Many people now live longer but live these extra years with two or more long-term conditions at the same time (“multimorbidity”). Two people of the same chronological age can therefore have very different levels of health. The aim of this project is to use UK Biobank data and artificial intelligence to develop a “biological age” that reflects the combined impact of multiple conditions on different organ systems, and to test whether this measure helps us better understand and predict outcomes in people with multimorbidity. We will bring together each participant’s existing health information in UK Biobank (for example, hospital diagnoses, procedures and key measurements) and convert it into a short, anonymised health summary in plain language. A large language model (a type of AI that works with text) will read this summary and estimate the biological age of major organ systems, as well as simple scores for external influences such as lifestyle and mental health. We will then examine whether participants with “older” biological ages, especially those with multimorbidity (two or more long-term conditions), have higher risks of death or of developing new serious diseases than people of the same chronological age but a “younger” biological profile. In subsets of participants with blood protein and genetic data, we will also explore which proteins and biological pathways are linked to an older biological age in the context of multimorbidity, to highlight possible mechanisms and future targets for prevention or treatment.
