FACTORS PREDICTING QUALITY OF LIFE IN DEMENTIA PATIENTS

by Dr.Sandra Burgener

SBurg@www.iupui.edu

This research is funded by the Institute of Nursing Research

National Institutes of Health Academic Investigator Award

NR00074-KO7

RESEARCH PLAN

1. SPECIFIC AIMS:

The patient with Alzheimer's Disease (AD) has received a great deal of attention from the research community, with the majority of the research being directed to the study of progression of AD; identification of dysfunctional behaviors; behavioral control, through pharmacological, environmental, and social interventions; and management of the disease process.1,2 As Lawton and colleagues3 and Cotrell and Schulz4 have recently explicated, the problem with this approach is the lack of attention to and understanding of both the patient's perspective of specific patient needs and the effectiveness of interventions in facilitating the patient's adaptation to AD and positive patient outcomes, rather than attention to negative aspects of the disease. Identification of specific patient needs and factors that facilitate positive patient outcomes is a necessary precursor to identifying additional variables for assessment and subsequent intervention development. Facilitating positive patient outcomes, although a worthwhile outcome in itself, also has the potential of influencing other aspects of this devastating disease process. For example, recent studies have supported the relationship between patient and caregiver behaviors, suggesting that more positive patient behaviors may result in a more positive caregiving experience.5-7 Considering the wealth of research identifying the influence of difficult patient behaviors on caregiver burden, interventions designed to increase positive patient behaviors should decrease caregiver burden, possibly influencing institutionalization rates as well.8 Lawton3 describes the "stripping of personhood" that is common with AD patients and the resulting demoralizing effect this has on both institutional and family caregivers. This lack of attention to the patient's personhood robs the family caregiver of a previously valued relationship as well. Lawton suggests that attention to the AD patient's personal states and needs can increase maintenance of the patient's personhood, provide positive feedback for caregivers, and enrich the total caregiving experience. Additionally, interventions designed to increase positive patient outcomes should decrease the prevalence of depressive disorders in AD patients, found to occur in as many as 87% of AD patients, with higher levels of depression noted in the early disease stages.9,10 Interventions designed to decrease depressive disorders have the potential to decrease the costs of care and, importantly, increase the patient's quality of life. This study addresses this previous lack of attention to the AD patient's needs and factors that influence the patient's adaptation to the disease through a systematic study of variables related to positive patient outcomes in the early stages of the disease process.

Given the inherent difficulty in the study of the AD patient's personal perspective and the state of development of objective outcome measures with AD patients, the purposes for this award include both research outcomes and developmental activities necessary to conduct the proposed descriptive, longitudinal study. Since both quantitative and qualitative research methods will be utilized to adequately address the complexity of data collection and analyses, inherent in the study are activities devoted to the development and testing of semi-structured interview guides, interview skills, and an objective outcome measure for use with cognitively-impaired persons. The research and developmental aims of the study reflect the varied approaches to the study's conduct and include:

Research Aim #1: Using Lawton's3 model of quality of life in cognitively impaired persons as a guiding framework, the central study purpose is to examine both patient-centered and external variables associated with quality of life indicators in AD patients during the first 18 months following the disease diagnosis.

Research Aim #2: Using qualitative approaches, the patient's perceptions of needs and subjective responses to the disease diagnosis and progression will be assessed and analyzed to identify additional factors, not included in the quantitative analyses, contributing to quality-of-life outcomes.

Research Aim #3: The third study purpose is to examine the reliability and validity of a newly developed, objective outcome indicator of the cognitively impaired person's adaptation to the disease process, the "Psychological Well-Being in Cognitively-Impaired Persons" (PWB-CIP) measure. Although initial testing of the PWB-CIP measure's psychometric properties is currently being conducted, a more extensive testing of its validity and reliability will be carried out with the target population, patients diagnosed with AD living in the home setting.

Developmental Aim #1: Expertise in constructing qualitative interview guides and qualitative analyses will be gained through consultation with Dr. Kathy Knafl, University of Illinois, a consultant for this award and a recognized expert in qualitative research methods. Dr. Knafl will also be consulted throughout the qualitative data collection and analyses to assure proper conduct of procedures and correct interpretation of the data. Additional expertise in qualitative research methods will be gained through coursework in qualitative research available through the Department of Sociology at Indiana University's Bloomington campus.

Developmental Aim #2: Development of appropriate interview skills will facilitate qualitative data collection and assure a competent, sensitive approach with patient and caregiver subjects. The award sponsors, Dr. Kathleen Buckwalter and Dr. Mary Austrom, possess knowledge relevant to this aspect of the study and provide the necessary expertise to assure the development of interview skills appropriate for the target population.

Developmental Aim #3: Expertise in conducting psychological assessments will be required to carry out patient assessments in the home. Clinical observation, supervised clinical experiences, and coursework will assist in development of assessment skills.

Developmental Aim #4: Increased expertise in instrument development and psychometric evaluation will be gained to facilitate meeting research aim #3. Consultation with Dr. Victoria Champion, recognized for her knowledge of instrumentation, and coursework available at the Indianapolis campus will assist in meeting this aim.

2. BACKGROUND/SIGNIFICANCE:

NEED FOR RESEARCH:

The health care needs of a rapidly growing elderly population are a nationally established priority. At present, 12% of the population is over 65, and it is expected that by the year 2020, 17% of the population will be over age 65.11 Of particular concern is the increasing prevalence of irreversible dementia in this growing elderly population. Twelve percent of persons over 65 have irreversible dementia and the percentage rises to 47.9% in elders over 85.12 Psychological responses of patients to the disease are currently reported mostly as occurrence of depression and behavioral consequences, usually defined as difficult behavioral problems. Depression rates in dementia patients are reported as high, estimated to occur in 40%13 to 87%9 of patients, with the most obvious depressive symptoms occurring in the early stages of the disease. Occurrence of behavioral problems in home-dwelling patients ranges from 50%14,15 to 90%.16

Development of appropriate outcome indicators for dementia patients continues to lag behind other areas of research, with the evaluations of the efficacy and appropriateness of most patient outcomes being from the researcher's, rather than the patient's, perspective.17 Without objective measures of important patient outcomes, such as well being or quality of life, evaluation of research outcomes for the dementia patient will remain ambiguous and narrowly focused, at best. Cotrell and Schulz4 describe the tendency of researchers to relegate the AD patient to the status of "object, rather than legitimate contributor" to the research process. They also point to the gains in understanding the disease process and appropriate care approaches resulting from a systematic study of the patient's perspective.

The dearth of research regarding the impact of AD on the person can be well understood considering the methodological and pragmatic constraints surrounding the study of persons with diminished ability to think and reason. Many traditional research instruments require the subjective, rational responses of subjects, and observational measures provide only limited information. Evidence, although limited, exists supporting the reliability of subjective responses from cognitively impaired persons and the meaningfulness of the subjective data.18-19 Given the many possible positive outcomes, including the development of interventions designed to meet patient-identified needs, the importance of understanding the impact of AD on the person from a personal, rather than a purely biophysical, perspective can no longer be ignored. Cohen17 hypothesizes that "with the exception of psychosis, many of the psychiatric symptoms and behavioral disturbances in early and middle stages of AD result when patients, family members, and health care professionals are unable to perceive and cope successfully with the deterioration or reactions to the deterioration" (p11). With AD patients, however, no systematic study of the patient's personal and psychological needs and responses or additional variables leading to adaptation and quality-of-life outcomes can be found, indicating the great need for longitudinal studies addressing this gap in knowledge.

THEORETICAL BASIS: The definition of quality of life in cognitively impaired persons evolving from the work of Lawton and colleagues3 serves as the organizing framework for this study. Lawton's formulation of "the good life" includes four sectors or domains: 1) psychological well-being, consisting of the person's affective states, which include anxiety, depression, agitation, and positive affective states; 2) behavioral competence, including physical health, functional ability, and cognition, along with the more complex domains of time use and social behavior; 3) the objective environment, consisting of structured events and the architectural features of the environment; and 4) perceived quality of life.

Lawton's3 conceptualization of quality of life contains one domain that requires subjective evaluation and a second domain, the objective environment, that may be influenced most by the caregiver or other family members, rather than the AD patient. Two quality-of-life domains, however, psychological well-being and behavioral competence, are reflected by individual factors within the patient, more likely to be amenable to interventions directed to positive patient outcomes, and can be assessed through objective measures. Positive and negative affect, representing psychological well-being, are reflected in behaviors available for assessment by an objective observer. The person-centered abilities (physical, functional and cognitive) inherent in the behavioral competence domain are readily available for assessment, along with the patient's social behaviors. Lawton3 argues for the need to clearly and distinctly measure cognitive function and both types of activities of daily living (ADL's), instrumental and functional, because one does not necessarily correlate well with the other. Lawton states that quality of life is an appropriate organizing framework for AD patients, not only because of the ability to operationalize the major concepts with this population, but because of the relevance of this outcome for both the patient and caregiver. He notes that negative patient responses or outcomes are not only deleterious for the patient, but can produce negative responses in others as well. He emphasizes the importance for both patients and caregivers of focusing on and examining positive patient outcomes.

