logo Living Dementia Consulting

alzheimer journal

The Relation Between the Perceived Role of Family and the Behavior of the Person with Dementia

Gwendolyn de Geest RN, BSN, MA

This paper was presented at the World Congress of Gerontology in Vancouver, BC, Canada in July, 2001

Abstract

The purpose of this study is to investigate the relation between the perceived role of family and the behavior of the person with dementia in a sample of older adults. Two instruments were used in data collection: The Family Perceptions of Caregiving Role (FPCR), and The Cohen–Mansfield Agitation Inventory (CMAI). A total of 40 persons with dementia were included in the study. Each had a family respondent and a nurse respondent. The data analyses consisted of three major parts. First, it was revealed that there was no significant difference between agitation levels of persons with dementia who received more frequent family visits and those who did not. Secondly, the study revealed that family caregivers did not perceive a partnering relationship with staff. Finally, correlational analysis provided little evidence of a link between the family’s perceived role in partnering, and the behavior of the person with dementia, specifically agitation levels.

Although no significant relation was established between the family’s perceived role in partnering and the behavior of the person with dementia, there was a high family and caregiver satisfaction. The clearest element of the family’s perspective was the desire that staff members care about the individual with dementia as a person.

Introduction

Although intuitively it might be expected that family play a significant role in the adjustment of persons with dementia to life in a care facility, relatively little research has addressed this issue. Tickle and Hull (1995) conducted a qualitative study, using a participation–observation methodology to determine the role and function of family members visiting in long–term care settings. They found that some family members want and need to be involved in caring for their relative in long–term care. When they do participate in caring, the assumption may be that residents are more satisfied. Certainly, it has been suggested that feeling unsafe and insecure is a common feature of the dementing process and that the presence of family may help persons with dementia to feel more satisfaction and less frustration. The main objective of the present study will be to examine the relation between the perceived role of the family and the behavior of the person with dementia.

Research has shown that family support of the resident may ultimately have a beneficial effect on the behavior of some agitated residents. Cohen–Mansfield, Marx, and Werner (1992)3 conducted a 3–month observational study of 24 agitated and severely cognitively impaired nursing home residents to document the typical ways in which residents spend their time and how time use relates to the manifestation of agitated behaviors. It was revealed that these residents were involved in no activity during 63% of the observations. Yet data analysis revealed that residents manifested a greater number of agitated behaviors when they were unoccupied and fewer agitated behaviors when involved in structured or social activities. This study demonstrated that residents who had visitors exhibited a significantly lower level of total agitation than residents who did not have visitors. Results of another study showed that 93% of a group of 408 nursing home residents manifested one or more agitated behaviors at least once a week.4 Furthermore, besides being a management problem, agitation seems to indicate discontent by the nursing home residents themselves.

Less clear and less well documented, however, are specific roles that families assume in the care of their elderly members in nursing facilities and the specific degree of family involvement. Family roles in nursing facilities have been ambiguous due to an apparent difference in attitudes and perceptions between the nursing home staff and family and friends of the residents.5 Researchers conducted a study, which examined role expectancies in essential nursing services between nursing home professionals and relatives of residents. Overall, the level of agreement was remarkable between staff and relatives as to the assignment of responsibility. The results of their study show that serious and sympathetic efforts on the part of staff to encourage increased family participation will in many cases help foster a partnership between staff and families and generate broader support for the resident. Duncan and Morgan (1994) examined family caregivers’ views of their relationships with nursing home staff. Their results point to the families’ desire for emotionally sensitive care and not just for technically competent performance of tasks. Bowers (1988) found that nursing–home based family caregivers emphasized the social and emotional aspects of care, and not merely a task-oriented division of labor. The clearest element of the family’s perspective was the desire that staff members care about the resident as a person.

The purpose of this study is:

  • to determine whether the behavior for persons with dementia, differs for those who receive family visits more frequently, compared to those persons with dementia who receive family visits less frequently
  • to examine the partnering relationship between family and staff caregivers
  • to examine the relation between perceived partnering role of family and the behavior of the person with dementia.

Method

Participants

Participants for this study included staff, families and patients in the Lower Mainland of Vancouver, British Columbia. A total of 40 persons with dementia were included in the study. Each had a family respondent and a nurse respondent. Thus, there were 40 family members and six nurses ( the nurse who best knew the patient). Two settings were used to obtain the study sample: a) a 75–bed extended long–term care facility in Vancouver, and b) a 75–bed intermediate care facility in Richmond. These facilities were used for data collection purposes to insure an adequate number of clients. The data were collected over a two–month period.

