Management of Nursing Home‐Acquired Pneumonia: Unresolved Issues and Priorities for Future Investigation

Patients residing in long-term care facilities develop pneumonia at a rate of approximately one infection per 1000 days of care.1 This rate is comparable to that experienced by patients admitted to acute care hospitals and represents a 10-fold increase over the rate of pneumonia in older people living in the community. Pneumonia is the leading cause of death among nursing home residents2,3 and a major reason for transfer to an acute care facility.4,5 Nursing home residents account for 10% to 18% of all who are hospitalized for pneumonia,6–8 with an average hospital charge of approximately $10,000 per admission.9 Against this background, it is distressing to observe that there is little consensus of opinion regarding the appropriate evaluation and management of nursing home-acquired pneumonia (NHAP). In addition, there are large variations in practice, including the extent of the evaluation of the patient, the rate of transfer to acute care, and the choice and route of administration of antimicrobial therapy. This state of affairs is caused, at least in part, by a lack of some very basic information. For example, data on the microbial etiology of NHAP is inconsistent and, in some cases, contradictory. The reported proportion of cases caused by Streptococcus pneumoniae ranges from 0% to 39% in reported series.1 This reflects, in part, the difficulty of obtaining adequate microbiologic specimens in this patient population. Microbiologic services comparable to those available to acute care facilities are often not available to long-term care facilities. Thus, empiric therapy rather than pathogen-directed therapy tends to be the rule. There are very few adequate clinical trials comparing different antimicrobial regimes in the treatment of pneumonia and even fewer addressing the appropriateness of oral versus parenteral therapy. Clinicians, therefore, have little solid information on which to base therapeutic decisions. Other variations in practice are less readily explained. The rate of hospital transfer for patients with nursing home-acquired pneumonia ranges from 9% to 51% in various reports.10–15 This is little evidence that hospitalization for pneumonia improves outcome in nursing home patients. Indeed, one study found that death and functional deterioration were more frequent in hospitalized patients than in those treated in the nursing home, even after controlling for confounding factors.16 A comparative study of two nursing homes found a twofold difference in hospitalization rates that was not readily explained by differences in the facilities or the patient populations;10 the mortality rates were identical. Until definitive answers are available, on what basis should clinical decision-making proceed when caring for a patient with NHAP? One approach is exemplified by the work of Naughton and Mylotte,17 reported in this issue of the Journal. They performed a review of 239 episodes of nursing home-acquired pneumonia from 11 facilities in a single community. They reviewed the relevant literature and the actual practice of clinicians treating patients with NHAP in local facilities and devised a tentative treatment guideline. This guideline was then refined after review and discussion with local practitioners. The resultant guideline addresses the antimicrobial agents used, use of parenteral versus oral therapy, timing of the switch from parenteral to oral treatment, and duration of treatment. Although this guideline does not have a firm foundation of adequate clinical trials (because few exist), it should nevertheless prove useful to clinicians. It complements recommendations derived from literature review1 and reflects the consensus of experienced local practitioners. It is therefore likely that these practitioners will adopt this guideline, which will help to reduce the variability in the delivery of care to patients with NHAP. See also p 82 One example of this variability is related to the decision regarding the route of initial therapy. By logistic regression, only very high (>104°F) or very low (<96°F) temperatures correlated with initial use of parenteral therapy. This predictor explained very little, however. Furthermore, the authors were unable to demonstrate any significant difference in the rate of hospital transfer or mortality between patients initially treated with oral therapy versus those initially treated with parenteral therapy. Because parenteral therapy is generally more uncomfortable and more expensive than oral therapy, avoidance of the former would decrease both patient discomfort and the cost associated with treatment. A key area not addressed by Naughton and Mylotte is the decision to hospitalize a patient with NHAP. Although a patient's likelihood of succumbing to pneumonia correlates well with indices of functional disability11,14 and with a prognostic index initially developed for use with community-acquired pneumonia,18 it is not at all clear, as previously noted, that hospitalization improves outcome. Indeed, the decision to hospitalize a nursing home patient for an acute illness is based on a variety of factors, including the wishes of the patient and family and the availability of a higher level of care at the facility. The organization of care within a facility seems to be an important factor in the decision to transfer. Nursing homes with special care units and greater availability of physicians or physician extenders (nurse practitioners/physician assistants) are significantly less likely to transfer patients to hospitals for acute episodes of illness.19 Evidence suggests that the number of nursing home primary care visits provided by patients’ health plans influences the frequency of hospitalization.20 In many facilities, a substantial proportion of patients with respiratory infection do not receive a physician visit, and many are managed over the telephone.21–23 Lack of appropriate initial evaluation is associated with an increased likelihood of hospitalization or death in NHAP.12 There is evidence that timely and appropriate management of acute illness within the nursing home reduces cost as well as the rate of hospitalization.22 What remains to be determined is which patients would benefit from hospitalization for NHAP. An additional critical area of uncertainty is that of prevention. With the possible exception of appropriate immunization against influenza and pneumococcus, strategies for the prevention of NHAP are undefined.1 Although a number of studies have identified risk factors for the development of NHAP, these include factors such as tendency to aspirate, dependence in feeding, functional debility, and chronic pulmonary disease. It is difficult to translate identification of patient risk into prevention unless the risk factors can be modified. The number of older people residing in nursing homes will grow from 1.6 million in 1990 to an estimated 5.3 million in 2030.20 If the incidence of pneumonia in this population remains constant, this will translate into 1.9 million episodes of NHAP annually. This represents an enormous burden of morbidity, mortality, and expenditure. The healthcare system will be ill-prepared to deal with what promises to be a major public health problem unless we can develop medically sound, cost effective strategies for the evaluation, management, and prevention of NHAP. Investigators in the fields of geriatrics, hospital epidemiology, and infectious diseases should consider NHAP to be an area of high priority for future investigations.