Objectives: To describe the findings of a retrospective study of the treatment of nursing home-acquired pneumonia (NHAP) in 11 nursing homes in one community and the development of a treatment guideline for NHAP using data from the retrospective study.
Design: A retrospective chart review of 239 episodes of NHAP occurring between November 1, 1997, and April 30, 1998, was performed. Data regarding antibiotic treatment of NHAP were used to revise a treatment guideline developed by the authors. Further refinements of the guideline were made based on small group discussions with physicians and nurse practitioners caring for the study population.
Setting: Residents with NHAP were identified among the populations of 11 nursing homes in the metropolitan Buffalo, New York area (Erie county). These 11 nursing homes had a total of 2375 beds, comprising nearly one-third of all nursing home beds in the county.
Participants: Nursing home residents with chest X-rays showing infiltrates and signs and symptoms of pneumonia.
Measurements: Antibiotic treatment (drug used, route of administration, and duration of treatment), location of initial treatment (nursing home or hospital), and status (alive or dead) of each resident were recorded 30 days after diagnosis of NHAP.
Results: Of the 239 episodes of NHAP, 171 (72%) were initially treated in nursing homes. Of these 171 patients, 105 (61%) were treated only with an oral regimen, whereas 66 (39%) were treated initially with an intramuscular antibiotic and subsequently with an oral regimen. There was no significant difference in 30-day mortality rates between those initially treated in nursing homes (22%) and those initially treated in hospitals (31%; P = .15) or between those initially treated with an oral regimen in nursing homes (21%) and those initially treated with an intramuscular antibiotic in nursing homes (25%; P = .56). There was no consistency in how physicians made the choice to use intramuscular antibiotics in nursing homes, and a logistic model for predicting this approach could explain very little. The frequency of the prescription of various antibiotic agents in nursing homes and in hospitals was tabulated as well as the duration of treatment; specific attention was paid to the timing of the switch to an oral agent among episodes initially treated with a parenteral agent. These data were used in the guideline to make specific recommendations regarding which agent to prescribe, the duration of parenteral therapy, the timing of the switch to an oral regimen, and the duration of treatment. In the setting of informal small groups, the guideline was discussed with physicians who cared for residents with NHAP in the study nursing homes. Revisions made to the guideline were based on these discussions.
Conclusions: A treatment guideline for NHAP was developed primarily on the basis of the practices of geriatricians in one community. These treatment practices were similar to those reported in the literature in terms of the proportion of patients treated in nursing homes and the antibiotics prescribed. The guideline also provided specific recommendations for timing of the switch to an oral agent after parenteral therapy and for duration of treatment. Studies are in progress to determine if use of this guideline will reduce some of the variation observed in the treatment of NHAP.