The accelerating trend toward athome care of the acutely and chronically ill has greatly expanded the role of families in caregiving (Brogna, 1998; Schumacher, Stewart & Archbold, 1998). Advances in medical technology, the changing demographic landscape, the growth of managed care and its diminishing reimbursement for support services, as well as strict guidelines for institutionalization have forced families to care for their disabled family members in the home setting and for longer periods. Fengler and Goodrich (1979) were the first to astutely label the spouses of disabled elderly men as "hidden patients," thereby warning the gerontological community that providing such care can place the caregiver at risk for negative physical, emotional, and social outcomes. Institutionalization of an ill older adult is frequently the only solution to relieving the burden of home caregiving. Because of the enormous physical, emotional, and financial costs to the caregivers, care recipients, health care system, and society as a whole, caregiver burden persists as a dominant theme in gerontological research.
Since the 1970s, many pertinent issues have been examined with respect to the burden of family caregivers of demented adults including: measurement of burden; correlates of burden; the process of caregiver selection; and to a lesser degree, evaluations of interventions designed to ease caregiver burden. Alzheimer's disease, an illness that places progressive and unrelenting demands on caregivers because of its severe and unpredictable behavioral and cognitive deficits, has been a focal point of disease-specific caregiving studies. The protracted nature of the illness further magnifies the psychiatric and physical morbidity effects of caregiving in this context. Furthermore, caregivers for individuals with dementia tend to be spouses who are especially vulnerable because of their own advanced age (Mastrian, Ritter, & Deimling, 1996). The purpose of this article is to explore the concept of burden itself and review the variety of tools to measure caregiver burden in the specific context of Alzheimer's disease and related dementing illnesses. Following a adescription of the tools and their psychometric properties, there will be a discussion of the research and practice implications of accurate assessment of caregiver strain.
CAREGIVER BURDEN DEFINED
Caregiver burden is a complex, multidimensional construct. Since the pioneering work of Zarit, Reever, and Bach-Peterson (1980) who first developed burden as a research construct, it has evolved to represent not only the physical tasks of caregiving but also the affective and cognitive responses that simultaneously exist (Stetz, 1992). Most descriptions of the phenomenon of burden include physical, psychological or emotional, social, and financial consequences.
George and Gwyther (1986) define caregiver burden as the "physical, psychological or emotional, social, and financial problems that can be experienced by family members caring for impaired older adults" (p. 253). It has been described as a burden, strain, and stressor that produces negative psychological responses (Poulshock & Deimling, 1984; Thompson & Doll, 1982; Zarit et al., 1980). Given and Given (1991) stated it is a "biopsychosocial reaction of the primary caregiver resulting from an imbalance of demands relative to resources available" (p. 84).
Burden is often characterized as the persistent hardships, stress, reactions of caring, or as the physical and psychological, social, and financial, problems experienced by family members providing care in the home (Montgomery, Gonyea & Hooyman, 1985). Klein (1989) added that the level of stress is determined by the caregiver's perceived responsibility for and relationship with the care recipient, and by the caregiver's level of moral development, which in turn impacts one's interpretation of the burden experience.
On the other hand, many authors acknowledge the positive impact of caregiving, which can include feelings of self-satisfaction or mastery, increased self-respect from fulfilling a responsibility, and coping with a personal challenge (Cohen, Gold, Shulman, & Zucchero, 1994; Given & Given, 1991; Horowitz, 1991; Kramer, 1997; Lawton, Kleban, Moss, Rovine, & Glickman, 1989; Motenko, 1989; Stetz, 1992). For the most part, however, there remains an inherent attributional process whereby caregivers have basic assumptions about caregiving being burdensome (George & Gwyther, 1986; Stull, Kosloski & Kercher, 1994). The term itself may be problematic. Care-giving does not always appear to produce solely negative outcomes. Others add that the effects of caregiving should not be viewed in the strictly burdensome sense but rather interpreted in terms of its impact on well-being (George & Gwyther, 1986; Stull et al., 1994). Analyzing the concept in this broader perspective - to include positive, negative, and neutral effects- more accurately elucidates the overall impact on the caregiver.
