| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to David M. Bass, Benjamin Rose, 850 Euclid Avenue, Suite 1100, Cleveland, OH 44114. E-mail: dbass{at}benrose.org
 |
Abstract |
|---|
 TOP Abstract MethodsFindingsDiscussionReferences |
|---|
From its inception in the 1920s and 1930s, the philosophy of managed care has placed high value on health information and support services to prevent crisis episodes and excessive utilization, and to help patients cope emotionally with illness (Lasker & Committee on Medicine and Public Health, 1997). Although information and support are part of the managed care philosophy, economic pressures lead many provider organizations to eliminate these nonmedical services or ration them only to health plan members at highest risk (e.g., those with multiple chronic conditions, or frequent hospitalizations and emergency department visits).
As an alternative strategy for dealing with economic pressures, some managed care health systems are maintaining their information and support services by establishing partnerships with community agencies that specialize in the provision of these nonmedical services (HMO Workgroup on Care Management, 1999). Low administrative overhead, reliance on volunteers, and use of philanthropic dollars allow these agencies to provide information and support services at a lower cost than possible in large health systems (Lasker & Committee on Medicine and Public Health, 1997).
This investigation examines outcomes for patients and family caregivers of an experimental partnership between a managed care health system and a community agency that provides information and support services. This partnership was established between Kaiser Permanente of Ohio and the Cleveland Area Alzheimer's Association to improve the quality of information and support services for patients with dementia and their family members. Because dementia often goes undiagnosed in primary care, patients and family members often do not receive sufficient information and support (Callahan, Hendrie, & Tierney, 1995). Even when diagnosed, patients and family members often do not receive enough information about the disease, treatments, and community services, and feel abandoned by health care providers (Beisecker, Chrisman, & Wright, 1997; Chenoweth & Spencer, 1986; Connell & Gallant, 1996; Glosser, Wexler, & Balmelli, 1985; Haley, Clair, & Saulsberry, 1992). This lack of information interferes with the effectiveness of treatments for dementia and other coexisting health conditions (Menzin, Lang, Friedman, Neumann, & Cummings, 1999), and may lead to more costly hospital stays (Coughlin & Liu, 1989; Gottlieb, 1999; Lyketsos, Sheppard, & Rabins, 2000).
Because of uncertainty about their ability to recognize, diagnose, and identify appropriate treatments for dementia, physicians and other providers report difficulties giving information and support to patients and families (Boise, Camicioli, Morgan, Rose, & Congleton, 1999; Small et al., 1997; Torian, Davidson, Fulop, Sell, & Fillit, 1992). Some acknowledge being uncomfortable dealing with behavioral symptoms (Glosser et al., 1985) and are not knowledgeable about available community services from agencies such as the Alzheimer's Association (Boise et al., 1999; Rubin, Glasser, & Werckle, 1987).
Experimental Partnership as an Intervention
Partnerships between managed care health systems and Alzheimer's Associations can take a number of different forms (National Chronic Care Consortium/ National Alzheimer's Association, 1998). The partnership tested in this investigation adds care consultation from the Cleveland Area Alzheimer's Association to usual managed care services offered to members of Kaiser Permanente.
Care consultation is a flexible, multicomponent intervention that builds on more than 10 years of research on interventions for family caregivers (Biegel & Schulz, 1999; Bourgeois, Schulz, & Burgio, 1996; Kennet, Burgio, & Schulz, 2000). It is a telephone intervention based on an empowerment conceptual framework (Guttierrez, GlenMaye, & DeLois, 1995). This framework assumes that patients and families have the capacity to make their own decisions if given sufficient information and coaching. Care consultants work with families in a collegial fashion to help identify personal strengths, as well as resources within the family system, health plan, and community. The goal is to provide tools to enhance patients' and caregivers' competence and self-efficacy. Care consultants also provide information about available community services, facilitate decisions about how to best utilize and apply for these services, and may contact service agencies on behalf of patients and caregivers.
Care consultants initiate the first contacts with patients and family caregivers. This strategy is intended to overcome delays in support and information service use or the use of these services only in times of crisis (Bass, McCarthy, Eckert, & Bichler, 1994; Costa et al., 1996). Care consultation is delivered by one of three Association staff members, two of whom are master's prepared licensed social workers.
