Reporting biases

Reporting biases occur when the spread of research findings is influenced by the nature and direction of results, either on the level of publications (publication bias, i.e. (non-)publication of research findings, or (rapid or delayed) timing of publication), or on the level of outcomes within studies (selective outcome reporting) (Chan 2004; Hopewell 2008; Page 2014). As systematic reviews are built upon existing research findings, they are susceptible for these kind of biases. Therefore, for a systematic review extensive searches should be undertaken (as described above) to identify as many relevant studies as possible, regardless of whether published or not. In addition, to identify possible selective outcome reporting trial register records can be used to look for discrepancies in the reported outcomes listed in the protocol and the reported outcomes in the publication of a study.