Artificial intelligence as a predictive tool for mental health status: Insights from a systematic review and meta-analysis
Künstliche Intelligenz als Vorhersagetool für den psychischen Gesundheitszustand: Einblicke aus einer systematischen Übersicht und Metaanalyse
PLOS One · 26.09.2025 · 20 (9) · e0332207
DOI: 10.1371/journal.pone.0332207 ↗Auf einen Blick
Die Übersichtsarbeit untersuchte KI-gestützte Tools, vor allem Conversational Agents, bei psychischer Belastung. Sie senkten Angst und Depression in der Metaanalyse signifikant.
Fragestellung
Wie wirksam sind KI-gestützte Tools, besonders Conversational Agents, zur Linderung psychischer Belastung und zur Verbesserung von Mental-Health-Outcomes?
Methode
Systematische Übersicht und Metaanalyse; Suche in PubMed, Google Scholar, Elsevier und Scopus für Studien von Januar 2000 bis Juli 2024; Random-effects-Metaanalyse mit standardisierten Mittelwertsdifferenzen.
Ergebnisse
20 Studien in der qualitativen Synthese, 6 in der Metaanalyse. Angst: Cohen’s d = 0,62, p < 0,01; Depression: Cohen’s d = 0,74, p < 0,001. Multimodale CAs waren wirksamer als textbasierte Systeme.
Einschränkungen
Langzeitwirkung inkonsistent wegen unterschiedlicher Follow-up-Dauern und methodischer Heterogenität; einige Studien ohne längere Beobachtung oder mit nachlassender Wirkung.
Schlussfolgerung der Autor:innen
KI-basierte Conversational Agents, besonders in mobilen Plattformen und multimodalen Interfaces, bieten skalierbare und ansprechende Unterstützung für die psychische Gesundheit.
Kernaussagen
- 20 Studien, davon 6 in der Metaanalyse
- Angst und Depression nahmen unter KI-basierten CAs ab
- Langzeitwirkung und Methodik blieben uneinheitlich
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Systematische Übersicht in der Zeitschrift PLOS One, online erschienen am 26.09.2025. Verfasst von Arsalan Humayun und 7 weiteren Autor:innen. Beteiligte Einrichtung: Community Dentistry Department BADC, SMBBMU, Larkana, Pakistan. Hauptthemen laut MEDLINE: Artificial Intelligence, Mental Health. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This systematic review and meta-analysis evaluates the effectiveness of AI-driven tools, particularly conversational agents (CAs), in alleviating psychological distress and improving mental health outcomes. The focus is on their impact across diverse populations, including clinical, subclinical, and older adults. A comprehensive search was conducted in PubMed, Google Scholar, Elsevier, and Scopus using specific MeSH terms and keywords such as “Artificial Intelligence,” “Machine Learning,” “Natural Language Processing,” “Depression,” and “Anxiety.” The timeframe included studies published between January 2000 and July 2024. Inclusion criteria comprised peer-reviewed original research articles, cohort studies, and case reports focusing on AI tools for mental health. Systematic reviews, secondary sources, and non-English publications were excluded. Random-effects meta-analysis was conducted using standardized mean differences, with effect sizes synthesized in forest plots. Twenty studies were included in the qualitative synthesis and six in the quantitative meta-analysis. The analysis demonstrated that AI-based CAs significantly reduce anxiety (Cohen’s d = 0.62, p < 0.01) and depression (Cohen’s d = 0.74, p < 0.001), with higher effectiveness observed in multimodal CAs compared to text-only systems. However, the long-term impact remains inconsistent due to variability in follow-up durations and methodological heterogeneity. Some studies lacked extended observation periods or reported diminished effects over time, highlighting a need for sustained intervention research. AI-based CAs, especially when integrated into mobile platforms and using multimodal interfaces, provide scalable and engaging support for mental health. While short-term benefits are evident, future studies should address long-term efficacy, methodological consistency, and ethical concerns like privacy and algorithmic bias to strengthen the utility and trust in AI interventions for mental health.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Community Dentistry Department BADC, SMBBMU, Larkana, Pakistan.
- School of Dental Sciences, Health Campus, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
- Paediatric Dentistry Unit, School of Dental Sciences, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
- Periodontology Unit, School of Dental Sciences, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
- Department of Statistics, Shahjalal University of Science & Technology, Sylhet, Bangladesh.
- Biostatistics Unit, School of Dental Sciences, Health Campus, Universiti Sains Malaysia, Health Campus, Kubang Kerian, Kelantan, Malaysia.
ORCID-Profile: Al Mahmud, Mohamad Arif Awang Nawi
Zitieren
APA
Humayun, A., Madawana, A. M., Hassan, A., Mahmud, A., Kamaruddin, N., Noor, S. H., Noor, S. H., & Awang Nawi, M. A. (2025). Artificial intelligence as a predictive tool for mental health status: Insights from a systematic review and meta-analysis. PLOS One, 20(9), e0332207. https://doi.org/10.1371/journal.pone.0332207
Vancouver
Humayun A, Madawana AM, Hassan A, Mahmud A, Kamaruddin N, Noor SH, et al. Artificial intelligence as a predictive tool for mental health status: Insights from a systematic review and meta-analysis. PLOS One. 2025;20(9):e0332207. doi:10.1371/journal.pone.0332207
BibTeX
@article{Humayun2025,
author = {Humayun, Arsalan and Madawana, Ashwini M and Hassan, Akram and Mahmud, Al and Kamaruddin, Noorshaida and Noor, Syed Husni and Noor, Syed Hatim and Awang Nawi, Mohamad Arif},
title = {Artificial intelligence as a predictive tool for mental health status: Insights from a systematic review and meta-analysis},
journal = {PLOS One},
year = {2025},
volume = {20},
number = {9},
pages = {e0332207},
doi = {10.1371/journal.pone.0332207}
}
EUVIMED