Care consultants follow a standardized protocol for service delivery that includes conducting a structured initial assessment, identifying problems or challenges, and developing strategies for using personal, family, and community resources. Care consultants collaborate with patients and family caregivers to create an individualized plan of care. The care plan outlines specific tasks to be completed; assigns patients, family members, or Association staff/volunteers to work on these tasks; and gives a time frame for task completion and reassessment. Tasks often include using other Association services, such as education and training programs, support groups, a respite reimbursement program, and a nationwide program to return wanderers safely home. Regularly scheduled follow-ups monitor progress and add new tasks to the care plan as needed. Follow-ups are initially done biweekly, decreasing to 1-month and 3-month intervals unless needs dictate more frequent contacts. In difficult periods, daily contact with care consultants may be necessary. Alternatively, if care consultants, patients, and caregivers agree and there are no problems that have not been addressed or discussed, trained volunteers make follow-up contacts, with care consultants on call.
Although all persons in the intervention group are offered care consultation, there is variation in the extent to which patients and families accept services. On average, care consultants have 12 direct communication contacts with patients and caregivers per year. Control group patients and caregivers are able to contact the Association independently and use any of its services other than care consultation. Use of Association services other than care consultation by both the intervention and control groups is incorporated into the analysis. All Association services are free-of-charge.
Outcomes and Hypotheses
The effects of care consultation on three categories of outcomes are examined: (1) utilization outcomes, including patients' use of hospital, emergency department, and physicians; patients' use of community services; and patients' and caregivers' use of non-Association information and support services; (2) caregiver satisfaction with managed care services, including satisfaction with types and quality of services provided; and information about memory problems; and (3) caregiver depression and care-related strain, including perceived health deterioration, role captivity, and relationship strain.
The primary study hypothesis is that the intervention group, compared with controls, will have lower utilization, increased satisfaction, and decreased depression and care-related strain over the 12-month study period. In addition to the primary hypothesis, four secondary modifying-effects hypotheses are tested. Modifying-effects hypotheses posit that the impact of care consultation will be intensified for intervention group patients and caregivers with certain characteristics (Biegel & Schulz, 1999; Bourgeois et al., 1996; Gitlin, Corcoran, Winter, Boyce, & Hauck, 2001). These secondary hypotheses are derived from the extensive body of research on family caregiving, which suggests why, for whom, and under what circumstances interventions may be expected to work (Hohmann, 1999).
The first modifying-effects hypothesis suggests care consultation will have greater benefits when it is used in combination with other Association services (i.e., support groups and education programs; Kosloski & Montgomery, 1995; Mittelman et al., 1993). In combination with care consultation, these other Association services form a service package that is expected to address more fully patient and caregiver needs.
The second modifying-effects hypothesis expects the benefits of care consultation to be greater when patients have only a diagnosis of memory problems rather than a specific dementia diagnosis (Mittelman, Ferris, Shulman, Steinberg, & Levin, 1996). Having a specific dementia diagnosis is one indicator that patients and caregivers have received an adequate assessment, which includes discussing symptoms, treatment options, and the likely course of the illness (Costa et al., 1996). Patients with memory problems, who have not received a specific dementia diagnosis, are less likely to have had a complete diagnostic assessment. This increases the chance that patients and caregivers have less complete information about symptoms, causes, treatments, and prognosis of memory problems. This heightened need for information is expected to increase the potential benefits of care consultation as an information resource.
The third modifying-effects hypothesis is that care consultation will lead to greater improvements in outcomes when caregivers consider patients' symptoms to be more severe. Perceived seriousness of symptoms has long been recognized as a key factor that motivates help-seeking and acceptance of interventions for health problems (Becker, 1974). Perceiving patients' memory symptoms as more serious at the start of the demonstration is expected to increase the chances that the intervention group will accept assistance offered by care consultants and thus experience greater benefits (Bass, McClendon, Brennan, & McCarthy, 1998; Lin, 1986).
