MSc (Res) Opportunity – Understanding Multimorbidity Combinations and End‑of‑Life Trajectories Using Bayesian Outcome‑Based Clustering
Project Title:
Understanding Multimorbidity Combinations and End‑of‑Life Trajectories Using Bayesian Outcome‑Based Clustering
Supervisor(s):
Primary Supervisor: Dr Sarah Mills (University of St Andrews, School of Medicine)
Secondary Supervisors: Dr Michail Papathomas (University of St Andrews, School of Mathematics and Statistics) and Professor Colin McCowan (University of St Andrews, School of Medicine)
Deadline:
Tuesday 19 May 2026
Project Description:
The NHS is facing growing pressure due to population ageing, rising multimorbidity, and increasing demand for unscheduled care. People in their final year of life are among the most frequent users of Emergency Departments (ED), yet they are often not recognised as being near the end of life. This can lead to reactive, fragmented care that does not reflect their needs or preferences. Improving our ability to identify individuals approaching the end of life—particularly those with complex multimorbidity—could transform these encounters into opportunities for more compassionate, anticipatory support.
This Masters project offers the chance to contribute to an emerging and highly impactful area of health data science. The project focuses on understanding which combinations of long‑term conditions are most predictive of being in the last year of life and of high unscheduled care use. Rather than simply counting conditions, you will examine the multiplicative effects of multimorbidity—for example, how dementia + COPD differs from dementia + cancer or dementia + heart failure in shaping risk. This approach recognises that certain disease pairings interact in ways that amplify vulnerability.
A further component of the project explores whether protected characteristics—such as age, sex, ethnicity, or socioeconomic status—modify these relationships. This will help identify groups who may be under‑recognised or underserved within current care pathways, contributing to a more equitable understanding of end‑of‑life identification.
What the project involves:
You will conduct a retrospective cohort study using linked Emergency Department data from NHS Tayside and NHS Fife, accessed through the Health Informatics Centre (HIC) Trusted Research Environment. Your analysis will draw on:
- ED attendances in the final year of life
- Diagnostic histories and multimorbidity profiles
- Prescribing patterns and outpatient activity
- Demographic and protected characteristic variables
- Timing and sequence of condition onset
The project uses Bayesian statistics with outcome‑based clustering. This approach allows you to identify combinations of conditions that predict outcomes, rather than simply describing patterns.
Why this project is distinctive
- First application of Bayesian outcome‑based clustering to multimorbidity in an end‑of‑life context
- Focus on synergistic, multiplicative interactions between diseases
- Integration of protected characteristics to explore inequity
- Real‑world potential to inform more anticipatory and compassionate care pathways
This project is ideal for students interested in health data science, epidemiology, applied statistics, or improving care for people with complex health needs.
School of Medicine Research Division:
Population and Behavioural Science
Funding Details:
The home fee for this opportunity is funded. Please see the university website for fee information.
How to Apply:
If you are interested in applying for this opportunity, please submit your application via the University’s online portal.
Please make sure your application is complete by Tuesday 19 May 2026.
Eligibility Criteria:
- Applicants should normally hold, or expect to obtain, a 2:1 Honours degree (or equivalent) in a relevant subject.
- Part time study for this project would be considered.
Contact:
Enquiries about the application process can be directed to Sandra Fleming at [email protected].