Review of use of the Trials within Cohorts (TwiCs) design approach

Contains fulltext : 209491.pdf (publisher's version ) (Open Access)
Introduction: Trials within Cohorts (TwiCs) is an innovative approach to the design and conduct of multiple randomised controlled trials (RCTs) (Relton et al, 2010). This approach utilises an observational cohort to recruit trial populations and obtain short and longer term outcomes. We describe what is currently known about the use of this design approach. Methods: An extension of the 2010 Consolidated Standards of Reporting Trials (CONSORT) Statements for RCTs using cohorts and/or routinely collected health data is in development, supported by a scoping review that includes publications of methods or reports ofprotocols or results from RCTs using cohorts, registries, electronic health records and administrative databases. Data sources for this scoping review included Medline and Cochrane Methodology Register and were limited to English language. This review of use of the TwiCs approach uses publications of methods or reports of protocols or results from RCTs that use cohorts to recruit identified in the scoping review. This is supplemented with information from topic experts. We report: (i) types of cohorts (setting, population, condition/ disease area), (ii) how the cohorts are utilised (identifying potential trial participants, recruitment, randomisation, process and outcome data collection including bespoke and/or routine health record data, types of trials conducted/ planned), (iii) approaches to informed consent, e.g. staged approach (Young-Afat et al, 2016), and (iv) any purported and/or real study design (in)efficiencies. Timing of Potential Results: Early results indicate 75+ eligible full text articles, including 23 trial protocols and 23 articles reporting the results of trials using cohorts. Full results will be available in August 2019 and presented at the conference. Potential Relevance and Impact: Standard approaches to trial design are often costly and frequently fail to recruit sufficiently large or representative samples. This review will help pro