Reasons for Admission for Geriatric Patients
The major diagnoses for which older adults are hospitalized are related to chronic diseases and respiratory conditions. The 15 most common conditions, accounting for 48% of the hospital admission diagnoses, are listed in Table 17-2. Also common and important to recognize, but less likely to be reported as the reason for admission, are conditions more likely to occur in older adults such as failure to thrive, falls, adverse drug effects, or change in mental status. In addition, older adults maybe admitted with an atypical presentation of another condition, such as when change in mental status is due to underlying fluid and electrolyte disorder or urinary tract infection (UTI). Often the reported diagnosis for a hospitalized older patient may not fully capture the underlying reasons that necessitated the admission and does not explain the hospital course and subsequent health status of the patient. While many of the 15 most frequent conditions reported as causes for hospitalizations among older adults represent acute exacerbations of chronic diseases, the reasons why a stable older adult with heart failure suddenly decompensates or a 90-year-old assisted living resident is admitted with a broken hip, often relate as much to the physical and/or social vulnerability of many older adults as to their complex health status.
Table 17-2 Most Frequent Conditions Causing Hospitalizations among Older Patients, 2003 ||Download (.pdf)
Table 17-2 Most Frequent Conditions Causing Hospitalizations among Older Patients, 2003
% OF ALL HOSPITALIZATIONS IN OLDER ADULTS
Acute myocardial infarction
Chronic obstructive pulmonary disease
Rehabilitation care, fitting prostheses, adjustment of devices
Fluid and electrolyte disorders
Urinary tract infection
Complication of medical device, implant, or graft
Total admissions for top 15 conditions
In addition to the primary problems that led to the admission, the effect of comorbid chronic diseases must be considered. Over 60% of Medicare patients have two or more major chronic diseases and 24% have four or more. In 2004, people 65 years and older admitted to the hospital had an average of 2.3 comorbid conditions. Comorbid chronic diseases have several consequences for the hospitalized elder and for the clinician. Multiple diseases often mean multiple outpatient physicians, complicating communication between inpatient and outpatient providers. Multiple diseases lead to the use of multiple medications. Multiple medications, even if indicated, can result in confusion about medications, difficulty with medication reconciliation and drug adherence, and adverse drug events (ADEs) including amplified side effects, drug–drug or drug–disease interactions, and errors in drug administration. A high burden of chronic disease can lead to self-care difficulties, patient and caregiver frustration and burnout, and physiological instability for the patient.
In older patients, especially those 75 years and older, common conditions such as vision or hearing impairment, mobility impairment and fall risk, poor nutrition, incontinence, depression, cognitive impairment, and functional impairment often occur in conjunction with the major chronic diseases that lead to hospital admissions. For example, 2004 data from the Health and Retirement Survey (HRS), a nationally representative health interview survey sponsored by the National Institute on Aging demonstrate that the geriatric conditions of falls and incontinence are common in older adults with heart failure, coronary heart diseases, and diabetes (Table 17-3).
Table 17-3 Proportion of Respondents Aged 65 Yrs and Older with Index Disease or Condition Who Have Other Diseases/Conditions* ||Download (.pdf)
Table 17-3 Proportion of Respondents Aged 65 Yrs and Older with Index Disease or Condition Who Have Other Diseases/Conditions*
INDEX DISEASE OR CONDITIONS (% OF TOTAL SAMPLE)
≥1 Other Condition
≥2 Other Conditions
Coronary Disease (8.7%)
Heart Failure (4.8%)
Conditions commonly seen in older patients often labeled as “geriatric” conditions can contribute to the need for the acute admission, and substantially influence the hospital course and discharge plans. Cognitive impairment, one such geriatric condition, is a major risk for delirium. At admission, it may be impossible to distinguish between delirium and dementia or to determine a patient's baseline cognitive performance. Delirium is associated with longer hospital length of stay, greater functional disability, and increased mortality following hospitalization. Several prospective studies have documented that hospital mortality is related to nutritional status, cognitive dysfunction, and functional disability. These factors independently predict mortality even when the comorbid diseases and diseases leading to hospitalization are considered.
Often physicians caring for an acutely ill, unstable or unsafe older adult who requires acute hospitalization are advised to identify or “screen” for frailty. There is as yet no universally accepted definition or measure of frailty. The term “frail” tends to be used to refer to an older adult who is physiologically or socially vulnerable. Recently, researchers have tried to develop empiric definitions of frailty, which, while promising, may not yet be clinically applicable. The idea of a vulnerable or at-risk elderly person may be clinically more helpful. The presence of comorbid conditions, functional decline, cognitive impairment, or inadequate or abusive social situations suggests vulnerability. Since functional decline and cognitive impairments increase with age, the advanced age of a patient (e.g., >75 years) may strongly factor into a clinician's decision to “screen” for risk of frailty and perform screens of memory, functional status, hearing, and sight, and take an in-depth social history during the hospitalization. These issues require attention from the admission process through discharge planning and postacute care transitions.
