| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Yale School of Medicine, Department of Epidemiology and Public Health, New Haven, CT
b Yale School of Medicine, Department of Internal Medicine, Dorothy Adler Geriatric Assessment Center, New Haven, CT
c Yale University, Program on Aging, New Haven, CT
Correspondence: Elizabeth H. Bradley, PhD, Yale School of Medicine, Department of Epidemiology and Public Health, 300 College Street, New Haven, CT 06520. E-mail: elizabeth.bradley{at}yale.edu.
Decision Editor: Vernon L. Greene, PhD
 |
Abstract |
|---|
 TOP Abstract MethodResultsDiscussionReferences |
|---|
Key Words: Goal-setting • Geriatric assessment • Outcomes
A fundamental challenge in evaluating the delivery of health care to older Americans is conceptualizing relevant outcomes of care for aging populations. In their recent reviews, both Binstock and Spector 1997
and Fennel and Flood 1998 argued that outcome evaluation studies are rarely designed to examine the chronic and multifaceted health issues common among older patients, many of whom are frail and have some degree of cognitive or functional impairment.
Geriatric assessment is a multidisciplinary effort to assess the medical, functional, social, and psychological needs of older people with multiple problems (Borgenicht, Carty, and Feigenbaum 1997), with the objective of identifying pertinent health problems and recommending the most effective care to achieve the best outcome. The literature evaluating outpatient geriatric assessment includes a wealth of empirical studies of selected outcomes of geriatric assessment (Boult et al. 1994; Burns, Nichols, Graney, and Cloar 1995; Epstein et al. 1990; Kerski, Drinka, Carnes, Golob, and Craig 1987; Morishita, Boult, Boult, Smith, and Pacala 1998; Rubin, Sizemore, Loftis, and Loret de Mola 1993; Silverman, Musa, Martin, Lave, Adams, and Ricci 1995; Tulloch and Moore 1979; Toseland et al. 1996, Toseland et al. 1997; Williams, Williams, Zimmer, Hall, and Podforski 1987; Yeo, Ingram, and Crapo 1987). These evaluative studies have been conducted in a variety of settings, including community-based practices (Boult et al. 1994, Burns et al. 1995; Kerski et al. 1987; Morishita et al. 1998; Rubin et al. 1993; Silverman et al. 1995; Yeo et al. 1987), Veterans Administration services (Toseland et al. 1996, Toseland et al. 1997; Williams et al. 1987), and HMO settings (Epstein et al. 1990). Among these studies, the most commonly examined outcomes have been (a) functional status; (b) health services use; (c) costs of care; (d) caregiver stress; and (e) patient and family satisfaction.
Results from the outcomes studies of geriatric assessment are equivocal. Although some studies (Burns et al. 1995; Yeo et al. 1987) have found that patients in geriatric assessment experienced improvements in functioning relative to patients in the control group, many found no long-term effects on functioning (Epstein et al. 1990; Silverman et al. 1995; Kerski et al. 1987; Tulloch and Moore 1979). Similarly, some studies have reported no reductions in health care use and costs associated with outpatient geriatric assessment (Epstein et al. 1990; Kerski et al. 1987; Silverman et al. 1995), whereas others have reported significant reductions in service use (Rubin et al. 1993; Toseland et al. 1997; Williams et al. 1987). Finally, although reductions in caregiver stress (Silverman et al. 1995) and enhanced patient satisfaction (Morishita et al. 1998; Silverman et al. 1995) have been found to be associated with outpatient geriatric assessment in some studies, others have detected no differences in these outcome measurements (Williams et al. 1987; Yeo et al. 1987).