The fourth modifying-effects hypothesis is that care consultation is expected to offer greater benefits for nonspouse caregivers. Nonspouse caregivers, who are mainly adult children, compared with spouse caregivers, are more likely to accept assistance and services, such as care consultation (Auslander & Litwin, 1990; Pruchno, Michaels, & Potashnik, 1990). Resistance to service use by spouses stems from feelings that it is their sole duty to provide care, and that services will complicate caregiving (Collins, Stommel, King, & Given, 1991) or force a change in the care arrangement, such as nursing home placement (Bass, Noelker, & Rechlin, 1996; Cicirelli, 1983; Pruchno et al., 1990).
|
Methods |
|---|
|
TOPAbstractMethodsFindingsDiscussionReferences |
|---|
Sample
The sample was drawn from Kaiser members whose medical records indicated they had either a specific diagnosis of dementia or a symptom code indicating memory loss. In addition, eligible patients had to be 55 years or older; reside outside of a nursing home at the start of the demonstration; and live in the Cleveland Area Alzheimer's Association's service area, which includes Cuyahoga County (including the city of Cleveland) and four adjacent counties.
Based on medical records, 525 patients met these eligibility criteria. As a first step in the consent process, primary care physicians sent personalized letters to eligible patients. The letter asked for written permission to release contact information to the evaluation team at Benjamin Rose so that patients and caregivers could be informed about the study and asked to participate. The evaluation team was permitted to contact patients and caregivers to ask for study participation only after Kaiser obtained written permission.
Written permission was obtained for 233 (44.4%) of the 525 patients. Limited available data showed no significant difference between patients who did and did not give permission to be contacted in whether they had a specific dementia diagnosis versus only a diagnosis of memory problems. Those who gave permission were significantly younger by an average of 2 years (76.4 vs. 78.3 years), and were significantly more likely to be males (40% vs. 32% males).
Of the 233 cases that agreed to be contacted by the evaluation team, consent for participation by patients, caregivers, or both was obtained in 210 cases (90%). Consenting cases compared with refusals did not differ in terms of having a specific dementia diagnosis, patient age, or patient gender. In 182 of the 210 cases, a primary family caregiver completed an initial T1 interview. The 27 other participating cases only include interviews with patients because they did not have or need a family caregiver, or their family caregiver refused to participate.
Of the 182 caregivers who completed T1 interviews, 157 also completed T2 interviews 1 year later and are included in analyses of outcomes constructed from interview data. These outcomes include selected utilization measures, satisfaction with managed care services, and caregiver depression and strain. Fourteen of the 25 caregivers who did not complete T2 refused to continue in the study, 6 could not be reached, 3 died prior to the interview, and patients of two caregivers died after the T1 interview but prior to beginning the intervention. Analysis comparing the 157 caregivers who completed T1 and T2 and the 25 who completed only T1 found those who participated in both interviews were significantly more likely to be Kaiser members, to be spouses of patients, to report more care-related health deterioration at T1, and to be caring for patients who had a specific dementia diagnosis. No differences were found in T1 levels of satisfaction with managed care services, caregiver depression, relationship strain, or role captivity. The significant differences may indicate that caregivers included in this analysis are more connected to the Kaiser health plan and may be at greater risk of negative health consequences.
In addition to outcomes constructed from interview data, Kaiser medical record data are used to create health care utilization outcomes for number of hospitalizations, emergency department visits, and physician visits, and whether there was a Kaiser case management visit. Kaiser mandated that a separate written consent be obtained to access utilization data. This separate consent was obtained for 120 patients whose caregivers completed T1 and T2 interviews. These 120 patients differed significantly from those who refused access to utilization data by being an average of 4 years younger, more likely to have caregivers who are Kaiser members, and more likely to have spouse caregivers. No differences were found in the likelihood of patients having a specific dementia diagnosis, caregiver satisfaction with the health plan, and caregiver depression and strain.
Measures
Intervention and Modifying Characteristics
The intervention and control groups are compared by including in the analysis a dichotomous variable that assigns a value of 1 to the intervention group and a value of 0 to the control group. Support for the primary hypothesis, which posits overall differences between the intervention and control groups, is evident by a significant relationship between this dichotomous variable and the dependent outcome measures.