Geriatric conditions can be the reason for admission. Examples include falls or “failure to thrive,” defined as poor nutrition and weight loss associated with diseases, dementia, functional disability, and sometimes inadequate caregiving. Assessment of comorbidities and geriatric conditions (e.g., falls, failure to thrive, dementia, urinary incontinence) upon admission and during the hospitalization with simple screening questions and physical evaluation will help the clinician operationalize the ideas of “vulnerable” and frail, and provide targets for therapy such as increased nutrition, physical therapy, glasses, and hearing aids. Social issues, such as inability to buy medications, inadequate caregiving and/or insufficient help at home, elder neglect and abuse, and elder self-neglect are unfortunately relatively common. They can be missed if they are not specifically considered and investigated.
At the time of admission, much of the focus is on evaluation and management of a disease-specific, perhaps life-threatening illness. However, the admission also provides an opportune time to screen for issues of importance in the care of elderly patients, particularly issues likely to affect the course, treatment, and prognosis of the illness that precipitated the hospitalization. Important components of the admission screen are described.
Communication with Family and Primary Care Providers
While many patients can provide accurate descriptions of their home situations and presenting symptoms, every effort should be made to discuss these issues with family members who can often provide additional information about social issues that may have contributed to admission or who may describe symptoms or events that help clarify the admitting diagnosis. Older patients often present with complex symptoms and atypical presentations of disease requiring increased attention to the factors that led up to the admission. Similarly, the patient's primary outpatient provider of care should be contacted. It is increasingly common for patients to be cared for by inpatient physicians, such as hospitalists, who do not care for patients in the outpatient setting. This means many elderly patients are being “handed-off” at admission. Studies have shown that most primary care providers want to be contacted at admission. Primary providers can help clarify the acute problem and can provide a more complete medical history including other comorbid conditions, previous diagnostic testing, and response to previous treatments. Depending on the reason for admission and the hospital course, communication with the patient's specialist physicians may also be necessary.
Hospital admission is an important time for medication review. Clarification of the patient's medications, often prescribed by multiple physicians, and identification of potential adverse drug reactions (ADRs) are two important aspects of medication review. ADRs may lead to hospital admission and are more common as numbers of medications and comorbid illness increase. While age alone is not an independent predictor of ADRs, older patients are more likely to have multiple comorbid conditions and be on multiple medications. In addition, there are certain medications or classes of medications that have been identified by expert consensus panels as being at high risk for ADRs in elderly patients with relatively low clinical benefit and often a safer alternative medication exists. These high-risk medications should likewise be identified and discussed with the primary care physician. Careful attention should be paid to medications most likely to lead to ADRs including analgesics, sedatives, cardiovascular medications, and psychoactive drugs. Regulatory bodies, such as the Joint Commission for Accreditation of Healthcare Organizations (JCAHO), are requiring that all institutions have a process in place to reconcile, or review, medication lists for high-risk or unnecessary medications. This process needs to happen throughout the hospitalization, but is especially important at admission. Medication reconciliation is a time-consuming, but important process involving a rigorous review of each medication for appropriateness in conjunction with comorbidities and the other medications the patient is taking and requires discussion with the primary care physician and, often, specialty physicians at the point of discharge for continuity of care, best drug choices, and safety. Aspects of medication reconciliation can be aided by computerized physician order entry systems and clinical pharmacists.
While precise definitions of frailty are elusive, studies have shown that patients of advanced age (e.g., >80 years) or with functional impairments are the most vulnerable and should be considered “frail.” At least one survey of a general medicine service in an urban medical center estimated 25% of elderly patients were frail or vulnerable, according to the Vulnerable Elderly Survey-13 tool that scores age, self-perceived health, and aspects of functional status to predict increased risk of morbidity and mortality. Frailty puts patients at risk for further functional and cognitive decline, delirium, prolonged hospitalization, increased costs, and mortality. Identification of frailty at admission should alert the hospital physician to the need to further evaluate for dementia and other geriatric conditions and can help frame discussions about prognosis. It also signals the need to start advanced discharge planning.
Functional measures are stronger predictors of mortality and contribute more to prognosis in the hospitalized older patient than comorbid illness, disease severity, and diagnoses. Assessing activities of daily living (ADLs) and instrumental activities of daily living (IADLs) are well-known measures of functional impairment. The hospital physician should also be comfortable in performing routine assessments of mobility, such as the “Get Up and Go” test (see Chapter 115). Any documented mobility or ADL impairment should trigger physical therapy and/or occupational therapy assessments and should signal the need to institute early mobilization and early institution of discharge planning.
Screening for dementia is particularly important in the elderly patient who is losing weight, noncompliant with medications, admitted from a nursing home or readmitted to the hospital. Impaired judgment and insight can impact a patient's ability to make health decisions, discuss end-of-life issues, and live independently after discharge. It also identifies patients at risk for the development of delirium in the hospital and readmission after discharge. While diagnosis of dementia is based on DSM-IV criteria, two common screening tools, the Mini-Mental Status Examination (MMSE) and the Mini-Cog (see Chapters 11 and 12), can be used to quickly identify patients at high risk for dementia. Both tests have similar sensitivities, but the MMSE is the only one to have been validated in the hospital setting, while the Mini-Cog is faster to perform. Impairments on either test should result in active planning for cognitive stimulation during the hospitalization, comprehensive discharge planning instituted at the beginning of the hospital stay, and family/caregiver involvement.
