ObjectivesPneumonia is a leading cause of morbidity and mortality in nursing home (NH) residents. Chest x-ray evidence is considered a key diagnostic criterion for pneumonia by the Infectious Disease Society of America (IDSA) diagnostic guidelines, the modified McGeer diagnostic criteria, and the Loeb criteria for initiating antibiotics; however, x-ray interpretation is often equivocal. We conducted chart audits of patients in NHs who had chest x-rays for new respiratory symptoms to determine the degree of ambiguity in the radiology reports and their relationship to antibiotic prescription decisions. DesignCross-sectional study. SettingThirty-one NHs in North Carolina. ParticipantsTwo hundred twenty-six NH residents who had a chest x-ray. MethodsMedical charts were abstracted to record (1) the patient's clinical presentation when a chest x-ray was ordered, (2) the verbatim report of the chest x-ray, and (3) the patient's course during the subsequent 7 days. To standardize the radiologist reports, a seven-category coding system was developed, which was further aggregated into three groups based on the radiologist's description of the likelihood of pneumonia. ResultsOf the 226 chest x-rays, 118 (52%) identified a very low likelihood of pneumonia, 67 (30%) indicated that pneumonia was present or highly likely, and the remaining 41 (18%) used a variety of terms to describe uncertainty regarding the presence of pneumonia. NH medical providers tended to treat ambiguous chest x-ray reports similarly to positive x-ray reports, prescribing antibiotic therapy to 71% of patients with ambiguous reports and 78% of positive reports. Also notable is that 40 (34%) of the 118 patients with a very low likelihood of pneumonia based on chest x-ray results were prescribed antibiotics, the majority of whom failed to meet criteria for a clinical diagnosis of pneumonia or chronic obstructive pulmonary disease exacerbation. ConclusionThe moderate rate of ambiguous x-ray interpretations in NH residents is likely a combination of the poor quality of portable x-rays, a high prevalence of chronic lung conditions, and conservative (ie, cautious) decision making by radiologists whose interpretation is based on little clinical information and a suboptimal quality film. As a result, data suggest that chest x-rays obtained in NHs may unnecessarily encourage antibiotic prescribing because a majority of readings are ambiguous or show a low likelihood of pneumonia, yet more than half of the patients are still treated. From an antibiotic stewardship standpoint, the apparent solution is to more closely rely on clinical signs and symptoms for diagnosis of pneumonia and to place less emphasis on the role of the chest x-ray given the high number of unclear readings.
Read full abstract