Hospitalization-Associated Disability: Title and subTitle Break“She Was Probably Able to Ambulate, but I’m Not Sure”

Ms N is a 70-year-old woman admitted to the medical service of an urban public hospital. She reported experiencing left labial pain and hematuria for 3 days. In the emergency department, she was in acute renal failure (serum creatinine, 10.8 mg/dL; potassium, 8.3 mEq/L).

She had a long history of type 2 diabetes mellitus, hypertension, chronic kidney disease (baseline creatinine, 3.5 mg/dL), coronary artery disease, peripheral vascular disease, and diabetic neuropathy.

Ms N emigrated from the Philippines in 1997 to join her husband, from whom she had since separated. Although her primary language was Tagalog, all communication with her during the hospitalization was in English. Her monthly income was $300 per month from the US Social Security Administration. One of her 6 children lives locally. Prior to admission, Ms N reported that she lived independently as a guest in a friend's home and was able to care for herself until 3 days before admission.

On admission, Ms N appeared frail, shivering with a temperature of 98.2° F (36.8° C), systolic and diastolic blood pressure of 155/42 mm Hg, pulse rate of 55 beats per minute, and a respiratory rate of 22 breaths per minute. There was a 3-cm mass in the left labium majorum that was tender and indurated. She was alert and oriented to person, place, and date and walked with a normal gait. In addition to renal failure, laboratory studies were notable for hematocrit at 19.9% and her albumin level was 3.2 g/dL.

On hospital day 1, hemodialysis was started and she received empirical treatment for a urinary tract infection. On day 3, a gynecology consultant noted a fluctuant, spontaneously draining 4 × 3 cm labial lesion that was incised and drained. Later that day, Ms N was transferred to the hospital's acute care for elders (ACE) unit. There, during the daily interdisciplinary rounds, her nurse noted Ms N had difficulty transferring from bed to commode due in part to shaking and jerking of her extremities. On the Mini-Cog test, she recalled 2 of 3 objects and drew a normal clock.

The ACE team felt her myoclonus was multifactorial and recommended discontinuation of gabapentin, which was started after admission for diabetic neuropathy. Five days later, her myoclonus had resolved and she was again independent in activities of daily living (ADLs) and walking independently using a walker. She continued receiving hemodialysis 3 times weekly. She was transferred from the ACE unit to a regular medical ward while awaiting an outpatient hemodialysis slot.

Over the next 2 weeks, she developed progressive difficulty with ambulation and experienced a slow and unsteady gait even with use of a walker. She also began to need help bathing and using the toilet. The unit social worker noted her cognitive status as “foggy” and it was felt she could not safely manage her medications or finances. She was discharged on hospital day 30 to a skilled nursing facility.

Ms N, the ACE geriatrician (Dr Z), and the inpatient social worker (Ms R) were interviewed separately by a Care of the Aging Patient editor between April and October 2010, while Ms N was in her postdischarge skilled nursing facility.

Ms N: They did all the services that I needed. They took care of my medicine, my bedding, and my wardrobe. . . . The social worker was very helpful in finding me a place to go when I left the hospital.

Dr Z (on Ms N's admission status): She was able to walk independently. She was able to take care of all of her ADLs and IADLs [instrumental activities of daily living] before she came to the hospital. She was a pretty independent person.

Ms R (on Ms N's status shortly after admission): She wasn't ambulating very much; she had some neuropathy in her legs. I think that dialysis was knocking her out a lot.

Ms N illustrates an unfortunately common scenario: an older adult living independently at home and independently managing all basic ADLs develops an acute illness requiring hospitalization. During hospitalization, the acute medical issues are treated. However, the patient is discharged with a major new disability that was not present before the onset of acute illness. The patient is no longer able to complete basic self-care activities, eg, bathing independently, and has difficulty ambulating, even with a walker.

