Nursing Home‐Acquired Pneumonia: Avoiding the Hospital

Lower respiratory tract infections remain a highly significant cause of morbidity and mortality among residents of long-term care facilities. Pneumonia is a leading cause of hospitalization from nursing homes, with some series reporting mortality rates of 40 to 50%.1 For at least the past 20 years, nursing home-acquired pneumonia (NHAP) has been considered a distinct clinical entity, defining a clinical syndrome of intermediate severity between community- (CAP) and hospital-acquired pneumonia (HAP).2 In Great Britain, British Thoracic Society guidelines exclude any patient in a nursing home from their definition of CAP.3 Given this characterization of NHAP, it is not surprising that there should be a low threshold for hospital admission for patients with NHAP. Recently, the concept that the clinical characteristics of NHAP are sufficiently discrete to dictate therapeutic decision for an entire sub-set of older persons with pneumonia has been challenged. In 1993 the American Thoracic Society officially adopted guidelines directed at evaluation and management of CAP.4 These guidelines provide a framework for the initial evaluation and therapy of the patient with CAP that are largely empirical and ignore such time honored approaches as extensive diagnostic testing, Gram's stains, chest X-rays, or classification into “typical” and “atypical” presentations. Initial decisions regarding selection of antibiotic therapy and need for hospitalization are based largely on whether patients are age 60 years or older and whether co-morbid illness exists. In this context, NHAP is not considered a distinct entity. Based on these guidelines, many nursing home residents with pneumonia would qualify for non-hospital therapy. These guidelines stress further that the most important determinants of the need for hospitalization are the presence of certain bedside signs, such as a respiratory rate greater than 30/minute, temperature greater than 38.3° C, and altered mental status. Given these widely accepted guidelines, along with the increasing capability of nursing homes to care for more seriously ill patients and the availability of broader spectrum antibiotics that can be administered orally, do the clinical outcomes of in-hospital care compared with care in the nursing home justify the assumption that pneumonia acquired in this setting is a distinct entity apart from CAP, necessitating more intense technology? The study by Fried and colleagues in this issue presents much needed information regarding clinically relevant outcomes of treatment of NHAP in the nursing home compared with the hospital setting.5 This study provides new information regarding the validity of a more individualized clinical approach to NHAP. Additionally, their conclusions add to a growing literature seeking to ascertain what kinds of highly skilled care traditionally provided in acute hospitals can be provided safely and effectively in long-term care facilities. In their initial paper, Fried and colleagues reported in a retrospective analysis of acute cases of pneumonia in a nursing home that immediate survival rates were comparable in patients treated either in the nursing home or in hospital.6 Simple bedside clues, especially the presence of a respiratory rate greater than 40/minute, were highly predictive of the need for hospitalization. They found that decisions to hospitalize were complex and often not based solely on severity of illness. For example, patients developing clinical symptoms of pneumonia in the evening were much more likely to be hospitalized than if those same symptoms developed during the day. Further analysis of this issue emphasizes that not only do the two venues of therapy produce about the same immediate events, those patients treated in the nursing home setting actually had better preservation of function and less total mortality in the 2 months following the acute pneumonia compared with the hospitalized cohorts. See also p 302 The rapidly growing national trend toward provision of higher technology and intensity of care in nursing homes, often labeled subacute care, makes it essential that there be well designed research to ascertain the kinds of highly skilled care, traditionally provided in acute hospitals, that can be provided safely and effectively in long-term care facilities. There is currently no universal agreement on the exact definition of subacute care and even less on how to best provide and finance it. However, strides are being made to attempt to delineate approaches to delivery of more skilled transitional care, usually meaning earlier discharge from hospital with continuation of the acute care modalities started in the hospital.7 It is clear that patients are increasingly being admitted to nursing homes with greater severity of chronic illnesses and complications, especially since the inception of DRGs — “quicker and sicker.”8 More technological support and more acute care nursing and medical skills are thus needed for both diagnosis and therapy. These include laboratory and radiological resources, capacity for intravenous and ventilator therapy, and more intense monitoring. The rationale for these changes, while in large part motivated by the cost-saving potential of reducing hospital care, is also based on the now well accepted realization that hospitals are not healthy places, especially for older, chronically ill people.9 In fact, patients with NHAP would be expected to be one of the highest risk groups for developing new disabilities in activities of daily living following any acute illness and hospitalization. The recent multicenter Hartford Foundation Hospital Outcomes Project for the Elderly (HOPE) study identified three patient characteristics that were independent predictors of functional decline in hospitalized older persons: increasing age, lower admission Mini-Mental State Exam scores, and lower preadmission IADL functional characteristics present in the vast majority of older persons hospitalized with NHAP.10 These objectives of cost reduction and hospital avoidance could presumably be served even better by going a step further than earlier hospital discharge, that is by treating acute episodes in the nursing home from their inception. This has been an area of interest in our region for some time and is currently the subject of a HCFA-supported randomized controlled trial.11,12 In the two pilot studies leading up to the current controlled trial, pneumonia accounted for 29% and 24% of cases. Bacterial infections in general accounted for 46% and 52%, the next most common general category being other cardiorespiratory conditions at only 11% and 16%. Although the Fried et al. study provides useful information that pneumonia can be managed successfully in a long-term care facility without apparent additional risk, there are a number of caveats regarding the generalizability of their results. First, their facility has a strong academic affiliation with staff physicians and geriatric fellows, daily physician rounds, and availability of intravenous therapy. This is clearly beyond the resources of the average nursing home. In fact, in a recent study evaluating the effectiveness of oral antibiotic therapy in NHAP, which included 31 community nursing homes in St. Paul, Minnesota, 31% of patients failed to improve with nursing home treatment and were subsequently hospitalized.13 Interestingly, most of these failures had elevated respiratory rates and would have been hospitalized initially according to the criteria proposed by Fried. Second, the Fried study was a retrospective cohort study in which the investigators made every possible attempt to control for confounding variables in the two groups, those hospitalized and those not. However, as they acknowledge, there could have been subtle variables not measured or documented that led to selection of sicker patients for hospitalization, who might not have fared as well in the nursing home. Thus, as they point out, “it is impossible to know from an observational study if the patients would have done differently if they had been treated at another site.”5 Clearly, randomized controlled trials are needed to demonstrate convincingly that care for usually hospitalized patients with acute conditions can be feasible, safe, and effective in long-term care facilities with equivalent, or better, short- and long- term outcomes compared with hospital care. In the meantime, this study strongly supports an individualized approach to the nursing home resident with pneumonia. Many, perhaps the majority, of patients with NHAP will not suffer adverse consequences by hospital avoidance, and may actually sustain better functional outcomes with lower risk of mortality. In this setting, perhaps another element, which has not been part of previous studies, should be considered: patient preference. When patient preferences are considered in individuals with CAP, most will choose care in the community.14 It is likely that a comparable, or even greater, proportion of individuals with NHAP would make the same choice if asked.