| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Cynthia Lindman Port, PhD, Department of Epidemiology and Preventive Medicine, University of Maryland School of Medicine, 660 West Redwood Street, Baltimore, MD 21201. E-mail: cport{at}epi.umaryland.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Nursing home residents • Cognitive impairment • Family involvement
The most commonly used approach for measuring family caregiver involvement in the NH is visit frequency (Port et al., 2003), which has been shown to relate positively to the provision of technical and nontechnical help by families (Dempsey & Pruchno, 1993). Previous research identifies characteristics of NH residents and family caregivers that relate to how frequently the family caregiver visits the resident. First-degree kin and those living closer to the facility visit more frequently, and residents utilizing Medicaid receive less contact (Bitzan & Kruzich, 1990; Greene & Monahan, 1982; Port et al., 2001; Yamamoto-Mitani et al., 2002). Evidence also indicates that involvement declines with increasing duration of stay (Bitzan & Kruzich, 1990; Yamamoto-Mitani et al., 2002). Finally, the resident's cognitive functioning may be important. Whereas one study found that residents with dementia received fewer contacts from family compared with residents without dementia (Port et al., 2001), a second study found no relationship between caregiver activity level and resident cognitive functioning (Dempsey & Pruchno, 1993). Unfortunately, characteristics such as proximity, kinship, payment source, duration of stay, and cognitive functioning are difficult to change, and thus they are likely to be poor candidates as targets for interventions to enhance family involvement.
Several potential barriers to family involvement that may be more promising for intervention have received little or no empirical investigation. Depressed mood and agitation in the resident may strain the caregiver–resident relationship and reduce visiting. The caregiver's physical or psychiatric health may interfere with involvement, whereas an available support network sharing the care burden may facilitate it. Even practical considerations, such as transportation, may determine whether a family caregiver is able to visit. Finally, having a strained relationship with staff may make visits distressing. Interventions such as education about coping with the dementing process, assisting caregiver's with health-based limitations, identifying and enlisting the assistance of the caregiver's larger social support network, solving transportation issues, and improving relationships with staff are possible ways of ameliorating these barriers. A handful of promising interventions have been proposed or conducted, although most were pilot or demonstration projects and none examined family involvement as an outcome (Anderson, Hobson, Steiner, & Rodel, 1992; Drysdale, Nelson, & Wineman, 1993; Hansen, Patterson, & Wilson, 1988; Kelley, Specht, & Maas, 2000; McCallion et al., 1999; Pillemer, Hegeman, Albright, & Henderson, 1998; Sancier, 1984). Prior to further evaluation and design of interventions, research is needed to identify which, if any, of these more changeable barriers relates to family involvement for today's caregiving families.
In the current study I examine barriers to family involvement in the care of NH residents with cognitive impairment, focusing on barriers with greater potential for improvement through intervention. The study asked whether variables with more potential for intervention are related to involvement after previously identified variables including distance, kinship, payment source, length of stay, and cognitive function are taken into account. I hypothesized that less frequent visitation would be related to a caregiver's having greater transportation and health problems, a poorer relationship with staff, a smaller network of supportive family and friends, lower visit enjoyment, and visiting a resident with mood or behavior problems. The study also presents family caregivers' responses to two open-ended questions. To provide an alternative approach to understanding barriers faced by NH family caregivers, the first question asked respondents to state the main thing that made it more difficult or unpleasant to be involved in care. The second question sought caregiver's suggestions for enabling them to be more involved in care.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
NH residents and informal caregivers (family member, friend, or guardian) were enrolled in pairs. Facilities provided the names of all residents 60 years or older with a minimum facility stay of 1 month and some degree of cognitive impairment. In most cases, the primary contact person from the resident's chart served as the informal caregiver. Before being enrolled, caregivers were asked if they were the person "most involved in the resident's care." A small proportion (less than 5%) indicated a second individual was most involved, in which case the second individual was enrolled. Of 241 informal caregiver–resident pairs approached for consent, 46% consented, 24% declined, and 30% did not respond or could not be contacted. Because some of the consenting residents died, a final participation rate of 42% was obtained, for a total of 100 resident–informal caregiver pairs. State-appointed guardians (7%) were removed from the current analyses for a total sample of 93 family caregivers (family members or friends). The Institutional Review Boards of the University of Maryland and all participating facilities approved the research protocol, and written consent was obtained for all resident and caregiver participants.
