|
|
| |||||||||||||||||||||||||||||||||
Background Older persons who are hospitalized for acute illnesses often lose their independence and are discharged to institutions for long-term care.
Methods We studied 651 patients 70 years of age or older who were admitted for general medical care at a teaching hospital; these patients were randomly assigned to receive usual care or to be cared for in a special unit designed to help older persons maintain or achieve independence in self-care activities. The key elements of this program were a specially prepared environment (with, for example, uncluttered hallways, large clocks and calendars, and handrails); patient-centered care emphasizing independence, including specific protocols for prevention of disability and for rehabilitation; discharge planning with the goal of returning the patient to his or her home; and intensive review of medical care to minimize the adverse effects of procedures and medications. The main outcome we measured was the change from admission to discharge in the number of five basic activities of daily living (bathing, getting dressed, using the toilet, moving from a bed to a chair, and eating) that the patient could perform independently.
Results Twenty-four patients in each group died in the hospital. At the time of discharge, 65 (21 percent) of the 303 surviving patients in the intervention group were classified as much better in terms of their ability to perform basic activities of daily living, 39 (13 percent) as better, 151 (50 percent) as unchanged, 22 (7 percent) as worse, and 26 (9 percent) as much worse. In the usual-care group, 40 (13 percent) of the 300 surviving patients were classified as much better, 33 (11 percent) as better, 163 (54 percent) as unchanged, 39 (13 percent) as worse, and 25 (8 percent) as much worse (P = 0.009). The difference between the groups remained significant (P =0.04) in a multivariable model in which we controlled for potentially confounding base-line characteristics of the patients. Lengths of stay and hospital charges were similar in the two groups.
Fewer patients assigned to the intervention group were discharged to long-term care institutions (43 patients [14 percent], as compared with 67 patients [22 percent] in the usual-care group; P = 0.01). Among the 493 patients discharged to private homes, similar proportions (about 10 percent) in the two groups were admitted to long-term care institutions during the three months after discharge.
Conclusions Specific changes in the provision of acute hospital care can improve the ability of a heterogeneous group of acutely ill older patients to perform basic activities of daily living at the time of discharge from the hospital and can reduce the frequency of discharge to institutions for long-term care.
Patients 65 years of age or older account for 31 percent of acute care hospital admissions in the United States and 45 percent of hospital expenditures for adults.11 These older patients are at high risk for loss of independence and institutionalization.12,13,14,15 Many interventions designed to improve the outcomes of acutely ill elderly patients have had disappointing results.15,16,17,18,19,20,21
We used complementary principles of quality improvement and comprehensive geriatric assessment to develop a new system of care for acutely ill older patients in our hospital.3,22 This program, which we call Acute Care for Elders, is designed to help patients maintain or achieve independence in basic activities of daily living through the combined effects of four key elements: a specially designed environment, patient-centered care, planning for discharge, and review of medical care (Table 1).22 In order to carry out the specific nursing protocols and to facilitate the work of an interdisciplinary team, our teaching hospital instituted this program in a single 14-bed unit.
|
From November 1990 through March 1992, we compared the outcomes of patients treated in this special unit with those of patients who received usual care at the University Hospitals of Cleveland, a private, nonprofit teaching hospital affiliated with Case Western Reserve University that has 874 beds.
During the study period, 1794 patients who were 70 or older were admitted for general medical care. Patients who were admitted to a specialty unit (e.g., intensive care, cardiology–telemetry, or oncology) were ineligible for the study (n = 2067). We randomly assigned 651 of the 1794 eligible patients either to the Acute Care for Elders program (n = 327) or to usual care (n = 324) in another general medical unit; the remaining 1143 eligible patients were not enrolled in the study because beds were not available in both the intervention and usual-care units at the time of their admission. Assignment according to computer-generated random numbers was performed at the time of admission by the admitting clerk. Informed consent was obtained orally from patients or their proxies (usually family members), according to procedures approved by the hospital's institutional review board.
Patient Care
In both the intervention and usual-care units, each patient was assigned a primary nurse, two resident physicians, and an attending physician. The intervention and usual-care units had the same hospital-supported staff-to-patient ratios (roughly one budgeted position for a registered nurse for each two beds) and used the same hospital-wide support services (for example, social work, physical therapy, and nutrition). Extramural grant support provided funds for increases in hours worked in the intervention unit by the medical and nursing directors, social worker, physical therapist, occupational therapist, and dietitian; these fractional increases totaled less than one additional full-time person per year. Under the leadership of the medical and nursing directors, the primary nurse assigned to each patient in the intervention group was responsible for assessing the patient's specific needs daily and implementing protocols for the prevention of disability and for rehabilitation.
