| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Mark Sciegaj, Rose Mary B. Fuss Center for Aging and Intergenerational Studies, Lasell College, 1844 Commonwealth Ave., Newton, MA 02466. E-mail: msciegaj{at}Lasell.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Consumer-directed care • Long-term care • Decision making
Younger disabled adults have long had more opportunities to exercise control over their care through the use of personal assistant services (Wallack, Sciegaj, & Long, 2002). Through these programs, disabled persons or their surrogates can hire, train, supervise, direct, and fire personal care attendants on their own (Benjamin & Matthias, 2001; Doty, Kasper, & Litvak, 1996). Traditionally, elders and their informal caregivers were not empowered to direct their own care, primarily because service providers and policy makers made assumptions about their level of interest and ability to exercise control. Over the last decade, research has questioned those assumptions, hypothesizing that in order for community long-term-care services to be effective for older adults, they have to match consumers' preferences, rather than agency service arrangements (Morris, Caro, & Hansen, 1998).
Even though consumer-directed care models are being advocated for use among elder populations, there are few data on the extent of elder interest in participating in the management of community long-term-care services, who they want involved in making these decisions, or their perceptions regarding the relative importance of different service choices. In addition, little is known whether elder preferences for consumer direction vary by race/ethnicity.
An early study by Cohen (1992) surveyed 57 elders regarding the importance of controlling various details of their care. Cohen's findings indicate the areas of service-delivery control considered important to younger adults with disability were not considered important to frail elders. For example, having control over the hiring, firing, and paying of caregivers was not considered an important element to the elders' sense of independence. What was important to the elders interviewed by Cohen was being able to participate and negotiate their care-services schedule. Cohen's study remains important because understanding the extent of elder preferences for control over different aspects of their community long-term-care services will ensure that elders or their families are not taking responsibility for choices that they do not really wish to make (Simon-Rusinowitz & Hofland, 1993).
In addition, the few published empirical studies reporting on the possible effects of race/ethnicity on preferences for control over the management and delivery of community long-term-care services suggest there may be differences based on race/ethnicity. In a survey of 491 young adult and elderly consumers, Simon-Rusinowitz, Mahoney, Desmond, Shoop, Squillace, and Fay (1997), assessing preferences for a cash-benefit option versus traditional agency-based services, found that Whites were half as likely (odds ratio =.49) to be interested in a cash benefit as African Americans (p <.01). Simon-Rusinowitz and Mahoney (2001) report that in a telephone survey (N = 2,140) also assessing preferences for a cash-benefit option versus traditional agency-based services, African Americans were 1.9 times as likely and Latinos were 1.6 times as likely to be interested in a cash-benefit option.
In addition, the findings of a study conducted by Blackhall, Murphy, Frank, Michel, and Azen (1995) that explored the effect of race/ethnicity on attitudes toward hospital-patient autonomy provide some insight into the importance of this area for understanding elder preferences for consumer direction in community long-term care. Blackhall and colleagues conducted interviews with 800 elders evenly divided among African Americans, European Americans, Korean Americans, and Mexican Americans. One topic area covered in these interviews regarded the issue of who should make the decision regarding life-prolonging technology (the elder or the family). Blackhall and colleagues report significant racial/ethnic differences, with Korean American and Mexican American elders favoring a more family-centered model of decision making rather than the more traditional patient-autonomy model. They conclude that physicians in these situations should ask the elder whether they prefer to make decisions independently or whether they prefer that their family be involved in this process.
Because of the growing heterogeneity of the elderly population, efforts to develop meaningful consumer-directed options need more knowledge on both the extent of elder preferences to assume control and whether there are possible racial/ethnic variations. The purpose of this study was to begin to explore these areas. The study surveyed 731 elders (200 African American, 200 Chinese, 131 Latino, and 200 White Western European) and examined whether their preferences for consumer-directed care differed. These four groups were selected based on the Blackhall and colleagues (1995) report of racial/ethnic group differences in medical decision making. Extending the findings of Blackhall and colleagues to community-based long-term-care services for elders, we hypothesized that White and African American elders would be more likely to choose consumer-directed models that increased individual decision making.
To test for potential differences, respondents were asked to select between three different models (Cash and Counseling Model, Negotiated Care Management or Social Health Maintenance Organization [SHMO]-Like Model, and Traditional Case-Management Model) and to rank them in terms of which one they liked the best overall. The three models were selected because they offer different levels of consumer control. The following is a brief description of each.
