Abstract
The relationships between caregiver burden as measured with the Burden Scale for Family Caregivers–short form and 6 characteristics of caregivers caring for patients with dementia were investigated for caregivers from England (n = 36), Finland (n = 42), and Greece (n = 46) using survey data. In all 3 countries, caregiver burden increases with physical problems of the caregiver, emotional problems of the caregiver, and weekly hours of care. Hence, in all 3 countries, special support for informal care is required when these characteristics are at high levels. When the caregiver is a spouse or long-term partner of the person with dementia, lives in the same house as this person, or spends fewer than 20 h/wk for other duties than care, this is associated with less caregiver burden in England but with more caregiver burden in Greece. Accordingly, special support is required for Greek caregivers with these characteristics, but the opposite is true for English caregivers.
Introduction
An essential component of care for people with dementia is performed by informal caregivers, such as close relatives or friends who did not choose caregiving as their profession but who took on this role because their close relative or friend had been diagnosed with dementia. Such informal caregivers can experience heavy caregiver burden (CB). To develop measures for alleviating CB and for identifying those caregivers who need most support, research identifying the conditions affecting CB is required. This includes research into cultural differences regarding CB because this might help in tailoring interventions to culture-specific needs. At present, hardly anything is known about such cultural differences. One of the few relevant studies stems from Konerding et al. 1 These authors found that English, Finnish, and Greek caregivers with the same level of CB endorse different items. English caregivers mainly endorse items regarding impairments in individual well-being, while Finnish caregivers endorse items regarding the conflict between the demands resulting from care and demands resulting from the remaining social life. Greek caregivers endorse items regarding impairments in their physical health. 1 These results indicate that CB is experienced differently in these 3 countries. Furthermore, cultural differences might show in the way the individual characteristics of caregivers affect CB. This is investigated here by further analyzing the data used in the study of Konerding et al. 1
Methods
The data for the original study stem from surveys performed in England, Finland, Germany, Greece, the Netherlands, and Spain as part of the research project MANAGED OUTCOMES. 2 These surveys addressed informal caregivers of patients with dementia who had been referred to hospitals for reasons other than their dementia and who were not yet living permanently in a nursing home. The caregivers were approached via the hospital administrations that identified eligible patients on the basis of the patient records. As the sample sizes in Germany, the Netherlands, and Spain were too small, only data from England, Finland, and Greece were analyzed. 1 In England, the survey was performed in March 2012, in Finland in November 2011, and in Greece in September 2011.
Questionnaire
The survey included questions addressing characteristics of the person with dementia (PWD) and the caregiver. The characteristics of the PWDs encompassed age and sex. The characteristics of the caregivers were age, sex, education (with having left school directly after the minimum school leaving age of the country classified as “low” and longer attendance at school as “high”), relationship between caregiver and PWD (“caregiver being spouse or long-term living partner” vs otherwise), living situation (“caregivers and PWD living in the same house” vs otherwise), weekly hours spent in providing care, and weekly hours spent for other duties. These other duties comprised (1) having paid work in addition to the care; (2) being in education, that is, being in vocational training, being at school, or studying; (3) having children younger than 16; and (4) additionally caring for someone else than the PWD addressed in the questionnaire.
The questionnaire also contained the EQ-5D-3L (EuroQuol, 5 dimensions, 3 levels) self-completion form and the EQ-5D-3L proxy (referring to the PWD). The EQ-5D-3L is an instrument for assessing health-related quality of life. It contains 5 questions addressing mobility, self-care, usual activities, pain/discomfort, and anxiety/depression with each dimension partitioned into the 3 levels: “no problems,” “some problems,” and “extreme problems.” 3 There are official versions for the EQ-5D-3L self-complete and the EQ-5D-3L proxy in English, Finnish, and Greek. 3 Moreover, there is evidence that the items of these 3 versions are understood in the same way in these 3 countries. 4
The questionnaire also contained the Burden Scale for Family Caregivers–short form (BSFC-s). The BSFC-s consists of 10 items with the answer categories: “agree,” “partly agree,” “mainly disagree,” and “disagree.” These categories are coded with numbers from 0 for “disagree” to 3 for “agree.” The corresponding sum scores range from 0 to 30 with higher scores indicating a greater CB. 5
The Investigated Caregiver Characteristics
Six characteristics of the caregiver were investigated: (1) physical condition, (2) emotional condition, (3) living situation (ie, whether or not living in the same house as the PWD), (4) personal relationship to the PWD (ie, whether being a spouse or long-term living partner or not), (5) extent of care given, and (6) extent of other duties in addition to providing care for the PWD. In the following text, these variables will be referred to as the “investigated caregiver characteristics.”
