The course of irreversible dementias, including Alzheimer's disease (AD) and vascular dementia, which are the most prevalent, is one of progressive impairment. Eventually, most individuals with dementia require placement in a nursing home to provide the needed care. For more than 2 decades, specifically structured environmental strategies, such as Special Care Units (SCUs), have been recommended as a solution for the care of individuals with dementia when the burden of care for family members becomes too great. SCUs have also been recommended to minimize excess disability and disruptive behaviors that are problematic in traditional nursing homes. Although a "predictable phase in the evolution of the long term care industry" (Maas, Hall, Specht, & Buckwalter, 1992, p. 46) and a useful strategy for caring for individuals with dementia in nursing homes, a variety of issues for residents, family members, and staff caregivers remain under-researched in SCUs.
Also, a number of nursing interventions for individuals with dementia in the home and in institutions need to be systematically tested in terms of outcomes for residents, families, and staff. To this end, several studies evaluating nursing interventions for individuals with dementia and their family caregivers have been conducted by faculty at the University of Iowa College of Nursing (Butcher, Holkup, Park, & Maas, 2001; Gerdner, Hall, & Buckwalter, 1996; Kelley, Specht, & Maas, 2000; Kelley, Swanson, Maas, & Tripp-Reimer, 1999; Maas et al., 1994; Maas et al., 2001; Maas et al., 2004; Maas, Specht, Waler, Buckwalter, & Turner, 1998; Schutte, Maas, & Buckwalter, 2003).
Reliable and valid measures of outcomes for residents with dementia and their family and staff caregivers are crucial to the systematic testing of intervention strategies. Developing and testing measures that are psychometrically sound for these populations, however, are complicated by a number of issues. Many research questions posed by nurse investigators examine phenomena that are not easily captured by instruments often borrowed from other disciplines. Few of the clinical problems and research questions encountered in working with elderly individuals are unidimensional, and most require indirect as well as multiple measures.
Eighty-five percent (85%) of all nursing home staff and an even higher percent (90%) of all nursing staff are non-professionals (Mezey, Lynaugh, & Cartier, 1989). Instruments may be overwhelming, confusing, and overly burdensome for these staff caregivers as well as for family members. Measurement tools often must be modified to reflect participants' interests, values, abilities, and limitations or they produce invalid responses.
Because of these issues, new, more appropriate and sensitive tools must be developed and tested. Each of the studies conducted by the Iowa investigators required the development or modification of instruments to measure outcomes for residents, family members, or staff. To share the instruments with other investigators who may find them useful and to encourage their use in gerontological research, this article describes six of these instruments, their development, and psychometric evaluation to date.
THEORETICAL FRAMEWORK AND RATIONALE FOR THE MEASURES
The theoretical framework for the intervention research related to individuals with dementia and their family and staff caregivers is derived from the person-environment fit/interaction concept (Kahana, 1975; Lawton, 1975). Person-environment fit/interaction suggests that dysfunctional and socially disruptive behaviors of individuals with dementia results from their inability to interpret and cope with environmental stimuli that overwhelm their compromised cognitive abilities (Hall & Buckwalter, 1987; Maas et al., 1994). The progressive course of dementia impairs ego-sensory, perceptual, and cognitive processes and affects the dementia resident s overall ability to interact successfully with the environment. Because of these impairments, individuals with dementia often exhibit behaviors that indicate disordered person-environment interactions, such as repetitive verbalizations and actions, catastrophic behaviors, and inappropriate behaviors (Roberts & Algase, 1989).
Based on the premise that these disordered interactions occur when environmental demands exceed the resident's ability to adjust, the Progressively Lowered Stress Threshold (PLST) is used to further guide programs of research (Hall & Buckwalter, 1987). The PLST model proposes that individuals with dementia need environmental conditions modified as they experience progressive cognitive decline so cues can be more easily processed with less stress. The PLST model further suggests that if the stress experienced is reduced, individuals with dementia will be less agitated and their cognitive and functional adaptive behaviors will be enhanced.
Family caregivers also experience disordered person-environment interactions when they provide care in the home and when it is necessary to institutionalize their relatives with dementia. In the home, the role of the family member must be adjusted as the loved one's dementia progresses, making the individual with dementia increasingly dependent, disruptive, and less able to assume their usual roles in the family unit. Family caregiver burden and stress from daily hassles and loss may result.