The Cash and Counseling Model represents one of the most unfettered forms of consumer direction, offering consumers cash allowances in lieu of agency-delivered services. Individuals in this model could use the cash to purchase services to meet their own needs from independent providers, including friends and family if so desired. Also available in this model is a counselor who provides the individual information, training, and advice to assist with planning and managing community long-term-care services. In addition, the consumer has the option of having a fiscal intermediary agent who attends to paying the long-term-care worker and performs other duties associated with being an employer.
The Negotiated Care Management or SHMO-Like Model gives the consumer a heightened level of control within an agency service-delivery framework. The mechanism for consumer direction in this model is an individual spending budget for services. How the budget is allocated is negotiated between an agency/care manager and the consumer. In this model, the care manager becomes a consultant to and resource for the consumer, helping him or her to make viable caregiving arrangements (Leutz, Capitman, MacAdam, & Abrahams, 1992). The meaningful choices of consumers are supported in this program not only by allowing them to express service prefer-ences but also by negotiating their levels of service in response to a known benefit limit (Leutz et al., 1992).
The Traditional Case-Management Model has been seen as the least consumer directed. Although in this model, the consumer is consulted regarding the type and timing of their services, the agency representative has final decision-making power over what services and schedule the consumer receives. The agency also takes responsibility for choosing, finding, and purchasing the services for the consumer.
Figure 1 illustrates the analytic framework used to examine elder preferences for consumer direction using the three management alternatives described above. Because the research in the area of elder preferences for consumer direction is not extensive, the study used this framework to pursue the following general study questions: What individual characteristics might determine an elder's preference regarding consumer direction in community long-term care? Are there significant racial/ethnic differences in elder preferences for consumer direction? Finally, are there significant interactions between race/ethnicity and individual characteristics on elder preferences for consumer direction?
|
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Procedure
The survey was administered through face-to-face interviews. Interviews lasted between 30 and 40 min. Master's-level social work and public health students were recruited and trained to administer the survey to the Chinese, African American, and White elder populations. Latino elders were recruited and trained to conduct interviews in that population. Because English was not the first language for either the Chinese or the Latino elder population, the interviews were conducted in the elders' native language with the appropriate dialect. Upon completion of the interview, participants received a small financial gift of $20.
Measures
The survey instrument developed for this study collected descriptive demographic data as well as health and functional status data. Questions were drawn from existing instruments or developed specifically for this study. Original questions were designed to measure elder preferences for consumer direction in community long-term care in the areas of service-delivery management, informal caregiver roles in long-term-care decisions, and consumer information needs. The participating community-based organizations translated the survey instrument into the appropriate dialects for their Latino and Chinese populations. Individuals not connected to the community-based organization then translated the survey back into English. In addition, small groups of elders from each of the four racial/ethnic groups were used to pilot test the survey instrument and to resolve any potential problems with the overall survey approach.
Dependent Variable (Elder Preferences for Consumer Direction)
This measure was defined as the level of control elders want in the organization, planning, and management of their community long-term care. Respondents were read three brief descriptions of new approaches to managing the delivery of their services (Table 1) and asked to identify the approach they liked the best overall. New Approach 1 described the Cash and Counseling Model, New Approach 2 described the Negotiated Care/SHMO-Like Model, and New Approach 3 described the Traditional Case-Management Model.
|
|
Health and Functional Status
Self-assessed health status was measured using the global health-assessment question and the Health-Perception Scale of the 36-Item Short Form (SF-36) (Ware, Snow, Kosinski, & Gandek, 1993). The global health question asks the respondents to assess their health from five possible choices ranging from excellent to poor. For analytic purposes, health status was also dichotomized (1 = excellent/very good/good, 0 = fair/poor). Functional status was measured using the SF-36 Physical Functioning Scale, which asks how limited a person is in 10 daily activities such as running, walking, lifting, carrying, climbing, bathing, and dressing. Depending on an individual's limitations, the scale ranges from 10 (very limited) to 30 (not limited). The psychometric properties of the scales making up the SF-36 are well documented, and the Cronbach's 
( =.87) found in this study is consistent with the reported literature (McHorney, Ware, & Raczek, 1993; McHorney, Ware, Rogers, Raczek, & Lu, 1992; Ware, Kosinski, Bayliss, McHorney, Rogers, & Raczek, 1995).