Variables
Family Caregiver
Family caregivers completed 1-hr telephone interviews with a trained interviewer. Demographics included caregiver age, gender, race, years of education, familial relationship, work status (retired vs. working) and number of dependents (not including the resident). Interviewers measured visit frequency by asking caregivers how frequently they visit the resident in person. Answers were given in terms of times per day, week, month, or year, whichever was appropriate, and were converted to yearly rates for analyses.
Barriers were assessed as follows. Interviewers obtained proximity by asking, "Approximately how many miles away do you live from [facility name]?" They rated transportation difficulty on a 5-point scale (1 = extremely difficult to 5 = very easy); respondents were asked, "In terms of transportation, how difficult is it for you to visit your loved one in the nursing home?" Interviewers measured caregiver physical health by using the 5-item general health scale from the 36-item Short-Form Health Survey (SF-36; Ware & Sherbourne, 1992). The scale ranges from 0 to 100; higher scores represent better health. Reliability for the current sample was 
=.76. Interviewers measured mental health by using the General Health Questionnaire (GHQ-28; Goldberg & Hillier, 1979), a widely used and validated 28-item measure that evaluates symptoms across four scales: depression, anxiety, social impairment, and hypochondriasis. The 2-point (present or absent) scoring method was used, such that each scale ranges from 0 to 14, with higher scores representing worse symptoms. Because reliability for the social impairment scale was low ( =.15), I omitted it from further analyses. Reliabilities for the remaining scales ranged from =.68 to =.71. For caregiving support network, caregivers indicated the total number of family members and friends involved in visiting the resident, not including staff or others visiting for professional reasons. Interviewers queried the quality of the relationship with staff by means of a 5-point scale (1 = very poor to 5 = very good) and the question, "How would you rate the quality of your relationship with the staff at the nursing home?" Finally, they rated caregivers' enjoyment of visits by using a 5-point scale (1 = very upsetting to 5 = very enjoyable) and the question, "In general, how do you find your visits to [name of resident]?"
Two open-ended items were also included in the interview. First, interviewers asked caregivers to state the "main thing that makes it difficult or unpleasant for you to be involved in care." I created categories on the basis of the content of the responses and barriers noted in the literature (Friedemann et al., 1997; Ross, Carswell & Dalziel, 2001; Monahan, 1995; Whitlatch, Schur, Noelker, Ejaz, & Looman, 2001). Two research assistants then independently coded responses into these categories. Coding disagreements occurred in less than 10% of responses and were discussed until 100% agreement was reached. Second, interviewers asked caregivers, "If someone wanted to help you to be able to be more involved in the care of [resident name], what would be some things they could do to help you?" Caregivers could provide multiple answers to this question. Results are presented descriptive only utilizing the same categories as in the first open-ended question.
Resident
I obtained resident age, gender, race, duration of stay, Medicaid status, cognitive functional level, and depression and behavior ratings from the Minimum Data Set (MDS; Morris et al., 1990). Residents funded in any part by Medicaid received a 1, whereas all others received a 0. For cognitive impairment, a Mini-Mental Status Exam (MMSE; Folstein, Folstein, & McHugh, 1975) was administered by a trained research assistant to 78 (84%) of the residents and estimated from the residents' most recent MDS–Cognition Scale scores (Hartmaier, Sloane, Guess, & Koch, 1994) for the remainder (Gruber-Baldini, Zimmerman, Mortimore, & Magaziner, 2000). Depression was measured by means of the MDS-based Depression Rating Scale (Burrows, Morris, Simon, Hirdes, & Phillips, 2000); behavior problems were a sum of the five MDS behavioral symptom items. Scale reliability was =.64 and =.88 for depression and behavior problems, respectively. Comorbidity was the number of conditions a resident had from a list of 12 diagnoses determined by means of chart review by an RN-level nurse researcher (see Table 1 for diagnoses).
