Alzheimer's disease (AD) currendy affects more than 4 million Americans (National Institutes of Health [NIH], 1999) and the effect of die disease on family caregivers' mental and physical health ahas been well-documented (Glaser & 1997; Schulz, O'Brien, Bookwala, & Fleissner, 1995). Although a number of medical treatments are currendy being tested for their ability to delay cognitive decline during early to middle stages of the disease (NIH, 1999), non-pharmacological interventions for impaired family members and their caregivers are still the treatment of choice. Caregiving has been considered a stressful experience for most caregivers, particularly for those who live with the care-receiver. More recent studies have suggested that behavioral symptoms, difficulties in providing care for activities of daily living (ADLs), communication problems, poor judgment, depression and anxiety, and, particularly, behaviors classified as agitated or aggressive, are the best predictors of caregiver distress and care-receiver institutionalization (Gallagher-Thompson, 1994; Gilley, Wilson, Beckett, & Evans, 1997).
The purpose of this pilot study was twofold:
* To test the feasibility and effectiveness of an intervention designed to assist caregivers when intervening with their relative's behavioral symptoms.
* To determine the feasibility of adding an immunological measure to a caregiver intervention study, examine the short-term stability of this measure, and determine its association with caregivers' depressive symptomatology.
The following hypotheses were identified:
* The Caregiver Skill Building Intervention (CSBI) will be effective in decreasing caregiver depressive care-receiver behavioral symptoms.
* The immunological measure will have shortterm stability and will have an inverse relationship with caregiver depressive symptomatology.
Relationship Between Caregiver Stress and Mental and Physical Health
Caring for a relative with AD has been found to affect caregiver mental and physical health. In shouldering much of the responsibility for providing care, it is not surprising that family members exhibit high rates of emotional distress and increased physical morbidity (Buckwalter et al., 1999; Schulz et al., 1995). Recent data suggest that physical and emotional stressors can have a negative influence on immune function, particularly cellmediated immunity (Glaser & Kiecolt-Glaser, 1997). Cell-mediated immunity plays a major role in the body's host defense against a variety of infectious agents and tumors. Defective cell-mediated immunity has been associated with an increased risk of cancer, life-threatening infections, immune deficiencies, and autoimmune disorders (Czarnecki, Zalcberg, Kulinskaya, & Kay, 1995).
Recent caregiver studies have focused increasingly on the problematic behaviors associated with AD as caregiver stressors because of their frequency of occurrence, caregiver negative reactions associated with these behaviors, and their consequences for both caregivers and carereceivers. Behavioral symptoms have been associated with caregiver depression and care-receiver institutionalization (O'Donnell et al., 1992). Existing research suggests that behavioral symptoms associated with personal care, such as bathing and dressing, are the most stressful to caregivers. Most notably, findings in both community and nursing home settings suggest that aggressive behaviors frequently occur in response to unwanted physical contact, are primarily defensive or fear-related, and primarily occur among individuals with advanced cognitive impairment and typically pose the greatest carerelated demand (Bridges-Parlet, Knopman, & Thompson, 1994; Gilley et al., 1997). Moreover, it has been reported that care-receiver behavioral symptoms, usually defined as depression, sleep disturbance, agitation, aggression, and psychosis, play a major role in predicting nursing home placement because they can tax family members' abilities to care for a relative at home (Logsdon et al., 1998).
Development of Caregiver Interventions
Earliest interventions for family caregivers of individuals with AD were based on the Stress Process Model (Lazarus & Folkman, 1984) and focused primarily on providing caregiver information and support around a broad range of issues (Toseland & Rossiter, 1989). Intervention modalities included group psychosocial care or day care services, or a combination of these approaches (Archbold & Stewart, 1996; Bourgeois, Schulz, & Burgio, 1996). Early intervention studies generally had positive subjective outcomes. However, there was mixed evidence regarding their efficacy to change objective outcomes such as emotional distress or coping skills. Effect sizes were generally lower for group interventions (.15 to .31) than for individual interventions (.41 to .58) (Gallagher-Thompson, 1994; Knight, Lutzky, & Macofsky-Urban, 1993). More recent studies have suggested that a variety of approaches may be effective in helping caregivers adjust to their caregiving roles and situations. Among these approaches are interventions that (Coon, Schulz, Ory, & the REACH Study Group, 1999):
* Increase knowledge and understanding.
