(EPSRC DTP) Developing Methods for Handling Missing and Irregularly Observed Data in Dynamic Clinical Prediction Models

Routine healthcare data are especially valuable because they include information collected about people over time. This means prediction models can be updated as new information becomes available, giving a more current estimate of a person’s risk. For example, a model might update someone’s estimated risk of cardiovascular disease as new blood pressure readings, blood test results or medication records are added to their health record. This is known as dynamic prediction modelling, and there are a range of methods available to develop such dynamic prediction models.

However, a major challenge is that electronic health record data are often incomplete. Some measurements may be missing, and others may be recorded at different times for different people, depending on when and why they use healthcare services. Researchers have studied how to handle missing data in non-dynamic prediction models, but much less is known about how to handle data that are both incomplete and recorded irregularly in dynamic prediction models.

This PhD will develop and evaluate new statistical methods for dealing with this problem. The project will examine how missing and irregularly recorded repeated measurements affect the accuracy of predicted risks for individual patients. It will compare different approaches for handling missing information and assess the compatibility of different approaches across the development, validation and implementation of a dynamic prediction model.

The project will mainly involve methodological development, supported by computer simulation studies and analysis of real-world electronic health record data. Clinical examples, such as cardiovascular disease prediction using primary care data, will be used to motivate and test the methods. The project will also produce open-source software so that other researchers can apply and build on the methods developed.

The project would suit a student with a strong background in mathematics, statistics, biostatistics, data science or a related quantitative subject, and an interest in applying these skills to healthcare. The student will receive training in clinical prediction modelling, statistical methods development, simulation studies and the analysis of electronic health record data.

Entry Requirements

Applicants are expected to hold (or about to obtain) a minimum upper second-class undergraduate honours degree (or equivalent) in (bio-)statistics, mathematics, data science or a relevant subject area. Research experience in prediction modelling, and its applications to healthcare data is desirable.

International applicants (including EU nationals) should ensure that they meet the academic eligibility criteria for the programme. Applicants who are offered a place will also need to meet the University's English language requirements before commencing their studies. Further information can be found on the University's English Language Requirements page.

Application Guidance

Candidates must contact the primary supervisor before applying to discuss their interest in the project and assess their suitability.

Apply directly via this link: https://tinyurl.com/mt2xfsvp or on the online application portal, select EPSRC DTP PhD as the programme of study.

You may apply for up to two projects within this scheme. If you wish to apply for two projects, you must submit a separate online application for each project, listing the relevant project title and main supervisor in each application.

Please ensure that your application includes all required supporting documents:

• Curriculum Vitae (CV)
• Supporting Statement
• Academic Certificates and Transcripts

Incomplete or late applications will not be considered. Further details are available on our website: EPSRC Doctoral Training Partnership | Biology, Medicine and Health | The University of Manchester

Equality, diversity and inclusion are central to the University’s activities. The full statement can be found here: https://www.bmh.manchester.ac.uk/study/research/getting-

Funding Notes

The EPSRC studentships provide full funding for tuition fees and a stipend at the Research Councils UK (UKRI) rate for 3.5 years starting in September 2027. Please note, candidates will need to cover the cost of relocation and any associated costs (e.g. visa, health surcharges).