Email Address: email@example.com
Other Research Theme Collaborations: DART ESH SGX
BSU Research overviewMany diseases and medical conditions, such as cancers, dementia and rheumatic disorders, are multi-factorial and may exhibit a range of complex biological phenotypes reflecting the contribution of a multitude of genetic and environmental components. Even with rare or ultra orphan diseases, such as Gaucher, which may be the result of mutations in a single gene, varying symptomatology, outcomes and treatment responses can be displayed across patients with the disease. The complexity of diseases leads to many challenges ranging from the understanding of disease mechanisms and disease susceptibility to risk prediction and the development and application of treatments. Stratified medicine, where "homogeneous" groups of people likely to respond similarly to treatment or have similar underlying disease mechanism or outcome risk are sought based on molecular (biomarker) information, is a first step towards reducing biological variability and lies within the continuum of "patient therapy", in which empirical or "all comer" medicine sits at one end and precision medicine at the other. The fact that genetic, molecular and imaging modalities have transformed biology from an observational science to a more data and computationally intensive quantitative science has meant that the need for those who are trained in handling data, accounting for uncertainty and making inference is ever more paramount. Many of the issues and challenges confronting stratified medicine are of a statistical nature. We are undertaking a programme of research that encompasses risk stratification, prediction and validation, integrative and joint modelling of molecular and clinical data of various kinds and complexities (and their conversion into meaningful outputs that can inform health care decisions), mechanistic understanding and causality, treatment strategies and the design of innovative/purposeful clinical trials for biomarkers and dynamic treatment regimes.Brian Tom's personal page
Selected PapersTom B D M, Farewell V T & Bird S M. (2014)Maximum and pseudo likelihood approaches for parametric time-to-event analysis for referral cohorts with informative entry times.
Annals of Applied Statistics 8: 726-746
Farewell V T & Tom B D M. (2014)The versatility of multi-state models for the analysis of longitudinal Data.
Lifetime Data Analysis 20: 51-75
van den Hout A & Tom B D M. (2013)Survival Analysis and the Frailty Model - The effect of education on survival and disability for older men in England and Wales.
The SAGE Handbook of Multilevel Modeling Chapter 30: 541-558
O'Keeffe A G, Tom B D M & Farewell V T. (2013)Mixture distributions in multi-state modelling: Some considerations in a study of psoriatic arthritis.
Statistics in Medicine 32: 600-619
Husted J A, Tom B D M, Farewell V T & Gladman D D. (2012)A longitudinal study of the bi-directional association between pain and depressive symptomatology in patients with psoriatic arthritis.
Arthritis Care & Research 64: (5): 758-756