| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address all correspondence to Dr. Laura N. Gitlin, Director, Community and Homecare Research Division, 130 South 9th Street, Suite 513, Philadelphia, PA 19107. E-mail: laura.gitlin{at}jefferson.edu.
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Clinical trial • Home modification • Home care
Future population projections suggest that families will continue to be the principal source of support for patients with dementia and that the home will remain the predominant setting for dementia care (Czaja, Eisdorfer, & Schulz, 2000; Wahl & Gitlin, 2003). Despite the proliferation of caregiver interventions (see Bourgeois, Schulz, & Burgio, 1996; Kennet, Burgio, & Schulz, 2000; Schulz et al., 2002), few studies test in-home supportive services that provide families with the requisite knowledge and skills to set up the home environment to manage daily care of persons with ADRD. Thus, developing and testing new approaches that teach families specific skills in providing quality home care and that, in turn, decrease caregiver burden and enhance well-being remains an important public health priority.
This article reports the results of a randomized clinical trial that tests the effectiveness of a home-based environmental intervention, the Environmental Skill-Building Program (ESP), and identifies subgroups of caregivers who benefit more than others do from the intervention. The ESP was tested as part of the National Institutes of Health initiative, Resources for Enhancing Alzheimer's Caregiver Health (REACH). The REACH project, a 6-year multisite cooperative study, was designed to characterize and test the feasibility of different promising, innovative interventions for family caregivers of persons with ADRD at 6, 12, and 18 months of study participation. The Philadelphia site, one of the six participating sites, developed and tested the ESP. The Philadelphia trial has two phases: the active phase, which occurs over the first 6 months of study participation; and the maintenance phase, which occurs over the subsequent 6 months. This article reports the results of the 6-month active phase of participation in the ESP, which reflects the primary endpoint for this intervention. Forthcoming articles will examine the sustainability of treatment effects of the ESP.
The ESP is designed to provide family caregivers with education about the disease process and the impact of environments on care recipient behaviors, problem-solving techniques to identify antecedents and consequences of targeted problem behaviors, and technical skills to modify the home. The ESP differs from psychosocial caregiver interventions with respect to its theoretical base, goals, and the specific caregiver concerns that are targeted for intervention. Based in a competence–environmental press framework (Lawton & Nahemow, 1973) and personal control theory (Schulz & Heckhausen, 1999), as described elsewhere in detail (Corcoran et al., 2002; Gitlin & Gwyther, 2002), the goal of the intervention is to lower the threshold or press of the social and physical environment for the caregiver and person with dementia, and provide the caregiver with primary strategies to maintain personal control and effective management of complex daily care problems. An environmental approach emphasizes the objective stressors or conditions in the caregiving situation (e.g., care recipient behaviors, home condition, and social network) that can be modified to enhance the ability of caregivers to carry out everyday caregiving tasks related to the care recipient's functional dependence and behavioral manifestations. This approach is distinct from, but complementary to, previously tested interventions for caregivers in its focus on the role of the home environment in supporting care recipient function and facilitating caregiving. Previous research has chiefly tested psychoeducational or counseling interventions for caregivers that emphasize the affective dimensions of the caregiving experience. These interventions principally seek to provide emotional support, stress management, and coping strategies. (For comprehensive reviews of these studies see Bourgeois, Schulz, & Burgio, 1996; Kennet, Burgio, & Schulz, 2000.)
Previous Research on Environmental Interventions for Family Caregivers
Although home safety recommendations have traditionally been a part of hospital and home care with dementia patients (Alzheimer's Association, 1997), only recently have systematic evaluations of approaches involving environmental modification occurred. Early exploratory and descriptive studies on home environments showed that family caregivers modified their homes and perceived these changes as helpful. These studies also showed that families tend to implement environmental strategies to cope chiefly with safety concerns and need professional guidance to apply strategies to other behavioral problems. Nevertheless, this first wave of studies with family caregivers used primarily single-case or panel designs, and outcomes were limited to utilization rates of environmental strategies and self-reported benefits (Olsen, Ehrenkrantz, & Hutchings, 1993; Pynoos & Ohta, 1991).
More recently, in a randomized controlled study with 202 dementia caregivers, we found that a 3-month intervention involving environmental recommendations resulted in improvements for both caregivers and care recipients (Gitlin, Corcoran, Winter, Boyce, & Hauck, 2001). Specifically, caregivers who received the intervention, compared with caregivers in usual care, reported less decline in instrumental activities of daily living, a trend toward less decline in self-care dependency, and fewer behavior problems at 3 months posttest. In addition, spouses who received the intervention reported less upset with behavioral manifestations, women reported enhanced self-efficacy in managing troublesome behaviors, and women and African American caregivers reported enhanced self-efficacy in managing functional dependency compared with their counterparts in the control group.