|
Regression models explored the relative contributions of resident and family caregiver characteristics to visit frequency. Because of missing data, I imputed means for two respondents for the SF-36 General Health scale and three respondents for the Anxiety and Depression scales of the GHQ. Because an analysis of variance indicated that one facility (Site 2) had significantly higher involvement, I entered the effect of this facility in the first step. In the second step, I entered variables previously shown to relate to visit frequency: distance, kinship, payment source, length of stay, and cognitive function. I dichotomized kinship into immediate kin (spouses, sons, and daughters) versus extended kin or friends. In the final step, I added the caregiver variables of transportation, relationship with staff, social support network, visit quality, general health, depression, and anxiety and the resident variables of mood and behavior problems.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Sample Characteristics
Sample characteristics are presented in Table 1. Caregivers were nearly 60 years of age on average and mostly female and White. Of non-Whites, one was Hispanic, and the rest were Black. Most had a high school education or more and worked full or part time. The largest kinship group was daughters, followed by sons and spouses and small percentages of other relatives and friends. The resident was the sole dependent of the caregiver for a minority of the sample. Caregivers visited 2.8 times a week on average, with a median frequency of twice a week. Mean transportation difficulty was close to "easy," and caregivers lived a mean of 39 miles (median 6 miles) from the facility. Results for general health are comparable with a large validation study of the scale, which found a mean of 57.0 (SD = 21.1; Rand Health Sciences Program, 1992). In this nonclinical sample, depression, anxiety–insomnia, and somatic symptoms were low. On average, about 6 persons were involved in visiting residents. The average quality of the relationship to staff approached "good," and the mean visit quality was average ("neither enjoyable nor upsetting").
Residents averaged 83.1 years and were mostly female. Racial data were identical to the caregivers and are not presented. Residents averaged 2.6 comorbidities and had been in the facility for 33.7 months (median 24 months). The majority (70%) utilized Medicaid. The mean MMSE score (9.34) was somewhat lower than that found for a large random sample of NH residents (12.7; Hartmaier et al., 1995). Mood and behavior problems were not common. Resident age and comorbidity are similar to a large stratified random sample of NH residents in Maryland (Magaziner et al., 2000). The proportion of non-White residents (37.6%) was higher than the state sample (17.0%), reflecting the high proportion of non-Whites (67.4%) in Baltimore.
Predictors of Visit Frequency
Table 2 presents the stepwise multiple regression analysis. In the first step, facility type is regressed onto visit frequency. The previously established variables relating to visit frequency are entered simultaneously in the second step. The remaining caregiver and resident variables are entered simultaneously in the third step. Most of the previously identified variables are again related to visitation; however, Medicaid status only approached significance (p <.08) and higher cognitive function was related to lower visit frequency. Three of the caregiver variables with greater potential for change through intervention were predictive of visit frequency in the hypothesized direction. Caregivers who visited less frequently also reported more transportation problems, a poorer relationship with staff, and a smaller support network. Contrary to hypotheses, caregivers reporting greater anxiety had a higher rate of visiting, and neither caregiver physical health, depression, and visit quality nor resident depression and behavior problems predicted visit frequency. In the complete model, transportation problems had the largest contribution to predicting lower visit frequency, followed by duration in the facility and kinship status. The remaining significant variables, including resident cognition, caregiver anxiety, visit quality, and the relationship with staff accounted for similar amounts of the variance in visit frequency. The model accounted for 60% of the total variance.
|
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
The variable most predictive of visit frequency was transportation difficulty. As in previous research, distance to the facility was negatively related to visit frequency. In the final regression model, however, transportation problems remained significant whereas distance did not. Given that the median distance to the facility in this and other studies was less than 10 miles (Greene & Monahan, 1982; Montgomery, 1982; Port et al., 2001), framing the issue in terms of transportation rather than distance may more accurately capture the problem. Indeed, of those reporting transportation difficulties, more than half lived 10 miles or fewer from the facility. Responses to the open-ended question also raise the issue of transportation problems. Nursing facilities and researchers alike have largely overlooked ride systems or other transportation solutions for family caregivers. A study of services offered to the families of NH residents in Michigan found that a mere 1% encouraged alternative types of transportation to the NH, 21% had options available upon request, and the remaining 78% had no available options (Friedemann et al., 1997). A search of the literature found no interventions that included facilitating transportation for families.
An issue that has garnered more research attention is the relationship between staff and families. Numerous studies support the need for improved relations and lament the lack of a truly collaborative approach to care (Bowers, 1988; Friedemann et al., 1997; Laitinen & Isola, 1996; and see Janzen, 2001 for a review). Other research has found that negative interactions with staff have a negative influence on resident and caregiver outcomes (Whitlatch et al., 2001). These results go a step further to demonstrate that caregivers who report problematic relationships with staff also visit less frequently. The study design precludes definitive statements about the direction of this link; however, it seems reasonable to suggest that programs to improve relations might also improve involvement levels.
