| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
University of North Carolina at Chapel Hill, Institute on Aging.
Address correspondence to Ishan C. Williams, University of Virginia School of Nursing, P.O. Box 800782, Charlottesville, VA 22903-0782. E-mail: icwilliams{at}virginia.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
As THE population continues to increase in longevity, the incidence of acquiring various diseases, such as Alzheimer's disease, is more prominent (Ferraro, 1997
; Hayflick, 2000), while the overall prevalence rates of dementia appear to be higher among Black Americans than for White Americans (Heyman et al., 1991). However, racial differences may vary by type of dementia and by measurement bias that may be present due to differences in education, socioeconomic status, or other cultural dimensions (Farrer, 2000; Froehlich, Bogardus, & Inouye, 2001).
Caregivers for older adults with dementia are often in the most stressful caregiving role, particularly when one considers the problematic behaviors that are typically present and often the most troubling to caregivers. Schulz, O'Brien, Bookwala, and Fleissner (1995) have noted, "dementia caregiving is perceived as posing some of the most extreme challenges faced by caregivers" (p. 171). The daily demands of providing such care can augment other problems that caregivers already face, such as having competing demands on their time (e.g., employment, children, spouse), limited resources (e.g., financial), and their own health problems (Ory, Yee, Tennstedt, & Schulz, 2000).
Given these challenges, providing care to an older adult with dementia may put certain caregivers at greater risk for negative emotional health (i.e., depressive symptoms) outcomes. Researchers found that Black caregivers were more likely to be in poorer health than their White counterparts (Dilworth-Anderson & Anderson, 1994), have limited economic resources (Hummer, 1996), and underutilize formal support to a greater extent (Hinrichsen & Ramirez, 1992; Tennstedt & Chang, 1998). In contrast, other studies have shown that Black caregivers report fewer burdens, less depression, and less stress related to caregiving than White caregivers (Dilworth-Anderson, Williams, & Gibson, 2002; Haley et al., 2004). Although previous research has made some attempt to disentangle the mechanisms and processes that affect emotional health, results have not been consistent because some studies do not have a large enough sample size to analyze why differences exist between the groups (Dilworth-Anderson et. al., 2002).
Researchers have illustrated that race often interacts with several variables known to affect well-being, which raises the importance of understanding how one's individual context can impact one's health outcomes (LaVeist, Arthur, Morgan, Plantholt, & Rubinstein, 2003). Due to their unique historical experience, Black caregivers may have the needed strengths and resources in dealing with the stresses of caregiving (e.g., use of extended family). In addition, Haley and colleagues (2004) found that Black caregivers showed more positive aspects of caregiving than White caregivers. Therefore, an understanding of the challenging caregiving role and the type of contextual resources caregivers bring to the role can provide insight into studying the emotional impact of dementia caregiving. The current study was fruitful in advancing knowledge on emotional health outcomes among Black and White caregivers. This study examined the ways that Black and White caregivers deal with the common stressors of caregiving and identified contextual factors that may be unique to Black and White caregivers.
Contextual Framework
Investigators in caregiving research have often used theoretical frameworks that are embedded in stress process models, which are useful in examining health outcomes among various types of caregivers. The present study used a contextual approach, which has shown utility in understanding diversity and caregiving (Dilworth-Anderson & Anderson, 1994). The contextual approach views "the mental health of caregivers as a function of the ecological niche that he or she occupies at any given time" (Dilworth-Anderson & Anderson, p.387; see Figure 1). In particular, understanding the influence of contextual factors (sociocultural, situational, temporal, interpersonal, and personal context) on emotional health outcomes among Black and White caregivers will serve to better inform research on caregivers with different contextual resources.
|
; Tirrito & Nathanson, 1994). In addition, researchers have shown that Black caregivers typically have much lower incomes than their White counterparts, which may impose additional stress on individuals and their families while also reducing the economic sharing between the extended social networks (Jarrett & Burton, 1999).