Although little research has been conducted on environmental modification with family caregivers, studies in other settings and with physically frail older adults provide empirical justification for the use of this approach in dementia households. For example, the severity of delirium in hospitalized older people has been associated with modifiable environmental risk factors (McCusker et al., 2001), and interventions to prevent delirium in the hospital involving environmental modification have been shown to be effective (Inouye et al., 1999). A randomized controlled study testing an assistive device and home modification intervention found lowered long-term care cost and postponement in relocation among the physically frail older people in the treatment group (Mann, Ottenbacher, Fraas, Tomita, & Granger, 1999).
Differential Effects of Caregiving Interventions
Caregivers represent a highly diverse population with wide-ranging and different needs, cultural beliefs and values, and caregiving preferences. Consequently, a given intervention approach may not be suitable for all caregivers. Thus, to advance an understanding of the impact of interventions for caregivers, it remains important to examine who may or may not benefit from different approaches (Biegel & Schulz, 1999). More specifically, an in-home environmental intervention requires caregivers to actively problem solve, change lifelong daily routines, use new care strategies, and adjust or remove material aspects of the environment that may have personal meaning. We thus asked whether there were subgroups of caregivers that respond more favorably to this approach than others.
In the present study, we examined possible interactions between the treatment and three characteristics of caregivers: gender, race, and relationship to the care recipient. These three characteristics have been shown in the research literature to account for significant amounts of variance in caregiver functioning, coping, and responses to caregiving demands. As to gender, women caregivers report more psychiatric symptoms in their care recipients and provide more hands-on involvement in caregiving (e.g., level of responsibility, number of tasks, and number of hours spent in hands-on care) than men caregivers (see Yee & Schulz, 2000, for a comprehensive review). In contrast, men caregivers tend to seek more assistance from family members and other informal services than women and express more objective than subjective burden (Neal, Ingersoll-Dayton, & Starrels, 1997; Schulz et al., 1995). Furthermore, although a sense of control is an expressed need for both genders, being assertive or exerting control is often accompanied by strain for women who are not used to assuming authority and may not have the skills to do so effectively (Miller, 1987). Little is known, however, about whether men and women caregivers derive distinct benefits from interventions. Our previous research showed that women were more likely to comply with a home environmental intervention than men (Gitlin, Corcoran, Winter, Boyce, & Marcus, 1999), and that they benefited more from participation (Gitlin et al., 2001).
Caregiving research has also consistently shown that spouses have higher rates of upset and depression than nonspouse caregivers (Pruchno & Resch, 1989; Schulz et al., 1995). However, little is known about how such differences affect participation in treatment and outcomes. Zarit, Stephens, Townsend, Greene, and Leitsch (1999) showed differences in adult day center use such that spousal caregivers tended to be briefer users than nonspousal caregivers. Our previous research showed that spouses in treatment expressed reduced upset with disruptive behaviors compared with nonspouses.
Whereas much research has shown racial differences in appraisals of caregiving, with African Americans showing lower levels of upset or depression, little is still known as to whether intervention outcomes vary by race or ethnicity. A few studies suggest differential effects for psychosocial interventions. Cox (1998) found that African American caregivers benefited more than White caregivers from a psychosocial intervention. Burgio, Stevens, Guy, Roth, and Haley (2003) also found gains for African American caregivers participating in an in-home skills program, whereas White caregivers benefited more from telephone contact. However, our previous research found that for an environmental intervention, minority caregivers, the majority of whom were African American, showed greater gains in only one caregiving domain, that is, self-efficacy in managing functional dependence, relative to controls. There were no differences between White and African American caregivers with regard to level of upset and other treatment outcomes. It is still unclear why differences occur and for which types of treatments, and whether socioeconomic status or educational level account for differential effects.
Present Research
The current Philadelphia REACH study systematically builds on previous research on home environmental research in several ways. First, the intervention we tested as part of the REACH initiative was standardized, enhancing its reproducibility. Second, the REACH ESP was more intense than our previous 3-month intervention, involving more occupational therapy time in the home, as well as the actual implementation of special equipment and other low-cost environmental strategies that had been recommended but not actually provided in the previous study. Third, in our previous work, care recipients had higher levels of functioning and less cognitive impairment, whereas in this study, we tested an environmental approach with care recipients at the moderate to severe stage of the disease process. Fourth, this study expands an understanding of the specific benefits of an environmental approach by including a broader range of measures than that previously examined. Finally, we sought to confirm our previous findings for women and spouse caregivers and advance an understanding as to whether certain caregivers derive more benefit than others do. This is important in order to help guide future research and develop targeted service programs involving an environmental approach.
