CONSORT adoption and quality of reporting of randomized controlled trials: a systematic analysis in two dermatology journals

Background  CONSORT (Consolidated Standards for Reporting Trials) guidelines were constructed to ensure optimal reporting quality of randomized controlled trials (RCTs).
Objectives  To determine the effect of the adoption of CONSORT on the reporting quality of RCTs, we performed a systematic evaluation of RCTs published in two dermatology journals pre- and post-CONSORT adoption.
Methods  The journals selected for the study were the Journal of the American Academy of Dermatology and the British Journal of Dermatology . We selected RCTs published in 1997 and 2006 using both Medline and hand searching. The following critical CONSORT criteria were recorded: sample size, type of disease studied, type of control, single-centre or multicentre study, type of funding, blinding, methods and type of randomization, definition of a primary endpoint, justification for sample size selection and power calculation, population for analysis, and adequacy of group comparison. A multivariable analysis was conducted to determine factors associated with optimal reporting quality.
Results  In total, 98 studies were included. Improvement in reporting quality was evident for the specification of the randomization method (20% in 1997 vs. 45% in 2006, P  <   0·01) and for the justification of sample size (22% in 1997 vs. 43% in 2006, P  = 0·027). The percentage of studies with optimal reporting quality increased from 11% in 1997 to 28% in 2006 ( P  =   0·03). Factors significantly associated with a good methodological quality were pharmaceutical industry funding and publication in 2006 vs. 1997.
Conclusions  There is a need to improve the reporting quality of RCTs published in dermatology journals.