Caregiving, care burden and awareness of caregivers and patients with dementia in Asian locations: a secondary analysis

Background This study investigated the differences in caregiver activity, caregiver burden, and awareness of both caregivers and patients with Alzheimer’s disease (AD) across different Asian locations. Methods This was a secondary analysis of a multi-national cohort study that aimed to assess caregiver activity and caregiver burden using the Caregiver Activity Scale (CAS) and Zarit Burden Interview (ZBI), respectively. Patients’ awareness of their dementia diagnosis was assessed by asking the following yes/no question: “Do you have dementia?” Caregivers’ awareness of the patient’s dementia diagnosis was assessed by asking the following yes/no question: “Does your patient have dementia?” Results In total, 524 caregivers of patients with AD from China, Hong Kong, South Korea, the Philippines, Singapore, Thailand, and Taiwan participated. The CAS and ZBI score were significantly different across most locations (p < 0.001 and p = 0.033, respectively). Overall, 56.6% of caregivers and 37.5% of patients had awareness of the dementia diagnosis, and the proportion of patients and caregivers with awareness were also different between each location (all, p < 0.001). Conclusions Caregiving, caregiver burden, and the awareness of caregivers and patients were different across many Asian locations. With understanding of cultural differences, further public education on dementia could help increase the awareness of patients and caregivers and reduce caregiver burden. Trial registration ClinicalTrials.gov, NCT02262975. Registered 13 October 2014,


Background
According to the World Alzheimer Report 2015, dementia currently affects more than 46 million people worldwide and this number is estimated to increase to 131.5 million by 2050 [1]. In comparison to 2009 estimates, the estimated prevalence of dementia has increased in Asia and Africa but decreased in Europe and in the United States [1]. In 2005, the Delphi Consensus Report estimated that 48% of dementia patients live in Asia and that this percentage is estimated to grow to 59% by 2050 [2]. Therefore, countries in Asia face a greater proportion of the global burden.
The World Alzheimer Report also reported that the majority of patients with dementia are still living at home, and that 94% of people living with dementia in low-and middle-income countries are cared for at home [1]. Therefore, those often directly involved in the care of dementia patients are their families, and as a result, dementia has a strong impact on the wellbeing, functioning and quality of life of not only home-dwelling dementia patients but also their family caregivers [1]. Efforts should be made to determine what factors affect caregiving activity and caregiver burden and what measures are necessary to reduce them.
A large proportion of patients with Alzheimer's disease (AD) have anosognosia (i.e., denial, impaired awareness or a lack of awareness of their cognitive deficits, functional limitations, and disease-related behavioral changes) [3][4][5]. The prevalence of anosognosia in AD varies between 20 and 80%, although this is dependent on factors such as the assessment method used, sample characteristics, and the severity of dementia [5]. In addition to the poor awareness of patients themselves, the awareness of caregivers can also affect the prognosis of patients with dementia as well as caregiver burden [6]. For example, Boise et al. found that caregivers delayed seeking a diagnosis because they lacked information about dementia; caregivers believed symptoms were signs of normal aging [7]. Caregivers were also generally overwhelmed by the situation and did not know which physician to ask [7]. Therefore, poor awareness of both the patient and caregiver may significantly affect the prognosis and caregiving of dementia patients.
Caregiver burden, caregiver awareness, and patient awareness are also influenced by cultural and socioeconomic factors [8,9]. In Asia, there is a cultural diversity that is likely to significantly affect caregiver burden and the awareness of dementia patients and their family caregivers. During the original study, the Caregiver Activity Scale (CAS), the Zarit Burden Interview (ZBI), and simple questions about awareness of the disease were conducted at baseline. A secondary analysis of the results of these surveys was conducted to investigate caregiver burden, caregiver awareness, and patient awareness in seven Asian locations, and whether cultural diversity had any effect.