In this article, we examine both main treatment effects on multiple outcomes and the interaction of treatment with caregiver characteristics. Specifically, we hypothesized that the ESP would have a stress reduction effect on measures of objective and subjective caregiver burden. We also expected that the ESP would have a quality of life enhancement effect for caregivers as measured by their perceived well-being, sense of mastery, and use of effective caregiving strategies. Furthermore, we evaluated the intervention's impact on behavioral disturbances and decline in daily functioning, as reported by the caregivers. We reasoned that environmental intervention strategies were designed to compensate for disablement and might therefore help support and sustain daily functioning of the care recipient. Finally, on the basis of our previous research and that of others, we explored the possibility that certain groups of caregivers would benefit more than others from the program. That is, we expected women in the intervention group to derive greater therapeutic benefits than men and spouses to report less upset than nonspouses. Interactions of treatment with gender and relationship were examined to replicate our previous findings. However, on the basis of our previous work and the extant literature, we did not anticipate treatment differences between African American and White caregivers for any of our outcome measures.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Caregivers were recruited primarily from the Philadelphia Corporation for Aging (PCA), the area agency on aging for Philadelphia County, and secondarily from media announcements. The PCA sent a letter and study brochure to potential eligible caregivers who were identified from an initial review of clients on a waiting list for in-home services. Caregivers who received the letter and were interested in participating in the study were invited to either contact the research team by telephone or mail a self-addressed, stamped response card. Caregivers who contacted the research team (n = 413) were then interviewed by telephone to determine their study eligibility and willingness to participate. Of the 413 persons who initially contacted the research team, 290 were eligible for study participation, of whom 255 (88%) were willing to participate. A comparison of the 255 caregivers who enrolled in the study with the 35 caregivers who were eligible but not willing to participate indicated no large or statistically significant differences in caregiver age, gender, relationship to care recipient, years providing dementia care, number of IADL or ADL dependencies of the care recipient, or care recipient's gender or age (Schinfeld, Klein, Winter, & Gitlin, 2001).
Of the 255 persons initially enrolled, 188 participated in the 6-month follow-up interview, representing a 26% attrition rate. Twenty subjects were bereaved, 16 had placed their family member in a long-term care facility, 24 missed the follow-up interview, and 7 had dropped out of the study. In accordance with clinical trial research principles, to reduce the number of subjects with missing data, we interpolated scores for 2 subjects who were unavailable at the 6-month follow-up but participated in the 12-month follow-up interview. The difference between the 12-month follow-up and baseline values was divided by time between follow-up and baseline to obtain a difference per day. This value was then multiplied by 182 (the number of days in 6 months) and added to the baseline score to derive a 6-month value. This resulted in a sample of 190 caregivers reported here.
A comparison of the background characteristics of the 190 caregivers included in the 6-month analyses to the 65 caregivers who did not have 6-month data revealed differences in race, relationship, and group assignment. Among the 65 caregivers without 6-month data, more were White, that is, 58% versus 41.5%, respectively (
2(1) = 3.65, p =.056); more were spouses, 49.2% versus 50.8%, respectively (2(1) = 3.98, p =.046); and 61.5% who were assigned to treatment discontinued compared with 38.5% of the control group caregivers, (2(1) = 4.19, p =.041). Although there was a higher rate of attrition in the experimental group compared with controls, the dropout rate for spouses and Whites was equivalent. In addition, the number of bereaved caregivers and caregivers who had placed their care recipient in a nursing home facility was equivalent in both the experimental and control group conditions.
Study Procedures
Following determination of study eligibility, trained interviewers met with caregivers in their homes, obtained signed informed consent on a consent form approved by the Institutional Review Board, and conducted the baseline interview. The interview consisted of a core battery of measures used by all REACH sites and described elsewhere (Wisniewski et al., in press), as well as several measures specific to the Philadelphia site. The entire interview took approximately 3 hr to administer. Following the baseline interview, caregivers were stratified by relationship (spouse or nonspouse) and race (White or non-White) and randomized into experimental or usual care control groups. These design variables were intended to ensure equivalence in race and relationship between experimental and control group subjects. The randomization schedule used random permuted blocks within each stratum and was prepared by the site biostatistician. Randomization was accomplished by opening opaque sealed envelopes in sequential order for the appropriate stratum. Participants were interviewed again at 6, 12, and 18 months postbaseline in accordance with the REACH protocol. Control group subjects received resource information at each testing occasion. The data reported here include follow-up data at the 6-month interview, which was the primary end point for the Philadelphia site.
The ESP
As already stated, the therapeutic premises of the ESP are a competence–environmental press framework (Lawton & Nahemow, 1973) and a particular conceptualization of the environment as physical, task, and social interactive layers (Barris, Kielhofner, Levine, & Neville, 1985), and personal control theory (Schulz & Heckhausen, 1999). Lawton and Nahemow (1973) posited that departures from an optimal match between the individual's competence and the demands of the environment produce problems in functioning and well-being. The theory was proposed for the elderly population in general but has particular relevance for individuals with ADRD, given the cognitive and functional impairments that the disease entails. The goal of the ESP is to help family caregivers reduce the disparity between environmental press and personal competence by providing skills to effectively manipulate dimensions of the home environment to manage daily problems associated with dementia care. Environmental modifications reflect primary control strategies to enhance the caregiver's perceived efficacy and lower appraisals of stress. The program consists of four components: education about dementia and the impact of the home environment on troublesome behaviors and ADL deficits; instruction in problem solving and developing effective approaches to manage caregiving concerns that involve manipulating the physical and social environment; implementation of environmental strategies tailored to the caregiver's context; and generalization of strategies to emerging problems.
