With over 90 staff and students, the MRC Biostatistics Unit (BSU) is one of the largest groups of biostatisticians in Europe, and a major centre for research, training and knowledge transfer in biostatistics. The Unit was founded in 1913 at the same time as the Medical Research Council. The BSU is internationally renowned for the strength of its research and its successful strategy of maintaining a unique balance between statistical innovation, dissemination and engagement with making impact in biomedical and public health issues. The BSU’s mission is to promote the development and application of innovative statistical methods in the health sciences for the improvement of health.
BSU researchers have worked extensively to make inference for complex data accessible to the scientific community and to produce innovative methodology related to trial designs, longitudinal and event history data, meta-analysis of clinical trials, missing data, evidence synthesis in public health and statistical genetics and genomics. We are recognised for our strength in Bayesian inference applied to biomedicine and public health. The BSU current research portfolio is organised into four main Themes that span the scientific research spectrum from basic science to population health and respond to current scientific needs in biomedicine.
- Statistical Omics (SOMX) is centred on developing and implementing new analytical strategies for integrative and translational genomics as well as epidemiological studies enriched with genetics data, opening up to statistical issues arising from newomics technologies. Issues of efficient inference and scalability of computations for big data are a particular focus.
- Precision Medicine and Inference for Complex Outcomes (PREM) focuses on the methodological issues arising from the development of precision medicine, the need to stratify complex disease phenotypes and to perform integrative clustering based on multiple data types, to tailor risk prediction and to analyse complex longitudinal data.
- Design and Analysis of Randomised Trials (DART) develops adaptive and novel trial designs that are ever more efficient at flexibly answering the key clinical questions of interest. It targets the translation of methods into practice with software production and dissemination activities.
- Statistical methods Using data Resources to improve Population Health (SURPH) is directed to tackle the statistical challenges in addressing causality (disease aetiology), estimating and predicting disease burden and evaluating public health policies using data from publicly-available sources which are of different types, potentially subject to selection biases and likely of large dimension.
Each of our themes is anchored in important scientific questions arising from our key partners, questions which both motivate methodological developments and ensure their impact. We work with leading trial centres and epidemiology groups and public health bodies such as Public Health England. We are developing links with major programmes in stratified medicine and in big data analytics such as Health Data Research UK and the Alan Turing institute.
Director: Professor Sylvia Richardson
“Statistics is applicable in all aspects of medicine, epidemiology and public health,” says Sylvia Richardson, Director of the MRC BSU, since 2012. “It is fundamental for designing clinical trials, modelling disease programmes, asserting the influence of the genetic make-up of our health, as well as evaluating the effectiveness of public health policies.”