CONSORT Extension for Trials Conducted in Existing Data Structures, Including Researcher-generated Cohorts, Registries, Electronic Health Records, and Administrative Databases

Large-scale randomized trials (RCTs) are complex and expensive to perform. Less than one-third of RCTs achieve their planned recruitment target, follow-up of participants is labor intensive, and results from trials often cannot be generalized to real-world settings. To overcome these challenges, an increasingly common approach involves conducting trials within existing data structures, including trials with participants from researcher-generated cohorts, registries, electronic health records and administrative databases. These designs, however, reflect relatively recent innovations, and published trial reports often do not adequately describe important aspects of trials, such as steps taken in recruitment and maintenance of the cohort, information provided or not provided to participants about trial participation, and how participants are selected and allocated to trial arms. Reporting guidelines increase the reporting quality and utility of research, but there is no guideline for trials embedded in existing data structures. The goal of the proposed project is to develop and disseminate a reporting guideline, as an extension of the Consolidated Standards of Reporting Trials (CONSORT) statement for trials conducted in existing data structures, including researcher-generated cohorts, registries, electronic health records, and administrative databases. We will systematically review guidance on conducting and reporting these types of trials and review published trials to identify reporting gaps. From this, we will generate a list of potential modifications to items in the standard CONSORT statement checklist and potential new items. These will be vetted via a Delphi exercise with team members and stakeholders and at a face-to-face meeting. The CONSORT extension for trials embedded in existing data structures, which we will produce and disseminate, will contribute to transparent reporting of these trials.