The strategies that are introduced modify the physical, task, or social dimension of the home environment (Barris et al., 1985). Strategies that modify the physical dimension may include the installation of equipment (e.g., grab bars or handrails); the removal, rearrangement, or labeling of objects; the use of color contrast; or the purposeful placement of objects (e.g., clothing or items for hygiene). Strategies that alter the task dimension may include simplification techniques such as effective communication (e.g., use of short verbal or written instructions), verbal and tactile cueing, planning a daily routine, and engaging the care recipient in graded activities. Strategies that modify the social environment may include helping caregivers learn to coordinate care among their social network, instructing them how to communicate with formal providers, and assisting them to develop consistency in interacting with the care recipient across care providers.
ESP involves two phases: active, which occurs over the first 6 months, and maintenance, which occurs over the subsequent 6 months. This article reports on the immediate and primary impact of treatment following the active phase only. This involves five 90-min home visits and one 30-min telephone contact. In the initial home visit, the occupational therapist reviews intervention goals and conducts a systematic needs assessment to identify which of 11 areas are difficult for the caregiver to manage and for which he or she wants to learn new strategies. The 11 domains considered are caregiver-centered concerns (e.g., caregiver feelings of upset, being overwhelmed, guilt, fatigue, concern for the future, and body aches and pains); communication issues with the care recipient; problems in coordinating care with formal providers or other family members; difficulties assisting in bathing, dressing, toileting, eating, or getting around; concern about home safety; difficulty distracting or engaging care recipient in meaningful activity; concern with wandering; difficulty managing incontinence; or difficulty managing catastrophic reactions. The needs assessment is conducted as a semistructured clinical interview in which the caregiver is encouraged to describe daily life, and the therapist follows up each area of concern with probing questions to identify the specific dimensions of the problem. Following the needs assessment, the occupational therapist and caregiver develop a targeted plan and identify the initial problem area to target. The occupational therapist then observes the caregiver or has the caregiver engage in role-play in order to evaluate his or her approach to the identified problem. In addition, a walk-through of the home is conducted to examine environmental setups and assess how these may affect the problem areas. Education about the disease is offered in this first session, as well as referral information if appropriate. Thus, the ESP protocol involves customization such that interventionists in collaboration with caregivers determine the specific problem areas to address. Strategies introduced in subsequent sessions are tailored to fit the caregiver's preferences and environmental setup. If a referral to community resources is necessary, interventionists consult with a social worker at the PCA.
In the second visit, the occupational therapist continues the education process and works with the caregiver to problem solve about the antecedents and consequences of the particular identified problem area. The occupational therapist helps the caregiver identify an achievable behavioral goal (e.g., reduce number of toileting accidents). Therapist and caregiver engage collaboratively in problem solving to identify alternate care strategies using an environmental perspective. Specific physical, task, and social modification strategies are introduced. Those accepted by the caregiver are practiced and then recorded on a form for the caregiver to review. Caregivers are asked to practice their use prior to the next home visit. In each subsequent home and telephone contact, the occupational therapist reinforces education about dementia through written materials and discussion, addresses a targeted problem area, observes the caregiver using previously recommended strategies, provides refinements to those strategies, and offers new recommendations. In the course of providing verbal instruction, the occupational therapist uses cognitive restructuring and validation to instill greater perceived control and confidence in the caregiver's own abilities to manage the problem and to develop more realistic expectations and appraisals of the caregiving situation and dementia-related behaviors. Helping caregivers reframe attributions and explain events is important to enable behavioral change and support the effective use of environmental strategies. Therapists serve as coaches and provide ongoing validation and reinforcement of the caregivers' use of environmental strategies. In addition, equipment and assistive devices are ordered and installed, if warranted, by the Housing Department of the PCA, who assumes responsibility for the quality and installation of materials. The occupational therapist schedules a home visit following installation or delivery of equipment to evaluate its appropriateness and provide instruction in its use. In the final visit, the occupational therapist reviews previously introduced strategies and how to apply them to future potential problems.
Interventionists participated in 25 hr of training that included both didactic and experiential opportunities. Several techniques to ensure adherence or treatment fidelity to the intervention protocol were implemented (Burgio et al., 2001). First, interventionists were monitored throughout the study by using formal case reviews on a biweekly to monthly basis. Second, a member of the research team directly observed randomly selected intervention visits with caregivers, using an investigator-developed checklist to identify both positive and negative interventionist behaviors. Constructive feedback was then provided to the interventionist and presented at group supervisory meetings. Third, treatment documentation was systematically reviewed and feedback given to interventionists on a continual basis. Finally, brief telephone interviews were conducted with caregivers to evaluate their satisfaction with the program.
Measures
Background characteristics of family caregivers that were measured as continuous variables included age, income, education, and number of years providing care. Gender, relationship to person with dementia (spouse or nonspouse), race (White or non-White) and marital status (currently married or currently not married) were measured as dichotomous variables.
Outcomes Related to Care Recipients
We examined three outcomes related to care recipient functioning: the occurrence of behavioral problems, dependence in ADLs, and dependence in IADLs.