Ory et al. (1985) took the definition of burden one step further by defining it in the specific context of Alzheimer's disease. They defined burden as the "impact of changes in cognition and behavior of the Alzheimer patient on the family, and the patient's subsequent need for care and supervision" (p. 623). Pearlin, Mullan, Semple, and Skaff (1990) added that the burden or stress commonly associated with caregiving is not an event or a unitary phenomenon, but rather a "mix of circumstances, experiences, responses, and resources that vary considerably among caregivers and that, consequently, vary in their impact on caregivers' health and behavior" (p. 591). It cannot be defined by any specified set of behaviors on the part of the caregiver; it merely begins with the older person's need for assistance and has no designated threshold regarding the minimal degree of involvement necessary for one to be classified as a caregiver (Malonebeach & Zarit, 1991).
There are two distinct types of burden: objective burden, relating to the disruption of family life and impact on caregiver in terms of actual impact on finances, role, and neighbor relations; and subjective burden, relating to the caregiver's feelings in response to the situation. It is not just the number or type of tasks performed for an ill person that makes caregiving challenging, but rather the impact of this experience on one's well-being and the caregiver's subsequent appraisal of such difficulties (Montgomery et al., 1985; Poulshock & Deimling, 1984; Thompson & Doll, 1982). As Stetz (1992) noted, the analysis of caregiving demands can be carefully accomplished by separating objective tasks from the subjective or experienced difficulty of demand. Miller, McFall, and Montgomery (1991) distinguished between two primary dimensions of burden. They described "personal burden" as caregivers' perceptions of problems in their personal actions and activities affected by caregiving and "interpersonal burden" as caregivers' perceptions of problems in the relationship between the caregiver and the care recipient. However categorized, a great deal is yet to be learned about caregiving for the cognitively impaired elderly population. It begins with the exact conceptualization of this elusive construct - the lack of which complicates the task's precise measurement and, in turn, compromises the development of specific therapeutic interventions.
OVERVIEW OF MEASURES
Not only has burden been broadly defined, it also has been differentially measured. Despite the variety of caregiver burden instruments noted in the literature, the need for reliable and precise methods for measuring this complex concept still exists. Of the 10 measures that follow, a number of the earlier measures are classified as unidimensional, focus strictly on caregivers' feelings and experiences, and tap similar facets. Others are multidimensional because they reflect various areas of the care-giver's well-being and function that may be differentially affected by the dementia of the relative (Novak & Guest, 1989).
The Burden Interview
By far, the most widely used instrument to measure caregiver burden is also one of the first developed. The Burden Interview (BI) (Zarit et al., 1980) is a 29-item self-report inventory that examines burden associated with functional and behavioral impairments in the home care situation. As the title implies, it is administered by structured interview (approximately 1.5 hours administration time). It is a unidimensional measure focusing only on the affective response of the caregiver to the consequences of caregiving. The respondent indicates on a threepoint scale how much discomfort the item or concern causes by choosing the most appropriate phrase from "not at all" to "extremely." A total burden score is then calculated. It has been used in a variety of studies, either in full or in part.
Psychometrically, the items are noted to possess content validity because they were derived from clinical and research experiences with caregivers of individuals with dementia and reflect common areas of concern, namely health, social Ufe, finances, and interpersonal relationships. Additionally, the tool has demonstrated high internal consistency (alpha = .91) and test-retest reliability of .71 (Vitaliano, Young, & Russo, 1991). Major strengths of the BI include the broad scope of situational stimuli considered to be sources of burden (e.g., life changes that might contribute to burden and affective responses) and its ease of administration. Furthermore, it can be used with both spouse and child caregivers. Its major drawback is its unidimensional nature and its subjective wording (i.e., simply noting the presence of a problem does not automatically imply distress).
The Caregiver Strain Index
The Caregiver Strain Index (CSI) (Robinson, 1983) is an example of an approach whereby burden is described as potentially severe distress or "strain." The CSI was developed inductively and consists of 13 items initially administered to a sample (n = 81) of spouses, family, friends, and neighbors providing varying degrees of care to patients age 65 and older recendy hospitalized for hip surgery or arteriosclerotic heart disease. Scoring is dichotomous (yes = 1, no = 0) and similar to the BI. Subjective burden can be inferred only by the endorsement of certain items (e.g., difficulties with sleep, physical strain, family and work adjustments, upsetting behaviors, feeling completely overwhelmed). The scale has demonstrated a high internal consistency with an alpha of .86.
Using a variety of other established instruments, which address the physical and emotional health of caregivers and their subjective perceptions of the caregiving relationship, construct validity was supported. Two advantages of this screening tool are its brevity and ease of administration. Its ability to quickly identify strain makes it useful in preventive clinical practice and in research (Robinson, 1983). However, it was used to measure common stressors experienced by informal care providers for patients postdischarge following acute hospitalization for hip surgery or heart disease. It was primarily developed and tested on caregivers for loved ones with dementia, thus limiting its applicability.