At each nursing home, the researcher discussed with the nurse clinician the nature and purpose of the study, the approach to be taken, and the expectations of the participants. The nurse clinician was asked to identify individuals with dementia who were known to have families who visited for invitation to participate in the study. The sample consisted of persons with dementia and their family members. The study utilized a convenience sample, examining persons with dementia.

Identification of the person with dementia was not made on the basis of age nor gender. The criteria for selection did not include agitated behavior. In fact, at the time of this selection process, the researcher had no knowledge whether or not the person with dementia was agitated. Forty persons with dementia were identified. For each of these individuals, one staff informant and one family informant participated in the study.

Letters of explanation were then mailed to the family member outlining the project. The family member then contacted the researcher if he/she wished to participate. These willing participants were asked to sign a consent form to participate in the study. Persons with dementia were not asked to sign a consent form. For persons who had a ‘committee of person’ in place, the legal guardian was contacted concerning participation in the study. If the guardian expressed interest in having the patient take part, the researcher then obtained consent.

The rationale for involving family members was that the family may be affected in some way by the outcomes of the person with dementia and they also may influence the outcome by their perception of their caregiving role with the staff. Of these family members, 35% of the visitors were daughters, 12.5% were sons, 22.5% were husbands, 10% were wives, 10% were nieces, and 10% were friends.

The nurse clinician also identified the nurses who best knew each person with dementia for invitation to participate in the study. A letter was then sent to each identified nurse outlining the project and the purpose of the study. Each identified nurse contacted the researcher if they were willing to participate. All staff members who were contacted agreed to take part in the study. Willing participants were then asked to sign a consent form to participate in the study. Each staff informant was then given a Cohen–Mansfield Agitation Inventory (CMAI) questionnaire to complete concerning the behavior of the person with dementia he/she was identified as knowing best.

Approval to conduct this research was applied for and received from the University of Victoria Human Research Ethics Committee.

Instruments

The measurement battery consisted of two instruments: 1) The Family Perceptions of Caregiving Role Tool (FPCR) and, 2) The CMAI.

The FPCR is intended to examine outcomes and measurement issues related to caregivers for persons with Alzheimer disease. This instrument was selected because it measured the family’s perception of the caregiving role. Separate dimensions of the instrument measure stress from the caregiver role and the quality of interactions with staff, the burden of the caregiver role, and loss resulting from the caregiver role.8 From research on 112 family members, Buckwalter et al. reported Cronbach’s alphas 0.90 for the total scale. Cronbach’s alpha is a test of internal consistency and compares each item in the scale simultaneously with each other. A coefficient of 0.90 means that the FPCR demonstrated reliability; the closer to the coefficient is, the more reliable the tool. Re–analysis of the FPCR suggested three subscales corresponding to Family Role, Captivity/Burden with the emphasis on Captivity, and Loss (David Reed, personal communication, August 20, 1999). Subscale composition is being re–evaluated, although alphas currently are greater than 0.80 for all three sub–scales. Test–retest reliability with 12 family members for a 3–week interval was 0.79 (p < 0.001). This means that when the FPCR was administered to the same subjects under similar conditions on two or more occasions, there was a correlation coefficient of .79; the magnitude of this correlation supports the idea that the FPCR has the attribute of stability.

In the present study, the first 20 questions and questions 41, 42, and 43 were selected from this tool. The rationale for selecting these questions from the original scale of 81 items is that these 23 items all deal with interactions between family and staff. These items measure the quality of interaction between family and staff described by Buckwalter et al. (1997). Each of the statements in the family interview described something about the family role in the care of the family member. Upon interview, families responded to each statement in terms of the extent to which they agreed or disagreed. For this reason, Questions 4, 7, 12–14, 17–19, 41, 42, and 43 are items that needed to be reversed before the scores were calculated. The last question on the FPCR was open–ended and allowed the family to comment on any personal insights into their perception of caregiving role. For the purposes of this study these comments were not analyzed.