Caregivers were classified as having physical problems when they had reported a problem for at least 1 of the first 4 EQ-5D-3L items. Accordingly, caregivers were classified as having emotional problems when they had reported problems for the last EQ-5D-3L item. The variable “extent of care provided” was constructed by dichotomizing weekly hours spent for care into “less than 20 hours” and “at least 20 hours.” The variable “extent of other duties” was constructed by dichotomizing the total number of weekly hours spent for other duties into “less than 20 hours” and “at least 20 hours.” The limit of 20 hours was chosen because this is the equivalent of a “half-time” working position.
Statistical Analyses
Descriptive statistics were computed for age and gender of PWD and caregiver, education of caregiver, and the 6 investigated caregiver characteristics. All computations were performed for the 3 countries separately as well as for the total sample. Country differences for age were tested using multiple linear regression models with dummy-coded countries as independent variables. For the other variables, a χ2 test was applied when the corresponding conditions were fulfilled. Otherwise, Fisher exact tests were computed.
To examine whether the investigated caregiver characteristics relate differently to CB as reflected by the BSFC-s sum score, 2 multivariate linear regression models with the sum score as dependent variable were computed for each characteristic. The first model contained dummy-coded countries and the respective characterstic as independent variables, the second model also contained terms for the interaction between countries and the respective characteristic. Both models were compared using an F test. Statistically significant better fit of the model with interaction terms was interpreted as evidence for country-specific differences in the relationships between the sum score and the characteristic. Additionally, the BSFC-s sum score means were computed for the 2 levels of each characteristic both separately for each country and for all countries together, and the differences between the 2 levels were statistically tested.
Results
In total, 36 English, 42 Finnish, and 46 Greek participants were finally included in the analyses. The caregivers were predominantly in their early 60s and female; the PWDs were predominantly in their early 80s and also female. The countries differed significantly with regard to 2 of the 6 investigated caregiver characteristics and 3 of the 5 additionally assessed characteristics of the caregivers or, respectively, the PWDs (see Table 1). The Finnish PWDs were distinctly younger than PWDs in England or Greece. Moreover, the Finnish caregivers were distinctly better educated, had fewer emotional problems, and spent less time in providing care than the caregivers from the other 2 countries. The remaining 2 country differences are mainly caused by the Greek sample. In comparison with the caregivers from the other countries, the Greek caregivers were younger and less frequently spouses or long-term living partners of the PWDs.
Features of the Study Sample.
Abbreviations: ns, not significant; PWD, person with dementia; SD, standard deviation.
aTest for statistical significance of the country differences: for nominal variables, χ2 test unless a different test was stated; for variables with at least interval scale-level, multivariate regression with dummy-coded country.
bFirst number represents mean, second number represents standard deviation, third number represents minimum, fourth number represents maximum, and fifth number the number of valid data.
For 2 of the 6 investigated caregiver characteristics, there is no statistically significant interaction between country and test variable (see Table 2), that is, these 2 variables relate in the same way to CB in all 3 countries. There is, however, a statistically significant main effect in both cases. Caregiver burden is stronger when caregivers have physical problems or spend at least 20 hours on providing care (see Table 2). For one of the investigated caregiver characteristics with significant interaction, that is, for emotional problems, there are statistically significant effects in the same direction in all 3 countries, that is, CB increases with emotional problems. There are, however, large differences in the size of these effects. The effect is largest for Greece and smallest for England (see Table 2). For the remaining investigated caregiver characteristics with significant interaction, there are country-specific tendencies—if not effects—in the opposite direction. Being a spouse or long-term partner of the PWD, living in the same house as the PWD, and spending fewer than 20 h/wk for other duties are associated with less CB in England but with more CB in Greece (see Table 2).