The situational context includes factors directly related to activities involved with caregiving, along with other demands posed as cognitive status of care recipients, assisting with activities of daily living (ADLs), and instrumental activities of daily living (IADLs). Researchers have suggested that cognitive functioning and problem behaviors of the care recipient are both associated with increased caregiver depression and well-being (Yates, Tennstedt, & Chang, 1999).
The temporal context focuses on the time period in the caregiver's life cycle when caregiving duties began (e.g., age). As Lazarus and Folkman (1984) discussed, there are "on-time" and "off-time" events that can be perceived as stressful. For instance, spouses may expect to assume a caregiving role of their partner, while adult child caregivers may think the caregiving role is another responsibility in the midst of other competing responsibilities (e.g., starting their own families, competing in the job market). However, the time at which the role starts may be perceived as stressful for both types of caregivers, which may increase negative emotional health outcomes.
The interpersonal context includes the support of family and friends and the quality of the relationships caregivers have with others in their support network. Studies have shown that positive appraisal of social support networks, even the perception that help from the network was available, could have a positive impact on caregivers' psychological health in times of stress (Haley et al., 1996; Picot, 1995).
Finally, the personal context consists of a set of psychological and psychosocial characteristics that caregivers may use to influence their emotional health. The personal context may include perceptions about caregivers' functional health, their ability to meet basic financial needs, and coping style (i.e., stressfulness of appraisals and positive aspects of caregiving) of caregivers. When coping responses do not eliminate or reduce stress, caregivers' health was negatively affected (Clark, Anderson, Clark, & Williams, 1999).
On the basis of the theoretical and conceptual argument, I conceptualized caregivers' contexts as providing the resources and support needed to continue in the caregiving role while reducing the impact of depressive symptoms; when stressors are present, higher levels of depressive symptoms will be experienced. The purpose of this study was to determine the utility of the contextual approach in predicting emotional health (i.e., depressive symptoms) and to determine whether racial differences existed in contextual factors associated with emotional health.
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
) study, a 6-year longitudinal multisite study, sponsored by the National Institutes of Health under the leadership of the National Institute on Aging (NIA) and the National Institute of Nursing Research (NINR). The current study data were collected through randomized clinical trials at four sites that recruited both Black and White participants: Thomas Jefferson University in Philadelphia, University of Tennessee–Memphis, University of Alabama–Birmingham, and the Research and Training Institute of the Hebrew Rehabilitation Center for Aged in Boston. My study sample consisted of 295 Black and 425 White dementia caregivers for whom baseline data were collected (N = 720). The Philadelphia site contributed 122 Black and 123 White caregivers, followed by Memphis with 97 Black and 143 White caregivers. Birmingham included 60 Black and 80 White caregivers, and Boston contributed 16 Black and 79 White caregivers.
The REACH study's main purpose was to test home and community interventions for enhancing family caregiving, particularly within minority families. The study recruited family caregivers living with persons with dementia for at least 6 months. REACH collected baseline data on 1,229 caregiver and care recipient dyads between 1996 and March 2000, of which 1,222 were randomized into 15 intervention and control conditions. Further details about the REACH design and cohort can be found elsewhere (i.e., Wisniewski et al., 2003). REACH recruited family caregivers from multiple community sites, health, and social agency settings, with special attention to enrolling diverse populations.
After study administrators obtained informed consent from those who were eligible, caregivers were administered survey instruments during structured in-person interviews and then were randomly assigned to an intervention condition. Detailed information about the overall REACH study and recruitment strategies is reported elsewhere (Burgio, Stevens, Guy, Roth, & Haley, 2003; Burns, Nichols, Martindale-Adams, Graney, & Lummus, 2003; Gitlin et al., 2003; Nichols, Malone, Tarlow, & Lowenstein, 2000; Tarlow & Mahoney, 2000). The present study used baseline data collected prior to the REACH interventions.
