Rural Caregivers for a Family Member With Dementia

Abstract

Forecasts of increasing prevalence of dementia in rural settings, coupled with reliance on family caregiver support, indicate that a greater understanding of caregiver distress in these contexts is necessary. The purpose of this study was to examine family caregiver burden and severity of distress on the day that a family member was diagnosed with dementia at a memory clinic that serves a rural population. Participants in this retrospective study were 231 primary family caregivers of a rural community-dwelling person with dementia. On the diagnostic day, women reported more burden and severity of distress than men and spouses reported more severity of distress than adult children. A structural equation model was not supported for the entire sample, but was supported for women caregivers only (n = 161). Caregiver distress related to dementia-specific behaviors explained both global distress and burden. Patients’ functional decline was related to caregiver burden.

Keywords

caregiving dementia gender rural and remote health care

According to the World Health Organization (WHO) and Alzheimer’s Disease International (2012), the current global prevalence of dementia is expected to double by 2030 and almost triple by 2050. The potential family and economic burden of this trend calls for proactive policies to address the needs of community-dwelling people with dementia and their caregivers. In Canada, the 2011 census revealed a higher rate of growth of seniors aged 65 and above in the previous 5 years, compared with younger age groups. The proportion of seniors was highest (17.2%) outside of communities with a population of 10,000 or more, which includes rural, small town, and remote areas (Statistics Canada, 2012). Along with a shift from institutional to community care, this pattern is evident on an international scale (WHO & Alzheimer’s Disease International, 2012) and highlights the pressing need for more understanding and support of family caregivers in rural and remote areas.

Despite the considerable research that has been conducted on challenges for family dementia caregivers (see reviews by Etters, Goodall, & Harrison, 2007; Yap, Seow, Henderson, & Goh, 2005), there are gaps in our knowledge of caregiver distress in rural communities (reviews by Goins, Spencer, & Byrd, 2009; Innes, Morgan, & Kosteniuk, 2011). To date, the majority of family caregiver studies have been conducted in urban settings.

The purpose of this retrospective study was twofold. First, we examined differences in rural family caregiver burden and severity of distress related to caregiver gender (male vs. female) and relationship (spouse vs. adult child) to the family member with dementia. Second, we developed a causal model of caregiver burden and severity of distress, based on data collected on the day of diagnosis. This research was conducted at a memory clinic located in an urban center that exclusively serves rural populations in a western Canadian province.

Caregiving and Gender

Since women have historically been socialized to assume caregiving roles (Brewer, 2001), it is not surprising that the emphasis in the literature has been on women as caregivers (Brody, 1990). Nevertheless, there has been an increased interest in men as caregivers in recent years (Baker & Robertson, 2008; Harris & Bichler, 1997). More generally, gender has implications for caregivers (e.g., Bédard et al., 2005). In this study, we define gender as “socially constructed relationships between women and men, among women, and among men” that develop through the process of social interactions (Gerson & Peiss, 1985, p. 317). Gender roles are subject to a variety of geographical (Little & Panelli, 2003), historical, and cultural factors.

In a study of women providing care for a family member with dementia, Neufeld and Harrison (2003) interpreted the caregivers’ perception of nonsupport and negative interactions from family and friends within symbolic interaction theory (Blumer, 1969), which posits that the self evolves through ongoing interaction with others. They note that the related identity theory (Hogg, Terry, & White, 1995) provides a “useful lens” because it suggests that “individuals negotiate self-conceptions through their interactions and in the context of what is socially valued for each of the roles they assume” (Neufeld & Harrison, 2003, p. 329). Within the caregiver–care recipient relationship, meanings are constructed that emerge and evolve over time. For example, in a study of caregivers who have a spouse diagnosed with Alzheimer disease, Calasanti and Bowen (2006) found that while both husbands and wives cross traditional gender boundaries in their caregiving tasks, they tend to maintain the gender identity of their spouse, which implies recognition of the importance of masculinity and femininity to the personal identity of the care recipient. This study illustrates a reconstruction of caregiver gender roles, while caregiver effort is made to maintain the gender identity of the spouse with dementia to preserve their dignity.

Research on subjective burden experienced in the care of a family member with dementia has generally revealed that women report more burden than men in similar situations (Bédard et al., 2005). Gender differences in perceptions of distress have been explained in a variety of ways, such as sex-role orientation (Wilken, Altergott, & Sandberg, 1996), coping styles (Lutzky & Knight, 1994), and care-recipient problem behaviors (Bédard et al., 2005). Kramer (1997) has suggested that there may be a difference in the way that men appraise strain and gain. In a study of gender differences in burden and depression (Gallicchio, Siddiqi, Langenberg, & Baumgarten, 2002), female caregivers of community-based elders with dementia reported more burden than their male counterparts, but there was no difference in the caregiver outcome of depression. Calasanti and King (2007) also suggest that men and women have gendered experiences of distress based on their repertoire of gendered ways of coping. A recent review article on male caregiving in dementia (McDonnell & Ryan, 2013) concluded that while there are similarities in the caregiver experience across gender, there is a need for more research on differences as a basis for the development of interventions specific to the gender of the caregiver.

Caregiver–Care Recipient Relationship

The literature on caregiving has focused more on spousal (husband, wife) care relationships (Braun et al., 2009) than intergenerational relationships, such as adult children providing care for a mother or father with dementia and bearing the additional concern about a familial risk to themselves (Jarvik & Blazer, 2005). Given the fact that sons have been less likely to serve as dementia caregivers than daughters, much of the research on filial caregiving has related to daughters (e.g., Depp et al., 2005; Ward-Griffin, Oudshoorn, Clark, & Bol, 2007). In a comparison of the coping strategies of wives and daughter as caregivers (Wilcox, O’Sullivan, & King, 2001), there were similarities in psychological measures (burden, stress, depression, anger) but differences in coping strategies, with daughters more likely than mothers to use adaptive problem-focused coping and less likely to use maladaptive coping strategies of blaming others and self-blame. The latter strategies were correlated with “anger-in” and “anger-control,” measured by Spielberger’s Anger Expression Scale. Expression of emotion, particularly anger, has been related to physiological outcomes. In an urban longitudinal study of female and male caregivers of a spouse with Alzheimer disease, Shaw et al. (2003) concluded that caregiving stress and emotional expressivity contribute independently to the risk of hypertension.

Barber and Pasley (1995) conceptualize predictors of strain in caregivers of a family member with Alzheimer disease in terms of time-variant factors (progressive dementia impairment) and time-invariant factors (caregiver gender and generational kinship). They argue that caregiver outcomes may differ if two time-invariant predictors are examined at the same time (e.g., gender and generational relationship). They studied the same four caregiving groups in the Midwest and western United States as we have in the present study in the Canadian prairies (i.e., wives caring for husbands, husbands caring for wives, daughters caring for parent[s], and sons caring for parent[s]). The theoretical potential with this approach is that our understanding of the impact on the caregiver is “sharpened by distinguishing caregiver gender, generational relationship [spousal vs. filial], and specific family kin relation [wife, husband, daughter, or son]” (Barber & Pasley, 1995, p. 189). They found that family strain, social restriction, and declining physical health varied according to the gender and generational relationship that the caregiver had with the care recipient.

The specific role designation (e.g., wife, son) may be less important in relation to perceived distress than the quality of the relationship that has evolved (Quinn, Clare, & Woods, 2009) or the course of the dementia, particularly related to development of behavioral symptoms (Black & Almeida, 2004). The impact on the caregiver of the cognitive, behavioral, and personality changes in the person with dementia, includes a loss of intimate exchange and loss of relationship quality that increases the risk of depression in the caregiver (Adams, McClendon, & Smyth, 2008).

