Multivariate models of subjective caregiver burden in dementia: a systematic review.

Ageing Research Reviews 15 (2014) 76–93

Contents lists available at ScienceDirect

Ageing Research Reviews journal homepage: www.elsevier.com/locate/arr

Review

Multivariate models of subjective caregiver burden in dementia: A systematic review Jacqueline van der Lee a,∗ , Ton J.E.M. Bakker b , Hugo J. Duivenvoorden c , Rose-Marie Dröes d a Psychiatric Skilled Nursing Home ‘DrieMaasStede’, Argos Zorggroep, Schiedam/VU Department of General Practice & Elderly Care Medicine, EMGO Institute for Health and Care Research VU University Amsterdam/VU University Medical Centre, Amsterdam, The Netherlands b Stichting Wetenschap Balans (Foundation for Scientific Research Geriatric Health Care), Rotterdam/VU Department of General Practice & Elderly Care Medicine, EMGO Institute for Health and Care Research VU University Amsterdam/VU University Medical Centre, Amsterdam, The Netherlands c Psychiatric Skilled Nursing Home ‘DrieMaasStede’, Argos Zorggroep, Schiedam, The Netherlands d Department of General Practice & Elderly Care Medicine, EMGO Institute for Health and Care Research VU University Amsterdam/VU University Medical Centre, Amsterdam/GGZ Ingeest, Amsterdam, The Netherlands

a r t i c l e

i n f o

Article history: Received 4 January 2013 Received in revised form 11 March 2014 Accepted 17 March 2014 Available online 25 March 2014 Keywords: Review Model Dementia Caregiver Burden

a b s t r a c t Background: Burden in dementia caregivers is a complex and multidimensional construct. Several models of burden and other representations of burden like depression or mental health are described in literature. To clarify the state of science, we systematically reviewed complex models that include both patient and caregiver determinants of caregiver burden. Objective: A review of determinant models of caregiver burden. Design: Systematic review. Data sources: Electronic databases PubMed, PsycInfo and EMbase were searched in December 2013. Study selection and analysis: Research studies with quantitative outcome measures of caregiver burden or burden-related concepts, including both patient and caregiver functional characteristics as determinants. We categorized the determinant variables in the models and calculated the percentages of proven determinants within each category. Results: We found 32 studies with burden models and 24 depression and mental health models. Patient behavioral problems, caregiver coping and personality traits and competence are most consistent determinants of caregiver burden, depression and mental health. Behavioral problems are more significant than cognitive disorders or lack of self-care. Of all measured caregiver personality traits, neuroticism has the strongest impact on caregiver burden. Regarding caregiver competences, feeling competent or enjoying higher self-efficacy in general diminish caregiver burden and promote caregiver mental health. © 2014 Elsevier B.V. All rights reserved.

Contents 1. 2. 3. 4. 5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Search strategy and selection criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author. Tel.: +31 104278139. E-mail addresses: [email protected], [email protected] (J. van der Lee), [email protected] (T.J.E.M. Bakker), [email protected] (H.J. Duivenvoorden), [email protected], [email protected] (R.-M. Dröes). http://dx.doi.org/10.1016/j.arr.2014.03.003 1568-1637/© 2014 Elsevier B.V. All rights reserved.

77 77 77 78 78 78

J. van der Lee et al. / Ageing Research Reviews 15 (2014) 76–93

5.1.

6. 7.

Models with burden as outcome variable (32 studies) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.1. Patient determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1.2. Caregiver determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Models with mental health and depression as consequences of burden (24 studies) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.1. Patient determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2.2. Caregiver determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Alzheimer’s disease and related disorders, like vascular dementia, affect approximately 30–40% of the population aged 80 or older (RIVM, 2010). The majority of dementia patients are cared for at home by family members, mainly spouses and adult children. Providing care for a dementia patient over a longer period of time is associated with emotional and physical health problems. Research performed by Netherlands Institute for Health Services Research Nivel (Peeters et al., 2007, 2010) in the Netherlands revealed that the prevalence of high levels of burden in informal caregivers of dementia patients is 20%. These caregivers feel so heavily burdened that they are at risk for health problems. Another 58% run the risk of emotional distress as a result of caregiving. The impact of dementia on the caregiver is frequently conceptualized in terms of caregiver burden. The burden concept has been the subject of research for several decades, distinguishing between objective aspects of burden, that is time spent on care and support, and subjective aspects of burden. Subjective caregiver burden is a complex and multidimensional construct, linked to caregivers’ psychological and physical well-being (Zarit et al., 1986). In the literature subjective burden is therefore related to other psychological concepts such as mood disorders, e.g. depression and anxiety, or caregiver quality of life. Conceptual differences in the definition of subjective burden lead to a diversity of measurement approaches and consequently to the development and use of different measurement instruments. Patient determinants of caregiver burden that are mentioned in the literature include cognitive function disorders, loss of autonomy and the related need of support in daily activities. However, most important are the neuropsychiatric symptoms accompanying dementia, like depression, agitation and apathy. Coping with behavioral problems has been reported to be one of the most stressful aspects of caregiving (Christie and Wood, 1988; Coen et al., 1997; Gaugler et al., 2000; Black and Almeida, 2004; Meiland et al., 2005; Allegri et al., 2006). Among the most heavily burdened caregivers, 86% of the experienced problems is connected to the disease-related behavioral changes in patients (Peeters et al., 2007). Caregivers vary widely in how they adapt to the demands of caregiving. Their characteristics make them more, or less, fit to deal with the challenge of caregiving and therefore also determine their susceptibility to stress, and burden (Pot et al., 1998; De Vugt et al., 2004; Helmes et al., 2005). Finally, the relationship between patient and caregiver can play a decisive role. Several models of burden, describing the interaction of both patient and caregiver determinants, are presented in the literature. For example, Lazarus and Folkman (1984) provided a conceptual model for understanding the caregiver stress process, which was used in further empirical research into caregivers’ adaptability by Haley et al. (1987). Pearlin’s et al. (1990) stress process framework focuses on contextual variables and distinguishes between stressors that are directly related to the caregiving role (e.g. care recipient behavioral problems), and indirectly related sources of stress (e.g. financial problems). Since Lazarus and Folkman (1984),

77

78 78 78 79 85 85 86 92 92 92

a basic principle of these stress models is the assumption that an emotional event does not automatically provoke stress, but that it depends on the individual appraisal of the event, and is therefore also mediated by personal, illness-related, social and material factors. Insight into the multiple determinants of caregiver burden has both scientific and clinical relevance: It may contribute to theory development by the construction of more comprehensive and valid theoretical models, as well as to clinical practice, by providing a rationale for the development of specific interventions to reduce caregiver burden. We therefore drew up an inventory of the existing determinants of subjective burden models described in the literature and we systematically analyzed the determinants included in these models. The central research question was: - Which patient and caregiver characteristics are significant determinants of subjective caregiver burden or burden-related concepts like caregiver mental health or depression in the different models?

2. Method 2.1. Search strategy and selection criteria An extensive search was performed of the literature on models of caregiver burden up to December 2013, utilizing the resources Medline (PubMed), PsycInfo and EMbase. The search strategy consisted of a combination of search items: burden, distress, stress, strain, overload, well-being, AND caregiver, family, spouse, informal, relative, home, AND psychogeriatrics, geriatric psychiatry, dementia, behavior, neuropsychiatric symptoms, OR theoretical, concept, predictors, prognostics, model, scale, limited by (old) age and human.1 The applied inclusion criteria were: (1) empirically tested models of burden in dementia caregivers, including both patient and caregiver determinant variables. We expected that this combination would provide more insight in their relative importance regarding caregiver burden. (2) Empirically tested models of burden-related concepts, such as (diminished) mental health or depression (as representations of burden) in dementia caregivers. Studies were excluded when the tested models included only biographical and no other functional characteristics of patients and caregivers. Additionally, references were checked for relevant publications (snowball method) and experts were consulted. Based on the content of the abstract the first inclusion of articles was performed by two independent reviewers. In case of doubt or disagreement, a third reviewer made the decision, after reading titles and abstracts. Finally, full articles were checked on the inclusion and exclusion criteria.

1

The search strategy can be obtained from the author.

78


Search in Medline (PubMed), PsyInfo and EMbase Keywords: [burden, distress, stress, strain, overload, well being] AND [caregiver, family, spouse, informal, relative] AND [psychogeriatrics, geriatric psychiatry, dementia, behaviour, neuropsychiatric symptoms] AND [theoretical, concept, predictors, prognostics, model, scale] [limited by (old) age and human]

process framework (Tables 1 and 2). All described models included two or more patient determinants such as: (1) behavioral problems or psychiatric symptoms, (2) related (lack of) self-care/need of support, (3) cognitive function disorders. The caregiver determinants in the models could be roughly categorized into: (1) caregiver role strains (Pearlin et al., 1990): social functioning, social network, social support, caregiver role; (2) caregiver physical and mental health; (3) intrapsychic strains (Pearlin et al., 1990): (sense of) competence, self-efficacy, perceived caregiving adequacy; and (4) coping or personality traits. 5.1. Models with burden as outcome variable (32 studies)

Selection of potentially relevant publications, based on title and abstract: n = 767 until December 2013

Exclusion based on content of the abstract: Publications relevant for more detailed evaluation: n = 129

n = 638

Exclusion based on main text: Studies included in

n = 73

review: n = 56 Fig. 1. Search and analysis strategy.

