ASSESSING AND MANAGING HOSPITALIZED OLDER PATIENTS
SYSTEMS OF CARE FOR OLDER HOSPITAL PATIENTS
ALTERNATIVES TO AND TRANSITIONS FROM HOSPITAL CARE
Older people are at disproportionate risk of becoming seriously ill and requiring hospital care, whether it is in an emergency department, on a medical or surgical ward, or in a critical-care unit. Persons aged 65 years or older, who make up only 13% of the U.S. population, account for 36% of acute-care hospital admissions and nearly half of hospital expenditures for adults. The impact of older persons on acute hospital care will increase rapidly with the aging of the population: the proportion of the U.S. population aged 65 years or older is expected to increase to as much as 19% by 2025, with those aged 85 years or older increasing most rapidly.
Hospital use rates vary as much as threefold for Medicare beneficiaries with the same illnesses across different regions of the United States. There is no evidence that these differences in practice patterns are explained by differences in disease rates or severity. Hospital use and the use of hospital resources is much lower among patients enrolled in capitated insurance plans than among those enrolled in fee-for-service plans; this difference in resource use has not been systematically linked to differences in patient outcomes.
During hospitalization, older patients tend to receive less costly care than do younger patients. In the large Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT), for example, seriously ill patients in their 80s were found to have received fewer invasive procedures and less resource-intensive, less costly hospital care than similar younger patients received. This preferential allocation of hospital services to younger patients was not based on differences in patients’ severity of illness or general preferences for life-extending care and is consistent with evidence regarding outpatient care. Differences in the aggressiveness of care have not been shown to explain differences between older and younger patients in survival or other outcomes. The best guides to assessment and management in the care of any older hospitalized patient are the clinical circumstances and the patient’s preferences, irrespective of the patient’s age.
There is also wide variability among seriously ill older persons in preferences for life-sustaining treatments. On the basis of findings of SUPPORT, it has been recognized that even though fewer older patients prefer aggressive care than do younger patients, many older patients want cardiopulmonary resuscitation and care that is focused on life extension. Moreover, patients’ families and physicians commonly underestimate older patients’ desires for aggressive care. Thus, in providing care for acutely ill older persons, it is essential to determine individual preferences for the site of care and to define with the patient the goals of care.
Many of the serious illnesses disproportionately experienced by older persons require hospital care for optimal management. The benefits of hospitalization can be remarkable: correcting serious physiologic derangements, repairing vascular obstructions and broken bones, using highly technical biomedical advances in the treatment of life-threatening illnesses. The hospital can also be a place where older persons are exposed to unintended hazards: deteriorating functional status, adverse drug reactions, and cognitive decline. A systematic approach to assessing and managing older hospitalized patients offers the best chance of reducing the risk for and consequences of these common hazards.
An initial comprehensive assessment of the hospitalized older patient includes an evaluation of function at the level of the organ system, the whole person, and the person’s environment. This assessment can identify needs for which targeted interventions may improve function or reduce risk for adverse outcomes. This approach complements the traditional medical assessment by highlighting problems that are common in hospitalized older persons, and it is similar in concept to comprehensive geriatric assessment conducted in other settings. (See Assessment.)
Table 13.1 lists 10 hazards and opportunities that are commonly overlooked in elderly hospitalized patients. These problems are selected on the basis of their importance in relation to other clinical issues, the quality of relevant evidence, and their specificity to older persons. Other important problems (eg, prevention of deep-vein thrombosis, the effects of alcohol or tobacco use, ameliorating pain, and advance directives) are not specific to older persons, and some problems specific to elderly persons (eg, age-related decline in renal function) are widely recognized. (See also Alcohol and Drug Abuse, Palliative Care, Legal and Ethical Issues, Kidney Diseases and Disorders, Perioperative Care.) Table 13.2 shows where assessment for these common geriatric problems can be incorporated into the routine of a hospital admission history and physical examination.