Service Satisfaction
Two measures of service satisfaction were originally created. First, a four-item scale was developed to measure the respondent's overall satisfaction with current community long-term-care services (Cronbach's =.82). Items included questions regarding satisfaction with amount, type, and timing of services and respondent's ability to make changes to service arrangements. Second, a three-item scale was created to measure elder satisfaction with their current control over current community long-term-care services (Cronbach's =.82). Items included questions regarding satisfaction with involvement in deciding the amount, type, and timing of services. Because these two individual scales were highly correlated (Pearson correlation =.68, p =.000), the scales were combined to make a measure of service satisfaction (Cronbach's =.84) (Table 3).
|
=.84). Items included questions regarding whether the respondent would like to receive information regarding types of services available to them, service agencies and workers, and the experiences of others with either these services or providers.
Locus of Control
A measure was created to gauge the level of control respondents felt they had over their lives. This measure was created because the study assumed that elder preferences are shaped in part by the particulars of the individuals' situations including (a) how they perceive their care needs, (b) their understanding of the range of choices available to them to satisfy these needs, (c) what makes for a meaningful choice among the options available, and (d) how their decisions are considered and respected (or not respected) by others. To measure the respondents' sense of general control in their lives, a five-item scale was created that included three items from a Mastery Scale developed by Pearlin, Lieberman, Menaghan, and Mullin (1981) as well as two additional new questions (Cronbach's =.69).
Desire for Control
Three measures were created to gauge the level of control desired by the respondents over their worker, service package, and community long-term-care service decisions. A four-item scale was created to measure elders' desire for control over decision making (Cronbach's =.83), incorporating questions regarding the recruitment, hiring, termination, and training of their service workers. A three-item scale measuring desire for control over service planning (Cronbach's =.91) was also created. Questions in this scale asked respondents whether they would want complete control in determining the types of services they receive, making decisions related to their services, and setting their service schedule. Finally, an eight-item scale measuring desire for worker supervision was also created (Cronbach's =.85). Questions in this scale asked respondents whether they wanted responsibility for worker recruitment, training, payment, and the paperwork associated with employment of an individual.
Correlation Between New Measures
To determine the relationship between the new measures created for the study, a correlation analysis was performed (see Table 3). As reported in Table 3, there were significant correlations between the different measures. For example, desire for information was significantly correlated with desire for worker control (p <.001), desire for service control (p <.001), and desire for decision control (p <.01). Likewise, desire for decision control was significantly correlated with desire for worker control (p <.001), desire for service control (p <.001), and locus of control (p <.001). Finally, desire for worker control was significantly correlated with desire for service control (p <.001), desire for decision control (p <.001), and locus of control (p <.01).
Certainly, these significant correlations should be expected (e.g., desire for worker control would have a significant relationship with service control and decision control). The question for the study at this juncture was whether any of the correlations were of sufficient magnitude that might indicate the measures were not independent constructs. As reported in Table 3, desire for service control has a strong correlation with desire for information (Pearson correlation =.42). Also, desire for service control and desire for decision control have strong relationships with desire for worker control (Pearson correlation =.48; Pearson correlation =.42, respectively). However, none of the correlation scores between these measures indicates they are measuring the same construct. As a result, the respondent's score for each of these scales was used as an independent variable.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
2 tests were performed to compare differences among the four race groups and between gender groups. In addition, Duncan post-hoc tests were computed to examine statistical differences between the four racial/ethnic groups. Significant racial/ethnic differences in life situation and control variables were found across the four racial/ethnic groups. Table 4 presents the findings from the 2 and ANOVA tests that were conducted on the categorical and continuous variables, respectively.
|
Significant differences were found among groups with respect to age, marital status, health and functional status, living situation, decision, service, worker control scores, and general locus of control. On average, the Chinese elders were the oldest (M = 80) and reported the highest percentage of being married (37%) and living alone (60%). The Latino elders reported being the youngest (M = 75) and having the most functional limitations (M = 13) and the fewest number (4%) reporting to be in "good health."
Among the four groups, the Chinese elders reported having the greatest sense of control in life (M = 17), whereas the Latino elders reported the least (M = 13). Perhaps because of this, it was not surprising to find that the Latino elders expressed the most desire for information about services and providers (M = 3) and control over their services (M = 2.4). African American elders expressed the most desire for control over their workers (M =.92) and for making decisions regarding their services without assistance from family or others (M =.72). There were no significant differences between the groups regarding satisfaction with current services.