Poulshock and Deimling's Model of Caregiving Impact
Poulshock and Deimling (1984) developed a multidimensional tool to examine subjective and objective burden related to functional/behavioral impairment, impact on life, and emotional responses of the caregiver or "burden" as the mediator in this relationship. Impairment is operationalized as dependency (number of activities of daily living [ADLs] requiring assistance). Mental impairment is operationalized as how upset caregivers are in relation to three indicators - sociability, disruptive behavior, and cognitive incapacity. Impact is operationalized as disruption in family life (e.g., job conflict, finances, interpersonal relationships) and is measured by using items similar to those in the BI.
The 13 -item scale includes a subscale of negative changes in family relationships (eight items, alpha = .88) and a subscale of restrictions in activity (five items, alpha = .80). Evidence of construct validity was provided through observed relationships between the burden measures and care recipient cognitive and ADL abilities; impact on family relationships and depression; and the decision to institutionalize based on disruptive care recipient behaviors, negative changes in relationships, and activity restrictions (Deimling & Bass, 1986). Although this tool is not used extensively, it does represent a major contribution to the conceptualization of burden in that it measured the construct in relation to actual caregiving problems and care recipient impairments. A limitation of the model is that the distress associated with the disruption of family life is not considered. It also fails to recognize psychological distress as a source of burden. Also cited is the noncontinuous scaling of the items (tiring, difficult, upsetting) which potentially diminishes its usefulness (Vitaliano et al., 1991).
Montgomery, Gonyea, and Hooyman's Inventories
Montgomery et al. (1985) are credited with extending the conceptualization and measurement of burden, making a distinction between objective and subjective burden via structured interviews (approximately 45 minutes in length) using a 22-item burden inventory. Objective burden, defined as changes and disruptions, includes 9 items, each rated on a 5point scale (ranging from a lot more, a little more, the same, a little less, a lot less). Respondents are asked to report the extent to which their caregiving behaviors had changed nine areas of their lives and personal freedom they had for themselves. Subjective burden, defined as attitudes and emotional responses of the caregivers, was assessed using 13items, many of which were derived from the BI.
Respondents report how often they experienced each of the 13 feelings. Again, each item rated on a fivepoint scale ranging from rarely or never, a little of the time, sometimes, often, and most of the time. The objective and subjective burden scales had internal consistency alphas of .85 and .86, respectively. Construct validity was supported only in relation to one ADL item, assisting with physical transfer. Because many items were extracted from the BI, it has similar shortcomings.
Unlike the majority of burden instruments designed for use in research studies, the Cost-of-Care Index (CCI) (Kosberg & Cairi, 1986) was primarily developed as a case management tool to assist in identifying actual or perceived problem areas of families as the primary informal caregivers of elderly relatives. The 20-item tool can be self-administered or case manager/researcher-administered. The items were initially extracted from a pool of variables associated with abusive behaviors in the provision of care for elderly individuals. There are five dimensions:
* Personal and Social Restrictions.
* Physical and Emotional Health.
Each is composed of four Likerttype items rated as "strongly agree," "agree," "disagree," or "strongly disagree." The coefficient alpha reported for CCI scores was calculated to be 0.91; no evidence of validity is provided. This tool takes a unique approach to assessing the burden or "cost" of caregiving. It has been used for family screening, intervention, and peer group interaction. Factor analysis was useful in fine-tuning the tool, but there is a question of the adequacy of sample size used for the number of items tested. Overall, the CCI appears to be a more practical and interpretable scale.
One might question the derivation of the instrument grounded in abusive behaviors research. Although it was tested in a sample of both physically and cognitively impaired patients, it is unclear how well their findings specifically fit caregivers of the cognitively impaired (Novak & Guest, 1989). The authors do note that it can assist in the identification of highrisk situations upon hospital discharge (e.g., poor care, propensity for institutionalization or elder abuse), but they are quick to add that it is not to be used as the sole determinant for any placement decision (Kosberg & Cairi, 1986; Kosberg, Cairi & Keller, 1990).
Caregiver Reactions Assessment
The Caregiver Reactions Assessment (CRA), developed by Given, King, Collins, and Given (1988), contains 27 Likert-type items that focus on general areas:
* Positive and negative reactions.
* Impact on schedule.
* Perceived impact on caregiver health.
Its development was based on a model whereby caregiver characteristics (e.g., health, relationships) and patient characteristics (e.g., memory, health, communication, incontinence) combine to determine involvement in care (ADLs and instrumental activities of daily living) which in turn, lead to caregiver "reactions."