The Rural Context

Caregiver distress may be related to accessibility of services in rural and remote areas, but research evidence has often focused more on urban settings than rural. Loos and Bowd (1997) studied the lived experience of 68 caregivers of persons with Alzheimer disease living in rural and isolated regions of northern Canada and found that a unique sense of personal loss and grief was central to their caregiving experience. Spousal caregivers tended to interpret their loss of control over events with “resignation and duty,” whereas adult children caregivers were more likely to “express anger and bitterness” with their shrinking social world (Loos & Bowd, 1997, p. 511). The loss was compounded in rural areas by lack of social support services.

By contrast, another Canadian study compared rural and urban informal caregivers for cognitively impaired older adults (Bédard, Koivuranta, & Stuckey, 2004) and found that, while rural caregivers had less access to formal services and were more likely to be nonspousal and employed, they did not differ from urban caregivers in reported burden or health status. Similarly, rural caregivers’ distress, burden, and coping did not differ based on caregivers’ geographic access to a major metropolitan center (O’Connell, Germaine, Burton, Stewart, & Morgan, 2013).

A study of male caregivers of wives or mothers with dementia and another chronic condition in rural regions in the Midwest United States (Sanders, 2007) revealed a range of experiences from informal supports to social isolation. This study differentiated the caregivers’ perceptions of support from their willingness to ask for help. The author notes that the experience of caregiver isolation was exacerbated by the demographics of rural communities where young people leave for urban settings and older people in the potential support network may be in caregiving roles themselves. In a rural–urban comparison of African American caregivers and care recipients, Chadiha, Feld, and Rafferty (2011) found that there were fewer informal supports for caregiving (i.e., secondary caregivers) in rural contexts than in urban settings. By contrast, Byrd, Spencer, and Goins (2011) found no rural–urban differences in caregiver burden or depressive symptoms despite assumed fewer formal supports in rural settings. They interpreted these findings as an adaptation to lack of formal services by reliance on informal supports. Clearly, there is need for more research in rural contexts to examine burden within the specific contextual factors that determine the availability and use of informal supports by dementia caregivers.

In a systematic review of research on the informal caregiving of family members with dementia living in rural and remote settings, Innes et al. (2011) found that there was “limited work examining the experiences of rural caregivers and their education and support needs” (p. 34). They concluded that most of the studies of rural family caregivers over the past two decades have been small and descriptive, rather than comparative, and that the understanding of caregiver needs in rural settings warrants further attention around the globe.

Another review article that focused more broadly on rural caregiving of older adults (Goins et al., 2009) underscored the need to examine the impact of living in a rural setting on the family caregiving experience. As well as drawing attention to the limited understanding of “what is rural about rural caregiving,” they identified limitations in the existing research such as cross-sectional rather than longitudinal designs, lack of operational definitions of variables, and absence of a theoretical orientation.

Theoretical Perspectives

A theoretical underpinning of the concept of caregiver burden is the cognitive model of stress and coping developed by Lazarus and Folkman (1984), with burden defined as negative cognitive appraisal that is viewed as an antecedent of physical and mental health problems such as depression (O’Rourke & Tuokko, 2003). The concept of burden has usually been considered a unidimensional construct, as operationalized by the Zarit Burden (ZB) Interview (Schreiner, Morimoto, Arai, & Zarit, 2006). More recently, an alternative view has emerged that burden is a multidimensional construct. For example, Savundranayagam, Montgomery, and Kosloski (2011) found three dimensions of burden for spouses and adult child caregivers: stress (anxiety), relationship (caregiver–care recipient dyad), and objective burden (time taken in the caregiver role). Furthermore, they found that caregiver time to help the care recipient with activities of daily living was only related to objective burden; whereas, care recipients with problem behaviors were associated with all three types of caregiver burden and with the intent to institutionalize.

In a critical review of research based on the stress and coping model, Kneebone and Martin (2003) concluded that while we know more about which coping styles are most effective (e.g., problem solving and acceptance), the 12 cross-sectional and four longitudinal studies reviewed do not use measures of coping strategies that are specific to the types of problems that arise for the caregiver as dementia progresses. Another perspective on coping draws on notions of psychological resources such as self-efficacy, competence, or personal mastery. For example, Adams, Smyth, and McClendon (2005) compared spousal caregivers with non-caregivers and found that a sense of personal mastery had a greater protective effect on mediating depression in caregivers than in non-caregivers, demonstrating the benefits of interventions that enhance a sense of control over one’s life and environment.

The day of diagnosis has been examined through qualitative research to build theory that is specific to the dementia experience. For example, Aminzadeh, Byszewski, Molnar, and Eisner (2007) developed a preliminary model of emotional reactions to the disclosure of a dementia diagnosis through the reports of the person diagnosed and their primary caregiver. As a theory-building strategy, they have used language consistent with other research conducted in the diagnostic phase and later as the disease progresses. The pre-diagnostic phase has been found to last from 1 to 3 years (Leung et al., 2011), which means that the caregivers have often been providing support for a considerable time before the diagnosis. In an Australian study that mapped out the personal and diagnostic events prior to the dementia diagnosis (Speechly, Bridges-Webb, & Passmore, 2008), the average “firm diagnosis” was 3.1 years after the first symptoms were noticed and a year after the person needed help with daily activities. Prior to diagnosis, families have often had a substantial amount of time living with the cognitive and behavioral aspects of dementia and uncertainty related to the cause of these symptoms, leading to perceptions of burden and distress when they arrive on the diagnosis day.

The Present Study

On the basis of literature from urban-based studies, we predicted that rural women would report more caregiver burden and severity of distress than rural men on the day of diagnosis (measured prior to the communication of the diagnosis to the family). Based on our univariate results and clinical experience, we further hypothesized a multivariate causal model of determinants of burden and severity of distress, controlling for the immediate pre-diagnosis stress which was constant over the entire sample. We examined the predicted gender effect by testing the model for the full sample, with gender of caregiver as an exogenous variable, and further testing by separating women from men caregivers. The exogenous variables in the model were as follows: severity of dementia-related behavior domains (agitation/aggression, apathy/indifference, disinhibition, psychotic symptoms, and sleep/nighttime behavior disorders), caregiver relationship (spouse vs. adult child caregivers), and functional assessment. Endogenous variables were caregiver distress linked to each of the five behavior domains.

Method

As part of the research program associated with the Rural and Remote Memory Clinic (RRMC), we conducted a two-phase analysis to improve our understanding of distress in rural family caregivers. First, we compared the four caregiver groups (wives, husbands, daughters, sons) on caregiver burden and severity of distress. Then we developed a causal model for these effects and tested it using structural equation modeling (SEM). Ethics approval for this study was obtained from the University of Saskatchewan Research Ethics Board.

Participants

All patients who were referred to the RRMC with their families (2004-2012, the clinic’s sixth data release) were eligible for inclusion. From a possible 363 caregivers, 231 family caregivers were selected. Inclusion criteria were as follows: (a) family relationship to the care recipient (spouse or adult child); (b) care recipient with diagnosis, based on interdisciplinary information, of mild cognitive impairment or dementia at the end of the initial in-person visit to the RRMC; (c) primary caregiver as designated by the family member who signed the informed consent to complete the family caregiver questionnaire; and (d) rural residence of care recipient, defined as living more than 100 km from major cities with tertiary care centers in the province of Saskatchewan, Canada. The primary caregiver sample included 86 wives, 75 daughters, 47 husbands, and 23 sons, who provided family dementia care in rural settings. For this analysis, we excluded data from caregivers with less commonly reported relationships, such as other relatives, friends, and unmarried partners.