Although the work of Lawton and colleagues3 has centered mostly around identification and measurement of quality-of-life domains in institutionalized AD patients, other research in quality of life and psychological well being suggests variables that may be related to quality-of-life outcomes. The meaning of a stressful event to the person has also gained support as being an important aspect of positive adaptation to the event, with more positive meaning being associated with more positive emotional and psychological outcomes.20-21 Although the role of social support in relation to positive adaptational responses is not clearly defined within the literature, both perceived and actual support (i.e. involvement in social networks and number of social contacts) have been hypothesized to influence positive outcomes in cognitively intact populations.22-24 Spiritual support, one aspect of social support, has been found to be associated with positive outcomes, including well being, and is increasingly recognized as a valuable component of social support.25-26 Continued participation in religious activities, both public and private, has been associated with higher morale,27 improved health, less anxiety, and lower rates of depression in cognitively intact persons.28 Recent writings have given some evidence of the importance of maintaining meaningful activities to AD patients, including the importance of recognizing and supporting previous skills and positive behaviors.29-30 Maintenance of the AD patient's self-care ability has been found to be related to increased functional behaviors in several studies, indicating the importance of facilitation of self care in patients.5,6,13 Also, the effect of the quality of the patient/caregiver relationship, especially a marital relationship, on patient outcomes has received initial support in the research literature.31 Although the literature on maintenance of personality in dementia has generally supported the existence of personality changes throughout the disease,32-33 findings of a recent study by Lawton and colleagues34 indicates a persistence and coherence of personality in highly impaired institutionalized patients, using personality traits measures from a variety of sources.

Using findings from the above studies within a quality-of-life framework suggests that both person-centered (meaning of the event, personality, maintenance of past activities) and external (social support, quality of patient/caregiver relationship, facilitation of self care) variables are relevant to quality-of-life outcomes. As many of these variables are amenable to intervention, examining the individual and combined influence of these variables on quality-of-life outcomes should lead to the development of interventions designed to increase positive patient and, ultimately, caregiver outcomes. Although personality represents a stable trait not readily amenable to alteration, understanding the influence of personality on subsequent behaviors in AD patients will increase knowledge regarding what types of patients are most likely to require more intense intervention, increasing the potential to target interventions for maximal effectiveness and use of resources.

The relationships among variables described above and quality-of-life indicators inherent in Lawton's model are displayed in Figure 1. It is expected the results from this study will inform the model's further development.

PAST RESEARCH: Although systematic studies of the variables predicting quality-of-life outcomes with AD patients were not found in the extant literature, several related areas inform the proposed study. First, studies will be reviewed examining the incidence and presentation of affective states, generally described as psychiatric symptoms such as depression, in AD patients. Next, studies describing behavioral responses related to psychological factors, such as loss of control, will be described. Finally, reports of the existence and relevance of individual patient characteristics and responses to AD will be reviewed.

Affective states in AD Patients: Studies investigating various affective states evident in AD patients have largely focused on the incidence, manifestation, and management of psychiatric diagnoses such as depression, with little attention directed to whether the patient's awareness, adaptation to increasing impairment, or increasing social isolation might be contributing factors to observed symptoms. One of the most common psychiatric diagnoses in AD patients is depression, reported to occur in a majority of patients. Merriam, Aronson, Gaston, Wey, and Katz9 found a high rate of psychiatric symptoms in 175 AD patients. A total of 87% of the patients were judged to be dysphoric, with symptoms of depression, sadness, fearfulness, excessive worrying, or hopelessness, and 93% exhibited a loss of interest and pleasure. Parmelee, Katz, Lawton35 found similar affective symptoms in patients defined as cognitively impaired, with depression rates ranging from 7.5% to 23% of institutionalized persons and 2% to 9% of congregate apartment residents. Pearson, Teri, Reifler, and Raskind10 examined depressive symptoms in 50 home-dwelling AD patients using DSM-III criteria. They found AD patients classified as depressed (n=20) were less cognitively impaired and earlier in the disease process that non-depressed patients. Controlling for cognitive status, the depression diagnosis was found to have a main effect on functional impairment, supporting the impact of negative affective states on other aspects of functioning. These findings support the earlier results of Reifler, Larson, and Hanley36 examining depression in 88 cognitively impaired outpatients. They found coexisting depression decreased significantly with increased severity in cognitive impairment, occurring in 33% of mildly impaired patients and decreasing to 12% of severely impaired patients. Increased levels of depression at earlier disease stages has been a consistent finding in AD patients.

Reifler and Larson37 make the distinction between depression as a cause of dementia and depression as a complication of dementia, emphasizing that depression in AD is not solely an emotional disorder on which dementia is superimposed. They propose that AD patients become increasingly depressed as they are faced with numerous losses, failure in self-care and functional ability, and loss of energy and interest in life. Furthermore, they found that psychosocial interventions were an important addition to pharmacologic treatment, evidenced by the effectiveness of the placebo treatment in the antidepressant clinical trials. These previous studies of affective states in AD patients have focused largely on evaluation of affective disorders in AD patients, with little attention to the benefits of interventions that affect the patient's experience of the disease and general well-being.

Behavioral Responses Associated with Psychological Factors: Behavioral problems in AD patients are well documented, with most research focusing on the manifestation of behaviors, prevalence rates, measurement issues, and the impact of patient behavioral problems on the caregiver. Reisberg and colleagues14 found the most commonly reported behavioral problems were delusions, nonspecific agitation, and sleep disturbances. Behavioral problems occurred in 50% of the patients, with a total of 23 different behavioral symptoms reported. Teri et al.13 examined behaviors of 56 home dwelling Alzheimer's patients and found that problems with cognitive functioning (memory, confusion, and disorientation) were most prevalent, occurring in 64% to 84% of patients, while high activity levels and emotional distress were reported in 20% to 43% of the sample. No caregiver reported an absence of behavioral problems. Decreased function in both instrumental and self-care activities of daily living was related to the total occurrence of behavioral problems (r =-.43 and r =-.26, respectively) and to the total duration of behavioral problems (r= -.33 and r =-.28).

Some empirical evidence exists concerning the role of psychological and environmental factors as precipitators and/or mediators of problem behaviors. Snyder and colleagues38 suggest that wandering behaviors, commonly exhibited in dementia patients, may be related to the patient's previous coping styles, a search for security, or previous activity patterns. Verwoerdt39 further suggests that many behavioral symptoms of dementia may be a combination of both mental impairment and the psychological defenses the patient chooses. Reisberg and colleagues14 interpret the patient's behavioral symptoms as reflective of adaptation or communication patterns.

In an outpatient setting, Swearer, Drachman, O'Connell, and Mitchell40 document aggression, angry outbursts, paranoia, wandering, repetitive manipulations of objects, and vegetative disorders such as disturbed sleep and incontinence. They noted one or more of these behavioral problems in 83% of 126 AD patients. Because aggressiveness correlated only weakly with degree of intellectual impairment, the findings suggest that behaviors are not just a simple product of intellectual deterioration. From observations of 680 patients in an outpatient diagnostic clinic, Cooper, Mungas, & Weiler41 found significant, but rather low, correlations between cognitive function and the presence of difficult behaviors in AD patients, r =-.09 to r =-.26, suggesting that cognitive function may account for only a small amount of the variance in difficult behaviors.41 More difficult behaviors were found at lower mental status levels, however.

Meddaugh6 found in 9 of 14 patients defined as aggressive that aggressive behaviors were connected to loss of freedom of choice.9 Whall used an observational method in a series of descriptive pilot studies to identify "triggers" for dysfunctional behaviors and support for a behavior-modification approach. She found that specific caregiving approaches, including giving the patient more freedom of choice, were related to reduced incidence of dysfunctional behaviors. In a study of 124 institutionalized patients, Ryden and colleagues42 found that 72% of aggressive behaviors were preceded by touch or invasion of personal space associated with caregiving episodes, such as movement, dressing, or toileting. In an observational study of 58 patient during caregiving episodes, Burgener and colleagues5 found the patient's increased self-care ability was positively related to functional behaviors, supporting the findings by Teri et al.13 in an out-patient setting. These combined studies indicate that patient perceptions and psychological factors, such as perceived loss of freedom of choice, invasion of personal space, and maintenance of self-care activities, are related to difficult patient behaviors, supporting the importance of an increased understanding of the patient's perceptions and interpretations.

Maintenance and Accessibility of Personal Responses and Characteristics: Evidence does exist for varying responses to the disease process, indicating that, while the disease causes generalized destruction to the brain itself, individuals can and do respond quite differently in behavioral and emotional outcomes. Burgener, Shimer, & Murrell28 and more recently Buckwalter, et al.,29 have presented numerous examples of how individuals retain and utilize past behaviors and skills, while expressing individual responses to the disease. The recent work by Quayhagen and Quayhagen43 using a cognitive stimulation program points to the patient's responsiveness through observed increased feelings of hope and confidence. Teri and Gallagher-Thompson44 have found that a cognitive therapy program for Alzheimer's patients results in patients generating more adaptive thoughts appropriate to their situation. Consistency in the completion of activities and increased concentration indicate the relevance of the activity to the patient. In a book entitled Loss of Self45 and more recently in an article17 supporting the subjective experience of AD, Cohen describes various ways that AD patients evidence human qualities such as faith, hope, the will to live, and love, even into the latter disease stages. She states that understanding of the patient's perceptions, beliefs, and desires is necessary to assure patient health and well being. She also points to the need for empirically based psychological, behavioral, and family therapeutic interventions designed to affect quality of life in patients and families. These combined studies provide evidence of individual responses to the disease process, the meaningfulness of patient perceptions and desires, and the ability of researchers to assess patient perceptions.