A version of the Revised Memory and Behavior Problem Checklist (RMBPC; Teri et al., 1992) that was modified by the REACH initiative was used. This version changed the response options from frequency of occurrence to presence or absence (yes or no) of a behavior (Roth et al., in press). As in the original validated scale, we used the three-factor structure reflecting three possible behavior types: nine depressive behaviors, eight disruptive behaviors, and seven memory-related behaviors. We expected reductions in the presence of memory-related and disruptive behaviors. However, we did not expect reductions in depressive behaviors given that ESP helps families manage the functional consequences of memory loss and disruptive behaviors, but does not target affective behaviors such as sadness or depression in the care recipient. Thus, the depressive subscale was not included in the analyses presented here. We computed two indices reflecting the number of behavioral occurrences within the domain of memory-related behaviors (Cronbach's alpha for our sample =.51) and disruptive behaviors (Cronbach's alpha for our sample =.59). High scores indicated the occurrence of a greater number of behaviors.
Family caregivers rated the level of ADL assistance required of the person with dementia by using a modification of the Functional Independence Measure (FIM; Granger & Hamilton, 1992), developed specifically for the REACH initiative (Gitlin et al., 2003). We used eight items from the mobility subdomain of the FIM (bathing, eating, dressing upper and lower body, toileting, grooming, getting around the house, and getting in and out of bed). Ratings ranged from complete independence (7) to complete dependence (1). A total score was computed by averaging scores across all items. We refer to this index as ADL Dependence (Cronbach's alpha for our sample =.91).
Caregivers also rated the level of assistance required for eight IADLs (Lawton & Brody; 1969). Level of assistance was measured by using the same 7-point modified FIM rating scale as already mentioned. A mean of the ratings on the eight items was computed to develop the index of IADL Dependence (Cronbach's alpha for our sample =.85). High scores indicated greater independence.
Outcomes Related to Caregivers
We examined caregiver ratings on seven measures of stress reduction and four measures reflecting enhancement effects. For stress reduction we considered both objective and subjective burden. Caregiver burden is a core construct in assessing the effect of caregiving (George & Gwyther, 1986). It has been conceptualized and measured in numerous ways (for reviews see Dunkin & Anderson-Hanley, 1998; Vitaliano, Young, & Russo, 1991). One conceptualization by Thompson and Doll (1982) distinguishes objective from subjective burden as distinct outcomes of providing care that may be differentially associated with caregiver characteristics. Objective burden refers to the disruption to family life and roles, amount of time providing supervision or oversight, or financial impact. Subjective burden refers to the emotional response to caregiving such as feelings of upset.
Objective burden was represented by three REACH items: vigilance, total hours of IADL help, and help received for ADLs. The vigilance item asked caregivers, "About how many hours a day do you estimate that you are actually doing things for CR?" (Mahoney et al., 2003). The total hours of IADL help item asked caregivers, "On average, how much time do you spend per day helping with IADL activities?" The final item, referred to as days receiving ADL help, asked caregivers, "How many days in a week have other family members or friends (not being paid) provided help (supervision, direction, or personal assistance) with [activities of daily living]?" Because these distinct measures of objective burden were single-item measures, Cronbach alphas were not computed.
Subjective burden was operationalized in terms of upset in four areas: upset with RMBPC memory-related behaviors, with disruptive behaviors, and with providing assistance with IADLs and ADLs. We used the caregiver reaction (upset) subscales from Teri and colleagues (1992) for memory-related and disruptive behaviors and applied this approach to each IADL and ADL item as well. For each item in these subscales, caregivers rated their level of upset on a 5-point scale (0 = no upset to 4 = extremely upset). For behaviors that do not occur or items that do not require assistance, a score of 0 (no upset) was assigned. A mean caregiver upset score was then computed for each index. We refer to these indices as Upset With Memory-Related Behaviors (Cronbach's alpha =.84), Upset With Disruptive Behaviors (Cronbach's alpha =.67), Upset With ADL Assistance (Cronbach's alpha =.82), and Upset With IADL Assistance (Cronbach's alpha =.81). Higher scores for each index indicated greater caregiver upset.
Enhancement was operationalized as caregiver mastery, skill enhancement, and well-being. To measure perceived mastery we used the Caregiving Mastery Index from Lawton, Kleban, Moss, Rovine, and Glicksman (1989), a six-item scale measuring the caregiver's appraisals of his or her ability to provide care (e.g., "How often do you feel you should be doing more for CR?"). The scale uses a 5-item Likert format ranging from 1 (never) to 5 (always). A higher score reflects greater mastery. Cronbach's alpha for this scale with our sample was.66.
Skill enhancement was measured with the Task Management Strategy Index (Gitlin et al., 2002), a scale developed specifically for this study. This 19-item scale measures the extent to which positive caregiving strategies were used to manage ADL dependence and problem behaviors. Caregivers indicate how often they used strategies such as visual and tactile cueing, simplifying routines, or short instructions, using a 5-point Likert format (1 = never to 5 = always). A mean task strategy use score was calculated, with high scores indicating greater use of such strategies (Cronbach alpha for our sample =.77).