Instruments

Clinical Dementia Rating (CDR)

The CDR (Morris, 1993) is a standardized clinician-based rating scale focusing on impairments in memory with secondary foci on rated impairments in instrumental and basic activities of daily living. CDR scores of 0 are consistent with no cognitive impairment, 0.5 is typical for pre-clinical or mild cognitive impairment, and scores of 1.0 are typical for persons with dementia. Higher scores (up to 3) represent more impairment. In the present study, a registered psychologist coded the CDR through an assessment of six areas of functioning: memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal care. These domains are referred to as “boxes” in the CDR. Summing the box scores of the CDR (CDR-SOB) provided a more detailed quantitative measure of global dementia severity, which is more sensitive than the traditional CDR at detecting changes in dementia severity over time. CDR-SOB scores range from 0 to 18, with larger scores suggesting severe impairment in all of the 6 domains assessed by the CDR. Criterion validity has been established for the individual domains and the global CDR (Morris, 1997). Although the CDR was developed for Alzheimer disease, the CDR-SOB score provides a global indicator of care needs over severity of dementia and has had neuropathological validation for the presence or absence of dementia (Morris, 1997).

Zarit Burden (ZB) Scale

The self-identified primary caregiver completed a family caregiver questionnaire, which included the short version (Bédard et al., 2001) of the ZB Scale (Zarit, Orr, & Zarit, 1985), at each clinic day over the three time intervals (diagnosis, 1-year follow-up, 2-year follow-up). This 12-item summated Likert-type scale had 5 response options (never, rarely, sometimes, quite frequently, nearly always) that were scored from 0 to 4. Items include a range of feelings related to taking care of the family member with dementia. The summation gives a total burden score from 0 to 48, with higher values representing increased burden. Although the ZB was originally used as an interview, we have used it here in the format of a self-report questionnaire, which was completed by the caregiver at the clinic. O’Rourke and Tuokko (2003) have provided psychometric support for the equivalency of the short version of the ZB versus the long version.

Brief Symptom Inventory (BSI)

This self-report scale (Derogatis & Melisaratos, 1983) was also embedded in the family caregiver questionnaire that primary caregivers completed at each of the three time intervals (initial visit and two annual follow-up visits). The BSI is a 53-item summated Likert-type scale with 5 response options (not at all, a little bit, moderately, quite a bit, extremely) that are scored from 0 to 4. The BSI has 9 subscales that measure symptom dimensions: depression, anxiety, hostility, somatization, interpersonal sensitivity, obsessive-compulsive, phobic anxiety, paranoid ideation, and psychoticism. The measure used in this study was the Global Severity Index (GSI), which was the sum of all responses divided by the number of items completed. Derogatis and Melisaratos (1983) provide evidence of good reliability (test–retest and internal consistency) and validity (convergent and construct) of this instrument and they recommend the GSI in relation to other BSI scoring methods as the “single best indicator of current distress levels” (p. 597).

Neuropsychiatric Inventory (NPI)

The NPI was developed to identify the presence of psychopathology in patients with brain disorders. It includes 12 domains of behavior known to occur in different types of dementia. The original NPI included 10 behavior content domains: delusions, hallucinations, dysphoria, anxiety, agitation/aggression, euphoria, disinhibition, irritability/lability, apathy, and aberrant motor activity (Cummings et al., 1994). Later, two additional behavior domains were added: nighttime sleep disturbances, and changes in appetite and eating behavior (Cummings, 1997). Content validity was established through a Delphi panel of national and international experts. Evaluation of concurrent validity was only possible with behavior domains that were part of other scales; some of the NPI domains (e.g., apathy and disinhibition) did not appear on other scales. The NPI was designed to be used in a structured interview with a caregiver. Each domain is rated on a 5-point scale for severity (1 = mild; 5 = severe) and associated caregiver distress (0 = not distressing at all; 5 = extremely distressing). Interrater reliability based on two raters, who observed and scored the same interviews, was high (89.4%-100% agreement) across 10 behavior domains (Cummings et al., 1994). Correlation conducted for test–retest reliability (3 week interval) were more variable in magnitude (r = .51-.98 for frequency; r = .51-.87 for severity of caregiver distress). However, the low scores were for behaviors that are known to be intermittent (e.g., agitation/aggression, anxiety, and irritability/lability).

For the present analysis, we selected six of the behavior domains for inclusion in our model building: agitation/aggression, apathy/indifference, disinhibition, sleep and nighttime behavior, delusions, and hallucinations. Based on the judgment of the clinical research team, these behaviors were considered to be most relevant to our sample, given the range of diagnoses and the different behavior profiles associated with different types of dementia. Since Cummings et al. (1994) found a low base rate of delusions and hallucinations, we collapsed these behaviors into a single item that we called “psychotic symptoms.” Cummings (1997) found that the NPI profile varied between patients with Alzheimer disease and frontotemporal dementia (e.g., the latter had more apathy and disinhibition). The NPI was completed as part of the baseline family caregiver questionnaire that was completed by all primary caregivers in our sample on the morning of the diagnosis day.

Functional Activities Questionnaire (FAQ)

The primary family caregiver completed the 10-item FAQ as part of the family caregiver questionnaire. Pfeffer, Kurosaki, Harrah, Chance, and Filos (1982) developed this measure of social function for community-dwelling older adults. The caregiver rates each of the 10 items from 0 (normal) to 3 (dependent) and the summated scores range from 0 to 30. The cutpoint of 9 (dependent in 3 or more activities) indicates impaired function and possible cognitive impairment. Concurrent validity of the FAQ was determined by comparison with multiple indicators of function including the Instrumental Activities of Daily Living Scale (IADL) and neurologist ratings of residual function. The sensitivity of the FAQ in discriminating between people with dementia and normal individuals was higher (.85) than the IADL (.57), while the specificity of the FAQ was slightly lower (.81) than the IADL (.92). Interrater reliability was high between neurologists (r = .97) and between a neurologist and nurse (r = .80).

Procedure

The RRMC is a weekly diagnostic clinic for patients referred from rural or remote areas (Morgan et al, 2009). A pre-clinic assessment is done by telehealth and blood work is completed at that time so that the results are available by clinic day. Patients come to the clinic in a tertiary care center, accompanied by their family member(s), and spend a full day in assessments. In the morning, the interdisciplinary team interviews the patient and family. Then the patient has neuropsychological testing done (approximately 2 hr). During this time, the family member completes scales related to themselves (ZB, BSI) and ratings of the patient’s functional and behavioral status. At some time during the day, the patient has a neurological examination, a physical therapy examination, and a CT brain scan. At the end of the day, the patients and family members meet with the neurologist and neuropsychologist who provide the diagnosis and plan for management. Follow-up is done by telehealth (videoconferencing) from the tertiary center to the closest community that has telehealth available (Morgan et al, 2011).

Statistical Analysis

The initial univariate phase of analysis used a factorial design with a 2 × 2 analysis of variance (ANOVA). The two factors were caregiver gender (women or men) and caregiver relationship (spouse or adult child) to the care recipient who was diagnosed with dementia. Separate ANOVAs were conducted for each of the two outcomes: burden (ZB) and severity of distress (BSI).