3. Sample From the original search in Medline (PubMed), PsycInfo and EMbase, 767 publications were selected as possible relevant studies based on their title. Subsequently, based on the abstracts, 638 publications were excluded, and, based on their main text, 129 were evaluated on the inclusion criteria. Finally, 56 publications met all inclusion criteria and were included in the review. 73 publications did not meet the inclusion criteria and were therefore excluded (Fig. 1). 4. Analysis We categorized the included patient and caregiver determinant variables of the selected multivariate models. We described the design and the study sample and checked whether the models were theory based. We also categorized the included studies based on the outcome variables in the model, that is burden and burden-related concepts like (diminished) mental health or depression. Furthermore, for each category of determinant variables, we analyzed the percentage of studies in which significant determinants were found for burden, mental health and depression respectively. Although our original intention was to also describe effect sizes, we were unable to as many studies did not provide this information. 5. Results In about 30% of the studies under review, the composition of the model was based on the Lazarus and Folkman (1984) stress-appraisal-coping model, the Poulshock and Deimling (1984) two-dimensional model of psychosocial morbidity, Haley’s (1979, 1987) stress and coping model, or Pearlin’s et al. (1990) stress

Models that included different types of patient and caregiver determinants were expected to provide the best insight into the relative importance of these determinants in the prediction of caregiver burden. We found several complex models, only two of which, however, included all seven of the above described categories of patient and caregiver determinants, namely the empirically tested models of Campbell et al. (2008) and van de Wijngaart et al. (2007), and two included six categories (Reis et al., 1994; Chappell and Reid, 2002). (Results are described in Table 1 and Fig. 2.) 5.1.1. Patient determinants All three categories of patient determinants, that is behavioral problems (lack of) self-care/need of support, cognitive function disorders, were included in the models in 17 of the 32 studies (Reis et al., 1994; Gold et al., 1995; Coen et al., 1997; Chappell and Reid, 2002; Gallicchio et al., 2002; Bédard et al., 2005; Gonyea et al., 2005; Aminzadeh et al., 2006; Davis and Tremont, 2007; van de Wijngaart et al., 2007; Campbell et al., 2008; Choi and Kim, 2008; Baker et al., 2010; Conde-Sala et al., 2010; Bergvall et al., 2011; Contador et al., 2012; Bruvik et al., 2013; Cheng et al., 2013). Behavioral problems or mood disorders, or (neuro)psychiatric symptoms, were the most frequently mentioned patient determinant in the models of caregiver burden: this conceptual category was included in 28 of the 32 studies. In all of these, except for van de Wijngaart et al. (2007), patients’ behavioral problems significantly determined caregiver burden. The category of ‘lack of self-care/need of support’ was included in 24 studies. In the models of Chappell and Reid (2002), Bédard et al. (2005), Aminzadeh et al. (2006), and Davis and Tremont (2007), Kim et al. (2012) and Bruvik et al. (2013) disabilities in activities of daily living and (lack of) self-care, increased caregiver burden. In and of themselves patient ‘cognitive function disorders’, included in the models of 25 studies, had no significance for caregiver burden. 5.1.2. Caregiver determinants The models of Campbell et al. (2008) and van de Wijngaart et al. (2007) included all four categories of caregiver determinants. In the model of Campbell et al. (2008) several caregiver determinants within these four categories were significant for burden: caregiver role captivity, caregiver overload, and caregiver neuroticism predicted higher burden while caregiver confidence was associated with decreased burden. The other models included three categories (Vitaliano et al., 1991; Reis et al., 1994; Chou et al., 1999; Chappell and Reid, 2002; Di Mattei et al., 2008) or less. Social functioning was included in 22 of the 32 studies. Less satisfaction with support predicted higher burden (Reis et al., 1994) and perceived positive support decreased burden (Shurgot and Knight, 2005). The availability of social support and the subjective appraisal of support was associated with decreased burden (Gold et al., 1995 and Coen et al., 1997 respectively). A high quality of the relationship with the patient decreased burden, while caregiver role captivity (i.e. the unwanted participation in a particular role) predicted higher burden (Campbell et al., 2008).


79

Categories of patient and caregiver determinants PATIENT cognitive function disorders (25 vars./25 st.)

CAREGIVER social functioning / support from others (27 vars./20 st.) 33%

0% PATIENT self-care / need of support (25 vars./24 st.) 36%

PATIENT behavioral problems / mood disorders (29 vars./28 st.)

79%

61%

6* 2** 1***

0*

CAREGIVER competence / self-efficacy (7 vars./7 st.)

3* 5** 6***

4* 3** 2***

6* 5** 12***

CAREGIVER health/ psychological problems (23 vars./19 st.)

3* 2** 1***

caregiver burden (32 studies)

86%

6* 3** 5***

82%

CAREGIVER coping / personality traits (17 vars./15 st.)

number of significant variables at level: * p < 0.05 ** p < 0.01 *** p < 0.001

Fig. 2. Caregiver burden and significant determinants.

Health was included in 19 studies. In the model of Di Mattei et al. (2008) only serious long-term health problems significantly increased the psychological distress of burden. Caregiver health-related determinants like physical ill-health together with a negative outlook on life (Vitaliano et al., 1991), low self-reported health and less enjoyment (Reis et al. (1994), low-rated healthrelated quality of life (Andrèn and Elmstahl, 2007), low self-rating of health (Baker et al., 2010), and (poor) physical and mental health (Conde-Sala et al., 2010) were all associated with higher burden. Furthermore, depressive symptoms (Gonyea et al., 2005), mood disorders (Davis and Tremont, 2007) and depression (Choi and Kim, 2008) were also associated with higher burden. However, subjective well-being significantly determined decreased burden in Aminzadeh et al. (2006). Competence or self-efficacy was included in seven studies. High self-efficacy significantly determined burden in the models of Chou et al. (1999) and Gonyea et al. (2005). Caregivers’ high self-esteem and the frequency of having a break were associated with (less) burden (Chappell and Reid, 2002). Self-efficacy and contingency expectancies were the best predictors of (a lower) caregiver burden (Contador et al., 2012). The same holds for self-efficacy (i.e. obtaining respite; controlling upsetting thoughts) in Cheng et al. (2013). In van de Wijngaart et al. (2007) however this relationship was not significant. Coping and personality traits were included in 15 studies. In two studies, problem-focused coping was associated with lower levels, and emotion-focused coping was associated with higher levels of burden (Chou et al., 1999; Riedijk et al., 2009). Kim et al. (2012) found that the more coping strategies caregivers used, the greater burden was reported. Most caregivers in this study tended to use emotion-focused strategies. However, in the model of Di Mattei et al. (2008) avoidant coping and problem-focused coping were not significantly associated with burden. Papastavrou et al. (2011) found a significant relationship of two out of five coping strategies, i.e. ‘positive approaches’ (positive) and ‘wishful thinking’ (negative) with burden. Caregivers with an internal locus of control are likely to be less burdened than caregivers not expecting to have control (Bruvik et al., 2013). Caregiver neuroticism, resp. higher scores on neuroticism predicted higher burden (Reis et al., 1994; Shurgot and Knight, 2005; Choi and Kim, 2008;

Melo et al., 2011), while extraversion and agreeableness decreased burden (Melo et al., 2011). Higher expressed emotion was also associated with increased burden (Vitaliano et al., 1991). In the model of Barusch and Spaid (1989) male spouses reported coping with interpersonal problems more successfully than female spouses. Remarkably female spouses reported higher subjective burden than male spouses, while they reported less caregiving tasks than men. A caregiver determinant not mentioned before, sense of coherence (SOC), or coping effectiveness, was significantly associated with decreased burden in the models of Andrèn and Elmstahl (2007) and Barusch and Spaid (1989). As our objective was to get insight in the relative importance of the multivariate determinants of burden, we inventoried the statistical measures (standardized regression coefficients and correlation measures) as far as they were available and study designs were not too dissimilar. We related these measures to the sample sizes, to get an impression of the magnitude of the effects. Independent of sample size behavioral problems had moderate effect sizes. (Baker et al., 2010; Bergvall et al., 2011), resp. n = 70 and n = 866. With respect to coping and personality and health we found similar, moderate effect sizes, both in smaller and larger sample sizes (Fig. 3).

5.2. Models with mental health and depression as consequences of burden (24 studies) Due to conceptual differences in defining the psychological consequences of caregiving, several studies in the review included models with different concepts representing caregiver burden, i.e.: depression, depressive symptoms or mental health-related outcomes. (Results are described in Table 2 and Fig. 3.) The most complex models consisted of six conceptual categories of patient and caregiver variables (Cooper et al., 2008; Rabinowitz et al., 2009). Several studies included more than one variable within a specific category, which was mostly the case within the categories of ‘caregiver coping and personality traits’ (Braithwaite, 1996, 2000; Goode et al., 1998; Hooker et al., 1998), ‘caregiver social functioning’ (Lawton et al., 1991; Crispi et al., 1997; Cooper et al., 2008) and ‘caregiver health’ (Braithwaite, 1996; Dröes et al., 1996;

80


Table 1 Multivariate models of caregiver burden. Author

Design/sample

Determinants

Based on theoretical model

Outcomes

Results

Aminzadeh et al. (2006)

Prospective study

P: cognition, self-care, behavior

No

Caregiver Global Burden

More problematic behavior (␤ 0.30) led to reduced well-being in caregivers. Providing assistance with personal care (␤ 0.13) was associated with higher levels of burden.

(Pre-post test) Outpatient/clinical n = 141 Cross-sectional

C: subjective well being

No

Burden (=general strain, isolation, disappointment, emotional involvement, environment)

Caregiver strain was determined by nursing home placement of the patient, but not by SOC. Emotional involvement was only related to SOC (␤ −0.33).

No

Caregiver strain

Memory and behavioral problems of the patient (␤ 0.517), (poor) self-rated health (␤ 0.265), duration of caregiving significantly contributed to personal strain in male caregivers.

Andrèn and Elmstahl (2007)

Outpatient

Baker et al. (2010)

n = 130 Cross-sectional

(Male caregivers) Outpatient n = 70

Barusch and Spaid (1989)

Convenience sample

(Spouses in support group) Outpatient

Bédard (1997)

n = 131 Pre-post design

Geriatric clinic Transmural n = 111 Bédard (2005)

Convenience sample

(Memory clinic) Transmural n = 557

Bergvall et al. (2011)

Cross-sectional

P: cognition, severity of dementia

C: subjective health, Sense of Coherence (SOC) P: memory and behavior problems

P: amount of support needed C: number of services used C: gender role conflict, gender identity C: self-rating of health P: memory and behavior problems

Gain

No

Burden

Higher subjective burden predominantly determined by patient memory and behavior problems (␤ 0.40), younger age and adverse social contacts. Male spouses cope with interpersonal problems (␤ 0.13) more successfully than female spouses.

No

Burden

For the male caregiver reduction of dysfunctional patient behavior, status and progression of dementia is associated with the reduction of burden at follow-up.

P: ADL-function, self maintaining C: time demands from caregiving C: support from other caregivers P: cognitive impairment

No

Burden

Less independence in IADL (␤ 0.33) and higher frequency of emotional and repetitive behavior (␤ 0.50) in patients associated with higher burden. Women caring for male patients scored higher on the burden scale. Younger caregivers reported higher burden.

