the symptoms that manifest varies from patient to patient, with no uniform expression in severity or prognosis. Because of this wide clinical spectrum, DM1 is seen as one of the most complex genetic disorders, which makes the standardization of care a demanding task [19]. Although a small population size poses a challenge to the study of disease-specific caregiver burden with respect to rare diseases, caregiving experience associated with distinctive characteristics of DM1 needs to be well understood in order to provide effective support to the caregivers. In this study, we aimed to clarify the factors related to caregiver burden in DM1 by collecting relevant data from both patients and their caregivers. Findings will contribute to shedding light on the type of potential support that is necessary for patients and help elucidate ways to reduce caregiver burden and promote carers’ physical and mental health.
MATERIALS AND METHODS
Participants
Patients with DM1 and their carers were recruited during outpatient visits to the National Hospital Organization, the Aomori Hospital, or the hospital of the National Center of Neurology and Psychiatry (NCNP) between October 2016 and November 2017. Participants were also recruited from websites and newsletters of patient organizations. The inclusion criteria were as follow: 1) patient has been diagnosed genetically; 2) patient is between 18 and 69 years of age; and 3) patient is autonomously mobile, with or without walking assistance. Patients were excluded as candidates if they did not meet the inclusion criteria.
Ethical considerations
Explanations of the research purpose and methods, participation, freedom of withdrawal, and protection of privacy were given to patients and their caregivers on the first visit, and consent was obtained from the participants. Approval was obtained from the ethics review committee of the NCNP and the Aomori Hospital (approval numbers: A2016-050 and 29-19, respectively). The study was conducted in accordance with the Declaration of Helsinki.
Survey items and questionnaires
Participating patients provided basic information and completed the MOS 36-Item Short-Form Health Survey (version 2; SF-36v2) for the assessment of health-related quality of life (HRQoL) [20–22], the Center for Epidemiologic Studies Depression Scale (CES-D) for measuring depression [23, 24], and the Epworth Sleepiness Scale (ESS) for assessing excessive daytime sleepiness [25, 26]. Conversely, caregivers were asked to complete the SF-36v2, CES-D, and ESS by reporting patient status as if they were reporting it from the patient’s perspective. Additionally, carers completed the Zarit Caregiver Burden Interview (ZBI) to report their caregiver burden [27]. Japanese versions with confirmed reliability and validity were used for all questionnaires [21, 27].
Demographic information
Information about patient’s age, sex, number of CTG repeats, and BMI were collected from the patients. Information about patient ADL, as measured by the Barthel Index (BI), was also collected [28]. BI is a measurement that assesses functional ability by measuring levels of independence in daily activities. These are: feeding, bathing, grooming, dressing, bowels and bladder control, toilet use, transfer ability, mobility, and ability to use stairs. Scores range from 0 to 100, with higher scores indicating more independence. Additionally, caregiver sex and relationship to the patient were provided by patients and caregivers or were collected from their medical records if they were patients of the Aomori Hospital or the NCNP.
Zarit Caregiver Burden Interview (ZBI)
ZBI consists of 22 items rated on a 5-point scale [9, 27]: questions 1 to 21 pertain to the care burden felt by the carer in various situations, and the final item asks about perceived difficulty of care overall. Scores range between 0 and 88, and a higher score signifies a “heavier” experienced burden. Although a universal cut-off score does not exist, a study which include the data of multiple diseases and of Japanese population reports the cut-off of 24 [29]. With reference to this study, we will define the score of 24 or more as a presence of significant care burden.
Short-Form Health Survey—version 2 (SF-36v2)
SF-36v2 is a widely used HRQoL evaluation scale that consists of 36 items divided into eight domains [20–22]; physical functioning, role limitations due to physical functioning, bodily pain, general health, vitality, social functioning, role limitations due to emotional function, and mental health. Moreover, the items can be summarized into two summary scores: a physical component summary (PCS) and a mental component summary (MCS). Scores can be compared to normative data for the Japanese population (mean = 50, standard deviation = 10) with the use of standardized norm-based scoring. Higher scores signify better perceived QoL.
Center for Epidemiologic Studies Depression Scale (CES-D)
The CES-D is a 20-item questionnaire that measures symptoms of depression experienced during the past week. The frequencies of symptoms related to depression are rated on a 5-point scale. A higher total score indicates stronger symptoms, with scores on or above the cut-off (16) implying the possibility of depression.
Epworth Sleepiness Scale (ESS)
The ESS is a self-report questionnaire that measures the degree of excessive daytime sleepiness. It questions respondent’s frequency of dozing off in particular situations, with scores ranging from 0 to 24. A higher score signifies a higher degree of daytime sleepiness. Generally, scores that exceed 10 are considered pathological [30]. Although there is a discussion that questions the reliability of this scale to be used on DM1 [31], we have chosen ESS as it is one of the most widely used scale to measure daytime sleepiness, and enables comparison with past DM1 studies as well as other diseases that share this symptom.
