Evidence shows that the key to preventing psychosis is to identify young people who might be at risk as early as possible. Clinicians can then offer preventative care which can help reduce symptoms and slow down or stop the onset of psychosis. However, currently only 5-12% of psychosis cases are detected by the NHS before the first episode. As many of these young people are already in contact with mental health services this is a missed opportunity for offering preventative therapies.

What have we done to address this?

Applied informatics researchers at ARC South London, led by Paolo Fusar-Poli, professor of preventive psychiatry at King’s College London, have developed a transdiagnostic risk calculator that helps non-specialist clinicians to detect people at high-risk of psychosis, so that they can be referred earlier to specialist mental health services for further assessment and support.

The calculator is integrated with the Clinical Records Interactive Search (CRIS) system at NIHR Maudsley Biomedical Research Centre, which provides secure access to anonymised information from electronic clinical records at South London and Maudsley (SLaM) NHS Foundation Trust.

The calculator uses natural language processing to analyse free text in electronic clinical records, and can spot mentions of psychosis symptoms and other risk factors such as substance use in clinical notes, letters and routinely collected data. By linking the calculator to other software, it can automatically alert clinicians if a patient is identified as being at high-risk of psychosis.

The tool is one of the first to have been tested for real-world use in mental healthcare. It works well with clinical records from SLaM and from other mental health centres in the UK and US.

What is the aim of this research?

The aim of this research is to evaluate the feasibility and acceptability of the calculator's use in clinical routine rather than its effectiveness.

How will the research be carried out?

The research will be carried out in two phases:

  • In the first phase the team will seek to integrate the risk calculator into the electronic health record system and develop filters for practical or ethical barriers.
  • In the second phase, once the tool is integrated and ready, the research team will begin recruiting eligible individuals for screening by the calculator from boroughs covered by SLaM.

Patients recruited from Lambeth and Southwark will be in the active group receiving screening from the tool, and patients from Croydon and Lewisham will be in the control group, receiving care as usual. This will enable the researchers to compare outcomes in the two groups.

Eligible individuals must be aged 14 years or over, accessing a SLaM service and have received a first diagnosis of a non-organic, non-psychotic mental disorder, with existing contact details.

The researchers will use surveys and focus groups to assess both service user and clinician perspectives on the tool. The main outcomes from the research will be an assessment of the acceptability and feasibility of the calculator. The researchers will also assess any changes in referrals to specialist clinics, incidence of psychotic disorders in detected individuals and economic feasibility.

What are the potential benefits?

The potential benefits of this research are being able to identify people at risk of psychosis more accurately and earlier, leading to more people receiving evidence-based preventative support. Researchers and clinicians will also gain access to new knowledge, skills and data, potentially leading to larger-scale research supporting precision psychiatry and similar mental health interventions.

This project has been adopted by ARC South London and is expected to finish in December 2024.