In the second phase of analysis, a multivariate model was developed to improve our understanding of rural caregiver burden and severity of distress, based on the clinical experience of the research team and the literature. SEM was used (Hayduk, 1987) and the model was tested using LISREL 9.1 (Joreskog & Sorbom, 2013). The prediction was that goodness-of-fit statistics would support the model, as evident from a nonsignificant maximum likelihood ratio χ² when the covariances from the data were compared to the covariances implied by the estimates from the model as conceptualized. Concepts of interest that were measured on the diagnosis day, but not related to the outcomes (hence, not in the model) were as follows: rurality, use of services in the past 12 months, frequency of contact between caregiver and care recipient, severity of dementia (CDR), and gender of care recipient. The first two variables were excluded a priori, based on our previous research (O’Connell et al, 2013) and the low base rate of service use evident in frequency data. The latter three variables were part of the model using all the data, but were excluded based on a lack of relationship to the outcomes. We planned to examine the overall model, with gender as an exogenous variable, and then test the model separately for each gender.

Results

The characteristics of the care recipients in this analysis are outlined in Table 1. Of note is the large range of ages and the diverse types of dementia, which can be explained by the fact that the memory clinic tends to have referrals for cases that are difficult to diagnose.

Table 1.

Characteristics of Care Recipients, 2004-2012 (n = 231).

Characteristic	n	%
Age, in years
<50	5	2.2
50-59	18	7.8
60-69	45	19.5
70-79	98	42.4
≥80	65	28.1
Gender
Male	96	41.6
Female	135	58.4
Marital status
Married	151	65.4
Living common-law	4	1.7
Separated	2	0.9
Divorced	9	3.9
Widowed	52	22.5
Missing	13	5.6
Highest level of schooling completed
Less than high school graduation	123	53.2
High school graduation	26	11.3
Beyond high school graduation	68	29.4
Missing	14	6.1
Research diagnosis
Alzheimer’s disease	95	41.1
Mild cognitive impairment^a	47	20.3
Frontotemporal dementia^b	26	11.3
Vascular dementia^c	16	6.9
Lewy body dementia	12	5.2
Dementia due to other etiologies^d	35	15.2

Mild cognitive impairment (MCI) includes MCI, MCI due to other, MCI nonamnestic, and MCI amnestic diagnoses.

Frontotemporal dementia (FTD) includes FTD frontal variant, FTD semantic, and FTD progressive nonfluent.

Vascular dementia includes vascular dementia and vascular mild cognitive impairment.

Dementia due to other etiologies includes Parkinson’s disease, Huntington’s disease, normal pressure hydrocephalus, alcohol-induced persisting dementia, alcohol-induced amnestic disorder, and dementia due to medical conditions.

There was no overall difference in severity of dementia across caregiver groups at the outset of the study based on the CDR scale; F(3, 212) = 0.82, p = .49; η² = .011. Although the families entered the study at different times over an 8-year period, the care needs at baseline (day of diagnosis) were similar within each of the four subgroups of caregivers: wives, husbands, daughters, and sons. There was some missing data so this assessment was approximate. All measures were obtained within the same time frame on the diagnostic day. The means and standard deviations of burden and severity of distress across caregiver groups are found in Table 2. The factorial analysis is presented in Table 3. As predicted, there was a main effect for gender with both measures. As compared with men, women (wives + daughters) reported more caregiver burden, F(1, 221) = 14.33, p < .0001, and greater severity of distress, F(1, 212) = 16.47, p < .0001. In the comparison of type of relationship, there was no difference in reported caregiver burden (ZB), but spouses (husbands + wives) reported greater severity of distress (BSI) than children (sons + daughters), F(1, 193) = 6.14, p = .014. There were no significant interaction effects.

Table 2.

Mean Scores (Standard Deviation) on Self-Report Measures of Caregiver Burden (Zarit Burden Scale) and Distress (Brief Symptom Inventory, Global Severity Index T-Score) Based on Relationship to Memory Clinic Patient on Diagnosis Day.

	Burden (n = 225)		Distress (n = 216)
Relationship	n	M (SD)	n	M (SD)
Husband	47	9.68 (9.29)	45	49.44 (10.40)
Wife	85	15.52 (9.31)	83	54.23 (8.89)
Son	22	11.45 (7.33)	21	44.52 (10.57)
Daughter	71	15.61 (8.18)	67	51.86 (8.93)

Note. Total number of participants in this study was n = 231 at the initial clinic day. Zarit Burden Scores were missing for 6 participants and Brief Symptom Inventory Global Severity Index T-score was missing for 15 participants at the clinic day.

Table 3.

Analysis of Variance (ANOVA) Results for Caregiver Gender and Relationship Main Effects and Their Interaction at Initial Clinic Day When Family Member Diagnosed With Dementia.

	Burden (n = 225)				Distress (n = 216)
Source	df	F	p	η²	df	F	p	η²
Between subjects
Gender (G)	1	14.33**	<.0001	.061	1	16.47**	<.0001	.072
Relationship (R)	1	0.55	.457	.003	1	6.14*	.014	.028
G × R	1	0.47	.495	.002	1	0.89	.347	.004
Within-group
Error	221	(77.09)			212	(88.72)

Note. Values enclosed in parentheses represent mean square errors.

p < .05. **p < .01.

Figure 1 illustrates the model for women only, which was supported by the LISREL analysis. The overall model did not have a good fit and there were too few men to conduct SEM for the men alone. The model for women fits the data in that the observed covariances match the model-implied covariances. The residual covariances are small and nonsignificant (χ² = 60.197 with 45 df; p = .064) and the model has an acceptable goodness-of-fit index (0.87). The modification indices were mostly under 4.0, with a few slightly higher but none over 6.5. The expected change in coefficients was negligible so we concluded that a further change would not substantially improve the model fit. Given that each of the concepts in our model has a single indicator, we assigned meaning to the concepts by adopting the fixed measurement error variance procedure recommended by Hayduk (1987, p. 191). All the indicators were assigned 5% measurement error variance with the exception of Caregiver Relationship, which was assigned a 0.5% error variance. The half-double procedure was used to check the assigned error variance by rerunning the model at 2.5% and again at 10% to see if there were important changes in the estimates or if the goodness of fit is improved, which was not the case (Hayduk, 1987; Hayduk & Littvay, 2012).

Figure 1.

Model of rural dementia caregiver burden and severity of distress (women only: n = 161).

Caregiver relationship was included as one of the seven exogenous variables and the same pattern existed as for the ANOVA results. The spouse caregiver dummy variable was assigned a score of 1 and adult children were 2. The negative coefficient represents the greater severity of distress of spouses (lower assigned number) compared to adult children caregivers. The other exogenous variables all come from the caregiver questionnaire (five severity of behavior indicators from the Neuropsychiatric Index and one functional indicator [FAQ]). The five endogenous variables are the caregiver distress ratings related to each of the five behavior severity scores. Table 4 provides the covariance matrix below the diagonal with means along the bottom; the variance is on the diagonal and a correlation matrix is above the diagonal. The significance level of the causal relationships, represented by directional arrows (causal influence from one variable to another), can be found in Table 5. The coefficients on Figure 1 are all maximum likelihood coefficients, which are reported in Table 5 with significance levels. The parallel standardized estimates vary in magnitude from the maximum likelihood estimates, but all of the significance levels are identical.

Table 4.

Indicator Means, Correlations (Upper Right), and Covariances (Diagonal and Lower Left).