Measures
Emotional health
Emotional health was assessed using the Center for Epidemiological Studies–Depression (CES-D) measure (Radloff, 1977). The CES-D, a 20-item scale, has four categories of assessment: depressive affect, positive affect, somatic signs, and interpersonal distress. Caregivers were asked to indicate how often they felt a particular way during the past week. Sample items included, "I was bothered by things that usually do not bother me," and "I felt that I could not shake off the blues, even with help from my family and friends." Responses ranged from 0 to 3 (0 = rarely to none of the time, less than one day; 1 = some or little of the time, 1 to 2 days; 2 = moderately or much of the time, 3 to 4 days; 3 = most or almost all the time, 5 to 7 days). Higher scores indicated greater depressive symptoms, with scores of 16 or greater indicating clinically significant levels of depression (current sample: 
=.72).
Sociocultural context
Participants were asked two single-item questions about gender and marital status. Gender was coded 1 for female and 0 for male. Marital status was also dichotomized (married = 1; not married = 0). I also assessed years of formal education as a continuous variable and measured employment using a dichotomous variable (employed = 1; homemaker or not employed = 0). Participants' yearly household income before taxes was measured using the NAM-Powers index. There were 10 income groups ranging from (0), less than $5,000, to (9), $70,000 or more. I treated responses for this item as continuous with higher scores indicating higher incomes.
Situational context
I examined the caregivers' situation by assessing the amount of reported problem behaviors, ADL and IADL responsibilities, and cognitive status of care recipients. I used the Revised Memory and Behavior Problem Checklist (RMBPC; Teri et al., 1992), to assess the number of potential problem behaviors of care recipients. It is a 24-item scale that consists of a list of problem behaviors that care recipients might have exhibited during the past week. It posed the standard question: "Within the past week, has (care recipient) done X?" For each "yes" item scored on the checklist, the participant received one point. Higher scores indicated more memory and problem behaviors ( =.77). A second question item assessed how bothered caregivers were about the behavior problems exhibited by care recipients. This item is discussed later in terms of appraisals (personal context).
The situational context also included ADL (Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963) and IADL (Lawton & Brody, 1969) measures. Questions about ADLs (7 items) assessed the care recipients' ability to perform tasks of day-to-day importance, and questions about IADLs (8 items) assessed whether care recipients could live independently in the community as well as the amount of caregiver assistance required in performing such tasks. I coded response options as 1 = yes and 0 = no. Sample ADL items include whether the care recipient "needed any kind of help getting into or out of bed, a chair, or wheelchair." Sample IADL items include "using the telephone" and "help with shopping." If caregivers reported that care recipients needed help to perform the activity, they were asked whether they helped with the activity. A score of 1 was assigned to each item when care recipients needed assistance and caregivers provided the assistance. I summed the responses to these items to create a single index of total ADL–IADL help needed ( =.84). The higher the score, the more activities caregivers assisted care recipients with.
Using the Mini-Mental State Examination Score (MMSE: Folstein, Folstein, & McHugh, 1975), I assessed the cognitive status of care recipients. The MMSE is composed of two parts. One part contains items that cover orientation, memory, and attention. The second part tests care recipients' ability to name, follow verbal and written commands, write a sentence spontaneously, and copy a complex polygon. The maximum total score is 30. Higher scores indicate better cognitive functioning exhibited by care recipients. I could not calculate Cronbach's alpha coefficient for this measure because individual item scores were not available in the data set provided by the Interuniversity Consortium for Political and Social Research (Schulz, 2001).
Finally, I added the vigilance measure (Mahoney et al., 2003) to the general assessment of caregiver burden and the costs of caregiving. A single item assessed the number of hours caregivers felt they had to "be there" in order to provide care for care recipients. The more hours caregivers had to be there with care recipients, the greater the burden and costs of caregiving.
Temporal context
I measured the time period in the caregiver's life, as indexed by age, as a continuous variable. I also asked caregivers, "What is your relationship to the care recipient (1 = spouse, 2 = adult child, 3 = other)?" In the current study, I conceptualized the relationship to care recipients to represent other competing roles.