After institutionalization, family caregiving continues (Kelley et al., 1999) and the family caregiver experiences role transition and adjustment along several continua. Family members become visitors rather than primary caregivers, they must adjust to having low control rather than the previous high control over the care of their loved ones, and they become outsiders in the caregiving process. When the resident with dementia is in a long-term care facility, staff caregivers assume the primary caregiving role and are the insiders with high control over the priorities and regulation of care (Buckwalter, Maas, & Reed, 1997). Therefore, the relationship between the family member and institutional staff is often problematic, limiting the quality of the resident's care (Butcher et al., 2001; Dellasega & Mastrian, 1995) and increasing the stress for the family member. Disruptions of family caregivers' interactions with the environment occur as their roles interface with the institution and staff. These disruptions are viewed as stressful for family caregivers (Collìns, Stommel, Wang, & Given, 1994).
Role adjustment is complicated by the institution's bureaucracy, requiring family members to assume an unfamiliar and unnegotiated role determined by the nursing home. Increased family stress and dissatisfaction with care are seen as manifestations of stress in response to discrepancies between the realities and expectations (role conflict, role ambiguity), whereas, satisfaction reflects congruence (Kelley et al., 1999; Maas et al., 1994).
Staff interactions and their fit with the environment can also be stressful and can be predicted from the theoretical framework. There are interactions within the institutional environment that have the potential to be stressful for staff caregivers. Role conflicts between staff and families forge negative attitudes about families and are sources of staff stress and dissatisfaction. The resident may be perceived by the staff as "theirs," with the family seen as having entrusted the resident to the nursing home staff (Clifford, 1985). Staff often resent family members' requests for care because staff believe they know best and do not like to have their care routines questioned or altered. Often family and staff assume a competitive or adversarial relationship as the family system and the institutional bureaucracy interface (Buckwalter & Hall, 1987; Dietsche & Pollman, 1982; Maas et al., 2004; Mobily, Maas, Buckwalter, & Kelley, 1992).
Staff stress is also generated from the burden of caring for individuals with AD. Burden is conceptualized as including the daily hassles of caring for progressively deteriorating residents, the behavioral management and the emotional responses associated with the very difficult care problems residents with dementia usually present. Given the demands for care and the limited available resources, role conflict, role ambiguity, and role overload are important concepts to assess in the measurement of staff stress.
The instruments presented in this article represent measures of outcomes for individuals with dementia, family caregivers, and staff caregivers for which the authors were unable to locate appropriate standardized measures. The measures were developed from the person-environment fit/interaction framework, role theory, and the PLST model. Reliable and valid measures of these outcomes are critical for assessing the effects of care strategies on individuals with dementia, their family members, and staff caregivers. Studies using the measure are listed in the Table, along with the purpose of each study.
Three instruments presented in this anide., the Functional Abilities Checklist (FAC), Family Perceptions of Caregiving Role tool (FPCR), and the Caregiver Stress Inventory (CSI), are available at the following website: http://www.nursing. uio wa, edu/hartford/ under Family Involvement in Care/instruments. Readers may also request copies of the instruments by firstname.lastname@example.org.
Functional Abilities Checklist (FAQ
The FAC is a 28-item instrument examining the AD patient in relationship to grooming, hygiene, toileting, bathing, eating, sleeping, exercising, and interacting with others in the environment. The FAC was developed by Maas and Buckwalter (1990) because other instruments in the literature did not include all areas relevant for assessing the functional abilities of individuals with AD. The FAC was used to measure four domains of the functional abilities of AD patients:
DESCRIPTIONS OF STUDIES
* Self -care ability.
* Inappropriate behavior.
* Cognitive status.
* Agitated behavior.
The instrument assesses those patient behaviors observed during the previous week and uses a 1 to 4 scale (1 = never; 4 = multiple times per day) to rate behaviors. A high score indicates a high degree of impairment.
Reliability. Test-retest reliability with a 2-week interval for 57 nursing home residents with dementia for Study 1 was .62 (p = .000). For 41 residents in subsequent research the test-retest coefficient was .77 (p = .000). Cronbach s alpha coefficients for the subscales using data from Study 2 ranged from .63 to .86 at pre-test, but were not stable across the baseline and post-test intervals (Maas & Buckwalter, 1990). Inter-item correlation coefficients were reassessed and the subscales were revised to five subscales:
* Inappropriate behavior.
* Cognitive status.
* Sexual behavior.
The three items of the sexual behavior subscale were deleted because of continued instability of alpha coefficients. An Ítem analysis led to the deletion of one more Ítem resulting in the 28-item, four-subscale format for the FAC. The Cronbach's alpha coefficients for Study 3 were: self care abilities subscale (.85); inappropriate behaviors subscale (.84); cognitive status subscale (.74); agitation subscale (.72); and FAC total (.85) (Maas et al., 2000). Item 6 in inappropriate behaviors was deleted because of the instability of the alpha coefficient.