	Severity of agitation/aggression	Severity of apathy/indifference	Severity of disinhibition	Severity of psychotic symptoms	Severity of sleep and nighttime behavior	Caregiver relationship	FAQ	CD relationship: Agitation/aggression	CD relationship: Apathy/indifference	CD relationship: Disinhibition	CD Relationship: Psychotic symptoms	CD Relationship: Sleep and nighttime behavior	Severity of caregiver distress	Caregiver burden
Severity of agitation/aggression	0.962	0.20	0.38	0.33	0.33	−0.05	0.32	0.86	0.31	0.41	0.37	0.39	0.34	0.58
Severity of apathy/indifference	0.200	1.022	0.31	0.15	0.32	−0.19	0.32	0.17	0.78	0.27	0.15	0.32	0.25	0.23
Severity of disinhibition	0.380	0.323	1.030	0.42	0.28	0.10	0.25	0.41	0.33	0.92	0.42	0.32	0.28	0.43
Severity of psychotic symptoms	0.500	0.233	0.645	2.331	0.30	0.01	0.36	0.39	0.19	0.43	0.93	0.39	0.13	0.38
Severity of sleep and nighttime behavior	0.361	0.359	0.314	0.508	1.247	−0.07	0.30	0.28	0.30	0.29	0.30	0.80	0.24	0.24
Caregiver relationship	−0.026	−0.098	0.051	0.005	−0.040	0.250	0.07	0.00	−0.24	0.03	0.02	−0.15	−0.13	0.00
FAQ	2.608	2.702	2.141	4.478	2.803	0.306	71.093	0.25	0.25	0.18	0.35	0.32	0.17	0.34
CD relationship: Agitation/aggression	1.396	0.288	0.682	0.998	0.520	−0.004	3.470	2.701	0.31	0.44	0.45	0.36	0.26	0.57
CD relationship: Apathy/indifference	0.428	1.105	0.456	0.410	0.460	−0.167	2.880	0.698	1.928	0.32	0.22	0.39	0.33	0.35
CD relationship: Disinhibition	0.612	0.411	1.394	0.997	0.480	0.023	2.239	1.086	0.667	2.245	0.43	0.37	0.30	0.48
CD relationship: Psychotic symptoms	0.889	0.376	1.014	3.451	0.819	0.025	7.053	1.778	0.721	1.549	5.818	0.43	0.13	0.39
CD relationship: Sleep and nighttime behavior	0.566	0.474	0.479	0.893	1.327	−0.111	3.916	0.869	0.804	0.812	1.542	2.171	0.39	0.39
Severity of caregiver distress	2.912	2.296	2.589	1.870	2.393	−0.560	13.044	3.835	4.075	4.091	2.801	5.229	80.172	0.48
Caregiver burden	4.988	2.094	3.860	5.109	2.356	0.021	25.234	8.279	4.375	6.435	8.450	5.141	38.146	38.146
M	0.761	1.129	0.665	0.877	0.987	1.463	14.222	1.212	1.291	0.907	1.367	1.053	53.119	15.561

Note. Ns range from 150 to 161. FAQ = Functional Activities Questionnaire; CD = caregiver distress.

Table 5.

Maximum Likelihood Estimates and Standardized Estimates for the Effects in the Rural Caregiver Burden and Severity of Distress Model (Women Only; n = 161).

Effect			R² or blocked-error-R^2a
To From	Maximum likelihood estimate	Standardized estimate	R² or blocked-error-R^2a
Caregiver distress relationship: Agitation/aggression			.84
Caregiver distress relationship: Psychotic symptoms	0.081**	0.118**
Severity of agitation/aggression	1.448**	0.865**
Caregiver distress relationship: Apathy/indifference			.72
Caregiver distress relationship: Agitation/aggression	0.152**	0.180**
Severity of apathy/indifference	1.091**	0.792**
Caregiver distress relationship: Disinhibition			.93
Severity of disinhibition	1.426**	0.965**
Caregiver distress relationship: Psychotic symptoms			.97
Caregiver distress relationship: Agitation/aggression	0.102**	0.070**
Severity of psychotic symptoms	1.510**	0.956**
Caregiver distress relationship: Sleep and nighttime behavior			.77
Caregiver distress relationship: Apathy/indifference	0.144**	0.136**
Caregiver distress relationship: Psychotic symptoms	0.108**	0.176**
Severity of sleep and nighttime behavior	0.988**	0.748**
Severity of caregiver distress			.35
Caregiver distress relationship: Agitation/aggression	−0.363	−0.067
Caregiver distress relationship: Apathy/indifference	0.568	0.089
Caregiver distress relationship: Disinhibition	0.502	0.084
Caregiver distress relationship: Psychotic symptoms	−0.671*	−0.181*
Caregiver distress relationship: Sleep and nighttime behavior	1.524**	0.252**
Caregiver burden	0.470	0.461
Caregiver relationship	−1.322	−0.076
FAQ (functional assessment)	−0.019	−0.018
Caregiver burden			.49
Caregiver distress relationship: Agitation/aggression	2.158**	0.405**
Caregiver distress relationship: Apathy/indifference	0.651	0.103
Caregiver distress relationship: Disinhibition	1.245**	0.213**
Caregiver distress relationship: Psychotic symptoms	0.112	0.031
Caregiver distress relationship: Sleep and nighttime behavior	0.450	0.076
Caregiver relationship	0.421	0.025
FAQ (functional assessment)	0.142*	0.137*

Note. The covariances among the exogenous variables are not presented since they are approximated by the corresponding data covariances.

The Blocked-Error-R² is explained by Hayduk (2006).

Coefficient exceeds 1.7 times its standard error, one-tailed p < .05.

Coefficient exceeds twice its standard error, two-tailed p < .05.

Discussion

This study examined the effects of gender and generational relationship on rural caregiver burden and severity of distress in caring for a family member with dementia. The finding of a gender effect has been a strong theme in the literature and this rural-based study lends further support that women report more burden and severity of distress than men in similar circumstances. In a secondary analysis of the Canadian Studies of Health and Aging, Gallicchio and colleagues (2002), also found a gender effect on burden using Zarit’s Burden Scale as a measure. While their study was urban-based, they found that women had higher odds than men of reporting burden and higher odds of burden were related to behavior concerns, as measured by the Dementia Behavior Disturbance scale, and perceptions of poor health reported by the caregivers. Studies such as this formed the rationale for our focus on the caregiver reports of behavior change and their reaction to these changes in the present study.

Our SEM model for rural dementia caregiver burden and severity of distress should be considered preliminary for two reasons. First, although it is supported by chi-square goodness of fit statistics (p > .05), which is interpreted as no difference between the available data and the theoretical model, the argument for causality is likely missing some important determinants. For example, while we examined the impact of behaviors that are present, we did not address some of the losses in the caregiver–care recipient relationship (Adams et al., 2008). Second, the data came from an existing data set and the retrospective aspect of the study again suggests that there may be other factors operating that we did not measure that are relevant to the rural situation. However, this is an important first step in understanding the complexity of burden and severity distress in a rural sample. One of the variables that was removed from the model was use of services, which had such a low base rate (low variance) that it hardly qualified as a variable. Likewise, our measure of rurality was not sensitive to the issues of access to services that we see in practice (O’Connell et al, 2013). We were not able to develop a contrasting model for men because the sample of males is small at this time, but clearly the addition of male caregivers in the model changed the parameters, which is why the model is only presented for female family caregivers. These data suggest determinants of burden and severity of distress are different for male caregivers, and future research on male caregivers in rural settings is required.

