To determine if there are any unique features of nursing home-acquired pneumonia we carried out a case-control study wherein each patient admitted with nursing home-acquired pneumonia was age- and sex-matched with a patient with community-acquired pneumonia. There were 36 men and 38 women in the nursing home group. The mean age of both groups was 74 years. The mortality rate for nursing home-acquired pneumonia it was 40.5%, whereas for community-acquired pneumonia it was 28% (P = NS). Patients with nursing home-acquired pneumonia had a significantly higher incidence of dementia and cerebrovascular accidents, and patients with community-acquired pneumonia were more likely to be smokers and to have chronic obstructive pulmonary disease. Aspiration pneumonia was more common among patients with nursing home-acquired pneumonia (P less than .001), and Hemophilus influenza pneumonia more common among the patients with community-acquired infection (P less than .01). Sputum for culture could be obtained in only 31 and 39% of the patients--contributory to the high rates of pneumonia of unknown etiology 63.5 and 56.1% for the nursing home group and the control subjects, respectively. Patients with nursing home-acquired pneumonia received cloxacillin and aminoglycosides more frequently than patients with community-acquired pneumonia (P less than .05), and patients with community-acquired pneumonia received erythromycin more frequently than patients with nursing home-acquired pneumonia (P less than .05). Complications were common during the hospital stay of these patients--the most frequent being congestive heart failure, urinary tract infection, renal failure, and respiratory failure.