Finally, caregiver well-being was operationalized with a 13-item scale developed for this study, the Perceived Change Index (Machemer et al., 2000). For each item, caregivers rated on a 5-point scale whether things have become worse (1) or improved a lot (5) over the past month. For example, caregivers were asked to appraise changes in their affect, ability to manage difficult behaviors, and somatic feelings (e.g., fatigue). An exploratory principal axis factor analysis, rotated to a varimax solution, yielded three interpretable factors with eigenvalues greater than 1.00. These corresponded to affect (Cronbach's alpha =.85), somatic feelings (Cronbach's alpha =.81), and management abilities (Cronbach's alpha =.76), and together they accounted for 53.7% of the variance. The overall scale was internally consistent (Cronbach's alpha =.90). We examined treatment effects for each subscale and for the overall index, referred to as Overall Well-Being or Perceived Change. We expected that participation in the ESP would be associated with improved scores on the total scale as well as the affect and management subscales, but not for the somatic subscale.
Data Analysis
Our first analytic step was to examine background characteristics of the 190 caregivers and the outcome variables, using chi-square and Wilcoxon rank-sum tests, as appropriate, to identify possible differences between experimental and control group participants at baseline. Second, we assessed treatment effects on the outcome measures by using analysis of covariance (ANCOVA). In these analyses, the outcome variable was the 6-month score. Covariates included the baseline value of the outcome measure and caregiver gender, race (White or non-White), and relationship (spouse or nonspouse). We tested the normality assumption for each dependent measure by examining the distribution of the residuals. The residual distribution was found to be skewed, and normality was improved with a transformation of the data for three outcome variables: vigilance, hours helping with IADLs, and days receiving ADL help. We report transformed results for these three outcomes converted back to the original scale.
Our third analytic strategy was to examine treatment interactions with caregiver race (White or non-White), gender, and relationship (spouse or nonspouse), using the same analytic approach as for the main effect models described. We considered the treatment interaction analyses with objective burden and the enhancement measures as exploratory or secondary to the initial main effect model. By contrast, the treatment interaction analyses with subjective burden, and care recipient behaviors and physical function, were considered to be confirmatory of our previous findings. In this article we report the interactions that reached statistical significance or were large in magnitude. The reported p values were not corrected for multiple end points. We conducted analyses of the main effects for measures reflecting three domains (stress reduction, enhancement, and care recipient functioning). The analyses of treatment interaction effects considered three interactions (gender, race, and relationship) for each outcome measure within these domains. All analyses were conducted by using SPSS version 10.0. The level of significance was set at.05.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
|
Effects of the ESP Intervention for Specific Subgroups
Separate ANCOVAs were conducted to examine treatment effects by specific caregiver characteristics (race, gender, and relationship) for each outcome listed in Tables 2–4. A number of statistically significant interaction effects with gender and relationship were found for some measures of stress reduction and enhancement. No statistically significant interaction effects were found for race or for any measures of care recipient functioning. Only large or statistically significant interaction effects are graphically presented. Figure 1 shows treatment by gender interaction effects for objective burden. As shown, women and men differed in vigilance or hours doing things for the care recipient (p =.011 for interaction). Men in the experimental group showed a significant reduction in time spent relative to men in the control group (p =.012). In contrast, women in both the experimental and control group conditions reported no change (p =.413). We found the opposite pattern with regard to days receiving ADL help (p =.036 for interaction). Women in the experimental group showed a significant reduction in the number of days receiving help with ADLs compared with control group women (p =.003), but men in both experimental and control groups showed no significant changes (p =.457) based on the pairwise comparison. Figure 2 shows treatment by gender interaction effects on enhancement variables. Men and women differed in overall well-being as measured by the Perceived Change Index (p =.020 for interaction). Specifically, experimental group women in comparison with control group women (p =.008) showed significant improvement, whereas both experimental and control group men showed less improvement (p =.231). The same pattern was found for two subscales, perceived change in affect (p =.024 for interaction) and ability to manage caregiving (p =.043 for interaction). That is, experimental group women in comparison with control group women showed significant improvement in affect (p =.003), whereas experimental and control group men showed no change (p =.345). Similarly, in comparison with control group women, experimental group women experienced significant improvement in their ability to manage caregiving (p =.029), whereas experimental and control group men showed no change (p =.263).
|
|
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Overall, the findings from this study support the main hypothesis that a home environmental approach reduces stress in targeted areas of objective and subjective burden and enhances select aspects of caregiver well-being. Furthermore, the findings confirmed our previous research that an environmental approach reduces upset with behaviors for spouses but does not appear to be effective for nonspouses, and that women gain more from this approach than men. However, in contrast to the findings from our previous study, for this REACH sample, we did not find main or interaction treatment effects for care recipient functioning. As anticipated, in this study, we did not find any differential treatment effects for White and non-White caregivers, the majority of whom were African American.