Two types of evidence suggest that these interventions are a good use of physician time. First, for each problem, there is compelling evidence (cited below) supporting the proposed intervention, either because the efficacy of the intervention is well established (eg, warfarin to prevent stroke associated with atrial fibrillation), or because the associated problem is common, often overlooked, and ameliorable with a safe and inexpensive intervention. Second, systematic approaches to the evaluation and management of acutely ill older persons can improve patient outcomes and reduce hospital costs.
Once hospitalized, older patients are at high risk for loss of independence and institutionalization. Among hospitalized medical patients aged 70 years or older, ~15% decline during hospitalization in their ability to perform basic self-care activities of daily living (ADLs), another ~20% are discharged without recovering their baseline prehospitalization abilities, and 15% of those admitted from home are discharged to a nursing home. Loss of personal independence is often hastened by the combined effects of the acute illness that led to hospitalization and underlying chronic illnesses and impairments. In addition, many older patients have lost their “bounce”—their ability to adapt and maintain the homeostasis of their physiologic, psychologic, and social systems in the face of the acute insults to these systems by illness and hospitalization. Functional decline during hospitalization and failure to recover baseline function have been independently associated with increasing age, lower preadmission function in instrumental activities of daily living (IADLs), and several admission characteristics, including cognitive impairment, symptoms of depression, and malnutrition. Optimal care for the hospitalized older patient requires the physician to manage acute illness and simultaneously intervene where necessary to promote or maintain independent functioning.
The ability to perform ADLs and IADLs is necessary if the older person is to live independently, and functional dependence is associated with worse quality-of-life outcomes, shortened survival, and increased resource use. The older patient’s ability to perform ADLs and IADLs, determined at the time of admission, will serve as a useful baseline. If functional dependence is uncovered, the causes can be explored (eg, dependence in IADLs is often associated with dementia), and strategies to maintain and improve functional capacity can be initiated (eg, physical and occupational therapy). These strategies may be best implemented effectively for many patients by ward staff without consultation or referral. Social work consultation and early involvement of family or other caregivers is often necessary to plan postdischarge care for those older patients who are functionally dependent. (See also Rehabilitation; Psychosocial Issues.)
Walking facilitates the performance of virtually all ADLs and IADLs. The ability to walk briskly and the habit of regularly walking 1 mile or more daily are associated with prolonged survival. Immobility during hospitalization, however, leads rapidly to deconditioning and subsequent difficulty walking. The major risk for deconditioned hospital patients of walking is falling, which can lead to serious injury. Falls and fall-related injuries in hospitalized patients are associated with cognitive impairment, new medications and multiple medications, environmental factors in the hospital, and abnormalities of gait, balance, and lower-extremity strength, as well as with multiple chronic medical conditions and depression.
It is helpful to assess the patient’s gait, balance, lower-extremity strength, ability to get up from bed, cognition, and mood during the initial physical examination. Persons able to walk independently should be encouraged to do so frequently during hospitalization. Those able to walk but unable to do so safely and independently can receive assistance from hospital staff while walking several times daily. Formal physical therapy may yield additional benefits. The initial physical examination is also a good time to assess a patient’s risk for falls by inquiring about a history of falls and by a careful musculoskeletal and neurologic examination. Interventions that reduce falls in other settings may also prevent falls in the hospital. Prudent preventive strategies include avoiding restraints and tethers, providing walking assistance for those who walk with difficulty, and providing physical therapy for those with weakness or gait abnormalities. If significant soft-tissue and bony abnormalities of the feet are discovered, a referral for podiatric care is appropriate. (See Falls; Gait Impairment; Physical Activity; Rehabilitation; and Diseases and Disorders of the Foot.)
Most hospitalized elderly persons have impaired vision or hearing, and these sensory impairments are risk factors for falls, incontinence, delirium, and functional dependence. Although most visual and hearing impairments are readily corrected by eyeglasses or hearing aids, these appliances are often forgotten in the hospital.
