Conduct and analysis of trials, meta-analyses and observational studies
We develop and evaluate methods for the statistical analysis of clinical trials, such that the results are robust and make best use of the available data.
Our methodology research aims to provide methodologically innovative and practically relevant solutions to the challenges faced in current clinical research.
These challenges include:
- Finding new trial designs to speed up the evaluation of therapies
- Identifying cost-effective, evidence-based improvements to trial conduct
- Improving the analysis of trials
- Speeding up reliable evidence synthesis
Below is a summary of highlights achieved by our four programmes: Design, Conduct, Analysis and Meta-analysis.
We created and pioneered the multi-arm multi-stage (MAMS) design. We are experts in setting up, running, analysing and communicating MAMS/platform trials. We also created and pioneered the MAMS-ROCI design for optimising a quantitative aspect of treatment (such as its duration), and the PRACTical design for ranking a set of treatments.
This work will be continued and expanded by the new MRC Centre of Research Excellence in Clinical Trial Innovation (CCTI)
Read more about the Design programme at the MRC Clinical Trials Unit at UCL
We generate evidence to determine the best ways to deliver clinical trials, amassing and sharing our knowledge on trial monitoring, running platform trials and how clinical trial data can be re-used in further research. We have highlighted the need to assess integrity and provenance of healthcare systems datasets and evaluate their utility for use in clinical trials. We have also developed guidance around communicating results to trial participants.
Read more about the Conduct programme at the MRC Clinical Trials Unit at UCL
We developed definitions of estimands for complex clinical trial designs and are working on how to estimate estimands that lack established methods of analysis. We introduced a new framework for planning and reporting simulation studies in statistical research and have developed better ways to handle missing data.
Read more about the Analysis programme at the MRC Clinical Trials Unit at UCL
We have developed and are applying a prospective and collaborative framework (FAME) to ensure faster, more reliable meta-analyses of emerging trial results. We are combining current best-practice analytical approaches for the analysis of individual participant data with the development of novel methods to address specific issues. This ensures we obtain reliable and nuanced answers to important clinical questions.
Read more about the Meta-analysis programme at the MRC Clinical Trials Unit at UCL