In open-ended questioning, those reporting problems with staff focused primarily on communication issues, such as getting better information and being listened to. Two promising interventions are available that center on improving communication to reduce sometimes adversarial relations between formal and informal caregivers. Partners in Caregiving (Pillemer et al., 1998) trains families and staff in communication techniques and conflict resolution skills while offering opportunities to influence facility practices. Participants felt the program improved their communication skills, but the effect on family involvement was not evaluated. Family Involvement in Care (FIC; Kelley et al., 2000) orients informal carers to the care environment, provides education about dementia and skills for negotiating partnerships with staff, negotiates a written agreement with staff regarding goals for care, and follows up to ensure the agreement is working well.
Having more family and friends who take part in visiting was also related to higher visit frequency. The importance of social support to caregivers is well documented, and decreases in support heighten the burden and isolation of community-dwelling (Zarit, Reever, & Bach-Peterson, 1980) and NH-based family caregivers (Harper & Lund, 1990; Kelley, Swanson, Maas, & Tripp-Reimer, 1999). Rates of visits and phone calls by family and friends to NH residents decline by nearly half following NH entry (Port et al., 2001), and this decrease probably manifests itself also in decreased support for the primary family caregiver. In both this and previous work, caregivers indicated that the difficulty of getting other family members to visit was a primary problem (Townsend, Deimling, & Noelker, 1988).
Family and friend support may be lost early in the institutionalization process because of characteristics of the NH setting or because of anxiety over how to be helpful during this stage of care. For NH social workers, making efforts to meet with family early in the institutionalization process (Greene, 1982) and with as many members of a resident's social network as possible may pay off in better support for residents and primary caregivers over time. Existing programs encouraging family participation include incorporating families into care planning (Anderson et al., 1992) and facilitating volunteer opportunities to provide activities, entertainment, and other services (Hansen et al., 1988). Facilities incorporating these types of programs send a clear message to the families of residents that their efforts are welcomed and valued. Then too, the problems with cleanliness and persistent odors noted by these family caregivers not only decrease the comfort of regular visitors but discourage less regular visitors as well.
The open-ended questioning highlighted difficult aspects of caregiving not captured by the quantitative measures. The most frequently cited problem was dealing with the declining status of the resident, with caregivers' comments centering on grief and loss over the changing relationship with their loved one. However, caregivers do not necessarily decrease visiting in the face of these difficulties. This is supported by the findings that neither the resident's cognitive functional level or mood and behavior problems, nor the quality of the visit was related to lower visiting and that caregiver anxiety related positively to visiting. Caregivers experiencing anxiety over the resident's decline and the responsibility for monitoring care closely are likely also those with the strongest emotional attachment to the resident. The anxiety scale may therefore have been capturing caregiver burden, with the most heavily involved caregivers also experiencing the highest burden (Monahan, 1995).
Interventions aimed at the psychoeducational needs of family caregivers may help to alleviate anxiety and burden (Monahan, 1995). A brief but comprehensive group is described by Drysdale and colleagues (1993) that offers information about aging, visiting, and resident and family adjustment to the facility, and that provides opportunities to ventilate feelings of guilt and grief. A similar format is presented by Sancier (1984). McCallion and colleagues (1999) focus on improving the experience of visiting and interacting with a NH resident with dementia in their Family Visit Education Program (FVEP). The FVEP involves observing families during visits and then providing training sessions about dementia and suggestions for promoting positive interaction. The FVEP reduced resident's problem behaviors and symptoms of depression and irritability and improved the way families communicated with residents (McCallion et al., 1999). The interest in psychoeducational opportunities among NH family caregivers is probably not being sufficiently met (Monahan, 1995). A study of Michigan NHs found that approximately half encouraged families to take part in family councils and support groups and one quarter encouraged participation in educational classes (Friedemann et al., 1997). Within Maryland, only 68% of NHs even offer a family council (Nursing Home Compare, 2003).
Though quantitative assessment of caregiver health was not related to involvement, results of open-ended questioning suggest that health can be an important barrier to involvement. Perhaps a measure of health-related functional ability would have better captured the types of health barriers some family caregivers face. Several interventions already discussed might help in lessening the negative impact of health on involvement, including transportation assistance and mental health services. Societal changes to improve workplace and childcare issues may be desirable, but there is likely little that can be done in terms of practical interventions to simplify the lives of these busy caregivers, most of whom are working and care for additional dependents.
