Statistical teams with extensive applied, computational and methodological expertise underpin the rigorous analysis and interpretation of large-scale data for which CTSU is renowned. The innovative work of CTSU in establishing large-scale clinical trials and large-scale biobanks over the last few decades has hinged on statistical and clinical vision and leadership; and our senior statistical scientists continue to shape the future direction of the Unit in both the development of research themes and of the skills required to exploit the unique datasets available. The statistical teams work in close collaboration with NDPH clinicians and epidemiologists and take key roles in population studies, such as the the China Kadoorie Biobank, the Mexico Biobank, and Cuban biobank as well as in large-scale clinical trials. These data resources provide unique opportunitues to undertake research in vascular-related diseases (including heart disease, stroke, diabetes, obesity, dementia) and cancer, using epidemiological, genetic and other ‘omics’ techniques. Our statistical teams, which include medical statisticians, genetic epidemiologists, statistical programmers, and bioinformaticians exploit their skills in various ways including:
Mega-trials, meta-analysis and trials methodology
Statisticians play a key role in all stages of our clinical trials, including design, conduct, data analysis, interpretation, publication and submission to regulatory authorities, as well as serving on Data Monitoring and Steering Committees. In addition, we undertake trials-related methodological research aimed at adopting statistical approaches to streamlining clinical trials. The findings from mega-trials are further extended by our meta-analysis groups that conduct collaborative individual patient meta-analyses to bring together randomised controlled evidence. Implementation of the findings from CTSU’s trials and meta-analyses is saving many tens of thousands of lives worldwide each year.
Genetic epidemiology & statistical genomics
The genetic epidemiology and bioinformatics teams at CTSU lead genomic research in clinical trials and population studies focusing on translational impact for patient care. In this highly collaborative field, we work closely with wider teams within NDPH and the Oxford Wellcome Trust Centre for Human Genetics, as well as other groups from across the UK and internationally, to push forward research in this fast moving area. Interests span the use of genome-wide association studies, large-scale sequencing, epigenetics, transcriptomics, and Mendelian randomization to explore questions related to vascular disease, cancer and pharmacogenomics. In our own studies, and as instrumental partners in large international consortia such as CARDIoGRAMplusC4D and METASTROKE collaboration of the ISGC, we undertake investigations to identify novel genetic associations with disease and to assess causal relationships between risk factors and disease outcomes.
‘Omics, imaging and machine learning
Diverse data sources including ‘omics’ data (metabolomics, lipidomics, proteomics) and imaging data (brain, body, heart, neck artery, eye) are becoming increasingly available in the various studies our statisticians work with. CTSU’s statistical teams are shaping and developing the integrated skills required and collaborating with specialist groups to establish appropriate techniques, such as machine learning, to exploit the resources for our population health aims.
Training and opportunities
CTSU offers a unique opportunity to develop an applied statistics career across a range of research themes in a valued and stimulating environment. We offer DPhil projects, 1-year statistical programming fellowships and departmental and externally sponsored post-doctoral fellowships, as well as statistical research opportunities.
Professor Sir Richard Peto: John Snow bicentennial lecture, LSHTM, 2013. Interpretation of large-scale randomised evidence: Need for reliable assessment of MODERATE effects on mortality
Associate Professor Jemma Hopewell: MRC Hubs for Trials Methodology Research, Webinar, Dec 2014. Monitoring Trials Efficiently: The role of central statistical monitoring.