Usual care consisted of services provided by physicians and nurses in other acute care medical units. The staff of the intervention unit was not involved in the care of patients receiving usual care, and none of the four elements of the program were implemented in usual-care units. However, attending physicians and resident physicians provided care to patients in both the intervention and usual-care groups.
Sources of Data
Data were obtained by means of interviews and from medical records. On admission, interviews were attempted with each patient, the patient's primary nurse, and a family member or other care giver. The patient and nurse were also interviewed at the time of discharge, and the patient and a family member or care giver were contacted three months after discharge. The interview covered sociodemographic characteristics, the ability to function in performing activities of daily living,23,24 the ability to walk, overall health status, the items on a geriatric depression scale,25 and the first 21 items of the Mini–Mental State Examination.26 The five basic activities of daily living that were included in the study were bathing, dressing, using the toilet, moving from a bed to a chair, and eating23; continence was not included because it is not reported as reliably as the other self-care activities.27 The seven instrumental activities of daily living were shopping, cooking, performing household chores, using transportation, managing money, managing medication, and using the telephone.24 On admission, patients and family members were asked about the patients' functional status and overall health two weeks before admission and at the time of admission. Three months after discharge, information was obtained about any stays in long-term care facilities, readmissions to a hospital, and health care services by paid providers at home.
All interviews were conducted by research assistants who were not involved in patient care. These interviewers received standardized training consisting of a written protocol, scripted scenarios, role playing, and observed interviews. Standardization of interviews was maintained during the study by weekly reviews of interview procedures. Interviewers were not blinded to the patients' group assignments. Interrater reliability for data obtained by the interviewers was assessed for 10 hospitalized patients (1.5 percent). The mean kappa statistics were 0.98 for the basic activities of daily living, 0.94 for the instrumental activities of daily living, 0.96 for the mental-status items, and 0.99 for items related to depression.
Clinical data were obtained from medical records. These included the reason for hospitalization,28 the Charlson comorbidity score,29 and the Acute Physiology and Chronic Health Evaluation (APACHE) II score30 on the day of admission.
Data on each patient's hospital charges as reported to the Health Care Financing Administration on the Universal Bill-1982, a standard billing form, were obtained from the hospital. Actual hospital costs, as estimated by the hospital's cost-accounting system, were available for 601 patients and correlated highly with charges (r = 0.96). The results of analyses of hospital costs and charges were similar and are not reported separately. We estimated the cost of the additional staff time paid for by the grant and the capital costs of special features of the intervention unit, which were not included in hospital reports of charges or costs.
Missing Data
Interview data were obtained primarily from the patients; 461 (71 percent) were interviewed on admission, 403 of the 603 surviving patients (67 percent) at discharge, and 328 of the survivors (63 percent) three months after discharge. When information about function was not available from the patients, data from proxy respondents were used; the primary nurse was the source of proxy data obtained at admission and discharge, and a family member or care giver was the source of preadmission and post-discharge data. Only patients' reports were used for the assessment of mental status and mood.
Statistical Analysis
The main outcome variable was the change from admission to discharge in the number of basic activities of daily living that the patients could perform independently. The patient's functional status at discharge was classified as better if the number of activities he or she could perform independently increased from admission to discharge, and worse if this number decreased. Changes from two weeks before admission to discharge were calculated similarly.
We examined differences between the intervention and usual-care groups in base-line characteristics, main end points and other outcomes at discharge, hospital charges and length of stay, and outcomes after discharge. Differences between the groups in the change in the ability to perform activities of daily living independently were evaluated with the chi-square test for linear trend.31 The results of ridit analyses32 confirmed those of the chi-square tests for linear trend and are not reported separately. We used the Wilcoxon rank-sum test to assess differences in continuous variables and the chi-square test for categorical variables, with the modification for linear trend when appropriate.31 The consistency of differences between the intervention and usual-care groups was examined in subgroups defined by several clinical characteristics: age, ability to perform basic and instrumental activities of daily living two weeks before admission, comorbidity score, and APACHE II score at admission. Stratified and multivariable ordinal logistic-regression analyses were used to control for potentially confounding factors.31,33 All statistical tests were two-sided, with P = 0.05 as the criterion to indicate statistical significance.