Original study design and cohort
The original cohort study was a multi-national trial to identify treatment discontinuation rates in de novo Alzheimer's disease patients who have been newly prescribed with DOnepezil in ASia (ADOS) (NCT02262975). In brief, the original study was conducted on patients who met the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association criteria [10] for probable AD and were aged between 50 and 90 years. Inclusion criteria included patients who were prescribed donepezil monotherapy for the first time and who had a caregiver who could accompany them to the hospital. In total, 527 patients from 38 institutions across seven countries in Asia (see Appendix) were included in the original study. Among them, three caregivers did not assess the caregiver questionnaire. Therefore, in this analysis, data from 524 caregivers were analyzed.
The original study was performed between July 2014 and August 2016 and conducted in accordance with Good Clinical Practice (GCP) and the Declaration of Helsinki as well as with the regulations of the International Conference on Harmonisation. The Institutional Review Board guidelines of each center were adhered to in all research centers, and written informed consent was obtained from all patients and caregivers who participated in the study.

Survey assessments
In the current analysis we investigated caregivers and patients with AD dementia from seven Asian geographic locations (China, Hong Kong, the Philippines, Singapore, South Korea, Taiwan, and Thailand). General demographic and background characteristics including caregiving activities, social support, and social activities of the caregiver were recorded at the first visit. The clinical characteristics of patients were also recorded including a mini-mental state exam (MMSE) score [11] and clinical dementia rating (CDR) [12].
This analysis used the CAS [13] and the ZBI [14] to assess caregivers of patients who participated in the ADOS study. The ZBI is one of the most commonly used scales for assessing the burden of caregivers who provide care for dementia patients, with a higher score indicating a greater burden [15]. The ZBI is a 22-item questionnaire measuring subjective burden, which has demonstrated high consistency and validity. The CAS was used to collect information regarding the time caregivers spent providing care during a typical 24-h period, with increased time indicative of greater burden. The clinical characteristics of the patients and caregivers were then used to determine the presence of any national or cultural differences in CAS and ZBI survey data. Survey information was collected at baseline and was not affected by whether or not participants completed a follow-up visit.
Additionally, patients' awareness of their dementia diagnosis was assessed by asking the following yes/no question: "Do you have dementia?" Caregivers' awareness of the patient's dementia diagnosis was assessed by asking the following yes/no question: "Does your patient have dementia?" The CAS and ZBI scores and patient and caregiver awareness were compared between Asian locations to determine which demographic and clinical characteristics of the caregivers and patients affected caregiver burden. These assessments were performed before the initial prescription of donepezil.

Statistical analyses
A two-sample t-test and an analysis of variance (ANOVA) were used as parametric hypothesis tests while analyzing differences between two and three groups, respectively. For non-parametric hypothesis testing, a Wilcoxon's rank sum test and a Kruskal-Wallis test were used to test differences between two and three or more groups, respectively. A Fisher's exact test with Monte Carlo simulation and Pearson's Chi-square test (χ2) were used to compare percentages. To analyze the association between patients, caregiver awareness, ZBI, and CDR scores, linear regression and logistic regression methods were used. The statistical significance threshold was set to α = 0.05 with a 95% confidence level. Statistical analyses were performed using SAS® software (version 9.3; SAS Institute Inc., Cary, NC, USA).

