COMMUNITY-ACQUIRED PNEUMONIA IN THE ELDERLY - A MULTIVARIATE-ANALYSISOF RISK AND PROGNOSTIC FACTORS

To assess the risk and prognostic factors of community-acquired pneumo nia occurring in the elderly (over age 65 yr) requiring hospitalizatio n, two studies, case-control and cohort, were performed over an 8-mo p eriod in a 1,000-bed university teaching hospital. We studied 101 pati ents with pneumonia (cases), age 78.5 +/- 7.9 yr (mean +/- SD). Each c ase was matched for sex, age (+/- 5 yr), and date of admission (+/- 2 d) with a control subject, without pneumonia during the preceding 3 yr , arriving at the emergency room. Etiologic diagnosis was obtained in 43 of 101 (42%) cases. The main microbial agents causing pneumonia wer e: Streptococcus pneumoniae (19 of 43, 44%), and Chlamydia pneumoniae (9 of 43, 21%). Cram-negative bacilli were uncommon (2 of 43, 5%). The multivariate analysis demonstrated that large-volume aspiration, and low serum albumin (< 30 mg/dl) were independent risk factors associate d with the development of pneumonia. Crude mortality rate was 26% (26 of 101), while pneumonia-related mortality was 20% (20 of 101). The at tributable mortality was 23% (odds ratio [OR]: 11.3; 95% confidence in terval [CI]: 3.25 to 60.23; p < 0.0001). The multivariate analysis sho wed that patients had a worse prognosis if they were previously bedrid den, had prior swallowing disorders, body temperature on admission was less than 37 degrees C, respiratory frequency was greater than 30/min or had three or more affected lobes on chest radiograph. Age by itsel f was not a significant factor related to prognosis. Among the signifi cant risk factors, only nutritional status is probably amenable to med ical intervention. The prognostic factors found in this study may help to identify, upon admission, those subjects at higher risk and who ma y require special observation.