Determining the reporting quality of RCTs in clinical pharmacology.

Abstract

The Consolidated Standards for Reporting Trials (CONSORT) are recommendations for improving the quality of reports of randomized controlled trials (RCTs). To determine the extent to which clinical pharmacology journals implement specific CONSORT recommendations. Analysis of RCTs published between May 2002 and May 2003 in four clinical pharmacology journals. Proportion of RCTs that published a participant flow diagram and that reported on randomization and restriction methods, allocation concealment, blinding, intention-to-treat analysis, ethical considerations, adverse events and source of funding. Of 482 clinical trials, 193 were RCTs. Healthy participants were involved in 129 [66.8%, 95% confidence interval (CI) 59.9, 73.1] trials, patients who required treatment in 61 (31.6%, 95% CI 25.4, 38.4) trials and both in three (1.6%, 95% CI 0.5, 4.4) trials. The following items were infrequently reported: sequence generation (17%), allocation concealment (3%), use of restriction (16%), description of blinding (26%), and flow diagrams of study participants (2%). In contrast, the following areas were often reported: use of intention-to-treat analysis (79%), description of withdrawals (92.2%) and description of adverse events (71%), ethics review (94%) and informed consent (95%). Sources of funding were reported in 56% of studies. The use of the selected CONSORT items is limited in these journals, possibly as many items may not be relevant to the types of studies published in clinical pharmacology journals. Further efforts are required to determine the applicability of CONSORT to RCTs in clinical pharmacology.