The CRA conceptualizes family members' responses to their caregiving roles as reactions rather than burdens in that these feelings may be both positive and negative and may vacillate over time. The CRA items, rated on a five-point scale, are divided into five subscales: positive reactions, negative reactions, financial responses, impact on daily schedule, and perceived caregiver health. The alpha coefficients for each of the subscales were .88, .84, .89, .66, and .86, respectively. It was also used in subsequent studies comparing caregivers of patients with Alzheimer's disease and those with cancer or other physical impairments. Construct validity was supported. The authors concluded that the CRA instrument is ideal for discovering both differences in the reactions of various groups of caregivers and changes in reactions over time. Although used specifically with caregivers of patients with Alzheimer's disease, it can also function as a more generic measure for other conditions as well.
Caregiver Burden Inventory
The Caregiver Burden Inventory (CBI) developed by Novak and Guest (1989) is another example of a multidimensional measure of caregiver burden. The 24 items were derived from review of the burden literature, previously published scales, and from the experiences of caregivers of individuals with Alzheimer's disease. Caregivers of cognitively impaired older people (n = 107) responded to items, rating them from 0 (not at all descriptive) to 4 (very descriptive). Factor analysis revealed five interpretable factors:
* Time-dependence burden.
* Developmental burden.
* Physical burden.
* Social burden.
* Emotional burden.
Internal consistency reliability of the five factors was good, ranging from .73 to .86. As with the BI, the items include both affective responses and task-related sources of burden, mixing objective and subjective burden. Despite the fact that the subscales cover a wide range of problems reported by caregivers of cognitively impaired individuals, evidence of construct validity is lacking. The CBI is a relatively short yet comprehensive measure of burden that is a practical tool for assessing, responding to, and designing therapeutic approaches (Novak & Guest, 1989; Vitaliano et al., 1991).
Caregiver Hassles Scale
The 42-item Caregiver Hassles Scale (CHS) (Kinney de Stephens, 1 989) measures the daily demands of caring for a family member with Alzheimer's disease. It is more globally oriented, based on a transactional perspective in which stress is conceptualized as hassles, or minor irritations, of daily living (Lazarus & Folkman, 1984). Hassles are both transient and chronic events appraised by individuals as threatening to their wellbeing. Individually, hassles exert a weak threat, but an accumulation of these stressors over time can have a significant impact on the individual's health. The scale consists of 42 items reflecting a variety of potential hassles that might have occurred in the previous week of caregiving. Five categories of hassles are represented: those associated with assistance with basic and instrumental ADLs (9 and 7 items, respectively), care recipient cognitive status (9 items) and behavior (12 items), and caregiver support network (5 items). Items are scored for occurrence during the past week (yes or no), and for those that occurred, rated on a four-point scales ranging from "not at all a hassle" to "a great deal of hassle." Cronbach's alpha and test-retest reliabilities (assessed at a 1-day interval) were high at .91 and .83, respectively. Construct validity was assessed in terms of other measures that measure similar constructs and in terms of three theorized outcomes of stress; correlations were high for two of the three subscales.
The CHS does seem to provide a reliable instrument for determining hassles associated with assistance in basic ADLs; assistance in instrumental ADLs; and care-recipients' cognitive status, behavior, and social network. One must note that the strong findings supporting internal consistency may be partly because of the greater length of this scale. Similarly, test-retest reliability was strong but measured only at a 1 -day interval. The presence of a hassle does not necessarily imply burden with respect to that experience. Another obvious conceptual concern is the decision to limit the responses to occurrence during the past week. Although designating a response frame enhances precision, limiting responses to such a short time period - particularly when issues related to dementia care tend to be pervasive worries - may compromise its precision (Vitaliano et al., 1991).
Caregiver Appraisal Scale
Lawton et al. (1989) devised the Caregiver Appraisal Scale (CAS) based on Lazarus and Folkman 's (1984) stress model. Rather than continuing to use the term burden, they suggested that the term "caregiving appraisal" be used because it is multifaceted and includes all cognitive and affective appraisals and reappraisals of the potential stressor, in addition to the efficacy of one's coping efforts. They emphasize that not all potential stressors are perceived as negative or burdensome; subjective responses can also be positive (Lazarus' concept of "uplifts") or neutral. The CAS includes aspects of other measures of burden, as well as potentially positive dimensions of the caregiving experience. The 47 items (which range from "embarrassed over impaired person's behavior" to "helping the impaired person has made me feel close to him/her") are scored by using one of two 5 -point scales (never to nearly always true or strongly agree to strongly disagree). Five dimensions (subjective caregiving burden, caregiver satisfaction, perceived caregiving impact, caregiver mastery, and caregiver ideology) were identified via factor analysis.
