Causal inference from observational data: trials, real world evidence and genomics to triangulate the evidence

PROJECT OUTLINE:

This project will involve the use of multiple methods and data assets to study the risk-benefit of medicines and/or medical devices.

Specifically, we will use observational routinely collected health data (also known as real world data), as well as genetic data and biobanks.

As part of the former, we will use target trial emulation methods [1], and related methods for the study of causal effects of medicines and/or devices. We will test different methods to minimise confounding, including propensity scores, inverse probability weighting, clone-weighting, and the G-formula. We will then identify the best performing methods based on metrics of observed and unobserved confounding, and using previously completed randomised controlled trials as a benchmark, in line with previous research [2].

As part of the latter, we will use genetic data linked to longitudinal cohort and electronic health records to triangulate the evidence generated by the studies conducted using previous randomised controlled trials and/or target trial emulation methods. As part of this, you will learn and apply pharmacogenomic methods including Mendelian randomisation [3].

Finally, you will apply the best performing methods and study designs to evaluate the risks and benefits of novel medicines, vaccines, or other medical innovations, potentially including surgery and/or implantable devices, and to advance the field of personsalised medicine.

RELATED PUBLICATIONS:

1: Wang SV, et al. Emulation of Randomized Clinical Trials With Nonrandomized Database Analyses: Results of 32 Clinical Trials. JAMA. 2023 Apr 25;329(16):1376-1385. doi:10.1001/jama.2023.4221

2: Hernán MA, Wang W, Leaf DE. Target Trial Emulation: A Framework for Causal Inference From Observational Data. JAMA. 2022 Dec 27;328(24):2446-2447. doi:10.1001/jama.2022.21383

3: Burgess S, et al. Guidelines for performing Mendelian randomization investigations: update for summer 2023. Wellcome Open Res. 2023 Aug 4;4:186. doi:10.12688/wellcomeopenres.15555.3

KEYWORDS:  

Causal inference; Health Data Sciences; Target Trial Emulation; Pharmacoepidemiology

RESEARCH GROUP AND SUPERVISION TEAM:

The Health Data Sciences (HDS) section of the Botnar Research centre includes a total of 35+ colleagues with a united vision to improve human health through the use of large data. We do this by studying and applying the best methods to the  most reliable data assets available nationally and internationally. We collaborate with many research groups globally, including the Observational Health Data Sciences and Informatics (OHDSI) community, Health Data Research UK (HDRUK), and the International Society of Pharmacoepidemiology (ISPE).

The HDS section has supervised many DPhil students to completion, and is currently hosting a total of 8 DPhil students.

Your main supervisor will be Prof Daniel Prieto-Alhambra. Dani is an academic clinical scientist with a long track record and experience in the field of HDS and pharmaco-epidemiology. He has published more than 380 indexed manuscripts, and supervised and mentored many students and junior staff. He is the overall lead for the Health Data Sciences section.

Your co-supervisors will include Dr Marti Catala-Sabate, Dr Danielle Newby, and Dr Junqing Xie.

Dr Marti Catala-Sabate is a senior researcher in health data sciences. He has extensive experience in the modelling of observational data, and in the programming of code and software for the analysis of real world data. He has expertise in the use of propensity scores, inverse probability of treatment weights, and other methods for causal inference.

Dr Danielle Newby is a senior postdoctoral researcher specializing in real-world evidence, with a diverse background in epidemiology, pharmacology, and machine learning. Her expertise includes triangulating real-world data from various sources, including genetic datasets from across the UK and Europe. Her expertise will equip the student with the essential skills and training required for the project.

Dr Junqing (Frank) Xie is a postdoctoral researcher in pharmacoepidemiology and pharmacogenetics, with a track record of impactful scientific publications. He has years of experience in analyzing global real-world data and is currently leading efforts to bridge large-scale phenotypes with genotypes in various Biobanks (UK Biobank, Our Future Health, All of Us, etc.) for innovative studies and evidence triangulation.

TRAINING:

The Health Data Sciences section is part of the Botnar Research Centre, University of Oxford. Training will be provided in health data techniques including epidemiology, biostatistics, common data models, causal inference, and real world evidence methods and multi-sourses data.

A core curriculum of lectures will be taken in the first term to provide a solid foundation in a broad range of subjects.  Students will also be required to attend regular seminars within the Department and those relevant in the wider University.

Students will be expected to present data regularly in Departmental seminars, the Health Data Sciences fortnightly lab meeting, and to attend external conferences to present their research globally, with limited financial support from the Department.

Students will also have the opportunity to work closely with our multiple collaborators nationally and internationally, including academic centres of excellence (Columbia University, Universitat Autonoma de Barcelona, Erasmus Medical Centre, among others), regulators (UK MHRA, European Medicines Agency), and industry.

Students will have access to various courses run by the Medical Sciences Division Skills Training Team and other Departments. All students are required to attend a 2-day Statistical and Experimental Design course at NDORMS and run by the IT department (information will be provided once accepted to the programme). We will offer our students attendance to our world famous Summer School in Real World Evidence using the OMOP Common Data Model. 

HOW TO APPLY:

Please contact the relevant supervisor(s), to register your interest in the project, and the departmental Education Team (graduate.studies@ndorms.ox.ac.uk), who will be able to advise you of the essential requirements for the programme and provide further information on how to make an official application.

Interested applicants should have, or expect to obtain, a first or upper second-class BSc degree or equivalent in a relevant subject and will also need to provide evidence of English language competence (where applicable). The application guide and form is found online and the DPhil will commence in October 2025.

Applications for this project should be made to the following programme, using the specified course code:

D.Phil in Clinical Epidemiology and Medical Statistics (course code: RD_NNRA1)

For further information, please visit http://www.ox.ac.uk/admissions/graduate/applying-to-oxford.

The Botnar Institute is a proud supporter of the Academic Futures scholarship programme, designed to address under-representation and help improve equality, diversity and inclusion in our graduate student body. The Botnar and the wider University rely on bringing the very best minds from across the world together, whatever their race, gender, religion or background to create new ideas, insights and innovations to change the world for the better. Up to 50 full awards are available across the three programme streams, and you can find further information on each stream on their individual tabs (Academic futures | Graduate access | University of Oxford).