The modifying-effects hypotheses are tested using product terms created by multiplying the dichotomous intervention variable and each of the four modifying characteristics. Three of the modifying characteristics are dichotomous: Whether patients or caregivers use any Association services in addition to care consultation during the 1-year period, whether patients' medical records listed a specific dementia diagnosis versus only memory problems, and whether caregivers are patients' spouses. The fourth modifying characteristic is a five-item index of the severity of patients' memory difficulties as reported by caregivers during T1 interviews. The Appendix displays the individual items that comprise this measure and Cronbach's alpha reliability.
Utilization Outcomes
Four utilization outcomes represent the use of Kaiser managed care services during the 1-year period. Three of these utilization outcomes are continuous: number of hospital admissions, number of emergency department visits, and number of physician visits. The fourth measure is dichotomous, indicating whether patients had any Kaiser case management visits during the 1-year period. A dichotomous rather than continuous measure is used for this service because of the 20% of patients who had any Kaiser case management visits; only 3% had more than one visit.
Two additional continuous utilization outcomes come from caregiver interview responses to questions about the use of nine types of community services. These services could have come from Kaiser or from some other community agency or hired helpers. Exploratory factor and reliability analyses indicated that the nine services form two indices representing the number of direct-care home and community services used by patients and the number of non-Association information and support services used by patients or caregivers. (See the Appendix for a list of specific services in each index and Cronbach's alpha reliability.) Both indices exclude any services provided by the Association.
Caregiver Satisfaction With Health Plan Outcomes
There are three multi-item indicators of satisfaction with Kaiser managed care services. Exploratory factor analyses of data from T1 and T2 interviews confirmed that these measures represent independent dimensions of satisfaction. The three measures include: a two-item index of satisfaction with the types of Kaiser services, an eight-item index of satisfaction with the quality of Kaiser services, and a five-item index of satisfaction with information received about the illness. Items, scoring, and Cronbach's alpha reliability for each index are displayed in the Appendix.
Caregiver Depression and Strain Outcomes
This category of outcomes is represented by four multi-item indices that come from caregiver interviews. Exploratory factor and reliability analyses of T1 and T2 data confirmed that items making up these measures form four distinct outcomes (see Appendix for items and Cronbach's alpha reliability).
Caregiver depression, a measure of general well-being (Stull, Kosloski, & Kercher, 1994), reflects the frequency of 11 symptoms of depression during the week preceding the interviews. The depression measure comes from a managed care demonstration being conducted in five communities across the United States (Montgomery, 1998). Nine of the 11 items are identical to those in the short version of the Center for Epidemiologic Studies–Depression (CES-D) scale (Kohout, Berkman, Evans, & Cornoni-Huntley, 1993). Two items from the short CES-D, reflecting the interpersonal domain (i.e., "people dislike me" and "people were unfriendly to me"), are omitted. These interpersonal items were dropped because of concerns about the cultural variation in the meaning among Hispanic respondents. Two other items from the full CES-D are used as substitutes (i.e., "bothered by things that don't usually bother me" and "trouble keeping your mind on what you were doing").
Caregiver strain is represented by three measures of the perceived negative effects specific to caregiving (see Appendix for items and Cronbach's alpha reliability). Four items are used to assess relationship strain between patients and caregivers. A measure of health deterioration is comprised of seven items that asked caregivers about any adverse health effects from caregiving (Bass, McClendon, Deimling, & Muhkerjee, 1994; Bass et al., 1996). Role captivity is constructed from three items that reflect feelings of being trapped in the caregiving role (Pearlin, Mullan, Semple, & Skaff, 1990).
Controls
A T1 version of each outcome measure is used as a covariate to control for preintervention levels, and ensure that multivariate analyses represent change in outcomes over the 1-year study period (Kessler & Greenberg, 1981). Two additional control factors are included. One is a dichotomous variable indicating if the patient died before completion of the 1-year study period. This variable is used when examining utilization outcomes in order to control for the more limited time period during which these patients could use services. The second control is a continuous variable representing the number of days between T1 and T2 interviews. Small variation in the projected number of days between interviews resulted from caregivers' availability in scheduling interviews.