Hospitalized older persons can be screened for sensory impairment by routinely asking if they have difficulty with seeing or hearing and whether they use eyeglasses or hearing aids. Physical examination including a test of visual acuity (eg, with a pocket card of the Jaeger eye test) and the whisper test of hearing, in which a short, easily answered question is whispered in each ear, is the next appropriate step in evaluation. For people with visual or hearing impairments, it is important to provide the appropriate assistive devices (eyeglasses or hearing aids brought from home), and staff may need to be instructed in the use of appliances to communicate more effectively. (See Hearing Impairment, and Visual Impairment.)
Depressive symptoms in hospitalized older persons are common, prognostically important, and potentially ameliorable. Major or minor depression occurs in roughly one third of hospitalized patients aged 65 years or older but is often undiagnosed. The presence of depressive symptoms is associated with increased risk for dependence in ADLs, nursing-home placement, and shortened long-term survival, even after controlling for baseline function and the severity of acute and chronic illness.
It is important to consider depression in all hospitalized older patients. Simply asking whether they feel down, depressed, or hopeless, or whether they have lost interest or pleasure in doing things, is a good place to start. A positive response to any one of these questions is likely sensitive to the diagnosis of depression (based on evidence from outpatients) and can be followed up by a formal assessment for an affective disorder. In hospitalized older persons, the presence of 3 or more of 11 depressive symptoms has been found to be 83% sensitive and 77% specific for a diagnosis of major depression.
Detection is the first and most important step in the management of depression. Psychotherapeutic interventions (environmental, behavioral, cognitive, and family) are safe and often effective in the initial management of patients with suspected depression. It is rarely necessary to begin pharmacotherapy during hospitalization for a medical or surgical condition, but follow-up shortly after discharge is critical. If pharmacotherapy is initiated, selective serotonin-reuptake inhibitors are often preferred because approximately 50% of older hospitalized patients have a contraindication to tricyclic antidepressants. (See Depression and Other Mood Disorders.)
Delirium is present in 10% to 15% of hospitalized older persons on admission, and it develops in up to 30% during the course of hospitalization. Delirium arising during the course of hospitalization is a predictor of prolonged hospital stay. Delirium is also associated with increased rates of in-hospital death and nursing-home placement. Patients who develop delirium may experience a worsening of a chronic cognitive impairment. Symptoms of delirium commonly persist for months following hospital discharge. Roughly one third of cases of delirium can be prevented by managing six risk factors for delirium appropriately: cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration. The diagnosis of delirium should be considered when any of the following is observed: fluctuation in mental status or behavior, inattention, disorganized thinking, and altered consciousness. The Confusion Assessment Method (CAM) is a useful screen for delirium. Prudent measures to prevent or ameliorate delirium include: avoiding medicines associated with delirium whenever possible; treating infection and fever; detecting and correcting metabolic abnormalities; frequently orienting patients with cognitive or sensory impairment; and avoiding excessive bed rest, room changes, and restraints. (See Delirium.) Prevention is the best strategy: a recent randomized trial examined the effect of systematic assessment for established delirium in hospitalized elderly patients on a medical service, with multidisciplinary care provided to delirious patients. Patients in the intervention arm had no better outcomes than those assigned to usual care.
Underlying cognitive impairment consistent with dementia is present on admission in 20% to 40% of hospitalized older persons, and it commonly goes undetected. Preexisting cognitive impairment is a risk factor for delirium, falls, use of restraints, and nonadherence with therapy. Also, there is intrinsic value in identifying previously undiagnosed dementia so that appropriate evaluation and management strategies can be implemented after discharge. Cognitive function can be assessed by the use of an established test of cognitive function, such as the Mini–Mental State Examination (MMSE) or the Mini-Cog test. The diagnosis of dementia should be considered in those a score of 24 or less on the 30-point MMSE. When dementia is a possibility, it is important to exclude reversible causes and to identify those patients for whom pharmacologic therapy and family-oriented interventions are warranted. (See Dementia.)