Relationships of Caregiver Characteristics With CB Measured by the BSFC-s Sum Score.
Abbreviations: BSFC-s, Burden Scale for Family Caregivers–short form; CB, caregiver burden; ns, not significant; PWD, person with dementia.
aCell entries are as follows: (first row) mean of BSFC-s sum score for the category given in the first cell of the row; mean of BSFC-s sum score for the alternate category; (second row) difference between the 2 means listed in the first row; (third row) sample size (variations due to missing values for the investigated characteristics of care); asterisks mark statistically significant deviation from 0.
bCell entries are as follows: (first and second rows) test statistic for interaction; (third row) result of the statistical test.
c P < .05.
d P < .01.
e P < .001.
Discussion
The findings that the Finnish PWDs were younger and that the Finnish caregivers spent less time in providing care are in line with PWDs being referred to a nursing home at an earlier stage of their dementia in Finland than in other countries. 1,6–7 Accordingly, those PWDs who have not yet been referred to a nursing home, need less care and are younger than nonhospitalized PWDs from other countries. The finding that the Greek caregivers were younger and less frequently spouses or long-term living partners of the PWDs is in line with PWDs in Southern Europe having a distinctly lower risk of being institutionalized within a given time period than PWDs from Northern Europe. 1,8 Accordingly, informal care must more often be provided by the PWDs’ children or children in law who are usually younger than the PWDs’ spouses or long-term living partners.
Further results must be discussed with the caveat that the 3 country samples have not been drawn at random from the populations of caregivers in these countries. Hence, the relationships between the investigated caregiver characteristics and CB might be different for those caregivers not included in the samples. However, there are hardly any studies regarding CB based on randomly drawn samples; hence, research in this area depends upon interpreting results from samples which might not be representative.
With the caveats just mentioned, the analyses regarding the impact of the investigated caregiver characteristics on CB suggest that CB increases in all cultures when caregivers have emotional or physical problems or when they spend much time for the care. The effects for physical and emotional problems have also been found in a sample of Irish caregivers, 9 and the effects for physical problems and time spent for care in a sample of South Korean caregivers. 10 The finding that emotional problems of the caregiver have the strongest effect on CB in Greece might be explained by the fact that there is hardly any support for informal caregivers in Greece. This explanation is corroborated by the result patterns for the 2 investigated caregiver characteristics without interaction effect. For both variables, the difference between the 2 compared groups is largest for Greece (see Table 2).
The further findings regarding the relationships between caregiver characteristics and CB indicate essential cultural differences. Being a spouse or long-term living partner, living in the same house and spending not much time for other duties attenuates CB in England but intensifies it in Greece. The characteristics just enumerated establish a close bond between the caregiver and the PWD. The results suggest that this close bond supports English caregivers in their care whereas the Greek caregivers suffer from it. An effect similar to that in Greece has also been found in a South Korean sample. 10
Conclusions
The results suggest that some caregiver characteristics relate basically in the same way to CB in different cultures. The CB seems to increase in all cultures with the number of care provided and with impairments of caregiver’s health state. On the other hand, some characteristics seem to have a different impact in different cultures. One of these characteristics is the relationship between caregiver and PWD. There might be further characteristics which affect CB in different cultures differently. Therefore, before developing public health programs for supporting caregivers in a specific culture, the manner in which different caregiver characteristics affect CB in this culture should be investigated. On the basis of such an investigation, specifically tailored programs could be developed or, if available, adopted from countries in which caregiver characteristics relate to CB in a similar way.
Footnotes
Acknowledgments
The authors would also like to express their gratitude to the participating health-care service provider organizations for their cooperation and support in enabling the study to take place. These were United Lincolnshire Hospitals NHS Trust in England, the Keski-Suomi Central Hospital in Finland, and the General Hospital of Syros in Greece.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 241741. The surveys have been performed with the special assistance of Apostolos Efkarpidis, Gill Garden, Ourania Kalogeraki, and Antero Vanhalla.