The CMAI is a nurses’ rating questionnaire consisting of 29 agitated behaviors. The CMAI measures how frequently episodes of agitation occur. Each item is rated on a 7–point scale of frequency (‘1’ indicates that the person never engages in the specific agitated behavior, and ‘7’ indicates that the person manifests the behavior on the average of several times an hour). In the nursing home, inter–rater agreement rates were calculated for each behavior on the CMAI (using 0–1 point discrepancy as agreement) for 3 sets of raters (in 3 units of a nursing home). These averaged .92 (n = 16), .92 (n = 23), and .88 (n = 31). Because the CMAI is an instrument that depends on direct observation of a behavior, it is important that it be tested for interrater reliability. The consistency or reliability of the observations between observers is extremely important. The above scores indicate a high degree of interrater reliability.

Although this questionnaire consists of 29 agitated behaviors, the present study utilized the short form version of the CMAI. This consists of 14 agitated behaviors, each rated on a 5–point frequency scale instead of a 7–point frequency scale. For the purpose of this study, the researcher adjusted the scale to read from 0–4 (‘0’ indicates that the person never exhibits the agitated behavior, and ‘4’ indicates that the person engages in the behavior a few times an hour or continuous for half an hour or more). This scale relies on subjective information given by the rater. For the short version of the CMAI, inter–rater reliability was as follows: exact agreement = .82; 0–1 point discrepancy = .93. These coefficients demonstrate an acceptable level of reliability, and assure the researcher that the CMAI is a reliable tool.

Procedure

The researcher interviewed 40 families. Each family member was asked to answer the 23 items from the FPCR relating to interactions between family and staff around the care of the person with dementia. In addition, the family members were asked how frequently they visited and also asked to comment on any other factors that might impact on their caregiving role. These comments were included as personal insights at the end of the interview.

The CMAI questionnaire was then completed by the nurse who best knew the patient. The nurse was asked to rate the frequency of occurrence of 14 behaviors of the person with dementia and also to comment on any other factors that might contribute to the behavior of the person with dementia. These comments were included as personal insights at the end of the questionnaire. The questions concerning these behaviors were then organized into two major categories or groups: verbal behavior and physical behavior. For analysis purposes, it is not useful to calculate a total score by adding all the categories.4 The rationale for using this combination of items relates to the behavior itself, and how the person with dementia is displaying the behavior (note that item #4 denotes both verbal and physical behavior).

Next, the frequency of visitation was established according to how the family pattern of visitation occurred. The distribution of the frequency of family visit was bimodal (see Figure 1). Given this distribution, two groups were formed with 19 residents receiving visits three or more times per week (Group 1) and 21 residents receiving visits twice per week or less often (Group 2). The rationale for selecting this breakdown for the groups was based on the fact that the group sizes needed to be fairly even and a bimodal distribution was representative of the family visitation pattern. The significance of these results will be discussed in the discussion section of the paper.

Figure 1. Frequency of distribution: bimodal distribution comparing 19 residents receiving visits three or more times per week (Group 1) and 21 residents receiving visits twice per week or less often (Group 2).

Results

The data analyses consisted of three major parts. First, a t–test was performed comparing Group 1 (3 or more visits per week) and Group 2 (visits twice per week or less often) on the CMAI scores to determine if there was any significant difference between the agitation levels of persons with dementia who receive more frequent family visits and the agitation levels of those persons who have less family visiting. Next, a distribution of scores on the FPCR was created to determine to what extent families feel they are partnering with staff. Finally, a correlation was undertaken between the FPCR and CMAI scores to determine if there was a relation between the family’s perceived role in partnering and the behavior of the person with dementia. The program Microsoft Excel 97 Data Analysis and the XY (Scatter Plot) Graphing Tools were used to analyze the data. The results will be presented for each hypothesis investigated.

A significance level of .05 was used for all statistical tests. This choice of standard for an acceptable probability is the risk one assumes of rejecting a null hypothesis when it is true. The higher the significance level used for testing a hypothesis, the higher the probability of rejecting a null hypothesis when it is true.

Hypothesis #1 states that persons with dementia who received more frequent family visits would experience less frequent episodes of agitation behaviors than those persons who received family visits less frequently. To investigate this issue, a t–test (One–Tailed Test for Difference Between Means) was performed. A one–tailed test was selected because the researcher expected that the family visitation would impact on the agitation levels of the person with dementia. This particular test was selected because it showed the difference between means and gave direction of the effect.

