Sir.—Gerber et al1 found no significant difference in the treatment failure rates between patients who received penicillin twice a day (BID) and those who received penicillin three times a day (TID) for group A β-hemolytic streptococcal pharyngitis. They concluded that their findings suggest that the BID regimen is as effective as the TID regimen. However, they failed to consider two other possible and important explanations for their findings.
The first consideration is chance. It is quite possible that the failure rate of the BID regimen was actually significantly greater than that of the TID regimen. However, due to a chance event in the performance of the study or in the selection of the patients, this difference was not detected.
The second consideration is size of the sample. The smaller the sample studied, the harder it is to detect a significant difference between two groups. Again, it is possible