Results
On admission, the patients randomly assigned to the intervention group were similar to those assigned to receive usual care in most sociodemographic characteristics, health-status measures, chief reasons for admission, and coexisting illnesses (Table 2). Forty-eight of the 651 patients (7 percent) died in the hospital — 24 patients each in the intervention and usual-care groups.
|
At the time of hospital discharge, 65 (21 percent) of the 303 surviving patients in the intervention group were classified as much better in terms of their ability to perform basic activities of daily living, 39 (13 percent) as better, 151 (50 percent) as unchanged, 22 (7 percent) as worse, and 26 (9 percent) as much worse. In the usual-care group, 40 (13 percent) of the 300 surviving patients were classified as much better, 33 (11 percent) as better, 163 (54 percent) as unchanged, 39 (13 percent) as worse, and 25 (8 percent) as much worse (P = 0.009) (Figure 1). The findings were similar (P = 0.009) in an analysis limited to the 382 patients who provided complete information about their functional status at both admission and discharge.
|
Table 3 shows the changes from hospital admission to discharge in the number of activities of daily living that could be performed independently by different groups of patients. In the intervention group, benefits were seen for patients less than 80 years old (P = 0.03), those who were able to perform independently fewer than five basic activities of daily living two weeks before admission (P = 0.04), and those with APACHE II scores of 0 through 14 (P = 0.02). Trends toward a benefit were seen in other groups.
|
With regard to changes from admission to discharge in the ability to perform individual activities of daily living, more patients assigned to the intervention group improved, and fewer became worse, in their ability to bathe and dress themselves (P = 0.006 and P = 0.02, respectively). More patients in the intervention group than in the usual-care group showed improvement in their ability to move from a bed to a chair and to use the toilet, but these differences were not significant (P = 0.2 and P = 0.3, respectively).
Fewer patients assigned to the intervention group were discharged to a long-term care institution (43 patients [14 percent], as compared with 67 patients [22 percent] in the usual-care group; P = 0.01) (Table 4). Among the 555 patients admitted to the hospital from private homes who survived to discharge, fewer patients assigned to the intervention group than to the usual-care group were discharged to long-term care institutions for the first time (9 percent vs. 16 percent, P = 0.02). Among the 493 patients discharged to private homes, similar proportions of the intervention and usual-care groups were discharged with plans for new paid health care at home, such as the services of a nurse, homemaker, or health aide (43 percent vs. 41 percent, P = 0.6).
|
Resources Used in the Hospital
The mean length of the hospital stay was 1 day shorter for patients assigned to the intervention group (7.3, as compared with 8.3 days for the usual-care group), but the median length of stay was the same (6 days) for each group (P = 0.4). Mean total hospital charges were $10,289 for patients in the intervention group, as compared with $12,412 for patients who received usual care; the median hospital charges were $7,057 and $7,839, respectively (P =0.3; 95 percent confidence interval for the difference between groups, -$1,212 to $392). The cost of additional hours worked by clinical personnel for the 17-month intervention period totaled $65,000. Capital costs allocated to the intervention group for the special features of the unit totaled $10,500. The total costs not reflected in hospital charges were thus approximately $75,500, or $231 for each of the 327 patients assigned to the intervention program.
Outcomes and Use of Resources during the Three Months after Discharge
During the three months after discharge, 82 patients died — 42 (14 percent) assigned to the intervention group and 40 (13 percent) assigned to the usual-care group; 6 patients (1 percent) were lost to follow-up. Three months after discharge, the intervention and usual-care groups did not differ significantly in the mean number of basic activities of daily living the patients could perform independently (4.0 and 3.8, respectively; P = 0.3), the mean number of instrumental activities of daily living they could perform independently (3.9 and 3.8, respectively; P = 0.5), or overall health status as reported by the patients or proxy respondents (P = 0.5).
Among the 493 patients discharged to private homes, similar proportions of the intervention and usual-care groups were admitted to long-term care institutions during the three months after discharge (24 of 260 patients [9 percent] vs. 23 of 233 patients [10 percent], respectively). Fewer patients assigned to the intervention group lived in long-term care institutions at any time during the three months after discharge (67 patients, vs. 90 patients in the usual-care group; P =0.03) (Table 4). Among the 603 patients who survived to discharge, similar numbers in the intervention and usual-care groups were readmitted to acute care hospitals during the three months after discharge (104 of 303 patients [34 percent] vs. 109 of 300 patients [36 percent], respectively; P = 0.6), and similar numbers received paid health care services from nurses, health aides, or homemakers (158 of 303 patients [52 percent] vs. 143 of 300 patients [48 percent]; P = 0.3).