Analytic Strategy
Ordinary least-squares regression is used for examining continuous outcomes and logistic regression for the dichotomous outcome. The modifying-effects hypotheses are tested by including four product terms created by multiplying the intervention variable by each modifying characteristic (Cleary & Kessler, 1982; Southwood, 1978). Only significant product terms are retained in the final equations. When product terms are significant, additional equations are estimated to describe the modifying effects further.
|
Findings |
|---|
|
TOPAbstractMethodsFindingsDiscussionReferences |
|---|
|
|
|
.01), whereas the effect of the intervention variable for the less impaired is not significant (b =.002; p =.99).
T2 Caregiver Satisfaction Outcomes
In Table 4, regression equations for the three measures of satisfaction with the health plan show that the intervention variable has significant main and interaction effects with the dementia diagnosis variable. Because of the significant interactions, significant main effects are interpreted as the impact of the intervention for patients who do not have a specific dementia diagnosis (b =.35,.31, and.40, respectively). These effects indicate increased caregiver satisfaction with quality of health plan services, types of health plan services, and information received about the illness when patients have not received a specific dementia diagnosis. The main and interaction effects together imply there is a significantly smaller relationship between the intervention and these outcomes for patients who have a specific dementia diagnosis [e.g., estimated slope for quality of health plan services (b = -.46 +.35 = -.11)].
|
The product term for the intervention variable and perceived memory difficulties also is significant for satisfaction with the types of Kaiser services (b =.34), with significant increases in satisfaction in the intervention group when patients have more severe memory difficulties. This is illustrated by estimating one equation for patients with above average severity of memory difficulties, and another equation for patients with average or below average severity. When severity is below average, the intervention's effect is not significant (b =.06; p =.74), whereas there is a significant intervention effect among the more severely impaired (b =.48; p =.05).
T2 Caregiver Depression and Strain Outcomes
The first equation in Table 5 is for depression and shows a significant main effect of the intervention (b = -.12). Caregivers in the intervention group have greater decreases in reported symptoms of depression over the 1-year study period.
|
The last two equations in Table 5 have significant product terms for the interaction of the intervention and other Association service use (b = -.33 and-.51, respectively). These effects indicate the intervention significantly reduces health deterioration and role captivity when patients and/or caregivers use other Association services. A separate equation for the subsample that uses other Association services confirms that the intervention significantly reduces caregivers' feelings of health deterioration (b = -.27; p =.03) and role captivity (b = -.38; p =.02). The intervention variable has a small, nonsignificant effect on health deterioration (b =.12; p =.36) and role captivity (b =.14; p = 35) among those who do not use other Association services.
|
Discussion |
|---|
|
TOPAbstractMethodsFindingsDiscussionReferences |
|---|
Utilization Outcomes
Three of the six service utilization outcomes are used significantly less often by the intervention group. For Kaiser case management and direct care community services, findings correspond to the primary hypothesis, with lower utilization in the intervention group as a whole. For non-Association information and support services, the difference between the intervention and control groups is consistent with the modifying-effects hypothesis. For patients with more severe memory difficulties, the intervention group has lower utilization than the control group.
The three services that are used less frequently by the intervention group provide many of the same types of assistance as care consultation, particularly in the areas of health information, health education, care planning and coordination, and emotional support. Lower utilization may occur because care consultation functions as an effective substitute for these services. Compared with many other services, particularly Kaiser case management, care consultation has advantages of specializing in needs of people with memory problems and the ability to provide care over an extended period of time.
Findings for traditional medical services, including emergency department visits, hospital admissions, and physician visits, do not show any significant intervention-control group differences. The lack of differences may indicate that care consultation is not sufficient to reduce potentially preventable emergency department visits, hospitalizations, and physician visits. Alternatively, there may be only a small number of potentially preventable utilization episodes for these services during a short 1-year time period, which may limit the ability to detect any impact of the intervention. Another explanation for the lack of differences may be that utilization of these medical services is more closely linked to other comorbid conditions rather than dementia or memory problems. It may be unrealistic to expect care consultation to impact service use not mainly caused by dementia or memory problems.
