The department has a number of methodological research interests and our members are engaged in a broad range of collaborative projects. The main research areas are summarised below.

  • Clinical trials and Methodology Research


    The Clinical Trials Research Centre (CTRC) provides a support and advice service to NHS professionals and academic researchers for NHS research. CTRC supports phase I - IV multi centre randomised controlled trials, systematic reviews, core outcome sets, service evaluation and other research. CTRC is part of the Liverpool Trials Collaborative, which is fully registered with the UKCRC.

    Visit the CTRC website for further information.

    The MRC North West Hub for Trials Methodology Research (NWHTMR) aims to facilitate and accelerate the translational process through the development of new and improved approaches for the design, conduct and analysis of clinical trials, particularly in methodologically challenging areas. 

The MRC NWHTMR is a collaboration between the Universities of Liverpool, Lancaster, Bangor and Manchester, involving experts in medical statistics, clinical trial methodology, pharmacoeconomics, sociology and clinical psychology.

    Visit the MRC NWHTMR website for further information.

  • Multivariate statistical modelling


    We have established a research group in the area of multivariate modelling aimed to develop methodology and disseminate concepts among researchers and clinicians. The Liverpool multivariate modelling group has expertise in multivariate methodologies, their application to clinical data, and the development of code in R and in other statistical packages.

    Multivariate data modelling is often applied to address research questions such as ‘are there a small number of underlying factors explaining the variables in the dataset?’, ‘is there a natural clustering in the dataset?’, ‘can we classify individuals into groups based on their characteristics?’ or ‘can we predict that a person will develop a particular condition/disease by looking at a range of symptoms?’

    Commonly applied multivariate techniques are discriminant function analysis, cluster analysis, principal components analysis, factor analysis, multidimensional scaling and MANOVA amongst others.

    Visit the Liverpool multivariate modeling group website for further information.

  • Meta-analysis


    Several members of the department are involved with methodological and applied research in the area of meta-analysis. Areas of particular interest to the group are

    • Issues surrounding meta-analyses of Individual Patient Data
    • Within-study selective reporting bias
    • Dealing with missing data at the patient level, including issues around non intention to treat analysis
    • Publication bias
    • Comparing individual patient data and aggregate data meta-analyses
    • Meta-analysis of method comparison studies
    • Meta-analysis of longitudinal data
    • Meta-analysis of time-to-event data
  • Pharmacogenetics


    Very few medicines are found to be effective for all individuals and there can be great variation in the degree to which people respond to them. Whilst for some they may be effective, others may experience adverse reactions, which in some cases may be fatal. There are several reasons why medicines may be ineffective or cause adverse side effects, such as inaccurate prescribing, lack of compliance by patients and interaction with other medications. However, it is also known that genetic make-up can influence a patient’s response to a drug.

    Pharmacogenetics is the study of how this genetic make-up determines drug response, looking in particular at genes that encode drug targets and enzymes that mediate drug metabolism and transport.

    Statistical methodology for investigating genetic association is a field that is developing at a rapid pace, however there is not as yet consensus as to the most effective methods. Although we can use some of the traditional statistical methodologies to analyse pharmacogenetic studies, these are not sufficient on their own to deal with the complexities associated with the modelling of relationships between genetic profile and drug response. We have research interests in the statistical analysis of pharmacogenetic studies, including the following:

    • Statistical analysis of pharmacogenetic studies in a time to event setting;
    • Meta-analysis of pharmacogenetic studies;
    • Quality assessment of pharmacogenetic studies.

    Examples of our application of statistical methods to pharmacogenetics studies are as follows:

    • Association between genetic and environmental factors and inter-individual variability in response to Warfarin
    • Pharmacogenetis of GABaergic mechanisms of benefit and harm in epilepsy
    • Genetic substudy of the following randomised controlled trial:
    • 'SANAD; Study of Standard versus New Antiepileptic Drugs'
    • Impact of genetic polymorphisms on the efficacy and toxicity of Tramadol
    • Metformin and genetic polymorphisms of organic cation transporters
    • Impact of variation at the HLA-B locus in Caucasian patients on carbamazepine hypersensitivity.
    • Genetic extension to the following randomised controlled trial: ‘The use of MElatonin in children with Neuro-developmental Disorders and impaired Sleep’
    • Clinical and Functional Evidence That Drug Transporter Ecpression Dictates the Response to Imatinib in Chronic Myeloid Leukaemia.
  • Statistical genetics


    The primary aim of research in the statistical genetics group is the development and evaluation of novel methodology for the analysis of the “next-generation” of genome-wide association studies (GWAS) of complex human traits. We consider common and rare genetic variation from diverse ethnic groups, interrogated through traditional GWAS genotyping arrays, supplemented by imputation, and via state-of-the-art whole-exome or whole-genome re-sequencing studies. Currently, we are addressing a number of key challenges in attempting to identify genetic variation contributing to the “missing heritability” of complex traits:

    1. Genome-wide analysis of rare variants, focussing on “burden testing” within genes, pathways, or other functional units;
    2. Trans-ethnic meta-analysis of GWAS from multiple ancestry groups;
    3. Fine-mapping of complex trait loci to localise the underlying causal variants;
    4. Methodology for the analysis of “non-standard” traits, including “time-to-event” outcomes and multiple correlated measures;
    5. Dissection of multi-trait association signals at GWAS loci.

    The methodology developed by the statistical genetics group is currently being applied to next-generation GWAS of a range of complex traits, with a focus on cutting edge studies investigating the genetic basis of: type 2 diabetes (T2D) and related metabolic phenotypes; women’s health disorders including endometriosis and polycystic ovary syndrome; and drug response including carbamazepine hypersensitivity.

    The findings of our research will contribute to evaluating the utility of next-generation GWAS of common and rare genetic variation for complex trait gene mapping, and will ultimately provide a more comprehensive view of the genetic architecture of common human diseases and drug response.

  • Stereology


    Stereology is a multidisciplinary science aimed at estimating, from a proper sampling design, geometrical parameters of a spatial structure such as volume, surface area, length, curvature, number of neurons or cells. The spectacular technical progress undergone by non-invasive scanning techniques (X-ray, computed tomography, magnetic resonance) has contributed much to the development and the implementation of stereological methods.

    Within the University of Liverpool, we have established close collaboration with the Magnetic Resonance and Image Analysis Research Centre (MARIARC) and the Department of Medical Imaging. We have also established national (Oxford, Edinburgh, Manchester) and international collaboration (Denmark, Spain, Chile, France, Turkey).

    In particular, stereological methods have been applied in combination with magnetic resonance imaging techniques to the following problems:

    • Volume estimation of internal human brain compartments.
    • Surface area estimation of human brain structure.
    • Quantification of tumour response to radiotherapy.
    • Estimation of body composition in muscular dystrophy.
    • Estimation of breast volume in patients with cyclical mastalgia.
    • Quantification of hippocampal volume reduction in patients with temporal lobe epilepsy.
  • Survival analysis


    The analysis of time-to-event data, commonly referred to as survival analysis, is relevant to many clinical studies where the outcome of interest relates to the time taken for some event to occur, e.g. time to first seizure or time to death following randomisation. Our main research in this area includes the meta-analysis of studies involving time-to-event outcomes, hierarchical Cox regression modelling, competing risks in anti-epileptic drug trials, indirect comparisons and general issues of analysis in epilepsy trials.