Validation and application of the pneumonia prognosis index to nursing home residents with pneumonia

OBJECTIVES: To evaluate the predictability of a pneumonia prognosis index i n nursing home residents with pneumonia and to use the index to account for acute severity of pneumonia before comparing the short-term outcome of res idents with pneumonia treated with intravenous antibiotic therapy in two di fferent settings: an inpatient geriatrics unit and a nursing home DESIGN: A retrospective chart review of 158 episodes of nursing home-acquir ed pneumonia treated initially with intravenous antibiotics; 100 episodes w ere treated in an inpatient acute geriatrics service (AGS), and 58 were tre ated completely in a nursing home (Nursing Home group) SETTING: The AGS is a 20-bed unit within a 400-bed, public, university-affi liate hospital. The Nursing Home group consisted of residents of two nonpro prietary nursing homes. PARTICIPANTS: Nursing home residents with radiographically proven pneumonia who had at least one of the following signs/symptoms: cough, fever, purule nt sputum, respiratory rate greater than or equal to 25 per minute, localiz ed auscultatory findings, or pleuritic pain. MEASUREMENTS: The pneumonia prognosis index was calculated for each residen t at the time of diagnosis of pneumonia; the index has been validated as a predictor of hospital outcome in patients with community-acquired pneumonia and is also considered a measure of acute severity of pneumonia. Status (a live or dead) of each resident at 30 days after diagnosis was the major dep endent variable RESULTS: Mean (+/- SD) duration of antibiotic therapy for the Nursing Home group (10.7 +/- 4.5 days) was not significantly different from that of the AGS group (9.6 +/- 3.4 days; P = .26). The pneumonia prognosis index strati fied the 158 episodes of pneumonia into low- and high-risk groups for 30-da y mortality; the mortality rates in each risk strata were not significantly different from those reported in the original derivation and validation st udies of the index. In addition, the distribution of episodes among the ris k strata of the index was not significantly different for the two study gro ups, which was an indication that the two groups were similar in terms of a cute severity of pneumonia. Thirty-day mortality was not significantly diff erent between the two groups: AGS, 21% and Nursing Home, 24.1% (P = .66). CONCLUSION: The pneumonia prognosis index seems to have the same capability for predicting the outcome in nursing home residents with pneumonia as in residents with community-acquired pneumonia. The index is also a measure of acute pneumonia severity. Nursing home residents with pneumonia, even thes e who are I-nest acutely ill, can be treated successfully with intravenous therapy in the nursing home; their 30-day mortality was no different than t hat of those with the same acute severity of illness who were admitted to a hospital for treatment.