Interpersonal context
I examined formal support and informal support of caregivers. I measured formal support by using the Formal Care Services questionnaire (Schulz, 2001). This measure included 21 yes or no questions that assessed the kinds of community-based services that caregivers and/or care recipients had used in the past month. The questionnaire asked participants about services such as homemaker, home health aide services, transportation, MRIs, and CAT scans. For this study, I assessed the number of formal support services used. Higher scores corresponded to greater use of formal support services by caregivers or care recipients.
I assessed informal social support (social support and social network) using a multidimensional measure from the work of Krause and Borawski-Clark (1995). The 11-item scale measured received support based on tangible, emotional, and informational support from friends, neighbors, or family members ( =.83). I assessed social networks—the actual number of available social contacts—with the 7-item Lubben Social Network Index (Lubben, 1988), which was modified by Newsom and Schulz (1996). This scale had separate items for family, friends, and confidants ( =.69).
Personal context
I examined caregivers' leisure time satisfaction, religiosity, financial difficulty to meet basic needs, stress appraisal, positive aspects of caregiving, and number of chronic illnesses. I used the leisure time satisfaction measure (Stevens et al., 2004) to assess information about the caregivers' ability to engage in desirable social activities. The index consisted of six items assessing how satisfied caregivers were with the amount of time they had to engage in activities, such as spending quiet time alone and taking part in hobbies, during the past month. Response options included: "not at all (0), a little (1), a lot (2)." In the current study, higher scores (range: 0–12) indicated more satisfaction with the level of participation in social activities ( =.80).
To assess religiosity, I examined information on the religious behaviors exhibited by caregivers and used two items to assess religious activities in their life: how often one attended religious services, meetings, and/or activities and how often caregivers prayed or meditated (5-point scale, ranging from never to nearly every day). I then summed the two items to create one variable to represent religious activity. Higher scores on this measure indicated more participation in religious or spiritual activities.
I assessed financial difficulty by asking caregivers one item about the difficulty of paying for the very basics like food, housing, medical care, and heating. Responses ranged from not difficult at all (1), not very difficult (2), somewhat difficult (3), to very difficult (4). I used financial difficulty as a continuous variable. Higher scores indicated greater financial difficulty.
The RMPBC second-question item assessed the appraisal of care recipients' problem behaviors reported. For each problem behavior reported, caregivers were asked, "How bothered or upset were you by this?" on a five point scale (not at all, a little, moderately, very much, extremely). The RMPBC Bothered score, or stressfulness appraisal, represented the average bothered score for each problem behavior identified by caregivers (Haley et al., 1996). The higher the score, the more bothersome caregivers appraised the care recipients' problem behaviors.
I also used the Positive Aspects of Caregiving scale (Tarlow et al., 2004) as an appraisal instrument to examine the caregivers' mental and affective state in relation to the caregiving experience and the caregivers' perception of benefits associated with the caregiving experience (11 items). Responses were on a five point agree–disagree scale. Sample items included "Providing help to [care recipient] has made me feel more useful, ... made me feel good about myself, and ... given more meaning to my life." The higher the score, the more positive appraisals caregivers had about caregiving (Cronbach's =.90).
Seven items assessed the number of health problems, or comorbidities, experienced by caregivers (Schulz, 2001). These items asked caregivers whether they had been diagnosed as having any of seven chronic health problems (e.g., arthritis, diabetes, and stroke). The higher the number of chronic health problems listed, the higher the level of comorbidity.
Data Analyses
Data analyses were conducted using JMP statistical software (SAS, 1996). For the purposes of these analyses, emotional health was operationalized with a measure of depressive symptomatology. I computed descriptive statistics (means and percentages) for demographic variables for caregivers and care recipients as well as descriptive statistics (means, standard deviations, percentages, chi-squares, and t tests) for depressive symptomatology and each of the predictor variables. A correlation matrix was also constructed to determine how the variables in the model were related to one another.
I calculated bivariate associations by race to determine the relationships between the variables in the contextual framework. A multiple regression was conducted to examine what factors were associated with emotional health. Interaction terms between race and each of the contextual variables were added to the model to test for homogeneous effects of the contextual variables on emotional health across race. The current study included two regression models: 1) full sample without race interaction terms and 2) full sample with race interaction terms. Due to potential multisite and regional differences, site was controlled in both models (Kraemer, 2000). Some of these site differences are noted elsewhere (Haley et al., 2004).