In a longitudinal study comparing rural and urban carers in Wales, Tommis and colleagues (2007) found gender differences that could be attributed to living in rural settings. Men who cared for a person with dementia or stroke in urban settings reported better mental health than comparable men in rural settings or women in either rural or urban settings. Based on their literature review of men as caregivers, Baker and Robertson (2008) concluded that less is known about husbands in a caregiving role than wives. They note that researchers have reported recruitment challenges with men who may not use support services, effectively limiting our knowledge about men who have difficulty coping with dementia care. Underlying differential caregiver distress may be gender differences in use of adaptive versus maladaptive coping (e.g., Lutzky & Knight, 1994). On the other hand, gender differences in coping are not consistent across studies (DeVries, Hamilton, Lovett, & Gallagher-Thompson, 1997). Bédard and colleagues (2005) found that the relationship between gender and burden is complex and is dependent on the domain of burden in question. Within the female subgroup, wives reported more role burden (demands of the role), whereas daughters reported more personal burden (difficulty fulfilling their role due to competing demands). Between genders, difficult behaviors concerned women more than men only if they were physically threatening or difficult to manage.

Winter, Gitlin, and Dennis (2011) found that the quality of the relationship between individuals with dementia and family caregivers affected the desire to institutionalize, with male caregivers who had a lower quality of prior relationship with the person with dementia, being more likely to want to institutionalize their family member with dementia than female caregivers with a lower quality premorbid relationship with the care recipient.

In our study on diagnosis day, the type of relationship was statistically significant for severity of distress, with spouses reporting greater severity of distress than adult children as primary caregivers prior to the diagnosis (Table 2). Clearly, some of this distress differential will be related to the fact that the spouses are cohabitating with the person being diagnosed for dementia, while most of the adult children are not living with their parent(s) as they might in other cultures. In a Japanese study (Nagatomo et al., 1999), spouses reported greater mental fatigue than offspring, but there was no difference in reported caregiver burden. Barber and Pasley (1995) found that there were different results when gender and generation were considered simultaneously in the same study as we have done here. Similar to our gender effect for burden, Barber and Pasley found that daughters reported the greatest relationship strain, followed by wives. They noted that spouses do not perceive their role as caregiver in the same way as filial caregivers because spouses see this as a continuation of their marital role. In our study, spouses reported more severity of distress than adult children caregivers, but there was no difference in perception of burden based on type of relationship (Table 3). However, examination of the pattern of means for burden (Table 2) suggests that further exploration is needed because husbands reported the lowest burden, while wives and daughters were almost identical. This pattern of means underscores the importance of continuing study with each of these family caregiver groups to understand the unique causal world that comprises their burden and severity of distress. Future studies should chart the trajectory of distress over time in these caregiver groups, with a view to developing interventions relevant to rural settings.

Another variable that we removed from the model, but that merits further study, is the gender of the parent within the group of adult child caregivers. Daughters have unique relationships with their mothers (Nagatomo et al., 1999; Perry, 2004) as compared with their fathers. At the same time, the behavior profile of men with dementia (i.e., their fathers) may differ from women. Cummings (1997) found a gender difference in behavioral expression related to Alzheimer’s disease, with the agitation/aggression domain more common in men.

The use of stress and coping theory (Lazarus & Folkman, 1984) was applicable to the measures of distress, particularly burden. However, although this framework was useful for cross-sectional analyses, we think that a blend of symbolic interactionism, identity theory (see Neufeld & Harrison, 2003), and feminist theory (Brewer, 2001) may be more useful for understanding changes in caregiver distress over time (i.e., gendered care, based on gender expectations) that evolves in the interaction context with the care recipient and other members of the family. We need new methods of studying coping that are specific to the daily caregiving experiences within the caring context (Kneebone & Martin, 2003).

Another issue with the study of burden is that it does not address the positive aspects of caregiving (Peacock et al., 2010). Van Ess Coeling and Biordi (2000) examined positive “work strategies” that were useful for caregiver–care receiver dyads: delegating, developing competence (self-efficacy) and positive problem orientation. These three strategies were associated with positive caregiver outcomes (mood and well-being), but they were more positive for spouses than nonspouses. The developing competence strategy was more positive for male caregivers than females, giving support for potentially differential caregiver subgroup benefit from an educational intervention. While the literature reveals many interventions that have been tested (e.g., review by Yap et al., 2005), the task for rural researchers is to come to a better understanding of the issues for specific family caregiver groups prior to intervention research. The functional assessment exogenous variable in the model for women was significantly related to burden but not to severity of distress, which can be explained by the nature of the two measures. Caregiver burden (ZB) is dementia-specific, while the GSI of the BSI is a general measure of psychological symptoms of the caregiver, but none of the items relate specifically to dementia. By contrast, all of the selected behavior domains related to both outcome measures, which suggests that it is useful to continue to use them together.

Limitations

The analysis reported here used an existing clinic database, so it has the limitations of a retrospective study. No specific gender questions were asked of the caregiver participants and the categorization for analysis was derived from their identification of their relationship to the patient, who they had accompanied to the clinic for diagnosis. The selection of behavior domains from the NPI for inclusion in the model was based on clinical experience and the literature, but other behavior domains may be more relevant for male caregivers. One of the features of this memory clinic was that referrals from rural and remote regions were often cases that were difficult to diagnose. This affects the representativeness of the data since there were proportionately fewer cases of Alzheimer disease in our sample than the general population, since family physicians in rural and remote areas were often diagnosing these cases without the assistance of the memory clinic. In future prospective studies, family caregivers could be asked about their distress and whether it is related in any way to rural living.

Conclusion

As hypothesized, rural women reported more caregiver burden and severity of distress than men. The development and testing of a model in this study provided greater understanding of the sources of burden and distress in women, while raising new questions about the distress of men in caregiving roles in rural settings. Future prospective research could combine interviews with standardized measures to gain an understanding of the differential needs of the four subgroups of caregivers. Further qualitative research could be done with caregivers who do not fit the categories examined in this study (e.g., friends, in-laws, unmarried partners, both same sex and opposite sex). In this study, the causal model developed for women indicated that distress related to dementia-specific behaviors explains both global distress and burden, while functional dependency is related to caregiver burden only. The lack of support for the model when men and women caregivers were both included provides evidence that gender differences exist. When larger samples of male caregivers are available, similar SEM modeling could be done to improve our understanding of their burden and severity of distress.

Footnotes

Acknowledgements

The authors extend their thanks to the family members who volunteered their time to complete the questionnaires on clinic days.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Canadian Institutes of Health Research (ACH #93185) and the Saskatchewan Health Research Foundation (#2104).

Author Biographies

Norma J. Stewart, PhD, RN, is a Professor in the College of Nursing at the University of Saskatchewan, Saskatoon, Canada. Her research interests are in rural dementia care and rural health services.

Debra G. Morgan, PhD, RN, is a Professor at the Canadian Centre for Health & Safety in Agriculture, University of Saskatchewan. She holds a Chair in Rural Health Delivery and a CIHR Applied Chair in Health Services & Policy Research (2009-2014) titled “Healthcare Delivery across the Continuum for Rural and Remote Seniors with Dementia.” She was the Principal Investigator of a New Emerging Team funded by CIHR and provincial partners (2003-2009), which provided the foundation for the Rural and Remote Memory Clinic.

Chandima P. Karunanayake, PhD, PStat, is a Professional Research Associate at the Canadian Centre for Health and Safety in Agriculture, University of Saskatchewan. As a Biostatistician, she has experience on applied linear models, longitudinal data analysis and complex survey data.

Joseph P. Wickenhauser, MA, is a Research Coordinator with the College of Nursing at the University of Saskatchewan. His master’s research focused on sexual and gender diversity among older adults in rural areas.

Allison Cammer, MSc, RD, is a Professional Research Associate in the Canadian Centre for Health and Safety in Agriculture, Department of Medicine, and a doctoral student in the College of Pharmacy and Nutrition, University of Saskatchewan. She works as a Registered Dietitian at the Rural and Remote Memory Clinic and her research focuses on rural aging and dementia.