Hospitalization presents many hazards for older patients. While the outcome of hospitalization is dependent on the severity and type of acute illness and the patient's baseline vulnerability, elderly patients are at five times increased risk for iatrogenic complications during hospitalization. Older patients have an average 35% risk of functional decline during acute hospitalization. In addition, they are at increased risk for the development of delirium. After discharge, they are at increased risk for needing institutionalization and hospital readmission. While the hospital may be considered “unsafe” for vulnerable elderly patients, hospitalization is often necessary to treat acute illness. Thus, considerable attention must be given to creating a systematic approach to preventing and treating common hospital complications in the geriatric population. This topic has appropriately generated considerable attention recently among researchers, payors, regulatory bodies such as JCAHO and the federal government, as well as patients and their families.
Common Problems in Hospitalized Elderly Patients
The incidence of delirium in hospitalized older patients is as high as 50% and is associated with increased mortality, hospital length of stay, and need for placement in long-term care. Because delirium in elderly patients can present atypically (such as the hypoactive form), it often goes unrecognized by physicians and nurses. Understanding risk factors, making the diagnosis, and instituting strategies for prevention of delirium are critical for the hospital physician.
In a study group of patients >70 years old, risk factors for delirium included severe illness, cognitive impairment (MMSE<24), and BUN / Cr ratio ≥ 18. Precipitating factors for delirium were use of restraints or a bladder catheter, ≥ 3 medications added, an iatrogenic event, and malnutrition (see Chapter 53). Patients at risk for delirium should have targeted strategies to prevent its development. Many institutions have put formal delirium prevention programs into place. These programs are designed to prevent cognitive impairment, sleep deprivation, immobility, dehydration, as well as vision and hearing impairment. The Hospital Elder Life Program (HELP) uses an orientation board and a program of cognitive stimulation to reduce the rate of confusion from 26% to 8%. Even in the absence of a formal program, the hospital physician should have the patient's family stay overnight if possible, remove unnecessary foley catheters, avoid restraints, eliminate unnecessary medications, order early mobilization and visual / hearing aids, and address dehydration. While attention to these issues by individual physicians is important, the case can also be made for a more systematic, interdisciplinary approach.
Over a 6-month period, a community hospital was able to demonstrate a 14% reduction in the rate of delirium and cost savings of over $600,000 for a 40-bed unit by implementing portions of the HELP program. The program showed sustained results including higher nursing and patient satisfaction. This program succeeded despite limited resources that required them to eliminate portions of the original HELP program.
During an acute hospital stay, the older patient is at high-risk for falls. These inpatient falls are not only common, but carry significant risk of short- and long-term adverse effects for the frail elderly patient. Estimates for inpatient falls among all hospitalized patients range from two to seven falls per 1000 patient-days. However, not all hospitalized patients face the same risk. In a single urban academic center, rates for medical patients were significantly higher (6.2 falls/1000 patient days) as compared to surgical patients (2.18 falls/1000 patient days).
Inpatient falls are frequently associated with injury, with estimates of one-third to almost one-half of falls resulting in injuries. Most concerning is that an estimated 8% of falls result in moderate to severe injuries. These falls and injuries are associated with significant in-hospital adverse outcomes. Patients who suffer falls with injury have longer lengths of stay and higher costs than similar patients who do not suffer a fall. In addition to the in-hospital effects, a fall with injury may lead to serious long-term health outcomes as well. Falls among elderly patients with and without injuries are risk factors for increased use of health care resources in the future, functional decline, loss of independence, higher rates of discharge to extended care facilities, and even death.
While all elderly patients are at risk for falling, as the risk of falls increases with age, in hospitalized patients, there are multiple well-described risk factors that can better identify patients at the highest risk (Table 17-4). These risk factors include patient-specific baseline characteristics, patient-specific effects of the acute illness superimposed on baseline risk, and environmental factors. As may be expected, multiple risk factors increase the risk of falling for patients. In one study, elderly patients with four risk factors had a 10-fold increased risk of falling as compared to those with no risk factors. Although this study was conducted in community dwelling elderly persons, it is likely that this increase in fall risk with each additional risk factor is similar for hospitalized elderly patients as well.
Table 17-4 Risk Factors for Falling among Hospitalized Patients ||Download (.pdf)
Table 17-4 Risk Factors for Falling among Hospitalized Patients
History of falls
Difficulty in balance, transfers, or walking
Use of sedatives/hypnotics
Lower extremity muscle weakness
High patient to nurse ratio
Given the well-described risk factors for falls among hospitalized elderly patients, a tool for identifying those patients at highest risk would prove very useful. Several tools designed to screen for risk of falls have been developed and are easy to use. However, these currently available tools have been found to have significant limitations when evaluated outside of the initial population used for development of the instruments. The major limitation is that the tools often have very high sensitivities and low specificities so that they tend to identify a large percentage of patients as high-risk. Understanding this limitation of risk assessment tools is important as a high sensitivity/low specificity tool is a reasonable method of screening for at-risk patients. The Morse Fall Scale, consisting of questions addressing history of falls, presence of multiple diagnoses, use of ambulatory aids, presence of intravenous therapy or heparin lock, gait and transfer ability, and mental status and the St. Thomas's Risk Assessment Tool in Falling Elderly Inpatients (STRATIFY) instrument, consisting of five items that address risk factors for falling: history of falling, patient agitation, visual impairment affecting everyday function, need for frequent toileting, and transfer ability and mobility are two of the most commonly used falls screens in the acute hospital setting.
