A Large Language Model–Based System for Depression Symptom Identification: Iterative Qualitative Co-Design Studies
Ein auf einem Large Language Model basierendes System zur Identifikation von Depressionssymptomen: iterative qualitative Co-Design-Studien
JMIR Formative Research · 2026-11-03 · 10 · e95422-e95422
DOI: 10.2196/95422 ↗Auf einen Blick
In 2 Co-Design-Phasen wurde ein KI-System zur Identifikation von Depressionssymptomen mit 6 und 8 Fachpersonen entwickelt. Die Rückmeldungen führten zu 10 Symptomkategorien und wurden als klinisch nützlich beschrieben.
Fragestellung
Wie kann iterative Co-Design-Entwicklung mit Fachpersonen die klinische Relevanz und ethische Grundlage eines KI-Systems zur Identifikation von Depressionssymptomen prägen?
Methode
2-stufige qualitative Co-Design-Studie. In Stufe 1 wurde DepressionSymp 1.0 mit 6 Fachpersonen in Online-Workshops geprüft; in Stufe 2 DepressionSymp 2.0 mit 8 weiteren Fachpersonen.
Ergebnisse
Fachpersonen bewerteten strukturierte Symptom-Ausgaben als hilfreich für geringere kognitive Belastung und Interventionsplanung. Die Kategorie Risikoeinstufung wurde als ethisch besonders sensibel genannt; in Stufe 2 wurden 10 Kategorien als klarer und nützlicher beschrieben.
Einschränkungen
Texteingabe ohne nonverbale Informationen; kulturelle und sprachliche Unterschiede in der Symptomausprägung; mögliche Automation Bias ohne ausreichende menschliche Aufsicht.
Schlussfolgerung der Autor:innen
Iteratives Co-Design mit Fachpersonen ist wertvoll für klinisch relevante und ethisch begründete KI-Tools. Qualitatives Feedback kann die Modellanpassung und das ethische Design schrittweise verbessern.
Kernaussagen
- 6 Fachpersonen prüften DepressionSymp 1.0, 8 weitere DepressionSymp 2.0.
- Strukturierte Ausgaben reduzierten die kognitive Belastung und unterstützten die Planung.
- Als sensibel galten Risikokategorisierung, fehlende nonverbale Infos und kulturelle Unterschiede.
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
At a glance
Journal article in JMIR Formative Research, published online on 2026-11-03. Written by Abeer Alotaibi and 2 co-authors. Institution: School of Computing and Communication, Lancaster University, InfoLab21, South Dr, Bailrigg, Lancaster, England, LA1 4WA, United Kingdom, 44 778740822…. Major MEDLINE topics: Depression. 131 references. The full text is freely available in PubMed Central.
Abstract (original language)
Background
AI technologies have considerable potential to support mental health practitioners in identifying depression symptoms. However, existing tools are often developed with limited practitioner involvement and evaluated primarily on technical performance, leaving questions of clinical relevance, ethical design, and practitioner trust insufficiently addressed.
Objective
This study aimed to examine how iterative co-design with mental health practitioners can inform the development of an AI-based model for identifying depression symptoms, focusing on how practitioner input shaped its clinical relevance and ethical foundations across 2 design cycles.
Method
We conducted a 2-stage qualitative co-design study. In stage 1, we developed DepressionSymp 1.0, a fine-tuned GPT-3.5 model trained on the CounselChat dataset, identifying 5 symptom categories based on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition ( DSM-5 ) criteria and the Linguistic Inquiry and Word Count (LIWC-22) psycholinguistic framework, reviewed and discussed by 6 mental health practitioners during online co-design workshops. In stage 2, this feedback informed DepressionSymp 2.0, expanded to 10 symptom categories and reviewed by 8 further practitioners in a second round of workshops.
Results
Practitioners valued the structured symptom outputs for reducing cognitive workload and supporting intervention planning. In stage 1, indirect indicators such as social withdrawal, financial difficulties, and major life changes were often classified as lifestyle factors rather than symptoms; this informed the expansion to 10 categories in stage 2, described as clearer and more useful for preassessment. Risk categorization was identified as the most ethically sensitive aspect across both stages, alongside broader concerns including the lack of nonverbal information in text-based assessments, cultural and linguistic differences in symptom expression, and the potential for automation bias without sufficient human oversight.
Conclusions
Iterative co-design with mental health practitioners is valuable for developing AI-based tools that are clinically relevant and ethically grounded, demonstrating how qualitative practitioner feedback can inform iterative model refinement and ethical design. Findings inform design implications around large language model (LLM) features, practitioner-in-the-loop oversight, and ethical considerations covering clinical standards, informed consent, and data governance. Future work should prioritize larger-scale independent validation, formal safety evaluation, and culturally diverse dataset development.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- School of Computing and Communication, Lancaster University, InfoLab21, South Dr, Bailrigg, Lancaster, England, LA1 4WA, United Kingdom, 44 7787408223.
Cite
APA
Alotaibi, A., Sas, C., & Rahmani, H. (2026). A Large Language Model–Based System for Depression Symptom Identification: Iterative Qualitative Co-Design Studies. JMIR Formative Research, 10, e95422-e95422. https://doi.org/10.2196/95422
Vancouver
Alotaibi A, Sas C, Rahmani H. A Large Language Model–Based System for Depression Symptom Identification: Iterative Qualitative Co-Design Studies. JMIR Formative Research. 2026;10:e95422-e95422. doi:10.2196/95422
BibTeX
@article{Alotaibi2026,
author = {Alotaibi, Abeer and Sas, Corina and Rahmani, Hossein},
title = {A Large Language Model–Based System for Depression Symptom Identification: Iterative Qualitative Co-Design Studies},
journal = {JMIR Formative Research},
year = {2026},
volume = {10},
pages = {e95422--e95422},
doi = {10.2196/95422}
}
EUVIMED