Accurate information on harms of medical interventions is essential for evidence-based practice. Most newly introduced treatments usually have small, incremental benefits, if any, against already available interventions, and differences in the profile of harms should play a key role on treatment choice. Randomized trials offer an excellent opportunity to evaluate harms of interventions using the most robust experimental design available in clinical research. However, several empirical evaluations (Table) have shown that many trials do not report harms or report them in a fragmented or suboptimal way. In this issue, an excellent study by Pitrou et al1 adds more evidence on this issue.
Ioannidis JPA. Adverse Events in Randomized TrialsNeglected, Restricted, Distorted, and Silenced. Arch Intern Med. 2009;169(19):1737-1739. doi:10.1001/archinternmed.2009.313