In reply
We appreciate this analysis of our study comparing 15% azelaic acid gel with 0.75% metronidazole gel. We believe that our study provides useful data about an important new topical therapy for our patients with rosacea.
As Drs Czernielewski and Liu point out, this comparative study showed a statistically significant difference in favor of 15% azelaic acid gel in terms of reduction of lesions and erythema. However, they question certain aspects of the prestudy assumptions, which they suggest affected the power of the study. It is important to understand that "the power of a trial is a useful concept when planning the [a] trial but has little relevance to the interpretation of its results."1 In planning this trial, the patient number was calculated using prestudy assumptions. The planned patient number ensured detection of assumed differences with a predetermined probability (power). However, as in any other trial, when the prestudy assumptions are not met, this may change the power, but does not invalidate the findings of statistically significant treatment differences, as were found in this study.