* Provide emotional and social support.
* Increase skills in managing carereceiver behavioral symptoms.
* Provide community and homebased resources.
* Modify the environment.
* Use technological approaches.
Only one caregiver intervention study, known to the authors, has included immunological measures as outcomes (Garand et al., 2002).
Development of the Caregiver Skill Building Intervention
The development of the CSBI was based on existing clinical and research-based knowledge in both home and institutional settings (Beck, Heacock, Mercer, Walton, & Shook, 1991; Hall & Buckwalter, 1996; Hellen, 1998). The goal of the CSBI was to assist caregivers to improve their skills in three areas:
* Assessment and appraisal of potential causes underlying behavioral symptoms associated with AD.
* Assessment and appraisal of care-receiver's abilities and inabilities.
* Implementation of targeted intervention strategies that decrease behavioral symptoms and accommodate care-receiver's abilities and inabilities.
This intervention moves away from a decrement-based model to one that focuses on care-receivers' remaining abilities (Hellen, 1998; Kielhofner, 1995), and encourages caregivers to be integrative and proactive in their approaches.
This cognitive-behavioral intervention was theoretically supported by social learning theory, which suggests that as individuals increase their self-efficacy or skills, they have a greater likelihood of experiencing lower levels of distress or depressive symptoms (Bandura, Adams, & Beyer, 1977; Perraud, 2000). It was hypothesized that the intervention would be effective because of two interactive elements of self-efficacy: outcome expectations and selfexpectancy efficacy. First, caregivers would believe that given behaviors on their part would lead to improved behaviors on the part of their family member. Second, that the CSBI would help caregivers be more confident that they could successfully execute behaviors that would modify their care-receivers' behaviors (Bandura et al., 1977).
Previous intervention research suggested the following:
* Usual 8- to 10- week intervention periods were not long enough to effect caregiver behavioral changes and to decrease stress (Toseland & Rossiter, 1989).
* Group formats could not respond to all caregiver concerns and uniquenesses (Mittelman, Ferris, Shulman, Steinberg, & Levin, 1996).
* Larger effect sizes were noted with individualized interventions (Knight, Lutzky, & MacofskyUrban, 1993).
To address these issues, the CSBI pilot protocol had two structural components: an initial 5-week group component focusing on common group problems (GRP), and a subsequent, 7-week individualized (IND) telephone component The CSBI targets specific behavioral symptoms that are distressing to caregivers and can potentially affect both caregiver and care-receiver outcomes. It also targets change in specific caregiver skills, such as assessing information concerning behavioral problems, acknowledging care-receiver strengths, and increasing the repertoire of effective interventions. Moreover, it included multiple learning modalities, including group and individual approaches.
DESIGN AND METHODS
Two separate pilot interventions were conducted. Both used a pretest/post-test research design. Descriptive statistics were used to describe the sample. Multivariate repeated measures analysis was used to determine if there were differences in caregiver depressive symptoms and care-receiver behavioral symptoms over time. Paired t tests were used to examine short-term stability of the delayed type hypersensitivity (DTH) skin test. Correlational analyses examined relationships between caregiver immune function and depressive symptoms. Qualitative descriptive procedures were used to describe caregiver concerns and to evaluate the intervention.
Description of the Caregiver Skill Building Intervention
Caregivers participated in a 12week intervention that included a 5week group component and a 7week individualized telephone component. In the group component, caregivers met weekly for 2.5 hours for a total of 5 weeks. Each session covered content on the full range of behavioral symptoms. Caregivers had the opportunity to share their personal experiences concerning behavioral symptoms and learn new skills for approaching these behaviors. Caregivers also received a customized manual and the book, Understanding Difficult Behaviors (Robinson, Spencer, & White, 1994). After completing the 5 -week component, caregivers were contacted weekly by telephone for 7 weeks. Telephone calls allowed caregivers to discuss individual concerns about their relatives' difficult behaviors.
Study participants were family caregivers who were providing home-based care for a person who had a progressive dementia. They were recruited through an AD diagnostic center, an adult day care program, Alzheimer's Association Support Groups, and a senior center. The majority of participants in both groups were women (70%) and all were high school graduates.
Participants in Study 1 consisted of 12 White caregivers with a mean age of 63.99 years (SD = 13.34). The 12 participants in Study 2 represented multicultural groups (African American [75%], White [17%], and Asian [8%]) and had a mean age of 60 years (SD = 10.68).