P: depression, freq problem behavior C: hours spent on caregiving C: external support C: presence of medical problem P: cognition, ADL

No

Burden

Psychiatric symptoms (␤ 0.38), strongest predictor of burden and more important than patient’s ADL-scores, independent of the hours informal care.

P: functional status C: Caregiving, use of (in)formal support C: Coping responses P: cognition, mood, behavior


81

Table 1 (Continued) Author

Bruvik et al. (2013)

Design/sample

Determinants

Outpatient

P: behavioral disturbances

n = 866

C: time informal care C: resource utilization C: health related quality of life

Cross-sectional

P: cognition, (I)ADL

Outpatient

P: psychiatric and behavioral symptoms C: Locus of Control (LoC), hours caring a day P: cognitive impairment, ADL, BPSD

n = 230 Campbell et al. (2008)

Cross-sectional

Transmural n = 74

Chappell and Reid (2002)

Randomly generated sample

outpatient n = 294

Cheng et al. (2013)

Choi and Kim (2008)

Cross-sectional

Outpatient n = 99 Convenience sample

P: BPSD C: self-efficacy P: cognitive impairment, functional status

(Interview)

P. neuropsychiatric symptoms C: caregiver’s personality traits

n = 47 Convenience sample

P: cognitive impairment

Outcomes

Results

Health related Quality of Life

No

Burden

(Internal) LoC (␤ 0.28) was most important factor associated with burden, even controlled for NPI-Q score (␤ 0.24), and carer’s hours a day to assist the patient (␤ 0.19).

No

Burden

Caregiver overload, role captivity, caregiver’s level of confidence in caregiver role (␤ -0.179), caregiver’s level of neuroticism, quality of relationship, experience of adverse life events determined 81.6% of variance in burden, with largest weights for (higher level of) role captivity (␤ 0.423) and caregiver’s sense of overload (␤ 0.296).

No

Burden

Increased behavioral problems (␤ 0.3), increased informal hours of care led (␤ 0.3) to increased burden. The frequency of having a break, high self-esteem (␤ 0.14) associated with less increased burden.

P: (freq) behavioral problems C: self-esteem C: hours caregiving a week/having a break C: hours formal services/perceived support P: cognition, ADL

Outpatient

Chou et al. (1999)

C: quality caregiver patient relationship C: cgv role captivity, cgv neuroticism C: adverse life events C: cgv overload, cgv confidence P: cognitive status, ADL


Well-being Overall burdened Perceived social support had direct and strong impact on well-being, no direct impact on burden.

No

Burden

No

Depression Gain Burden

Caregivers with higher self-efficacy (␤ 0.242) in controlling upsetting thoughts had more positive gains and less burden when confronted with more behavioral problems (␤ 0.296).

Caregiver neuroticism, depression, and neuropsychiatric symptoms (rho 0.37) explained 19% variance of burden.

Depression

Yesa

Caregiver burden (4 indicators)

Physical and psychological demands (␤ 0.35) no direct effect on burden. Caregiver involvement and emotion-focused coping styles had (increasing) effect on burden.

82



Coen et al. (1997)

Conde-Sala et al. (2010)

Design/sample

Determinants

Outpatient, clinical n = 150

P: demands of care P: disturbing behavior C: caregiving involvement

Convenience sample

C: filial obligation C: self-efficacy C: coping strategies P: cognitive functioning, IADL

Transmural n = 50

P: behavioral disturbance C: supportive behaviors family/friends C: subjective appraisal of support

Cross-sectional

P: cognition, functional assessment

(2 cohorts, children, spouses) Outpatient

P: BPSD


Outcomes

Higher levels of caregiver self-efficacy (␤ 0.73) and problem-focused coping associated with lower levels of burden.

No

Caregiver burden

Behavior disturbance (F-ratio 24.24) highly predictive of burden. Both social support measures were also predictive of burden, though not to the same extent as behavior disturbance. Appraisal of social support (F-ratio 8.68) also predictive of burden, but less.

No

Burden

In children: burden determined by disinhibition, irritability (␤ 0.16), living with patient, time spent on ADL, and mental health (␤ 0.36).

C: physical and mental health

n = 251 Contador et al. (2012)

Cross-sectional

P: behavioral problems, cognition, ADL

Outpatient

C: depression, personal control

Results

No

Burden

No

Perceived burden

In spouses: burden determined by apathy, irritability, appetite disorders, time spent on ADL, and mental health. Self-efficacy and contingency (personal control)(␤ 0.32) best predicted burden, while behavioral problems did not. Burden was associated with higher levels of depression in caregiver, severity of patient behavioral problems.

n = 130 Davis and Tremont (2007)

Di Mattei et al. (2008)

Convenience sample

P: apathy, disinhibition,

(self-report questionnaire) Transmural n = 72 One measurement

P: executive disfunction, ADL C: mood

Cross-sectional

P: availability of additional support C: physical and mental health C: coping strategies P: cognitive impairmant, functional status

Outpatient

Gallicchio et al. (2002)

n = 112 Multicenter study

Transmural n = 327

P: cognition

P: behavioral disturbances C: formal services use C: self-rated health

Behavioral problems (F-ratio 12.54) predicted caregiver burden after controlling for dementia severity and caregiver depression.

Apathy was less burdensome. No

Burden

Avoidance coping increased and problem-focused coping decreased psychological distress of burden. Emotional exhaustion and health impairment (1st), emotional burden (2nd) patient’s support dependency (3rd) most important determinants of burden.

No

Burden

Female caregivers did not have a significant higher probability of high level of depression. A higher level of behavioral disturbances (OR 27.3), poorer perceived health was significantly associated with burden.

Depression


83


Design/sample

Determinants


Outcomes

Results

Gold et al. (1995)

Cohort

P: cognition, extent of assistance

No

Feelings of burden

Burden was determined by behavior disturbances (t 5.15), satisfaction with social support (t 2.15), premorbid relationship.

Two measurements Outpatient

P: memory and behavioral disturbances C: social support, social interaction

Gonyea et al. (2005)

n = 118 RCT

Outpatient n = 80

Kim et al. (2012)

Lewis et al. (2005)

P: ADL, IADL

Correlational

C: number. informal helpers, hours help a week C: coping strategies

Melo et al. (2011)

P: cognitive status, functional capacity

Clinical n = 65 (22 depr/43 dem) Cohort in RCT

P: depression C: care and support


Outpatient n = 105

Mohamed et al. (2010)

Papastavrou et al. (2011)

No

Caregiver burden

Caregiver depressive symptoms (␤ 0.51) most significant predictor of burden, followed by self-efficacy (␤ 0.33) and caregiver age.

Neuropsychiatric symptoms (␤ 0.24) alone accounted for only 6% of the variance in burden. Yesb

Caregiver burden

The more impairments in (I)ADL (␤ 0.34), the more hours spend on caregiving (␤ 0.32), the greater burden. The more coping strategies (mostly emotion-focused) (␤ 0.37), the higher burden.

No effect of number of helpers.

Comparative

Qualitative exploration Outpatient

Enjoyment

P: neuropsychiatric symptoms C: depressive symptoms, self-efficacy

Cross-sectional

Outpatient n = 274

Leinonen et al. (2001)

P: dementia severity, functional abilities

Impaired health

No

Yesc

P: BPSD, cognition, stage of dementia

No

longitud (6 months outpatient

P: psychiatric/behavioral symptoms C: intensity of care


P: behavioral problems

Burden Physical and Emotional Health

C: personality (5 traits)

P: cognitive functioning, ADL, quality life

Caregivers in dementia group were more burdened, more frequent psychological distress.

Psychological distress

P: cognitive status, behavioral problems C: subjective responses to caregiving

RCT effectiveness

Burden

Burden

Qualitative analysis: 4 framing categories of caregiving (Instrumental, Reactive, Relational, Role Acquiring). Relational: lower on burden, role captivity, and distress. Higher on positive feelings. BPSD no direct determinant in the model. Neuroticism (␤ 0.42) increased both burden and depression, extraversion (␤ 0.04) decreased burden and depression. Agreeableness (␤ 0.24) decreased burden.

Depression Distress (as result of BPSD)

No

Caregiver burden

Influence of BPSD on distress mediated by burden. Neuropsychiatric symptoms and patient quality of life (F-ratio 16.1) predicted majority of variance in 3 burden measures.

Caregiver depression

No

Caregiver distress

Patient severity of depression (F-ratio 68.4) predicted caregiver distress.

Burden

Caregivers appraisals of patients behavioral problems (rho 0.54) were related to higher levels of burden. Burden is significantly related with 2 out of 5 coping strategies: positive approachers (rho −0.20), wishful thinking (rho 0.16), the same for depression.

84



Reis et al. (1994)

Design/sample

Determinants

Correlational

C: depression, coping strategies

Outpatient clinic n = 172 Convenience sample

Longitud (2 years)

Riedijk et al. (2009)

Shurgot and Knight (2005)

Outpatient/clinical n = 213 Group comparison

C: health-related quality of life

n = 153 (63 FTD/90 AD)

C: coping styles

One sample


Secondary analysis

C: frequency of social support, familism C: positive and negative social support C: neuroticism

n = 77

Outcomes

Results

Depression

No

Burden

Burden predicted by more behavioral problems (B 0.44), higher neuroticism (B 0.14), being female, less time for recreational activities, less enjoyment, less satisfied with support.

No

Burden due to neuropsych. sympt.

Burden was predicted by coping strategy ‘seeking distraction’ (␤ 0.20) Caregivers of Fronto-Temporal Dementia (FTD) patients living at home more burdened due to neuropsychiatric symptoms (␤ 0.24) than caregivers of FTD patients residing in nursing home.

Burden of caregiving

No

Burden

Neuroticism (not presented) and behavioral problems (␤ 0.24) significantly positive relation to burden. Positive perceived social support (␤ 0.26) inversely related to burden.

