menu

Pragmatic Clinical Trials Unit

Research menu

Trials within cohorts (TwiCs)

Trials within Cohorts (TwiCs) is an innovative approach to research that combines the strengths of observational research (cohorts) and experimental research (the randomised trial).

Most randomised trial designs in healthcare use a single trial approach, i.e. for each trial a bespoke infrastructure is created in order to identify, recruit and retain participants and obtain outcomes. At the end of each trial, all this infrastructure is disbanded. In contrast, the TwiCs approach to randomised trial design uses a single cohort infrastructure which enables participants to be identified and outcomes obtained for multiple trials.

The TwiCs approach was first described in the BMJ in 2010 as the ‘cohort multiple’ RCT (cmRCT) design (Relton & Nickol et al). In this design:

A. Participants enrol in a longitudinal observational cohort with regular outcome measurement. This cohort then provides a framework for the implementation of multiple RCTs. Then for each RCT embedded in the cohort, those eligible for the RCT are identified, and a random selection of RCT eligible patients are allocated to the intervention. Outcomes of those randomly selected to the intervention group are compared to outcomes of RCT-eligible patients not randomly selected to the intervention, i.e. those who receive usual care.

B. As all participants have agreed to participate in longitudinal research, additional consent is only sought from participants randomised to any non-standard care intervention. This ‘patient centred’ approach aims to replicate informed consent practices as they would be applied in clinical care, i.e., patients are only informed about interventions if they are able to access  them (Relton et al, 2010). The ‘staged-consent approach’ (Young-Afat et al, 2016) further refined the ‘patient’ centred approach by seeking consent at cohort enrolment for participants to not be informed if they are allocated to any RCT usual care control group at a later date. This they describe as the 'staged consent approach'.

The TwiCs approach is particularly well suited to pragmatic trials – trials which help choose between care options and help evaluate the effect of treatment in routine clinical practice. There is growing use of the TwiCs approach in both clinical and population health research. An extension to CONSORT Reporting guidelines for Trials Using Cohorts and Routinely Collected Data are currently being developed (Kwakkenbos et al, 2018).

We also holds talks and introductory courses on TwiCs. Our most recent international meetings were Symposium on the Efficiency and Analysis of TwiCs, (2019) and Symposium on the Ethics of TwiCs. 

https://www.qmul.ac.uk/pctu/courses-and-events/

For more information contact c.relton@qmul.ac.uk or b.nickolls@qmul.ac.uk

Specific topics of our work include:

  •  Recruitment & Retention
  •  Informed Consent & Ethics
  •  Efficiency and Analysis
  •  Reporting Guidelines
  •  Mega-cohorts
Return to top