Limitations of Previous Studies:

While the research evidence to date points to probable relationships between patient perceptions and emotional and behavioral responses, there are no studies systematically examining the patient's perspective and these relationships. The literature does support the existence of individual patient responses, at times based on past abilities and skills, indicating the continued existence of individual characteristics, despite the effects of the dementing process. Affective patient responses are documented, with a clear indication of the extent and manifestation of depressive symptoms. What is not clear, however, is how depressive symptoms are influenced by individual patient losses, perceptions, and adaptational responses. It is also not clear how external factors, such as family and social support, reassignment of roles, changes in family relationships, or the family's facilitation of the patient's self-care ability, influence the patient's adaptational and quality of life outcomes. The relationship between loss of functional ability and difficult behavioral problems has been documented through empirical studies, giving support to the importance of maintaining personal abilities and possibly reflecting the relationship between self-esteem and personal control and resulting behavior. These collective findings support the need to understanding the patient's perspective, especially in the early and middle disease stages, and how the patient's adaptation is affected by both external and person-centered factors. No studies currently funded by either National Institute of Aging or National Institute for Nursing Research were found that focused on assessment of the relationships among person-centered variables, including the patient's perception of the impact of the disease; external factors; and quality-of-life outcomes.46

3. PRELIMINARY STUDIES:

The principal investigator (P.I.) has completed four studies related to the proposed research, including two studies done during a post-doctoral fellowship at the University of Rochester, Rochester, NY. These studies support the strength of the P.I.'s background in conducting the proposed study and effectively disseminating its results. The proposed study is a logical next step in this program of research. The P.I. furthered her knowledge and development at Wayne State University where she completed coursework in two cognate areas: gerontology and interpersonal communication. In addition, the P.I. is a certified American Nurses' Association Gerontological Nurse Practitioner and has practiced with elderly patients since 1986. This clinical practice has provided first-hand knowledge of patient responses to the disease.

STUDY #1: The first study conducted by the P.I. was a dissertation study focusing on the relationship between the behaviors of caregivers and difficult behaviors of cognitively impaired elders, funded through an individual National Research Service Award, #3-30804. A dementia specialty unit in one long-term care facility was used. Naturally occurring caregiving interactions (assistance with dressing) and interpersonal contacts were observed, allowing for analysis of 239 caregiver/patient interactions. Although the central purpose of this study was to examine the relationship between caregiver and patient behaviors, it also was the P.I.'s first extensive exposure to the varying response patterns of AD patients. A lack of attending to the patient's emotional responses, other than controlling of difficult behaviors, by institutional caregivers was evident from the observations in the study. Two publications resulted from the dissertation study:

*Burgener, S., & Barton, D. (1991). Nursing care of cognitively-impaired, institutionalized elderly. Journal of Gerontological Nursing, 17, 37-43.

Burgener, S., & Barton, D. (1991). Caring for cognitively-impaired elderly persons in public institutions: The effect of environmental and interactive variables on behavior. Proceedings of the First Annual Conference for Public Sector Psychosocial Nursing, Univ. of Washington, Seattle.

STUDY #2: The dissertation study results indicated the need for further examination of the relationship between caregiver and patient behaviors, using a larger, more diverse sample and a variety of variables. A study was, therefore, conducted as part of a post-doctoral experience funded through the Robert Wood Johnson Foundation. Dr. Kathleen Buckwalter, Professor, University of Iowa, served as an external advisor to the study and will serve as a sponsor to the proposed study.

Although the central purpose of the post-doctoral study was to further examine the relationship between caregiver and patient behaviors, a secondary purpose was identified after several months of observation in two long-term care institutions, one large, public facility and a second, smaller, private facility. Both special dementia units and general units were used at both study sites with a total sample size of 116 subjects (58 patient and 58 caregiver subjects). Through observations of caregiving tasks, a continued lack of attention to the patient's emotional responses by institutional caregivers was again evident, even with a much larger sample and more diverse settings, including a small, religiously affiliated facility. It was also noted that clergy and rabbis were not observed visiting the patient units, offering spiritual support to patients. As a result of these observations, semi-structured interviews were conducted with a subset of 10 patients to elicit personal responses to the disease, with specific questions about the importance of practicing past religious behaviors, such as praying. Interviews occurred at several points in time, usually at 2-week intervals.

The results of the interviews were surprising for several reasons. Patients who were fairly advanced in the disease process, usually at a middle to late-middle stage of the disease, often provided very coherent and meaningful responses to the interviews. A consistency in responses was also found in most patient responses, with the same patterns of response, or even the same exact wording, provided on repeated interviews. One patient, for example, talked consistently about his wife and son and how much missed being with them. A second patient openly talked about the comfort she experienced when praying, often saying, "When I pray, I can feel my mother pat my head (she would demonstrate this), like she used to when I was a child". Most patients stated they prayed every day, giving some indication of the importance of the spiritual aspects of their lives, despite the disease process. These findings point to the likelihood of obtaining meaningful data from patient interviews, especially in light of the consistency of responses from patients over several interview sessions.

Several other findings support the importance of examining the patient's response to the disease process. The correlation between the patient's calm/functional self-care ability and patient functional behaviors was moderate, but significant (r =.36). A relationship between the patient's self-care ability and more functional behaviors was also noted in casual observations. Especially during bathing, when patients would become increasingly agitated, allowing the patient to do some self-care, such as washing his/her face or assisting with dressing, seemed to result in almost immediate calmer behaviors. This consistency in the quantitative and observational findings support the potential benefits of examining patient responses to the disease, since the personal variables, such as maintenance of situational control and self-care ability, were found to be related to behavioral responses, reflective of coping responses in adaptational models.

Two of the following publications resulting from the post-doctoral study describe the central quantitative findings from this study, while the Image: Journal of Nursing Scholarship and Journal of Gerontological Nursing articles focus on the interview and observational findings described above:

*Burgener, S. C., Jirovec, M., Murrell, L., & Barton, D. (1992). Caregiver and environmental variables related to difficult behaviors in institutionalized, demented elderly persons. Journals of Gerontology, 47, P242-P249.

*Burgener, S., & Chiverton, P. (1992). Conceptualizing psychological well-being in cognitively impaired elderly persons. Image, 24, 209-213.

*Burgener, S., Murrell, L., & Shimer, R. (1993). Expressions of individuality in cognitively impaired elders: The need for individual assessment and approaches to care. Journal of Gerontological Nursing, 19(4), 13-22.

*Burgener, S., & Shimer, R. (1993). The relationship between environmental and person-centered variables and caregiver behaviors. Research in Nursing and Health, 16(3), 193-202.

*Refereed Journal.

STUDY #3: A second study was conducted as part of the post-doctoral fellowship at University of Rochester that focused on the perceived support for caregivers and patients from religious organizations. A questionnaire containing a religiosity scale, caregiver stress scale, caregiver well-being scale, and specific questions about support from the religious community was sent to caregivers identified by the local Alzheimer's Association chapter. Of the 600 questionnaires sent, 182 were returned, with 135 respondents currently providing care representing the caregiver sample. Caregivers were found to have a lower level of met need for contact with the church/synagogue compared to control subjects. Also, significant inverse relationships were found between the extent to which the caregiver's need for contact with the church/synagogue was met and life upset, indicating lower levels of stress as the need for contact was met. The same trend was found between the extent to which the caregiver's spiritual needs were met and life upset. Current levels of church/synagogue attendance were related to both general well-being and social functioning. Perceived meeting of need for contact with the church/synagogue was positively related to well-being and mental health. Extent to which caregivers perceived their spiritual needs were met was associated with increased well-being and improved mental health.

More importantly, several caregivers expressed a desire for attention and participation by the AD patient in church/synagogue activities, including more frequent visits by clergy/rabbis to the patient. One respondent stated, "What frustrates me so is that the Alzheimer's `victims' are written off at diagnosis. They become non-persons. People do not talk to them anymore - they talk over them or about them, etc. My mother needs peer friendships, religious opportunities. She needs people in her life besides me and the aides we hire. We are not peers; we provide care, not friendships." This desire for more involvement by and attention to the AD patient was reflected in similar comments by caregivers, suggesting the difference in satisfaction with contact with the church or synagogue may be in relation to the patient rather than the caregiver.

While the central purpose of this study was to examine the relationship between religious support and caregiver well-being and stress, the study findings also support both the importance of religious support as an important component of social support for older persons in general and the importance of attending to the patient's need for continued support from and contact with the religious community. One publication resulted from this study:

Burgener, S. C. (1994). Caregiver religiosity and well-being in dealing with Alzheimer's dementia.

Journal of Religion and Health, 33(2), 175-189.

STUDY #4: Based on results of previous studies, interventions designed to facilitate patient care in the home setting were developed into a bipartite intervention model for testing with family caregivers. Forty-nine caregiver/patient pairs who completed the study were placed in one of three treatment groups (group 1: received both intervention components; group 2: received the educational program only; group 3: received the caregiving modification booklet and role-modeling of caregiver approaches associated with calm, functional patient behaviors and increases in the patient's self-care behaviors). A fourth group, a delayed-treatment group, served as a control. One important aspect of the intervention was the focus on increasing the patient's involvement in care, particularly approaches to caregiving allowing the patient opportunities for self-care, control during caregiving procedures, and maintenance of past behaviors.