The frequency of responses to the CMAI observed for families with more or less visitation are presented in Appendix A.

Appendix A

Frequency of Responses to Cohen–Mansfield Agitation Inventory

Note: The frequency responses are shown for each item within the verbal and the physical subscales of the CMAI within each visitation group.

Each item within the verbal and physical subscale of the CMAI is shown for each visitation group. It is interesting to note that the percentage of zero scores was modestly higher for Group 2 than for Group 1. In other words, a slightly larger percentage of the people with dementia who had less frequent family visiting showed agitated behavior. The t–test comparing the groups on the scores for verbal behavior category of the CMAI did not detect a significant difference between the groups; t (1,2) = .78, p= .15. Similarly, no significant differences between the groups were detected by the t–test used to examine the scores on the physical category of behavior on the CMAI ; t (1,2) =– 1.02, p= .22.

For the verbal category of behavior, the upper limit of acceptance was 1.7 x 7.4 = 12.6 (see Figure 2). This Figure illustrates that the difference between the two sample means (1.6) lies within the acceptance region. In order to reject the null hypothesis, the observed difference of sample means would need to fall sufficiently high in the right tail of the distribution. Thus, the null hypothesis was accepted.

For the physical category of behavior, the upper limit of acceptance was 1.7 x 6.8 = 11.5 (see Figure 3). This Figure illustrates that the difference between the two sample means (–1.1) lies within the acceptance region. In order to reject the null hypothesis, the observed difference of sample means would need to fall sufficiently high in the right tail of the distribution. Thus, the null hypothesis was accepted.

Figure 2. One–tailed test for difference between means: for the verbal category of behavior on the Cohen–Mansfield Agitation Inventory, one can see that the difference between the two sample means lies within the acceptance region.

Figure 3. One–tailed test for difference between means: for the physical category of behavior on the Cohen–Mansfield Agitation Inventory, one can see that the difference between the two sample means lies within the acceptance region.

It appeared from the results that one must accept the null hypothesis. There was no significant difference between the agitation levels of persons with dementia who received more frequent family visits and agitation levels of those persons who received family visits less frequently.

Hypothesis #2 stated that family caregivers viewed themselves as sharing the caring role. That is, family caregivers perceived a partnering relationship with staff. In order to investigate this issue, a distribution of scores on the FPCR was created. This is illustrated in Figure 4.

The group scores were derived from how the respondent answered the item on the FPCR. It is important to note that if the individual responded 1 to 23 items, then 23 would be the lowest possible score (and also denotes strongly disagree on the scale). If the individual responded 7 to 23 items, then 161 would be the highest possible score (and denotes strongly agree). The largest number of responses (21) came from the group score of 47–69, which actually denoted an average response on the FPCR of 2 or 3. This was leaning towards the strongly disagree end of the scale. The mean for this distribution of scores was 69. The median was 65.5, which denoted the mid–point of the data set. The mode was 47. This represented the most frequent recurring total score illustrated on Figure 4, and was represented by the highest point in the distribution curve. This positively skewed distribution showed the researcher that the most responses came from family members who responded at the strongly disagree end of the scale. In this case the median (65.5) was the best measure of location, and showed the researcher that the family’s responses were leaning towards the strongly disagree end of the scale. In other words, family caregivers did not perceive a partnering relationship with staff.

Figure 4. Distribution of scores on Family Perceptions of Caregiving Role Tool: determines to what extent families feel they are partnering with staff. It is important to note that the largest number of responses are leaning towards the strongly disagree end of the scale.

92* indicates: a neutral score.

Hypothesis #3 stated that those family caregivers who felt they were in a partnering relationship with staff, were families of patients with less agitated behavior. This would mean that high scores on the FPCR (reflecting partnering) would be related to low scores on the CMAI (reflecting less agitation). That is, a significant negative correlation between the two measures was anticipated. Both the size of the correlation and the size of the sample affected the determination of significance; as either increased, the probability of significance also increased. To investigate this issue correlations between the FPCR score and each of the CMAI scores were calculated. Because the CMAI contained a diversified group of behaviors, the verbal and physical behaviors were first categorized.