Background characteristics of caregivers and patients
In the original study, 524 caregivers of patients with AD were recruited from China (n = 101), Hong Kong (n = 10), South Korea (n = 259), the Philippines (n = 23), Singapore (n = 52), Thailand (n = 6), and Taiwan (n = 73) ( Table 1). The overall mean age of caregivers was 59.17 ± 15.35 years with the oldest age in Hong Kong (80.20 ± 6.88 years) and the lowest age in the Philippines (46.30 ± 14.23 years). In total, 61.3% of caregivers were female; however, there were more male caregivers in China (53.5%) and Taiwan (52.1%). The mean duration of education of caregivers was 11.58 ± 5.20 years, with longer durations of education in the Philippines (15.17 ± 2.87 years) and shorter durations of education in Hong Kong (3.00 ± 4.55 years) and Thailand (3.00 ± 1.73 years). There were significant differences across all seven locations for mean age, sex, marriage status, duration of education, employment status, and relationship to the patient (all, p < 0.001). In general, caregivers were the spouse (35.3%), daughter (29.4%), or son (21.5%). The percentage of children as caregivers was 58.9%, if son, daughter, or son-or daughter-in-law were included; however, there were significant differences between locations (p = 0.005). Table 2 describes the demographic and clinical characteristics of patients. Overall, patients mean age was 75.68 ± 7.07 years. The mean MMSE score was 18.66 ± 5.45 (n = 504), and the mean CDR (global) was 0.94 ± 0.53 (n = 522; range 0.5-3) in the overall population.
Caregiver activity scale and Zarit burden interview score Figure 1 illustrates the CAS and ZBI scores obtained in each location. The CAS score (p < 0.001) and ZBI score (p = 0.033) were significantly different between each location.
Linear regression analysis showed that in all studied locations, the CAS score was generally higher for patients who were male, overweight, and/or tall and who were cared for by an older caregiver, unemployed caregiver, poorly educated caregiver, non-child caregiver, or a caregiver who resided with them (Table 3). Additionally, in China, the CAS score was high if the patient had a severe CDR (n = 96; estimate 137.96 ± 58.37, p = 0.02). In Singapore, the CAS score was high if the caregiver was unmarried (n = 49; estimate 392.76 ± 110.21, p = 0.001), or if the patient was aware of their illness (n = 49; estimate 209.15 ± 93.88, p = 0.031). In the Philippines, the CAS score was higher according to the lower MMSE score (n = 23; estimate − 62.60 ± 14.81, p < 0.001) and the more severe CDR (n = 23; estimate 678.90 ± 183.87, p = 0.001) of the patients.
Generally, in all studied locations, the ZBI score was high if the patient was male or old, with a low MMSE score and a high CDR, and if the caregiver was female or cohabiting with the patient ( Table 3). The patient's awareness of the illness was unrelated, but the ZBI score was high if the caregiver was aware of the patient's dementia. Specifically, in South Korea, the ZBI score was high if the caregiver's duration of education was short (n = 247; estimate − 0.51 ± 0.19, p = 0.008), and the caregiver was unemployed (n = 256; estimate 4.71 ± 2.05, p = 0.023). In Singapore, the ZBI score was high if the patient was unaware of the illness (n = 52; estimate 16.08 ± 3.61, p < 0.001). In Hong Kong, the ZBI score was high if the caregiver was spouse of the patients, not the child (n = 8; estimate − 33.50 ± 12.58, p = 0.037).

Awareness of caregivers and patients
The awareness of patients and caregivers of the patients' dementia diagnosis differed by location (Fig. 2). Overall, 56.5% of caregivers and 37.4% of patients had awareness; however, in Hong Kong and the Philippines, 100.0% of caregivers had awareness. In contrast, 49.0 and 41.7% of caregivers in China and South Korea had awareness, respectively. The proportion of patients with awareness were 80.0 and 78.3% in Hong Kong and the Philippines, respectively, followed by Thailand (66.7%), Singapore (57.7%), Taiwan (52.8%), China (35.7%), and South Korea (23.6%), in descending order (p < 0.001).
Logistic regression analysis showed that, in general, patients were likely to have awareness if they were female and if their caregiver was young, educated or their child (Table 4)   Caregivers were more likely to have awareness if the patient was female, older, and severely demented and if the caregiver was young, had a longer duration of education, and had a high ZBI score (Table 4). Caregivers' awareness was inversely proportional to patients' awareness (Table 5), and caregivers were

