Azithromycin, a recently introduced antibiotic, offers the potential advantages of short-course administration and lower toxicity compared to other macrolides. Approved for the treatment of mild pneumonia, this drug was investigated in a study of patients hospitalized for community-acquired pneumonia. In an open-labelled randomized study, oral azithromycin was compared with intravenous benzylpenicillin in patients suspected to have pneumococcal pneumonia. Azithromycin was also compared with erythromycin, both administered orally, in all other patients. Three hundred thirty-four patients with community-acquired pneumonia were hospitalized, 108 of whom were randomized; 104 could be evaluated. A need for intravenous therapy was the most common reason for exclusion. In the pneumococcal group, 35 patients received azithromycin and 29 benzylpenicillin. The clinical and radiological success rate achieved with azithromycin (83%) was considerably higher than that achieved with benzylpenicillin (66%), though the difference was not significant. In the non-pneumococcal group, 19 patients received azithromycin and 21 erythromycin; no differences in the success rate were found (79% and 76%, respectively). Eight patients on azithromycin had a blood culture positive for Streptococcus pneumoniae; in three of these patients therapy was changed. None of the five patients with pneumococcal bacteraemia who received benzylpenicillin required a change in therapy. It is concluded that oral azithromycin, administered as short-course therapy, is an appropriate antibiotic for treating patients with community-acquired pneumonia. However, it is not yet certain that azithromycin is a good choice for patients with pneumococcal bacteraemia.