Theme: Systems Diagnostics

Project Description

Young people with psychosis develop obesity, diabetes and cardiovascular disease at strikingly high rates, contributing substantially to a 15-year average reduction in life expectancy. Prevention can make a difference if we understand who is most likely to benefit, from which intervention, and how to make those conversations meaningful.

PsyMetRiC (https://psymetric.app/) is the first prediction tool in psychiatry available for routine clinical use, as a regulated software medical device. PsyMetRiC predicts risk of stakeholder-preferred short, medium and longer-term cardiometabolic outcomes in 16–35-year-olds with psychosis, with its web-based implementation co-designed by lived experience contributors.

This PhD tackles the next question: can we move from personalised prediction of risk (“what is likely to happen?”) to personalised prediction of treatment benefit (“how might my outcome differ if I receive this intervention rather than not?”)

The Fellow will lead an interdisciplinary programme spanning causal inference, real-world data, clinical trials and digital health. First, they will develop individualised treatment-effect predictions using large-scale, longitudinal, anonymised electronic health records. They will learn to address the problems that make this work exciting: confounding, treatment selection, missing data, uncertainty and whether a model works fairly across populations.

The Fellow will then be supported to test these predictions prospectively in a pragmatic pilot randomised study with young people experiencing psychosis, building the foundations for a definitive future validation trial. Finally, they will work with patients and clinicians, using co-design, think-aloud methods and usability testing to develop clear, honest ways to communicate benefit, harm and uncertainty in the PsyMetRiC web application. Lived experience engagement will be embedded through the existing PsyMetRiC Lived Experience Advisory Panel.

This is an exciting opportunity to help shape precision prevention in early psychosis, from rigorous methods to a tool that could support genuinely person-centred clinical decisions.