Mental health trials are still running on outdated designs that slow down access to better treatments. This research programme aims to fix that by adapting the efficient trial methods already used in cancer research—such as adaptive designs that can test multiple treatments at once or switch course mid-trial—for use in mental health, particularly for psychosis and complex psychosocial interventions. The problem is that mental health trials often rely on simpler, slower designs that waste time and resources. The researchers will systematically identify why these efficient designs are not being used, then develop new statistical frameworks and guidelines tailored to mental health. They will also tackle tricky analytic issues like missing data, non-compliance, and how to combine multiple outcomes into a single meaningful measure. If successful, this work could dramatically speed up how quickly new mental health treatments reach patients. It would also produce publicly available guidance and training materials, and establish two national networks to embed these methods permanently. The result is not a new drug or therapy, but a faster, more reliable engine for testing whatever treatments come next.
View original technical description
Background: Patients, clinicians and funders need efficient randomised trials. This applies across all clinical disciplines, in both academia and industry, and both nationally and internationally. Some disciplines, notably oncology, have pioneered more efficient trial designs to address this issue. However these trial designs are not yet fully used in mental health or for complex or psychosocial interventions. Psychosis is an area where implementing them would bring large patient benefit. Aims: To improve the health of patients by introducing efficient trials, with applications in mental health. This programme includes seven objectives: 1. Review the literature on efficient trials and identify barriers to their use and uptake in mental health 2. Adapt existing and develop novel efficient trial designs for patient stratification, multiple treatments and patient-care pathways 3. Incorporate causal modelling into the analysis of efficient trials 4. Produce a unified approach for the statistical analysis of post-randomisation variables in trials 5. Develop a new framework and guidelines for performing efficient trials in mental health 6. Form two new national groups to engage with clinical and methodological communities 7. Undertake a bespoke leadership programme and build research capacity to deliver world-leading clinical trials methodology Methods: There are two methodological work-streams, targeting complementary elements of trials WS1: Improving trial designs We will define efficient trials, and systemically review the current literature. This will include (but is not limited to): adaptive designs for dose-finding, multi-arm multi-stage designs, umbrella trials and sequential multiple assignment randomized trials. We will perform a Delphi survey with patients, clinicians, academics, funders and other stakeholders to identify barriers to implementing these in mental health and for complex interventions. We will use the results to propose extensions to existing designs or create novel designs, on which we will seek further feedback. We will develop a platform for the implementation of these designs. WS2: Improving trial analysis We will review and unify the currently separate analytic approaches for post-randomisation variables, including missing data, non-compliance, mediation and surrogate outcomes. We will consider how analysis of non-compliance and mediation can be performed in efficient trials, in which analysis currently focuses on the intention-to-treat question. We will consider how to develop weighted combinations of multiple outcomes as a primary outcome, assess which methods for interim analysis are appropriate for use with complex interventions, and extend these for intermediate outcomes including neuroimaging and intensive longitudinal data. Methodological advances will be made using statistical theory, Monte Carlo simulation studies, and by application to existing trial datasets. Anticipated impact and dissemination: We will produce guidance documents and online training materials for performing efficient trials in mental health, drawing on the MRC Complex Interventions guidance, but updating it to reflect new methodology. The applicant will have a highly visible leadership role, forming and leading two national groups to work together to implement the research in this proposal and building research capacity. This work addresses the translational gap from developing methodology to implementation of methods for evaluating new treatments, in order to deliver them more expeditiously and effectively.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know