Trial results for a study investigating an Artificial Intelligence-based clinical training tool, CBTpro, for conditions including Bipolar Disorder, were posted on ClinicalTrials.gov on 2026-02-27, with 601 participants enrolled.
Background
Psychosis, a symptom present in conditions like schizophrenia, schizoaffective disorder, bipolar disorder, and depressive disorder, significantly impacts quality of life. Cognitive Behavioral Therapy for Psychosis (CBTp) is an evidence-based intervention, but its widespread availability and consistent quality can be limited. The development of scalable and effective training tools for clinicians is crucial to address this gap, ensuring more patients have access to high-quality CBTp. This study aimed to develop and evaluate such a tool, CBTpro, designed to support skills training for clinicians using simulated patients and spoken language technologies.
Trial design
This completed study, designated as Phase NA, enrolled 601 participants across various conditions, including Psychosis, Schizophrenia, Schizo Affective Disorder, Bipolar Disorder, and Depressive Disorder. The primary objective was to develop and evaluate CBTpro, an AI-based clinical training tool, to support high-quality skills training in CBT for psychosis (CBTp). The trial aimed to assess how this training tool could scale and sustain high-quality CBTp in routine care settings. Patient outcomes were measured in groups where clinicians either utilized the CBTpro training tool or received Treatment as Usual (TAU).
Key results
The trial reported several key measurements using the Revised Green Paranoid Thoughts Scale (R-GPTS) Mean Score, which assesses paranoid thoughts. The results are presented for patients whose clinicians were in either the CBTpro training group or the Treatment as Usual (TAU) group:
- For the CBTpro group, R-GPTS mean scores were observed as:
- 12.3 (Standard Deviation 8.8)
- 11.4 (Standard Deviation 8.1)
- 11 (Standard Deviation 8.3)
- 14.1 (Standard Deviation 12.6)
- 10.7 (Standard Deviation 10.7)
- 11.4 (Standard Deviation 11.3)
- For the Treatment as Usual (TAU) group, R-GPTS mean scores were observed as:
- 12.1 (Standard Deviation 8.7)
- 12.1 (Standard Deviation 8.9)
- 11.1 (Standard Deviation 8.4)
- 12.7 (Standard Deviation 11.1)
- 13.3 (Standard Deviation 12.3)
- 11.9 (Standard Deviation 12.2)
What this means
The posted results provide specific mean scores and standard deviations on the Revised Green Paranoid Thoughts Scale for patients whose clinicians received either CBTpro training or Treatment as Usual. While the data offers insights into the range of outcomes for paranoid thoughts across these groups, no comparative statistical analyses or p-values were provided. These measurements contribute to the evaluation of CBTpro as a training tool for clinicians, potentially informing future strategies for scaling high-quality CBTp in clinical practice for conditions such as Bipolar Disorder and other psychoses.
Source
The information regarding these trial results was obtained from ClinicalTrials.gov, a public database of clinical studies. The results for the study NCT05127837, titled "CBTpro: Scaling up CBT for Psychosis Using Simulated Patients and Spoken Language Technologies", were posted on 2026-02-27 on clinicaltrials.gov.