Validity. Validity of the FAC was assessed by correlating the subscale total scores with Geriatric Rating Scale (GRS) scores (Plutchìk et al., 1984) and by performing a discriminant function analysis. As an estimate of concurrent criterion validity, data collected bimonthly in the study to evaluate a SCU were used (Maas & Buckwalter, 1990). The 31-item GRS is a standardized observational measure of activities of daily living rated by nursing staff. Correlation coefficients were .52 or greater (p = .000) for the self-care Ítem grouping and the FAC self care subscale, but were smaller or inverse for inappropriate behavior, cognitive status, and agitation, indicating that the FAC measured some different aspects of function than the GRS.
For the application of discriminant function analysis, 62 nursing home residents were divided into two groups of 31 in an impaired group (1 to 2 ratings) and 31 in an unimpaired group (4 to 5 ratings), and 67% of the cases were selected randomly. Analysis of the 42 (67%) randomly selected cases was significant (24 = 34.725,/i < .001) with 85.7% of the 42 cases classified correctly. Analysis of the 20 unselected cases resulted in 95% classified correctly. The results supported the discriminating effectiveness of the four subscales of the FAC in differentiating between impaired and unimpaired individuals with AD. The instrument also discriminated between residents in a residential long-term care facility who were functionally able and AD residents in an intermediate care facility who were functionally impaired (p < .05). These data supported the revised FAC s construct validity.
Family Perceptions of Care Tool (FPCT)
The FPCT, developed by Maas and Buckwalter (1990) for the study to evaluate the effects of an SCU, is a selfreport questionnaire measuring family member satisfaction with the care of the individual with dementia. Each of 51 items is assessed on a 7-point Likert scale with 7 indicating highest satisfaction and 1 indicating highest dissatisfaction. Item scores are summed to form 4 subscale scores and a total FPCT score.
Family members typically complete the instrument in 20 to 30 minutes, although some family caregivers, particularly elderly spouses, require longer or need some assistance to rate the 51 items. The questionnaire also contains an area where the family member is asked to write any other comments or concerns they have about the care of their relative with AD. Based on internal consistency reliability analyses for Study 3, which evaluated the Family Involvement in Care (FIC) intervention (Kelley et al., 2000) the instrument was revised to include four subscales: staff consideration of family and resident, management effectiveness, physical care, and activities.
Reliability. During Study 1, test-retest reliabilities for the subscales ranged from .78 to .90 (all/; < .05), based on data from 15 family members of institutionalized individuals with AD, with an interval of 10 days. Using data from Study 2 (Maas & Buckwalter, 1990), internal consistency reliability coefficients for the FPCT (Version 2) were computed for each of the six bimonthly data collections before and after the opening of the SCU. Cronbach's alpha coefficients across Study 2 pre-intervention data collections ranged from:
* .77 to .92 (satisfaction with environment).
* .9 1 to .93 (satisfaction with nursing care).
* .74 to .92 (satisfaction with AD resident, staff, and family relationships).
* .87 to .96 (satisfaction with overall care), and .92 to .97 (total FPCT).
In Study 2 post-intervention data, internal consistency reliability coefficients (Cronbach's alpha) ranged from:
* .83 to .94 (satisfaction with environment).
* .82 to .89 (satisfaction with nursing care).
* .87 to .92 (satisfaction with AD resident, staff, and family relationships).
* .87 to .96 (satisfaction with overall care).
* .92 to .97 (total FPCT).
For Study 3, internal consistency reliability coefficients figured on the baseline data were .97 for physical care, .87 for activities, .88 for management effectiveness, and .85 for consideration (Maas et al., 2000). The Cronbach's alphas for Study 4 conducted with Korean family caregivers (Park, 2001) were .83 for physical care, .8 1 for activities, .60 for management effectiveness, .70 for consideration and .94 for the total scale. Greater variance in scores was found for Ítems that refer to care that family members would like their relatives to have if there were more staff and more resources available.
Validity. Content validity was assessed by a panel of gerontologicai nurses and social workers who reviewed and revised the instrument Item selection was based on a review of the limited literature about family members of institutionalized relatives and were derived from the theoretical framework. A review of literature at the time the tool was developed revealed only instruments for measuring family caregrver perceptions for non-institutionalized individuals with dementia. Thus, no criterion measure was found for further validation.
Analysis of the family members responses related to additional comments and concerns validated findings from the FPCT. The majority of the family members who returned the completed FPCT with comments (N = 70), indicated that they had no asdditional concerns or that they were very satisfied with the care. However, 14 family members wrote about specific concerns, including relationships with staff and feeling left out of the AD relative's care (n = 6), proper care (n = 4), and concern that the patient did not have enough social and recreational diversions (n = 6). A check of the responses of these family members on the FPCT showed that the ratings of satisfaction and dissatisfaction were consistent with their qualitative comments.
