EUVIMEDObjevte EUVIMED
ValidierungsstudieOpen access

Use of a Large Language Model to Reveal Narrative Architectures of Veteran Transition Stress: Development and Validation Study

Isaac R Galatzer-Levy, Xi Pan, Roland P Hart, George A Bonanno

JMIR Mental Health · 2026-04-30 · 13 · e90155-e90155

DOI: 10.2196/90155 ↗

At a glance

Validierungsstudie in JMIR Mental Health, published online on 2026-04-30. Written by Isaac R Galatzer-Levy and 3 co-authors. Institution: Google DeepMind, New York, NY, United States. Major MEDLINE topics: Large Language Models, Veterans, Stress, Psychological. 22 references. The full text is freely available in PubMed Central.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

Background

The stress caused by multiple aspects of veterans’ transitions from military to civilian, termed transition stress, represents a unique source of psychological impact that is underresearched due to its qualitative nature. The assessment of this complex psychological phenomena has thus relied on laborious interviews designed to extract quantitative information from qualitative narratives of the transition to civilian life. We sought to determine if large language models (LLMs) could be used as valid measurement tools to extract relevant information from open-ended narratives.

Objective

This study sought to develop and validate a generative artificial intelligence (AI) approach to automate the quantification and subsequent thematic analysis of veteran transition stress.

Methods

Utilizing transcripts from interviews of a sample of US military veterans, we developed an LLM to rate transition stress severity and examined the model’s reliability in relation to human coders and validity in relation to a set of related questionnaire measures. Next, we used the LLM scores to quantitatively define high and low transition stress groups, enabling a targeted, automated analysis of themes related to narrative identity and life transition themes that might differentiate the two groups.

Results

LLM ratings of transition stress correlated highly with the human expert ratings and showed significant, theoretically congruent correlations with measures of clinical symptoms, reintegration difficulties, and veterans’ self-ratings of transition difficulty. Critically, the AI-derived thematic analyses of the narratives from high and low transition stress veterans revealed clearly distinct and informative patterns.

Conclusions

These findings suggest that generative AI offers a robust, scalable, and reliable method for multidimensional analysis of complex, narrative-based psychological constructs.

Subjects

MeSH terms (MEDLINE)

Large Language ModelsVeteransStress, PsychologicalHumansReproducibility of ResultsNarrationGenerative Artificial IntelligenceMaleAdultMiddle AgedFemale

Author keywords

large language modelsnatural language processingpsychological assessmentthematic analysistransition stressveteran mental health

Authors and institutions

  • Google DeepMind, New York, NY, United States.
  • Department of Clinical Psychology, Teachers College, Columbia University, 525 West 120th Street, Box 303, New York, NY, 10027, United States, 1 2126783468.

Cite

APA

Galatzer-Levy, I. R., Pan, X., Hart, R. P., & Bonanno, G. A. (2026). Use of a Large Language Model to Reveal Narrative Architectures of Veteran Transition Stress: Development and Validation Study. JMIR Mental Health, 13, e90155-e90155. https://doi.org/10.2196/90155

Vancouver

Galatzer-Levy IR, Pan X, Hart RP, Bonanno GA. Use of a Large Language Model to Reveal Narrative Architectures of Veteran Transition Stress: Development and Validation Study. JMIR Mental Health. 2026;13:e90155-e90155. doi:10.2196/90155

BibTeX
@article{GalatzerLevy2026,
  author = {Galatzer-Levy, Isaac R and Pan, Xi and Hart, Roland P and Bonanno, George A},
  title = {Use of a Large Language Model to Reveal Narrative Architectures of Veteran Transition Stress: Development and Validation Study},
  journal = {JMIR Mental Health},
  year = {2026},
  volume = {13},
  pages = {e90155--e90155},
  doi = {10.2196/90155}
}
Start for free →