Sign in Subscribe
Youth

Bipolar Disorders Research Highlights Potential of Speech Analysis for Early Identification

Strategic Insights

— 2 min read

Researchers from the Douglas Mental Health University Institute have made a significant contribution to the understanding of bipolar disorders by systematically reviewing longitudinal studies assessing speech/language markers as predictors of major mental disorder onset or symptom progression in youth. The study, which was funded by the Canadian Institutes of Health Research, Canada First Research Excellence Fund, and Fonds de Recherche du Quebec - Sante, among others, aimed to comprehensively review the available research on this topic.

Key Takeaways:

  • The study found that existing studies highlight the potential of applying Natural Language Processing methods to speech samples from both youth and parents for early identification of bipolar disorders and other mental health conditions.
  • A total of 11 studies met the inclusion criteria, covering major depressive disorder, psychosis, and attention deficit hyperactivity disorder (ADHD).
  • The studies used both manual and computational speech analyses, with speech samples from parents and youth, and identified predictive speech/language markers such as parental expressed emotion, formal thought disorder, and acoustic/linguistic features.
  • The study quality was moderate to good, with a mean score of 5.45/8 on the Newcastle-Ottawa Scale.
  • Externally validated longitudinal studies on the predictive value of speech/language markers of youth-onset mental disorders are scarce, restricted to a few target disorders, and do not allow for variations due to the developmental stage of the samples.
  • The study suggests that speech analysis has the potential to be a useful tool for early identification and intervention in bipolar disorders and other mental health conditions.

Statistics:

  • 2260 articles were screened for inclusion in the study.
  • 11 studies met the inclusion criteria, covering major depressive disorder (n = 3), psychosis (n = 5), and ADHD (n = 3).
  • The studies included speech samples from a total of 1230 participants (parents and youth).
  • Speech analysis methods used included manual and computational analysis of acoustic and linguistic features, as well as parental expressed emotion and formal thought disorder.
  • The mean study quality score was 5.45/8 on the Newcastle-Ottawa Scale.

Sources:

  • NewsRx. Studies from Douglas Mental Health University Institute Yield New Information about Bipolar Disorders (Speech and Language Markers as Longitudinal Predictors of Youth Mental Health: A Systematic Review). Mental Health Weekly Digest. October 20, 2025; p 1078.
  • Early Intervention in Psychiatry. Speech and Language Markers as Longitudinal Predictors of Youth Mental Health: A Systematic Review. October 2025;19(10).
  • Canadian Institutes of Health Research. (Source of funding for the research).
  • Canada First Research Excellence Fund. (Source of funding for the research).
  • Fonds de Recherche du Quebec - Sante. (Source of funding for the research).
  • Douglas Mental Health University Institute. (Research institution).
  • Early Intervention in Psychiatry. (Journal where the study was published).

Read more