In an earlier study of 214 children with acute otitis media (AOM) randomly assigned to a 14-day course of either cefaclor or amoxicillin, 55.7% of cefaclor-treated subjects were effusion-free compared with 41.2% of amoxicillin-treated subjects at the end of treatment. The present study was conducted to determine whether, in a 1-year period, subjects treated with cefaclor for each episode would have middle ear effusion for significantly less time than those treated with amoxicillin for each episode. One hundred fifty-seven subjects with AOM were enrolled. No statistically significant differences were found between the cefaclor-treated and amoxicillin-treated groups in number of subjects effusion-free immediately after first treatment (47.9% vs. 42.3%, respectively), average percent of time with effusion (22.2% vs. 23.4%, respectively), or in rate of new episodes of AOM (3.05 vs. 3.26, respectively). We conclude that there is no clinically significant advantage to the use of cefaclor rather than amoxicillin as the routine first line drug in the treatment of AOM.