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Table 2 shows the correlation matrix of all the variables used in the conceptual model. I evaluated the variance inflation factors (VIF) to determine whether multicollinearity might be influencing the regression findings, and I found all to be below 3. These were well below the cut-off of 10 for substantial multicollinearity (Belsky, Kuh, & Welsh, 1980). Table 3 shows the bivariate analysis means, percentages, and standard deviations of the contextual variables included in the conceptual and theoretical model by race. Black caregivers have fewer years of education, were more often employed, have less income, and were less likely to be married than White caregivers. Black and White caregivers were helping care recipients with similar ADLs and IADLs, but Black caregivers were responsible for care recipients who have poorer cognitive status, while White caregivers provided care to care recipients who have significantly more memory and behavior problems than Black caregivers.
|
|
Results of the multiple regression models are shown in Table 4. In Model 1, income (ß = –.09, p <.05) was the only sociocultural context factor significantly associated with fewer depressive symptoms. Findings show that among situational factors, caregivers who reported more memory and behavior problems (ß =.11, p <.01) reported more depressive symptoms. Among temporal context factors, younger caregivers (ß = –.26, p <.001) and being a spouse caregiver (vs adult child; ß =.14, p <.01) were significantly associated with higher depressive symptoms. Also in Model 1, caregivers who used more formal support services (ß =.10, p <.01), who indicated more health problems (ß =.14, p <.001), and who were more bothered by care recipients' problem behaviors (ß =.28, p <.001) reported more depressive symptoms. Caregivers who reported a larger social network (ß = –.18, p <.001), more satisfaction with their level of social activities (ß = –.15, p <.001), and more positive aspects of caregiving (ß = –.09, p <.01) also reported fewer depressive symptoms. The predictor variables explained 36% of the variance in depressive symptomatology (see Table 4).
|
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
), the current study examined the emotional health of Black and White dementia caregivers. Specifically, this study provides information about a critical issue facing family caregivers in our aging population. Dementia caregiving poses several challenges to caregivers, one of which is compromising their emotional health. While emotional well-being research has explored several caregiver factors, other contextual factors (i.e., personal factors) have been neglected. In the current study, I conducted regression analyses with both Black and White caregivers to determine the contextual factors influencing depressive symptoms. Finally, I performed regression analyses using both Black and White caregivers and including interaction terms with race and contextual factors to determine what effect race had on depressive symptomatology. Results of the regression analyses reveal that caregivers pay a major emotional price to care for demented loved ones. Specifically, caregivers who used more formal support services (interpersonal) had more difficulty meeting basic financial needs (personal), and caregivers who reported more stressful appraisals of the care recipients' behavior (personal) and who had more chronic illnesses (personal) also reported more depressive symptoms. However, caregivers who were older (temporal), who had a larger social network (interpersonal), were satisfied with leisure time activities (personal), and who had more positive aspects of caregiving (personal) reported fewer depressive symptoms.
Results also show that the difficulty in dementia caregiving comes not only from the actual stressors that were faced but also from how caregivers interpreted the situation and whether the caregiver had available resources and coping mechanisms (i.e., level of social activities, social network size, appraisal) to help them deal with the stressors. Similar to previous studies, Black caregivers reported more positive aspects of caregiving (Foley, Tung, & Mutran, 2002; Haley et al., 2004), fewer stressfulness appraisals, and more religious activities (Haley et al.) than White caregivers. But, more importantly, this study shows that the burden of dementia caregiving was greater for Black caregivers of care recipients who reported more memory and behavior problems.
Further, within the personal context, researchers have shown that when caregivers' health fails, they often exhaust all informal support and utilize formal support out of necessity (Blieszner & Alley, 1990). While Black caregivers reportedly use formal support services less often, this study's results do not show any differences by race. Given the changing family structure among Black families, utilizing formal support services may become more acceptable to meet the challenging needs of a demented older adult.