The number of drugs prescribed to hospitalized patients is directly proportional to their age. Moreover, hospitalization is a period of rapid turnover in drug therapies for older patients: One study found that 40% of drugs prescribed before admission were discontinued during hospitalization and 45% of drugs prescribed at discharge were started during hospitalization. Although older patients are at increased risk for inappropriate drug therapy, adverse drug effects, and drug-drug interactions, they may also be undertreated when effective therapies are not used or are used in inadequate doses. In one study 88% of older hospitalized patients were found to have had at least one or more clinically significant drug problem, and 22% had at least one potentially serious and life-threatening problem. Consultation by clinical pharmacists can improve appropriate prescribing and improve the older patient’s adherence to prescribed therapy.
A hospital admission is an ideal time to completely review a patient’s medication regimen and to discontinue those that are unnecessary or have low therapeutic value (eg, sedative hypnotics). During hospitalization and at discharge, a medication review is useful to identify prescribing errors in six common categories: inappropriate choice of therapy, incorrect dosage, incorrect schedule, drug-drug interactions, therapeutic duplication, and allergy. (See Pharmacotherapy.)
Nonvalvular atrial fibrillation is present in 5% or more of hospitalized older persons, often as an incidental finding. Consistent and compelling evidence from randomized trials shows that the risk of stroke in persons with atrial fibrillation can be reduced approximately two thirds by treatment with warfarin (eg, from 4.5% per year to 1.4% per year). Moreover, the beneficial effect of warfarin is maintained in persons aged 75 years or older and in those with other risk factors for stroke. Nonetheless, many older patients with atrial fibrillation are discharged from hospital without warfarin therapy, even when warfarin is indicated. Failure to prescribe warfarin is likely due to underestimation of the benefit of therapy, overestimation of its potential risk, and the difficulty of implementing, monitoring, and modifying therapy to minimize adverse effects.
In every hospitalized older patient, the history and physical examination can be targeted to identify the presence of chronic or paroxysmal atrial fibrillation. When atrial fibrillation is diagnosed, it is important to exclude valvular disease and hyperthyroidism as causes. In the absence of a strong contraindication, anticoagulation therapy is indicated to prevent stroke, usually with warfarin (see Perioperative Care, and anticoagulation information in the Appendix). A large randomized trial compared the long-term effects of ventricular rate control and anticoagulation to rhythm control (in which an attempt is made to convert patients to and maintain them in normal sinus rhythm). Adverse drug effects were less frequent with rate control and anticoagulation, and mortality at 5 years was somewhat lower (23.8% versus 21.3%, P = .08). New joint guidelines from the American Academy of Family Physicians and the American College of Physicians recommend rate control with anticoagulation as the preferred strategy for most patients. (See also the section on atrial fibrillation in Cardiovascular Diseases and Disorders.)
Serious deficiencies of macronutrients and micronutrients are common in hospitalized older patients. Key macronutrients are protein, calories, salt, water, and fiber. On admission, severe protein-calorie malnutrition is present in approximately 15% of patients aged 70 years or older, and moderate malnutrition is present in another 25%. Moreover, 25% of older patients suffer further nutritional depletion during hospitalization. Even after controlling for the underlying illness, its severity, and comorbid illnesses, malnutrition is associated with increased risk for death, dependence, and institutionalization.
In addition to other deficiencies of vitamins and electrolytes that may develop with protein-calorie malnutrition, vitamin-D deficiency is especially common among older hospitalized patients. In one large hospital, nearly two thirds of medical inpatients aged 65 years or older were found to be vitamin-D deficient; vitamin-D deficiency was nearly as common in inpatients without a risk factor for vitamin-D deficiency and in those taking multivitamins as in other patients. These data regarding the high prevalence of vitamin-D deficiency in hospitalized older persons complement evidence that vitamin D and calcium supplementation reduce by half the incidence of nonvertebral fractures in men and women aged 65 years or older. (See Osteoporosis and Osteomalacia, and Malnutrition.)