Discussion
The present study showed that 56.6% of caregivers had awareness of their patient's dementia and 37.4% of patients were aware of their dementia, and that patients' and caregivers' awareness were mutually influential. Patients were more likely to be aware of their dementia if they were female and the caregiver was young, highly educated, and one of the patient's children. From the caregivers' perspective, high awareness was achieved if they were young, female, highly educated, and one of the patient's children, and if the patient was highly demented.
The CAS score was more likely to be high if the patient was male, overweight, and tall and if the caregiver was old, poorly educated, unemployed, the spouse, the primary caregiver, or cohabiting with the patient. The ZBI score was more likely to be high if the patient was male, old, severely demented, or with dysfunctional cognitive skills and if the caregiver was female, cohabiting with the patient, the primary caregiver, and if the caregiver was aware of the patient's illness. Interestingly, patients' awareness of their illness did not significantly influence caregiver CAS and ZBI scores; however, the ZBI score was high when the caregiver was aware of the patient's illness.
The factors that influence CAS and ZBI scores were similar to what has been previously published in other studies. Caregivers who are of an advanced age [16,17], are female [18][19][20], and are cohabiting with the patient,  are more likely to experience a greater burden [21]. Several studies have also reported that spousal caregivers experience the highest level of burden [16,21,22]. Furthermore, some studies have also suggested that moderate-to-severe disability, which affects a patient's ability to perform basic daily activities, is related to a higher caregiver burden [16,21,23,24]. However, care burden about dementia patients in Asia is somewhat different, compared to the Western countries. Ikels followed a sample of 200 elders from 1987 to 1991 in Canton, China and reported that the experience of giving care to patients with dementia was psychologically less threatening than for people in the United States [25,26]. This is possibly because in China there are cultural concepts that preserve the patients' sense of self for longer and caregiving offers a greater intrinsic reward to the family caregiver.
Caregiver burden has previously been defined as "a multidimensional response to physical, psychological, emotional, social, and financial stressors associated with the caregiving experience" [27]. This description differs from caregiving, which is referred to as the "activities and experiences involved in providing help and assistance to relatives who are unable to provide for themselves", which does not include the psychological distress that may come from the act of caregiving [28]. In this study we used both CAS and ZBI to distinguish caregiving activity from caregiver burden in each location.
The factors influencing awareness also differed by location. The patients' and caregivers' awareness were low, below 50%, in South Korea and China compared to other locations. This is presumably because South Korea and China have a relatively more traditional Confucian culture, which might prevent dementia from being recognized as an illness [29]. In previous studies, there have been reports that there is a paucity of information on dementia in South Korea and China [30,31]. One longitudinal prospective study of cognitive impairment among elderly adults showed that the prevalence rate of dementia in persons aged 65 years and older was 4.6% in Shanghai [32]. This is much lower than reports from Western countries where researchers have reported difficulty in recruiting Asian caregivers who perceive a sense of shame for having a relative with AD [33][34][35]. Furthermore, Asian caregivers have been shown to delay seeking an AD diagnosis because they lacked information about dementia and believed the symptoms were signs of normal aging [7]. A large proportion of patients with AD have anosognosia [5,6], and in our study the greater the severity of dementia the more their awareness deteriorated. With higher MMSE scores observed in South Korea and China, it might also be considered that antidementia medications could be prescribed earlier before patients and their family recognize their cognitive dysfunction. Interestingly, in this study, the awareness in patients was 20-30% lower than the awareness in caregivers, even though each country has differences in culture and education. When exploring the emotional impact of an Alzheimer's disease diagnosis, although the majority of patients display signs of an emotional crisis, a range of responses including lack of insight, active denial, and positive coping responses have been reported [36]. In addition, a lack of awareness of cognitive functioning deficits is a complex and multidimensional phenomenon that is also related to a lack of awareness of functional activity impairments, age, and caregiver burden [37]. Further analysis of our findings would be helpful for understanding and developing strategies to lessen caregiver burden in different Asian countries.
A caregiver's awareness is strongly influenced by personal preferences and practicality does not seem to play a substantive role when making a decision related to caregiving. Strong family-related emotions can also often overwhelm rational decision-making, and therefore, in the case of dementia, medical professionals are required to play a more explicit role in the decision-making process by anticipating transitions in care and outlining options for care and treatment in advance. The assessment of awareness has a high clinical relevance in cases of dementia, particularly when considering the impact this has on patients and their family as well as the social implications of their increased need for medical, social, legal, and financial support [38,39].
One of the limitations of this study was the differences in both patient and caregiver sample sizes in each country, which is likely to affect the accuracy of any national comparisons. In particular, the number of South Korean patients was considerably higher than that of other locations, which may have influenced our findings. In addition, the differences in sample sizes from each country could be due to differences in refusal rate, which could impart selection bias; however, we were unable to analyze the refusal rate because of differences between the sample sizes of each country. Finally, although we aimed to assess awareness of dementia before the initial prescription of donepezil, it is possible that some patients were diagnosed with dementia and provided with information on their disease before our assessment. Therefore, these patients are likely to have a higher level of awareness of their dementia diagnosis at baseline.

Conclusions
This study examined factors that affect caregiver burden as well as patient and caregiver awareness of dementia across seven Asian locations. This analysis offers a unique perspective on patients with dementia and caregiver management. Ultimately, these data may be useful when planning future multinational studies by incorporating the location and cultural diversity that can influence the care of patients with dementia.