Family Perceptions of Caregiving Role (FPCR)
The FPCR tool is a 61-item self-report tool to measure multiple dimensions of family member stress associated with an institutionalized relative with dementia. This tool was developed by Maas and Buckwalter (1990), and some items were adapted from the Burden Inventory (Zarit, Todd, & Zarit, 1986). Items related to grief, role deprivation, and role capacity were adapted from measures developed by Pearlin, Mullan, Semple, and Skaff (1990). The content of the instrument was derived from the theoretical frameworks of PersonEnvironment Fit and Interaction (Kahana, 1975) and Role Theory (Hardy & Conway, 1978; O'Neill & Ross, 1991). Items are rated on a 7-point Likert-type scale (1 = strongly disagree to 7 = strongly agree). Originally, the instrument was designed to measure stress and burden from the caregiver role, and the loss resulting from the caregiver role. Factor analysis of data from Study 3 yielded four factors defined by the dimensions of:
* Loss of aspects of the relationship with relatives with dementia.
* Guilt from perceived failure in caregiving.
* Captivity resulting from obligations of caregiving.
* Conflict with staff related to caregiving.
Reliability. In Study 3 pre-testing with 50 family members, Cronbach's alphas were .93 for total stress, .93 for caregiving role stress, .94 for burden, and .87 for loss. Test-retest reliability with 12 family members with a 3-week interval was .79 (p < .001). Reported Cronbach's alphas for Study 3 were .73 for the loss subscale, .70 for the guilt subscale, .81 for the captivity subscale, and .84 for the conflict subscale (Maas et al., 2000). The Cronbach's alphas for Study 4 were .83 for loss, .74 for guilt, .80 for captivity, .60 for conflict, and .94 for the total scale (Park, 2001).
STAFF CAREGIVER INSTRUMENTS
Caregiver Stress Inventory (CSI)
The 43-item CSI, developed by Maas and Buckwalter (1990) measures the extent of stress experienced by staff associated with the care of AD residents. Staff caregiver stress is the response individual staff caregivers have to the incidents occurring in the daily care of individuals with AD. Each Ítem is self-rated by staff members on a 7-point Likert scale (1 = not stressful, 7 = extremely stressful). The instrument requires 15 to 20 minutes to complete. The questionnaire was developed using a "critical incidents technique" by asking 50 nursing home employees to describe a highly stressful event involving a resident with AD that had occurred to them during the previous week. Factor analysis using data from Study 3 yielded four factors representing staff stress from catastrophic behavior, inappropriate behavior, resident safety, and resource deficiency.
Reliability. Internal consistency reliability (Cronbach's alpha) ranged from .90 to .95 for the six pre-intervention data collections for Study 2. The ranges of internal consistency coefficients for the subscales were: verbaVphysical = .92 to .96; emotional/mental = .94 to .97; and knowledge/resources = .87 to .92. Internal consistency coefficients for the post-intervention data collections of Study 2 were stable and ranged from .98 to .99 (total score), .94 to .98 (verbaVphysical), .95 to .98 (emotional/ mental), and .92 to .96 (knowledge/resources) (Maas & Buckwalter, 1990). Reported Cronbach's alphas of each subscale for Study 3 were .91 for the catastrophic behavior subscale, .94 for the inappropriate behavior subscale, .83 for the resident safety subscale, .82 for the resources deficiency subscale, and .96 for the total scale (Maas et al., 2000). The Cronbach's alphas for Study 4 with Korean staff members were .87 for the catastrophic behavior subscale, .91 for the inappropriate behavior subscale, .86 for the resident safety subscale, .85 for the resources deficiency subscale, and .95 for the total scale (Park, 2001).
Validity. Content validity of the CSI is supported by the critical incident approach used to develop Ítems of the instrument. To assess construct validity of the CSI, subscale and total scores were correlated with subscale scores of the Maslach Burnout Inventory (MBI) (Maslach & Jackson, 1981). The MBI ìs a standardized, selfreport questionnaire that measures the frequency and intensity of perceived burnout of individuals in helping professions. Hypothesized correlations of the subscales with other measures of theoretically related variables have supported the validity of the MBI (Hackman & Oldham, 1975; Iwanicke & Schwab, 1981; Maslach & Jackson, 1981). Test-retest reliability estimates ranged from .53 to .82 for all six subscales (Maslach & Jackson, 1981).
As hypothesized, correlations between the extent of stress measured by the CSI and the Emotional Exhaustion and Depersonalization dimensions of the MBI were moderate and positive, whereas the extent of stress and the Personal Accomplishment dimension of the MBI were moderately and negatively correlated throughout Study 1 (Maas & Buckwalter, 1990).
