Send to

Choose Destination
J Infect. 1997 Jan;34(1):41-8.

The epidemiology of community-acquired pneumonia among hospitalized adults.

Author information

Division of Internal Medicine, Soroka Medical Center, Faculty of the Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel.



To identify and characterize the aetiological agents of community-acquired pneumonia (CAP) among hospitalized patients, as an aid in therapeutic decision-making.


A prospective 1 year study of all patients hospitalized with CAP in the Negev region of Israel. The aetiology was determined by blood and pleural fluid cultures, and specific serological testing for pathogenic agents. Eighty-nine percent of the patients underwent follow-up for a month after discharge.


The study included 346 patients (53% males, mean age 49.3 +/- 19.5, range 17-94). A single aetiologic agent was identified in 146 patients (42.2%), multiple agents were found in 133 (38.4%), and no agent was identified in 67 (19.4%). Among the common pathogens were Pneumococcus sp. in 148 patients (42.8%). Mycoplasma pneumoniae (101, 29.2%), Chlamydia pneumoniae (62, 17.9%), Legionella sp. (56, 16.2%), viruses (35, 10.1%), Coxiella burnetti (20, 5.8%). Haemophilus influenzae (19, 5.5%), and other bacteria (21, 6.1%). Approximately 70% of the patients infected with M. pneumoniae and C. burnetti were younger than 45 years (P < 0.05). In contrast, about 50% of the patients with C. pneumoniae (TWAR) were over the age of 65 (P = 0.03). The presence of comorbidity was associated with a greater frequency of bacterial aetiologies (57% vs. 44%, P = 0.02), and fewer infections with M. pneumoniae (15% vs. 36%, P = 0.0004), or C. burnetii (2% vs. 8%, P = 0.02). Specific causative agents were associated with specific seasons: viruses between December and April (P = 0.03), and Legionella sp. from July to October (P = 0.003). In contrast, no seasonal variation was associated with pneumococcus, M. pneumoniae, or C. pneumoniae (TWAR).


Patients are hospitalized with CAP throughout the year. Since the pathogen is usually unknown at hospitalization, epidemiological data is important for choosing medication. The findings of this study point to the importance of macrolides alone or in combination with cephalosporins, as the treatment of choice for patients in our region.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center