Background
The prevalence of dementia in the elderly in South Korea was reported to be 10.7%, and the number of patients with dementia is estimated to exceed 3 million by 2050 [
1]. Alzheimer’s disease (AD) is the most common type of dementia, accounting for 60–70% of all patients with dementia [
2].
The prevalence of non-cognitive behavioural and psychological symptoms of dementia (BPSD) in patients with AD is 56–98% in the community and up to 91–96% in hospitals or long-term care facilities [
3]. BPSD exacerbates cognitive decline and physical dysfunction in patients with AD [
4] and imposes a great burden and stress on the caregivers of patients with dementia [
3,
5].
In particular, family caregivers of patients with dementia were found to be associated with social isolation, physical health deterioration, and psychological disorders such as depression. Likewise, these people were also found to experience financial difficulties [
6]. Owing to the prolonged disease course, the burden on family caregivers has led to emerging social problems such as conflicts within the family, deterioration of the patient–caregiver relationship, patient abuse, and suicides. Therefore, studies on family caregiver burden are warranted.
Factors related to caregiver burden as reported in previous research are cognitive function [
5], stages of dementia [
7], depression [
7], activities of daily living (ADL) [
8], instrumental activities of daily living (IADL) [
9], and BPSD [
5,
10]. However, it is reported that cognitive function [
11], disease progression [
12], depression [
11], ADL [
11], and IADL [
13] are related to BPSD. In addition, some studies have reported on the relationships between BPSD, cognitive function, caregiver burden, and therapeutic effects [
14]. However, causal relationships between such factors have not yet been established. Furthermore, since caregiver burden and BPSD have been studied with a focus on a single symptom, BPSD expression characteristics have not been fully reflected. As BPSD manifests with the co-occurrence of various symptoms, a trend to investigate BPSD as symptom clusters or groups of related symptoms in a wide range [
15] has been established. Previous studies have reported on the classification of the clusters of BPSD symptoms in patients with dementia [
15‐
17]. Garre-Olmo et al. [
16] classified BPSD into psychotic, emotional, and behavioural syndromes. In South Korea, a study reported on factor analyses of BPSD in patients with AD living in an institutional setting [
18], and another study reported the effects of BPSD in elderly patients with dementia in nursing homes [
5]. However, studies regarding the causal relationship between BPSD in community-dwelling patients with AD and caregiver burden have not yet been established. There is also a lack of research that classifies BPSD as symptom clusters to determine how each cluster affects caregiver burden and which clusters have the greatest impact on burden. If we determine which symptom cluster affects caregiver burden and which BPSD makes up these clusters, then the same treatment and nursing approach could be employed for these symptoms [
15].
The present study aimed to (i) classify BPSD in patients with AD, (ii) identify the causal relationship between BPSD and the factors affecting family caregiver burden, and (iii) provide empirical data for the development of systematic therapeutic and nursing intervention programmes to address family caregiver burden.
Discussion
Since BPSD is the most difficult problem to deal with for family caregivers, who are the primary caregiver for the elderly with dementia, figuring out how BPSD affects the family caregivers’ burden is critical in that it can provide not only the well-being of the family caregivers but also further basic data that can contribute to improving the quality of life (QOL) of AD patients. Our study focuses on classifying BPSD as a symptom cluster and, subsequently, linking these factors that affect BPSD to each cluster, which differentiates this study from previous studies.