In this article, we address hospitalization-associated disability, which includes patients (like Ms N) who develop disability for the first time when hospitalized, as well as those who were disabled before hospitalization and develop further disability when hospitalized. We review the literature on risk factors for hospitalization-associated disability, discuss interventions that may prevent it, and recommend management practices that clinicians can implement.

We conducted a literature review to determine which risk factors help identify older adults most likely to develop hospitalization-associated disability. We searched PubMed, CINAHL, and EMBASE from 1990 to 2011 using the Medical Subject Headings terms aged, hospitalization or hospitals, activities of daily living or recovery of function, and prospective studies or cohort studies or longitudinal studies or odds ratio. We did not search middle aged. We retained studies that either used multivariate analysis to assess the impact of a single risk factor or that considered multiple risk factors to develop a risk stratification tool. We focused on articles that compared ADL function at or shortly before hospital admission (within 1 month) to function at or shortly after hospital discharge (within 1 month).

Hospitalization-associated disability manifests as the loss of ability to complete 1 of the basic ADLs needed to live independently without assistance: bathing, dressing, rising from bed or a chair, using the toilet, eating, or walking across a room. Individuals who develop hospitalization-associated disability cannot successfully live without assistance; they become reliant on the help of caregivers or require long-term care.^{1 - 2} Hospitalization-associated disability develops between the onset of the acute illness and discharge from the hospital. At least 30% of patients older than 70 years and hospitalized for a medical illness are discharged with an ADL disability they did not have before becoming acutely ill.^{3 - 5} Older adults who are frail, such as those with slow gait speed, have even higher rates of hospitalization-associated disability.⁵ The acute illness is often seemingly minor, such as an uncomplicated pneumonia or exacerbation of congestive heart failure that responds to usual medical management.⁵ Quiz Ref IDApproximately 50% of disability among older adults occurs in the setting of medical hospitalization.⁶ Although Ms N's hospitalization was unusually long, rates of hospitalization-associated disability are high even after brief hospitalizations.³ One year following discharge, fewer than half of older adults have recovered to their preillness levels of functioning and rates of nursing home placement and death are high.^{7 - 8}

Although it is obvious how directly disabling illnesses such as hip fracture or stroke can lead to long-term disability, it is less obvious how Ms N's illness led to disability. Most of the acute metabolic and infectious problems that led to hospitalization showed dramatic improvement by discharge and hemodialysis should have led to an improvement in uremic symptoms. Yet her functional trajectory followed a downward spiral. This discordance between the improving trajectory of the medical diagnoses and the failure to recover functional loss occurring shortly before admission, as well as new functional loss after admission, is a common sequelae of hospitalization.⁹

Although older age disability can develop suddenly (ie, from a stroke), its onset is usually insidious,^{10 - 11} meaning an individual gradually accumulates impairments that make it more difficult to independently perform an ADL. Acute illness and hospitalization then precipitate a transition to frank disability. Before hospitalization, a patient may note that getting dressed is taking greater effort but can be done without assistance. After a hospitalization, a patient might only be able to get dressed with the assistance of a caregiver.

Hospitalization-associated disability can be understood through the paradigm of geriatric syndromes as it shares many features with other geriatric syndromes¹² such as falls,¹³ delirium, and incontinence.¹⁴ As is typical of geriatric syndromes, hospitalization-associated disability can rarely be explained by a single cause. Rather, it occurs in vulnerable older adults who accumulate impairments in multiple domains. These impairments encompass a range of vulnerabilities such as comorbid diseases, cognitive impairment, and psychosocial factors such as depression and limited social support. Precipitating events such as acute illness and hospitalization then trigger the full syndrome.^{3 ,5} Processes of hospital care including iatrogenic risks, immobility,^{15 - 16} polypharmacy,¹⁷ and lack of adaptive accommodations can both inhibit the recovery of functional loss that occurred immediately before hospitalization and lead to additional functional loss during the hospitalization.¹⁸ The Figure describes how these factors interact to increase the risk of hospitalization-associated disability and loss of independence. Because hospitalization-associated disability shares many characteristics with other geriatric syndromes, it may be considered the hospitalization disability syndrome.

