Medical-AI trial: using AI to gauge psychosis risk early and act ahead — staged diagnosis and risk-adapted treatment (pronia.ai) (NCT05813080)
A multicenter randomized controlled trial in people at increased clinical risk for psychosis, comparing an arm with AI-staged early diagnostics and risk-adapted treatment (RAB) against treatment-as-usual (TAU). The primary outcome uses the Structured Interview for Psychosis-Risk Syndromes (SIPS). 260 patients; completed.
Trial overview (primary data)
- StatusCompleted
- ConditionsPsychotic Disorders, Prevention
- InterventionsDEVICE: "pronia.ai" medical device for high risk psychosis prognosis, OTHER: Treatment-as-usual (TAU)
- SponsorHeinrich-Heine University, Duesseldorf
- Target enrollment260 participants
- Period2023-05-01 〜 2025-12-16
Key points
- A multicenter RCT comparing AI-staged early diagnostics plus risk-adapted treatment (RAB) against treatment-as-usual (TAU) in people at increased risk for psychosis.
- The primary outcome uses the Structured Interview for Psychosis-Risk Syndromes (SIPS). 260 patients; completed.
- Psychosis has a prodromal phase before onset; identifying high-risk people may allow acting early.
- Early intervention means acting ahead in the prodromal phase rather than after onset; AI grades risk from multifaceted information in stages.
- Risk prediction is uncertain and needs care about labeling and overtreatment; an effect-evaluation stage, not an establishment of effectiveness.
Psychosis, as in schizophrenia, can be preceded before clear onset by warning signs (a prodromal phase) such as anxiety and subtle cognitive or perceptual changes. If people at high risk can be identified at this stage and acted on early, onset might be prevented or softened. But telling who is truly high-risk is hard, which has raised the question of whether AI can help.
This study tests that early identification and pre-emptive treatment, as a multicenter randomized controlled trial (RCT). Per the registry summary, in patients at increased clinical risk for psychosis, it compares an interventional arm of AI-staged early diagnostics and risk-adapted treatment (RAB) against a control arm of treatment-as-usual (TAU).
The primary outcome uses the Structured Interview for Psychosis-Risk Syndromes (SIPS). The size is 260, at the completed stage.
In psychiatry there is strong interest in early intervention, acting ahead by gauging risk in the prodromal phase rather than responding after onset. If AI can grade risk from multifaceted information in stages, it could direct limited resources to those who truly need support and lead to adaptive treatment that is neither excessive nor insufficient.
That said, predicting psychosis risk carries substantial uncertainty, and concerns about labeling and overtreatment require care. This study evaluates the effect of AI-staged diagnosis and risk-adapted treatment via SIPS and the like; it does not establish their effectiveness.
Why it matters
Psychiatry has strong interest in early intervention, acting ahead in the prodromal phase rather than after onset. If AI can grade risk in stages from multifaceted information, it could direct limited resources to those who truly need support and enable adaptive treatment, though prediction uncertainty and labeling concerns need care (this study is an effect-evaluation stage, not an establishment of effectiveness).
FAQ
What is the prodromal phase of psychosis?
Does the AI predict psychosis?
Sources (primary)
Source: ClinicalTrials.gov (U.S. NIH/NLM, public domain). This site does not provide medical advice. Verify the latest and exact details with the official source. This site is not endorsed or certified by the NIH/NLM.
- ClinicalTrials.gov (study record, original)
- NCT ID: NCT05813080