More specifically, the data show that caregivers who received ESP improved over 6 months in three areas: objective burden, as measured by the amount of help received by family and friends for ADL assistance; subjective burden, as measured by the level of upset with memory-related behaviors; and perceived change in affective well-being. In contrast, control group participants showed declines in each of these areas over 6 months. Caregivers did not, however, show statistically significant improvements in other outcome variables, as we initially hypothesized. We expected but did not find improvements in burden with ADL and IADL assistance. We also did not find a decline in time spent overseeing IADL performance or, importantly, a slower rate of decline in functional dependence among care recipients, nor did we find that caregivers reported fewer disruptive behaviors. The lack of significant change in level of upset may be due to a floor effect such that caregivers reported very little burden at baseline with functional dependence. Consequently, there may have been little room for improvement in this area. Another explanation may be a mismatch between the intervention and the outcome measures; outcome measures may not have been specific or sensitive enough to detect the environmentally related impact of the ESP.
Regarding the impact of the intervention on care recipient ADL dependence, our previous research indicated that all caregivers in treatment reported a slower rate of decline in care recipients' IADLs, and female caregivers in treatment reported a slower decline in care recipients' ADLs. However, this was not the case with our REACH sample. This may be the result of sample differences: the Philadelphia REACH sample care recipients had greater impairment levels than our previous sample. In the present investigation, caregivers reported complete dependence of the care recipient in IADLs and moderate to complete dependence of the care recipient in ADLs, whereas caregivers in our previous study reported only mild IADL and ADL impairments of care recipients. This suggests that an environmental approach to help caregivers sustain residual abilities may be effective but only if introduced early in the illness trajectory. We were also unable to show a change in the frequency of behavioral occurrences. This may be due to the use of a dichotomized measure of behavioral occurrence that was not sensitive to reductions in frequency of occurrence that may have occurred. Further research is warranted using a more sensitive scale of behavioral occurrences given that the magnitude of change from baseline to posttest for the number of memory-related behaviors was in the expected direction, albeit not statistically significant.
One reason that we found only select main effects may be the complex pattern of significant outcomes observed for various subgroups. Treatment effects varied by gender and relationship of the caregiver to the care recipient. Specifically, female experimental participants tended to use more effective task strategies, which may have resulted in improved mastery, improved ability to manage, improved affective well-being, and less need for and use of ADL assistance from family and friends. Male experimental participants reported a decrease in objective burden (vigilance) but did not report less subjective burden or affective enhancements. These findings are consistent with research on gender differences in caregiving. This literature suggests that women are much more involved in hands-on aspects of caregiving than men (Neal et al., 1997) and tend to perform personal care, household tasks, and spend more hours in care tasks than male caregivers. Thus, women might be more invested in the ESP in that it provides them with practical strategies to provide daily assistance more effectively and efficiently. This, in turn, results in an improved sense of mastery and less need for assistance from others for caregiving duties. Conversely, the approach of men to caregiving is more pragmatic and they express burden with the objective aspects of caregiving. Thus, the ESP provided men with the necessary skills to decrease objective burden with regard to their vigilance activities, but treatment did not result in improved affect. Further, men and nonspousal caregivers were less responsive to the present intervention than women and spouses, suggesting that this approach should be modified to address the special needs of nonresponsive groups. These subgroup findings, however, must be interpreted cautiously given small sample sizes (e.g., there were 22 experimental group men and 23 control group men). In addition, as suggested by Weissert and Hedrick (1994), subjects enrolled in a study represent a subgroup already selected from a larger pool who has met study criteria. Study criteria are designed to identify a subgroup that is believed to most likely benefit from a program.
Our findings fall within the range of effect sizes reported by Sorenson, Pinquart, and Duberstein (2002) in their recent review of 78 caregiver intervention studies. Overall, they found caregiver improvement ranging from 0.14 to 0.41 SDs in caregiver burden and other measures of well-being. Our effect sizes ranged from small for reduced upset with memory-related behaviors (p =.027; d =.23) to moderate for reduced upset with disruptive behaviors in spouses (p =.022; d =.47). Furthermore, Sorenson and colleagues (2002) found that caregivers of persons with dementia in these studies benefited less from interventions than did caregivers of older adults without dementia. Moreover, most of these other studies were not carried out as randomized clinical trials; nor were data analyzed by using intention-to-treat statistical methods.
Several general conclusions can be drawn regarding treatment effects for home environmental interventions for family caregivers. Foremost is that the relationship of treatment to outcome is complex. The findings of this study underscore that the difficulties caregivers confront are multifaceted, and both interventions and measurement strategies should be multidimensional as well. Moreover, the findings highlight the dimension-specific impact of an environmentally based intervention. That is, the ESP had multiple but selective benefits. It did not result in improvements uniformly across all hypothesized outcomes. Related to this point is that subscale analyses are important in caregiver intervention research and offer a more nuanced understanding of the precise domains in which treatment gains are evinced. For example, we found as anticipated that treatment effects were associated with less upset with memory-related and disruptive behaviors, but not with upset with depressive behaviors, using the subscales of the RMBPC scale. A related point that can be drawn from this study is that subgroup analyses are important to fully understand who benefits from interventions.