Finally, the relationship between lower resident cognitive functioning and higher visit frequency merits discussion. As described previously, the results of earlier studies are somewhat contradictory in this regard (Dempsey & Pruchno, 1993; Port et al., 2001). In this study, the zero-order correlation between MMSE score and visit frequency was low (r = –.13), but the relationship became significant in the regression model. Although more research is needed in this area, one possibility is that caregivers increase involvement for residents who are less capable of advocating for their own care needs.
The family caregiver sample in this study was typical in terms of caregiver age, gender, education, kinship status, work status (Dempsey & Pruchno, 1993; Duncan & Morgan, 1994; Farber, Brod, & Feinbloom, 1991; Friedemann, Montgomery, Rice, & Farrell, 1999), and visit frequency (Bitzan & Kruzich, 1990; Greene & Monahan, 1988; Stephens, Kinney, & Ogrocki, 1991). Residents were typical in terms of length of stay, age, and Medicaid funding (Bitzan & Kruzich, 1990; Dempsey & Pruchno, 1993; Ross, Carswell, & Dalziel, 2001; Tornatore & Grant, 2002; Whitlatch et al., 2001). Although MMSE scores ranged from 0 to 26, caution should be used when these results are applied to caregivers of cognitively intact NH residents, as the barriers and difficulties they face may differ.
Study generalizability is limited also by the sample and participation rate. Future research should include a representative sample of facilities of varying size, quality, and rural versus urban setting. The participation rate of 42%, although somewhat low, is comparable with other studies of informal caregivers to NHs (Bowers, 1988; Karner et al., 1998; Ryan & Scullion, 2000). Furthermore, although a wide range of involvement levels were obtained (0–730 visits per year) and no differences were found in resident demographics between participants and nonparticipants, it is possible that less involved caregivers were also less likely to participate, leading to inflated involvement levels. The study's correlational design precludes causal interpretations. Controlled intervention studies are needed to determine whether easing transportation difficulties, improving family–staff relations, and increasing support to family caregivers will facilitate family visitation in the NH. Finally, responses to the open-ended questions may have been biased by preceding interview questions. The finding that most of the responses to the first open-ended question cited an issue not directly assessed quantitatively is reassuring in this regard; however, a more rigorous application of qualitative methods is needed in the area of barriers to family involvement in long-term-care settings.
Despite theses limitations, the finding that several more changeable characteristics of caregivers may be as or more important than those previously thought to relate to involvement provides some empirical basis for existing intervention programs and suggests new approaches as well. Caregiver grief and the difficulty of communicating with residents and staff are fairly universal concerns when caregivers are institutionalizing a loved one, and these concerns may be best handled through orientation–psychoeducational groups for all interested families early in the transition to institutionalized care. Discussing basic questions such as transportation and enlisting assistance from additional friends and family may also be critical. Early attention to problems may also prevent later problems. For example, caregivers who have an outlet for their grief or who learn effective communication skills for NH settings may make fewer unreasonable demands of staff and minimize the development of adversarial relationships. Similarly, family caregivers who develop new support networks with other NH family caregivers may be less affected when their own family and friend networks do not provide sufficient support. In the end, a comprehensive approach for understanding and addressing the individualized barriers that family caregivers face will probably prove the most effective method for encouraging greater family involvement in long-term care.
|
Footnotes |
|---|
1 Department of Epidemiology and Preventive Medicine, University of Maryland School of Medicine, Baltimore.
Decision Editor: Linda S. Noelker, PhD
Received for publication December 17, 2003. Accepted for publication March 29, 2004.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Robison, L. Curry, C. Gruman, M. Porter, C. R. Henderson Jr., and K. Pillemer Partners in Caregiving in a Special Care Environment: Cooperative Communication Between Staff and Families on Dementia Units Gerontologist, August 1, 2007; 47(4): 504 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Port Informal caregiver involvement and illness detection among cognitively impaired nursing home residents. J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2006; 61(9): 970 - 974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Port, S. Zimmerman, C. S. Williams, D. Dobbs, J. S. Preisser, and S. W. Williams Families Filling the Gap: Comparing Family Involvement for Assisted Living and Nursing Home Residents With Dementia Gerontologist, October 1, 2005; 45(suppl_1): 87 - 95. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