Staff Perceptions of Caregiving Role (SPCR)
The SPCR is a 58-item self-reported instrument developed by Maas and Buckwalter (1990) to measure staff perceptions of family and staff caregiving roles as sources of stress. Many of the items are analogous to items in the FPCR instrument. Various aspects of staff caregiver role and burden are assessed as well as interactions and negotiations with family members. Some Ítems were adapted from the Burden Inventory (Zarit et al., 1986). Various aspects of staff caregiver role and burden are assessed, including interactions and negotiations with family members. Factor analysis using data from Study 2 yielded four dimensions: task burden, role inadequacy, dominion, and exclusion.
Task burden represents the costs of caregiving, such as feeling angry or overworked. Role inadequacy is the feeling that there are no benefits from caregiving. Dominion reflects the attitude that staff should be in control of caregiving. Exclusion means the disposition that family caregivers should not participate in caregiving within the facility. It is conceptually distinct from dominion, because family caregivers conceivably may participate extensively without being involved in decision-making regarding care for their relatives with dementia.
Reliability. Pre-testing with 50 staff for Study 3 yielded Cronbach s alpha coefficients of .88 (total stress), .72 (role stress), and .88 (burden). Reported Cronbach *s alphas for Study 3 were .71 for the dominion subscale, .70 for the exclusion subscale, .84 for the task burden scale, and .82 for the role inadequacy subscale (Maas et al., 2000). The Cronbach's alphas for Study 4 were .64 for the dominion subscale, .70 for the exclusion subscale, .61 for the task burden scale, and .70 for the role inadequacy subscale (Park, 2001).
Validity. Construct validity for the CSI and SPCR was supported by analyzing the correlations between subscales of the two instruments using staff data from Study 3. The correlations for all subscales were statistically significant ranging from .325 to .498, except correlations between resident safety in CSI, and dominion and exclusion in SPCR. SPCR subscales of dominion and exclusion had low correlations with the resident safety subscale in the CSI (r = .121, - .177, respectively).
Attitudes Toward Families Checklist (AFC)
The AFC is a 16-ìtem self-report instrument used to assess staff attitudes about family members of residents with dementia, including family visitation, family requests regarding care of their relative, and family participation in the care of their relative. Maas and Buckwalter (1990) developed the AFC. Staff rate the items on a 7-point Likerttype scale (1 = strongly disagree; 7 = strongly agree). A high score indicates that the staff caregiver has a positive attitude toward regular visitation from family members, family requests concerning care of the resident with dementia, and the family caregiving role for their relatives. The AFC consists of three subscales: calming (reflecting whether family caregivers are seen as having a calming effect or a disruptive effect on residents), partner (reflecting whether family caregivers are seen as equal partners in care or not), and relevant (reflecting whether family caregivers are seen as relevant to residents and staff).
Reliability. The AFC was pretested with 50 nursing home staff from Study 3 and yielded an internal consistency reliability coefficient (Cronbach's alpha) of .91. Test-retest reliability with 15 staff at a 3 -week interval was r = .84. Cronbach's alphas of the subscales with the data from Study 3 were .64 for the calming subscale, .58 for the partner subscale, .61 for the relevant subscale, and .74 for the total scale (Maas et al., 2000). The Cronbach's alphas for Study 4 were .56 for the calming subscale, .63 for the partner subscale, .55 for the relevant subscale, and .70 for the total scale (Park, 2001).
Content validity for the FPCR, SPCR, and AFC were assessed by a panel of gerontologìcal nurses and social workers who reviewed and revised the instruments. Item selection was based on a review of the limited literature related to family and staff members in SCU and was derived from the theoretical framework. A review of literature at the time the tool was developed revealed instruments that only measured staff caregiver perceptions for general long-term care settings versus SCUs. Thus, no criterion measures were found for further validation of these three tools.
CLINICAL APPLICATION OF INSTRUMENTS
Patient Measure: FAC
The FAC provides a way for nurses to assess functional performance and behaviors of residents with dementia. The instrument can easily be used to document progress or deterioration following nursing intervention, because it requires 10 minutes or less to rate. Single subscales of the instrument, such as the agitation subscale, require even less time and can be used to evaluate the effectiveness of specific interventions for more specific problems.
The results of assessment using the FAC are also useful for care planning by helping to identify problem areas that are amenable to intervention. Because the FAC combines measures of self-care performance, cognitive status, and behaviors commonly encountered among individuals with dementia, it is especially useful in planning and evaluating the care of these long-term care residents. Completion of the FAC by nursing home staff during each shift to identify differences in function and behavior throughout a 24-hour period improves the comprehensiveness of data used to complete the Resident Assessment Instrument (RAI, Centers for Medicaid and Medicare Services, 2005). The use by each shift enhances the validity of RAI data and increases the relevance of the plan of care for staff throughout the 24-hour care cycle.