Only the quantitative findings of interest will be addressed here. On the six-month post-intervention assessment, a slight increase in the patients' mean functional self-care scores was evident for the treatment group receiving the caregiving modification booklet and role-modeling (+1.63), while scores decreased slightly for the group receiving the education intervention and control group patients. The patients' mean difficult behaviors scores increased for all three groups, with a larger increase for control patients (+3.0), compared to treatment group patients (+1.32).

Although effect sizes are not large, the findings do support the effectiveness of interventions designed to increase the patient's self-care ability and control. These interventions were also somewhat effective in limiting the patient's development of difficult behaviors, compared to a control group. These findings point to the possible benefits of an increased understanding of the patient's response to the disease and attending to the patient's personal need for autonomy and control. Of the 49 caregivers interviewed at several points over the 6-month study period, however, only 3 caregivers indicated they openly talked with the patient about his/her response to the disease. Most caregivers expressed the belief that talking with the patient would not be helpful due to the patient's limited cognitive ability. One caregiver, a daughter, who did talk with her father regularly about his responses to the disease (her mother, the primary caregiver, refused to talk openly with the patient), stated that her father appeared relieved to have an opportunity to talk about the disease and about his responses and needs. The daughter also stated she felt this type of open discussion was somewhat responsible for her father's positive adjustment to the disease, including her father's continued use of effective coping strategies, such as exercise, humor, and maintaining interest in past activities. Manuscripts to be submitted for publication from this study are currently under preparation.

Summary of outcomes from past studies: The findings from the P.I.'s previous studies support the need and methods for the proposed study. Across studies, both in institutional and home settings, an inattention to the patient as a person, especially lack of concern for the impact of the disease on the patient's emotional responses and needs, was evident. Instead, patients became "non-persons", as described explicitly by one caregiver, and activities and care were done for the patient, not by or with the patient. The quantitative findings from these previous studies support the importance of allowing the patient control and self-care opportunities in varying caregiving situations. The findings also support a relationship between increasing patient control and self-care and lower levels of difficult behaviors, giving evidence of the relationship between the patient's personal responses (need for autonomy) and resulting behaviors. These collective findings point to the need to examine the patient's perspective and give recognition to the importance of the patient in the disease process. Institutional and family caregivers will not change current practice until evidence is available to support the importance of recognizing the patient as a person, giving credence to his/her need for psychological, social, and spiritual support. Additionally, the findings suggest that meeting the patient's need for autonomy and recognition (evidenced by responses to increased control and self-care), can result in positive behavioral responses, supporting the possible positive outcomes of designing interventions to meet the patient's psychological and quality of life needs.

4. RESEARCH DESIGN AND METHODS:

RESEARCH QUESTIONS: To address the three study research aims, the following research questions will guide the study:

Research Question #1: What are the relationships among person-centered variables (meaning of event, personality traits, maintenance of past activities), external variables (social support, quality of caregiver/patient relationship, facilitation of patient self care), and quality-of-life outcomes in AD patients over the first 18 months following a dementia diagnosis?

Research Question #2: What are common patient responses to the dementia diagnosis in terms of meaning of the illness, psychological needs, perception of control, behaviors of others, and perceived loss or benefit?

Research Question #3: How do patient responses to the disease process change over time, specifically the first 18 months following diagnosis?

Research Question #4: What is the validity and reliability of the PWBCIP measure when used with AD patients in the early and early-middle disease stages?

DESIGN: Two research approaches will be used to address the above research questions. A descriptive, repeated-measures design will be used to address the relationships among variables described in research question one. All predictor and outcomes variables will be measured within two months of the initial dementia diagnosis (T1) (providing a baseline measure) and at 6 (T2), 12 (T3), and 18 months (T4) following the dementia diagnosis, outlined in Table 1. Qualitative methods, incorporating a semi-structured interview, will be used to answer research questions two and three. This triangulation of methods provides a comprehensive approach to address the complexity of the research questions, considering the inherent difficulties in objectively assessing the variables of interest in persons with impaired cognition.4 Although initial testing of the PWB-CIP's psychometric properties will be carried out prior to the study, further evaluation of the PWBCIP's effectiveness in measuring psychological well-being with homedwelling patients will be carried out during the study.

Prior to initiating the study procedures, the developmental phase of the study will be completed, including development and initial testing of the semi-structured interview guides for use with both AD patients and their primary caregiver. The interview guides will be developed under the guidance of Dr. Kathleen Knafl, a consultant for this study. Additionally, the P.I. will complete training in interviewing skills specific for AD patients under the direction of both project sponsors, Drs. Kathleen Buckwalter and Mary Austrom.

Table 1

Entry and Assessment of Subjects

                  Time Interval*           Reason for Visit       Time for Visit         

Initial Visit     One Month after          Obtain Written         60 Minutes             
                  Diagnosis                Consent Review                                
                                           Caregiver-Rated                               
                                           Instruments                                   

T1  Assessment    One Month                 Patient/Caregiver      90-120 Minutes        
                                           Assessment                                    
                                           Semi-Structured                               
                                           Interviews                                    

T2  and T3        Six Months               Patient/Caregiver      90-120 Minutes         
Assessment                                 Assessment                                    
                                           Semi-Structured                               
                                           Interviews                                    

T4 Final Visit      Six Months             Final                  90-120 Minutes         
                                           Patient/Caregiver                             
                                           Assessment                                    
                                           Semi-Structured                               
                                           Interviews                                    


*Time interval is time since previous visit

SUBJECTS AND SETTING: Home-dwelling AD patients and their primary caregiver will be purposively sampled through the Indiana Alzheimer's Disease Center located at the Indiana University Medical Center. Patients newly diagnosed with either Alzheimer's, multi-infarct, or mixed dementia will be recruited primarily from this out-patient setting. The AD Center currently evaluates approximately 30 new patients monthly, providing a sampling frame of 360 AD patients per year. Patients come to the center from many parts of the state, including both rural and urban settings, providing a diverse patient population. If an adequate sample cannot be obtained from the AD Center, several diagnostic centers located in central Illinois (Carle Clinic in Urbana, Illinois, and Southern Illinois University School of Medicine Alzheimer's Treatment Center in Springfield, Illinois) will provide additional sources of newly diagnosed AD patients. These combined resources, if required, provide a sampling frame of newly diagnosed patients representing both urban and rural populations. The P.I.'s previous intervention study in the Central Indiana area included approximately 30% of subjects residing in rural settings, the expected rate for this study. The minority population rate is 13.8% in the greater Indianapolis area, indicating a probable minority inclusion rate of 13% to 14%.47 The P.I.'s previous study included a 12.5% minority inclusion rate, representative of the general population. AD patients are primarily female, around 65%, the expected inclusion rate for female subjects in the proposed study and the same rate of female subjects evidenced in the P.I.'s previous studies.12

The attrition rate for the P.I.'s most recent intervention study was 12%, with the length of time in the study being 6 months. As most of the attrition in the intervention study was due to institutionalization or death and the patient subjects were at various stages in the disease process, it is expected the attrition rate will not be as high for the proposed study, considering all patient subjects will be newly diagnosed with the disease, lessening the likelihood of institutionalization or death. However, with an expected 10% attrition rate, an initial sample of 85 patient subjects will be required to allow for 75 patient subjects completing the study. A power analysis, described in the Data Analysis section for multiple regression analysis, indicates that 75 patient subjects will be required for a power between .68 to .99. A letter of support from Dr. Bernardino Ghetti, Director, Indiana AD Center, is included as Appendix A.

Patient subjects must meet the following criteria: 1) The patient must have a confirmed, recent (within 2 months) diagnosis of Alzheimer's, multi-infarct, or mixed dementia, using standard diagnostic criteria that include: physical examination, CAT or MRI scan, thyroid dysfunction tests, urinalysis, neuropsychological testing, and blood chemistry. Since the AD Treatment Center uses this accepted diagnostic criteria, it is expected all patient subjects will have undergone standard diagnostic evaluation. Because Alzheimer's and multiinfarct dementia progress with a similar pattern of neuropsychiatric pathology, unlike other types of irreversible dementias, including patient subjects in both diagnostic groups provides a patient sample with similar patterns of behaviors and physical changes throughout the dementing process. 2) Informed, written consent of the family member most responsible for the patient's care, legal guardian, or involved individual designated by the patient is required for participation. 3) The patient will provide verbal assent to participate. 4) The patient's primary caregiver must be a family member, rather than a paid assistant. No age limitations are applied to the patient sample to allow inclusion of both early- and late-onset AD patients.

Caregiver subjects will be sampled to include the primary caregiver for patient subjects. George and Gwyther's48 definition of primary caregivers will be used: persons providing the majority of care to the home dwelling patient subject. This definition will be used more narrowly in this study, as only family members providing the majority of care will be included. Caregiver subjects will be elicited concurrent with consent for patient subjects, equalling the same number (85 total) as patient subjects. Since caregivers will contribute to the study through completing some of the study instruments, written consent will also be obtained for the caregiver's participation.

MEASURES/OPERATIONAL DEFINITIONS:

Predictor/Mediating Variables:

Patient-Centered Variables: Variables characterizing patient subjects will be used to describe the patient sample and to analyze the relationships among these potential predictor variables and patient quality-of-life outcomes. A demographic data sheet will be completed with the primary caregiver and patient on the initial (T1) assessment and includes background information for both caregiver and patient subjects. Information to be collected includes: age, gender, education, occupation, medical diagnoses for both caregivers and patients, and medications the patient is currently taking. Information will also be collected at subsequent data collection points (T2 - T5) about any situational changes that may influence outcomes, such as changes in either the caregiver's or patient's health status, stressful family events, or conflicts, such as having to terminate the patient's driving privileges.

