[Correspondence] The BLISTER study: possible overestimation of tetracycline efficacy

The BLISTER study, by Hywell C Williams and colleagues (March 6, p 1630),1 indicated that a 25% decrease in the efficacy of tetracycline in the early control of blisters would be acceptable to most UK dermatologists, if accompanied by a reduction of at least 20% in long-term serious side-effects compared with prednisolone. The efficacy of the tetracycline doxycycline was acceptable according to the study's primary effectiveness measure at 6 weeks (upper limit of 90% CI of adjusted difference between treatments [UB], 26 ยท1%, within the 37% predefined acceptable non-inferiority margin [AM]); however, we question the aspects of the study design that appear to favour doxycycline.
Source: LANCET - Category: General Medicine Authors: Tags: Correspondence Source Type: research