Use of the FAC also adds needed detail for selecting appropriate interventions for residents whose self-care limitations are closely associated with cognitive status, agitation, and difficult to manage behaviors. The FAC is helpful in improving shared perceptions of residents and their care needs and in training staff to observe and document resident function and behavior. Comparison of registered nurse and assisting staff ratings provides excellent content for the registered nurse to hear staff perspectives about residents, teach observation and care strategies, and to provide the leadership for staff that is so badly needed in long-term care settings.
Family Measures: FPCT and FPCR
The FPCT is an instrument to measure family member satisfaction with care as part of a quality improvement program. Ratings on the instrument could be one measure of quality of life because families often must serve as proxies for at least a portion of the measurement of quality of life when residents have dementia (Montgomery, 1994). The FPCT can be completed for individuals on SCUs or individuals with dementia throughout the facility or a representative number of randomly selected residents. Providing a way for anonymous responses will likely increase the honesty of responses because retribution from staff for saying negative things is often a fear of family members. The four subscales help to identify specific areas of satisfaction or dissatisfaction, allowing the facility staff to focus on particular areas for improvement and sharing areas of excellence identified by families with staff. The instrument is an effective interdisciplinary measure of satisfaction and could be useful to the whole team in the measurement of quality.
Family caregivers continue to play an important role in the care of institutionalized individuals with dementia (Spencer, 1991). Often, family member visitation and participation with care of the individual with dementia declines as the disease progresses and makes the individual less socially accessible. Assessing family member Stressors with the FPCR provides information about the sources and severity of stress perceived by family members. This information can be used to help family members decrease their stress through education and support groups, referrals for additional services, work with family-staff relationships, and individual counseling. Assisting family members to cope with and decrease Stressors will not only help the family members, but also will help them remain involved in care of their relatives with dementia. The conflict scale can be used as a quality indicator for family-staff relationships. The use of the FPCR conflict subscale coupled with the AFC gives both family and staff perspectives and strengthens assessment of the quality of family and staff relationships.
Staff Measures: CSI, SPCR, and AFC
The clinical applications of the staff instruments are their usefulness for quality assurance programs and for assessing the effectiveness of plans of correction/interventions identified as part of the quality improvement plan. In addition, all three instruments provide excellent information about content needed in training and development programs and are also helpful in evaluating the usefulness of the programs presented.
The CSI and SPCR are useful in determining behaviors of individuals with dementia and interactions and expectations of families that are most stressful for staff caregivers. The results of the survey can then be used to develop training and development programs for staff to better prepare them to work with resident behaviors and with families to reduce staff stress. With the current shortages of nursing home staff and high turnover rates, identifying and addressing workplace Stressors is critical.
The AFC is a good quality indicator for staff relationships with families, with higher scores representing more positive attitudes. If scores are low, educational programs, stafffamily meetings, and/or the Family Involvement in Care Protocol, which was developed to facilitate partnership between family and staff caregivers (Kelley et al., 2000) can be implemented. Use of the AFC provides a very useful and easy way to evaluate and improve staff and family relationships.
CONCLUSION AND RECOMMENDATIONS
The instruments described in this article measure resident, staff, and family outcomes. They have been tested and improved with multiple studies for use with residents with dementia and the staff and family caregivers who care for them. One objective of this program of research, testing nursing interventions for individuals with dementia, family, and staff caregivers, was to develop reliable and valid measures of outcomes when suitable standardized tools were not available. The testing indicated that the psychometric properties and ease of use of all of the instruments are reliable. Qualitative measures (semistructured interviews with family and open-ended questions) supported the quantitative findings of the instruments, adding validity to the measures (Butcher et al., 2001; Kelley et al., 1999). The authors believe that other investigators will find these six instruments useful, and anticipate further psychometric assessment of them by other researchers.
The measures of satisfaction and burden of family and staff members (i.e., FPCT, FPCR, SPCR, CSI, AFC) are especially important because they fill a void not previously addressed - that is, measuring family and staff caregivers' outcomes in long-term care settings. The satisfaction and burden instruments are multi-dimensional, and when used together capture the perceptions of staff and families. Assessing similar dimensions for staff and family caregivers enhances the understanding of care and the care environment from both perspectives that are needed to achieve quality outcomes. Another advantage of the instruments is their usefulness for both research and clinical applications, including assessment, quality improvement, and training.