While all clinicians caring for elderly hospitalized patients should be aware of the risks and implications of falls, systematic interventions have proven most successful to date in lowering the risk of falls. These systematic interventions rely on identifying at-risk patients using baseline risk, preventing or minimizing the effects of the acute illness/injury on risk of falls, and affecting environmental changes that limit the falls as well as reducing the risk of injury from any falls that do occur. Reductions in total in-hospital falls have been reduced by as much as 19% with an even more dramatic 77% reduction in falls with injuries with reductions sustained for as long as 2 years. These interventions have utilized a multipronged approach and it is unclear as to what aspects of the programs are most effective. It is important to note that not all published reports have been able to demonstrate this level of effectiveness. While additional work is required to better understand the components of an effective falls and injury prevention program, it is appropriate for hospitals and clinicians caring for elderly patients in the acute care setting to develop local programs modeled on successful programs and then to evaluate their programs locally.
Poor nutritional status is common among hospitalized elderly patients. Studies have estimated that up to 50% of all hospitalized older patients are nutritionally at risk and up to 25% meet criteria for malnourished. The causes of these nutritional deficiencies are usually multiple and are related to issues including chronic host issues, effects of the acute illness, and environmental factors prior to hospitalization. Among hospitalized older patients, poor nutritional status is associated with worse clinical outcomes than matched nutritionally intact patients. During the hospitalization, nutritionally at-risk patients are more likely to suffer from a hospital-acquired complication, particularly infection. Compared with patients with good nutritional status, poor nutritional status is associated with longer length of stay, higher readmission rates, increased likelihood of being discharged to an extended care facility, and higher mortality rates.
Given the prevalence and importance of poor nutritional status in the older hospitalized patient, strategies aimed at early identification and intervention should be instituted. Several screening tools have been developed included the Chandra scale, the Nutrition Screening Initiative, and the Mini Nutritional Assessment. The Mini Nutritional Assessment has been shown to be predictive for in-hospital mortality, longer length of stay, and greater likelihood to be discharged to a long-term care facility. None of these instruments should be used as diagnostic tools; they are reasonable to use as part of a system of nutritional screening that can be followed up by a more thorough assessment.
Interventions on at-risk patients can be effective at improving nutritional markers. These interventions include dietary counseling, oral supplementation, or enteral feeding in select populations. When possible, oral feeding is the optimal method for nutritional repletion and maintenance. While these interventions can improve nutritional markers, there is less clear evidence that they affect important clinical outcomes. Despite this lack of evidence in the hospitalized older patient, it is reasonable to provide these nutritional interventions given the positive effects seen on intermediate outcomes.
A difficult situation that often arises in the hospital setting is the question of whether to place a gastrostomy tube (or equivalent) for long-term nutritional support. These feeding tubes do have a role in patients who have reversible deficits in oral feeding ability (e.g., postesophageal surgery) and in patients who are cognitively intact but may have a long-term need for enteral feedings due to mechanical issues. However, often the decision in the acute hospital setting involves older patients who have underlying cognitive deficits, most often dementia, who are no longer able to maintain adequate oral nutrition even with oral supplements. It is important to recognize that this is a marker of end-stage dementia and that there is harm associated with placing feeding tubes in these patients with no benefit to cognitive or functional status and can negatively impact quality of life. This can be a difficult decision for family members to make and can be a trigger to begin or continue discussions about goals of care in these patients.
Preventing Nosocomial Infections
Infections are one of the most common adverse events in health care, affecting over 2 million people, contributing to over 90 000 deaths, and costing over $4.5 billion yearly in the United States. Up to 10% of patients admitted to acute care hospitals develop hospital-acquired infections; this rate is rising. The majority of hospital-acquired infections are due to UTIs, pneumonia, surgical site infections, and bloodstream infections. A full discussion of the management of all these infections is beyond the scope of this chapter, but older patients are disproportionately affected by UTIs, pneumonia, and increasingly, Clostridium difficile infections. Putting strategies in place to address these infections (and in particular, preventing them) and practicing good infection control is critical for inpatient physicians focused on the care of elderly patients.
Hospital-Acquired Urinary Tract Infections
UTI is the most common hospital acquired infection, accounting for up to 40% of all nosocomial infections. Hospitals acquired UTIs lead to bacteremia in up to 4% of patients, and carry a mortality rate between 15% and 30%. Females, older patients, and patients with severe underlying illness are at greatest risk. Urinary catheterization is a contributing factor in the majority of infections in hospitalized patients. The National Nosocomial Infection Surveillance (NNIS) System reports a median rate of catheter associated UTI per 1000 catheter days of 3.9. While classic UTIs present with dysuria, frequent urination, and urgency, these symptoms are less common in older patients, and are difficult to evaluate in the patient with a catheter in place or with cognitive impairment. In addition, asymptomatic bacteriuria in hospitalized patients is a common finding. Its significance in patients with a catheter remains unknown, and most experts recommend against treatment in the absence of symptoms. When symptomatic, most cases of UTI in the hospital are caused by Escherichia coli, Enterococcus, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Candida species.
