Funded PhD: Predicting frailty using integrated health and social data to support preventive, community-based care

• Principal investigator(s) Prof. Hamde Nazar, Prof. Noura Al-Moubayed
• Research theme Population Health and Health Services

This project explores how we can better support people to live healthier, more independent lives for longer by predicting frailty earlier using large, real-world datasets. Frailty often develops gradually in the community and is a major driver of hospital admission, loss of independence, and reduced quality of life. However, it is frequently recognised too late, when opportunities for prevention are limited.

The 'hospital at home' model delivers hospital-level care in a patient’s home, helping people avoid unnecessary admissions while receiving appropriate monitoring and treatment in a familiar environment. Predicting who is at risk of deterioration or increasing frailty is key to making this approach safe, effective, and scalable.

Building on advances in machine learning, early warning scores, and the use of clinical data, this project extends the focus beyond hospital settings into the community. It will explore how combining multiple sources of data –rather than relying on a single score or dataset – can provide a richer, more accurate picture of an individual’s health trajectory.

Key objectives include:

• Understanding frailty in a community and home-based care context: The project will examine how frailty develops over time and how it impacts care decisions, particularly in models like hospital at home.
• Integrating diverse data sources: In addition to traditional healthcare data (e.g. vital signs, diagnoses, hospital records), the project will explore the value of wider datasets such as social care information, activity levels, behavioural patterns, and potentially environmental or lifestyle factors.
• Developing personalised prediction approaches: By analysing these combined datasets, the project will investigate how risk prediction can be tailored to individuals, supporting more targeted interventions rather than a one-size-fits-all model.
• Supporting preventive and community-based care: A key focus will be on how predictions could be used in practice – to trigger early support, guide care planning, or enable safe delivery of care at home.

What makes this project particularly exciting for a young researcher is its strong connection to real-world impact and future healthcare models. It brings together data science, clinical insight, and community care, highlighting how innovative use of data can shift healthcare from reactive treatment to proactive support.

The project also encourages creative and independent thinking. Working with complex and varied datasets presents both challenges and opportunities, allowing the researcher to explore new ideas and approaches. Importantly, it emphasises that better care is not just about better technology, but about understanding people in the context of their daily lives.

Ultimately, this project aims to demonstrate how combining healthcare, social and behavioural data can enable earlier, more personalised interventions. By supporting models like hospital at home, it contributes to a vision of healthcare that helps people remain well, independent, and supported in their own communities for as long as possible.

Tenure: Three years

Start date: October 2026

Specification

Minimum requirements

• An undergraduate degree (first class or upper second class honours, or equivalent) in a relevant area such as: computer science, mathematics, biomedical engineering, physics, health informatics, etc.
• A master’s degree in a related field is desirable, but not essential.
• Strong programming skills in Python.
• Familiarity with machine learning and AI techniques, including deep learning, statistical modelling, or generative AI.
• Experience with data analysis and handling large or complex datasets.
• Interest in healthcare applications of AI, such as digital health, clinical decision support, electronic health records, predictive analytics, etc.
• Ability to work independently and collaboratively within a multidisciplinary research team.
• Good written and verbal communication skills in English.

Desirable candidate specifications

• Prior research experience demonstrated through a dissertation, publications, internships, or research projects.
• Knowledge of responsible AI, explainability, fairness, privacy or healthcare regulations.
• Experience working with healthcare or biomedical datasets.

Application process

Please apply for the research project through the link below.

Applications must include:

• A completed application form
• English language requirements