Duane Minish is a Research Assistant at the Canadian Centre for Health & Safety in Agriculture. He is also a student completing his undergraduate degree in the Department of Psychology at the University of Saskatchewan.

Megan E. O’Connell, PhD, RD Psych, is as Assistant Professor in the Department of Psychology at the University of Saskatchewan and a Clinical Psychologist working on the neuropsychology team in the Rural and Remote Memory Clinic, Saskatoon, SK. She researches issues related to rural dementia care; including use of telehealth videoconferencing for rural health care delivery, use of technology to support rural dementia care, neuropsychological assessments, psychometric theory, cognitive rehabilitation, and caregiver support.

Leslie A. Hayduk, PhD, is a Professor in the Department of Sociology at the University of Alberta. He has authored several articles and two books on structural equation modeling methodology. His other publications are eclectic and have addressed: children’s early school experiences, personal space, sexual orientation, knowledge transfer in nursing contexts, natural killer cell activity, and several other topics.

References

Adams

K. B.

McClendon

M. J.

Smyth

K. A.

(2008). Personal losses and relationship quality in dementia caregiving. Dementia, 7, 301-319. doi:10.1177/1471301208093286

Adams

K. B.

Smyth

K. A.

McClendon

M. J.

(2005). Psychosocial resources as moderators of the impact of spousal dementia caregiving on depression. The Journal of Applied Gerontology, 24, 475-489.

Aminzadeh

Byszewski

Molnar

F. J.

Eisner

(2007). Emotional impact of dementia diagnosis: Exploring persons with dementia and caregivers’ perspectives. Aging & Mental Health, 11, 281-290. doi:10.1080/13607860600963695

Baker

K. L.

Robertson

(2008). Coping with caring for someone with dementia: Reviewing the literature about men. Aging & Mental Health, 12, 413-422.

Barber

C. E.

Pasley

B. K.

(1995). Family care of Alzheimer’s patients: The role of gender and generational relationship on caregiver outcomes. Journal of Applied Gerontology, 14, 172-192.

Bédard

Koivuranta

Stuckey

(2004). Health impact on caregivers of providing informal care to a cognitively impaired older adult: Rural versus urban settings. Canadian Journal of Rural Medicine, 9(1), 15-23.

Bédard

Kuzik

Chambers

Molloy

D. W.

Dubois

Lever

J. A.

(2005). Understanding burden differences between men and women caregivers: The contribution of care-recipient problem behaviors. International Psychogeriatrics, 17, 99-118. doi:10.1017/S1041610204000857

Bédard

Molloy

D. W.

Squire

Dubois

Lever

J. A.

O’Donnell

(2001). The Zarit Burden Interview: A new short version and screening version. The Gerontologist, 41, 652-657.

Black

Almeida

O. P.

(2004). A systematic review of the association between the behavioral and psychological symptoms of dementia and burden of care. International Psychogeriatrics, 16, 295-315.

10.

Blumer

(1969). Symbolic interactionism. Englewood Cliffs, NJ: Prentice Hall.

11.

Braun

Scholz

Bailey

Perren

Hornung

Martin

(2009). Dementia caregiving in spousal relationships: A dyadic perspective. Aging & Mental Health, 13, 426-436. doi:10.1080/13607860902879441

12.

Brewer

(2001). Gender socialization and the cultural construction of elder caregivers. Journal of Aging Studies, 15, 217-235.

13.

Brody

E. M.

(1990). Women in the middle: Their parent-care years. New York, NY: Springer.

14.

Byrd

Spencer

S. M.

Goins

R. T.

(2011). Differences in caregiving: Does residence matter? Journal of Applied Gerontology, 30, 407-421. doi:10.1177/0733464810371098

15.

Calasanti

Bowen

M. E.

(2006). Spousal caregiving and crossing gender boundaries: Maintaining gendered identities. Journal of Aging Studies, 20, 253-263.

16.

Calasanti

King

(2007). Taking “women’s work” “like a man”: Husbands’ experiences of care work. The Gerontologist, 47, 516-527.

17.

Chadiha

L. A.

Feld

Rafferty

(2011). Likelihood of African American primary caregivers and care recipients receiving assistance from secondary caregivers: A rural-urban comparison. Journal of Applied Gerontology, 30, 422-442. doi:10.1177/0733464810371098

18.

Cummings

J. L.

(1997). The Neuropsychiatric Inventory: Assessing psychopathology in dementia patients. Neurology, 44(Suppl. 6), S10-S16.

19.

Cummings

J. L.

Mega

Gray

Rosenberg-Thompson

Carusi

D. A.

Gornbein

(1994). The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology, 44, 2308-2314.

20.

Depp

Sorocco

Kasl-Godley

Thompson

Rabinowitz

Gallagher-Thompson

(2005). Caregiver self-efficacy, ethnicity, and kinship differences in dementia caregivers. American Journal of Geriatric Psychiatry, 13, 787-794.

21.

Derogatis

L. R.

Melisaratos

(1983). The Brief symptom Inventory: An introductory report. Psychological Medicine, 13, 595-605.

22.

DeVries

H. M.

Hamilton

D. W.

Lovett

Gallagher-Thompson

(1997). Patterns of coping preferences for male and female caregivers of frail older adults. Psychology and Aging, 12, 263-267.

23.

Etters

Goodall

Harrison

B. E.

(2008). Caregiver burden among dementia patient caregivers: A review of the literature. Journal of the American Academy of Nurse Practitioner, 20, 423-428.

24.

Gallicchio

Siddiqi

Langenberg

Baumgarten

(2002). Gender differences in burden and depression among informal caregivers of demented elders in the community. International Journal of Geriatric Psychiatry, 17, 154-163.

25.

Gerson

J. M.

Peiss

(1985). Boundaries, negotiation, consciousness: Reconceptualizing gender relations. Social Problems, 32, 317-331.

26.

Goins

R. T.

Spencer

S. M.

Byrd

J. C.

(2009). Research on rural caregiving: A literature review. Journal of Applied Gerontology, 28, 139-170. doi:10.1177/0733464808326294

27.

Harris

P. B.

Bichler

(1997). Men giving care: Reflections of husbands and sons. New York, NY: Garland Publishing.

28.

Hayduk

L. A.

(1987). Structural equation modeling with LISREL. Baltimore, MD: The Johns Hopkins University Press.

29.

Hayduk

L. A.

(2006). Blocked-Error-R2: A conceptually improved definition of the proportion of explained variance in model containing loops or correlated residuals. Quality & Quantity, 40, 629-649.

30.

Hayduk

L.A.

Littvay

(2012). Should researchers use single indicators, best indicators, or multiple indicators in structural equation models? BMC Medical Research Methodology, 12, 159.

31.

Hogg

M. A.

Terry

D. J.

White

K. M.

(1995). A tale of two theories: A critical comparison of identity theory with social identity theory. Social Psychology Quarterly, 58, 255-269.

32.

Innes

Morgan

Kosteniuk

(2011). Dementia care in rural and remote settings: A systematic review of informal/family caregiving. Maturitas, 68, 34-46.

33.

Jarvik

L. F.

Blazer

(2005). Children of Alzheimer patients: An overview. Journal of Geriatric Psychiatry and Neurology, 18, 181-186. doi:10.1177/0891988705281859

34.

Joreskog

K. G.

Sorbom

(2013). LISREL 9.10. Chicago, IL: Scientific Software International.

35.