Most of the focus on preventing nosocomial UTIs is on mitigating the risk imposed by urinary catheters. Indwelling urinary catheter (IUC) use varies by hospital unit and patient type, but they are used in up to 25% of elderly hospitalized patients. In addition to contributing to infection risk, catheters are known to increase the risk of delirium and falls and have been referred to as a “one-point restraint.” While systematic reviews have shown that some types of antimicrobial catheters (nitrofurazone or silver alloy coated) reduce rates of bacteriuria in hospitalized patients, most effort has been spent on trying to avoid catheters in the first place. Appropriate use of catheters includes patients unable to void, postanesthesia use, monitoring of urine output in patients unable to comply with collection, protection of an open wound in an incontinent patient, and use in palliative care. Unnecessary use is common; some reports have suggested that up to 30% of hospital physicians are unaware that their patients even have a catheter. One of the most important risk prevention strategies is avoiding unnecessary catheter use. In institutions with computerized order entry, clinical decision support, electronic reminders, and automatic stop orders have been effective at reducing the overall use and duration of use of urinary catheters. Paper reminders are also effective in institutions lacking computerized order entry. In male patients who require a urinary catheter (other than for urinary retention), recent work has shown that condom catheters in addition to being rated as more comfortable, reduce the composite risk of bacteriuria, symptomatic UTIs, and death. Of note, most of the benefit was in the reduction of asymptomatic bacteriuria. This protective effect of condom catheters maybe diminished in patients with dementia who frequently pull off their catheters. Each time the condom catheter is replaced, swabbing of the meatus is required which may increase the risk of bacteriuria and subsequent infection when it occurs repeatedly.
Hospital-acquired pneumonia, also referred to as nosocomial pneumonia, includes both ventilator-associated pneumonia and nonventilator-associated pneumonia that develops 48 or more hours after hospitalization. Almost all the data on hospital-acquired pneumonia comes from intensive care units (ICUs) and the study of ventilator-associated pneumonia. Extrapolation to the non-ICU population may not be appropriate. A third class of nosocomial pneumonia that is increasingly common and important for the geriatric population is health care-associated pneumonia. Patients at risk for health care-associated pneumonia include patients receiving home IV antibiotics, home nursing or home wound care, residence in a nursing home or long-term care facility, patients who have been hospitalized for ≥2 days in the past 90 days, and patients who have received dialysis or IV therapy at a hospital-based clinic in the past 30 days. Health care associated pneumonia develops in the community, but has a spectrum of causative organisms, and an approach to management that is similar to hospital-acquired pneumonia.
Pneumonia is the second most common nosocomial infection with an incidence of 5 to 10 cases per 1000 hospitalizations. The rates in patients with endotracheal tubes are up to 20 times higher. The mortality attributable to nosocomial pneumonia is debated, but maybe as high as 30%. An episode of nosocomial pneumonia clearly increases hospital length of stay and costs.
Nosocomial pneumonia results from microbial invasion of sterile lung parenchyma as a result of microaspiration of contaminated oropharyngeal or gastric secretions. A defect in host defenses, aspiration of a large inoculum of organisms, or aspiration of a particularly virulent organism may contribute to parenchymal infection. Risk factors for the development of nosocomial pneumonia that are commonly found in older patients are listed in Table 17-5. Understanding the risk factors is critical to implementing effective prevention strategies. Many prevention strategies for ventilator-associated pneumonia have been well-described elsewhere and are best directed at the ICU physicians and other ICU personnel. However, there are several important strategies to prevent nosocomial pneumonia that are important for physicians caring for hospitalized elders in the non-ICU setting.
Table 17-5 Risk Factors for Nosocomial Pneumonia in Elderly Patients ||Download (.pdf)
Table 17-5 Risk Factors for Nosocomial Pneumonia in Elderly Patients
Impaired host defenses/increased aspiration
Enteral feeding tubes
Impaired mental status
Large inoculum of organisms
Gastric alkalinization (enteral feeds/H2-receptor blockers, proton pump inhibitors)
Iatrogenic (forced hand ventilation)
Contaminated respiratory equipment
Overgrowth of virulent organisms
Prolonged antibiotic use
Iatrogenic (inadequate hand washing)
Central venous lines
Prolonged hospital stays
One of the best ways to prevent pneumonia is to avoid intubation in patients with respiratory failure. Increasingly, the use of noninvasive ventilation has been effective in reducing the need for intubation. In selected patients with acute COPD or heart failure exacerbations, noninvasive ventilation reduced the need for intubation and was associated with less pneumonia, shorter hospital length of stay, and in some studies, lower mortality. Patients with facial trauma and altered mental status are generally not recommended for noninvasive ventilation due to increased risk of complications such as aspiration.
