I am currently Reader in Medical Statistics in the Arthritis Research UK Epidemiology Unit. I have been with the Unit since 1999, although I spent the academic year 2006-2007 on sabbatical with the Pharmacoepidemiology Group of Harvard University's Medical School, studying causal inference.

Having completed a B.Sc. in mathematics from Warwick University in 1985, and a PGCE in 1987, I started my career as a secondary school maths teacher, first in Liverpool and then in Liberia. However, my teaching career was interrupted by the civil war in Liberia, so I spent a short while working for Medicins sans Frontieres. During this time, I was faced with a number of interesting questions that could have been answered using statistics (e.g. is the increase in cases of diarrhoea in this village normal seasonal variation, or do we need to dig a new well ?) but I did not know enough stats to answer them. So on returning to England, I took and M.Sc. in Medical Statistics in London School of Hygiene and Tropical Medicine.

My first job after qualifying in 1993 was  as study statistician with the European Prospective Osteoporosis Study (EPOS), based in Cambridge. My main focus at this time was on the epidemiology of osteoporosis: identifying risk factors for reduced bone density and vertebral fractures. At this time I began my Ph.D. with the Open University, on statistical methods of identifying vertebral fractures. I completed this in 2003, after moving to Manchester in 1999. I was promoted to Senior Lecturer in 2004.

My primary research interest is estimating the effect of treatment from observational studies (i.e. looking at subjects who receive different treatments for clinical reasons, rather than from a randomised clinical trial). In a randomised trial, a given person is equally likely to receive either treatment. However, in an observational study, subjects with more active disease are more likely to receive treatment, and are also more likely to have a poor outcome. This can lead to the treatment correlating with poor outcomes, even if it does provide benefit to the patient. This is called "Confounding by indication" (confounding is when two variables are correlated not because one causes the other, but because they are both caused by a third variable).

In order find out what effect the treatment has, we need to compare the outcome in treated subjects to the outcome they would have had if they had not been treated. Finding ways to estimate their expected outcome if they had not been treated is the focus of my research. Most established methods for doing this revolve around the propensity score (the probability that a given person will receive treatment, calculated from all of the potential confounders that were measured). Two subjects with the same propensity score can be expected to have the same outcome if they were not treated, so comparing treated and untreated subjects with the same propensity score can give an unbiased estimate of the effect of treatment.

Postgraduate Opportunities

Although the performance of various propensity-score based estimators is know when the predictor variables follow known distributions, less is known about how the estimates degrade when the predictors do not follow the assumed distribution. Since real data is likely to deviate from the theoretical distribution to some extent, it would be useful to know how these deviations affect different estimators, and to compare the robustness of estimators.

I run a course entitle "Statistical Modelling in Stata", which covers the basics of statistical inference and fitting and interpreting statistical models. This course runs from September to December each year, and is intended primarily as in-service training for researchers in the Arthritis Research UK Epidemiology Unit, but is open to other staff and students as well. All of the course materials are available here.

In addition, I teach statistics sessions on the M.Sc Rheumatology course and the Translational Medicine M.Res, as well as the "Introduction to Epidemiology, Genetic Epidemiology and Biostatistics" course run by the Arthritis Research UK Epidemiology Unit.

Most of my work revolves around estimating the effect of some kind of exposure (often a treatment of some description) on an outcome. Generally, there will be some confounding (i.e. the exposure and the outcome are both affected by a third variable, such as age or disease activity). Methods used to get around the confounding include generalised linear models, propensity methods and instrumental variables. In addition, in most studies, there is some missing data, so I am interested in methods that can eliminate the bias that can arise when the probability that data is missing is affected by the outcome, exposure or confounding variables.