Local adaptation and validation of a transdiagnostic risk calculator for first episode psychosis using mental health patient records

---

This study evaluated whether a transdiagnostic risk calculator for psychosis could be successfully adapted for use in a new healthcare region in Southeast England.

Using electronic health records and advanced natural language processing (NLP), researchers analysed data from more than 63,000 patients to assess how accurately the model could identify individuals at increased risk of developing psychosis. The findings showed that the tool maintained good predictive performance in a different clinical setting, supporting its potential for wider implementation across healthcare systems.

The study highlights the importance of validating digital tools in diverse populations and demonstrates how artificial intelligence and electronic health records can contribute to earlier detection and more personalised mental healthcare.

---