One limitation of this study was its use of multiple end points, resulting in three statistically significant main effects. Because we found main effects on measures that were not evaluated in our previous study, further confirmation of these findings is required. Another limitation is that it is unclear whether the positive change that occurred in the treatment group is due to social attention by a health professional or the environmental content of the intervention given that our comparison was a usual care group. However, it seems unlikely that reported improvements particularly in perceived skill level (e.g., mastery and ability to manage on a day-to-day basis) among women in the treatment group are a consequence of attention. Nevertheless, more research is warranted to substantiate this claim. Also of interest is a comparison of an environmental approach with other traditional caregiver interventions to evaluate the relative benefits of each (for a comparison of the REACH interventions, see Gitlin et al., in press, and Belle et al., in press).
Another consideration for future research is the importance of determining the differential impact of treatment in order to effectively target caregiver services. Specifically, future research on environmental interventions for ADRD caregivers should anticipate interactions with gender and family relationship and provide sufficient statistical power to detect intervention effects. In summation, the Philadelphia REACH study provides further evidence of the benefits of an environmental approach for caregivers and demonstrates that treatment effects are not uniform but vary by the domain of measurement and caregiver characteristic, particularly by caregiver gender and relationship. Given the targeted stress reduction and enhancement benefits of the ESP for caregivers, and the additional gains for women and spouses found in this study, we believe there is preliminary evidence to suggest that this type of intervention should be integrated into home and community-based services for families.
|
Footnotes |
|---|
1 Community and Homecare Research Division, Thomas Jefferson University, Philadelphia, PA.
2 Department of Occupational Therapy, Thomas Jefferson University, Philadelphia, PA; now with the School of Medicine and Health Sciences, George Washington University, Washington, DC.
3 Department of Medicine, Thomas Jefferson University, Philadelphia, PA.
Decision Editor: Laurence G. Branch, PhD
Received for publication May 9, 2002. Accepted for publication September 6, 2002.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. N. Gitlin, L. Winter, M. P. Dennis, and W. W. Hauck Variation in Response to a Home Intervention to Support Daily Function by Age, Race, Sex, and Education J. Gerontol. A Biol. Sci. Med. Sci., July 1, 2008; 63(7): 745 - 750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Devor and M. Renvall An Educational Intervention to Support Caregivers of Elders With Dementia American Journal of Alzheimer's Disease and Other Dementias, June 1, 2008; 23(3): 233 - 241. [Abstract] [PDF]
|
|
|
|
|
|
M. J. L. Graff, M. J. M. Vernooij-Dassen, M. Thijssen, J. Dekker, W. H. L. Hoefnagels, and M. G. M. OldeRikkert Effects of Community Occupational Therapy on Quality of Life, Mood, and Health Status in Dementia Patients and Their Caregivers: A Randomized Controlled Trial J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2007; 62(9): 1002 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. K. Chee, L. N. Gitlin, M. P. Dennis, and W. W. Hauck Predictors of Adherence to a Skill-Building Intervention in Dementia Caregivers J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2007; 62(6): 673 - 678. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J.L. Graff, M. J.M. Vernooij-Dassen, J. Zajec, M. G.M. Olde-Rikkert, W. H.L. Hoefnagels, and J. Dekker How can occupational therapy improve the daily performance and communication of an older patient with dementia and his primary caregiver?: A case study Dementia, November 1, 2006; 5(4): 503 - 532. [Abstract] [PDF]
|
|
|
|
 |
|
 L. N. Gitlin, K. Reever, M. P. Dennis, E. Mathieu, and W. W. Hauck Enhancing quality of life of families who use adult day services: short- and long-term effects of the adult day services plus program. Gerontologist, October 1, 2006; 46(5): 630 - 639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. C. Williams Emotional Health of Black and White Dementia Caregivers: A Contextual Examination J. Gerontol. B. Psychol. Sci. Soc. Sci., November 1, 2005; 60(6): P287 - P295. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. N. Gitlin, W. W. Hauck, M. P. Dennis, and L. Winter Maintenance of Effects of the Home Environmental Skill-Building Program for Family Caregivers and Individuals With Alzheimer's Disease and Related Disorders J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2005; 60(3): 368 - 374. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Mittelman, D. L. Roth, W. E. Haley, and S. H. Zarit Effects of a Caregiver Intervention on Negative Caregiver Appraisals of Behavior Problems in Patients With Alzheimer's Disease: Results of a Randomized Trial J. Gerontol. B. Psychol. Sci. Soc. Sci., January 1, 2004; 59(1): P27 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Schulz, L. Burgio, R. Burns, C. Eisdorfer, D. Gallagher-Thompson, L. N. Gitlin, and D. F. Mahoney Resources for Enhancing Alzheimer's Caregiver Health (REACH): Overview, Site-Specific Outcomes, and Future Directions Gerontologist, August 1, 2003; 43(4): 514 - 520. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
experimental group males; 
usual care group females; 
experimental group females