One review examined the evidence from randomized controlled trials of the benefits of oral nutritional supplements for older persons at risk of malnutrition. (Of the almost 2500 subjects included in the review, 22% were from studies of hospitalized patients.) Nutritional supplementation was found to be associated with reduced mortality and shortened length of hospital stay, although the reviewers call for additional research to substantiate these findings, as most of the studies included in their analysis were of poor quality. Beyond prescribing supplements, clinicians should assess malnourished older hospitalized patients for remediable factors such as difficulty chewing, or insufficient time or encouragement to eat. (See Eating and Feeding Problems.)
The maintenance of water and electrolyte balance requires special attention in older persons during and after fluid administration because of their decreased capacity to achieve and maintain homeostasis. Initial efforts can be directed toward achieving normovolemia and correcting electrolyte abnormalities. Subsequent efforts to maintain fluid and electrolyte balance are based on estimates of daily metabolic requirements. The intracellular volume is about 25% to 30% of body weight for men aged 65 to 85 years weighing between 40 and 80 kg (88 to 176 lb) and about 20% to 25% of body weight for women in the same age and weight ranges. The daily metabolic requirements, as a proportion of intracellular volume, can be estimated as follows: water in L, 10%; energy in 1000 kcal, 10%; protein in g, 0.3%; sodium in mol, 0.3%; and potassium in mol, 0.2%. Thus, for a 75-year-old woman weighing 60 kg (intracellular volume 12 to 15 L), the daily maintenance requirements are 1.2 to 1.5 L of water and nutrients providing 1200 to 1500 kcal, 36 to 45 mmol of sodium, and 24 to 30 mmol of potassium. Administration of fluids and electrolytes must be adjusted on the basis of daily physical examination and of serum values of electrolytes and renal function, as needed.
Hospitalization of older persons is sometimes precipitated by mistreatment, which includes physical or psychologic abuse, neglect, self-neglect, exploitation, and abandonment. Elder mistreatment was not recognized in the medical literature until 1975; it is now estimated to affect 700,000 to 1.2 million Americans annually. In a large prospective cohort study, the annual incidence of referral to protective services for mistreatment was found to be approximately 1% among persons aged 65 years or older. Those referred for abuse, neglect, or exploitation were found to have a lower rate of survival over 13 years of follow-up (9%) than those referred for self-neglect (17%) and those not referred (40%). Most older persons referred to protective services because of physical abuse have been seen in hospital emergency departments, and many emergency visits lead to hospitalization.
Universal screening for mistreatment has been recommended and can be implemented by asking each older patient, “Do you feel safe returning to where you live?” (The sensitivity and specificity of this and other screening approaches are unknown.) Further questions can explore the living situation and specific settings or aspects of mistreatment. It is important to consider the diagnosis of mistreatment when there are physical or psychologic stigmata, such as unexplained injury, dehydration, malnutrition, social withdrawal, or recalcitrant depression or anxiety. When mistreatment is suspected, most states require that Adult Protective Services or the equivalent state agency be contacted. (See Elder Mistreatment.)
All persons aged 65 years or older should be asked at the time of admission to the hospital whether they have received influenza or pneumococcal vaccination. During the fall and winter months, influenza vaccination can be administered to those who have not already received it. Pneumococcal vaccination can be administered to older hospitalized patients who do not recall having received it in the past 10 years. (See also Prevention.)
Three systematic approaches have been demonstrated in controlled trials to improve hospital care of older persons. These approaches involve comprehensive multicomponent interventions, two of which were implemented on designated medical units.
Geriatric evaluation and management (GEM) units for elderly patients who have stabilized during an acute hospitalization were developed and pioneered in Veterans Affairs medical centers. These units incorporate comprehensive geriatric assessment (including screening for geriatric syndromes, and assessment for and treatment of functional, cognitive, affective, and nutritional problems) with interdisciplinary team-based care. A multicenter randomized trial demonstrated improved ADL function and physical performance, relative to usual hospital care, for veterans assigned to GEM units. Some measures of health-related quality of life were also superior for patients treated on GEM units. These units did not affect mortality and were cost-neutral after consideration of both initial hospitalization costs and the costs of care after discharge.