Data analysis
To understand the characteristics of patients with DM1 and their carers, descriptive statistics were calculated for demographic and clinical attributes and ZBI, SF-36v2, CES-D, and ESS scores. Correlation analysis was performed to investigate the relationship between caregiver burden and other variables, using the Spearman’s rank correlation coefficient test. Correction for multiple testing was performed to adjust p values by using Benjamini-Hochberg method. Furthermore, multiple linear regression analysis was performed to investigate the predictors of caregiver burden. Logarithmic transformation was performed on the ZBI scores to have them conform to normality. The order of the variables in the regression model was determined using Akaike’s Information Criterion (AIC) [32]. These analyses were performed using R software version 3.5.1 for Windows, with p values of less than 0.05 considered significant.
RESULTS
Characteristics of participants and caregivers
Forty-three patients with DM1 (21 men, 22 women; mean age±standard deviation: 44.7±10.8) participated in the study. Patient characteristics are shown in Table 1. The mean score of the BI was 89.0±10.6 and the mean number of CTG repeats was 950.8±439.6. Mean BMI was 23.7±4.2. The mean PCS score (27.3±15.4) was lower than the Japanese average, while the mean MCS score (52.0±9.5) exceeded the national average. The mean CES-D score was 15.0±8.7, which was below the cut-off of 16. The mean score of the ESS was 9.0±5.2. The details of the characteristics of this patient population have been reported previously in a study by Endo et al. [33].
Characteristics of participants
Notes: Data are presented as mean ± SD. aSF-36v2, CES-D and ESS in the ‘Caregivers’ column were reported by caregivers on the state of the patient to whom they provide care. CTG, cytosine-thymine-guanine.
The forty-three caregivers consisted of 20 men and 23 women. Caregivers’ relationships to the patients included parent (21), spouse (18), child (2) and other (2). The mean score on the ZBI, which measures caregiver burden, was 20.7±17.4, with the lowest score being 1 and the highest being 79. 32.6% of the caregivers reported ZBI score of 24 or more. Caregiver-reported SF-36v2 questionnaires on patient HRQoL had mean scores of 23.2±15.3 for PCS and 51.6±11.1 for MCS. These mean scores indicated a trend similar to that of the patient-completed responses, with lower than average physical QoL and higher than average mental QoL. Mean scores on the CES-D and the ESS were 14.1±10.1 and 10.3±5.9, respectively. The score difference between patients’ and caregivers’ assessment on the patients’ condition was the largest at 6.6 for Social Functioning subscale of SF-36v2.
Strength of correlation among caregiver burden and other factors
Table 2 shows the correlation between ZBI scores and other characteristics of patients and caregivers. A positive correlation with caregiver-reported patient CES-D scores (ρ= 0.60, p < 0.001) was observed. No significant association was found between ZBI scores and any other factors. In Table 3, intercorrelation among genetic and ADL information, patient-reported scores and caregiver-reported scores of the questionnaires are shown. Both patient-reported and caregiver-reported PCS of patient correlated with Barthel index score (ρ= 0.53, p < 0.01 and ρ= 0.46, p < 0.01, respectively). PCS reported by patients and caregiver also had a significant correlation with each other (ρ= 0.60, p < 0.01). On the contrary, patient-reported and caregiver-reported MCS, CES-D and ESS were not significantly correlated. The number of CTG repeats associated with neither patient-reported nor caregiver-reported HRQoL, CES-D and ESS. Patient-reported CES-D had significant correlations with other patient-reported items, such as with PCS (ρ= –0.51, p < 0.01), MCS (ρ= –0.49, p < 0.01), and with ESS (ρ= 0.46, p < 0.05). The same applies for caregiver-reported CES-D and other caregiver-reported items, as significant associations were observed with PCS (ρ= –0.37, p < 0.05), MCS (ρ= –0.44, p < 0.01), and with ESS (ρ= 0.46, p < 0.05). However, patient-reported CES-D did not correlate with caregiver-reported PCS, MCS and ESS, and vice versa for caregiver-reported CES-D.
Correlation between caregiver burden and other variables
Notes: *p < 0.001. aRelationship to patient was coded as younger generation (child) = 0, same generation (spouse, siblings) = 1, older generation (parent) = 2. bPatients’ state was reported by caregivers in these questionnaires. BMI, body mass index; CES-D, Center for Epidemiologic Studies depression scale; CTG, cytosine-thymine-guanine; ESS, Epworth sleepiness scale; MCS, mental component summary; PCS, physical component summary; ZBI, Zarit Caregiver Burden Interview.