In addition to the inconsistent results reported among these studies, only four to five outcomes have been examined systematically. These outcomes likely reflect beliefs on the part of clinicians, health services researchers, and policymakers about what constitutes a successful medical encounter for this population of older patients, including reduced utilization and costs and patient and family satisfaction. However, they have not been based on empirical investigations of what families, patients, and clinicians have as goals of geriatric assessment or what they are trying to accomplish by this mode of practice. Because goals are what people want to achieve, understanding goals of care is essential for identifying meaningful outcomes against which to measure the success of clinical practice. The objective of this study was to describe goals of geriatric assessment as expressed by patients, family caregivers, physicians, and case managers in order to assess the range of objectives fundamental to the practice of outpatient geriatric assessment. Understanding the breadth of these goals may help elucidate outcomes of care to be assessed in future evaluations of outpatient geriatric assessment and better link the process of geriatric assessment to outcomes of care.
|
Method |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
Sample
To be potentially eligible for the study, patients had to be new patients to the center during the study period (July 1997–February 1998) and had to be seen for a comprehensive geriatric assessment by one of the center's geriatricians. During the study period, 242 potentially eligible patients were seen at the center. In 2 cases, both the patient and family caregiver were excluded from being interviewed (because the patient was too cognitively impaired to be interviewed and no family caregiver existed); thus, neither the physician nor the case manager were given surveys to complete, resulting in a total of 240 eligible sets of patients, family caregivers, physicians, and case managers. Among the 240 sets of potential respondents, surveys were not distributed to clinicians as a result of administrative error in 12 cases, and clinicians did not complete the survey in 2 cases, resulting in 226 completed surveys for both clinician groups (response rate among physicians and among case managers, 94.2%).
Patients were excluded from interview if they had a Folstein Mini-Mental State Exam (MMSE) score (Folstein, Folstein, and McHugh 1975) of less than 18 
if they did not speak English 
, if they were unable to comprehend the interview 
, or if the case manager had requested the patient not be approached 
, leaving 171 potential patient respondents. The MMSE cutoff of 18 was chosen to maximize the likelihood of meaningful patient participation in the interview while still including patients with mild to moderate dementia. Of the 171 potential patient respondents, 35 refused and 2 were missed, resulting in 134 completed interviews (response rate among patients, 78.4%).
Family caregivers were not interviewed if the case manager requested the family not be approached 
), and in 17 cases, there was no family caregiver, leaving 217 potential family caregiver interviews. Of these, 4 family caregivers could not be reached, and 13 family caregivers refused to be interviewed, leaving 200 completed interviews (response rate among family caregivers, 92.2%).
Data Collection and Measurement
Patient interviews were conducted in person at the center immediately following the assessment. Family caregiver interviews were conducted on the telephone within 1 week of the geriatric assessment. This interval was chosen to provide adequate time for the family caregiver to reflect on findings and recommendations from the assessment. Physicians and case managers were given a self-administered questionnaire to complete within 1 week of the assessment. A minority of assessments can require 2–3 visits to complete. In these cases, interviews were still conducted on the basis of the first assessment visit.
The questionnaires focused on the participant's goals for the patient's care. Questions were developed after an extensive qualitative study of goal setting in geriatric medicine, which defined both a taxonomy of goals (Bogardus, Bradley, and Tinetti 1998) and a broader theory of goal setting in clinical medicine (Bradley, Bogardus, Tinetti, and Inouye 1999). This earlier qualitative research identified a set of goal domains including safety, independence in day-to-day functioning, social support, caregiver stress, medical issues, and behavioral and emotional issues. Thus, within each of these six domains, all participants, including all participating patients, were asked to describe their goals concerning the patient's care with a single open-ended question: "What are your goals for [patient's name] care regarding his/her ability to do day-to-day activities?" Participants were also asked if they had any other goals, in addition to those mentioned within the six prespecified domains. Responses to these questions were recorded verbatim by the interviewer, and all open-ended responses were then coded by means of content analysis (Neuman 1994), a standard approach to coding open-ended responses. After extensive training in use of the coding key, three coders completed the content analysis of open-ended responses. Kappa coefficients for all three coders exceeded 0.6, indicating good interrater reliability (Cicchetti and Sparrow 1981; Fleiss 1981).
