Anne L. Glowinski

Background: Large language models (LLMs) and, more recently, large reasoning models (LRMs) have rapidly garnered significant interest for application in psychiatry and behavioral health. However, recent studies have identified significant shortcomings and potential risks in the performance of LLM-based systems, complicating their application to psychiatric diagnosis. Two promising approaches to addressing these challenges and improving the efficacy of these models are simulated reasoning (SR) and self-verification (SV), in which additional “reasoning tokens” are used to guide model output, either during or after inference. Objective: We aimed to explore how the use of SR (via LRMs) and SV (via supplemental prompting) affects the psychiatric diagnostic performance of LLMs. Methods: 106 case vignettes and associated diagnoses were extracted from the DSM-5-TR (Diagnostic and Statistical Manual, Version 5, Text Revision) Clinical Cases book, with permission. Both an LLM and an LRM model were selected from the latest available model generation for each of the two vendors studied (OpenAI and Google). Two inference approaches were developed: a Basic approach that directly prompted models to provide diagnoses and a SV approach that augmented the Basic approach with additional prompts. All case vignettes were processed by the two LLMs, two LRMs, and two inference approaches, and diagnostic performance was evaluated using the sensitivity and positive predictive value (PPV). Binomial generalized linear mixed models were used to test for significant differences between the model vendors (OpenAI, Google), type (LLM, LRM), and the addition of an SV prompt. Results: All vignettes were successfully processed by each model and inference approach. Sensitivity ranged from 0.732 to 0.817, and PPV ranged from 0.534 to 0.779. The best overall performance was found in the o3-pro LRM using SV, with a sensitivity of 0.782 and a PPV of 0.779. No statistically significant fixed effects were found for sensitivity. For PPV, a statistically significant effect was found for prompt type (SV, P =.007) and model type (LRM, P =.009). No significant interaction effects were identified. Conclusions: We found that both SR and SV yielded statistically significant improvements in the PPV, without significant differences in the sensitivity. The addition of the manually specified SV prompt improved the PPV even when simulated reasoning was used. This suggests that future efforts to apply language models in behavioral health could benefit from manually crafted reasoning prompts and automated SR.

Background: Large language models (LLMs), such as OpenAI’s GPT-3.5, GPT-4, and GPT-4o, have garnered early and significant enthusiasm for their potential applications within mental health, ranging from documentation support to chat-bot therapy. Understanding the accuracy and reliability of the psychiatric “knowledge” stored within the parameters of these models and developing measures of confidence in their responses (ie, the likelihood that an LLM response is accurate) are crucial for the safe and effective integration of these tools into mental health settings. Objective: This study aimed to assess the accuracy, reliability, and predictors of accuracy of GPT-3.5 (175 billion parameters), GPT-4 (approximately 1.8 trillion parameters), and GPT-4o (an optimized version of GPT-4 with unknown parameters) with standardized psychiatry multiple-choice questions (MCQs). Methods: A cross-sectional study was conducted where 3 commonly available, commercial LLMs (GPT-3.5, GPT-4, and GPT-4o) were tested for their ability to provide answers to single-answer MCQs (N=150) extracted from the Psychiatry Test Preparation and Review Manual. Each model generated answers to every MCQ 10 times. We evaluated the accuracy and reliability of the answers and sought predictors of answer accuracy. Our primary outcome was the proportion of questions answered correctly by each LLM (accuracy). Secondary measures were (1) response consistency to MCQs across 10 trials (reliability), (2) the correlation between MCQ answer accuracy and response consistency, and (3) the correlation between MCQ answer accuracy and model self-reported confidence. Results: On the first attempt, GPT-3.5 answered 58.0% (87/150) of MCQs correctly, while GPT-4 and GPT-4o answered 84.0% (126/150) and 87.3% (131/150) correctly, respectively. GPT-4 and GPT-4o showed no difference in performance (P=.51), but they significantly outperformed GPT-3.5 (P<.001). GPT-3.5 exhibited less response consistency on average compared to the other models (P<.001). MCQ response consistency was positively correlated with MCQ accuracy across all models (r=0.340, 0.682, and 0.590 for GPT-3.5, GPT-4, and GPT-4o, respectively; all P<.001), whereas model self-reported confidence showed no correlation with accuracy in the models, except for GPT-3.5, where self-reported confidence was weakly inversely correlated with accuracy (P<.001). Conclusions: To our knowledge, this is the first comprehensive evaluation of the general psychiatric knowledge encoded in commercially available LLMs and the first study to assess their reliability and identify predictors of response accuracy within medical domains. The findings suggest that GPT-4 and GPT-4o encode accurate and reliable general psychiatric knowledge and that methods, such as repeated prompting, may provide a measure of LLM response confidence. This work supports the potential of LLMs in mental health settings and motivates further research to assess their performance in more open-ended clinical contexts.