No

Burden

Burden at follow-up predicted by baseline burden, ADL, vulnerability, and resources. (␤’s not presented)

Longitudinal, 1 group

P: cognition, ADL

(15 month follow-up) Outpatient

C: social network, anger,

Anxiety

C: physical health, health service utilization

Depression severity

Each component of model was unique significant predictor of burden

Burden

(High) burden predicted by problems in social functioning, behavior problems (greater) self-efficacy. Decreased burden determined by better functional health. Caregiver appraisal (perception of caregiver role as a threat) more likely to report increased burden.

n = 95 van de Wijngaart et al. (2007)

C: extent of assistance, network C: self-reported health C: personality traits P: neuropsychiatric symptoms

Clinical and outpatient

Outpatient

Vitaliano et al. (1991)

P: cognition, behavior problems


Explorative

Group interval trial Outpatient n = 95

C: outlook on life, personality, coping P: functional disability

P: severity dementia-related problems C: support, emotional aspect resources C: health status, self-belief competence C: coping

Yesa

Coping not related to burden. (No effect sizes presented.)

P = patient, C = caregiver; cgv = caregiver; (I)ADL = (Instrumental) Activities of Daily Living; BPSD = behavioral and Psychological Symptoms of Dementia. a Lazarus and Folkman (1984) stress-appraisal-coping model. b Integration of Pearlin et al. (1990) and Schulz and Martire (2004) into stress process model. c 7 Constructs derived from literature.


β

1

0.9 0.8

behavioural problems

0.7

self-care

0.6

social funconing

0.5 health

0.4 competence

0.3 personality

0.2 0.1

Sample size

0 0

100

200

300

400

500

600

700

800

900

1000

Fig. 3. Effect size of significant determinants of burden, related to sample size.

Crispi et al., 1997; Edwards et al., 2002; Haley et al., 2003; Meiland et al., 2005; Pioli, 2010) 5.2.1. Patient determinants Seven out of 24 studies (Pot et al., 1998; Mahoney et al., 2005; Meiland et al., 2005; Gilliam and Steffen, 2006; Lee et al., 2006; Cooper et al., 2008; Wu et al., 2009) included all three categories of patient determinants (behavioral problems, lack of self-care, cognitive function disorders) in their models. Behavioral problems were most frequently included in 18 of 24 studies. Behavioral problems, (neuro)psychiatric symptoms or disruptive behavior positively predicted psychological or emotional distress (Dröes et al., 1996; Pot et al., 1998; Meiland et al., 2005; Lee et al., 2006), patient depressive symptoms (Goode et al., 1998; Alspaugh et al., 1999; Rabinowitz et al., 2009) and increased anxiety and depression (Mahoney et al., 2005; Pioli, 2010) had a negative impact on the mental health of the caregiver (Hooker and Bowman, 2002) or was associated with caregiver negative affect (Rapp and Chao, 2000). In two studies, however, behavioral and psychological symptoms of dementia (BPSD) were not related to caregiver mental health (Wu et al., 2009) or to caregiver depression (Gilliam and Steffen, 2006). Lack of self-care, or need of support was included in 16 studies. In the models of Meiland et al. (2005) and Pot et al. (1998) the degree of assistance or support required by dementia patients in the activities of daily living (ADL) positively determined caregiver distress. In other studies ADL and related measures were the only determinant of depressive symptoms (Au et al., 2009), mental health (Braithwaite, 1996), and caregiver well-being (Lawton et al., 1991). However, in Wu et al. (2009), the frequency of visiting patterns to the patient residing in a nursing home (considered a measure of patient assistance in this study) was positively related to the mental health of the caregiver. Cognitive function disorders were included in the models of 17 studies. In none of these models did (severity of) cognitive function disorders in and of themselves impact on mental health outcomes in caregivers. 5.2.2. Caregiver determinants The model of Rabinowitz et al. (2009) included all four categories of caregiver determinants and was the most complex in this respect. The other models included three categories or less; outcome variables varied from depression and depressive symptoms to mental health and psychological well-being.

85

The category social functioning was included in 16 of the 24 studies, and some of them included more than one variable from this category in their models. Negative social interactions and a limited social network were significant determinants of depression (Haley et al., 2003). Attachment preoccupation (which refers to anxiousness, closely monitoring and staying by the side of the patient), together with caregiving difficulties (such as daily irritants and hassles that may affect caregivers), were predictors of psychological symptomatology (Crispi et al., 1997). Determinant variables within this category were rather diverse: Role captivity (the unwanted participation in a particular role) determined depression and anxiety in a negative way (Pioli, 2010) and played a part in the development of depressive symptoms (Alspaugh et al., 1999). The quality of the relationship with the patient was associated with anxiety and depression in caregivers (Mahoney et al., 2005). Social support was positively related to experiencing less depressive symptoms, with self-efficacy as a mediator (Au et al., 2009). Additionally, caregiver burden (belonging to the health category), and a dysfunctional relationship with the patient overlap as predictors of caregiver well-being and mental health. (Braithwaite, 2000). The category of health was included in 15 studies. Caregiver poor health led to caregiver depression (Lawton et al., 1991; Haley et al., 2003; O’Rourke and Tuokko, 2004; Mahoney et al., 2005). For spouses, poor physical health was a predictor of caregiver mental health problems (Wu et al., 2009). Caregiver worry and strain were positively related to depression (Edwards et al., 2002), and strain to caregiver negative affect (as outcome variable) (Rapp and Chao, 2000). Competence and self-efficacy were included in seven studies. For caregivers with a low sense of competence, psychiatric symptoms of the patient were a major predictor of emotional distress (Meiland et al., 2005). The sense of competence has been found to be one of the important determinants of psychological and psychosomatic symptoms in caregivers (Dröes et al., 1996). Increased mastery, and a decreased overload in caregivers determined the reduction of depression (Mausbach et al., 2007). In the model of Rabinowitz et al. (2009) ‘controlling negative thoughts’ (as one of three measures of self-efficacy) was a direct determinant of (less) depressive symptoms. Coping and personality traits were included in the models of ten studies. Emotion-oriented coping led to experiencing less, and problem-focused coping to experiencing more anxiety (Cooper et al., 2008). This finding, however, contradicted the researchers’ initial hypothesis that problem-focused coping would protect against the impact of burden on anxiety. Caregiver self-esteem (viewed in this study as a measure of coping), and a reinterpretation coping strategy promoted mental health (Braithwaite, 1996). In a next study by Braithwaite (2000), resiliency as a coping strategy was proven to promote mental health. Personality traits like neuroticism and optimism were related to a worse or better mental health respectively (Hooker et al., 1998). Stressful appraisals and changes in coping were related to negative changes in mental and physical health (Goode et al., 1998). Dröes et al. (1996) found that the coping strategy ‘expression of feelings’ moderated psychological and psychosomatic symptoms in caregivers. Coping and social support are the mediating factors between stressors such as patient cognitive functioning or behavior and psychological distress in a study of Lee et al. (2006). Relating effect sizes to sample sizes of the studies, we found quite large differences in effects between the different measures of behavior problems in Alspaugh et al. (1999) and Meiland et al. (2005), in larger and smaller sample sizes (n = 188; n = 85). For competence similar moderate effects were found in Dröes et al. (1996) and Au et al. (2009), resp. n = 80 and n = 134 (Figs. 4 and 5).

86


Categories of patient and caregiver determinants PATIENT

CAREGIVER

CAREGIVER

cognive funcon

social funconing / support from others

health/ psychological problems

disorders

(29 vars./16 st.)

(17 vars./17 st.)

(19 vars./15 st.) 66%

0% PATIENT

0*

self-care / need of support (19 vars./16 st.)

21%

58%

competence / self-efficacy 77%

(13 vars./7 st.)

6* 2** 2***

caregiver mental health

6* 4** 1***

CAREGIVER

4* 5** 3***

1* 2** 2***

PATIENT behavioral problems / mood disorders

63%

8* 6** 5***

CAREGIVER 10* 6** 2***

coping / 95%

personality traits (19 vars./10 st.)

(19 vars./18 st.)

(24 studies)

number of significant variables at level: * p < 0.05 ** p < 0.01 *** p < 0.001

Fig. 4. Caregiver mental health and significant determinants.

6. Discussion We searched the literature for determinant models of caregiver burden and consequences of burden such as caregiver mental health or depression. We selected combined models that include both patient and caregiver variables and we investigated whether patient and caregiver characteristics were predictive of burden. Our first conclusion is that only a minority of the traced determinant models are clearly based on a theoretical framework. In several studies, the composition of the model was clearly based on a theoretical model like Pearlin’s et al. (1990) stress process framework, the Lazarus and Folkman (1984) stress-appraisal-coping model, Haley’s (1979, 1987) stress and coping model or the Poulshock and Deimling (1984) two-dimensional model of psychosocial morbidity. Most models were conceptually designed based on earlier clinical research.

β

behavioural problems

1

0.9

self-care

0.8

social funconing

0.7

health

0.6

competence

0.5 personality 0.4 0.3 0.2 0.1

Sample size

0 0

50

100

150

200

250

300

Fig. 5. Effect sizes of significant determinants of mental health, related to sample size.

We found relatively large percentages of confirmed determinants in the models with caregiver burden as outcome. Regarding patient determinants, ‘patient behavioral problems’ proved to be a significant determinant of burden, positively related to increased burden, in 79% of the models that included this variable. Behavioral problems (79%), that is, ‘neuropsychiatric symptoms’, ‘behavioral and psychological symptoms of dementia’ (BPSD), ‘behavior problems’ or ‘disturbances’, outweigh ‘(lack of) self-care’ (32%) and ‘cognitive function disorders’ (0%) in terms of significance. Patient ‘mood disorders’ (e.g. depression) as a specific neuropsychiatric symptom were highly significant for burden (Bédard et al., 1997, 2005). Based on the results of this study, the category of ‘patient behavioral problems may therefore be considered the most important patient determinant of caregiver burden. The conceptualizations and measurement instruments for ‘(lack of) self-care’ were diverse. High significance was found for ‘number of hours caregiving a week’ (Chappell and Reid, 2002), and ‘need of higher maintenance’ (Vitaliano et al., 1991) both determining a (higher) burden. It must be noted that the research samples in almost all studies consisted of patients who attended outpatient or transmural services, whose caregivers provided most of the daily care. Although we can assume that his is a better performing group of patients, here also there appeared to be a clear relation between the degree of assistance needed by the patients and the burden of their caregivers. Regarding caregiver determinant variables, the following categories were most important for burden: ‘caregiver competence’ (86%), ‘caregiver coping and personality traits’ (82%), and ‘caregiver health’ (61%). If personality traits were included in the model, ‘neuroticism’ determined higher burden (Reis et al., 1994; Shurgot and Knight, 2005; Choi and Kim, 2008). If ‘emotion-focused’ and ‘problem-focused coping’ styles were included in a model, these coping styles were important determinants for higher and lower caregiver burden, respectively (Chou et al., 1999; Kim et al., 2012). Within the category of ‘caregiver social functioning’ we found the lowest percentage of significant determinant variables (33%). It should be noted that this category was also rather diverse conceptually. Here ‘caregiver role captivity’ was the most striking determinant of high burden (Campbell et al., 2008). In general,


87

Table 2 Multivariate models of caregiver depression and mental health. Author

Design/sample

Determinants


Outcomes

Results

Alspaugh et al. (1999)

Cohort, longitudinal

P: memory and behavior problems

No

(Development of) depression

Higher levels of behavioral problems (rho 0.71), role captivity (rho 0.52) and overload at T1 (rho 0.53) distinguished caregiver at risk for experiencing a depression over 1 year.