Mean scores for BPSD were higher than those for family caregiver burden in patients with AD. These scores of our study were higher than those reported in a study [
38] of patients with AD living in veterans’ homes in Taiwan using the same tool. However, these scores were lower than those reported in a study of outpatients with dementia in Portugal [
39]. The reason that the scores for BPSD and caregiver burden were lower in the study by Hsu et al. [
38] than in our study could be because the subjects of the former were living in veterans’ homes and their caregivers were nurses, whereas the patients of this study were diagnosed with AD at the outpatient clinic at a tertiary general hospital and their caregivers were family members. It is also thought that the subjects of the study conducted by Melo et al. [
39] had various types of dementia, including AD, frontotemporal dementia, and vascular dementia. The findings of this study indicate that a higher BPSD is associated with a higher family caregiver burden. However, because there is a difference in the subjects, their residence, and the characteristics of caregivers between this study and previous studies, it is thought that there may be a variable difference in the levels of BPSD and caregiver burden. Nevertheless, 71.8% of the subjects were under-educated people, and it is necessary to check the impact of this education level. In the study of García-Alberca et al. [
40], years of education, one of the AD patient general characteristics showed correlation with a range of depression in the tests of measurements taken to determine caregivers’ burden. However, another research [
41], which showed no correlation between years of education and caregiver burden, the other [
42] showed a faster cognitive decline in AD patients with higher education levels. Further studies with repetition of the analyses will be needed to determine whether or not patient’s cognition and caregiver burden are related to the patient’s education level.
The results of the factor analysis of 12 symptoms of BPSD showed that BPSD can be classified into three symptom clusters:
hyperactivity, psychosis, and
physical behaviour symptom clusters.
Hyperactivity symptoms consisted of disinhibition, irritability, and agitation/aggression, which were consistent with the cluster name and pattern reported in a study conducted by Kang et al. [
18] In addition, studies with the same cluster name and pattern [
15] and studies with similar cluster patterns but different cluster names [
17] were also identified.
Psychosis symptoms consisted of hallucinations, anxiety, elation/euphoria, delusions, and depression/dysphoria. Except for depression/dysphoria, the remaining symptoms were found to be symptoms included in the
psychosis symptom cluster identified in previous studies [
15‐
18,
43]. However, in these studies, depression was also classified as
mood [
17,
43],
emotional [
16], and
affective clusters [
15,
18]. Moreover, in other studies, euphoria has not been grouped together with other symptoms but has been classified as a single symptom cluster [
15,
17].
Physical behaviour symptoms consisted of appetite and eating abnormalities, apathy/indifference, aberrant motor behaviour, and sleep and night-time behaviour disturbances. A few previous studies [
17,
18,
43] reported similar cluster patterns with different cluster names than those of this study, while other studies have reported that aberrant motor behaviour was classified as a common symptom belonging to a symptom cluster of the same name “physical behaviour symptoms.” [
16,
17] The diversity and variation in cluster classifications and nomenclature, with varying agreements, similarities, and differences between studies, are thought to be due to differences in the characteristics of the subjects, sample size, and analytical methods. While this study found correlations between each related symptom, further studies are required to confirm its BPSD-symptom-cluster classification, determine generalised BPSD clusters, and propose intervention programmes through cluster-specific approaches.
Testing the model of this study showed that the three symptom clusters had a direct effect on the family caregiver burden, whereby the
psychosis symptoms cluster was the most influential. The results also found that the stages of dementia, ADL, and IADL indirectly affected family caregiver burden through hyperactivity, psychosis, and physical behaviour symptoms, indicating that BPSD had a total mediating effect. Garre-Olmo et al. [
44] reported that BPSD in patients with AD had a direct effect on caregiver burden and ADL had an indirect effect on caregiver burden through BPSD, which was consistent with the results of this study. In addition, a study by Onishi et al. [
45] showed that the severity of dementia and physical disability had a direct effect on caregiver burden, which was similar to the results of this study. In the modified model of this study, cognitive function demonstrated no effect on BPSD and caregiver burden. In the study conducted by Kang et al. [
41], cognitive impairment status, sub-domains of neuropsychological test’s memory function, and frontal executive function showed correlation with caregiver burden measured using K-MMSE. The results are similar to those that did not act as an influence factor in the regression analysis. However, in the meta-analysis for family caregivers of home-dwelling elderly people with dementia [
46], the factors such as family caregiver’s characteristics, problematic behaviour, cognition (memory), stages of dementia, and ADL have an effect on the caregiver burden. In particular, ADL [
47,
48] and IADL [
48] are also factors that affect the QOL for dementia patients. Therefore, when providing interventions for caregiver burden due to BPSD in patients with AD, it is necessary to explore strategies and algorithms that consider the different stages of dementia to be able to develop the necessary interventions, while focusing on slowing disease progression and improving functional independence in daily living. Furthermore, repeated studies are needed to determine whether cognition acts as an influencing factor on caregivers’ burden by utilising various scales that are used to measure cognition.