Satisfaction Outcomes
For each dimension of satisfaction (i.e., satisfaction with quality of services, types of services, and information about the illness), caregivers in the intervention group whose patients had not received a specific dementia diagnosis before the start of the demonstration have increased satisfaction with managed care services. These findings support the modifying-effects hypothesis. Patients who had not received a specific dementia diagnosis before the demonstration have a greater need for information about symptoms, course of treatment, and care (Costa et al., 1996). Although they do not diagnose dementia, care consultants provide a wealth of information about memory symptoms, causes, treatments, and strategies for managing symptoms. Care consultants also guide patients and families to use managed care services effectively. Increased satisfaction with managed care may result because patients and family members in the intervention group become more knowledgeable consumers and are empowered to use the health plan more effectively. Increased satisfaction also may reflect favorable opinions about care consultation, which from consumers' points of view is an added part of managed care services.
An additional significant effect among the satisfaction outcomes also supports the modifying-effects hypothesis: Intervention group caregivers who indicate their patients have more severe memory symptoms are more satisfied with the types of managed care services offered by the health plan. Caregivers of patients with severe symptoms are more likely to have sought assistance for memory problems from the health plan before the demonstration. These prior experiences may increase caregivers' awareness of how the partnership with the Association has improved available services for people with memory problems and enabled the managed care system to expand its service options.
Caregiver Depression and Strain
Caregiver depression is an indicator of general well-being not specifically linked to caregiving. Results for this outcome support the primary hypothesis, with intervention group caregivers showing a significant decrease in depression compared with the control group. Care consultation may decrease depression by attending to the needs of caregivers, which contrasts to the typical health care emphasis solely or primarily on patient needs. Care consultants also work to increase the size and effectiveness of caregivers' informal support networks, thus reducing social isolation that may cause or exacerbate depression. Additionally, this assistance may enable caregivers to give higher quality care to their patients, which may help avoid crisis episodes that are emotionally distressing.
Caregiver strain outcomes are specifically linked to perceptions of negative caregiving consequences. Findings for caregiver strain support modifying-effects hypotheses, with intervention group nonspouse caregivers having reduced strain in the relationship with patients, and caregivers who use other Association services in combination with care consultation having reduced health deterioration and role captivity.
The intervention's ability to reduce relationship strain among nonspouse caregivers may reflect caregiver receptiveness to outside assistance from service providers, such as care consultants. Nonspouse caregivers, compared with spouses, are more open to and accepting of services to address care-related problems (Collins et al., 1991; Noelker & Bass, 1995). Receptivity to services may allow care consultants to facilitate a more positive relationship between patients and caregivers by providing information about the impact of memory problems on patients' moods and behaviors, distinguishing symptoms of illness from intentional negative behaviors, offering strategies for dealing with difficult symptoms, and suggesting ways of fostering positive interactions despite the memory problems. Less impact of care consultation on relationship strain for spouse caregivers also may reflect greater stability in long-standing patterns of interaction within a marriage and may limit possible changes caused by a caregiving intervention (Auslander & Litwin, 1990; Pruchno et al., 1990).
The ability of the intervention to reduce health deterioration and role captivity when other Association services are used in conjunction with care consultation may result because these services reinforce and complement suggestions and recommendations given by care consultants. Other Association services are based on a similar philosophy as care consultation, which promotes consumer empowerment. Care consultants and staff/volunteers who provide other Association services share a common orientation about how to best assist caregivers. Additionally, the care plan developed through the collaboration of care consultants and families is used by other Association services as a blueprint for guiding interactions with caregivers and focusing on common service goals and objectives. Care consultants and other Association staff and volunteers also may jointly discuss cases and establish unified and consistent strategies for working with families.
Results of this demonstration suggest that care consultation delivered within a partnership between a managed care system and Alzheimer's Association may be a promising strategy for improving care for dementia patients and their caregivers. The beneficial impact of the intervention is evident in reduced use of selected managed care and community services, increased satisfaction with managed care services, and decreased caregiver depression and care-related strain. Some beneficial effects apply to the intervention group as whole, whereas others are evident only when modifying characteristics are incorporated into the analysis. Only a small number of prior intervention studies have explicitly tested for modifying effects (Hohmann, 1999). In this investigation, several beneficial effects of the intervention would have been overlooked had the analysis only tested for evidence of the primary hypothesis.
