Multinational Study Using Smartphone Speech Analysis to Predict Psychosis Relapse Risk Observational Research in Adults Aged 18 to 65 Speaking Six Languages

What this Trial Is About

Researchers are investigating the use of speech and self-report data to estimate relapse risk in people with psychotic disorders. This observational, multinational study involves individuals at risk of psychotic relapse and healthy controls, aiming to assess the feasibility of collecting speech data in six different languages over a 12-month period. The study explores new speech markers for predicting relapse while minimizing burden on participants and supports future research into clinical decision support systems. Participants will complete weekly speech assessments through a smartphone app, which collects speech and self-report data to be securely transferred for AI-based analysis. Two groups are involved individuals at risk of psychotic relapse and healthy controls. The study includes three in-person visits at baseline, six months, and twelve months to evaluate system usability and functional outcomes. Independent clinical evaluations ensure data quality and validate AI-generated risk scores, which are not shared with treating clinicians. During the study, participants will provide weekly speech recordings and self-reports, with quality control performed by designated reviewers checking audio and transcription accuracy. Researchers measure user adherence, transcription quality, usability of recordings, and the performance of the relapse prediction system. Follow-up assessments include clinical evaluations, functional and quality-of-life measures, and exploration of novel speech markers. The total participation lasts one year, with safety and usability monitored throughout.

Speech-based Assessment of Relapse Risk in People With Psychosis