The efficacy and safety of clarithromycin, roxithromycin and erythromycin stearate in mild pneumonia were compared in an open randomized trial. Eighty-six male patients, doing their obligatory military service, ranging between 19 and 24 years of age (mean 20), were randomly treated: 29 with clarithromycin 500 mg 12-hourly, 30 with roxithromycin 150 mg 12-hourly, and 27 with erythromycin stearate 500 mg 6-hourly, each course being administered for 10 days. Seventy-eight patients were able to be evaluated for efficacy, 28 receiving clarithromycin, 28 roxithromycin, and 22 erythromycin stearate. There were no significant differences among the groups in terms of clinical success rates (clinical cure or improvement: 89% for clarithromycin, 82% for roxithromycin, and 73% for erythromycin stearate, p = 0.32). However, we found that there were significant differences among the groups in terms of clinical cure rates (75% for clarithromycin, 64% for roxithromycin, and 41% for erythromycin stearate, p = 0.04). Adverse events, mostly gastrointestinal, caused discontinuation of treatment in 3.4% of the patients in the clarithromycin group, in 6.6% of the patients in the roxithromycin group, and in 18.5% of the patients in the erythromycin stearate group. The results indicate that there were no statistically significant differences among the three treatment groups in terms of clinical success rates, but that clarithromycin and roxithromycin were better tolerated.