Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models

Background: We evaluated whether a large language model could assist in selecting psychopharmacological treatments for adults with treatment-resistant depression. Methods: We generated 20 clinical vignettes reflecting treatment-resistant depression among adults based on distributions drawn from elec...

Full description

Saved in:
Bibliographic Details
Main Authors: Roy H. Perlis, Pilar F. Verhaak, Joseph Goldberg, Cristina Cusin, Michael Ostacher, Gin S. Malhi, Carlos A. Zarate, Richard C. Shelton, Dan V. Iosifescu, Mauricio Tohen, Manish Kumar Jha, Martha Sajatovic, Michael Berk
Format: Article
Language:English
Published: Elsevier 2025-12-01
Series:Journal of Mood and Anxiety Disorders
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2950004425000392
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849317140190986240
author Roy H. Perlis
Pilar F. Verhaak
Joseph Goldberg
Cristina Cusin
Michael Ostacher
Gin S. Malhi
Carlos A. Zarate
Richard C. Shelton
Dan V. Iosifescu
Mauricio Tohen
Manish Kumar Jha
Martha Sajatovic
Michael Berk
author_facet Roy H. Perlis
Pilar F. Verhaak
Joseph Goldberg
Cristina Cusin
Michael Ostacher
Gin S. Malhi
Carlos A. Zarate
Richard C. Shelton
Dan V. Iosifescu
Mauricio Tohen
Manish Kumar Jha
Martha Sajatovic
Michael Berk
author_sort Roy H. Perlis
collection DOAJ
description Background: We evaluated whether a large language model could assist in selecting psychopharmacological treatments for adults with treatment-resistant depression. Methods: We generated 20 clinical vignettes reflecting treatment-resistant depression among adults based on distributions drawn from electronic health records. Each vignette was evaluated by 2 expert psychopharmacologists to determine and rank the 5 best next-step pharmacologic interventions, as well as contraindicated or poor next-step treatments. Vignettes were then presented in random order, permuting gender and race, to a large language model (Qwen 2.5:7B), augmented with a synopsis of published treatment guidelines. Model output was compared to expert rankings, as well as to those of a convenience sample of community clinicians and an additional group of expert clinicians. Results: The augmented model prioritized the expert-designated optimal choice for 114/320 vignettes (35.6 %, 95 % CI 30.6 %–41.0 %; Cohen’s kappa = 0.34, 95 % CI 0.28–0.39). There were no vignettes for which any of the model choices were among the poor or contraindicated treatments. Results were not meaningfully different when gender or race of the vignette was permuted to examine risk for bias. A sample of community clinicians identified the optimal treatment choice for 12/91 vignettes (13.2 %, 95 % CI: 7.7–21.6 %; Cohen’s kappa = 0.10, 95 % CI 0.03–0.18), while an additional group of expert psychopharmacologists identified optimal treatment for 9/140 (6.4 %, 95 %CI: 3.4–11.8 %; Cohen’s kappa = 0.03, 95 % CI 0.01–0.08). Conclusion: An augmented language model demonstrated moderate agreement with expert recommendations and avoided contraindicated treatments, suggesting potential as a tool for supporting complex psychopharmacologic decision-making in treatment-resistant depression.
format Article
id doaj-art-41c99ac2b45749858c1b0fa87c8fb535
institution Kabale University
issn 2950-0044
language English
publishDate 2025-12-01
publisher Elsevier
record_format Article
series Journal of Mood and Anxiety Disorders
spelling doaj-art-41c99ac2b45749858c1b0fa87c8fb5352025-08-20T03:51:19ZengElsevierJournal of Mood and Anxiety Disorders2950-00442025-12-011210014210.1016/j.xjmad.2025.100142Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language modelsRoy H. Perlis0Pilar F. Verhaak1Joseph Goldberg2Cristina Cusin3Michael Ostacher4Gin S. Malhi5Carlos A. Zarate6Richard C. Shelton7Dan V. Iosifescu8Mauricio Tohen9Manish Kumar Jha10Martha Sajatovic11Michael Berk12Center for Quantitative Health and Department of Psychiatry, Massachusetts General Hospital, Boston, MA, United States; Department of Psychiatry, Harvard Medical School, Boston, MA, United States; Correspondence to: Massachusetts General Hospital, 185 Cambridge Street, 6th Floor, Boston, MA 02114, United States.Center for Quantitative Health and Department of Psychiatry, Massachusetts General Hospital, Boston, MA, United States; Department of Psychiatry, Harvard Medical School, Boston, MA, United StatesDepartment of Psychiatry, Mt. Sinai School of Medicine, New York, NY, United StatesDepartment of Psychiatry, Harvard Medical School, Boston, MA, United StatesDepartment of Psychiatry, Stanford University School of Medicine, Stanford, CA, United StatesAcademic Department of Psychiatry, Kolling Institute, Northern Clinical School, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia; CADE Clinic and Mood-T, Royal North Shore Hospital, Northern Sydney Local Health District, St. Leonards, NSW, Australia; Department of Psychiatry, and Oxford Uehiro Centre