Keeping patients upright, or semirecumbent, has been shown to reduce rates of pneumonia. In one study, elevating the head of the bed >45° reduced rates of ventilator-acquired pneumonia significantly. While little data exist for this practice outside the ICU, head of bed elevation is likely to reduce episodes of hospital-acquired pneumonia outside the ICU, especially in patients at risk for aspiration. Stress ulcer prophylaxis increases gastric pH and allows for colonization of the gastrointestinal tract by potentially pathogenic organisms. Studies have suggested that use of medications that raise gastric pH (primarily H2-antagonists) is associated with an increased risk of pneumonia. While these medications are necessary in some patients, they are likely overused and may contribute to episodes of hospital-acquired pneumonia. In general, the only patients shown to benefit from stress ulcer prophylaxis are those with shock, respiratory failure, and coagulopathy. Data on risks for the use of proton pump inhibitors (PPI) in hospitalized patients is limited. However, studies in the community setting link these agents to increased rates of community-acquired pneumonia suggesting similar mechanisms. In the hospitalized elder, every effort should be made to assure an appropriate indication exists for continued use of H2-antagonists or PPIs. Given the role of colonizing oropharyngeal bacteria, it would seem that improving oral hygiene could prevent nosocomial pneumonia. While no study has been performed in hospitalized patients, improved oral hygiene in nursing home residents—by the use of antiseptic mouthwash, brushing teeth after meals, and weekly plaque removal—has been linked to reduced rates of aspiration pneumonia. Improved oral hygiene in hospitalized elderly patients (mouthwash and brushing), is low cost, can involve the family in resource-constrained environments, and may prevent pneumonia.
Clostridium difficile is the most common cause of health care–acquired diarrhea, occurring at a rate of 61 cases per 100 000 discharges in 2003. In the past few years, several outbreaks in hospitals and nursing homes have been described. These outbreaks have been associated with high morbidity and mortality. Concurrent with recent outbreaks are reports of poor response to therapy, and high rates of relapse. Many of the epidemic outbreaks that have been associated with highly morbid disease have been linked to a previously uncommon strain, B1/NAP1.
The biggest risk factors for infection include antibiotic use and hospitalization. Additional risk factors relevant to the care of a geriatric population include advanced age, severity of illness, use of proton pump inhibitors, and use of tube feeding. While clindamycin is classically associated with C. difficile, any antibiotic (including metronidazole and vancomycin used to treat the disease) can cause infection. Cephalosporins are widely used and commonly cause C. difficile infection. More recently, fluoroquinolone use has been associated with outbreaks of the B1/NAP1 strain.
Prevention of C. difficile requires both infection control measures and antibiotic stewardship. Outbreaks have been successfully controlled by patient isolation in a single room, implementation of contact precautions (gown and gloves for anyone entering the room) and, because clostridia spores are not easily killed by alcohol-based hand rubs, liberal handwashing with soap and water for all persons entering and leaving the patient's room. The implementation of formal antibiotic rotation programs and restriction of particular antibiotics during outbreaks may also be required to combat C. difficile.
Prevention of hospital-acquired infections also includes careful attention to hand hygiene and preventing the spread of antimicrobial resistance. With careful hand washing before and after patient contact, the incidence of nosocomial infection is reduced. Existing data suggest that plain soap and water, antibacterial soaps (chlorhexidine based), and alcohol-based hand rinses have equivalent effect. The use of alcohol-based rinses in bedside dispensers may improve compliance with hand hygiene recommendations.
The numbers of organisms in U.S. hospitals displaying resistance and the mechanisms by which they develop resistance have both increased dramatically in recent years. Particular problem organisms include among the gram positives, methicillin-resistant Staphylococcus aureus (MRSA), and vancomycin-resistant enterococcus (VRE), and among the gram negatives, drug-resistant Pseudomonas aeruginosa, Acinetobacter baumannii, and extended spectrum beta-lactamase-producing Enterobacteriaceae. Infections with these organisms are associated with increased morbidity, mortality, and costs. Physicians should be familiar with their institution's resistance profile and understand when infection with these organisms is possible in order to give timely, appropriate antibiotic therapy. Delay in the initiation of adequate antibiotic therapy (i.e., antibiotics that are active against the organism that is ultimately isolated) has been associated with increased mortality in many types of infection.
Preventing the development and spread of these organisms is critical. The use and misuse of antibiotics has been associated with the development of resistance. Considerable attention has been given to antibiotic stewardship programs that focus on development of appropriate antibiotic formularies and treatment recommendations, restricted use of certain agents, and feedback to providers on prescribing practices. Many of these programs have been effective in controlling the emergence of resistance and in stopping outbreaks of highly resistant organisms. Prevention of the transmission of these organisms once they emerge is also critical. Preventing person-to-person spread has been most successful with gram-positive organisms. Contact precautions (i.e., gowns, gloves) and isolation of infected patients in private rooms are the usual strategies. Caution is warranted when caring for elderly patients who are placed in isolation. Studies have shown that patients in isolation are less likely to be examined by health care providers. This potentially places these patients at increased risk of a missed diagnosis. It can also mean less physical therapy, less contact with nursing and family, and the potential for development of delirium. Extra effort may be required to continue the best practices of geriatric care for elderly patients who are in contact isolation.