To test for reliability, the CAS was administered to home caregivers via structured interview and then readministered 1 8 months later to those participants who had since placed the impaired individual in a nursing home (n = 239). Internal reliability of the subscales ranged from .65 to .87. Construct validity was established in relation to affective states, quality of relationship with care recipient, and emotional burden.
Advantages of this scale are its longitudinal perspective and its broad scope in addressing the complexities of the interrelationships between positive and negative aspects of caregiving. Particularly noteworthy is its recognition of "caregiving mastery" (which addresses positive feelings of accomplishment and self-esteem) and "caregiver ideology" (which addresses the principles of reciprocity, role modeling for children, and tradition) as potential benefits of caregiving. Its primary drawback may be its length, especially if used in the spousal caregiver population.
Screen for Caregiver Burden
The 25-item Screen for Caregiver Burden (SCB) was designed by Vitaliano, Russo, Young, Becker, and Maiuro (1991) to assess objective and subjective burden among caregivers of spouses with Alzheimer's disease. It was offered as either a substitute for or as a complement to more lengthy approaches. The development and testing of the SCB was undertaken in three investigations (n = 191). The most frequently cited stressors were translated into structure items which tap several domains, including care recipient behaviors, disruptions in family and social life, and caregiver affective responses. The SCB yields two scores: objective burden (OB) and subjective burden (SB).
The former consists of a prevalence count of caregiving experiences, whereas the latter reflects ratings (from 1 to 4) of distress in relation to each experience. As far as its psychometric properties, internal consistency alphas were .85 and .89 for OB and SB, respectively. Test-retest reliabilities were good (.70, ? > .001; and .64, ? > .001 for SB and OB, respectively), especially considering the 15- to 18month interval versus the 1-day to 3week intervals used with other measures. Both construct validity and criterion validity were supported.
The SCB is a brief measure of upsetting caregiver experiences and does not purport to sample the full domain of caregiving experiences, only those possibly related to distress. In addition to its strong evidence of reliability, validity, and sensitivity to change, the SCB is not related to gender or education - a desirable quality because scores may not have to be adjusted accordingly for these variables. It is one of the few instruments to be tested longitudinally but is not without limitations. One of its drawbacks is that certain items represent appraisals (e.g., "it is exhausting...," "I am upset..."), so even objective items involve an element of subjectivity. Consequently, its developers suggested a more accurate set of labels such as, "prevalence of experiences" to replace "objective burden" and "appraisal of distress from experiences" instead of "subjective burden." These labels may more closely parallel components of the cognitive-phenomenological stress model (Lazarus & Folkman, 1984).
Although many burden measures exist in the literature, there is much debate regarding their usefulness. First, there is the problem with the precise definition of the construct. Conceptualization of burden has yielded diverse representations which can obviously cloud the measurement picture. There still remains little consensus regarding a universally acknowledged definition, hence how to best assess its impact. Is burden a unidimensional or multidimensional phenomenon? Does its definition vary depending on whether it is studied as a predictor or an outcome? Can all the potential consequences of caregiving be fully captured in a single measure of burden? Should burden instead be viewed as a dimension of wellbeing so that its positive aspects are simultaneously recognized?
Since the development of Zarit's (1980) interview tool (considered by many to be the premier, albeit unidimensional, measure of caregiver burden), confounding of objective and subjective sources of burden has remained problematic in both its conceptualization and measurement. The development of several multidimensional instruments in the meantime has lent support for the crucial distinction between objective aspects of burden and the subjective response to the experience. Approaching burden unidimensionally not only fails to recognize its multifaceted nature, but it confines one's understanding of caregiving to only the negative effects (Stetz, 1992) versus the often overlooked positive aspects of the experience (Cohen et al, 1994).
Another related and important focus identified in the literature is the obvious challenge of quantifying the concept of burden via a measure. George and Gwyther (1986) delineated three problems pertaining to the measurement of caregiver burden:
* The explicit focus on caregiving, so measures cannot be administered to noncaregivers.
* Extant measures of burden explicitly require respondents to relate caregiving to its impact, an unwelcome kind of confounding occurs (i.e., the presumed stressor and its outcomes become intertwined).