Pearson product–moment correlation coefficients were calculated between the variables (FPCR score and each of the CMAI scores). This is identified in Figures 5 and 6. The dependent variable (CMAI score) is on the vertical axis (Y–axis) and the independent variable (FPCR score) is on the horizontal axis (X–axis). These scatter diagrams portray the relation between these two variables of interest. Note that if the correlation was weak, there was considerable scatter about a straight line drawn through the centre of the data. The correlation coefficient value for the CMAI verbal score was: r = 0.087128, and for the CMAI physical score was: r = 0.128065. These coefficients of correlation showed that there were weak positive correlations between the FPCR and each of the CMAI scores (i.e., verbal and physical). In other words, although the family perceived there was some partnering relationship, the person with dementia remained agitated.

It appeared from the results that one must accept the null hypothesis: there was no significant relation between the family’s perceived role in partnering and the behavior of the person with dementia, specifically the agitation levels.

Figure 5. Correlation between the Family Perception of Caregiving Role score and the Cohen–Mansfield Agitation Inventory verbal score. The correlation coefficient is: r=.087128. This denotes a weak positive correlation between these two variables.

Figure 6. Correlation between the Family Perception of Caregiving Role score and the Cohen–Mansfield Agitation Inventory physical score. The correlation coefficient is: r=0.128065. This denotes a weak positive correlation between these two variables.

Discussion

A limitation of this study centers on the instruments themselves. The FPCR was selected primarily because it measured the family’s perception of the caregiving role. The original FPCR scale contains 81 items. In the present study the researcher selected 23 items all dealing with interactions between family and staff. Due to this modification of the original version, it is difficult to measure the validity and reliability of the shortened version. However, in testing with data from current research ( N= 112 family members), Cronbach’s alphas were .90 for the total scale. Subscale composition is being re–evaluated, although alphas currently are greater than .80 for all three sub–scales. Test–retest reliability with 12 family members for a 3–week period was 0.79 (p <0.001).8 A good methodological approach for future studies might be to administer the FPCR to the same subjects under similar conditions on two or more occasions to support this attribute of stability.

Similarly the modification of the CMAI may present limitations to the conclusions drawn in this study. This questionnaire consists of 29 agitated behaviors; the present study utilized 14 agitated behaviors. Because the CMAI is an instrument that depends on direct observation of behavior, it is important that it be tested for inter–rater reliability. In testing the psychometric properties of the short version of the CMAI, inter–rater reliability was as follows: exact agreement = .82; 0–1 point discrepancy = .93.4 Inter–rater reliability was not assured in the present study. A suggestion for future studies might involve interviewing the family caregiver as well. The goal is to achieve the most accurate reflection of the frequency at which these agitated behaviors occurred.

Results from the analyses provide evidence relevant to the questions of interest in this study. This study examined agitated behaviors for the person with dementia who received family visits more frequently compared with those who received family visits less frequently. It also examined relations between the perceived partnering role of the family and the agitated behavior of the person with dementia. Although it was hypothesized that more frequent visitation and a higher perceived level of partnering would be related to lower agitation levels, these relations were not observed in this study. These findings will be discussed within the context of each hypothesis investigated in this study.

Relation Between Family Visitation and Behavior

The results do not support the relation between the frequency of family visitation and the agitation behaviors of those persons with dementia. Although previous research suggests a positive correlation between visiting and its relationship with level of agitation,9 this phenomenon was not evident in the present study. In Cohen–Mansfield’s study, 24 study participants were selected on the basis of their high agitation and cognitive impairment. This was not a criteria for the present study, only that the person had dementia. Also, in Cohen–Mansfield’s study a research assistant observed each resident over the course of 3 months. During each observation, a research assistant recorded the number of times each resident manifested each of the agitated behaviors. In the present study the nurse observed and recorded the behavior of the person with dementia one time only.

A myriad of other factors may have influenced the individual’s behavior. For example, verbally disruptive behaviors may be associated with discomfort such as pain. Others, such as requests for attention and aggressive behaviors, may be adaptive, and may signal an attempt to communicate needs.10 In a subsequent study, Cohen–Mansfield et al. (1990)11 examined the link between agitated behaviors and cognitive functioning in a study of 408 nursing home residents. Results showed that cognitively impaired residents manifested aggressive behaviors and physically nonaggressive behaviors.11 Several possible factors which may have impacted on the behavior in the present study include: 1) environmental influence on agitation, 2)time use related to manifestation of agitated behaviors, 3) past experience of the individual as a predictor of agitation, and 4) the relation between sleep and agitation.