Transmural

C: role captivity, overload Depressive symptoms

Partial mediation effect of self-efficacy (␤ 0.44) between social support (␤ 0.26) and depressive symptoms.

Mental health

The strongest predictor of burden was the appraisal of dysfunctional caregiving (␤ 0.58). Resiliency (␤ −0.27) was significantly associated with less burden. Burden and dysfunctional relationship overlap as predictors of symptoms/well-being.

Au et al. (2009)


Transmural

Braithwaite (2000)

n = 134 One sample

Transmural

P: cognition, ADL

C: social support, self-efficacy P: memory loss, disorientation

No

C: coping (6 dimensions)

Burden

n = 144 Braithwaite (1996)

One sample

P: memory loss, orientation, dependency

Yes: to open the debate on the stress paradigm

Mental Health

Burden predicted by self-esteem (␤ −0.30), acceptance and reinterpretation coping strategy (␤ 0.25). By adding crisis of decline variables, burden was higher when dealing with cognitive, social, emotional degeneration.

No

Depression and anxiety

Depression was associated with spending more time caring, health (␤ -0.21), role, physical functioning and burden. Anxiety determined by burden, emotion-oriented coping (␤ −0.20), problem-focused coping (␤ 0.30). Coping mediator between burden and depression NOT proved.

No

Psychological symptomatology

Attachment style (␤ -0.36) was best predictor in all outcomes. Attachment preoccupation (␤ 0.24) was predictor psychological symptoms, not caregiving difficulty. Disease severity no significant predictor.

Transmural n = 144

Cooper et al. (2008)

C: crisis of decline C: physical health/well being/social support C: stressors, personal/social/material resources Longitudinal(LASER- P: cognition, ADL, AD) neuropsych. symptoms

Transmural

n = 91

Crispi et al. (1997)

Cross-sectional Clinical (nursing home)

C: social readjustment, social support C: health status, quality of life, burden C: coping strategies P: perceived disease severity C: attachment style/preoccupation

n = 108 Dröes et al. (1996)

Multicenter

P: behavioral problems/ADL

Outpatient

C: loneliness/social support C: financial problems/professional support C: coping/sense of competence

n = 65

Yesa

Caregiving difficulty Distress

Distress determined by financial problems (␤ 0.32), loneliness, patient dependency, with a moderator effect of coping: seeking distraction (␤ 0.27), expressing feelings (␤ 0.37). Psychological/psychosomatic symptoms determined by sense of competence (␤ −0.44) with moderator effect of coping strategy (expression of feelings) (␤ 0.44), behavior problems (␤ 0.19), loneliness (␤ 0.62). Psychological/psychosomatic symptoms

88



Design/sample

Determinants


Outcomes

Results

Edwards et al. (2002)

Longitudinal


No

Depressive symptomatology

Worry and strain (0.37; t 3.49) significant main effect on depression, interaction worry and strain × work conflict not significant.

Cross-sectional

P: Performance(ADL), time history C: work-caregiving role conflict

Outpatient n = 202 Gilliam and Steffen (2006)

Sample (spouses)

P: cognition, ADL, behavior problems

(intervention study)

C: length of caregiving, self-efficacy

No

Depression

Transmural/n = 74

Goode et al. (1998)

Haley et al. (2003)

P: impairment, memory/behavior

One group Transmural

C: social support C: coping strategies, appraisals

2 groups Clinical n = 80

Hooker et al. (1998)

Explorative

2 groups (AD-PD) Outpatient n = 175 cgv

Hooker and Bowman (2002)

P: memory/behavior problems/self-care

Explorative

P: neuropsychiatric symptoms C: cgv experiences/stress appraisal P: severity of symptoms

baseline data RCT Outpatient

n = 285 vs n = 272

C: caregiver assistance C: cgv appraisal/personal resources

Depression

Self-care, memory, behavior problems not significantly related to decline in mental, physical health. Stressful appraisals (␤ 0.19) related to decline in mental, physical health. Changes in coping (␤ 0.29) directly related to changes in mental, physical health.

Physical health

Yesc

Depression

Depression predicted by poor health (␤ −0.24), female, negative interactions (␤ −0.27), low benefits of caregiving (␤ −0.27), self-care problems (␤ 0.24), low network (␤ −0.27).

Life satisfaction

Mental health

Personality (neuroticism/optimism) significantly related to low social support, high perceived stress, and worse mental health.

Physical health

C: appraisals of social support C: neuroticism/optimism C: perceived stress

P: cognitive impairment

2nd analysis of

Yesc

C: social interactions/activities/network C: satisfaction with support C: health, perceived benefits caregiving P: cognitive No impairment

Longitudinal

Clinical/n = 64

Lawton et al. (1991)

Neither cognitive impairment nor behavior problems significantly predicted depression.

Longitudinal

n = 122 Comparative

Role overload (0.29; t 2.62) significant main effect on depression. Interaction role overload × role conflict not significant. Self-efficacy was not a moderator between patient cognitive impairment, behavior problems and caregiver depressive symptoms.

High levels of stress significantly related to worse mental health. Personality (␤ 0.38) and perceived stress (␤ 0.28) strongest effect in the model. Yesa

Physical health

Increase in problem behaviors (␤ 0.26) and caregiving for patients in long-term facility strongly associated with caregiver physical and mental health. These effects are mediated by caregiver stress appraisal (␤ 0.65).

Mental Health

Yesa , d

Satisfaction

Psychological well-being

Depression

Symptom severity resulted in more care being given and greater burden for spouses, but not related to satisfaction of psychological well-being. Poor caregiver health (␤ 0.38) associated with greater burden, more depression. Caregiver satisfaction led to positive affect, burden led to depression.


89


Design/sample

Determinants


Outcomes

Results

Lee et al. (2006)

One group

P: cognitive functioning/behavior

Yesb


Caregivers of severe dementia group reported highest level of objective and subjective stressors: behavioral problems (␤ 0.60) related to caregiver psychological distress. Same level of stressors affects caregivers differently. Mediating factors: coping (␤ −0.38), social support.

Outpatient

C: reaction to subjective stressors C: subjective burden C: coping/social support P: cognit functioning/ADL

n = 122

Mahoney et al. (2005)

Cohort, longitudinal

LASER-AD study Outpatient/clinical n = 153 Mausbach et al. (2007)

Cohort, longitudinal Outpatient/clinical n = 130 (spouses)

Meiland et al. (2005)

O’Rourke and Tuokko (2004)

Pioli (2010)

Pot et al. (1998)

P: cognitive functioning/hours of care C: stress/mastery

P: psychiat. symptoms/care/assististance C: financial expenditure/duration of care

n = 85 (partners)

C: objective/experienced support C: sense of competence P: problematic behavior/ADL

Outpatient n = 137 Sample (spouses) Outpatient

C: health problems

n = 200

C: global mastery/cgv mastery

AMSTEL-study

P: behavior problems/care tasks

(epidemiological)

C: caregiving appraisal

Outpatient/clinical n = 175

No

Anxiety and depression

Female caregiver, living with patient, worse relationship with patient, poor health, level of ADL impairment (OR 1.04) significant independent predictors of anxiety. Higher NPI scores (OR 1.18), caregiver poor health (OR 1.01), poor qualilty of relationship were predictors of depression.

Yesb

Depressive symptoms

Increased mastery (1.26; t-value: -3.18), decreased overload (052; t-value: 1.35) determined reduction in depressive symptoms.

Caregiver Health

Significant mastery × overload interaction; relation between overload and depression was conditional on level of mastery.

Yesa

Emotional impact

Psychiatric symptoms major predictor of emotional impact (52% var expl.) Higher emotional impact of psychiatric symptoms (␤ 0.73) on caregivers with low sense of competence (␤ −0.29), more financial expenditure, patient needing less care (␤ −0.38).

No

Depressive symptoms

Burden (␤ 0.29) is significant antecedent of depressive symptomatology for caregivers who live together with patient over time.

Yesb

Depression

Significant main effect of global mastery in all depression models. Problem behavior (␤ 0.28), role overload (␤ 0.32), role captivity (␤ 0.26) all positive related to depression. Positive association between role overload and depression is weaker at higher levels of caregiver mastery (␤ −0.26).

P: neuropsych sympt/depression C: quality of relationship C: physical health

Multicenter study Outpatient

Longitudinal

State anxiety

P: problem behaviors/ADL C: role overload/role captivity

Anxiety

Yesa


Perceived pressure (rho 0.39) was mediator between behavioral problems (rho 0.23) and psychological distress, and between care tasks (rho 0.24) and psychological distress in non-spouse caregivers.

Perceived pressure was only mediator between care tasks and psychological distress in spouse caregivers.

90



Design/sample

Determinants


Outcomes

Results

Rabinowitz et al. (2009)

Explorative/multisite

P: disruptive behavior

No

Depressive symptoms

Age, spouse, caregiver health (t = −7.44), level of caregiver activity, avoidant coping and patient ADL predicted depressive symptoms.

(part of REACH) Outpatient

C: activity/health C: self-efficacy/coping

Rapp and Chao (2000)

n = 256 (female) Explorative

Outpatient

Wu et al. (2009)

n = 65 Multicenter

Clinical

P: memory, behavior problems

Main effect for memory and behavior problems (t = 3.08) and self-efficacy (t = 4.51). No

Caregiver strain (␤ −0.24) accounts for more of the variance in negative affect than gender, memory/behavior problems or gain. None of the models significantly predicted positive affect.