Table 1 illustrates that the risk for hospitalization-associated disability is defined by a broad range of factors, as is typical for geriatric syndromes. Age is the most potent risk factor.³ More than 50% of adults older than 85 years will leave the hospital with a major new ADL disability.³ Depression and cognitive dysfunction are additional potent risk factors for hospitalization-associated disability.^{19 ,23 ,25}

Studies by Mehta et al,²⁷ Inouye et al,²⁸ and Sager et al²⁹ demonstrate that integrating information from brief multidimensional prognostic assessments identifies older adults most likely to develop hospitalization-associated disability (Table 2).^{27 - 29} Use of such risk instruments may facilitate early assessment of postdischarge needs and prevent the rushed planning that often occurs immediately before discharge, when it is discovered that patients are unable to adequately care for themselves.³⁰ Ms N had many risk factors for being disabled at the time of discharge, including limitations in activities such as doing housework and use of transportation, poor social functioning, and mobility difficulties. The Mehta index would have been particularly useful for Ms N as it focuses on older adults independent in all ADLs before the onset of the acute illness and likely would have suggested at least moderate risk for hospitalization-associated disability.

Dr Z: The most important thing is assessing what the level of function was before the illness. Was the patient independent with taking medicines and remembering when to take them? Does the patient remember appointments? Know what the level of independence is at baseline and if there is any change in function either cognitively or physically. Try to determine if the change is part of the acute illness.

Ms R (referring to postadmission functioning): She was alert . . . but having a lot of brain fog. . . . She was probably able to ambulate but I’m not sure . . . I think she was just hanging out in bed.

Quiz Ref IDAlthough hospitalization-associated disability results, in part, from the illness necessitating hospitalization, hospital processes may play a role both in inhibiting recovery of functional loss that occurred shortly before admission and in accelerating additional functional decline during hospitalization.¹⁸ For example, most hospitalized older adults spend the majority of time in bed; even short periods of bedrest accelerate muscle wasting.^{15 ,31 - 32} Hospitalization provides opportunities for iatrogenic complications, including medication adverse effects and infections, especially in patients with indwelling devices such as urinary catheters.³³ Nutritional status often deteriorates during hospitalization.³⁴ This may be partially related to excessive use of nothing by mouth orders and restrictive unpalatable diets. Many of these processes also precipitate delirium, a potent risk factor for hospitalization-associated disability.³⁵

A number of interventions have been implemented to reduce the incidence of hospitalization-associated disability. They include geriatric inpatient units (acute care of elders or ACE units), geriatric inpatient rehabilitation (also known as geriatric evaluation and management [GEM] units), geriatric inpatient consultation, Hospital at Home (HAH), and the Hospital Elder Life Program (HELP). Table 3 describes some of the outcomes associated with these models in clinical trials and meta-analyses.

The strongest evidence that redesigned care systems improve outcomes stems from ACE units.^{36 ,41 - 42} ACE provides interdisciplinary care on geriatric-focused units with protocols for prevention and rehabilitation of disability.¹⁸ Quiz Ref IDA meta-analysis suggested that ACE units reduce the incidence of functional decline at discharge and increase the likelihood of discharge to home.³⁶ ACE units can reduce hospital length of stay and the resulting cost savings may be greater than the added costs of the unit.⁴³ One study showed increased satisfaction among patients, nurses, and physicians.⁴¹ Many of the processes adopted by ACE units are also facilitated by Nurses Improving Care for Healthsystem Elders (NICHE), a program that provides resources to nurses to improve care of hospitalized older adults.⁴⁴ Ms N regained independence in ADL function during her brief stay on the ACE unit, but her independence subsequently declined after she was transferred to a regular medical floor despite the continued stabilization of her acute medical problems.

