A new clinic-ready web-based risk prediction tool called PsyMetRiC is now available to forecast the risk of young people with psychosis developing cardiometabolic disorders such as obesity, metabolic syndrome and diabetes.
The algorithms behind PsyMetRiC have been specifically tailored for young people with psychosis, and predict clinically significant weight gain within one year, metabolic syndrome within six years and type 2 diabetes within 10 years. They were developed and tested using routine anonymised health data from over 25,000 young people with psychosis from the UK, who were followed up for over 20 years. The science underlying this work is published today in The Lancet Psychiatry .
On average, people with severe mental illness die 15 years younger than the rest of the population, mostly explained by a heightened risk of developing these highly preventable physical health conditions. While prediction tools have been used for many years by GPs to help them decide who may need statins to lower risk of heart attacks or strokes in the future, these tools weren’t designed for younger populations, and don’t work accurately in people with psychosis.
The net result is that the very people who could benefit most from early interventions are not getting them.
Dr Benjamin Perry, Associate Clinical Professor of Psychiatry from the University of Birmingham, who led the development of PsyMetRiC, said: “Psychotic disorders are typically diagnosed when people are in their late teens and early twenties, and the impact on lifestyles can be profound. People with psychosis may be less able to eat healthy food, or exercise, and may be more likely to smoke. Additionally, antipsychotic medicines could have side effects making patients feel hungrier, or more sedate, contributing to weight gain. People with psychosis may also experience healthcare and other inequalities, preventing them from receiving the same standard of physical healthcare as the rest of the population.”
“Consequently, factors that predispose people to obesity, diabetes and heart disease, are detectable from the onset of psychosis – many years earlier than would be typical for the rest of the population.”
Although it may be possible to reverse the early signs of physical health conditions with lifestyle changes and medication, young people with psychosis often fall under the detection radar.
PsyMetRiC is intended to change this, by prompting meaningful discussions between patients and health professionals, encouraging shared decision-making about preventative measures, interventions, and behavioural change. It is suitable for use in both primary and secondary care.
Predictive power and fairness
The web application provides a read-out showing the risk of metabolic syndrome within six years, and type 2 diabetes within ten years of onset of psychosis.
The PsyMetRiC web application is based on a refined and updated version of a prediction model that was first developed and validated in the UK in 2021 . It has so far been tested in Spain, Switzerland, Finland, Netherlands, Canada, Hong Kong, and Australia, and the researchers have recently received funding to test in the USA.
Known as PsyMetRiC1, its ability to discern the up-to-six-year risk of people with psychotic spectrum disorders developing metabolic syndrome – a cluster of changes including high blood pressure, weight gain, high cholesterol and high blood sugars - has been confirmed in academic studies in many different countries.
Now, through access to much larger UK-based population-level datasets, the research team were able to substantially refine PsyMetRiC, improving its predictive power, to create PsyMetRiC2.
The team prioritised feedback from stakeholders including clinicians, carers and young people with lived experience of psychosis. The McPin Foundation , in collaboration with The Centre for Mental Health and Equally Well, worked with those with lived experience to ensure the app would deliver outcome measures that are meaningful in clinical practice and desired by patients.
PsyMetRiC is designed to be as simple and easy to use in clinical practice as possible, and requires only simple, routinely-recorded information to make predictions.
In the validation study, used to explore the generalisability of the PsyMetRiC algorithms, the researchers took careful steps to test whether PsyMetRiC would work fairly for people from different, particularly under-served, backgrounds.
Dr Perry said: “We were inspired by the work of the STANDING Together collaboration, led by Birmingham researchers including Dr Joseph Alderman and Professor Alastair Denniston, who highlighted such an important issue – if inequalities in society get baked into health datasets, then tools developed using those datasets may inherit those biases, potentially making things worse”.
Meaningful in clinical practice
The PsyMetRiC app is designed specifically to inform rather than dictate clinical decisions, and the outcome measures selected for the tool predict the outcomes preferred by doctors and patients – an essential consideration where behaviour change may be more important than prescribing.
Dr Perry said: “The original PsyMetRiC study predicted metabolic syndrome. We received feedback that many clinicians aren’t aware of what that means, let alone patients. If PsyMetRiC is to change behaviour – both health professional and patient – then the outcomes need to be meaningful to both groups.”
Here the work with focus groups came into its own, shaping every aspect of the research and how information can be best communicated to patients.
Dr Perry said: “Input from the lived experience panel was critical to designing risk communication guides that are accessible and motivating for patients, showing how each risk factor contributes to an overall risk score, how this overall risk changes over time, and how interventions can reduce the overall risk.”
PsyMetRiC presents risk information in variety of graphical and numeric formats, including a representation of uncertainty, and an adaptation of the ‘heart age’ score approach that has shown promise in motivating changes such as reducing smoking, improving diet, or increasing physical activity.
All of this is explicitly focused on improving understanding, promoting deeper and more meaningful clinician-patient health-risk conversations. The website also houses a downloadable co-produced risk communication guide for health professionals.
Dr Perry said: “We are hoping to increase vigilance about the physical health risks for young people with psychosis, and expand the conversation between doctors and patients, so these risks are mitigated and premature mortality is reduced by earlier intervention and preventative measures.”
The PsyMetRiC web application is certified with the UK’s Medicines & Healthcare products Regulatory Agency (MHRA) as a Class 1 Medical Device. It is among the first prediction tools in psychiatry to achieve this status, meaning it is one of the first that can be used in clinical practice.
Users can sign up for a free account to use the app, after agreeing use terms, and confirming they are a health professional.
Like most software, it will receive iterative updates over time, with improvements and updates shaped by new research, stakeholder input, and the results of upcoming qualitative, health economic and impact evaluation studies. The researchers also have plans to extend to ex-UK settings in the future.
The Lancet Psychiatry
Data/statistical analysis
People
Cardiometabolic prediction models for young people with psychosis spectrum disorders in the UK (PsyMetRiC 2.0): a retrospective, multicohort clinical prediction model study
11-Mar-2026