Measures were administered at baseline, 5 weeks, and 12 weeks. Measures in Study 1 included general demographic information and caregiver depression. In Study 2, carereceiver behaviors and caregiver immune status were added. Measures were limited in Study 1 because of funding and time limitations.
Caregiver depressive symptoms. The Center for Epidemiological Studies Depression Scale (CES-D) (Radloff, 1977) measured caregiver depressive symptomatology. This self-report measure has been used extensively in previous epidemiologic community surveys with a wide range of populations and in previous caregiver studies (Schulz et al., 1995). Internal consistency reliability for this measure ranges from r = .85 to .90, and test-retest reliability ranges from r = .45 to .70. Psychometric properties are comparable with shortened versions of the CES-D and have been highly correlated with the full 20-item CES-D (r = .96) (Schulz et al., 1997). A shorter, 10-item version of the CES-D with Yes or No responses was administered in Study 1 (Kohout, Berkman, Evans, & CornoniHuntley, 1993). A 19-item version of the CES-D, using a 4-point Likerttype response scale, was administered in Study 2. Caregiver depression data were not aggregated for the two pilot studies, because different forms of the depression measure were used.
Care-receiver behavioral symptoms. The short form of the CohenMansfield Agitation Inventory (CMAI) (Cohen-Mansfield, 1986) assessed care-receiver's behavioral symptoms in Study 2 only. The CMAI is a 14-item measure that focuses on agitated behaviors and rates each behavior on a 5-point ordinal scale for frequency of occurrence (1 = never to 5 = a few times per hour). The CMAI has three subscales: aggressive behavior, non-aggressive behavior, and verbally agitated behavior. Internal consistency rehability (r = .86 to .91), inter-rater agreement (.88 to .92), and test-retest rehability (r = .83) have been established (Cohen-Mansfield, Marx, & Rosenthal, 1989; Finkel, Lyons, Anderson, 1992; Koss et al., 1997).
Caregiver immune status. In Study 2, the DTH skin test for cell-mediated immunity (CMI) was administered using the Multitest CMI device. (Kniker, Anderson, McBryde, Roumiantzeff, & Lesourd, 1984). This is a commercially available kit (Merieux Institute, Miami, FL) containing a battery of seven recall antigens (i.e., old tuberculin, tetanus, streptococcus, diphtheria, candida, trichophyton, proteus) and a glycerin control pre-loaded on a puncture applicator. The DTH skin test is an in-vivo measure used to assess the overall functional integrity of the cellular immune system in an intact organism. It is a sensitive index of CMI that represents an in-vivo correlate of in-vitro lymphocyte proliferation responses (Smith & deShazo, 1992). The basis of the DTH response is the ability of the host, in particular the host's T-lymphocytes, to recognize an antigen as foreign. The intensity of cutaneous DTH reactions reflects the adequacy of antigen-specific CMI and is an important indicator of the immune system's ability to resist a specific immunologic challenge (Dhabhar & McEwen, 1996). Findings in an earlier descriptive study of wives who cared for husbands with dementia noted that the DTH skin testing was feasible and that there were inverse relationships between DTH antigen and induration scores and caregiver stressor and distress variables (McCann, 1991).
The CMI device is applied to the volar surface of the right or left forearm using the procedure described by the manufacturer. Skin-test results are evaluated 48 hours after the application (± 1 hour) and follow a standardized protocol. Because of the potential for subjectivity in reading skin-test results, two trained data collectors independently read each skin test. Inter-rater reliability was high, with a Kappa of .80 (p = .005). A positive reaction is defined as induration at an antigen site (mean of two perpendicular diameters) which exceeds (s* 2 mm) induration at the glycerin control site. Erythema without induration is not considered a positive reaction.
Two types of scores are recorded: an antigen score, or the total number of positive reactions; and the total induration score, or the sum (mm) of the average diameters of all positive reactions. Scores are compared to normative standards for age and gender in United States populations (Kniker et al., 1984). A normal response constitutes a positive reaction to two or more antigens. An anergic individual has no positive responses to any of the seven recall antigens. Relative anergy or a hypoergic response extends below the fifth percentile for an individual's age and gender.