Quiz Ref IDInpatient GEM units differ from ACE units in that they usually admit patients after their acute problems are stabilized.³⁷ GEM units provide a greater focus on rehabilitation than on prevention of hospitalization-associated disability. They are also more selective in their choice of patients, with a focus on those most likely to benefit from rehabilitation. ACE and GEM units share an important similarity, which is the oversight of a multidisciplinary team and the use of comprehensive geriatric assessment. A meta-analysis suggested that GEM units increase the likelihood of functional improvement by the time of discharge and lower the need for nursing home care.³⁷ Subgroup analyses suggest that GEM units are particularly promising following hip fracture.

In contrast to these models, consultative inpatient geriatric assessment has failed to demonstrate the same effect on outcomes as the interventions in which the multidisciplinary team takes a primary role in patient care.³⁹ The lessons from ACE and GEM units suggest that the usual scenario, in which physical therapy, occupational therapy, and geriatrics are provided as isolated consultation services, is less effective than integrating all of the disciplines into the multidisciplinary team.

Some of the best evidence that hospital processes contribute to poor outcomes derives from randomized studies that reengineer these processes. A study of the HELP, which targets delirium prevention by instituting protocols to promote sleep, provides aids for sensory impairment and reorientation, promotes hydration, and promotes daytime activity, showed that delirium incidence could be reduced by one-third.³⁸

Some studies have proposed moving as much hospital care as possible to the home setting.^{40 ,45} Although evidence is limited, a nonrandomized comparison suggested that this approach is safe and may lead to modest reductions in disability.⁴⁰

Assessing and Documenting Disability, Mobility, and Cognition. By the end of her hospitalization, Ms N lost her ability to function independently and required care in a nursing home. Her nursing home care, paid for by Medicaid, resulted in substantial public expense.⁴⁶

Although Ms N's functional status was an essential measure of her clinical well-being, the medical record suggests that it was sporadically assessed during her hospitalization. Failure to assess functional status in hospitalized patients is the norm, with 1 study showing that more than 50% of major functional limitations are not documented.⁴⁷ The admission evaluation included little evaluation of functional status, and the functional course that led to Ms N's inability to return home is unclear. However, several of her caregivers noted an inability to bathe independently at discharge, along with very limited mobility.

Functional status can change frequently between the onset of the acute illness, hospitalization, and discharge. These changes in functional status are powerful prognostic tools, predicting mortality and other health outcomes during and after hospitalization. Clinicians should view functional status over the course of hospitalization as a vital sign that can help guide care and serve as a guidepost of clinical well-being. Quiz Ref IDWe suggest that the key domains that should be documented include baseline ADL function, mobility, and cognition. Although other domains such as IADLs and screening for depression may be considered, our recommended assessment is guided by a belief that by keeping the screen simple and focused, clinicians will be better able to implement the screen in practice.

Considerable information about a patient's level of functioning can be obtained without a large expenditure of time. Core measures should be assessed on admission and in daily rounds (Table 4). The assessment we recommend is not comprehensive, but is guided by pragmatism and the needs of the busy clinician. This screening has not been tested prospectively for effectiveness but the components yield a functional portrait that is prognostically useful⁴⁸ and provide a much richer portrait of the clinical condition of the older patient.

First, at admission, learn from the patient (or caregiver) about the ability to perform ADLs. Typically, before the ability to independently perform an ADL is lost, the older adult may report difficulty completing the activity.⁴⁹ Therefore, it is useful to inquire as to whether the patient has difficulty, and if so, to learn whether the activity can still be completed without assistance. To determine how the current illness has affected ADL functioning, we suggest asking about the patient's ADL functioning on the day of admission and before the onset of the acute illness.