Finally, the beginning tests with another culture demonstrate the usefulness of the measures cross-culturally (i.e., for Korean caregivers). This evidence is important because of the increasingly diverse ethnic groups represented among residents, family members, and staff caregivers in longterm care facilities. Instruments that can be translated to accurately measure resident, family, and staff needs, and outcomes with cultural sensitivity are extremely important given the global nature of health care.
Nurse investigators are urged to meet the challenge of measurement that is specifically relevant for nursing research with individuals with dementia and their family and staff caregivers. This challenge is crucial to the development of the knowledge base for the discipline. There is a particular need for nursing research with vulnerable elderly and long-term care populations where a high proportion of clinical problems fall within the scope of nursing practice. Without effective nursing care, the care of older individuals is clearly compromised (Cohen & Spector, 1996; Dellefield, 2000). The instruments reported in this article contribute to the reliable and valid measures of outcomes available for systematically testing and evaluating the effectiveness of nursing interventions for elderly individuals with dementia and their family and staff caregivers.
- Buckwalter, KC, Sc Hall, G. (1987). Families of the institutionalized older adult: A neglected resource. In TH. Brubaker (Ed.), Aging, health, and family (pp. 176-196). Newbury Park, CA: Sage Publications.
- Buckwalter, K., Maas, M., Sc Reed, D. (1997). Assessing family and staff caregiver outcomes in Alzheimer disease research. Alzheimer Disease and Associated Disorders, í(Suppl. 6), 105-116.
- Butcher, H., Holkup, P., Park, M., St Maas, M. (2001). A thematic analysis: The experience of making a decision to place a family member with Alzheimer's disease in a special care unit. Research in Nursing & Health, 24(6), 470-480.
- Clifford, A. (1985). Your mother is ours now. Journal of Gerontological Nursing, ./0(1), 44.
- Centers for Medicaid and Medicare Services. (2005). Revised Long Term Care Resident Assessment Instrument User's Manual for the Minimum Data Set (MDS) -version 2.0. Retrieved May, 16, 2005, from http://www.cms. hhs.gov/quafity/mds20/
- Cohen, J., St Spector, W (1996). The effect of Medicaid reimbursement on quality of care in nursing homes. Journal of Health Economies, 75(4), 23-48.
- Collins, C., Stommel, M., Wang, S., St Given, GW. (1994). Caregiving transitions: Changes in depression among family caregivers of relatives with dementia. Nursing Research, 43(4),22Q-225,
- Dellasega, C, St Mastrian, K. (1995). The process and consequences of institutionalizing an elder. Weitern Journal of Nursing Research, 17(2), 123-140.
- Dellefield, M.E. (2000). The relationship between nurse staffing in nursing homes and quality indicators. Journal of Gerontological Nursing, 26(6), 14-28.
- Dietsche, L., 8c Pollman, J.N. (1982). Alzheimer's disease: Advance in clinical nursing. Journal of Gerontological Nursing, 8(2), 97-100.
- Gerdner, L.A, Hau, G.R, St Buckwalter, KC. (1996). Caregiver training for people with Alzheimer's based on a stress threshold model. Image: The Journal of Nursing Scholarship, 28(1), 241-246.
- Hackman, J.R., St Oldham, G-R. (1975). Development of the Job Diagnostic Survey. Journal of Applied Psychology, 60, 159-179.
- Hall, G.R., Oc Buckwalter, K.C. (1987). Progressively lowered stress threshold: A conceptual model for care of adults with Afzheimer's disease. Archives of Psychiatric Nursing, 1(6), 399-406.
- Hardy, M.E., St Conway, M. (1978). Role theory: Perspectives for health professionals. New York: Appleton-Century-Crofts.
- Iwanicke, E.F., St Schwab, RL. (1981). A crossvalidation study of the Maslach Burnout Inventory. Educational and Psychological Measurement, 41(4), 1167-1174.
- Kahana, E. (1975). A congruence modelfor care of person-environment interaction. In P. Windley, T Byerts, St F. Ernst (Eds.), Theory development in environment andaging (pp. 181-214). Washington, DC: Gerontologicat Society.
- Kelley, L.S., Specht, J.K, St Maas, M.L. (2000). Family involvement in care: A research-based protocol for persons with dementia or other chronic illnesses. Journal of Gerontological Nursing, 26(2), 13-21.
- Kelley, L., Swanson, E., Maas, M., & Tripp-Reimer, T (1999). Family visitation on special care units. Journal of Gerontological Nursing, 25(2), 14-21.
- Lawton, M.P. (1975). Competence, environmental press, and the adaptation of older people. In P. Windley, T Byerts, St F. Ernst (Eds.), Theory development in environment and aging (pp. 13-88). Washington, DC: Gerontological Society.