Negative affect

C: appraisals of gain/strain P: cognition/BPSD

Positive affect

No

C: physical health/visiting patterns

Mental health

Spouse (␤ 0.22), physical health (␤ 0.41) are significant predictors of mental health. Culture(=Shanghai) significant predictor of depression only. Every day visits (␤ 0.23) related to high depressive scores.

Depression

n = 149 BPSD and caregiver psychological morbidity not associated. P = patient; C = caregiver; cgv = caregiver; (I)ADL = (Instrumental) Activities of Daily Living; BPSD = behavioral and Psychological Symptoms of Dementia; NPI = Inventory of Neuropsychiatric Symptoms (Cummings et al., 1994). a Lazarus and Folkman (1984) stress-appraisal-coping model. b Pearlin’s et al. (1990) stress process framework. c Haley’s (1979, 1987) stress process model. d Poulshock and Deimling’s (1984) 2-dimensional model of psychosocial morbidity.

a larger ‘caregiver network’ and ‘support’ were related to lower burden (Reis et al., 1994; Gold et al., 1995; Shurgot and Knight, 2005). It should be noted that for the category ‘caregiver competence’ we found only five studies. This is due to the fact that before the mid-1990s caregiver competences were not considered to be a relevant determinant of burden. In this review, caregiver competences and self-efficacy were related to lower burden in four out of five models (Chou et al., 1999; Chappell and Reid, 2002; Gonyea et al., 2005; Campbell et al., 2008). Only one study did not confirm this relationship (van de Wijngaart et al., 2007). The conceptualizations and measurement instruments for ‘health’ were rather diverse, predominantly including mental health concepts as determinants of burden. Most striking were (less) ‘subjective well-being’ (Aminzadeh et al., 2006), more ‘adverse life events’, a higher ‘caregiver stress overload’ (Campbell et al., 2008), ‘serious long-term (mental) health problems’ (Di Mattei et al., 2008), ‘depressive symptoms’ (Gonyea et al., 2005) and (diminished) ‘physical health and health service utilization’ (Vitaliano et al., 1991), all promoting burden. Complex models are expected to provide most insight into the relative importance of the determinants of burden. However, the sample sizes in the traced complex models in this review were not always large (e.g. 74 participants in Campbell et al., 2008; 47 participants in Choi and Kim, 2008), which makes these findings sensitive to coincidence. For instance, in the model of Campbell et al. (2008), none of the patient variables were significant, not even behavioral problems. In the complex model of Choi and Kim (2008) ‘caregiver neuroticism, and depression’, ‘patient neuropsychiatric symptoms’ promoted burden, but only 19% of the variance in burden was explained. Another striking observation is that in studies with a large sample size, ‘patient behavioral problems’ proved to be the most important determinant of burden, irrespective of the

inclusion of the caregiver determinant variables ‘health-related quality of life’ and ‘resource utilization’. This also holds true for the large study of Bergvall et al. (2011) with 866 participants. In the extensive studies of Bédard et al. (2005) with 557 participants, and Gallicchio et al. (2002) with 327 participants, only patient determinant variables, depressive symptoms, Activities of Daily Living (ADL), and behavior disturbances, were significant for burden. In their review of literature and guidelines for assessment and intervention on dementia caregiver burden, Dunkin and AndersonHanley (1998) also found that behavioral problems appeared to have a greater effect on burden than cognitive impairment or functional limitations. It should be noted that patient cognitive function disorders, although not proven to be a determinant of burden or mental health in this review, can indeed be relevant for burden in the future. The patients in the studies in our review predominantly used outpatient facilities and we may assume that their cognitive impairment (mean MMSE-score: ±17; mean range ±11 to ±24) was not yet severe. However, taking into account the policy2 that people who suffer from mild dementia must live at home as long as possible, and be taken care of by their relatives, this will inevitably lead to burden in their caregivers in the long run. Regarding the models with mental health or depressive symptoms as consequences of caregiver burden, we also found large percentages of significant determinants within the category ‘patient

2 Today government policy in the Netherlands (Social Support Act, WMO) stimulates citizen autonomy. In the context of the WMO it is expected that even people with disabilities can participate in society and live independently. Municipalities determine in consultation with their citizens what kind of support is needed and what can be provided by their own social environment (family, friends). In practice, it means that even elderly people in need of support are expected to look for solutions themselves. In case of intensive care, municipalities supports caregivers in their caring responsibilities.


behavioral problems’ (62%), making it more important than ‘(lack of) self-care’ (21%). As with the burden models, there is no proof that ‘cognitive function disorders’ in and of themselves are determinants for depression or mental health. The most important caregiver determinant variables were: ‘caregiver coping and personality traits’ (86%), ‘caregiver competence’ (79%), ‘caregiver social functioning’ (63%) and ‘caregiver health’ (63%). Percentages here are a little different from the models with caregiver burden as outcome. Within the category of ‘personality traits’ and ‘coping styles’, ‘neuroticism’ and ‘optimism’ as personality traits are related to poorer and better mental health respectively (Hooker et al., 1998). This is in line with studies of the burden models. ‘Emotion-focused coping’ and ‘problem-focused coping’ are related to anxiety (Cooper et al., 2008), with the striking finding that the use of emotionfocused coping is assumed here to be a protective strategy against depression and anxiety. This differs from the positive relation that was found between ‘emotion-focused coping’ and (high) burden. In a review, Cooper et al. (2007) found that caregivers experiencing high burden were more anxious if they use ineffective coping strategies. Here, no relation was found between coping styles and depression. As with the burden models, having a ‘social network’ has a positive effect on mental health, i.e. the caregiver feels less depressed (Haley et al., 2003). Most striking were the relations between ‘role overload’ (Edwards et al., 2002), ‘role captivity’ (Alspaugh et al., 1999; Pioli, 2010) and the presence of depression or depressive symptoms. Furthermore, both the outcome variables and determinants within the category ‘caregiver social functioning’ were very different and difficult to compare, and no general conclusions can be therefore drawn about their predictive value. Regarding the ‘caregiver health’ category, the most striking relations were found between ‘caregiver stress’ (Mausbach et al., 2007) and ‘caregiver health’ (Rabinowitz et al., 2009) and depressive symptoms as outcome. Because of their conceptual diversity, no general conclusions can be drawn about the predictive value of specific caregiver health determinants for depression or mental health outcomes. Within the ‘caregiver competence’ determinant category, findings are very consistent and they all point in the same direction: high ‘sense of competence’ (Dröes et al., 1996; Meiland et al., 2005), ‘mastery’ (Mausbach et al., 2007; Pioli, 2010) or ‘self-efficacy’ (Gilliam and Steffen, 2006; Au et al., 2009; Rabinowitz et al., 2009) are associated with better mental health or less depressive symptoms. The sample sizes of the most complex models with depression as outcome varied from a minimum of 80 participants (Haley et al., 2003) to a maximum of 256 (Rabinowitz et al., 2009). Both complex models include several determinant categories. In Haley et al. (2003), unlike caregiver determinant variables (mostly social network), none of the included patient determinants were of significance for ‘depression’, which may be explained by the fact that the research group consisted not of ‘common’ dementia patients but of terminally ill dementia patients who lived in a hospice. However, a large study by Lawton et al. (1991) with 285 participants also did not find a relationship between patient determinant variables (severity of symptoms) and ‘depression’. Regarding the complex models with depression or mental health as outcome, no general conclusions can be drawn about the relative importance of patient determinants as compared to caregiver determinants. Although our main aim was to perform a systematic review, we also paid some attention to a possible link between the effect sizes of significant determinants of burden and mental health and the sample size of these studies. As in larger sample sizes random errors are smaller, we expected less uncertainty about the magnitude of the effects in studies with a larger sample size, and

91

herewith in the relative importance of the multivariate determinants of burden, resp. mental health. Though most of the included studies had limited sample sizes (less than 150), there was also a substantial amount of studies with larger samples. The magnitude of the effect was less pronounced as the sample size increased, and these results applied to both burden- and mental health studies. The specific inclusion criteria we used could be considered a limitation of our study. Our review only examined models that included both patient and caregiver determinant variables. It is therefore not possible to generalize our findings to all models of caregiver burden or mental health. Another limitation concerns our categorization of the determinant variables. The categories of ‘caregiver social functioning’ and ‘caregiver health’ were conceptually rather diverse, whereas the category of ‘competence’ was very specific and was included in only 14 studies with a high rate of significant determinants (ranging from 86% in burden models and 78% in mental health models). A different categorization might have led to other percentages of proven determinants. A third limitation concerns the databases we searched. Although Pubmed, PsycInfo and EMbase together cover about 90% of the scientific medical literature, it is possible we missed some relevant studies. Moreover, we only looked for articles and not for books on this subject. This study is a review and not a meta-analysis. We would recommend performing a meta-analysis of studies with the same determinant and outcome variables with a patient sample in a specific setting (for instance a day-care facility compared to a clinical setting) to gain more insight into the relative importance of determinant variables. Pinquart and Sörensen (2003) conducted a very thorough meta-analysis of the association of stressors of caregiving for demented and physical frail older adults with burden and depression of caregivers. The difference with our review, besides that we included recent studies, is that we only selected studies which were performed among demented elderly and caregivers. Our objective was to trace the factors most strongly associated with burden and mental health and expected to find these by including multivariate models with both patient and caregiver measures, the last mostly considered as moderator variables. This means that we laid more emphasis on the diversity of potential caregiver determinants of burden. However, as Pinquart and Sörensen, we also found that the behavioral problems of dementia patients were consistently related with caregiver burden and mental health. Besides, in accordance with Pinquart and Sörensen, we can conclude that further research is needed regarding caregiver moderator variables such as gender and coping. The scientific relevance of our study is that it provides insight into the relative importance of patient and caregiver determinants for burden, depression and mental health by measuring the percentages of proven determinants. We assume we have included the most important determinant models of caregiver burden, depression and mental health in our review, and have not overlooked other existing models. We recommend that further research into models including the here proven determinants, such as patient behavioral variables, caregiver personality traits and coping styles and caregiver competences, should be performed with a larger sample size. The clinical relevance of the study is that insight in the proven determinants of caregiver burden and depression or mental health provides starting points for the design of treatment and support programs with combined interventions for dementia patients and their caregivers. Treatments that focus on patient behavioral problems as well as improving caregiver competences and coping. With respect to caregivers at risk for burden, specific attention has to be paid to the development of prevention programs. Caregiver burden has been studied only since a few decades, due to an increasing understanding that the effects of caring for others may have negative consequences and the less pleasant aspects should also be clarified. The Zarit Burden Interview (Zarit