Meaning of Event: The interview schedule developed and tested in the study's developmental phase will provide information about the patient's perceptions of the impact of the illness on his/her life, with an emphasis on the patient's needs. Following procedures consistent with qualitative data analysis, themes or patterns of responses will be identified, indicating how patients commonly perceive and experience AD. Questions adapted from the Management of Meaning scale described by Pearlin and colleagues21, measuring the ways in which caregivers manage the meaning of the dementia, will be used to assess meaning in both patient and caregiver subjects. Initial questions adapted from the Management of Meaning scale are included in Appendix B.

Personality Traits: Family members will be asked to complete the Adult Personality Rating Scale (APRS), developed and used by Lawton and colleagues in several studies with AD patients and caregivers to assess the caregiver's past and current perceptions of the patient's personality.34,49 This 50-item inventory represents 4 factors, accounting for up to 78% of the total variance in past studies. Each item in rated on a 5-point scale, from "very low" to "very high", reflecting the family member's rating of the patient on each trait. The factors include extraversion, hostility, task assertiveness, and neuroticism, which corresponds with four of the major five factors inherent in most personality measures. Summated scores are created for each of the factor-defined scales. The validity of the APRS has been supported through significant correlations with a health professional's independent ratings of aggression and excess disability criterion. Kleban, et al.,49 found the APRS useful for measuring change in personality in older adults, with defined personality traits accounting for as much as 60% of the predicted improvement in excess disabilities in institutionalized patients. In a recent study, Lawton and colleagues34 found missing data were close to nonexistent for family ratings of the AD patient's personality. The APRS has the advantage over other instruments in that it was developed for and tested with family members as a rating of change in personality in older adults. Family members in this study will be asked on the T1 assessment visit to rate both their perspective of the patient's previous personality (one year prior to the onset of AD) and the patient's current personality. On subsequent assessment visits, caregivers will be asked only to rate the patient's current personality. The APRS requires approximately 25 minutes to complete and is found in Appendix C.

Maintenance of Past Activities: The Pleasant Events Schedule-AD (PES-AD), Appendix D, will be used to operationalize past and current activities for the AD patient. Although the PES-AD was developed to assist caregivers in identifying activities pleasant for AD patients at various points across the disease process, the 53-item instrument contains activities common to older adults.50 The PES-AD is based on two earlier Pleasant Events Schedules, one designed for general adult and one designed for older adult populations.51, 52 The original scales had excellent psychometric properties, with an alpha coefficient of .98 and a correlation with one another of r =.96. The PES-AD scale contains three ratings: 1) how many times an activity occurred during the past month, using a 3-point scale from "not at all" to "7 or more times"; 2) the availability of the activity during the past month using the same 3-point scale; and 3) how enjoyable the activity is rated on a 2-point scale of either "now enjoys" or "enjoyed in the past". A total score can be obtained for each rating, providing a measure of how the patient's involvement in, access to, and experience of enjoyable activities changes over time. The scale can also assist in identifying enjoyable activities that are lost to the AD patient early in the disease, providing direction for development of appropriate interventions to facilitate enjoyable activity participation. Teri and Logsdon50 found the scale could be completed by caregivers in approximately 30 minutes without assistance.

External Variables:

Social Support: Actual social support, rather than perceived social support, will be measured as an indicator of support received by the patient. As found in Appendix E, information regarding social contacts and formal support services will be recorded on each assessment, reflecting support received during the previous week. Although the influence of "actual" rather than "perceived" social support on outcomes is inconsistent in the research literature, actual social support has been hypothesized to predict positive outcomes in cognitively intact populations, giving initial support to the utility of actual social support as an adequate indicator of social support with AD patients.22,24 Indicators of actual social support include: number of contacts with family and friends providing six categories of social support (intimate interaction and confiding, positive feedback, advice, material aid, physical assistance, and social participation), group activities attended, telephone calls received from family/friends, and letters or cards received. Formal support services includes: adult day care, outpatient health care services, number of contacts with paid assistants providing direct care, visits from professionals in nursing care and social service agencies, attendance at formal support groups, telephone contacts from health care providers, individual counseling, and contacts with paid escorts.53 Spiritual support will be measured through visits and telephone contacts from representatives from religious organizations (lay or clergy); attendance at religious events, including formal religious services; and facilitated private religious activities in the home, such as prayer or Bible reading. A total "score" or value will be obtained for each item. Additionally, scores for each category - informal social contacts, formal support services, and spiritual support - will be calculated by totalling item scores within each category. The caregiver will be provided the social support recording form prior to the assessment interview and will be asked to record information on the form during the week prior to the visit. Information recorded on the form will be reviewed by the researcher with the caregiver to assure the accuracy of the information.

Quality of Patient/Caregiver Relationship: Most measures of "quality of relationship" examine the marital relationship, rather than more globally measuring the patient's relationship with all close family members. One exception is the Elder-Caregiver-Family Relationships (ECFR) subscale of the caregiving burden scale developed and tested by Poulshock and Deimling.54 This 11-item scale reflects the negative changes in elder-caregiver-family relationships, with items rated on a 3-point scale from 1 = "not at all" to 3 = "a great deal". Examples of items include, "Caregiver relationship with the elder is strained" and "Caregiver feels resentful toward the elder". Descriptive data on 614 families indicated that 52% of caregivers reported two or more areas of impact on patient-caregiver relationships. The ECFR's validity was supported through significant relationships between ECFR scores and measures of caregiver burden, r = .46, and caregiver depression (r = .40). The ECFR will be completed by the patient's primary caregiver, taking approximately 10 minutes to complete. The ECFR is included as Appendix F.

Facilitation of Patient Self-Care: Extent to which the primary caregiver facilitates the AD patient's self-care will be assessed in two ways. The OARS IADL-Self Care Skills (SCS) (Appendix G) measure used to operationalize the patient's functional ability contains 15 items reflecting the patient's capacity for both instrumental and functional activities of daily living. Each item is rated on a 3-point scale from 0 = "unable to perform" to 2 = "able to perform without assistance". The psychometric properties of the OARS IADL-SCS scale are described in a later section. While the OARS IADL-SCS scale is completed by the researcher, a companion scale will be completed by the primary caregiver, rating each item on the extent to which the patient actually performs the task at the indicated level. The caregiver's ratings will also be made on a three-point scale from 0 = "never performs" to 2 = "performs all the time". A "difference" score can be calculated for each of the two subscales inherent in the OARS IADL-SCS instrument, giving some indication of the extent to which actual ability relates to performance of activities. Through a comparison of ability and actual performance ratings, discrepancies in self-care ability and performance can be detected. As discrepancies are detected, the semi-structured interview will be used to further clarify reasons for the observed discrepancies. Although the interview will be designed to elicit information regarding other aspects of the caregiving experience, information regarding the obstacles to facilitating patient self-care will provide a basis for targeted interventions to optimize self care in the home setting.

Patient Quality-of-Life Outcomes:

Psychological Well-Being: Two instruments will be used to operationalize psychological well being in AD patients. The Cornell Scale for Depression in Dementia (CSDD) measures a major aspect of the patient's well-being, affect, specifically, depressive state.55 The CSDD is a 19-item instrument administered by a trained researcher or clinician. Items are rated on 4-choice response format including a = "unable to evaluate", 0 = "absent", 1 = "mild or intermittent", and 2 = "severe". The 19 items reflect five aspects of the patient's affective state: mood-related signs, behavioral disturbances, physical signs, cyclic functions, and ideational disturbances. Each item is rated by the caregiver and by the clinician through a interview with the patient. If significant discrepancies appear in the ratings, the clinician interviews the caregiver to clarify the reason for the disagreement. Total administration time is approximately 30 minutes. The CSDD's reliability has been well-supported in patients at various levels in the disease process, with interrater reliability correlations ranging from r = .64 to r = .99. Internal consistency was adequate with a coefficient alpha of .84 and a mean interitem correlation of .24 in one study.55 The CSDD's validity was supported through its ability to distinguish demented subjects with no depression, minor depression, probable major depression, and definite major depression in both hospitalized patients and nursing home residents. The CSDD is included as Appendix H.