Intercorrelation between patient-reported and caregiver-reported SF-36v2, CES-D and ESS
Notes: Data are presented as Spearmans’ ρ. *p < 0.05, **p < 0.01. CES-D, Center for Epidemiologic Studies depression scale; ESS, Epworth sleepiness scale; MCS, mental component summary; PCS, physical component summary.
Associated factors of ZBI scores
The results of the multiple linear regression analysis are shown in Table 4. The combination of caregiver-reported patients’ depressive symptom (CES-D score), caregiver-reported psychological QoL of patient (MCS score), and number of CTG repeats produced a moderately significant model for predicting ZBI score (R2 = 0.44, p < 0.001). However, caregiver-reported MCS and CTG repeats alone did not produce a significant correlation with caregiver burden.
Multiple regression model for caregiver burden
Notes: CES-D, Center for Epidemiologic Studies depression scale; CTG, cytosine-thymine-guanine; MCS, mental component summary; ZBI, Zarit Caregiver Burden Interview.
DISCUSSION
The mean score for caregiver burden among the study participants was 20.7. In previous studies, mean scores were 28.3 for cerebrovascular disorder [14], 38.03 for dementia, and 23.4 for neuromuscular disease [4, 17]. These results demonstrate that the care burden felt by the caregivers in our study is close to that observed in carers of individuals suffering from neuromuscular disease. Although diverse symptom variety exists in neuromuscular disorders, muscular symptoms such as weakness and atrophy that are common across different diseases may have a similar impact on how care burdens are felt. Mean care burden score in this study was less than the cut-off of 24, but there still were 14 caregivers (32.6%) who scored 24 or more. This suggests a significant level of burden. Yet at the same time, the participants of this study were relatively well-functioning with mean Barthel index score at 89.0, and majority were maintaining the ability to walk at the time of the visit. This means that carers of DM1 patients with more severe symptoms may feel heavier burden. Additionally, it is recognized that being female in itself is generally regarded as a risk factor for a higher perceived care burden [34]. The fact that our study included carers of both sexes almost evenly (female: 53%, male: 47%) may have reduced the ZBI score compared to other studies which included higher percentage of women in their samples [7, 35]. A higher number of female caregiver participants may have resulted in higher felt burden. These results indicate the importance of providing support for caregivers to reduce care burden in DM1.
Correlation analysis showed that ZBI score had a significant positive association with caregiver-reported CES-D score of patient’s depressive symptom. Furthermore, the multiple regression analysis revealed that caregiver-reported CES-D score, caregiver-reported MSC score, and CTG repeat number were factors in the best-fitting model for predicting ZBI score. Conversely, other characteristics such as patient age, sex, ADL, BMI, caregiver sex, and caregiver relationship to patient did not show significant associations with care burden. With regards to Barthel index score, previous research has produced mixed results. As did this study, some prior research argues for a lack of relationship between ADL ability and care burden [7, 35]. However, others reported a significance of this relationship [12, 36–38]. The same reason as described above may apply to our result. The relative functional independence of our participants may be a factor behind why the caregiver did not feel the burden in helping the patients with daily activities. The positive correlation observed between Barthel index and patients’ physical QOL (PCS reported by both patients and caregivers) supports this explanation, as patients with better ADL level had higher physical QoL in the eyes of both the patients themselves and the caregivers. Therefore, if a patient’s symptoms progressed to further limit ADL abilities, a caregiver may experience heavier burden. Hence, the necessity of providing support for caregiving in order to help the caregiver accommodate the increased workload cannot be ruled out. While this study examined general ADL ability using the total BI score, it will also be possible in the future to investigate the impact of individual ADL items on caregiving burden. This will help in providing support tailored to each caregiver’s needs.
With respect to CES-D scores, care burden was greater for caregivers who noticed more symptoms of depression in patients they cared for. On the other hand, patient-reported depressive symptoms did not have a significant correlation with caregiver burden. Moreover, unlike the scores for patient’s physical QoL wherein associations were found between patients’ and caregivers’ reports, patients and caregivers did not have a significant correlation on patient’s psychological QoL. One explanation for this result may be related to disease awareness of DM1 patients. It is recognized that DM1 affects the patients’ perception of symptoms through its impact on cognitive abilities. One study has found a tendency of patients to understate their difficulties in psycho-social aspects of their daily life [39]. It has also shown that scores on the patients’ social function had the largest mismatch between the patients’ and caregivers’ view. If we assume caregivers’ report as an accurate observation, it can be said that our study participants are restricted in disease awareness, especially when self-assessment of mental and social aspects are concerned.