In accordance with the “Global action plan on the public health response to dementia (2017–2025)” announced by the World Health Organization [
49], the Dementia Friendly Communities emerged, supported by the government’s Dementia Management policy. The Korean Ministry of Health and Welfare has proposed a dementia system, and they plan to institute dementia care centres in villages as a national imperative, which is part of its continual expansion strategy [
50]. However, this is a long-term and complex project that requires appropriate preparation, budget allocations, and consensus among community members. It is, therefore, recommended that the government authorities fully support the development and establishment of systematic strategies to expand programmes and actively publicise them, so that patients with dementia and their family caregivers can utilise them.
Since patients with dementia are in chronic and disabling conditions and the family caregivers can be seen as a hidden patient [
51], strategies to reduce caregivers’ burden due to BPSD and improve caregivers’ QOL should also be considered. In a previous study [
52], if the caregiver burden was high, the occurrence of AD in patients with neuropsychiatric symptoms (NPS) was more frequent and severe. Engagement coping strategies of the caregiver showed a mediating effect between caregiver burden and NPS. Another study [
53] that measured coping strategies of the AD patient’s caregiver utilised the Coping Inventory for Stressful Situations (CISS). This study reported that the success of the caregiver is related to task-focused and avoidance-focused strategies. The results of the other study showed that emotion-focused coping strategies decreased BPSD and caregiver burnout [
54]. Heavy burden and stress among caregivers who take care of dementia patients can lead to poor QOL of AD patients [
55]. A correlation was observed between caregiver burden and QOL [
56] because BPSD and caregivers’ burden have an effect on the QOL of the caregiver [
48,
57]; adequate coping strategies for reducing the burden and stress are required for improving the QOL of the caregiver [
58]. In particular, Korean dementia families tend to use passive attitudes, avoidance, and anger [
59]; therefore, they need an active intervention to train their families’ effective problem-oriented coping.
Although there is a need for customised intervention according to the preference of patients and their family caregiver, in Korea, it is difficult to apply a person-centred care model that individually supports the target, as it is based on specialised medical models such as doctors and social workers. In addition, the institutional foundation for universal welfare based on the social model is not yet firmly established, as the cost of health and social services is eased, but it remains an economic burden to dementia patients and their family members. Therefore, it is necessary to decide the direction of dementia policy through philosophical discussion in Korea, deviating from simply comparing and applying dementia policy cases in other countries [
60]. Discussion on person-centred dementia care has recently expanded [
61,
62] to ensure that dementia patients and their family caregivers are the main agents of the policy.
The findings of this study may not be applicable to the general population as the participants in this study were outpatients. Thus, these findings are limited in terms of generalisation and interpretation. In addition, this study used data previously collected for diagnostic and therapeutic purposes and was limited in describing family caregivers’ burden using a multidimensional approach that included characteristics, such as social position, monthly income, work position, caregiving time demands, home care service costs, medical device costs, educational level, and psychological factors [
63‐
65]. Longitudinal studies covering the time span of disease progression, commencing from the time of the diagnosis of dementia, are recommended to determine changes in the factors affecting the classification of BPSD and family caregivers’ burden as the disease progresses. Finally, it is necessary to analyse how the coping strategies of the caregiver medicate the BPSD symptom clusters, the caregiver burden, and the QOL of the patient and family caregiver while controlling the general characteristics of the patient and the caregiver, such as sex, age, and education level.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.