The Family’s Partnering Relationship with Staff

Previous research has indicated that there may be role ambiguity between staff and family members caring for persons with dementia.5 Staff have been encouraged to view the resident’s family as client,15 as needed resource for resident care,21 and as a partner in care.16 However, Bowers (1988)7 and Duncan and Morgan (1994)6 agree that family members do not completely set aside their roles and responsibilities when their relative moves to a nursing home. In fact, families appear willing to share or assume responsibility for over half the tasks. In Bower’s study, it was suggested that biographical knowledge about the person with dementia is necessary, in order to understand what makes that person unique, including knowledge about personal preferences which might influence the person’s activities of daily living.

The present study revealed that families do not perceive a partnering relationship with staff. However, in examining the distribution of scores for each item on the FPCR, families do perceive there is a partnering relationship around some issues, and other issues where they do not feel a partnering relationship with the staff. The data reveal that families do perceive partnering around visitation, being asked to remain when care is provided, being included in priority setting for their relative, feeling comfortable making suggestions to the staff about their relative’s care, having a sense that they still are the primary caregiver for their relative, and feeling included in the care of their relative. On the other hand, family members feel they should be partnering and are not being included in decision–making, providing personal care, feeling a sense of control over the care their relative receives, offering input about what care will be provided for their relative and how, feeling a concern that their relative is not being cared for as an individual, nor feeling they can interact purposefully with staff about the care of their relative.

Family involvement is recommended as a goal in the holistic care of patients with dementia who reside in long–term care institutions.19 In order to recognize the personal histories, values and preferences of these residents, the staff needs to enhance the roles of the family in facilitating the individualization of care.20 Cutillo–Schmitter (1996)21 agree that incorporating family members in the care of patients with dementia provided pertinent psychosocial data, led to mutual decision–making regarding care, and produced changes in the responses of the residents with dementia, as well as in the family and nursing staff. The authors discuss the importance of fostering affectionate connections and competency among family members as they readjust to changing circumstances. For many families, fostering this connection may involve partnering with the nursing staff in the setting of the nursing home.

The discrepant finding for Hypothesis #2 may result from several factors. Firstly, transition to the nursing home was an extremely difficult time for family members. These family caregivers demonstrated uncertainty about how to perform their changing roles. Secondly, in order to successfully implement a partnership between family and staff caregivers, there needed to be a more consistent staffing pattern. Family members voiced concerns to me that almost each time they visited there seemed to be a different nurse working. Finally, it was apparent during the course of this study that the family members were engaging with the nursing assistants when they had information to share regarding the condition of their loved one. It was unclear if this information was being communicated to the registered nurse. Family members felt that although the nursing assistants were indeed listening to them, they did not see this information being integrated into the care plan of their loved one.

Relation Between Perceived Role of Family and Behavior

The results show that there is no significant relation between the family’s perceived role in partnering and the behavior of the person with dementia, specifically the agitation levels. However, the literature suggests otherwise. Buckwalter and Hall (1987)21 maintain that families are a neglected resource for providing quality care to their relatives with Alzheimer disease. They suggest that family and staff should form a partnership with clearly delineated cooperative role behaviors designed to maximize the cognitive and functional abilities of the patients, while enhancing satisfaction and reducing stress of families. This relationship was not evident in the present study. It was expected that high scores on the FPCR (reflecting partnering) should be related to low scores on the CMAI (reflecting agitation). In fact there was a weak positive correlation between the family’s perception of partnering and the agitation levels of the person with dementia. This means that although the family is partnering with staff, the person with dementia remains agitated

In addition to the influences on behavior previously discussed by Cohen–Mansfield et al. (1992)9, interventions by family and staff can also have a profound effect on behavior. 21 Grant’s study (1996)20, supports the need to acknowledge the importance of families in recognizing individual functional, cognitive, and emotional changes in the individual, and the importance of family and staff sharing this information. Droes, Breebaart, Ettema, van Tilburg, and Mellenbergh (1999)22 concur that integrated family support, in which patients and caregivers are both supported by the professional staff, is more effective in influencing behavior problems and mood of the dementia patient than nonintegrated support

In the present study, when family members were asked to comment on any personal insights, they stated that they did not always feel acknowledged by the nursing staff, nor did they feel that staff would share important information with them. This is supported by the empirical data. When the families were asked the items on the FPCR, such as being included in the decisions about the care of their relative, having a secondary role since their family member was admitted to the nursing home, feeling control over the care their relative receives, and agreement on care priorities for their relative, the results reflected a non–partnering role.