Kneebone

I. I.

Martin

P. R.

(2003). Coping and caregivers of people with dementia. British Journal of Health Psychology, 8, 1-17.

36.

Kramer

(1997). Differential predictors of strain and gain among husbands caring for wives with dementia. The Gerontologist, 37, 239-249.

37.

Lazarus

R. S.

Folkman

(1984). Stress, appraisal, and coping. New York, NY: Springer.

38.

Leung

K. K.

Finlay

Silvius

J. L.

Koehn

McCleary

Cohen

C. A.

. . .Drummond

(2011). Pathways to diagnosis: Exploring the experiences of problem recognition and obtaining a dementia diagnosis among Anglo-Canadians. Health and Social Care in the Community, 19, 372-381. doi:10.1111/j.1365-2524.2010.00982.x

39.

Little

Panelli

(2003). Gender research in rural geography. Gender, Place and Culture: A Journal of Feminist Geography, 10, 281-289.

40.

Loos

Bowd

(1997). Caregivers of persons with Alzheimer’s disease: Some neglected implications of the experience of personal loss and grief. Death Studies, 21, 501-514.

41.

Lutzky

S. M.

Knight

B. G.

(1994). Explaining gender differences in caregiver distress: The roles of emotional attentiveness and coping styles. Psychology and Aging, 9, 513-519.

42.

McDonnell

Ryan

(2013). Male caregiving in dementia: A review and commentary. Dementia, 12, 238-250. doi:10.1177/1471301211421235

43.

Morgan

D. G.

Crossley

Kirk

D’Arcy

Stewart

Biem

. . .McBain

(2009). Improving access to dementia care: Development and evaluation of a rural and remote memory clinic. Aging & Mental Health, 13, 17-30.

44.

Morgan

D. G.

Crossley

Kirk

McBain

Stewart

N. J.

D’Arcy

. . .Basran

(2011). Evaluation of telehealth for pre-clinic assessment and follow-up in an interprofessional rural and remote memory clinic. Journal of Applied Gerontology, 30, 304-331.

45.

Morris

J. C.

(1993). The Clinical Dementia Rating (CDR): Current version and scoring rules. Neurology, 43, 2412-2414.

46.

Morris

J. C.

(1997). Clinical dementia rating: A reliable and valid diagnostic and staging measure for dementia of the Alzheimer type. International Psychogeriatrics, 9, 173-176.

47.

Nagatomo

Akasaki

Uchida

Tominaga

Hashiguchi

Takigawa

(1999). Gender of demented patients and specific family relationship of caregiver to patients influence mental fatigue and burdens on relatives as caregivers. International Journal of Geriatric Psychiatry, 14, 618-625.

48.

Neufeld

Harrison

M. J.

(2003). Unfulfilled expectations and negative interactions: Nonsupport in the relationships of women caregivers. Journal of Advanced Nursing, 41, 323-331.

49.

O’Connell

M. E.

Germaine

Burton

Stewart

Morgan

D. G.

(2013). Degree of rurality is not related to dementia caregiver distress, burden, and coping in a predominantly rural sample. Journal of Applied Gerontology, 32, 1015-1029. doi:10.1177/0733464812450071

50.

O’Rourke

Tuokko

H. A.

(2003). Psychometric properties of an abridged version of the Zarit Burden Interview within a representative Canadian caregiver sample. The Gerontologist, 43, 121-127.

51.

Peacock

Forbes

Markle-Reid

Hawranik

Morgan

Jansen

. . .Henderson

S. R.

(2010). The positive aspects of the caregiving journey with dementia: Using a strengths-based perspective to reveal opportunities. Journal of Applied Gerontology, 29, 640-659.

52.

Perry

(2004). Daughters giving care to mothers who have dementia: Mastering the 3 R’s of (Re)Calling, (Re)Learning, and (Re)Adjusting. Journal of Family Nursing, 10, 50-69.

53.

Pfeffer

R. I.

Kurosaki

T. T.

Harrah

C. H.

Chance

J. M.

Filos

(1982). Measurement of functional activities in older adults in the community. Journal of Gerontology, 37(3), 323-329.

54.

Quinn

Clare

Woods

(2009). The impact of the quality of relationship on the experiences and well-being of caregivers of people with dementia: A systematic review. Aging & Mental Health, 13, 143-154. doi:10.1080/13607860802459799

55.

Sanders

(2007). Experiences of rural male caregivers of older adults with their informal support networks. Journal of Gerontological Social Work, 49(4), 97-115. doi:10.1300/J083v49n04_06

56.

Savundranayagam

M. Y.

Montgomery

R. J. V.

Kosloski

(2011). A dimensional analysis of caregiver burden among spouses and adult children. The Gerontologist, 51, 321-351.

57.

Schreiner

A. S.

Morimoto

Arai

Zarit

(2006). Assessing family caregiver’s mental health using a statistically derived cut-off score for the Zarit Burden Interview. Aging & Mental Health, 10, 107-111.

58.

Shaw

W. S.

Patterson

T. L.

Semple

S. J.

Dimsdale

J. E.

Ziegler

M. G.

Grant

(2003). Emotional expressiveness, hostility and blood pressure in a longitudinal cohort of Azheimer caregivers. Journal of Psychosomatic Research, 54, 293-302.

59.

Speechly

C. M.

Bridges-Webb

Passmore

(2008). The pathway to dementia diagnosis. Medical Journal of Australia, 189(9), 487-489.

60.

Statistics Canada. (2012). The Canadian population in 2011: Age and sex (Report No. 98-311-X2011001). Ottawa, Ontario, Canada: Author.

61.

Tommis

Seddon

Woods

Robinson

C. A.

Reeves

Russell

I. T.

(2007). Rural-urban differences in the effects on mental well-being of caring for people with stroke or dementia. Aging & Mental Health, 11, 743-750. doi:10.1080/13607860701365972

62.

Van Ess Coeling

Biordi

(2000). The relationship between work strategies used by informal caregivers and care receivers and clinical outcomes. Journal of Applied Gerontology, 19, 264-283. doi:10.1177/073346480001900302

63.

Ward-Griffin

Oudshoorn

Clark

Bol

(2007). Mother-adult daughter relationships in dementia care: A critical analysis. Journal of Family Nursing, 13, 13-32. doi:10.1177/1074840706297424

64.

Wilcox

O’Sullivan

King

A. C.

(2001). Caregiver coping strategies. Clinical Gerontologist, 23(1), 81-97. doi:10.1300/J018v23n01_08

65.

Wilken

C. S.

Altergott

Sandberg

(1996). Spouses’ self-perceptions as caregivers: The influence of feminine and masculine sex-role orientation on caring for confused and non-confused partners. American Journal of Alzheimer’s Disease & Other Dementias, 11, 37-42.

66.

Winter

Gitlin

L. N.

Dennis

(2011). Desire to institutionalize a relative with dementia: Quality of premorbid relationship and caregiver gender. Family Relations, 60, 221-230. doi:10.1111/j.1741-3729.2010.00644.x

67.

World Health Organization and Alzheimer’s Disease International. (2012). Dementia: A public health priority. Retrieved from http://www.who.int/mental_health/publications/dementia_report_2012/

68.

Yap

L. K. P.

Seow

C. C. D.

Henderson

L. M.

Goh

Y. N. J.

(2005). Family caregivers and caregiving in dementia. Reviews in Clinical Gerontology, 15, 263-271. doi:10.1017/S0959259806001900

69.

Zarit

S. H.

Orr

N. K.

Zarit

J. M.

(1985). The hidden victims of Alzheimer’s disease: Families under stress. New York, NY: New York University Press.