Inconsistency and low transparency were found between core outcome set protocol and full text publication: a comparative study

Ming Liu,Ya Gao,Yuan Yuan,Shuzhen Shi,Kelu Yang,Cuncun Lu,Jiarui Wu,Junhua Zhang,Jinhui Tian
DOI: https://doi.org/10.1016/j.jclinepi.2020.11.009
IF: 7.407
2021-03-01
Journal of Clinical Epidemiology
Abstract:<h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Objectives</h3><p>To assess inconsistencies between individual protocols and associated full-text publications in the development of core outcome sets (COSs).</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Study Design and Setting</h3><p>Protocols and subsequent full-text publications were retrieved by searching the following electronic databases: PubMed, Embase, Web of Science (WoS), and the Core Outcome Measures in Effectiveness Trials (COMET) database from inception to October 1, 2019. We summarized changes in the general and methodological characteristics by comparing the protocols with the full-text publications, and reported change as information frequency and proportion.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Results</h3><p>A total of 24 protocols and 32 corresponding full-text publications that encompassed fourteen study topics were identified from there databases. In the identified initial list of outcomes, five COSs (20.8%) changed the included study type, none of which explained the reasons for these changes. Additionally, eight COSs showed inconsistencies between the protocols and full-text publications in the searched databases, of which, only two studies explained the reasons for these changes. Compared to the protocols, three COSs changed the number of Delphi rounds, eight COSs changed the participants (stakeholder groups), and three COSs changed the consensus definition of the Delphi survey. Only two COSs explained the reason for changing the number of Delphi rounds, and none of the studies explained why the participants changed. For the face-to-face consensus meeting, we found that nine COSs changed the participants and none explained the reasons for these changes.</p><h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Conclusion</h3><p>Our study found many inconsistencies between protocols and the full-text publications concerning COS development. These inconsistencies related to the included study types, databases searched, Delphi surveys, and face-to-face consensus meetings. As it is necessary to publish protocols before developing COSs, transparency regarding any changes to the methods is needed.</p>
public, environmental & occupational health,health care sciences & services
What problem does this paper attempt to address?