Discussion
This randomized trial provides evidence that specific changes in the provision of acute hospital care can improve the ability of a heterogeneous group of older patients hospitalized with acute illnesses to perform basic activities of daily living at the time of discharge. More patients assigned to the intervention group — as opposed to those who received usual care — improved from admission to discharge in their ability to perform basic activities of daily living, and fewer became worse, despite their somewhat shorter hospital stays and shorter time for recovery before discharge. Other differences observed at discharge were consistent with this main finding, and the differences between the groups remained significant in multivariable and stratified analyses in which we controlled for potentially confounding base-line characteristics of the patients. Fewer patients assigned to the intervention group were discharged to long-term care institutions or lived in them during the three months after discharge. For every 15 patients treated in the intervention unit, 1 more patient was returned home than from the usual-care units, and 1 less patient was admitted to a long-term care institution. The beneficial effects we observed were apparently achieved without increasing in-hospital or post-discharge costs. Nevertheless, it is important to emphasize that the functional status of the majority of patients in both groups was unchanged or worse at the time of discharge. Three months after discharge, the groups did not differ significantly in terms of their ability to perform basic or instrumental activities of daily living.
Comparison with Past Studies
Despite the dramatic benefits of geriatric evaluation and management at the end of acute care hospital stays for at least some patients,34,35 consultative and unit-based interventions to improve the functional outcomes of acutely ill hospitalized older persons have had little benefit.15,17,18,19,20,21,36,37,38,39,40,41,42,43 The intervention we studied differed from most earlier interventions in several ways, notably by incorporating the physical redesign of the hospital unit, the key role assigned to nurses in initiating assessment and case management, and the scheduling of daily rounds by a multidisciplinary team. Two earlier studies44,45 also involved nurse-initiated interventions; both of these studies found evidence of beneficial effects.
Methodologic Considerations
We recognize potential limits to the validity of our findings. The impracticality of blinding patients and interviewers to the treatment assignments may have biased the reports of outcomes. To obtain reports about health status for all patients, proxy reports were required in many cases, and reports from patients and proxies may differ.27,46 Nonetheless, several factors support the validity of our findings. Different outcome measures consistently indicated that the intervention had a beneficial effect. The better function in the intervention group than in the usual-care group at the time of discharge was consistent in subgroup and multivariable analyses. Finally, bias in the design and conduct of the study was reduced by the random assignment of patients, the use of measures with established validity and reliability, and the complete follow-up data for the main outcome variable.
Our study was designed to test the efficacy of the Acute Care for Elders program as a whole, rather than to determine the relative efficacy of its different components, its cost effectiveness, or its long-term effects. Further evaluation will be necessary to address these issues and to test the effectiveness of this approach in other settings.
Implications
The loss of functional independence is not an inevitable consequence of acute illness and hospitalization among older patients. The intervention we studied can serve as a model for improving aspects of overall function — indicated by the ability to care for oneself — that are not specifically related to a particular disease or treatment. Although functional outcomes are rarely the focus of conventional medical care, they may be critical determinants of the quality of life, independence, cost of care, and prognosis among older patients.46,47,48 Functional outcomes are especially important in acutely ill, hospitalized older patients, who are often frail, chronically ill, and at high risk for functional decline and institutionalization for long-term care.12,14,48 Furthermore, an intervention program such as ours does not preclude, and may complement, disease-specific or treatment-specific efforts to improve patients' outcomes.4,5,6,7,8,49
Supported by grants from the John A. Hartford Foundation (88277-3G) and the National Institute on Aging (AG-10418-03) to the Claude D. Pepper Older Americans Independence Center at Case Western Reserve University. Dr. Landefeld's work was supported in part by a Career Development Award from the Health Services Research and Development Service, Department of Veterans Affairs, and a George Morris Piersol Teaching and Research Scholarship of the American College of Physicians.
We are indebted to Michael Vender, M.S.S.A., L.I.S.W., the geriatric resource nurses, the staff of the Unit for the Acute Care of Elders, and Barbara Juknialis, M.A., Linda Quinn, M.S., and Miriam Rosenblatt, B.S., for their assistance; we are also indebted to Mary-Margaret Chren, M.D., Sidney Katz, M.D., Donna Regenstreif, Ph.D., David B. Reuben, M.D., Gary E. Rosenthal, M.D., and Mark Sager, M.D., for their comments and suggestions.
Source Information
From the Division of General Internal Medicine and Health Care Research (C.S.L., R.H.F.) and the Division of Geriatrics (C.S.L., R.M.P., R.H.F., J.K.), Department of Medicine, Case Western Reserve University; the University Hospitals of Cleveland (C.S.L., R.M.P., D.M.K., R.H.F., J.K.); and the Cleveland Veterans Affairs Medical Center (C.S.L., J.K.) — all in Cleveland. Presented in part at the annual meetings of the American Federation for Clinical Research, Baltimore, May 1–4, 1992, and Washington, D.C., April 30–May 1, 1993.
Address reprint requests to Dr. Landefeld at the Division of General Internal Medicine and Health Care Research, 111 G(W), Cleveland Veterans Affairs Medical Center, 10701 E. Blvd., Cleveland, OH 44106.
References
| |||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
