An analysis of the safety data emanating from several clinical trials of the use of azithromycin in children has been undertaken. The studies compared azithromycin with standard regimens of various comparator agents: co-amoxiclav, amoxycillin, penicillin V, erythromycin, dicloxacillin and flucloxacillin, in the treatment of otitis media, pharyngitis and skin infections. Side-effects, based on the investigators' assessments of the severity and possible relationship to the study drugs, were measured within 35 days of the start of therapy. The total number of side-effects reported was significantly lower in the azithromycin-treated patients (46/606; 7.6%) than in the comparator-treated patients (72/523; 13.8%; P = 0.001). Two (0.3%) and five (1.0%) patients, respectively, withdrew from treatment in each group. The most frequently reported side-effects were gastrointestinal with an incidence of 5.6% (34 patients) in the azithromycin group and 11.7% (61 patients) in the comparator group (P < 0.001). The only other organ system with more than 0.5% of subjects reporting side-effects was the skin, with nine (1.5%) patients in the azithromycin and 11 (2.1%) patients in the comparator group complaining of dermatological reactions. Azithromycin was equally well tolerated in children of all ages. Inspection of individual studies showed no significant differences in laboratory test abnormalities between treatment groups. Overall, the most frequently (> or = 1% incidence) reported treatment-related laboratory test abnormalities were changes in neutrophil count (3.2%), lactate dehydrogenase (1.5%), blood urea (1.3%) and potassium (1.2%) in the azithromycin group; and in neutrophil (3.2%) and eosinophil count (2.1%) in the comparator group.(ABSTRACT TRUNCATED AT 250 WORDS)