References

1. Tickle, E. H., & Hull, K.V. (1995). Family members’ roles in long–term care. Medsurg Nursing, 4, 300–304.

2. Miesen, B. (1992). Attachment theory and dementia. In G.M.M. Jones and B.M.L. Miesen (Eds.), Caregiving in dementia. (pp. 38–56). London: Routledge.

3. Cohen–Mansfield, J., Marx, M.S., & Werner, P. (1992). Agitation in elderly persons: An integrative report of findings in a nursing home. International Psychogeriatrics, 4, 221–240.

4. Cohen–Mansfield, J., Marx, M.S., & Rosenthal, A.S. (1989). A description of agitation in a nursing home. Journal of Gerontology: Medical Sciences, 44, M77–84.

5. Schwartz, A.N., & Vogel, M.E. (1990). Nursing home staff and residents’ families role expectations. The Gerontologist, 30, 49–53.

6. Duncan, M.T., & Morgan, D.L. (1994). Sharing the caring: Family caregivers’ views of their relationships with nursing home staff. The Gerontologist, 34, 235–244.

7. Bowers, B.J. (1988). Family perceptions of care in a nursing home. Gerontologist, 28, 361–368.

8. Buckwalter, K.C., Maas, M., & Reed, D. (1997). Assessing family and staff caregiver outcomes in Alzheimer disease research. Alzheimer Disease and Associated Disorders, 11, 105–116.

9. Cohen–Mansfield, J. Marx, M.S., & Werner, P. (1992). Observational data on time use and behavior problems in the nursing home. The Journal of Applied Gerontology, 11, 111–121.

10. Cohen–Mansfield, J., & Werner, P. (1995). Environmental influences on agitation: An integrative summary of an observational study. The American Journal of Alzheimer’s Care and Related Disorders and Research, 1, 32–39.

11. Cohen–Mansfield, J., Marx, M.S., & Rosenthal, A.S. (1990). Dementia and agitation in nursing home residents: How are they related? Psychology and Aging, 5, 3–8.

12. Cohen–Mansfield, J., & Marx, M.S. (1989). Do past experiences predict agitation in nursing home residents? International Journal of Aging and Human Development, 28, 285–294.

13. Beck, C.K., & Vogelpohl, T.S. (1999). Problematic vocalizations in institutionalized individuals with dementia. Journal of Gerontological Nursing, 25, 17–26.

14. Cohen–Mansfield, J., Werner, P., & Freedman, L. (1995). Sleep and agitation in agitated nursing home residents: An observational study. Sleep, 18, 674–680.

15. Montgomery, R. (1982). Impact of institutional care policies on family integration. The Gerontologist, 22, 54–58.

16. Spencer, B. (1991). Partners in care: The role of families in dementia care units. In D.H. Coons (Eds.), Specialized dementia care units (pp. 189–204). Location: Baltimore, MD: John Hopkins University Press.

17. Maas, M., Buckwalter, K.C., Swanson, E., Specht, J., Tripp–Reimer, T., & Hardy, M.A. (1994). The caring partnership: Staff and families of persons institutionalized with Alzheimer disease. The American Journal of Alzheimer Care and Related Disorders and Research, 12, 21–30.

18. Travis, S. (1996). Formal long–term care networks: Forming a partnership with gerontological nurses. Journal of Gerontological Nursing, 22 (12), 21–24.

19. Tickle, E. H., & Hull, K.V. (1995). Family members’ roles in long–term care. Medsurg Nursing, 4, 300–304.

20. Rowles, G.D., & High, D.M. (1996). Individualizing care: Family roles in nursing home decision–making. Journal of Gerontological Nursing, 22, 20–25.

21. Hadley, C., Brown, S., & Smith, A. (1999). Evaluating interventions for people with severe dementia: Using the Positive Response Schedule. Aging and Mental Health, 5, 234–240.

22. Droes, R.M., Breebaart, E., Ettema, T.P., van Tilburg, & Mellenbergh, G.J. (1999). Effect of integrated family support versus day care only on behavior and mood of patients with dementia. International Psychogeriatrics, 12, 99–115.

B O O K

C A R E W H E E L