Caregivers were tested at the end of the group intervention (Week 5) and again at the completion of the individual telephone follow-up session (Week 12). Each test administration required two visits, one for administration of the skin test and one for reading the results 48 hours later. All participants in Study 2 complied with the test protocol that required a total of four visits to the data collection site.
Concerns expressed during 7week individualized telephone calls. More detailed information concerning the individual telephone intervention component was collected during Study 2. Caregiver concerns in the following areas were reported:
* Instrumental ADLs and personal ADLs.
* Difficult behaviors.
* Other health related concerns.
Program evaluation. Following Study 2, a research assistant who was not involved in the intervention collected information at 5 and 12 weeks related to whether the group was helpful, if caregivers gained new knowledge and information, and whether written materials were helpful.
Caregiver Depressive Symptoms
Changes were noted with caregiver depressive symptoms in both studies. In Study 1 (10-item, shortform CES-D with Yes or No responses), a significant difference existed among the measures at the three time points using multivariate repeated measures analysis (F (2, 9) = 4.71, p = .04). The largest decrease in mean scores occurred between baseline (M = 3.36, SD = 1.43) and 5 weeks (M = 1.45, SD = 1.69), with a slight non-significant increase in mean scores occurring between Week 5 and Week 12 (M = 2.34, SD = 2.16). The effect size in this group was .66. These changes were noted despite the ongoing stressors caregivers experienced while attending the program (e.g., hospitalization of two patients, a wife caregiver was selling her home to move to a retirement center).
In Study 2 (19-item CES-D with a 4-point Likert-type scale), a consistent, but non-significant decrease in caregiver depressive symptoms occurred over three time points (F (2,10) = 2.00, ? = .18) (Baseline M = 3.08, SD = 2.15; Week 12 M = 1.33, SD = 1.67). The effect size for this group was .91 (Table).
Care-Receiver Behavioral Symptoms
Care-receiver CMAI scores were relatively stable with no significant differences over time in the aggression, non-aggression, verbal agitation, or total scores. Multivariate repeated measures analysis revealed no significant changes in total aggressive behaviors (F (2,8) = .34, ? = .72) (Baseline M = 21.60, SD = 3.56; and Week 12 M = 22.8, SD = 3.52).
Caregiver Immune Status
A total of 8 participants (66%) in Study 2 volunteered to receive the DTH skin test at two time points. Despite the small sample, there was a great deal of between-person variability in mean antigen and induration scores. Antigen scores ranged from 1 to 6 at both times (possible range: 0 to 7). An antigen score of 1 indicates a hypoergic or diminished response and an antigen score of 0 indicates cutaneous anergy. Of note, four caregivers (50%) had an antigen score of 1 at Time 1, and three caregivers (38%) had an antigen score of 1 at lime 2; however, only two caregivers had antigen scores of 1 at both time points. No caregivers were completely anergic. Induration scores ranged from 3.25 to 48.4 at Time 1 and from 5.75 to 22.83 at Time 2.
DESCRIPTION OF KEY VARIABLES BY STUDY
As expected, within-person variability in antigen and induration scores was noted between the two time points (Turk, 1980). However, using paired t tests, these differences between time points were not statistically significant. Mean antigen scores were similar at Time 1 (M = 2.25, SD =1.83) and Time 2 (M = 2.5, SD = 1.93), and there were no significant differences in antigen scores between the two time points (t = .33, ? = .75). Results were similar for induration scores. The mean induration score at Time 1 was 14.04 (SD = 14.50) and at Time 2 was 10.61 (SD = 6.37). Again, differences in induration scores between the two time points were not significant (t = - .67; p = .52), suggesting short-term stability of this measure.
The authors hypothesized that higher depression scores in caregivers would be associated with lower antigen and induration scores in caregivers, indicating poorer cellular immune function. Results of the correlational analysis showed negative associations between depression scores (CES-D) and antigen and induration scores. These associations were in the expected direction, but were not statistically significant. Although not identified in the authors' original hypotheses, correlations between the agitated behaviors (CMAI) and antigen and induration scores also were not significant.