Data regarding race, marital status, education, age, functional status, and cognitive status were abstracted from the patients' medical records at the outpatient geriatric assessment center. Functional status was measured by using the number of both basic and instrumental activities of daily living (ADL) impairments (Katz, Ford, Moskowitz, Jackson, and Jaffe 1963; Lawton and Brady 1969); cognitive status was measured by means of the Folstein MMSE score (Folstein et al. 1975), recorded as part of the assessment.
Data Analysis
The total number and the prevalence of each type of goal was described, using standard descriptive statistics. For purposes of summarizing the data, specific goals were described in broader categories of goals. For instance, improving driving and monitoring driving were both described in the broader goal category of driving. As examples of verbatim goals and how they were coded, one caregiver said that her goals for the patient's care were "to get [the patient] to take her medicine better and to have her eat more." This was coded as two separate goals, under the category of medications issues for the first desired outcome mentioned and under the category of health behaviors for the second. In another case, a patient stated this goal: "I just want to feel better, like I used to." This goal was coded as general health and well-being.
Differences in the mean number of goals among types of respondents were tested by using t tests, with the family caregiver group as the reference category. The association between the mean number of goals and patient's cognitive status (dichotomized as MMSE of less than 24 vs. 24+) was assessed by using paired t tests, for each of the respondent types. In a targeted subgroup analysis, differences in the prevalence of goals in the more common goal areas for patients with and without cognitive impairment (i.e., MMSE less than 24 vs. 24+) were compared using chi-square test or Fisher's exact test where appropriate.
|
Results |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
|
. Further, case managers had the most goals of any group, including family caregivers 
reported fewer goals on average than those without cognitive impairment 
However, clinicians (physicians and case managers) reported more goals on average in cases of patient cognitive impairment (physicians 4.1 vs. 3.1 and case managers 5.9 vs. 4.2, respectively, p < .01). The mean number of goals reported by family caregivers did not vary by patient's cognitive status.
|
|
|
Targeted subgroup analysis of the more common goal areas indicated that physicians and case managers were more likely to have goals in two areas, supervision and education/referrals, for patients with versus without cognitive impairment (p < .01). Physicians were also more likely to report goals in the area of social/family relations and activity for patients with versus without cognitive impairment (p = .02). Patients with versus without cognitive impairment were more likely to report goals in the area of supervision (p = .03). The prevalence of goals reported by family caregivers was not associated with patient's cognitive status for any of the goal categories.
|
Discussion |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
), case management (Gwyther 1995; Shapiro 1995), and more recently, geriatric medicine literature (Fischer, Hillel, Stoeckle, and Emanuel 1997; Mold 1995; Post and Whitehouse 1998; Rockwood, Stolee, and Fox 1993; Sachs 1998; Yip et al. 1998). Because complex, chronic illnesses such as dementia currently have no cure, goals of care may be difficult to describe and may encompass desired outcomes unrelated to the alleviation of disease. Nevertheless, understanding goals for care is essential to identifying what it is that patients and families want to achieve with medical care. Thus, describing goals of care is a necessary component of designing, implementing, and evaluating effective and appropriate outcomes-based systems of care. This study is the first of which we are aware that provides a general description of the goals of care from the perspectives of patients undergoing outpatient geriatric assessment, many of whom have moderate or mild dementia, and their family caregivers. The findings are limited to the extent that the study was conducted at a single site and that this site delivers care primarily for patients with varying degrees of dementia. Conducted in other types of geriatric assessment centers, the study might reveal additional goals among patients, families, and clinicians. Despite this limitation, the study demonstrates that most patients, even those with dementia, are able to articulate goals, and, in many cases, patients may have multiple goals. Further, the study finds that both formal and informal caregivers typically have several goals for care rather than a single goal or desired outcome. Finally, the study reveals the breadth of goals that exist among patients, family caregivers, and clinicians. The results suggest that, although functioning and general health and well-being are among the more common goals expressed, many other goals are also deemed equally important by families, patients, and clinicians in the setting of geriatric assessment.