- Maas M., St Buckwaíter, K.C. (1990). Nursing evaluation research: Alzheimer's care unit [NR01689]. Bethesda, MD: NIH Center for Nursing Research.
- Maas, M., Buckwalter, KG, Swanson, E., Specht, J.,Hardy,M, St Reimer, T.T. (1994). The caring partnership: Staff and families of persons institutionalized with Alzheimer's disease. Journal of Alzheimer's Care and Related Disorders and Research, 9(6), 21-30.
- Maas, M., Hall, G., Specht, J., St Buckwalter, K.C. (1992). Dedicated, not isolated: Development of long-term care Alzheimer's units. In K.C. Buckwalter (Ed.), Geriatric mental health nursing (pp. 44-60). Thorofare, NJ: Slack Publishing.
- Maas, M.L., Reed, D., Park, M., Specht, J., Schutte, D., Kelley, L.S., Swanson, E.A, Tripp-Reimer, T, St Buckwalter, KC. (2004). Outcomes of family involvement in care intervention for caregivers of individuals with dementia. Nursing Research, 53(2), 76-86.
- Maas, M., Reed, D., Specht, J., Swanson, S., TrippReimer, T, Buckwalter, K, Schutte, D, St Kelley, D. (2001). Family involvement in care: Negotiated family-staff partnerships in special care units for persons with dementia. In S.G. Funk, E.M. Tornquist, J. Leeman, M.S. Miles, Sc J.S. Harrell (Eds.), Key aspects of preventing and managing chronic illness (pp. 330-345). New York: Springer.
- Maas, M., Specht, J., Weiler, K, Buckwalter, K, ôc Turner, B. (1998). Special units for persons with Alzheimer's disease: Only for the privileged few? Journal of Gerontological Nursing 24(3), 28-37.
- Maas, M., Swanson, E., Buckwalter, K.C., Lenth, R, Specht, J.P., Tripp-Reimer, T, Tranci, D., Reed, D, Broffitt, B., Brenneman, D., Peters, J., Rose, D., Kelley, L., Pottinger, J., Ramler, C, Schutte, D, Sung, C., St Park, M. (2000). Final report: Nursing interventions for Alzheimer's family role trials [ROl NR01689NCNR]. Rockville, MD: National Institutes of Health.
- Maslach, G, St Jackson, S. (1981). Maslach Bumout Inventory. Palo Alto, CA: Consulting Psychologists Press.
- Mezey, MD., Lynaugh, J.E., St Cartier, M.M. (1989). Nursing homes and nursing care: Lessons from the teaching nursing homes. New York: Springer.
- Mobily, P, Maas, M., Buckwalter, K, St Ketley, L. (1992). Taking care of the caregivers: Staff stress and burnout on a special Alzheimer's unit. Journal of Psychosocial Nursing, 30(9), 25-31.
- Montgomery, R. (1994). Family measures in the special care unit context. Alzheimer Disease and Associated Disorders, #(Suppl. 1), S242-S246.
- O'Neill, G, St Ross, M.M. (1991). Burden of care: An important concept for nurses. Health Care for Women International., 12, 1 1 1-122.
- Park, M. (2001). Korean and U. S. family and staff caregivers' perceptions of care in dementia care units: A cross-cultural triangulation approach. Unpublished doctoral dissertation, University of Iowa, Iowa City, IA.
- Pearlin, L.I., Mullan, J.T., Semple, S.J., St Skaff, MM. (1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30(5), 583-594.
- Plutchik, R., Conte, H., Lieberman, M., Baker, M., Grossman, J-, St Lehrman, N. (1984). Geriatric rating scale. In DJ. Mangen, St WA. Peterson (Eds.), Research instruments in social gerontology, Vol. 3: Health, program evaluation, and demography (pp. 4043). Minneapolis, MN: University of Minnesota Press.
- Roberts, B.L., St Algase, D.L. (1989). Victims of Alzheimer's disease and the environment. Nursing Clinics of North America, 35(2), 113-118.
- Schutte, D.L., Maas, M., St Buckwalter, KC. (2003). A LRPAPl intronic insertion/deletion polymorphism and phenotypic variability in Afzheimer disease. Research and Theory for Nursing Practice: An International Journal, 77(4), 301-319.
- Spencer, B. (1991). Partners in care: The role of families in dementia care units. In D.H. Coons (Ed.), Specialized dementia care units (pp. 189209). Baltimore: The Johns Hopkins University Press.
- Zarit, S.H., Todd, PA, St Zarit, J.M. (1986). Subjective burden of husbands and wives as caregivers: A longitudinal study. The Gerontologist, 26, 260-266.
DESCRIPTIONS OF STUDIES