Concerns Expressed During 7-Week Individualized Telephone Calls
As frequently noted in the literature, caregivers most frequently reported having difficulty with ADL-related care such as bathing (50%), incontinence (50%), and dressing (42%) (Bridges-Partlet et al., 1994; Gilley et al., 1997; Ryden, Bossenmaier, McLachlan, 1991). Concerns related to IADLs, such as taking medications, poor housekeeping skills, and difficulty getting out of a chair, were reported by fewer caregivers (8% to 25%). Caregivers most frequently reported difficulties with behavioral symptoms such as insomnia (42%), wandering, threatening behavior (42%), and verbal aggression (25%). Confusion was the most frequently encountered memory-related behavior (33%). Mood-related behaviors included apathy, anxiety, and depression, with apathy being most frequently encountered (25%). Of note is that more than 50% of caregivers reported concerns about the care-receivers' physical health.
Study 2 participants reported they benefited from the group interaction (92% good to excellent), gained new knowledge (85%), learned new approaches (75%), and appreciated the written material they received (70%). AU participants reported learning new skills after the 5-week program. Improved and new skills adopted by caregivers included increased skills in assessing disruptive behaviors and understanding underlying causes of these behaviors, increased understanding related to care-receiver abilities and inabilities, and increased number of interventions used. A total of 82% of the caregivers reported they had changed their behavior toward their care-receiver as a result of the intervention. At Week 12, following the 7-week telephone intervention, caregivers reported having a better understanding of why aggression occurs, of how to be more patient with their relative, and of how to blame themselves less for what was happening (70%).
DISCUSSION AND FUTURE DIRECTIONS
These two pilot studies assessed the effectiveness of the CSBI to decrease caregiver depressive symptoms, decrease care-receiver behavioral symptoms, and assess caregiver immune function. The first hypothesis, that caregiver depressive symptoms would decrease after the intervention, was generally supported. Significant decreases were noted in Study 1, and scores in Study 2 decreased, but not significantly. Decreases in caregiver depressive symptoms are supported by other cognitive-behavioral caregiver intervention studies (Bourgeois et al., 1996; Mittelman et al., 1996). Effect sizes in both pilot studies were considerably higher than the effect sizes of an earlier educational support group intervention conducted by the investigators (-.31 to .20) (Farran & Keane-Hagerty, 1994).
Despite a decrease in CES-D scores over time, scores at each time point were considerably lower than those usually reported in the literature. Caregiver mean scores for the 20-item CES-D often range from 14 to 18, with a score of 16 or above indicating that the individual is at risk for clinical depression (Schulz et al., 1995). Further work is needed to determine intervention effectiveness in caregivers who have higher initial levels of depressive symptoms and to determine the clinical significance of these changes with low initial levels of symptoms (Kazdin, 1999).
Pilot study findings did not support the hypothesis that behavioral symptoms would decrease after the intervention. Rather, findings suggested that care-receiver behavioral symptoms were stable during a 12week period of time. Findings concerning care-receiver behaviors in other studies are mixed. One study noted that 1 -month test-retest reliability of the CMAI for patients with AD was .83 (p < .001) (Koss et al., 1997). Another community-based intervention with caregivers of individuals with dementia who successfully learned the intervention reported lower levels of selected behavioral symptoms (Buckwalter et al., 1999). One difficulty in interpreting results across studies is that there is no consistent method by which these behaviors are assessed.
Assessment of caregiver immune status demonstrated stability during a 7-week period, supporting the authors' original hypothesis. This was expected because another cognitive behavioral intervention study conducted with individuals who had cancer noted that immune changes were not observed until 6-months after the intervention (Fawzy et al., 1990). Somewhat unexpected was the lack of significant relationships between immune function and caregiver depressive symptoms, because low to moderate inverse relationships have previously been reported (-.26 to -.42) (McCann, 1991). The lack of significant findings may be caused by the small sample size or the very low mean levels of caregiver depressive symptoms in this study. Findings suggested that the DTH skin test was feasible to use with a multicultural group of family caregivers.
This study also provided preliminary information about concerns expressed by caregivers during the 7week individualized telephone intervention component. Caregivers most frequently reported difficulties with ADL- and IADL-related concerns, yet had concerns in other domains as well (i.e., behavioral symptoms, memory- and mood-related behaviors, other health related concerns). These data support the need to further test an intervention directed toward assisting caregivers with ADL/IADL concerns, but also support the need for including information about problem behaviors and care-receiver health-related issues (Bourgeois et al., 1996; Toseland & Rossiter, 1989).