Table 4 also reports elder preferences for the different new approach models described earlier. With the exception of the Chinese elders, the overwhelming majority of African American (n = 146; 73%), Latino (n = 116; 88%), and White Western European (n = 152; 76%) elders preferred New Approach 3 (Traditional Case-Management Model). Only half of the Chinese elders (n = 100; 50%) preferred this model. Because of low numbers in New Approach 1 (Cash and Counseling) and New Approach 2 (Negotiated Care Model), the study was not able to use a multinomial logistic regression to distinguish between the three consumer-directed models. The study opted to combine elder preferences for New Approach 1 (Cash and Counseling) and New Approach 2 (Negotiated Care Model) (n = 217; 30%) and analyzed elder preferences for these more consumer-directed approaches against those preferring New Approach 3 (Traditional Case-Management Model; n = 514; 70%).
Multivariate Results
The original analysis plan for the study was to use a multinomial logistic regression to distinguish between the three consumer-directed models. As described above, this approach was dropped, and a logistic regression model, using the Statistical Package for the Social Sciences (SPSS, Chicago, IL), was then created. For the multivariate analysis, responses to which new approach model the respondents liked best were converted into the following dichotomous dependent variable: Traditional Case-Management Model versus the more consumer-directed models of Cash and Counseling Model and Negotiated Care/SHMO-Like Model (1 = New Approach 3; 0 = New Approaches 1 and 2). Figure 2 illustrates the revised analytic model.
|
Table 5 presents the hierarchical results predicting the odds of wanting New Approach 3 (Traditional Case-Management Model) versus the combined New Approach 1 (Cash and Counseling) and New Approach 2 (Negotiated Care Model). When Block 1 variables were entered, elders who either desired more control over their services or their worker or felt more control in their lives were less likely to select the Traditional Case-Management Model. However, elders who desired more information about their service package or providers were more likely to choose this option.
|
In the full model (respondent characteristics, race membership, race interactions with significant characteristics), good health drops from significance, whereas desiring control over services returns to significance. Being either Chinese or Latino remains a significant predictor for being less likely to select the Traditional Case-Management Model. However, when considering the interaction of race/ethnicity with significant model characteristics, Chinese elders with a greater sense of control in their lives and Latino elders who desired more control over their services were more likely to choose the case-managed model. Other significant race/ethnicity interaction terms included Latino and African American elders who desired more control over their workers and were less likely to choose the case-managed model.
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
First, the respondents were recruited from three community-based agencies noted for their cultural competency. Elders receiving services from these organizations typically have case managers and direct care workers who can communicate in their native language or are of the same race/ethnicity. It was not surprising, therefore, that the percentage of elders in this study who were either satisfied or very satisfied with their care or their involvement with decisions surrounding their care was 85%. Furthermore, there was no significant difference in the percentage of elder satisfaction between racial/ethnic groups. This might explain why 71% of elders in this study selected the Traditional Case-Management Model. However, neither general service satisfaction nor satisfaction with the level of involvement with care decisions was a significant predictor of selecting any of the new approach models.
Stemming from the first limitation, a second limitation of this study is the fact that the respondents had been receiving services for at least 12 months. The study design wanted individuals who had sufficient experience with receiving formal care services so they could state a preference regarding the amount of control they desired. However, their established relationship with agencies and agency representatives may have influenced their preference for the Traditional Case-Management Model. Other studies have reported much higher preferences for consumer-directed care among potential users of community long-term-services care users (Simon-Rusinowitz & Mahoney, 2001; Simon-Rusinowitz et al., 1997) than were found in this study. It is possible that familiarity with their existing service delivery influenced the preferences of the study respondents.
A third study limitation is that the elders were presented with only hypothetical models regarding consumer direction in community long-term care. If, however, these elders were selected to participate in the Cash and Counseling Demonstration, they would have received more information than provided in this survey. They also would have had an opportunity to discuss their options with trained case managers. It is not clear whether elder preferences for the different consumer-direct models used in this study would be the same if the elders were, indeed, required to make decisions regarding consumer direction in community long-term care or if they had more time to reflect on the differences in each model. Even with these limitations, the findings of this study do provide some insight for individuals interested in the development of consumer-directed care models for elders and/or for closing the gap between elder perceptions and preferences and professional perceptions and judgments.
Consumer Direction Is a Continuum
At its essence, consumer direction attempts to give individuals greater control over the management and delivery of their community long-term-care services. When examining the effects of control, research on elders often overlooks the possibility that individuals may feel as though they can exercise more control in some domains of their life than in others (Bisconti & Bergeman, 1999). For example, Diwan, Berger, and Manns (1997) report community long-term-care recipients may reject one service because they want a family member to do it, but that same client may want to accept other services. Certainly, the findings of this study suggest that elders who desire more control in one area of their community long-term-care service delivery may not desire more control over all areas.