* Many measures of burden generate total scale scores and reliance upon such scores can mask dimension-specific patterns of caregiving impact and precludes identification of the different correlates or antecedents of its specific dimensions.
While the first is an analysis issue concerning the use of a generic measure of burden, the others can and to some degree have been addressed via the ongoing refinement of multidimensional instruments.
Specificity in the measurement of caregiver burden is an equally salient concern. To enhance instrument precision, items should reflect common problems that are characteristic features of that specific illness in the target population (Vitaliano et al., 1991). For example, measures designed for caregivers of individuals with terminal cancer may not be appropriate for caregivers of individuals with dementia, and vice versa.
Another component regarding specificity concerns the caregiver-care recipient relationship. Caregiver burden has been shown to vary by the nature of the relationship between the caregiver and the recipient (Cantor, 1983; Schulz, O'Brien, Bookwala, & Fleissner, 1995), as well as the living arrangements, which determine whether care is continuous or episodic in nature (Vitaliano et al., 1991). This issue of specificity must be also addressed or the validity of such measures will be compromised.
Likewise, measures should be specific to the setting and the phase of illness. The type and degree of distress in a caregiver tending to the needs of their loved one in the home environment appears to be substantially different than that experienced after the care recipient is placed in a long-term care facility. Although the intensity of burden may be somewhat diminished with transfer of the care-recipient from the home to the nursing home, the source of stress presumably shifts as the caregiver's feelings of resentment of the heavy physical caretaking tasks and constant vigilance are replaced by feelings of guilt over long-term care placement, coping with sudden changes in living arrangements, and possible feelings of failure.
Furthermore, because the content validity of burden measures designed for one type of care recipient or caregiver population is questionable for use in different populations, the practice of adapting scales warrants careful consideration (Vitaliano et al., 1991). Therefore, the controversy lingers regarding whether more global measures can be used interchangeably for comparison purposes versus one designed for a specific caregiving situation. Lastly, practical considerations surface in the development and use of any tool. In the case of any caregiver burden instrument, brevity and ease of administration are of particular significance, especially when used with spousal caregivers. Minimizing respondent burden is of utmost concern in this population who presumably not only are distressed from their circumstances, but also are advanced in age. They are more likely to be frail, perceptually impaired, and more easily fatigued (Alt-White, 1995).
Continued documentation of the impact of caregiving on the caregivercare recipient dyad is essential to the subsequent development and testing of nursing intervention models. Both positive and negative effects must be explored so concerns most amenable to nursing interventions can be identified. One such framework that has emerged in the past decade is the progressively lowered stress threshold model (PLST) developed by Hall and Buckwalter (1987). This conceptual model serves as a specific nursing intervention for dementia management via a caregiving training protocol (Gerdner, Hall & Buckwalter, 1996). Despite such efforts to provide conceptual frameworks on which to base nursing care, clearly much more work needs to be performed to develop and refine coping strategies to support the caregiver. Precise measurement of the type and degree of burden can greatly augment efforts to design practical and effective interventions to assist caregivers in these challenging situations.
IMPLICATIONS FOR NURSING PRACTICE AND RESEARCH
There are several instruments available to quantify the complex concept of caregiver burden. Thus far, they have demonstrated varying degrees of practicality and psychometric stability when used in the Alzheimer's caregiver population. As a group, the psychometric properties have not been systematically and rigorously tested crosssectionally or longitudinally (Malonebeach & Zarit, 1991; Schulz et al., 1995; Vitaliano et al., 1991). Furthermore, longitudinal analysis is of paramount importance particularly when studying the documented detrimental effects of Alzheimer's disease given its progressive, potentially lengthy, and unpredictable course.
Without adequate knowledge of the correlates of burden and the caregivers' differential experiences during the course of the illness, effective strategies to reduce the potentially damaging consequences of caregiving, both observed and perceived, cannot be formulated (Montgomery et al., 1985).
While these instruments are not without flaws, they all have contributed in some way to explicating caregiver burden. A thorough grasp of the experience of caring for a cognitively impaired family member does have profound implications for gerontological nurses, as well as other health care professionals. There are also implications for policymakers who, under managed care, have "become increasingly concerned that the stresses experienced in caring for a disabled family member may erode the capacity of families to continue providing needed services" (Motenko, 1989, p. 166). That is, burden specifically has "indirect policy relevance because of its association with the family's decision to relinquish the caregiving role and seek institutional forms of care" (Braithwaite, 1992, p. 11).