These preliminary data suggested the CSBI was feasible and demonstrated preliminary effectiveness in reducing caregiver depressive symptoms with both White and AfricanAmerican caregivers. These data support the need to fully implement and test the intervention in a randomized clinical trial with a larger sample of family caregivers and a control group to examine intervention effects on both caregivers and care-receivers. Results suggested that caregivers were open to participating in studies that include more physiologically "invasive" measures. The data also support the feasibility of using the DTH skin test as a measure of caregiver immune function, and support its short-term stability with caregivers of individuals with AD who participate in a cognitive-behavioral intervention.
The majority of existing caregiver intervention studies focus on decreasing caregivers' psychological distress as an outcome (Buckwalter et al., 1999; Farran & Keane-Hagerty, 1994; Toseland & Rossiter, 1989), with one caregiver intervention study, known to the authors, suggesting a cognitive behavioral intervention had positive immunological effects (n = 37) (Garand et al, 2002). An immunologic measure, such as the DTH skin test, provides an opportunity to examine the mechanisms underlying the effect of a cognitive-behavioral intervention on immune outcomes. A larger study including both psychological and immune measures would facilitate the study of these issues. Intervention studies with other clinical populations, such as cancer patients, have demonstrated that behavioral intervention programs can contribute to improved immune function at 6 months (Fawzy et al., 1990) and lower recurrence rates 5 to 6 years later (Fawzy et al., 1993).
A larger study that follows caregivers during a longer period of time and uses additional immunological measures is needed to determine if cognitive- behavioral interventions are effective in improving immunological parameters, and also to determine if interventions can affect physical health outcomes (Archbold & Stewart, 1996). Such a study would address a broad spectrum of nursing issues, including caregiver depressive symptoms; care-receiver behavioral symptoms; and the relationships among caregiver stress, distress, immune, and physical health outcomes.
A number of study limitations should be noted. The sample size was very small and the study did not include a control intervention. The caregiver group was heterogeneous related to the relationship of the caregiver to the care recipient. Given the small sample size, it was not feasible to assess response to caregiver stress based on gender, relationship, or race, nor is it appropriate to offer clinical implications of these findings. The depression scores were also remarkably low. Future studies with larger samples are needed to further test and address these unanswered questions.
As the number of elderly individuals increases, nurses are much more likely to be involved in the care of individuals who have AD, be it in community-based, in-patient, or nursing home settings. If caregiver stress, emotional distress, and changes in physical health are not attended to, the nurse may see two kinds of patients - the family member with AD and the family caregiver. Although the course of AD cannot be modified with currently available medical treatments, it is important for nurses to realize that many interventions can maintain the care recipients' quality of life. Caregivers can be taught how to develop effective skills in coping with the care recipients' cognitive impairment, with difficult behaviors associated with the disease, and with how to provide personal care such as bathing and toileting. Caregivers can also benefit by using support from family and friends and from formal services, such as day care, chore workers, and other respite care.
An area of caregiver intervention that has been most neglected, however, is caregiver self-care. It is important for caregivers to develop a "twotrack" life - one devoted to caring for their family member and die other focusing on self-care. Caregivers need increasing support to attend to their own need for sleep, healthy nutrition, and adequate exercise. When nurses advocate this approach, it is more likely caregivers will experience improved mental and physical health.
- Archbold, P.G., & Stewart, BJ. (1996). The nature of the family caregiving role and nursing interventions for caregiving families. In E.A. Swanson, & T. Tripp-Reimer (Eds.), Advances in gerontological nursing: Volume 1. Issues for the 21st Century (pp. 133-156). New York: Springer Publishing Company.
- Bandura, ?., Adams, N.E., & Beyer, J. (1977). Cognitive processes mediating behavioral change. Journal of Personality and Social Psychology, 35(3), 125-138.
- Beck, C, Heacock, P., Mercer, S., Walton, CG., & Shook, J. (1991). Dressing for success: Promoting independence among cognitively impaired elderly. Journal of Psychosocial Nursing, 29(7), 30-35.
- Bourgeois, M.S., Schulz, R., & Burgio, L. (1996). Interventions for caregivers of patients with Alzheimer's disease: A review and analysis of content, process, and outcomes. International Journal of Aging and Human Development, 43(1), 35-92.
- Bridges-Parlet, S., Knopman, D., & Thompson, T. (1994). A descriptive study of physically aggressive behavior in dementia by direct observation. Journal of American Geriatric Society, 42, 192-197.