The Psychological WellBeing in CognitivelyImpaired Persons (PWBCIP) measure (Appendix I) will be included as a second measure of the patient's psychological well being if initial testing supports adequate reliability and validity. The PWB-CIP a 16item, observerrater instrument developed by the P.I., based on the extensive work of Andrews and Withey25 on measurement of well being and the work of Lawton and colleagues3,56 defining psychological well being as consisting of positive and negative affect. The PWBCIP represents positive and negative affective states and congruence between desired and attained goals, represented by congruence with the environment, described by Lawton56 as being central to psychological well being. Additional items specific to the AD patient's psychological well being include the patient's ability to maintain interest in the environment. Eight PWBCIP items represent a positive state of psychological well being and 8 items reflect a negative state of psychological well being, with one additional item providing a global assessment of the patient's overall psychological well being. The observer is asked to rate each item on a scale from 1 (never) to 5 (always), indicating how often the response or behavior has been observed during the last 24 hours. A total PWBCIP score is obtained by subtracting item ratings indicating a negative state of psychological wellbeing from item ratings indicating a positive state of psychological well being to yield an overall score, similar to the procedure used by Bradburn57. Following psychometric evaluation, if subscales emerge in the instrument, a total subscale score can be calculated for each subscale, using the same method as described for the total instrument score. Items 16 indicate positive and negative affective states, with descriptors given for further clarification of responses that might otherwise be ambiguous. Items 710 measure the patient's congruence with and freedom in the environment, representing Lawton's56 congruence between desired and obtained goals. Items 1116 measure the patient's response to activities and social encounters. Item 17 is the global assessment of the patient's affective state, using 7 faces ranging from a marked smile (7) to a marked frown (1). The observer indicates which "face" most represents the patient's psychological wellbeing during the previous 24 hours. The overall PWBCIP score can be compared to the global assessment score as further validation of the observer's ratings of the patient's behaviors. Initial testing of the PWB-CIP's content validity is being carried out with national nursing experts in care of AD patients. Content validity has been supported for the 16 items, with CVI scores ranging from .80 to .95.

Behavioral Competence:

Physical Health: The Cumulative Illness Rating Scale (CIRS) was designed as a brief, yet comprehensive, instrument for assessing physical impairment.34,58 The CIRS is completed by a health professional competent in physical examination techniques and assesses 13 areas of physical impairment, grouped under major body systems. The 13 areas are assessed using a 5-point rating scale, with 0 = "none" to 4 = "extremely severe". Total pathology is represented by adding the sums for the 13 items, producing a total scale score. Interrater reliability estimates for 110 ratings for 20 patients was reported as high, ranging from .83 to .91. Validity has been supported through the CIRS's ability to predict survival in 472 persons. The CIRS will be completed by the P.I., a Gerontological Nurse Practitioner skilled in physical assessment techniques, with the patient at each assessment point. The CIRS measure is included as Appendix J.

Functional Ability: The Instrumental Activities of Daily Living (IADL) and Self Care Skills (SCS) from the larger OARS instrument will be used as the measure of the patient's functional ability.59 The total instrument includes 15 items, 8 indicating IADL's and 7 SCS items measuring ability to perform self-care. All items are rated by an observer (for this study, the researcher or assistant) on a three-point scale: 0=unable to perform, 1=some help is required, 2=able to perform without help. A mean score is obtained for each of the separate sets of activities, instrumental activities of daily living and self-care skills. Items indicating IADLs include activities related to use of telephone, preparing meals, taking medications, and handling finances. Self-care skills include activities of eating, dressing, bathing and showering, performing personal hygiene, walking, toileting, and continence. Test-retest reliability for the IADL and SCS portions of the OARS instrument is reported as r =.82. Inter-rater reliability ranges from .62 to .70.65 Discriminant validity for the IADL and SCS scales is supported through differences in scores for community, clinic, and institutional populations in the expected direction, with institutional populations representing the least functional subjects. Information will be elicited from the primary caregiver and patient regarding the patient's functional ability. The IADL and SCS scales were found to be sensitive to change over a six-month time period in the P.I.'s intervention study.

Cognitive Function: The MiniMental State Exam (MMSE) was developed and tested by Folstein and colleagues60 as a short, easytoadminister measure of mental status and screening tool for dementia (see Appendix K). It consists of asking subjects 11 questions or commands such as "What is the year?" or "Write a sentence". The instrument yields a total score of 30, with scores below 23 suggesting dementia. The MMSE takes only 5 to 10 minutes to administer and assesses 7 categories of cognitive function including orientation to time and place, registration of three words, attention and calculation, recall of three words, language, and visual construction. The MMSE does not require extensive training to administer. The MMSE has been used extensively in research involving dementia patients and found to be an acceptable measure of mental status. The MMSE's reliability has been reported as consistently high, with testretest coefficients ranging from .83 to .98. Criterionrelated validity is evident in correlations of .66 to .78 with standard measures of cognition.60

Social Behavior: The Functional Behavior Profile (FBP) is a newly developed indicator of positive behaviors in dementia patients.61 The FBP includes 27 items describing patient behaviors rated on a 5point scale from 0 (never) to 4 (always) indicating the frequency of the observed behavior over the previous week. The 27 items represent three subscales, task performance, problem solving, and social interaction, with Cronbach's alphas for the subscales ranging from .94 to .96. Total FBP scores were significantly correlated with the ADL Index, r = .85, the Memory and Behavior Checklist, r = .83, and the Blessed Dementia Scale, r = .83, as indicators of the FBP's construct validity.61 FBP scores were also found to decrease across the various stages of dementia, indicating the ability of the FBP to indicate level of performance in patients at different disease stages. The FBP requires approximately 15 minutes to complete and can be administered through an interview format or may be rated independently by a caregiver.61 The FBP scale is included as Appendix L.

The study variables and operational definitions are provided in Table 2.

Table 2                                                                                
Study Variables with Operational                                                       
Indicators           
Variable                                           Measure                   

Predictor/Mediating Variables:                                                         

Patient-Centered Variables:                                                            

   Patient Characteristics                 Demographic Data Sheet                      

   Personality Traits                      Adult Personality Rating Scale              

   Maintenance of Past Activities          Pleasant Events Schedule-AD                 

   Meaning of Event                        Semi-Structured Interview                   

External Variables:                                                                    

   Social Support (including spiritual     Actual Social Support Measure               
support)                                                                               

   Quality of Patient/Caregiver            Elder-Caregiver-Family Relationships Scale  
Relationship                                                                           

   Facilitation of Patient Self-Care       Caregiver Ratings of OARS IADL/SCS          
                                           Behaviors                                   

                                              Semi-Structured Interview                

Quality-of-Life Outcome Indicators:                                                 

   Psychological Well-Being                Cornell Scale for Depression in Dementia    

                                              Psychological Well-Being in Cognitively  

                                              Impaired Persons                         

Behavioral Competence:                                                                 

   Physical Health                         Cumulative Illness Rating Scale             

   Functional Ability                      OARS IADL and Self-Care Skills              

   Cognitive Function                      Mini-Mental State Exam                      

   Social Behavior                         Functional Behavior Profile                 

PROCEDURES:

Procedures to Address Developmental Aims: Prior to implementing the research procedures, the P.I. will consult with Dr. Kathleen Knafl at the University of Illinois to develop the interview guide, review procedures for qualitative analysis, and refine interviewing skills and procedures. The P.I. will also take coursework in qualitative methods, S659 Qualitative Methods in Sociology, offered through the Department of Sociology at Indiana University and review suggested readings in development of interview guides in qualitative methods.

Interviewing skills will be refined through supervised experiences with both Dr. Kathleen Buckwalter at the University of Iowa and with Dr. Austrom. Dr. Buckwalter's expertise in care of both institutionalized and home-dwelling AD patients will be especially valuable in assuring that a competent, sensitive approach is used throughout the interview procedures. During an initial one-month supervised internship with Dr. Buckwalter, sample interviews will be audiotaped and reviewed by Drs. Buckwalter, Knafl, and Austrom as they are obtained. Feedback from Dr. Knafl and the study sponsors will be obtained to improve interview strategies, facilitate patient responses, and develop additional questions or approaches to increase meaningful patient responses. Although Dr. Buckwalter is located at the University of Iowa, communication following the internship will be facilitated through telephone and electronic mail conferencing and visits by the P.I. to the University of Iowa. Additionally, Dr. Buckwalter's current research study is simultaneously being conducted at the University of Iowa and Indiana University, facilitating Dr. Buckwalter's presence in Indianapolis. Telephone conferencing with Dr. Knafl will facilitate communication during the entire interview development process.

To gain competency in psychological assessment, the P.I. will initially observe this aspect of the initial patient assessment in the Indiana AD Center. The P.I. will then perform the assessment under the supervision of Dr. Austrom and the clinic staff. Additionally the course, P540 "Principles of Psychological Assessment and Prediction", will be taken at Indiana University's Bloomington campus. The combined supervised clinical experiences and coursework will assure the P.I. possesses adequate skills for the assessment of the patient's affective states. Additionally, the P.I. will meet with Dr. Victoria Champion, a project consultant with experience in instrument development, on a monthly basis for assure that proper evaluation of the PWB-CIP's psychometric properties is carried out. Also, the P.I. will take a course offered by Dr. Champion on instrument development early in project. Ongoing activities include participation in the AD Center's grand rounds, weekly staff meetings, and attendance at recognized national conferences, such as the Gerontological Society of America. These continuing and combined activities will allow the P.I. to consult with other professionals regarding the study's activities and outcomes. A timetable for completing activities inherent in meeting the study's developmental aims is included as Table 3.