Conversely, it is also possible to make an assumption that the caregivers’ view on patient’s symptoms and conditions are to a certain extent, affected by particular factors, as patients’ self-reported CES-D was in its own terms associated with self-reported SF-36v2 (better QOL relates to lower CES-D) and self-reported ESS (less sleep problem relates to lower CES-D). There have been different opinions about the cause of depressive symptoms in DM1 [40–42]. While expression of apathy as reduced motivation and interest, which is common in DM1, may appear as symptoms of depression [40, 41], physical symptoms of depression such as disordered sleep and fatigue overlap with DM1 symptom characteristics [42]. The caregivers’ ratings for patient’s CES-D may have been done keeping in mind the patients’ apathy, sleep disorder and fatigue, rather than at depression as a psychological expression. Indeed, a significant correlation was observed between caregiver-reported CES-D and ESS (ρ= 0.46, p < 0.05). If this was the case, caregivers who reported higher care burden may have found these symptoms of the patients as being difficult to face. However, this interpretation needs to be treated with care as we do not have the data for apathy and fatigue of the patients. Another possible factor that can influence how caregivers view patient’s condition could be the caregiver’s psychological state. Previous studies have found associations between the mental status of caregivers, such as the presence of depression or anxiety, and caregiving burden [12, 44]. Anxiety, distress, low marital satisfaction and low psychological QoL are also common in partners and relatives of DM1 patients [5, 8]. It is understood that caregiver ratings or their perceptions of patients’ physical and mental health status are associated with the caregivers’ perceived burden [35, 45]. As caregiver-reported patient’s depressive symptoms had a significant correlation with care burden in our study participants (who were all either next to kin or partner of patient and not occupational caregiver) while patient-reported CES-D did not, there may be a possibility that caregivers’ psychological condition reflected on their view of the patient. Nevertheless, this discussion remains speculative as data on the caregivers’ psychological status is not available in this study. In the future, we should compare caregiver’s mental health with caregiver’s perception on patient’s symptoms in order to further the understanding of this effect.
As for the number of CTG repeats and the caregiver-reported MSC of the patients, these variables alone did not have a significant correlation with caregiving burden. However, a multiple regression analysis showed that caregiver-reported CES-D scores, caregiver-reported MSC scores, and number of CTG repeats were factors selected in the best-fitting model predicting ZBI scores. It is understandable that MSC is included in the model with CES-D, as they are both measuring a similar concept and was associated with each other. The number of CTG repeats extends to more than 100 in patients with DM1, while the number of repetitions among the general population is around 5 to 37 or less [2]. Generally, associations are observed between small expansions of 50–90 and mild symptoms, and between expansions of over 1,000 and congenital DM1 presenting severe symptoms [46, 47]. Thus, it is natural to suggest from this association that, severe DM1 symptoms are a factor that influence care burden. However, further research is necessary in order to consider the relevance between symptom severity and care burden, as the full spectrum of severity does not associate exactly with the number of CTG repeats [18]. Nevertheless, taking disease severity into account when planning for support of carers is crucial for improving the burden experienced when caring for the patients.
With the results of this study, we clarified the factors related to caregiver burden in DM1. However, these results are not without limitations. First, we must be cautious when generalizing these results to all patients with DM1 and their carers in Japan, as study participants were recruited mainly from northern Japan. Future studies should be conducted at multiple locations in order to include a regional variety of DM1 caregiving realities. Additionally, because support for caregivers depends largely on system and regulations on health care, the result of this study may not completely apply to other countries where care systems are different from that of Japan. Second, there were factors that could not be covered in this study due to resource issues, such as patient’s cognitive and behavioral tendencies, caregiver’s mental, physical, and economic status, marital satisfaction, time spent on care, presence or absence of a care counselor, and the availability of accessible welfare services. An analysis that includes these important factors will be necessary in future research. Finally, analyzing longitudinal data concerning caregiving burden during the process of disease progression will contribute significantly to the standardization of support and intervention methods to reduce the burden of care.
CONCLUSION
This study aimed to clarify the factors related to caregiving burden in DM1. The study found that the burden was felt among the caregivers of DM1 despite the fact that the patients were being relatively functional. There may have been a mismatch in evaluating the psychological status of patients between patients and caregivers. Possible explanations for this may be related to the disease awareness of the patients and how carers viewed patients’ moods. Genetic characteristics of patients also predicted care burden when modelled together with carer perception of patient mood. These factors should be taken into account when planning and providing support for both patients suffering from DM1 and their caregivers.
CONFLICT OF INTEREST
The authors have no conflicts of interest to report.
Footnotes
ACKNOWLEDGMENTS
The authors would like to thank the participant patients and caregivers, as well as patient advocate groups for their understanding and support during this study. This work is funded by JSPS Kakenhi grant number JP16K09735.
