Keywords: Psychotic Disorders, Clinical Decision Support Systems, Electronic Health Records

Background:

Psychotic disorders are a major public health concern due to early onset, delayed diagnosis, and significant functional impairment. Although the Clinical High-Risk for Psychosis (CHR-P) framework has advanced early detection, its implementation in primary care remains limited. P-Risk, a digital tool developed in the UK using electronic health records (EHR), offers a scalable method to support general practitioners (GPs) in identifying individuals at risk of developing psychosis.

Research questions:

Can the P-Risk algorithm be externally validated and adapted for use in Catalan primary care to predict the onset of psychosis using routine EHR data?

Method:

A retrospective cohort study will be conducted using the SIDIAP database, covering over 5.8 million individual EHR in Catalonia. The cohort includes patients aged 17 and older with consultations or prescriptions related to non-psychotic mental health conditions between 2005 and 2025. Psychosis onset will be identified through diagnostic codes or antipsychotic prescriptions over a six-year follow-up. The P-Risk model’s performance will be assessed using discrimination and calibration metrics (e.g., Harrell’s C-index, sensitivity, specificity).

Results:

As this is a study protocol, results remain pending.

Conclusions:

If validated, P-Risk could enhance early psychosis detection in primary care, enabling timely interventions and reducing the duration of untreated psychosis. Integration into GP workflows may support clinical decision-making and streamline referrals to mental health services.

Points for discussion:

This project exemplifies the application of digital innovation in general practice, supporting precision psychiatry through scalable, EHR-based tools.

It reinforces the role of family medicine in early detection of severe mental illness and highlights the value of cross-country collaboration in mental health research.