Table 3                                                                                 
Activities to meet Developmental Objectives                                             


Objective                 Activity             Time Commitment         Methods      

1.  Develop              Directed study with Dr.   One month on site      on-going         
semi-structured          Kathleen Knafl                                   consultation     
interview guide/enhance                                                   for interview    
expertise in                                                              telephone        
qualitative                                                               contacts         

                         -                         One semester           Coursework:      
                                                                          S659             
                                                                          Qualitative      
                                                                          Methods          
                                                                          selected         
                                                                          readings in      
                                                                          qualitative      
                                                                          methods in       
                                                                          methods in       
                                                                          Sociology        

                                                   Monthly                guide            
                                                                          refinement/quali 
                                                                          tative data      
                                                                          analysis         

2.  Develop appropriate  Consultation with Dr.     One month on site      -observation of  
interviewing skills      Kathleen Buckwalter                              interviewing     
                                                                          techniques       

                                                    Monthly contacts      -supervised      
                                                                          interviews with  
                                                                          AD patients      

                         Consultation with Dr.     One day/week in        -supervised      
                         Mary Austrom                                     interviews with  
                                                                          AD Center        
                                                                          patients         

3.  Increase skill in    Supervised clinical        One day/week in AD                     
psychological            experiences with Dr.      Center                                  
assessments              Mary Austrom                                                      

                         -Attendance at team       Weekly during entire                    
                         meetings research grand   project                                 
                         rounds seminars                                                   

                         -Coursework:  P540        One semester                            
                         Principles of                                                     
                         Psychological Assessment                                          
                         and Prediction                                                    

4.  Increase knowledge    Consultation with Dr.    Monthly contacts                        
of instrument            Victoria Champion                                                 
development                                                                                

                         -Coursework:  D752        One semester                            
                         Instrument Development                                            


Procedures Addressing Research Questions:

Consent Procedures: The P.I. will contact the clinical team at the AD Center to initiate procedures for referral of patient subjects. Potential patient and caregiver subjects will be identified by the clinic staff during the visit on which the diagnosis of dementia is given to the patient and family. As possible patient and caregiver subjects are identified, the P.I. will contact the caregiver by telephone to explain the purpose and conduct of the study. If the caregiver indicates a willingness to participate, a time for an initial visit in the home will be arranged and a written consent form will be forwarded to the caregiver for completion. During the initial home visit, within one month of the initial diagnosis, the written consent will be finalized. This initial consent visit also allows the caregiver and patient to become familiar with the P.I. and research assistant, hopefully increasing comfort during the subsequent assessment visits.

Additionally, verbal assent for the patient's participation will be elicited. Although AD patients may not be legally competent to give consent, they may give assent or indicate a desire not to participate. To insure patient subjects have the opportunity to refuse participation, the study procedures will be explained to the patient on the first home visit using a standard statement of the purpose and conduct of the study. It is acknowledged that the patient may give verbal assent to participate without clearly understanding the nature of the agreement. Therefore, to further protect the patient's rights, patients may withdraw from the study at any time. In addition, care will be taken to inform caregivers subjects of the option of withdrawing from the study at any time.

Rater Training Procedures: Prior to administration of study instruments, participating research assistants will receive training in administering the MiniMental State Exam and other study instruments. The MMSE's administration procedures will be reviewed with the assistants, and the assistants will have the opportunity to observe the principal investigator, who has used the MMSE extensively, administer the MMSE to at least three patient subjects. The principal investigator will then observe each research assistant administering the MMSE to no fewer than three patient subjects to assure consistency in ratings. Proper administration procedures for the other study instruments will be reviewed with the assistants, assuring consistency in carrying out study procedures.

Data Collection

Visit 1: T1 Assessment: Within two months of the initial dementia diagnosis, one month following the initial visit to obtain written consent, baseline assessments will be made in the subject's home. Baseline assessments will include: caregiver and patient demographic data, patient predictor and outcome variables, and external variables. A research assistant will accompany the P.I. on all assessment visits to facilitate data collection. The research assistant will complete with the caregiver the demographic data sheets, Adult Personality Rating Schedule, Elder-Caregiver-Family Relationship Scale, the caregiver's component of the OARS IADL & SCS scale, and Pleasant Events Schedule-AD. The Actual Social Support, Functional Behavior Profile, and Psychological Well-Being in Cognitively-Impaired Persons scales will be explained to the caregivers during the initial consent visit and completed prior to the first assessment visit. While the research assistant is completing the caregiver-rated instruments with the caregiver, the P.I. will conduct the Cumulative Illness Rating Scale, OARS IADL & SCS scale, Mini-Mental State Exam, Cornell Scale for Depression in Dementia, and semi-structured interview with the patient. Each interview will be tape recorded, using a small, pocket-sized recorder. The P.I. will record any non-verbal patient responses that may lend meaning to the verbal data. The P.I. will review with the caregiver any discrepancies in the Cornell Scale for Depression in Dementia ratings. A semi-structured interview will also be conducted with caregiver subjects. The entire assessment will take approximately 90 to 120 minutes to complete. Following each interview, the tape-recorded patient statements will be transcribed using the matrix approach described in the Data Analysis section.

T2 to T4 Assessments: The T2 to T4 visits will include the same data collection procedures and instruments outlined for the T1 assessment visit, with the exception of omitting the demographic data collection, although information regarding changes in status, such as hospitalizations, loss of employment, etc., will be sought. Information will be collected on events that may have affected the patient's adaptation during the prevailing time period, such as major illnesses or stressors. The research assistant will again be primarily responsible for collecting data with the caregiver, while the P.I. conducts the patient assessments and semi-structured interviews. Interpretations of patient responses from the previous interview will be verified with the patient to assure consistent and accurate interpretation of the qualitative responses. Care will be taken to conduct all assessments at the caregiver's and patient's convenience. The six-month time interval between the T2, T3, and T4 assessments allows for identification of changes in adaptation in patient subjects without taxing subjects. Some patients have been observed continuing to deny the illness as long as 12 months following diagnosis, and, therefore, following patients for as long as 18 months following diagnosis may identify adaptational changes not evident at 12 months. All subjects will be entered into the study 16 months following the study's initiation, allowing adequate time to complete all follow-up assessments and analyze findings. Approximately 8 patient subjects will be entered each month to meet the required total sample, a reasonable expectation in view of the number of potential subjects in the sampling frame.

DATA ANALYSIS

Using procedures described by Cohen,62 statistical power was estimated using means and standard deviations of the outcome variables obtained from sample values in previously published studies that used the instruments in similar populations. The analysis allows for a maximum of five variables to be included in any one statistical analysis. In regression and correlational analyses, with a moderate to large effect size, (.18 to .456), alpha of .05, and five variables, a sample size of 75 should yield of power of .68 to .99.

Prior to statistical analysis, psychometric evaluation of the study instruments will be conducted. The relationships among outcome measures will be examined to detect possible overlap in measures and inter-correlations among means. Reliability checks will occur at data collection intervals throughout the study. Descriptive statistics will periodically be calculated to determine if adequate variability in ratings is occurring or if altering of study procedures is indicated. Careful cleaning of the data will be carried out throughout the study. Data cleaning procedures within the SPSS-X and SAS statistical packages will be used. In addition, data will be reviewed by the P.I. and research assistant to assure that accurate entry has occurred prior to data analysis. The P.I. and the biostatistician will take primary responsibility for data management, although the research assistant will assist with data coding, cleaning, and entry.

Demographic data will be analyzed using descriptive statistics. Relationships among demographic, predictor/mediating, and outcome variables will be assessed using correlational procedures. Statistical procedures will be described for each research question or hypothesis. A significance level of p =.05 will be used for all analyses.

Research Question No.1: What are the relationships among person-centered variables, external variables, and quality-of-life outcomes in AD patients over the initial 18 months following the AD diagnosis? Using correlational analyses, the relationship between the predictor and mediating variables for each data collection point will first be analyzed for significant relationships between variables. The relationships among the predictor, mediating, and outcome variables for each data collection point will then be determined using correlational analyses. Using hierarchial multiple regression analysis, mean scores for variables representing the person-centered and external variables will be evaluated as predictor variables, with the mean scores for patient quality-of-life indicators entered as dependent variables. The regression analyses will be carried out for each data collection point to determine predictors of quality-of-life outcomes changes across the first 18 months following the AD diagnosis. Demographic variables significantly related to outcome variables will be entered in step one of the multiple regression analysis to control for their effects. Situational factors, recorded in anecdotal form throughout the study, will be analyzed to determine additional factors associated with quality-of-life outcomes.

Research Questions No.2 and 3: What are common patient responses to the dementia diagnosis in relation to meaning of the illness, psychological needs, perception of control, behaviors of others, and perceived loss or benefit? How do patient responses to the disease process change over the first 18 months following the AD diagnosis? Using responses to the semi-structured interview questions, tape-recorded patient and caregiver responses to specific interview questions will be transcribed using the matrix technique described by Miles and Huberman63 and Breitmayer and colleagues.64 The matrix approach to qualitative data structures and analysis allows for systematic evaluation of the relationship between subject responses to interview questions and quantitative measures of interest. For this study, patient responses will be compared to scores on the Cornell Scale for Depression in Dementia and Functional Behavior Profile, while caregiver responses will be compared to scores on the Quality of Patient/Caregiver Relationship scale, providing for a triangulation of the data. Categories of responses, such as expression of loss, frustrations, concerns, behaviors of others, feelings of control/loss of control, change in role, or change in self image, will be used to cluster the responses. The coding scheme will be revised as data are analyzed. Segments of the data (one of every 15 interviews) will be double-coded by the P.I. and research assistant to assure clarity and reliability of coding. The accepted level of reliability will be greater than 70% agreement.63 The memoing technique will also be used to record the P.I.'s impressions and thoughts, not reflected in the audiotaped interviews. The matrix approach allows for analysis of responses across individuals as well as within individuals. This analysis will be carried out at each assessment point (T1 to T4) to observe for changes in the relationships over time. Additionally, the data will be reviewed to detect rival explanations and negative evidence to confirm or rule out initial conclusions. If the meanings of some responses are difficult to interpret, the P.I. will confirm the interpreted meaning with the patient or caregiver on the following assessment visit. Dr. Knafl will be consulted throughout the data entry and analysis to assure accurate interpretations of the qualitative data.