In addition, this study illustrates the importance of social activities. Caregivers who were more satisfied with the amount of time they spent in leisure activities reported fewer depressive symptoms. Similarly, prior research using stress process models has identified social activities as an important coping resource (e.g., Haley et al., 1996; Schulz, Gallagher-Thompson, Haley, & Czaja, 2000; Stevens et al., 2004). As noted by Stevens and colleagues, satisfaction with social activities was positively associated with emotional well-being.
As previously mentioned, racial differences identified from this study are noteworthy. Black caregivers may be providing care to care recipients with more memory and behavior problems, yet potential interpersonal and personal resources (e.g., social network and positive aspects of caregiving) may contribute to why these caregivers report fewer depressive symptoms than White caregivers. These findings may also provide evidence of strong internal coping resources (Haley et al., 2004). Despite similar experiences in providing care to demented care recipients, Black and White caregivers have consistent differences on reporting memory and behavior problems of the care recipient and depressive symptoms.
Results from this study illustrate that context is very critical when trying to understand the role of dementia caregivers and their emotional health outcomes. Specifically, findings show that although there was no main effect for race, after running interactions I found that differences existed. This may mean that race is possibly masking the more important contextual resources, which in this study is the personal context. It appears that personal factors (e.g., being able to financial provide for family, get some leisure time away, etc.) are essential in understanding the caregivers' emotional well-being. This is an important finding as researchers struggle to assess the meaning of race in our models. Using race as a proxy for personal context, culture, or socioeconomic conditions adds very little to our understanding of what race-related differences or disparities mean and through which pathways these differences are represented.
Strengths and Limitations of the Study
Several strengths are noteworthy regarding this study. First, the sample was recruited across the United States, allowing for a large number of Black as well as White caregivers to be examined. In addition, the data have allowed researchers to take important steps forward in examining Black and White caregivers and depressive symptoms within a contextual framework. However, the sample included care recipients who were already diagnosed as having Alzheimer's disease or experiencing related dementias. For some, this may raise concerns about external validity. However, dementia is a rather common and progressive disease that is impacting older adults and their caregivers.
In addition, there was a limitation in the contextual model. There was not a temporal context variable that sufficiently measured whether the caregiving event began at an expected or unexpected time in their lives. The proxies used in the current study (i.e., age and relationship to care recipient) were reasonable alternatives; however, these variables may not address whether caregiving was expected or not. As seen from this study, caregivers' perception of their role was influential in how they emotionally responded regardless of the time that caregiving began.
Future Research
Understanding the context that a dementia caregiver embodies can assist researchers in alleviating the emotional toll caregiving may cause. As indicated in this study, personal context factors are essential in reducing the depressive symptoms among Black and White caregivers. Although not included in the current study, future research should consider the addition of a cultural socialization assessment to the current battery of instruments to assess caregiving health outcomes. Specifically, a measure is needed to examine acculturation, assimilation, and cultural values, norms, and beliefs that may dictate how and why caregivers from different racial and ethnic backgrounds provide care under challenges that ultimately impact their own health. This is not to suggest that caregiving-related stressors should be downplayed. Rather, it suggests that the complexities around caregiving as experienced by different ethnic and racial groups could be understood by examining cultural factors more closely (Dilworth-Anderson et al., 2002).
Although this study provides evidence that context is important, more research is needed to replicate these findings. Replication of these findings with other large samples of Black and White caregivers is necessary to validate the utility of the variables used for explaining the differences in depressive symptoms. As stated by Gallo (1990), support given to the caregivers will not only benefit their well-being but will also optimize the care given to the care recipients, which might offset premature and costly residential long-term-care service use.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication August 17, 2004. Accepted for publication May 11, 2005.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. W. Williams, C. S. Williams, S. Zimmerman, J. Munn, D. Dobbs, and P. D. Sloane Emotional and Physical Health of Informal Caregivers of Residents at the End of Life: The Role of Social Support J. Gerontol. B. Psychol. Sci. Soc. Sci., May 1, 2008; 63(3): S171 - S183. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