Venous Thromboembolism Prophylaxis
Prevention of venous thromboembolism (VTE), which includes deep venous thrombosis (DVT) and pulmonary embolism (PE), is an important consideration in hospitalized patients, leading to serious morbidity and mortality. Venous thromboembolism, while not uncommon in hospitalized medical patients, is often unrecognized. While there is no consensus on quantifying patient-specific risk factors, patients admitted with heart failure, severe respiratory disease, or malignancy, patients confined to bed, and older patients are considered to be at increased risk for VTE. Multiple studies, including several meta-analyses, have demonstrated the efficacy of pharmacologic prophylaxis against VTE in high-risk hospitalized patients. These studies have shown that both low molecular weight heparin (LMWH) and unfractionated heparin reduce the risk of both DVT and PE without a significant increased risk of major bleeding (minor bleeding episodes, including development of hematomas, maybe more common in patients receiving pharmacologic prophylaxis). Prophylaxis has not been associated with a mortality benefit and its cost-effectiveness has not been well studied. For patients at high risk of bleeding, the use of intermittent pneumatic leg compression is a reasonable substitute for heparin products.
The use of standard order sets and reminders built into computerized order entry systems facilitate routine use of VTE prophylaxis in high-risk patients. The effect of avoiding bedrest and encouraging frequent mobilization on reducing the occurrence of DVTs and PEs has not yet been studied.
Hospitalized older adults are more likely than hospitalized younger adults to suffer iatrogenic complications of hospitalization because older adults have longer, more complex hospitalizations, and greater physiological vulnerability. Since the Institute of Medicine Report focusing on patient safety, some of these “iatrogenic” complications are now understood as patient safety problems. For example, infectious complications of catheters and devices are common among older patients, who often have urinary catheters, other lines, and devices. Thromboembolic events, nosocomial infections, falls, ADEs, and drug errors due to incorrect reconciliation of a complex medical regimen, all are more common in older hospitalized patients and are targets of patient safety improvement. National Patient Safety Goals, approved by the Joint Commission on the Accreditation of Health Care Organizations in 2008, address many risks faced by older adult hospital patients. The goals include reducing risk of health care-associated infections, reconciling medications accurately across the care continuum, reducing falls and fall risk, preventing pressure ulcers, and assessing risk of pressure ulcer development. Geriatricians and other physicians and providers caring for older hospitalized patients should be actively involved with planning and implementing patient safety interventions in the acute hospital setting.
In contrast to the ideal vision of death in which most people see themselves dying at home with family, many Americans die in hospitals. As recently as 2001, it was estimated that nearly 50% of people who died did so in hospitals. In some cases death is unexpected, but in most cases it is not. Despite the fact that many people die in hospitals, we do not yet do a good job of delivering end-of-life care. Surveys of patients and families have shown that patients dying in hospitals believe they do not have enough contact with physicians, emotional support, information about what to expect of the dying process, and unfortunately, a substantial percentage report moderate or severe pain in the last 3 days of life. Delivering high-quality end-of-life care requires experience, as well as a substantial investment in time. Providers working in today's hospitals recognize these problems, yet struggle with increasingly complex patients, as well as demands for both increased quality and efficiency of care. As a result, many hospitals have developed formal palliative care programs to adequately address the needs of dying patients. Most palliative care programs focus on maximizing comfort and quality of life rather than delivering curative therapy. Formal palliative care programs focus on pain and symptom control, communication challenges, addressing spiritual needs, and facilitating care transitions. As of 2002, it was estimated that 26% of academic medical centers and 17% of all hospitals had hospital-based palliative care programs. While many hospitals use a variety of practitioners to staff their programs, geriatricians, hospitalists, and oncologists are well-represented in most programs. There now are formal palliative care fellowships and plans for board certification in this discipline. Any comprehensive approach to care of the hospitalized elder must address issues of end-of-life care. A full discussion of end-of-life care can be found in Chapter 31.
The care of hospitalized elders is occurring within a dynamic hospital environment that is changing, in part to address financial and other challenges for providers but also to better meet the needs of the geriatric population, and the increasing demands to provide safer, higher quality, and more cost-effective care. In this section of the chapter, we address some of the changes in models of hospital care relevant for the healthcare provider focused on the high-quality care of hospitalized elders.
Concurrent with the increased attention to the needs of hospitalized elders has been the emerging presence of hospitalists. The hospitalist physician, usually a general internist, focuses clinical activity almost exclusively on caring for hospitalized patients. Varied forces have driven the hospitalist movement, including improved efficiency of hospital care, physician ownership of inpatient quality and safety initiatives, and the dual clinical and financial challenges of managing patients in both the office and hospital. Teaching hospitals also employ hospitalists to mitigate pressures resulting from resident work hour reductions. The rapid growth of the hospitalist field has been supported, in part, by direct financial support from physician group practices, hospitals, and academic medical centers.
There are currently 15 000 practicing hospitalists in the United States, with numbers expected to double by 2010 (Figure 17-1). The American Hospital Association (AHA) found that nearly 30% of U.S. hospitals had hospital medicine groups; the percentage exceeded 60% for hospitals with more than 200 beds. Similar penetration exists at academic medical centers (AMCs). Over half of teaching hospitals and 66% of major teaching hospitals (defined as a member of the Council of Teaching Hospitals and Health Systems) have developed hospital medicine groups. The hospitalist programs at these academic centers tend to be quite large, with an average of 17 hospitalists per program.