The demonstration has a number of features that limit generalizablity of findings and highlight the necessity of broader replication studies. For example, the sample does not represent all caregiving situations, even within the context of managed health care. Although this sample of caregivers is larger than 85% of recent intervention research (Bourgeois et al., 1996; Kennet et al., 2000), it may underrepresent the largest segment of patients whose memory symptoms go unnoticed and undiagnosed in health systems (Callahan et al., 1995). Examples include caregiving situations in which symptoms are mild, and there is little or no contact with providers, or when patients have fewer comorbid conditions that could be complicated by dementia. Despite these limitations, the sample has some advantages over prior studies by including the selection of patients from primary care rather than specialty care clinics, and not being restricted to patients and caregivers already using services targeted to persons with Alzheimer's disease or dementia.
Generalizablity may be limited also because the study was conducted in only one staff-based managed care system. It is unknown whether this type of partnership can be established and yield similar benefits in other types of managed care environments or whether this type of partnership can benefit Medicare patients not enrolled in managed care. Further study is needed to assess whether the nearly 200 other Alzheimer's Association chapters in the United States can provide care consultation as part of a partnership within a managed care system. The Cleveland Area Alzheimer's Association is one of the largest chapters and has extensive experience with local and national demonstration projects. It is not clear whether smaller and/or less experienced Associations could maintain a partnership with large health systems.
Several other issues not addressed in this analysis would further clarify the impact of care consultation. First, more detailed information on the types of assistance provided by care consultants, as well as the amount and timing of that assistance, would help identify the exact causes of improvements in outcomes. Second, psychosocial outcomes for patients, such as depression and satisfaction with services, were not examined, although they will be considered in future analyses. Third, the demonstration would have been strengthened by having more detailed medical information about changes in patients' memory difficulties over the course of the intervention. Although caregivers report on patients' illness during interviews, after the initial recruitment period, clinical data from providers were not available. Fourth, this study did not examine the impact of care consultation on physicians and other providers, including whether the intervention altered interactions between patients, caregivers, and health care providers.
|
|
Footnotes |
|---|
This demonstration was funded by grants from The Cleveland Foundation, The Elisabeth Severance Prentiss Foundation, The Abington Foundation, The Eva L. and Joseph M. Bruening Foundation, the Alzheimer's Association, The Laura Horn Memorial Trust, the Ohio Department of Aging, and the Administration on Aging. We thank the staff of Kaiser Permanente of Ohio and the Cleveland Area Alzheimer's Association for developing and implementing the intervention, with special thanks to Linda Choken, Wendy Dahar, Diane Graham, Dr. Kathleen Grieser, Lynn Hodge, Dr. Alan Khoury, Katie Maslow, Ann Scott, and John Wallin. The authors also thank demonstration co-investigators Dr. Richard Fortinsky, the University of Connecticut, Department of Family Medicine, and Dr. Kathleen Smyth, Case Western Reserve University, University Memory and Aging Center. An earlier version of this article was presented at the 54th Annual Meeting of The Gerontological Society of America, Chicago, IL, November 2001. 
1Margaret Blenkner Research Institute/Benjamin Rose, Cleveland, OH.
2Cleveland Area Alzheimer's Association, Cleveland, OH.
Received for publication February 27, 2002. Accepted for publication May 28, 2002.
|
References |
|---|
|
TOPAbstractMethodsFindingsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. B. Hogan MD, P. Bailey MD, S. Black MD, A. Carswell MSc PhD, H. Chertkow MD, B. Clarke MD, C. Cohen BA MD, J. D. Fisk PhD, D. Forbes RN PhD, M. Man-Son-Hing MSc MD, et al. Diagnosis and treatment of dementia: 4. Approach to management of mild to moderate dementia Can. Med. Assoc. J., October 7, 2008; 179(8): 787 - 793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Sebern and C. J. Whitlatch Dyadic Relationship Scale: A Measure of the Impact of the Provision and Receipt of Family Care Gerontologist, December 1, 2007; 47(6): 741 - 751. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. G. Vickrey, B. S. Mittman, K. I. Connor, M. L. Pearson, R. D. Della Penna, T. G. Ganiats, R. W. DeMonte Jr., J. Chodosh, X. Cui, S. Vassar, et al. The Effect of a Disease Management Intervention on Quality and Outcomes of Dementia Care: A Randomized, Controlled Trial. Ann Intern Med, November 21, 2006; 145(10): 713 - 726. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
