Home Page >
Research areas >
Methodology >
Applied statistical methodology
Applied statistical methodology
The validity of the findings of clinical trials and other
clinical studies largely depends on the use of appropriate
statistical methods to analyse the data collected. Inappropriate
methods can lead to misleading conclusions, with consequences for
patients, clinicians and policy makers.
It is also essential to develop and adopt new statistical
techniques to make the design and analysis of trials more efficient
in practice.
Therefore, in recent years, we have established a programme of
applied statistical methodology. This benefits from close links
with other world-renowned academic institutions such as
UCL Statistical Sciences
Department, UCL Research Department of Infection and
Population Health, MRC Biostatistics Unit, and
researchers at the London School of Hygiene and Tropical
Medicine, University of Bristol, University of Melbourne
(Australia), Harvard University, University of
California and University of San Francisco.
The topics covered are of particular relevance to the Unit’s
research programmes and studies:
Improving trial design
Developing more flexible and informative methods to
analyse survival data
Modelling to identify prognostic and predictive
patient factors
Avoiding bias in the analysis of longitudinal
data
Answering questions that cannot be answered by
randomised trials
Reducing bias and properly reflecting uncertainty
from missing data
Design and analysis of trials involving
biomarkers