92


et al., 1980, 1986), which was used in most studies of this review, operationalizes the negative consequences of caring and makes it possible to study these separately. The concept of burden is gaining importance and has scientific, clinical and societal implications. After all, the increased individualism of citizens goes together with a more or less government-imposed responsibility to care for other people, especially relatives. Having various concurrent responsibilities, including caring tasks, inevitably leads to high pressure for the individuals involved. Caregiver burden and its determinants remain an important subject for further research. 7. Conclusion It may be concluded from this review that patient’s behavioral problems or mood disorders are consistently reported as important determinants of caregiver burden, depression and mental health. Especially for burden, most studies show that behavioral problems are more significant than cognitive disorders or lack of self-care. Caregiver resources, for example personality traits, coping styles, and competences, are also strong determinants, and may be considered mediators between the impact of patient’s behavioral problems and caregiver burden, depression and mental health. Of all measured caregiver personality traits, neuroticism has the strongest impact on caregiver burden. Regarding caregiver competences, feeling competent or enjoying higher self-efficacy in general diminish caregiver burden and promote caregiver mental health. The predictive strength of caregiver social functioning and caregiver health in determining caregiver burden or mental health is less evident, due to the conceptual diversity of these determinants. Acknowledgements This study was funded by a grant from the Netherlands Organization for Health Research and Development (ZonMw) (www.zonmw.nl), which is a national organization that promotes quality and innovation in the field of health research and health care, initiating and fostering new developments. The majority of ZonMw’s commissions come from the Ministry of Health, Welfare and Sport (VWS). The Foundation for the Advancement of Scientific Research in Nursing Homes [Stg. Wetenschaps Bevordering Verpleeghuiszorg], The Netherlands. References Allegri, R.F., Sarasola, D., Serrano, C.M., Taragano, F.E., Arizaga, R.L., Butman, J., Lon, L., 2006. Neuropsychiatric symptoms as a predictor of caregiver burden in Alzheimer’s disease. Neuropsychiatr. Dis. Treat. 2 (1), 105–110. Alspaugh, M.E., Liming, Stephens, M.A., Parris, Townsend, A.L., Zarit, S.H., Greene, R., 1999. Longitudinal patterns of risk for depression in dementia caregivers: objective and subjective primary stress as predictors. Psychol Aging. 14 (1), 34–43. Aminzadeh, F., Byszewski, A., Dalziel, W.B., 2006. A prospective study of caregiver burden in an outpatient comprehensive geriatric assessment program. Clin. Gerontologist 29 (4), 47–60. Andrèn, S., Elmstahl, S., 2007. Relationships between income, subjective health and caregiver burden in caregivers of people with dementia in group living care: a cross-sectional community-based study. Int. J. Nurs. Stud. 44 (3), 435–446. Au, A., Lai, M-K., Lau, K-M., Pan, P-C., Lam, L., Thompson, L., Gallagher-Thompson, D., 2009. Social support and well-being in dementia family caregivers: the mediating role of self-efficacy. Aging Ment. Health 13 (5), 761–768. Baker, K.L., Robertson, N., Connelly, D., 2010. Men caring for wives or partners with dementia: masculinity, strain and gain. Aging Ment. Health 14 (3), 319–327. Barusch, A.S., Spaid, W.M., 1989. Gender differences in caregiving: why do wives report greater burden? Gerontologist 29 (5), 667–676. Bédard, M., Molloy, D.W., Pedlar, D., Lever, J.A., Stones, M.J., 1997. Associations between dysfunctional behaviors: gender, and burden in spousal caregivers of cognitively impaired older adults. Int. Psychogeriatr. 9 (3), 277–290. Bédard, M., Kuzik, R., Chambers, L., Molloy, D.W., Dubois Lever, J.A., 2005. Understanding burden differences between men and women caregivers: the contribution of care-recipient problem behaviors. Int. Psychogeriatr. 17 (1), 99–118.

Bergvall, N., Brinck, P., Eeek, D., Gustavsson, A., Wimo, A., Winblad, B., Jonsson, L., 2011. Relative importance of patient disease indicators on informal care and caregiver burden in Alzheimer’s disease. Int. Psychogeriatr. 23 (1), 73–85. Black, W., Almeida, O.P., 2004. A systematic review of the association between the behavioral and psychological symptoms of dementia and burden of care. Int. Psychogeriatr. 16 (3), 295–315. Braithwaite, V., 1996. Understanding stress in informal caregiving: is burden a problem of the individual or of society? Res. Aging 18 (2), 139–174. Braithwaite, V., 2000. Contextual or general stress outcomes: making choices through caregiving appraisals. Gerontologist 40 (6), 706–717. Bruvik, F.K., Ulstein, I.D., Ranhoff, A.H., Engedal, K., 2013. The effect of coping on the burden in family carers of persons with dementia. Aging Ment Health 17 (8), 973–978. Campbell, P., Wright, J., Oyebode, J., Job, D., Crome, P., Bentham, P., Jones, L., Lendon, C., 2008. Determinants of burden in those who care for someone with dementia. Int. J. Geriatr. Psych. 23 (10), 1078–1085. Chappell, N.L., Reid, R.C., 2002. Burden and well-being among caregivers: examining the distinction. Gerontologist 42 (6), 772–780. Cheng, S.T., Lam, L.C., Kwok, T., Ng, N.S., Fung, A.W., 2013. Self-efficacy is associated with less burden and more gains from behavioral problems of Alzheimer’s disease in Hong Kong Chinese caregivers. Gerontologist 53 (1), 71–80. Choi, Y., Kim, M.J., 2008. The correlation between caregiver’s personality traits and caregiver burden in dementia. Eur. Neuropsychopharmacol. 18 (S4), S504. Chou, K-R., LaMontagne, L.L., Hepworth, J.T., 1999. Burden experienced by caregivers of relatives with dementia in Taiwan. Nurs. Res. 48 (4), 206–214. Christie, A.B., Wood, E.R.M., 1988. Age, clinical features and prognosis in S.D.A.T. Int. J. Geriatr. Psych. 3 (1), 63–68. Coen, R.F., Swanwick, G.R.J., O’Boyle, C.A., Coakley, D., 1997. Behaviour disturbance and other predictors of carer burden in Alzheimer’s disease. Int. J. Geriatr. Psych. 12 (3), 331–336. Conde-Sala, J.L., Garre-Olmo, J., Turró-Garriga, O., Vialta-Franch, J., López-Pousa, S., 2010. Differential features of burden between spouse and adult-child caregivers of patients with Alzheimer’s disease: an exploratory comparative design. Int. J. Nurs. Stud. 47 (10), 1262–1273. Contador, I., Fernandez-Calvo, B., Palenzuela, D.L., Migueis, S., Ramos, F., 2012. Prediction of burden in family caregivers of patients with dementia: a perspective of optimism based on generalized expectancies of control. Aging Ment Health 16 (6), 675–682. Cooper, C., Balamurali, T.B.S., Livingston, G., 2007. A systematic review of the prevalence and covariates of anxiety in caregivers of people with dementia. Int. Psychogeriatr. 19 (2), 175–195. Cooper, C., Katona, C., Orrell, M., Livingston, G., 2008. Coping strategies: anxiety and depression in caregivers of people with Alzheimer’s disease. Int. J. Geriatr. Psych. 23 (9), 929–936. Crispi, E.L., Schiaffino, K., Berman, W.H., 1997. The contribution of attachment to burden in adult children of institutionalized parents with dementia. Gerontologist 37 (1), 52–60. Cummings, J.L., Mega, M., Gray, K., Rosenberg-Thompson, S., 1994. The neuropsychiatric Inventory, comprehensive assessment of psychopathology in dementia. Neurology 44, 2308–2314. Davis, J.D., Tremont, G., 2007. Impact of frontal systems behavioral functioning in dementia on caregiver burden. J. Neuropsychiatr. Clin. Neurosci. 19 (1), 43–49. Di Mattei, V.E., Prunas, A., Novella, L., Marcone, A., Cappa, S.F., Sarno, L., 2008. The burden of distress in caregivers of elderly demented patients and its relationship with coping strategies. Neurol. Sci. 29 (6), 383–389. Dunkin, J.J., Anderson-Hanley, C., 1998. Dementia caregiver burden: a review of the literature and guidelines for assessment ad intervention. Neurology 51 (Suppl. 1), S53–S60. Edwards, A.B., Zarit, S.H., Stephens, M.A.P., Townsend, A., 2002. Employed family caregivers of cognitively impaired elderly: an examination of role strain and depressive symptoms. Aging Ment. Health 6 (1), 55–61. Gallicchio, L., Siddiqi, N., Langenberg, P., Baumgarten, 2002. Gender differences in burden and depression among informal caregivers of demented elders in the community. Int. J. Geriatr. Psych. 17 (2), 154–163. Gaugler, J.E., Davey, A., Pearlin, L.I., Zarit, S.H., 2000. Modeling caregiver adaptation over time: the longitudinal impact of behavior problems. Psychol. Aging 15 (3), 437–450. Gilliam, C.M., Steffen, A.M., 2006. The relationship between caregiving self-efficacy and depressive symptoms in dementia family caregivers. Aging Ment. Health 10 (2), 79–86. Gold, D.P., Cohen, C., Shulman, K., Zucchero, C., Andres, D., Etezadi, J., 1995. Caregiving and dementia: predicting negative and positive outcomes for caregivers. Int. J. Aging Hum. Dev. 41 (3), 183–201. Gonyea, J.G., O’Connor, M., Carruth, A., Boyle, P., 2005. Subjective appraisal of Alzheimer’s disease caregiving: the role of self-efficacy and depressive symptoms in the experience of burden. Am. J. Alzheimers Dis. Other Demen. 20 (5), 273–280. Goode, T., Haley, W.E., Roth, D.L., Ford, G.R., 1998. Predicting longitudinal changes in caregiver physical and mental health: a stress process model. Health Psychol. 17 (2), 190–198. Haley, W.E., LaMonde, L.A., Han, B., Burton, A.M., Schonwetter, R., 2003. Predictors of depression and life satisfaction among spousal caregivers in hospice: Application of a stress process model. J. Palliat. Med. 6 (2), 215–224. Haley, W.E., Levine, E.G., Brown, S.L., Bartolucci, A.A., 1987. Stress, appraisal, coping and social support as predictors of adaptational outcome among dementia caregivers. Psychol. Aging 2 (4), 323–330.