Consequently, gerontological nurses have much to derive from a keen understanding of all elements of the caregiving situation (Bergman-Evans, 1994). Nurses, via their opportunities for unique and productive interactions with the patient and their families, are in the best position to anticipate, assess, and intervene in problems related to or caused by stressful caregiving situations (Gaynor, 1990). Without this valuable information, obtained in part via methodologically sound and appropriate caregiver assessment tools, interventions cannot be developed and implemented to assist this extremely vulnerable group. Interventions can range from educational strategies focused on managing difficult behaviors (Burgener, Bakas, Murray, Dunahee, & Tossey, 1998; Hewett & Chang, 1998), specific skills training (Brogna, 1998), and tapping the caregiver's own internal and external coping resources (Stevens, Walsh, & Baldwin, 1993) to ways to maintain or augment the social support system (Mastrian et al., 1996). Equally crucial to the caregiver is the nurse's support and guidance in identifying and accessing services via community referrals as indicated.
Repeated use and psychometric testing of burden instruments that accurately and consistently assess all dimensions of caregiving, specifically for spouses with dementia, can contribute immensely to the knowledge base in this relevant research area. Improvements would include further development and use of existing instruments that concurrently measure the enjoyable and rewarding aspects of caregiving, such as the aforementioned CAS.
Such data will provide gerontological nurses with a more comprehensive understanding of the caregiving role, thereby enhancing the development of effective intervention strategies to ameliorate the deleterious effects and highlight the positive dimensions of caregiving in the context of Alzheimer's disease. Not only will this optimize the caregiving environment for both the caregiver and care-recipient, but it will benefit society in terms delayed institutionalization as a result of an intact family network more willing and able to endure the challenges of home caregiving.
- Alt- White, A.C. (1995). Obtaining "informed" consent from the elderly. Western Journal of Nursing Research, /7(6), 700-705.
- Bergman-Evans, B.F. (1994). Alzheimer's and related disorders: Loneliness, depression, and social support of spousal caregivers. Journal of Gerontological Nursing, 20, 6-16.
- Braithwaite, V. (1992). Caregiving burden: Making the concept scientifically useful and policy relevant. Research on Aging, 14(1), 3-27.
- Brogna, L. (1998). Caring for caregivers: Home care nursing's challenge. In J.A. Allender & CL. Rector (Eds.), Readings in Gerontological Nursing (pp. 515-522). Philadelphia: Lippincott.
- Burgener, S.C., Bakas, T., Murray, C, Dunahee, J., & Tossey, S. (1998). Effective caregiving approaches for patients with Alzheimer's disease. Geriatric Nursing, 19(3), 121-126.
- Cantor, M.H. (1983). Strain among caregivers: A study of experience in the United States. The Gerontologist, 23, 597-604.
- Cohen, CA., Gold, D.P., Shulman, KJ, & Zucchero, CA. (1994). Positive aspects of caregiving: An overlooked variable in research. Canadian Journal on Aging, 13(3), 378-391.
- Deimling, G.T., & Bass, D.M. (1986). Symptoms of mental impairment among elderly adults and their effects on family caregivers. Journal of Gerontology, 41(b), 778-784.
- Fengler, A.P., & Goodrich, N. (1979). Wives of elderly disabled men: The hidden patients. The Gerontologist, 19, 175-183.
- Gaynor, S.E. (1990). The long haul: The effects of home care on caregivers. Image: Journal of Nursing Scholarship, 22(4), 208-212.
- George, L.K., & Gwyther, L.P. (1986). Caregiver well-being: A multidimensional examination of family caregivers of demented adults. The Gerontologist, 26, 253-259.
- Gerdner, L. ?., Hall, G. R, & Buckwalter, K.C. (1996). Caregiver training for people with Alzheimer's based on a stress threshold model. Image: Journal of Nursing Scholarship, 28(3), 241-246.
- Given, B.A, & Given, CW. (1991). Family caregiving for the elderly. Annual Review of Nursing Research, 9(5), 77-101.
- Given, B.A, King, S.K, Collins, C, & Given, C. (1988). Family caregivers of the elderly: Involvement and reactions to care. Archives of Psychiatric Nursing, 2, 281-288.
- Hall, G.R, & Buckwalter, K.C (1987). Progressively lowered stress threshold: A conceptual model for care of adults with Alzheimer's disease. Archives of Psychiatric Nursing, 1(6), 399-406.
- Hewett, LJ, & Chang, FL. (1998). Alzheimer's disease, a hopeful note for the closing of a century: Challenges for the new millenium. In J.A. Allender & CL. Rector (Eds.), Readings in gerontological nursing (pp. 484-499). Philadelphia: Lippincott.