The respondents in this study were asked to select a primary model of consumer-directed care that reflected the level of control they desired in the organization, planning, and management of their community long-term care. The study found that elders who desired control over their services were more likely to select the Traditional Case-Management Model, suggesting, perhaps, that consumer direction (for these elders) could be achieved through existing agency-based models. Perhaps, for this group, having meaningful participation in making service-delivery decisions would suffice for their sense of consumer direction. This finding is certainly in keeping with other published observations suggesting elders' preferences for the amount of control they want in service planning and management are complicated and not a simple all-or-nothing proposition (Capitman & Sciegaj, 1995; Cohen, 1992; Eustis & Fischer, 1992).
Not surprisingly, the study found that elders who desired control over their workers would be less likely to select the Traditional Case-Management Model. Indeed, based on the findings of this study, desire for worker control seems to be the strongest consistent predictor for level of consumer direction preferred by elders. When race membership is added, desire for worker control remains significant (when other previously significant variables do not). And desire for worker control produces a significant interaction with African American membership as a negative predictor for the traditional model. These two findings suggest a need for further analyses of the study data to identify level of elder preferences for control in the different service-delivery areas (e.g. planning, decisions, worker).
Cultural Competency and Consumer Direction
The study findings reveal that whereas racial/ethnic group membership might predict the elder's preference for one of the three models (e.g., being Chinese was a negative predictor for the Traditional Case-Management Model), heterogeneity also exists within the groups. The findings of heterogeneity within racial/ethnic groups illustrate the need for cultural competency as service systems consider the development of consumer-directed service provision. Had the study analyses focused on race/ethnicity only as a series of dichotomous variables, the findings would have concluded that elders who desired control over their services or elders who were Chinese or Latino were less likely to want the Traditional Case-Management Model. However, as noted by LaViest (1994), a dichotomous racial/ethnic variable does not capture within-group differences that may influence outcomes.
For this reason, the study created models to test racial/ethnic interactions. We found that if the Latino elders happened to desire greater service control or the Chinese elders had a greater sense of overall control in their lives, then they, as a subgroup within the racial group, were more likely to want the Traditional Case-Management Model. Our findings seem to confirm the importance of looking beyond global racial/ethnic variables and support recent publications regarding the importance of examining the interaction of race/ethnicity with other life situational or attitudinal factors (Angel & Angel, 1997; Burlingame, 1999; Mui, Choi, & Monk, 1998).
As Angel and Angel (1997) articulate, race/ethnicity may give elders a structure through which to view the world; however, it does not necessarily dictate behavior. Three recent reviews on this topic conclude that race/ethnicity competes with other factors in determining individual preferences for how elders live their lives, address illness and care, and what kind of help they desire (Burlingame, 1999; Mui et al., 1998). These authors point out that a truly sophisticated view of race/ethnicity and its impact on an elder's needs and preferences requires an understanding of its interaction with other factors that have a significant impact on the older adult's life and social context.
The findings of this study of racial/ethnic differences in elder preferences regarding the amount of control (and areas of control) over community long-term-care services suggest a need for service providers to become better informed about the complexities of what it means to be "autonomous" and how these understandings interact with a person's racial/ethnic background, traditions, and culture. The study findings also indicate that it may be important to provide elder consumers with a range of options that might allow for greater control in some areas of community long-term-care service delivery (and not necessarily in all areas of service delivery). In doing so, care planners can work with elders and their families to assess the potential trade-offs between enhanced consumer control and safeguards for community long-term-care service provision.
|
Footnotes |
|---|
1 RoseMary B. Fuss Center for Aging and Intergenerational Studies, Lasell College, Newton, MA.
2 Schneider Institute for Health Policy, Heller School, Brandeis University, Waltham, MA.
3 Hofstra University, Long Island, NY.
Decision Editor: Linda S. Noelker, PhD
Received for publication September 11, 2002. Accepted for publication January 20, 2004.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. M. Wiener, W. L. Anderson, and G. Khatutsky Are Consumer-Directed Home Care Beneficiaries Satisfied? Evidence From Washington State Gerontologist, December 1, 2007; 47(6): 763 - 774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. E. Alkema, J. Y. Reyes, and K. H. Wilber Characteristics Associated With Home- and Community-Based Service Utilization for Medicare Managed Care Consumers Gerontologist, April 1, 2006; 46(2): 173 - 182. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