The study has implications for both clinical practice and research concerning outpatient geriatric assessment. In terms of clinical practice, the study suggests that older patients, with varying degrees of cognitive impairment, can often articulate goals for their own care. The findings also highlight the diversity of goals among family caregivers, patients, and clinicians, suggesting that explicit discussions of goals may be necessary to meet the needs of patients and their families and to focus care on meaningful outcomes. Measuring the extent of goal concordance or discordance among family caregivers, patients, and clinicians and identifying the clinical impact of such concordance or discordance are important topics for future research.
The implications of the study for research concerning outpatient geriatric assessment are substantial. As already described, the outcomes assessed in previous evaluative studies of outpatient geriatric assessment are limited in scope. These outcomes include patient functioning, health services use, costs of care, caregiver stress, and patient and family satisfaction. Use and costs are clearly important outcomes from a policy perspective, and functioning and satisfaction represent broad areas of health and well-being. However, the findings of our study indicate that these broad and policy-relevant measures of success in geriatric assessment may neglect fundamental goals of care as expressed by patients, families, and clinicians. Patient functioning, commonly evaluated in previous outcomes studies of geriatric assessment, was noted as a common goal of care among all respondent groups. However, other outcomes typically evaluated were rarely identified as goals. Caregiver burden was described as a goal by only 14.2% of patients, about 10% of family caregivers and physicians, and 4% of case managers. Future health services use and costs of care were mentioned as a goal by fewer than 2% of any group of respondents. In contrast, other goals were far more common, including obtaining education, information, or referrals; improving social and family relationships; and enhancing patient supervision. The degree to which these more commonly identified goals are met has not been a part of previous evaluations of outpatient geriatric assessment, yet meeting these goals of care may be fundamental to meeting the needs of patients and their families.
It is notable that the goals mentioned by patients and family caregivers were often process goals, as opposed to the outcomes typically assessed in health services research. This finding highlights the difference between the patient's and family's perspective, which may focus on goals more proximate to individuals' needs, and the health services research perspective, which focuses on more distal, and perhaps more policy-relevant, goals.
As previously noted, existing evaluations of geriatric assessment are equivocal as to the value of outpatient geriatric assessment in terms of assessed outcomes. This study suggests that the fundamental research question may not be whether geriatric assessment promotes these outcomes but rather whether these are the right outcomes to be measuring, given what patients and family caregivers state as their goals for care. Perhaps a better measure of success may be the degree to which patients' and caregivers' goals are met, regardless of whether these goals are to maintain the patient's functioning, to obtain an appropriate education and/or referrals, to improve social relationships, or to address specific medical and behavioral issues. This study was not designed to determine the reasonableness of individual goals for individual patients. However, absent a validated measure of goal reasonableness, this study demonstrates a set of goals that have face validity and may be important elements to the appropriate care of patients and their families in the context of geriatric assessment. Future studies might assess the degree to which success in meeting goals articulated by patients and family caregivers enhances more traditional outcomes such as functioning, caregiver burden, appropriate health care use, and cost effectiveness of care.
|
Acknowledgments |
|---|
We would like to thank Mary Tinetti, MD, for her support and guidance throughout this project, as well as Vasum Peiris, Sheel Pandya, and Evelyne Gahbauer for their research assistance.
Received for publication January 14, 1998. Accepted for publication November 30, 1999.
|
References |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. J. Cook and G. A. Rooke Priorities in Perioperative Geriatrics Anesth. Analg., June 1, 2003; 96(6): 1823 - 1836. [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Cohen, J. R. Feussner, M. Weinberger, M. Carnes, R. C. Hamdy, F. Hsieh, C. Phibbs, D. Courtney, K. W. Lyles, C. May, et al. A Controlled Trial of Inpatient and Outpatient Geriatric Evaluation and Management N. Engl. J. Med., March 21, 2002; 346(12): 905 - 912. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