- Kinney, J.M., & Stephens, M.A.P. (1989). Caregiving Hassles Scale: Assessing the daily hassles of caring for a family member with dementia. The Gerontologist, 29, 328-332.
- Klein, S. (1989). Caregiver burden and moral development. Image: Journal of Nursing Scholarship, 21(2), 94-97.
- Kosberg, J.I, & Cairi, R. (1986). The Cost of Care Index: A case management tool for screening informal caregivers. The Gerontologist, 26, 273-278.
- Kosberg, J.I, Cairi, R.E, Keller, D.M. (1990). Components of burden: Interventive implications. The Gerontologist, 30, 236-242.
- Kramer, BJ. (1997). Gain in the caregiving experience: Where are we? What next? The Gerontologist, 37, 218-232.
- Lawton, M.P, Kleban, M.H, Moss, M, Rovine, M, & Glicksman, A. (1989). Measuring caregiver appraisal. Journal of Gerontology, 44(3), P61-P67.
- Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
- Malonebeach, E.E, & Zarit, S.H. (1991). Current research issues in caregiving to the elderly. International Journal of Aging and Human Development, 32(2), 103-114.
- Mastrian, KLG, Ritter, C, & Deimling, CT. (1996). Predictors of caregiver health strain. Home Healthcare Nurse, 14(3), 209-217.
- Miller, B, McFall, S, & Montgomery, A. (1991). The impact of elder health, caregiver involvement, and global stress on two dimensions of caregiver burden. Journal of Gerontology, 46(1), S9-19.
- Montgomery, RJ, Gonyea, J. G, & Hooyman, N.R. (1985). Caregiving and the experience of subjective and objective burden. Family Relations, J4(issue), 19-26.
- Motenko, A.K. (1989). The frustrations, gratifications, and well-being of dementia caregivers. The Gerontologist, 29, 166-172.
- Novak, M, & Guest, C (1989). Application of a multidimensional Caregiver Burden Inventory. The Gerontologist, 29, 798-803.
- Ory, M.G, Williams, T.F., Emr, M, Lebowitz, B, Rabins, P, Salloway, J, Sluss-Radbaugh, T, Wolff, E, & Zarit, S. (1985). Families, informal supports, and Alzheimer's disease: Current research and future agendas. Research on Aging, 7(4), 623-644.
- Pearlin, L.I, Mullan, J.T, Semple, SJ, & Skaff, M.M. (1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30, 583-594.
- Poulshock, S. W, & Deimling, CT. (1984). Families caring for elders in residence: Issues in measurement of burden. Journal of Gerontology, 39(2), 230-239.
- Robinson, B.C. (1983). Validation of a Caregiver Strain Index. Journal of Gerontology, 38(3), 344-348.
- Schulz, R, O'Brien, A.T, Bookwala, J, & Fleissner, K. (1995). Psychiatric and physical morbidity effects of dementia caregiving prevalence, correlates, and causes. The Gerontologist, 35, 771-791.
- Schumacher, KL, Stewart, BJ, & Archbold, RC (1998). Conceptualization and measurement of doing caregiving well. Image: Journal of Nursing Scholarship, 30(1), 63-69.
- Stetz, K. M. (1992). Response to caregiving demands: Their difficulty and effects on the well-being of elderly caregivers. Scholarly Inquiry for Nursing Practice: An International Journal, 6(2), 129-133.
- Stevens, CL, Walsh, R.A, & Baldwin, B.A. (1993). Family caregivers of institutionalized and noninstitutionalized elderly individuals. Nursing Clinics of North America, 28(2), 349-352.
- Stull, D. E, Kosloski, K, & Kercher, K. (1994). Caregiver burden and generic wellbeing: Opposite sides of the same coin? The Gerontologist, 34, 88-94.
- Thompson, E.H, & Doll, W (1982). The burden of families coping with the mentally ill: An invisible crisis. Family Relations, 31(3), 379-388.
- Vitaliano, P.P., Russo, J, Young, H.M, Becker, J, & Maiuro, R.D. (1991). The screen for caregiver burden. The Gerontologist, 31, 76-83.
- Vitaliano, P.P., Young, H.M, & Russo, J. (1991). Burden: A review of measures used among caregivers of individuals with dementia. The Gerontologist, 31, 67-75.
- Zarit, S.H, Reever, K.E, & Bach-Peterson, J. (1980). Relatives of the impaired elderly: Correlates of feelings of burden. The Gerontologist, 20, 649-655.