A second approach involves a system of care designed to help acutely ill older patients to maintain or achieve independence in ADLs and IADLs; this system has been called Acute Care for Elders (ACE). This approach has been adapted in many acute-care hospitals where acutely ill elderly patients are admitted to an ACE unit. The four components of this intervention are as follows:
In a randomized trial involving 651 medical patients aged 70 years or older in a university teaching hospital, ACE was found to be associated with greater independence in ADLs at discharge, less frequent nursing-home discharge, and somewhat shorter and less expensive hospitalization. In a second randomized trial involving 1531 community-dwelling persons aged 70 years or older in a community teaching hospital, ACE was found to be associated with substantial differences in the satisfaction of patients, family members, physicians, and nurses but with only modest differences ADL function. These findings demonstrate that ACE is a promising approach to improving outcomes and reducing hospital costs for acutely ill older general medical patients, but the effects of ACE on patient outcomes are likely sensitive to factors dependent on the function of the interdisciplinary team.
A third systematic approach (the Hospital Elder Life Program) involves a multicomponent intervention to prevent delirium in hospitalized older patients. The intervention consists of protocols to manage six risk factors for delirium: cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration. Elderly patients receiving this intervention are not segregated on a special hospital ward or unit. In a prospective controlled study, the incidence of delirium was found to be reduced by one third, from 15.0% to 9.9%. The intervention was also associated with significant improvement in the degree of cognitive impairment among patients with cognitive impairment at admission and a reduction in the rate of use of sleep medications among all patients. Among the other risk factors, there were trends toward improvement in immobility, visual impairment, and hearing impairment. The intervention was not associated with a reduction in the severity, duration, or recurrence of delirium, and the effects of the intervention on other outcomes have not yet been reported. Nonetheless, these findings suggest that primary prevention of delirium is probably the most effective treatment strategy.
It is often assumed that older persons would prefer to be treated for acute illness at home rather than in the hospital whenever possible. The safety and feasibility of this approach for acutely ill older persons who would usually be hospitalized has been demonstrated. This approach, sometimes called the home hospital, requires intensive resources for medical and nursing care at home that are not yet widely available.
Older persons’ preferences for home rather than hospital care vary widely. In a study of community-dwelling older persons, virtually all preferred care in the site that would provide the higher probability of survival. When home care and hospital care provide equivalent probabilities of survival, roughly half of patients prefer care in each site, with those preferring home care being more likely to be white, better educated, living with a spouse, deeply religious, and dependent in two or more ADLs. The major difference perceived by older persons between home care and hospital care was feeling safer in the hospital than at home.
Almost one quarter of hospitalized patients aged 65 and older are discharged to another institution. Although it is labor-intensive, meticulous discharge planning may maximize the probability that patients maintain the clinical and functional benefits achieved by hospitalization, and probably reduces the risk of early readmission and the use of emergency services. Discharge planning ideally begins at hospital admission, with a projection of medical, nursing, rehabilitative, and functional support required by the patient at the time of discharge. The following items should be communicated to patients (or their caregivers) who are being discharged directly home: follow-up appointments; warning symptoms or signs to watch for with instructions on whom to contact; clinical disciplines (eg, nursing, physical therapy) contracted for provision in the home; and a reconciled medication list, with clarification of which prehospital medications are to be continued. Patients being discharged to other care venues should be oriented with respect to the nature of the institution, the identity of a new attending physician, and the expected frequency of physician visits. If the provider at the receiving institution differs from the hospital clinician, then clear and prompt communication is essential. Some items of information (critical but pending study results, nuances of goals of care or family dynamics) call for direct communication between sending and receiving clinicians. Otherwise, a pithy and prompt discharge summary containing the following will suffice: summary of hospital course with care provided; a list of problems and diagnoses; baseline physical functional status; baseline cognitive status; medication list (with termination dates for time-limited drugs such as antibiotics); allergies; tests results still outstanding; follow-up appointments; and information related to goals, preferences, and advance directives.
