«There is strong evidence that much of the information on adverse events remains unpublished and that the number and range of adverse events is higher in unpublished than in published versions of the same study»: this is the conclusion of a study recently published on the Open Access journal PLOS Medicine, supported by the National Institute for Health Research. «The extent of “hidden” or “missing” data prevents researchers, clinicians, and patients from gaining a full understanding of harm, and this may lead to incomplete or erroneous judgements on the perceived benefit to harm profile of an intervention» write Su Golder and colleagues from the University of York, in the UK.
The researchers conducted a systematic review to measure the adverse events reported in peer-reviewed publications and in unpublished documents such as regulatory websites, trial registries, as well gray literature such as conference proceedings and press releases. Out of the total of over 5.000 records retrieved, 28 studies met the inclusion criteria from 31 publications: the studies that looked at differences regarding adverse events all concluded that information on adverse events is more often present in unpublished materials than in published material. «There were 8 included studies (from 9 publications and representing 10 comparisons) that compared the number of studies reporting adverse events in the published and matched unpublished documents. The percentage of studies reporting adverse events was higher in all cases in the unpublished versions than the published versions. The median percentage of published sources with adverse events information was 46% compared to 95% of the corresponding unpublished documents. A similar pattern emerged when the analysis was restricted to serious adverse events» the authors write. In detail, peer-reviewed articles failed to report 43% to 100% (median, 64%) of adverse events associated with medical treatments, including 2% to 100% of serious ones.
![(click to enlarge) Percentage of matched published and unpublished studies with adverse event information. *Classified adverse events information as either “completely reported” versus “incompletely reported.” Incompletely reported adverse events could lack numerical data or include only selected adverse events, for example. Maund 2014a [16] and Maund 2014b [16] compare published trials to registry reports and clinical study reports (CSRs), respectively. Riveros 2013 [35] compares trials with number of adverse events reported and classifies adverse events information as either “completely reported” or “incompletely reported.” http://dx.doi.org/10.1371/journal.pmed.1002127.g002](https://archive.cancerworld.net/wp-content/uploads/2016/09/journal.pmed_.1002127.g002.png)
*Classified adverse events information as either “completely reported” versus “incompletely reported.” Incompletely reported adverse events could lack numerical data or include only selected adverse events, for example. Maund 2014a [16] and Maund 2014b [16] compare published trials to registry reports and clinical study reports (CSRs), respectively. Riveros 2013 [35] compares trials with number of adverse events reported and classifies adverse events information as either “completely reported” or “incompletely reported.”
http://dx.doi.org/10.1371/journal.pmed.1002127.g002
Leave a Reply