The mission of the Health Data Science group is to produce clinically actionable insights from observational health data by enabling data-driven healthcare. Improved interoperability of data is a necessary pre-requisite for this mission.

Read More


We perform methodological research in clinical characterisation, population-level effect estimation, and prediction modelling. We develop open-source analytical tools that can be applied on the OMOP Common Data Model.

Read More


We believe more education for young health data scientists, medical students, and healthcare professional, is needed to train them in the opportunities and limitations of big data in healthcare.

Read More

Latest News

Thursday, September 09, 2021

We are showcasing our work next week at the virtual OHDSI symposium, we are well-represented at plenary sessions and over 15 lightning talks and posters. This includes work on trends in patient-level prediction, data quality, frequent pattern mining, treatment effect heterogeneity, attention-based neural networks, explainable models, and several new-open source tools.

For an overview of our work:

posters symposium2021

 For the full program and registration: click here.

Monday, June 14, 2021

Important work, with contributions from our group, characterising the background incidence rates of adverse events of special interest (AESIs) in eight countries was published today in BMJ.

This large, international study found large variations across age groups, countries, and males/females in the observed rates of AESIs associated with covid-19 vaccines. This shows the need for stratification or standardisation before using background rates for covid-19 vaccin safety surveillance and suggests caution when interpreting the differences between observed and expected rates.