Background: randomized controlled trials (RCTs), when appropriately designed, conducted, and reported, are the gold standard evidence synthesized in systematic reviews of treatment efficacy and effectiveness. To be able to critically evaluate and use the results of RCTs included in systematic reviews and meta-analyses, readers require complete and transparent information with respect to what was planned, what was done, and what was found. Inadequate reporting of RCTs is well documented in the medical literature, particularly for study outcomes. Key information about the selection process, definition, measurement, and analysis of outcomes are often missing or poorly reported in trial protocols and reports. This enables the introduction of bias by facilitating selective outcome reporting and switching. It also impairs reproducibility, critical appraisal, systematic review production, undermining uptake of trial results into clinical practice.
Objectives: this international project developed the SPIRIT- and CONSORT-Outcomes reporting guidelines providing harmonized guidance for describing RCTs’ outcomes in protocols and reports, respectively.
Methods: we developed the SPIRIT- and CONSORT-Outcomes reporting guidance using the EQUATOR framework for health research reporting guidelines. This included the generation and evaluation of candidate outcome reporting items via expert consultations, a scoping review, a three-round international Delphi survey, and a two-day, in-person, expert consensus meeting.
Results: we identified 133 outcome reporting items from the scoping review and expert consultations, the majority of which are not currently included in CONSORT or SPIRIT. Items were consolidated into 67 candidate items for Delphi voting, which was completed by 124 participants from 22 countries. After Delphi, 19 items met criteria for further evaluation at the Consensus Meeting for trial protocols, 30 for trial reports. The consensus meeting yielded 21 extension (added to checklists) and elaboration (on existing statements) items: eight for SPIRIT-Outcomes (e.g. composite outcomes components, outcome assessors, outcome timepoint analyses, threshold cut-offs, effect size relevance) and 13 for CONSORT-Outcomes (e.g. outcome definition, rationale, choice of instrument, assessor training, missing outcome data causes and solutions, unplanned analyses).
Conclusions: SPIRIT-Outcomes and CONSORT-Outcomes provide new frameworks aimed to enhance trial transparency in areas known to be associated with outcome reporting bias.
Patient or healthcare consumer involvement: the patient member, the public member, research funder, and industry representatives attended the consensus meeting, helped finalize item wording, and gave advice on implementation and endorsement strategies. The team leaders met with the patient member and public member several times before the consensus meeting to discuss their participation and ensure their questions were clearly and fully answered.