The frequent ADL changes that occur over the course of hospitalization may be best observed by nonphysician members of the health care team such as nurses and physical and occupational therapists. These individuals often observe functional status over the full day, while physicians may be primarily observing patients during morning rounds. Regular review of notes and good interdisciplinary collaboration could help physicians better monitor this functional status vital indicator. As hospitals develop electronic medical record systems, efforts to better collate and use functional information has the potential to improve care.⁵⁰

Second, we suggest a brief mobility assessment to assess whether the patient can sit up in bed without assistance, stand, and walk a few steps. This should be done on admission and on daily rounds, as impairments in mobility are strongly associated with an increased risk of hospitalization-associated disability.²¹ Our suggested assessment of mobility, a functional neurologic assessment, takes less time than the assessment of strength, sensory function, and reflexes and usually will be more useful.

We recommend that all older patients receive a brief assessment of cognitive function on admission. The Mini-Cog, which consists of the clock draw test in addition to the 3-item delayed recall, is a brief and useful screening test—as negative screens markedly reduce the likelihood of cognitive impairment.⁵¹ The screen is considered negative if the patient can remember all 3 items or if the patient can remember 1 to 3 items and has a normal clock draw result. This classification has a sensitivity of 99% and a specificity of 93% for a diagnosis of dementia.⁵¹ Identifying dementia can help clinicians identify patients at high risk for developing delirium, an acute change in mental status marked by features such as disorientation and inattentiveness.⁵² Since an abnormal Mini-Cog could also be due to delirium, it should be followed up with a tool such as the Confusion Assessment Method (CAM) to determine if the patient has delirium. The CAM has a sensitivity of 96% and specificity of 93% in diagnosing delirium.²⁶

Like 30% of hospitalized older adults, Ms N almost certainly had undiagnosed delirium, which was described as “brain fog.”⁵² It takes little clinical skill to recognize the hyperactive form of delirium in the agitated patient. However, the hypoactive form is equally dangerous, at least as common, but much less likely to be recognized.⁵³ It is important to be alert to the possibility of delirium in the very quiet patient. During rounds, the patient may seem confused, inattentive, lethargic, or give tangential answers to questions. Review of nursing notes may reveal that mental status and alertness may increase and decrease over the course of a day.

Ms R: Like what happens with a lot of older people, she probably did very well in the ACE unit because they're specially trained to get her out of bed multiple times a day and have her up for meals, but then went to a unit where she was feeling crummy and was not encouraged to get up multiple times a day. . . . [it] was a really good unit, but they just don't have the training.

To describe steps hospitals and clinicians can consider to prevent hospitalization-associated disability, we categorized the common characteristics of ACE units by reviewing the 5 randomized controlled studies^{41 - 42 ,54 - 56} included in a meta-analysis³⁶ (Table 5).

Most inpatient geriatric units create significant environmental modifications. For example, carpeted floors are used to increase ambulation. The hospital's often loud, disorienting environment is mitigated by interventions to reduce noise and provide aids to improve orientation. Processes that restrict mobility such as excessive bed rest orders are avoided. In addition, the common practice of restricting patient mobility to avoid falls is replaced with a focus on enabling safe mobility.⁵⁷ A focus is often placed on the avoidance of functional restraints such as poles for administering intravenous medications and indwelling urinary catheters (the 1-point restraint).³³ For example, urinary catheters are seldom indicated for monitoring volume status as it is unlikely that the value of the information gained justifies the increased risks of infection and immobility.

Preventive and rehabilitative protocols are in place for each of the basic ADLs and capable patients are encouraged to continue performing each ADL independently without assistance. Patients who are able continue to bathe or dress independently and are encouraged to dress in their street clothing instead of hospital gowns. Physical and occupational therapy interventions are in place for each ADL for which patients need help.

Most inpatient geriatric units conduct daily reviews of medical care, focusing on both pharmacologic and nonpharmacologic care. Daily review of medications often focuses on reducing unnecessary psychoactive and anticholinergic medications. Dietary and no food by mouth orders are also reviewed, avoiding overly restrictive diet orders.

