Dr Matthew Smith
Research Fellow in Statistics
I am a Postdoctoral Research Fellow of Biostatistics within the Inequalities in Cancer Outcomes Network at the LSHTM. I obtained my PhD from LSHTM in 2021, which focused on the socioeconomic inequalities in survival of patients with non-Hodgkin lymphoma. Prior to this, I obtained my MSc in Medical Statistics from LSHTM and a BSc in Mathematics whilst at Nottingham.
Currently, I am funded by the Medical Research Council (Better Methods, Better Research award) to develop, implement, and disseminate double-robust causal inference methodology amongst applied researchers in public health, health economics, clinical sciences, and epidemiology.
I hold an honorary research fellow position within the Institute of Health Informatics at University College London.
I am a co-module organiser for the Causal Inference and Missing Data module taught on the MSc Medical Statistics course.
I also facilitate practical sessions for several modules on the MSc Medical Statistics course, namely (i) Generalised Linear Models and (ii) Analysis of Hierarchical and Other Dependent Data.
I supervise one PhD student conducting research into the relationship between mechanical ventilation and health outcomes for patients within intensive care units.
I am a tutor for MSc Medical Statistics students and a supervisor for their summer projects.
My experience is in applying advanced statistical methods to answer epidemiological research questions regarding cancer patient health outcomes, incorporating the investigation of the patient and healthcare system interactions, evaluating interventions to reduce inequalities, and recommending key policy decisions.
In 2021, I received the Better Methods, Better Research award, jointly funded by the Medical Research Council and the National Institute of Health Research. During this 3-year programme I will advance methods in causal inference (e.g., targeted maximum likelihood estimation) for wider dissemination and reproducibility in applied research.
My additional interests are in estimating causal effects within the relative survival setting, emulating target trials, handling missing data by using multiple imputation in observational studies, and investigating socioeconomic inequalities in the management of patients with non-Hodgkin lymphoma.