BACKGROUND Large language models (LLMs), such as OpenAI’s GPT-3.5, GPT-4, and GPT-4o, have garnered early and significant enthusiasm for their potential applications within mental health, ranging from documentation support to chat-bot therapy. Understanding the accuracy and reliability of the psychiatric “knowledge” stored within the parameters of these models and developing measures of confidence in their responses (ie, the likelihood that an LLM response is accurate) are crucial for the safe and effective integration of these tools into mental health settings. OBJECTIVE This study aimed to assess the accuracy, reliability, and predictors of accuracy of GPT-3.5 (175 billion parameters), GPT-4 (approximately 1.8 trillion parameters), and GPT-4o (an optimized version of GPT-4 with unknown parameters) with standardized psychiatry multiple-choice questions (MCQs). METHODS A cross-sectional study was conducted where 3 commonly available, commercial LLMs (GPT-3.5, GPT-4, and GPT-4o) were tested for their ability to provide answers to single-answer MCQs (N=150) extracted from the Psychiatry Test Preparation and Review Manual. Each model generated answers to every MCQ 10 times. We evaluated the accuracy and reliability of the answers and sought predictors of answer accuracy. Our primary outcome was the proportion of questions answered correctly by each LLM (accuracy). Secondary measures were (1) response consistency to MCQs across 10 trials (reliability), (2) the correlation between MCQ answer accuracy and response consistency, and (3) the correlation between MCQ answer accuracy and model self-reported confidence. RESULTS On the first attempt, GPT-3.5 answered 58.0% (87/150) of MCQs correctly, while GPT-4 and GPT-4o answered 84.0% (126/150) and 87.3% (131/150) correctly, respectively. GPT-4 and GPT-4o showed no difference in performance (P=.51), but they significantly outperformed GPT-3.5 (P<.001). GPT-3.5 exhibited less response consistency on average compared to the other models (P<.001). MCQ response consistency was positively correlated with MCQ accuracy across all models (r=0.340, 0.682, and 0.590 for GPT-3.5, GPT-4, and GPT-4o, respectively; all P<.001), whereas model self-reported confidence showed no correlation with accuracy in the models, except for GPT-3.5, where self-reported confidence was weakly inversely correlated with accuracy (P<.001). CONCLUSIONS To our knowledge, this is the first comprehensive evaluation of the general psychiatric knowledge encoded in commercially available LLMs and the first study to assess their reliability and identify predictors of response accuracy within medical domains. The findings suggest that GPT-4 and GPT-4o encode accurate and reliable general psychiatric knowledge and that methods, such as repeated prompting, may provide a measure of LLM response confidence. This work supports the potential of LLMs in mental health settings and motivates further research to assess their performance in more open-ended clinical contexts. CLINICALTRIAL

Telemedicine has emerged as an innovative platform to diagnose and treat psychiatric disorders in a cost-effective fashion. Previous studies have laid the functional framework for monitoring and treating child psychiatric disorders electronically using videoconferencing, mobile phones (smartphones), and Web-based apps. However, phone call and text message (short message service, SMS) interventions in adolescent psychiatry are less studied than other electronic platforms. Further investigations on the development of these interventions are needed. The aim of this paper was to explore the utility of text message interventions in adolescent psychiatry and describe a user feedback-driven iterative design process for text message systems. We developed automated text message interventions using a platform for both depression (EpxDepression) and autism spectrum disorder (ASD; EpxAutism) and conducted 2 pilot studies for each intervention (N=3 and N=6, respectively). The interventions were prescribed by and accessible to the patients' healthcare providers. EpxDepression and EpxAutism utilized an automated system to triage patients into 1 of 3 risk categories based on their text responses and alerted providers directly via phone and an online interface when patients met provider-specified risk criteria. Rapid text-based feedback from participants and interviews with providers allowed for quick iterative cycles to improve interventions. Patients using EpxDepression had high weekly response rates (100% over 2 to 4 months), but exhibited message fatigue with daily prompts with mean (SD) overall response rates of 66.3% (21.6%) and 64.7% (8.2%) for mood and sleep questionnaires, respectively. In contrast, parents using EpxAutism displayed both high weekly and overall response rates (100% and 85%, respectively, over 1 to 4 months) that did not decay significantly with time. Monthly participant feedback surveys for EpxDepression (7 surveys) and EpxAutism (18 surveys) preliminarily indicated that for both interventions, daily messages constituted the "perfect amount" of contact and that EpxAutism, but not EpxDepression, improved patient communication with providers. Notably, EpxDepression detected thoughts of self-harm in patients before their case managers or caregivers were aware of such ideation. Text-message interventions in adolescent psychiatry can provide a cost-effective and engaging method to track symptoms, behavior, and ideation over time. Following the collection of pilot data and feedback from providers and patients, larger studies are already underway to validate the clinical utility of EpxDepression and EpxAutism. Clinicaltrials.gov NCT03002311; https://clinicaltrials.gov/ct2/show/NCT03002311 (Archived by WebCite at http://www.webcitation.org/6qQtlCIS0).