A hallmark of ACE units is an intensified emphasis on planning, beginning on the day of admission, for the patient's needs after leaving the hospital. Several units have replaced the concept of discharge planning (ie, getting the patient out of the hospital) with planning to go home, replacing the utilization management focus with a patient-centered focus.^{41 - 42} Planning to go home focuses on the patient's expected functional trajectory and resulting needs, as well as the needs of the caregiver(s).

What can clinicians who cannot admit patients to an acute geriatrics unit do to prevent hospitalization-associated disability? Although no studies examine the effect of any individual clinician action on the risk of hospitalization-associated disability, many of the elements of ACE unit care can be implemented by individual clinicians (Table 5). In addition, Malone et al⁵⁸ has demonstrated that many elements of ACE care can be implemented at hospitals without ACE units or geriatricians on staff. We recommend several high-priority actions that clinicians may consider to prevent hospitalization-associated disability. Although these actions have not been studied in randomized controlled trials (RCTs) as isolated interventions, many of them have been included in RCTs of multicomponent interventions.

First, better recognition of functional status and more direct efforts to communicate with other disciplines such as nursing, social work, and physical therapy can mimic some elements of interdisciplinary care. Second, clinicians should walk patients, encourage exercise if the patient is able, and encourage patients to ambulate from bed to the chair as much as possible. This should include removing any unnecessary tethers such as oxygen, bladder catheters, and intravenous catheters, creating hospital norms that patients are expected to walk regularly if they are able, and requesting physical therapy consultation when assistance may be helpful. Third, make sure that patients are eating adequately. This should include attention to actual intake, assurance that food is available when the patients are on the ward, availability of food from home, and minimization of no food by mouth and restrictive diet orders. Fourth, sedative hypnotics and antihistamines should be avoided, including at bedtime. Safer nonpharmacologic protocols to aid patients with difficulty sleeping should be considered.³⁸

Disability associated with hospitalization is a sentinel event that often has profound effects on the patient and family long after hospital discharge. Whether patients will be able to live at home will depend not only on their capacity but also on their social support, resources, and environment.⁵⁹

Planning for a patient's return to home should include assessing whether the patient can successfully accomplish ADLs alone or with available support and recognizing that many patients able to perform these activities before acute illness may be unable to perform them upon returning home. For a patient like Ms N, who needed help bathing, had very limited mobility, and whose cognitive limitations necessitated help taking her medications, it is important to specifically assess how these tasks will be completed and whether temporary home assistance or equipment could help the patient to return home. For example, could providing support bars and a shower chair restore independence in bathing? Does the patient need only supervision while bathing or is an aide needed to help the patient in and out of the shower? A matter as simple as the number of stairs the patient must scale may make going home unsafe. Although Ms N was able to walk short distances with a walker, this level of disability was made more serious by her need to reliably use transportation to attend dialysis. Patients dealing with new disabilities may benefit from a home equipment evaluation, completed by a physical or occupational therapist visiting the home, a service covered by the Medicare Skilled Home Care benefit.⁶⁰

Ms N's delirium was another important consideration in the assessment of her ability to function at home. It is increasingly recognized that delirium can be stubbornly persistent and often does not resolve by discharge.⁶¹ Although efforts to reconcile medications and instruct patients in their use at discharge are laudable, they may be futile in delirious patients. The patient's ability to understand medication instructions should be assessed and caregivers may need to be included in discharge education. Assistance with medicines is a skilled nursing need that will justify Medicare payment of postdischarge home nursing support.⁶⁰ Beyond medication concerns, cognitive concerns will also effect the ability to manage finances.⁶²

Finally, for Ms N, poverty and her lack of social support was another precipitant of loss of independence.⁶³ Others with Ms N's level of disability, but greater financial resources, likely would have been able to pay for home health aide assistance that would have made staying at home possible.

The care of Ms N illustrates the importance of considering the ability of the patient's caregivers.⁶⁴ Family caregiving often comes at significant physical, financial, and emotional costs.⁶⁵ Caregivers often give up employment to provide care and report very high rates of depression.⁶⁶ This patient's friend and daughter had apparently reached their respective limits. Considering Ms N's capacity in the context of her support system and resources made returning home unsafe.