for Practical Ethics, Faculty of Philosophy,University of Oxford, Oxford, UKSection on the Neurobiology and Treatment of Mood Disorders, National Institute of Mental Health, National Institutes of Health, Bethesda, MD, United StatesDepartment of Psychiatry and Behavioral Neurobiology, Heersink School of Medicine, University of Alabama at Birmingham, Birmingham, AL, United StatesNYU School of Medicine and Nathan Kline Institute, NY, United StatesDepartment of Psychiatry and Behavioral Sciences, University of New Mexico School of Medicine, United StatesDepartment of Psychiatry, UT Southwestern Medical Center, Dallas, TX, United StatesUniversity Hospitals Cleveland Medical Center, Case Western Reserve University School of Medicine, Cleveland, OH, United StatesSchool of Medicine, Deakin University and Barwon Health, AustraliaBackground: We evaluated whether a large language model could assist in selecting psychopharmacological treatments for adults with treatment-resistant depression. Methods: We generated 20 clinical vignettes reflecting treatment-resistant depression among adults based on distributions drawn from electronic health records. Each vignette was evaluated by 2 expert psychopharmacologists to determine and rank the 5 best next-step pharmacologic interventions, as well as contraindicated or poor next-step treatments. Vignettes were then presented in random order, permuting gender and race, to a large language model (Qwen 2.5:7B), augmented with a synopsis of published treatment guidelines. Model output was compared to expert rankings, as well as to those of a convenience sample of community clinicians and an additional group of expert clinicians. Results: The augmented model prioritized the expert-designated optimal choice for 114/320 vignettes (35.6 %, 95 % CI 30.6 %–41.0 %; Cohen’s kappa = 0.34, 95 % CI 0.28–0.39). There were no vignettes for which any of the model choices were among the poor or contraindicated treatments. Results were not meaningfully different when gender or race of the vignette was permuted to examine risk for bias. A sample of community clinicians identified the optimal treatment choice for 12/91 vignettes (13.2 %, 95 % CI: 7.7–21.6 %; Cohen’s kappa = 0.10, 95 % CI 0.03–0.18), while an additional group of expert psychopharmacologists identified optimal treatment for 9/140 (6.4 %, 95 %CI: 3.4–11.8 %; Cohen’s kappa = 0.03, 95 % CI 0.01–0.08). Conclusion: An augmented language model demonstrated moderate agreement with expert recommendations and avoided contraindicated treatments, suggesting potential as a tool for supporting complex psychopharmacologic decision-making in treatment-resistant depression.http://www.sciencedirect.com/science/article/pii/S2950004425000392Major depressionArtificial intelligenceMachine learningPsychopharmacologyExpert consensus
spellingShingle Roy H. Perlis
Pilar F. Verhaak
Joseph Goldberg
Cristina Cusin
Michael Ostacher
Gin S. Malhi
Carlos A. Zarate
Richard C. Shelton
Dan V. Iosifescu
Mauricio Tohen
Manish Kumar Jha
Martha Sajatovic
Michael Berk
Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models
Journal of Mood and Anxiety Disorders
Major depression
Artificial intelligence
Machine learning
Psychopharmacology
Expert consensus
title Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models
title_full Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models
title_fullStr Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models
title_full_unstemmed Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models
title_short Clinical decision support for pharmacologic management of treatment-resistant depression with augmented large language models
title_sort clinical decision support for pharmacologic management of treatment resistant depression with augmented large language models
topic Major depression
Artificial intelligence
Machine learning
Psychopharmacology
Expert consensus
url http://www.sciencedirect.com/science/article/pii/S2950004425000392
work_keys_str_mv AT royhperlis clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT pilarfverhaak clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT josephgoldberg clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT cristinacusin clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT michaelostacher clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT ginsmalhi clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT carlosazarate clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT richardcshelton clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT danviosifescu clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT mauriciotohen clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT manishkumarjha clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT marthasajatovic clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels
AT michaelberk clinicaldecisionsupportforpharmacologicmanagementoftreatmentresistantdepressionwithaugmentedlargelanguagemodels