J. van der Lee et al. / Ageing Research Reviews 15 (2014) 76–93 Helmes, E., Green, B., Almeida, O.P., 2005. Individual differences in the experience of burden in caring for relatives with dementia: role of personality and mastery. Aust. J. Ageing 24 (4), 202–206. Hooker, K., Monahan, D.J., Bowman, S.R., Frazier, L.D., Shifrin, K., 1998. Personality counts for a lot: predictors of mental and physical health of spouse caregivers in two disease groups. J. Gerontol. Psychol. Sci. 53B (2), P73– P85. Hooker, K., Bowman, S.R., Padgett Coehlo, D., Rae Lim, S., Kaye, J., Guariglia, R., Li, F., 2002. Behavioral change in persons with dementia: relationships with mental and physical health of caregivers. J. Gerontol. Ser. B: Psychol. Sci. Soc. Sci. 57 (5), P453–P460. Kim, H., Chang, M., Rose, K., Kim, S., 2012. Predictors of caregiver burden in caregivers of individuals with dementia. J Adv Nurs 68 (4), 846–855. Lawton, M.P., Moss, M., Kleban, M.H., Glicksman, A., Rovine, M., 1991. A two-factor model of caregiving appraisal and psychological well-being. J. Gerontol. 46 (4), P181–P189. Lazarus, F., 1984. Stress, Appraisal, and Coping. Springer, New York. Lee, H.S., Kim, D.K., Kim, J.-H., 2006. Stress in caregivers of demented people in Korea: a modification of Pearlin and colleagues’ stress model. Int. J. Geriatr. Psych. 21 (8), 784–791. Leinonen, E., Korpisammal, L., Pulkkinen, L-M., Pukuri, T., 2001. The comparison of burden between caregiving spouses of depressive and demented patients. Int. J. Geriatr. Psych. 16 (4), 387–393. Lewis, M.L., Hepburn, K., Narayan, S., Nelson Kirk, L., 2005. Relationship matters in dementia caregiving. Am. J. Alzheimers Dis. Other Demen. 20 (6), 341–347. Mahoney, R., Regan, C., Katona, C., Livingston, G., 2005. Anxiety and depression in family caregivers of people with Alzheimer disease: the LASER-AD study. Am. J. Geriatr. Psych. 13 (9), 795–801. Mausbach, B.T., Patterson, T.L., van Känel, R., Mills, P.J., Dimsdale, J.E., Ancoli-Israel, S., Grant, I., 2007. The attenuating effect of personal mastery on the relations between stress and Alzheimer caregiver health: a five-year longitudinal analysis. Aging Ment. Health 11 (6), 637–644. Meiland, F.J.M., Kat, M.G., van Tilburg, W., Jonker, C., Dröes, R.M., 2005. The emotional impact of psychiatric symptoms on partner caregivers: do caregiver, patient, and situation characteristics make a difference? Alzheimer Dis. Assoc. Disord. 19 (4), 195–201. Melo, G., Maroco, J., De Mendonca, A., 2011. Influence of personality on caregiver’s burden, depression and distress related to the BPSD. Int J Geriatr Psychiatry 26 (12), 1275–1282. Mohamed, S., Rosenbeck, R., Lyketsos, C., Schneider, L.S., 2010. Caregiver burden in Alzheimer disease: cross-sectional and longitudinal patient correlates. Am. J. Geriatr. Psych. 18 (10), 917–927. O’Rourke, N., Tuokko, H.A., 2004. Caregiver burden and depressive symptomatology: the association between constructs over time. Clin. Gerontol. 27 (4), 41–52. Pearlin, L.I., Mullan, J.T., Semple, S.J., Skaff, M.M., 1990. Caregiving and the stress process: an overview of concepts and their measures. Gerontologist 30 (5), 583–594. Peeters, J., Francke, A., van Beek, S., Meerveld, J., (in Dutch) 2007. Which dementia caregivers feel most burdened? Results of the national programme in

93

dementia care (Welke groepen mantelzorgers van mensen met dementie ervaren de meeste belasting?). NIVEL and the Alzheimer Foundation. Peeters, J.M., Van Beek, A.P.A., Meerveld, J.H.C.M., Spreeuwenberg, P.M.M., Francke, A.L., 2010. Informal caregivers of persons with dementia, their use and needs for specific professional support: a survey of the National Dementia Programme. BMC Nurs. 9, 9. Pinquart, M., Sörensen, S., 2003. Associations of stressors and uplifts of caregiving with caregiver burden and depressive mood: A meta-analysis. J Gerontol B Psychol Sci Soc Sci. 58 (2), 112–128. Pioli, M.F., 2010. Global and caregiving mastery as moderators in the caregiving stress process. Aging Ment. Health 14 (5), 603–612. Poulshock, S.W., Deimling, G.T., 1984. Families caring for elders in residence: issues in the measurement of burden. J. Gerontol. 39 (2), 230–239. Pot, A.M., Deeg, D.J.H., van Dyck, R., Jonker, C., 1998. Psychological distress of caregivers: the mediator effect of caregiving appraisal. Patient Educ. Couns. 34 (1), 43–51. Rabinowitz, Y.G., Mausbach, B.T., Gallagher-Thompson, D., 2009. Self-efficacy as a moderator of the relationship between care recipient memory and behavioral problems and caregiver depression in female dementia caregivers. Alzheimer Dis. Assoc. Disord. 23 (4), 389–394. Rapp, S.R., Chao, D., 2000. Appraisals of strain and of gain: effects on psychological wellbeing of caregivers of dementia patients. Aging Ment. Health 4 (2), 142–147. Reis, M.F., Gol, D.P., Andres, D., Markiewicz, D., Gauthier, S., 1994. Personality traits as determinants of burden and health complaints in caregiving. Int. J. Aging Hum. Dev. 39 (3), 257–271. Riedijk, S.R., De Vugt, M.E., Duivenvoorden, H.J., Niermeijer, M.F., Van Swieten, J.C., Verhey, F.R.J., Tibben, A., 2009. Sense of competence in a Dutch sample of informal caregivers of frontotemporal dementia patients. Dement. Geriatr. Cogn. Disord. 27 (4), 337–343. RIVM., 2010 Towards Better Health: Core Report of Public Health and Forecasts. (Report in Dutch) Bilthoven The Netherlands. Shurgot, G.R., Knight, B.G., 2005. Influence of neuroticism, ethnicity, familism, and social support on perceived burden in dementia caregivers: pilot test of the transactional stress and social support model. J. Gerontol. B: Psychol. Sci. Soc. Sci. 60 (6), P331–P334. Vitaliano, P.P., Russo, J., Young, H.M., Teri, L., Maiuro, R.D., 1991. Predictors of burden in spouse caregivers of individuals with Alzheimer’s disease. Psychol. Aging 6 (3), 392–402. De Vugt, M.E., Stevens, F., Aalten, P., Lousberg, R., Jaspers, N., Winkens, I., Jolles, J., Verhey, F.R.J., 2004. Do caregiver management strategies influence patient behaviour in dementia? Int. J. Geriatr. Psych. 19 (1), 85–92. Wu, H.Z.Y., Low, L-F., Xiao, S., Brodaty, H., 2009. Differences in psychological morbidity among Australian and Chinese caregivers of persons with dementia in residential care. Int. J. Geriatr. Psych. 24 (12), 1343–1351. van de Wijngaart, M.A.G., Vernooij-Dassen, M.J.F.J., Felling, A.J.A., 2007. The influence of stressors, appraisal and personal conditions on the burden of spousal caregivers of persons with dementia. Aging Ment. Health 11 (6), 626–636. Zarit, S.H., Reever, K.E., Bach-Peterson, J., 1980. Relatives of the impaired elderly: correlates of feelings of burden. Gerontol. 20 (6), 649–655. Zarit, S., Todd, P., Zarit, J., 1986. Subjective burden of husband and wives as caregivers: a longitudinal study. Gerontol. Soc. Am. 26, 260–266.

Caregiver burden for informal caregivers of patients with dementia: A systematic review.

Caregiver burden: a clinical review.

Dementia Caregiver Burden: a Research Update and Critical Analysis.

Predicting perioperative mortality after oesophagectomy: a systematic review of performance and methods of multivariate models.

From Subjective Cognitive Complaints to Dementia: Who is at Risk?: A Systematic Review.

Enteral nutrition in dementia: a systematic review.

Subjective assessment versus ultrasound models to diagnose ovarian cancer: A systematic review and meta-analysis.

Caregiver burden in end-of-life care: advanced cancer and final stage of dementia.

Caregiver burden: the strongest predictor of self-rated health in caregivers of patients with dementia.

Effectiveness of Caregiver Interventions on Patient Outcomes in Adults With Dementia or Alzheimer's Disease: A Systematic Review.

Caregiver burden and hemodialysis.

Lightening the load: an overview of caregiver burden in dementia care.

Is sense of coherence helpful in coping with caregiver burden for dementia?

Comparative Analysis of Informal Caregiver Burden in Advanced Cancer, Dementia, and Acquired Brain Injury.

Estimating the burden of early onset dementia; systematic review of disease prevalence.

Factors Underpinning Caregiver Burden in Frontotemporal Dementia Differ in Spouses and their Children.

Factors Associated with Caregiver Burden in Dementia: 1-Year Follow-Up Study.

Neuropsychiatric symptoms in Parkinson's disease dementia are associated with increased caregiver burden.

Self-efficacy moderation and mediation roles on BPSD and social support influences on subjective caregiver burden in Chinese spouse caregivers of dementia patients.

Effectiveness of coping strategies intervention on caregiver burden among caregivers of elderly patients with dementia.

Caregiver burden characterization in patients with Alzheimer's disease or vascular dementia.

Memantine improves goal attainment and reduces caregiver burden in Parkinson's disease with dementia.

Oxytocin for frontotemporal dementia: a systematic review.

Development of a measure of caregiver burden in paediatric chronic kidney disease: The Paediatric Renal Caregiver Burden Scale.