■ Cohen HJ, Feussner JR, Weinberger M, et al. A controlled trial of inpatient and outpatient geriatric evaluation and management. N Engl J Med. 2002;346(12):905–912.
Using a two-by-two factorial design, the authors performed a randomized study of both inpatient and outpatient geriatric evaluation and management (GEM) at 11 Veterans Affairs (VA) medical centers. Eleven VA sites were involved, and 1388 elderly veterans were enrolled and followed. Significant benefits were observed from the GEM unit (acute care) intervention in terms of subjects’ physical performance and basic activities of daily living at hospital discharge. Subjects in the intervention arm also demonstrated higher ratings for health-related quality of life in physical functioning, bodily pain, energy, and general health.
■ Coleman EA. Falling through the cracks: challenges and opportunities for improving transitional care for persons with continuous complex care needs. J Am Geriatr Soc. 2003;51(4):549–555.
This article discusses the challenges of managing care transitions, provides statistics on numbers of elderly patients transferred from one care venue to another, lists characteristics of effective care transitions, and proposes a research agenda for this understudied area of geriatric medicine.
■ Counsell SR, Holder CM, Liebenauer LL, et al. Effects of a multicomponent intervention on functional outcomes and process of care in hospitalized older patients: a randomized controlled trial of Acute Care for Elders (ACE) in a community hospital. J Am Geriatr Soc. 2000;48(12):1572–1581.
This study reports the second large randomized trial of Acute Care for Elders (ACE), a multicomponent unit-based intervention designed to improve functional outcomes and the process of care for hospitalized older patients. In contrast to the first randomized trial, in this trial self-reported measures of function were not found to differ at discharge between the intervention and usual-care groups by intention-to-treat analysis. Nonetheless, the composite outcome of decline in activities of daily living from baseline or nursing-home placement occurred less in the intervention group at discharge (34% versus 40%; P = .027) and during the year following hospitalization (P = .022). Nursing care plans to promote independent function were more often implemented in the intervention group (79% versus 50%; P = .001), physical therapy consults were obtained more frequently (42% versus 36%; P = .027), and restraints were applied to fewer patients (2% versus 6%; P = .001). Satisfaction with care was higher for the intervention group than for the usual-care group among patients, caregivers, physicians, and nurses (P < .05 for each group). This study provides further evidence that ACE can improve the process of hospital care for acutely ill elderly patients but also demonstrates that ACE interventions will vary among hospitals and over time in their effects on patient outcomes.
■ Milne AC, Potter J, Avenell A. Protein and energy supplementation in elderly people at risk from malnutrition (Cochrane Review). The Cochrane Library. Issue 1. Chichester, UK: John Wiley & Sons, Ltd; 2004.
This review summarizes evidence from trials of oral nutritional supplementation in elderly patients. The authors selected randomized and quasi-randomized trials, and synthesized the findings of 31 trials including 2464 subjects. Most subjects were residents of long-term care or subacute wards, but 22% were hospitalized patients. Overall, nutritional supplementation appears to produce weight gain and may reduce mortality and shorten hospital stays. There was little evidence of benefit in terms of functional outcomes.
■ Walter LC, Brand RJ, Counsell SR, et al. Development and validation of a prognostic index for one-year mortality in older adults after hospitalization. JAMA. 2001;285(23):2987–2994.
The authors developed a prognostic index in 1495 patients (mean age 81 years) discharged from the general medical service of a tertiary care hospital, then validated it in 1427 patients (mean age 79) from a community teaching hospital. The index employs a point system that assigns patients to one of four risk groups on the basis of six variables: gender, dependency in activities of daily living, presence of two comorbid conditions (heart failure or cancer), and two laboratory values drawn at admission (serum creatinine, serum albumin). The index demonstrated good discrimination and predictive ability.
William L. Lyons, MD
C. Seth Landefeld, MD