Growth of hospital medicine. Source: Society of Hospital Medicine.
As hospital medicine programs have grown, so have the roles of hospitalists. The hospitalist's primary task has not changed since the formation of the field, namely, the day-to-day clinical care of complex medical patients, in both academic and nonacademic settings. However, in many institutions, they have expanded their scope of practice by establishing palliative care programs as well as quality improvement and patient safety programs all of particular applicability and potential benefit to the hospitalized elderly patient. In addition, hospitals have supported the creation of surgical comanagement programs that focus on both outpatient preoperative assessment and inpatient postoperative management. With the hospital setting as their “office,” hospitalists are often in leading administrative roles within hospitals and academic medical centers.
Teaching hospitals need to support residency programs; this need has catalyzed hospitalist growth in those settings. Driven by the need for increased educational oversight and Medicare billing requirements, academic medical centers have increasingly turned to hospitalists to staff resident teaching services. More recently, in an effort to prevent adverse outcomes related to overly tired trainees, in July 2003, the Residency Review Committee (RRC) began enforcing rules that prevent residents from working more than 80 hours in 1 week or for more than 30 hours in a single shift. Faced with increasing bed occupancy, increasingly complex patient populations, and the perception that the floats, hand-offs, and patchwork coverage solutions were adversely affecting patient care, academic medical centers looked for other options. Many chose to decompress the numbers of patients seen on resident services to nonresident services, which are now commonly staffed by hospitalists.
As the hospitalist model of care has spread, the field has changed the way clinical problems are approached and managed. New research questions have been raised regarding the clinical approach to hospitalized patients, implementation of best practices, care transitions, and how to improve patient safety, again with potentially significant benefit to the medically complex and/or potentially vulnerable elderly patient. Representative research topics currently being explored by hospitalist investigators include identifying biomarkers predictive of perioperative outcomes, evaluating methods to prevent health care-associated infection, examining the utility of a variety of care processes for common inpatient diagnoses like pneumonia, studying the intersection between resident education and inpatient safety, and exploring techniques to improve quality during transitions of patient care.
Hospitalists are both ideally suited and situated to play a central role in improving quality and safety of the care of hospitalized elders. As a site-defined specialty, hospitalists spend more time in the hospital than do many other physicians, allowing them to gain insight into the problems commonly encountered in geriatric care and some potential solutions. By working closely with hospital personnel and administration, hospitalists can take the lead in the development and refinements of interdisciplinary teams designed to enhance the quality and safety of inpatient geriatric care.
Acute Care of the Elderly Units
Acute Care of the Elderly (ACE) units were described and studied in community and academic settings in the mid-1990s. The ACE unit intervention consisted of environmental interventions (rugs, lighting, handrails), patient-centered care model that emphasized independence and rehabilitation from the beginning of the hospital stay, and a multidisciplinary care team consisting of nurses, physicians, therapists, social workers, and nutritionists and pharmacists providing coordinated care within a dedicated inpatient unit. Findings from the randomized controlled trial in 1995 found significantly improved functional status and decreased admission to nursing homes in patients randomized to the ACE unit interventions. Lengths of stay and hospital charges were similar in both groups. Recent retrospective studies have confirmed ACE unit benefits on function and quality of life, without increased hospital charges. Despite favorable results, ACE units are not widely disseminated in U.S. hospitals, largely due to the capital expenditures incurred in creation of such a unit, difficulty in recruiting the qualified specialist leaders, and future concerns with not having enough room for the rapidly expanding elderly patient population in such a space-defined unit. A review in 2003 surveyed hospitals associated with medical schools that had a geriatrics program and found that only 16 had ACE units. The number of community hospitals with ACE units is unknown. Some hospitals or academic medical centers have preferred to develop a virtual ACE team, rather than a dedicated ACE unit. The virtual ACE team is a multidisciplinary geriatrics consult team involving physician, nurse, social work, pharmacy and physical therapy that is designed to improve patients' hospital course and expedite discharge planning and transitional care. The virtual ACE team model provides the expertise and systems approach to the care of the elderly hospitalized population, particularly the vulnerable and medically complex outside the confines of a defined unit.
Stroke units use a care delivery model analogous to ACE unit models. A stroke unit is a multidisciplinary team specifically dedicated to the care of the stroke patient. Physicians, nurses, and therapists are involved in a coordinated effort to support the care and rehabilitation of stroke patients. The stroke unit maybe a dedicated hospital unit, a mobile team than manages patient along with the primary inpatient team, or may function within a rehabilitation unit. Stroke units were reviewed in a Cochrane report in 2001, which analyzed results from 23 randomized and quasirandomized trials and concluded that there was evidence that stroke units improved function at discharge, rates of discharge to home, and stroke survival. There was some evidence that dedicated stroke units had better outcomes. Whether there are aspects of poststroke care that can be generalized and systematized to improve elderly patient health care outcomes outside such a unit needs further investigation.