- Buckwalter, K.C., Gerdner, L.A., Hall, G.R., KeUy, A., Kohout, E, Richards, B., & Sime, M. (1999). Effects of family caregiver home training based on the progressively lowered stress threshold model. In S.H. Gueldner, Sc L.W. Poon (Eds), Gerontological nursing issues for the 21st century: A muhidisàplinary dialogue commemorating the International Year of Older Persons (pp 8198). Washington, DC: Center Nursing Press, a Division of Sigma Thêta Tau International.
- Cohen-Mansfield, J. (1986). Agitated behaviors in the elderly: Preliminary results in die cognitively deteriorated. Journal of American Geriatric Society, 34, 722-727.
- Cohen-Mansfield, J., Marx, M.S., & Rosenthal, A.S. (1989). A description of agitation in a nursing home. Journal of Gerontology: Medical Sciences, 44(3), M77-84.
- Coon, D.W., Schulz, R., Ory, M.G., & the REACH Study Group. (1999). Innovative intervention approaches for Alzheimer's disease caregivers. In D.E. Biegel, & A. Blum (Eds.), Innovations in practice and service delivery across the lifespan (pp. 295325). New York: Oxford University Press.
- Czarnecki, D., Zalcberg, J., Kulinskaya, E., & Kay, T. (1995). Impaired cell-mediated immunity of apparently normal patients who had multiple skin cancers. Cancer, 76, 228-231.
- Dhabhar, ES., & McEwen, B.S. (1996). Stressinduced enhancement of antigen specific cell-mediated immunity. Journal of immunology, 156, 2608-2615.
- Farran, C.J., & Keane-Hagerty, E. (1994). Multi-modal intervention strategies for caregivers of persons with dementia. In E. Light, G. Niederehe, & B.D. Lebowitz (Eds.), Stress effects on famUy caregivers of Alzheimer's patients: Research and interventions (pp. 242-259). New York: Springer Publishing Company.
- Fawzy, I.F., Fawzy, N.W., Hyum, CS., Elashoff, R., Guthrie, D., Fahey, J.L., & Morton, D. L. (1993). Malignant melanoma: Effects of an early structured psychiatric intervention, coping, and affective state on recurrence and survival 6 years later. Archives of General Psychiatry, 50(9), 681689.
- Fawzy, F.I., Kemeny, M. E., Fawzy, N. W., Elashoff, R., Morton, D., Cousins, N., & Fahey, J. (1990). A structured psychiatric intervention for cancer patients: II. Changes over time in immunological measures. Archives of General Psychiatry, 47, 729-735.
- Finkel, S.I., Lyons, J.S., & Anderson, R.L. (1992). Reliability and validity of the Cohen-Mansfield agitation inventory in institutionalized elderly. International Journal of Geriatric Psychiatry, 7, 487-490.
- Gallagher-Thompson, D. (1994). Direct services and interventions for caregivers. In M.H. Cantor (Ed.), Family caregiving: Agenda for the future (pp.102-122). San Francisco, CA: American Society on Aging.
- Garand, L., Buckwalter, K.C., Lubaroff, D.M., Tripp-Reimer, T., Frantz, RA., & Ansley, T. (2002). A pilot study of immune and mood outcomes of a community-based intervention for dementia caregivers: The PLST Intervention. Archives of Psychiatric Nursing, 16(4), 156-167.
- Gilley, D.W., Wilson, RS., Beckett, LA., & Evans, D. A. (1997). Psychotic symptoms and physically aggressive behavior in Alzheimer's disease. Journal of the American Geriatric Society, 45, 1074-1079.
- Glaser, R., & Kiecolt-Glaser, J.K. (1997). Chronic stress modulates the virus-specific immune response to latent herpes simplex virus type 1 . Annals of Behavioral Medicine, 19(2), 78-82.
- Hall, G.R., & Buckwalter, K.C. (1996). Progressively lowered stress threshold: A conceptual model for care of adults with Alzheimer's disease. Archives of Psychiatric Nursing, 1, 399-406.
- Hellen, C. (1998). Alzheimer's disease: Activityfocused care (2nd ed.). Stoneham, MA: Andover Medical Publishers.
- Kazdin, A.E. (1999). The meanings and measurement of clinical significance. Journal of Consulting and Clinical Psychology, 67(3), 332-339.
- Kielhofner, G. (1995). A model of human occupation: Theory and application (2nd ed.). Baltimore, MD: Lippincott Williams and Wilkins.
