Dr Thomas Cowling


Assistant Professor
in Clinical Epidemiology

15-17 Tavistock Place
United Kingdom

I am Assistant Professor in Clinical Epidemiology in the Department of Health Services Research and Policy, Faculty of Public Health and Policy.

My research currently focuses on measures of patient morbidity and the prognosis of cancer patients, using national datasets of electronic healthcare records (see 'Research' below). This work is supported by a Medical Research Council (MRC) Skills Development Fellowship (2018-2021).

I joined LSHTM in 2017 from Imperial College London where I completed degrees in Biomedical Sciences (BSc 2011), Public Health (MPH 2012), and Health Services Research (PhD 2016). My PhD was funded by a National Institute for Health Research (NIHR) Doctoral Research Fellowship.


Department of Health Services Research and Policy
Faculty of Public Health and Policy


Centre for Global Chronic Conditions
Centre for Statistical Methodology


MSc Health Data Science: I am part of the Programme Committee that leads this MSc and am a Module Organiser for the Data Challenge module.

MSc Public Health: I am a personal tutor and have taught on the Health Care Evaluation, Health Services, and Principles and Practice of Public Health modules.

I supervise three PhD students:

  • Matthew Parry (Specialist Registrar in Urology) - Radiotherapy treatment strategies for patients with locally advanced prostate cancer: a national study using routinely collected hospital data and patient-reported outcome measures
  • Jemma Boyle (Specialist Registrar in General Surgery) - Using national registry data to describe current practices and outcomes in the use of systemic anti-cancer therapy in the management of colorectal cancer in England
  • David Wallace (Specialist Registrar in General Surgery) - Liver transplantation as treatment for hepatocellular carcinoma: a national study using linked electronic healthcare records

Current and prospective students are very welcome to contact me.


My research is contextualised by three main trends:

  • Patients' health profiles have become more complex, on average, due to population ageing and greater levels of multimorbidity
  • In the 'Big Data' era, large datasets of electronic healthcare records are increasingly available for research and linked to other datasets
  • Flexible methods for estimating models, from both the statistics and machine learning communities, are receiving greater interest

I am investigating new approaches to modelling patient morbidity in electronic healthcare records, using more of the available data than is used by conventional methods.

This work has included approaches for selecting a small set of diagnosis codes from much larger sets of codes and for modelling many interactions between health conditions using machine learning approaches.

I am applying this methodological work to develop prognostic models for cancer patients, using national cancer registry and hospital records from England. These models have so far focused on colorectal cancer.

The PhD students whom I supervise are studying the factors that predict the treatments received by prostate, colorectal, and liver cancer patients and how outcomes differ between treatment groups.

I am a member of the Project Teams of the National Bowel Cancer Audit and National Prostate Cancer Audit.

Research Area
Clinical care
Clinical databases
Health care policy
Health inequalities
Health outcomes
Health services research
Statistical methods
Electronic health records

Selected Publications

Logistic regression and machine learning predicted patient mortality from large sets of diagnosis codes comparably.
Cowling TE; Cromwell DA; Bellot A; Sharples LD; van der Meulen J
Journal of Clinical Epidemiology
Determinants of Variation in the Use of Adjuvant Chemotherapy for Stage III Colon Cancer in England.
Boyle JM; Kuryba A; Cowling TE; Aggarwal A; Hill J; van der Meulen J; Walker K; Braun MS
Clinical oncology
One-year mortality of colorectal cancer patients: development and validation of a prediction model using linked national electronic data
Cowling T; Bellot A; Boyle J; Walker K; Kuryba A; Galbraith S; Aggarwal A; Braun M; Sharples L; Van Der Meulen J
British Journal of Cancer
A novel approach selected small sets of diagnosis codes with high prediction performance in large healthcare datasets.
Cowling TE; Cromwell DA; Sharples LD; van der Meulen J
Journal of Clinical Epidemiology
Patient-Reported Functional Outcomes After Hypofractionated or Conventionally Fractionated Radiation for Prostate Cancer: A National Cohort Study in England.
Nossiter J; Sujenthiran A; Cowling TE; Parry MG; Charman SC; Cathcart P; Clarke NW; Payne H; van der Meulen J; Aggarwal A
Journal of clinical oncology
Assessing the Time-Dependent Impact of Performance Status on Outcomes After Liver Transplantation.
Wallace D; Cowling T; McPhail MJ; Brown SE; Aluvihare V; Suddle A; Auzinger G; Heneghan MA; Rowe IA; Walker K
Hepatology (Baltimore, Md)
Physical multimorbidity, health service use, and catastrophic health expenditure by socioeconomic groups in China: an analysis of population-based panel data.
Zhao Y; Atun R; Oldenburg B; McPake B; Tang S; Mercer SW; Cowling TE; Sum G; Qin VM; Lee JT
The Lancet Global Health
See more Publications