Hospitalization-associated disability has a poor prognosis. Over the ensuing year, many will make transitions in and out of disability.⁶⁷ In a study of older adults who had developed hospitalization-associated disability, 41% died by 1 year, 29% remained disabled at 1 year, and only 30% returned to their preillness level of function.⁷ These outcomes, for what may often appear to be nondisabling acute illness episodes, are similar to those reported for older adults hospitalized for hip fracture.⁶⁸

Prognostic stratification can be challenging, but assessment of prognosis at discharge may be useful in guiding care and counseling patients and families. The Walter Prognostic Index can be used to estimate prognosis in hospitalized patients at discharge (Table 6).⁴⁸ Ms N's Walter score was 5 (2 points for dependence in 1 ADL, 2 for a high creatinine level, and 1 for low albumin), suggesting a 34% 1-year mortality risk. Her prognosis for return to independent functioning was poor, with the generally low likelihood of functional recovery made worse by the need for dialysis and nursing home placement.^{7 ,69} One study suggests that nursing home residents who initiate dialysis have accelerated functional loss after dialysis initiation and seldom improve in ADL functioning within 1 year of dialysis.⁶⁹

At the time of hospital discharge, in a newly disabled older patient, it is useful to consider to what extent the disability is a reversible manifestation of the acute illness and the deconditioning effects of hospitalization, vs a manifestation of ongoing progressive disease that may be leading toward the end of life. In most cases, it will be difficult to fully answer this question until the posthospitalization course is revealed. There is little research to determine which patients with hospitalization-associated disability will benefit from rehabilitation. In patients in whom restoration of function is important, a trial of outpatient or inpatient physical therapy and rehabilitation seems reasonable.

The often poor prognosis for hospitalization-associated disability suggests that simultaneous pursuit of both rehabilitative care and palliative approaches be part of the care of patients who develop disability. For example, for Ms N, would equipment such as a motorized scooter have made life easier? Would home modifications have been helpful? What could have been done to assist her caregivers? Would community resources such as meals on wheels or paratransit services have been useful? Which of her symptoms needed treatment? Disabled older adults often have unrecognized and untreated pain, which goes unnoticed as the hospital discharge plan is formulated because it may be unrelated to the acute illness. For example, in the last 2 years of life, arthritis is a stronger determinant of pain than the diagnosis leading to death.⁷⁰

Unfortunately, the palliative needs of patients are often not considered unless a patient is referred for hospice care, which in practice usually occurs in the final days of life.⁷¹ However, palliative care is appropriate for all patients with serious, incurable chronic illnesses.⁷² For example, Ms N would not meet traditional hospice criteria, and might benefit from both life-prolonging therapies such as dialysis as well as aggressive symptom management. It is a common mistake to believe that patients must choose between traditional medical care and palliative care.

Hospitalization-associated disability is an unfortunately common occurrence in hospitalized older adults with significant consequences for patients and caregivers. Clinicians should supplement the traditional disease-focused approach to hospital care with one that recognizes that changes in functional status are a clinical vital sign and the most important manifestation of illness in older adults across admitting diagnoses. Reengineered hospital care that focuses on function, including assessment on admission and throughout the hospital stay, promoting physical activity, avoiding hospital processes and complications that impair functional recovery, and planning for discharge home with the support needed to complement a patient's functional capacity, may reduce the incidence of hospitalization-associated disability. Adoption of models of care that can reduce hospitalization-associated disability should be a high priority for hospitals and clinicians caring for older patients.

Ms N was unable to return home and 6 months following her hospitalization was still living in a nursing home. However, she was adjusting well to her dialysis and her new home situation. At her most recent outpatient primary care clinic visit, Ms N reported only some dizziness following hemodialysis. Her arteriovenous fistula had matured and was ready for use. She reported good appetite and stable weight.