- Knight, B.G., Lutzky, S.M., & MacofskyUrban, F. (1993). A meta-analytic review of interventions for caregiver stress: Recommendations for future research. The Gerontologist, 33(2), 240-248.
- Kniker, WT, Anderson, CT, McBryde, J.L., Roumiantzeff, M., & Lesourd, B. (1984). Multitest CMI for standardized measurement of delayed cutaneous hypersensitivity and cell-mediated immunity. Normal values and proposed scoring system for healthy adults in the USA. Annals of Allergy, 52(2), 75-82.
- Kohout, F.J., Berkman, L.F., Evans, D.A., & Cornoni-Hundey, J. (1993). Two shorter forms of the CES-D depression symptoms index. Journal of Aging and Health. 5(2), 179-193.
- Koss, E., Weiner, M., Ernesto, C, CohenMansfield, J., Ferris, S.H., Grandmati, M., Schafer, K., Sano, M., Thal, L.J., Thomas, R., & Whitehouse, P.J., & the Alzheimer's Disease Cooperative Study. (1997). Assessing patterns of agitation in Alzheimer's disease patients with the Cohen-Mansfield Agitation Inventory. Alzheimer Disease and Associated Disorders, 11(2), S45-S50.
- Lazarus, RS., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer Publishing Company, Inc.
- Logsdon, R. G., Teri, L., McCurry, S.M., Gibbons, L.E., Kukull, WA, & Larson, E.B. (1998). Wandering: A significant problem among community-residing individuals with Alzheimer's disease. Journal of Gerontology: Psychological Sciences, 53B(5), P294-P299.
- McCann, JJ. (1991). Effects of stress on spouse caregivers' psychological health and cellular immunity. Dissertation Abstracts International, 52, 4114B (UMI No. 9200779).
- Mittelman, M.S., Ferris, S.H., Shulman, E., Steinberg, G, & Levin, B. (1996). A family intervention to delay nursing home placement of patients with Alzheimer's disease. Journal of the American Medical Association, 276(21), 1725-1731.
- National Institutes of Health. (1999). Progress report on Alzheimer's disease (NIH Publication No. 99-4664). Silver Spring, MD: U.S. Department of Health and Human Services, Public Health Service.
- O'Donnell, B.F., Drachman, DA., Barnes, HJ., Peterson, K.E., Swearer, J.M., & Lew, RA. (1992). Incontinence and troublesome behaviors predict institutionalization in dementia. Journal of Geriatric Psychiatry and Neurology, 5(1), 45-52.
- Perraud, S. (2000). Development of the depression self-efficacy scale (DCSES). Archives of Psychiatric Nursing, 14, 276-284.
- Radloff, L.S. (1977). The CES-D scale: A selfreport depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.
- Robinson, A., Spencer, B., & White, L. (1994). Understanding difficult behaviors: Series on Alzheimer's Disease and related illnesses. Ypsilanti, MI: Eastern Michigan University.
- Ryden, M.B., Bossenmaier, M., & McLachlan, C (1991). Aggressive behavior in cognitively impaired nursing home residents. Research in Nursing and Health, 14, 87-95.
- Schulz, R., Newsom, J., Mittlemark, M., Burton, L., Hirsch, C, & Jackson, S. (1997). Health effects of caregiving: The caregiver health effects study: An ancillary study of the cardiovascular health study. Annals of Behavioral Medicine, 19(2), 110-116.
- Schulz, R., O'Brien, AT, Bookwala, J., & Fleissner, K. (1995). Psychiatric and physical morbidity effects of dementia caregiving: Prevalence, correlates, and causes. The Gerontologùt, 35(6), 771-791.
- Smith, D.L., & deShazo, RD. (1992). Delayed hypersensitivity skin testing. In N.R. Rose, E.C. de Macario, J.L. Fahey, H. Friedman, & GM. Penn (Eds.), Manual of clinical Uboratory immunology (4th ed., pp. 202-206). Washington, DC: American Society for Microbiology.
- Toseland, RW, & Rossiter, CM. (1989). Group interventions to support family caregivers: A review and analysis. The Gerontohgist, 29(4), 438-448.
- Turk, J.K. (Ed.). (1980). Delayed hypersensitivity (Research Monographs in Immunology, Vol. 1). New York: Elsevier/North-Halland.
DESCRIPTION OF KEY VARIABLES BY STUDY