Research Question #4: What is the reliability and validity of the PWB-CIP instrument when used with AD patients in the early and middle disease stages? Ratings from family caregivers will be using to further assess the PWBCIP's psychometric properties. Descriptive statistics will first be calculated to assure adequate variability and a normal distribution in the obtained ratings. Following review of the descriptive data, the PWBCIP's construct validity will be assessed through significant correlations in the anticipated direction with the Functional Behavior Profile and Cornell Scale for Depression in Dementia. Also, factor analysis of the PWBCIP's items will be conducted using a principalcomponents analysis to first determine the common variance among the variables. A varimax rotation solution will then be obtained to maximize the factor loadings and interpret the factor structures. If multiple factors are inherent in the PWBCIP, internal consistency reliability will be determined using Cronbach's alpha for each subscale and the total instrument scores, with an alpha level >.80 being considered adequate. Reliability will be further assessed for rater consistency using inter-class correlations between caregivers' ratings taken simultaneously, with an r >.75 being considered acceptable.

LIMITATIONS:

Unexpected difficulties, such as postponement of a visit due to unanticipated conflicts for caregiver or patient subjects, will occur and cannot be totally controlled for in the design or study procedures. If a visit postponement is required, the visit will be rescheduled as close to the original visit time as possible, with the actual visit time being recorded. The problem of eliciting personal responses from dementia patients was addressed previously, but remains a possible limitation of the study. By using a variety of approaches to data collection, including caregiver ratings and patient perspectives, as well as quantitative measures and qualitative approaches, the reliability and validity of the study outcomes are optimized.

5. HUMAN SUBJECTS

1. Home-dwelling AD patients and their primary caregivers will be included as subjects, totalling 85 subject pairs. The expected age range for patients subjects is 60 to 85, with a probable mean of 72, based on previous in-home studies. AD patients are primarily female, around 65%, the expected inclusion rate for female patient subjects in the proposed study. The minority population rate in the greater Indianapolis area is 13.8%, the expected rate of inclusion for both minority patient and caregiver subjects, consistent with the P.I.'s past studies. No patient subject will be excluded from the study due to ethnic origin or gender. Most patient subjects will have some health impairment, with an expected mean of 4.5 medical diagnoses. AD patients are considered to be a vulnerable population. As the proposed study will provide essential information for the development interventions to increase positive patient outcomes in dementia patients, the inclusion of dementia patients is necessary. Patient subjects will be included as they meet the following criteria:

Inclusion Criteria:

A. The patient must have a confirmed, recent (within 2 months) diagnosis of Alzheimer's,

multi-infarct, or mixed dementia.

B. Informed, written consent from the family member most responsible for the patient's care

legal guardian, or involved individual designated by the patient is required for participation.

C. The patient must provide verbal assent to participate.

D. The patient's primary caregiver must be a family member, rather than a paid assistant.

Caregiver subjects will be primary caregivers for patient subjects. The majority of primary caregivers are female, 60%, the expected rate for female caregivers in this study. Caregivers are expected to have varying levels of health, with expected high rates of stress and depression. No caregiver subject will be excluded due to ethnic origin or gender.

2. Data will be obtained through questionnaires, open-ended interviews, observation of patient behaviors, and physical assessment. Caregiver subjects will be asked to provide written responses to several questionnaires, including demographic information, a measure of the patient/caregiver relationship, a rating of the patient's personality, received social support, a patient psychological well being measure, and patient depression and functional behavior profile. All data will be used specifically for research purposes.

3. Patient and caregiver subjects will be recruited through an outpatient treatment center, the Indiana Alzheimer's Disease Center, located at the Medical Center at Indiana University. The researcher will contact potential subjects identified by the Center's staff through a telephone contact. The purpose and conduct of the study will be explained over the phone. Written consent forms will be sent to the patient's primary caregiver, with caregivers providing written consent for participation for both themselves and the patient. On the first home visit, after written consent is obtained, the study instruments will be explained to the caregiver and arrangements for the second visit for the baseline assessment will be made. Verbal assent from patient subjects will also be obtained on the first home visit, after a careful explanation of the study procedures and purpose. If the patient indicates h/she is distressed by any research procedure, the procedure will be terminated and scheduled at a different time. All subjects will be informed of the confidential nature of the data and of their right to withdraw from the study at any time without negative consequences.

4. The major risk to subjects is invasion of privacy due to being asked personal questions about the patient/caregiver relationship, activities, and personal characteristics of the patient. The time required of caregivers to complete the study instruments and procedures may impose another requirement in an already full schedule. Allowing family caregivers ample time to complete instruments by mailing them to the home prior to the visit and scheduling visits at the caregiver's and patient's convenience should lessen potential stress. Caregivers in the P.I.'s previous in-home study often expressed satisfaction and some relief due to the researcher's visits, noting it was encouraging that someone was interested in their welfare.

5. In addition to the precautions described above, only the P.I. and research assistants will know the subjects' identity. All instruments will be coded, so no name will appear on any form, except for the initial consent forms. All consent forms and study instruments will be locked.

6. The risks described may be minimal in view of the potential benefits to both patients and caregivers from knowledge gained regarding factors predicting to positive patient outcomes. Caregiver and patients may benefit from decreased patient stress, increased knowledge of positive interventions, improved affective states in patients, improved self care, and increased preservation of the patient's "personhood".

6. VERTEBRATE ANIMALS

None.

7. CONSULTANTS/COLLABORATORS

Sponsor: Kathleen Buckwalter, Ph.D., R.N., F.A.A.N.

Professor and Associate Director

Office for Nursing Research

Development and Utilization

The University of Iowa

Iowa City, Iowa 52242

Co-Sponsor: Mary Guerriero Austrom, Ph.D.

Assistant Professor

Director, Education Core

Indiana Alzheimer's Disease Center

Indiana University

Indianapolis, Indiana 46202

Consultant: Kathleen Knafl, Ph.D.

Associate Dean for Research

University of Illinois

College of Nursing

Chicago, Illinois

Victoria Champion, D.N.S., R.N., F.A.A.N.

Professor and Associate Dean for Research

Indiana University

School of Nursing

1111 Middle Drive

Indianapolis, Indiana 46202

8. CONSORTIUM/CONTRACTUAL ARRANGEMENTS

None.

Also, no program income is anticipated during the period for which the grant is requested.

10. REFERENCES

Letter of Support:

Angela Barron-McBride, Ph.D., R.N., F.A.A.N.

Distinguished Professor and University Dean

Indiana University

School of Nursing

Indianapolis, Indiana 46202

Letters of Reference:

Jean Johnson, Ph.D., R.N.

Professor

Univesity of Rochester

School of Nursing

Rochester, New York 14642

Mary Jirovec, Ph.D., R.N., F.A.A.N.

Associate Professor

Wayne State University

School of Nursing

5557 Cass Avenue

Detroit, Michigan 48202

Victoria Champion, D.N.S., R.N., F.A.A.N.

Professor and Associate Dean for Research

InAPPENDIX C

PSYCHOLOGICAL WELLBEING IN COGNITIVELYIMPAIRED PERSONS

Indiana University

School of Nursing

1111 Middle Drive

Indianapolis, Indiana 46202

Jane Backer, D.N.S., R.N.

Assistant Professor

Indiana University

School of Nursing

1111 Middle Drive

Indianapolis, Indiana 46202

APPENDIX I

PSYCHOLOGICAL WELL BEING IN COGNITIVELY IMPAIRED PERSONS

Please rate the following patient behaviors on how frequently you have observed the patient displaying the behavior during the last 24 hours. The response choices are: 1=never; 2=a little; 3=sometimes; 4=frequently; 5=always.

1 2 3 4 5

1. Smiles or laughs in response to contact from others.

2. Relaxed facial expression (appears tranquil or at

ease), jaw not clenched or tightened.

3. Positive verbal response to contact from others

(not angry or disapproving).

4. Frightened facial expression (appears scared,

troubled, or alarmed).

5. Attempts to strike or hit others in angry manner.

6. Saddened facial expression (frowns or cries openly).

7. Moves freely in the environment.

8. Freely manipulates objects in the environment.

9. Asks repeatedly to go elsewhere.

10. Pulls or moves away in response to contact

from others.

11. Sleeps excessively throughout the day.

12. Involved in an activity for more than 5 minutes.

13. Calm when involved in activities.

14. Frustrated or agitated when asked to complete

an activity.

15. Agitated or fearful response when touched by

another person.

16. Calm response when touched by another person.

17. Please circle the face that best represents the

patient's overall psychological wellbeing during

the past 24 hours.

7 6 5 4 3 2 1

APPENDIX B

CAREGIVER AND PATIENT DEMOGRAPHIC SHEET

Caregiver Age: Patient Age:

Sex: Sex:

Caregiver(C)/patient(P) race: White/caucasian Hispanic

Black Other

Specify:

Caregiver Education: Total years (including high school):

Patient Education: Total years (including high school):

Length of time in caregiving role: years months

Length of time patient diagnosed with dementia: years months

Caregiver medical diagnoses: (please list)

Patient medical diagnoses: (please list)

Medications patient is currently taking:

Other sources of caregiver support: (for example, family members, home care nurses,

support groups, including length of time support received)