©
Department of Psychiatry, University of Oxford
It is the first time ever a mental health clinical prediction tool has been demonstrated as effective.
The PETRUSHKA tool, developed by the University of Oxford and funded by the National Institute for Health and Care Research (NIHR), supports clinicians and patients to choose the antidepressant that is better tailored to each individual. It uses AI to combine clinical and demographic information with patient preferences, particularly around side effects, to support treatment personalisation and shared decision-making in routine care.
The tool was tested in a large international randomised clinical trial across Brazil, Canada and the UK involving more than 500 adults with major depressive disorder (MDD) and launched in 2024. The findings, published in the Journal of American Medical Association (JAMA), show that people whose antidepressant was selected using PETRUSHKA were significantly more likely to continue their treatment and experienced better mental health outcomes compared with usual care up to six months.
Choosing the right antidepressant remains a major challenge in everyday practice, with many people stopping treatment early because of side effects or a lack of perceived benefit. PETRUSHKA was designed to address this by using evidence from clinical trials and real‑world data alongside what matters most to patients, helping to identify the most suitable options from the outset.
In the trial, participants using PETRUSHKA were around 40 per cent less likely to discontinue their antidepressant within the first eight weeks of treatment. Fewer people stopped treatment because of adverse effects, and by 24 weeks those in the PETRUSHKA group also reported greater improvements in depressive and anxiety symptoms.
Read the full story on the Department of Psychiatry website.