Literature Collection

BACKGROUND: Mental health care systems worldwide face critical challenges, including limited access, shortages of clinicians, and stigma-related barriers. In parallel, large language models (LLMs) have emerged as powerful tools capable of supporting therapeutic processes through natural language understanding and generation. While previous research has explored their potential, a comprehensive review assessing how LLMs are integrated into mental health care, particularly beyond technical feasibility, is still lacking. OBJECTIVE: This systematic literature review investigates and conceptualizes the application of LLMs in mental health care by examining their technical implementation, design characteristics, and situational use across different touchpoints along the patient journey. It introduces a 3-layer morphological framework to structure and analyze how LLMs are applied, with the goal of informing future research and design for more effective mental health interventions. METHODS: A systematic literature review was conducted across PubMed, IEEE Xplore, JMIR, ACM, and AIS databases, yielding 807 studies. After multiple evaluation steps, 55 studies were included. These were categorized and analyzed based on the patient journey, design elements, and underlying model characteristics. RESULTS: Most studies assessed technical feasibility, whereas only a few examined the impact of LLMs on therapeutic outcomes. LLMs were used primarily for classification and text generation tasks, with limited evaluation of safety, hallucination risks, or reasoning capabilities. Design aspects, such as user roles, interaction modalities, and interface elements, were often underexplored, despite their significant influence on user experience. Furthermore, most applications focused on single-user contexts, overlooking opportunities for integrated care environments, such as artificial intelligence-blended therapy. The proposed 3-layer framework, which consists of the L1: LLM layer, L2: interface layer, and L3: situation layer, highlights critical design trade-offs and unmet needs in current research. CONCLUSIONS: LLMs hold promise for enhancing accessibility, personalization, and efficiency in mental health care. However, current implementations often overlook essential design and contextual factors that influence real-world adoption and outcomes. The review underscores that the self-attention mechanism, a key component of LLMs, alone is not sufficient. Future research must go beyond technical feasibility to explore integrated care models, user experience, and longitudinal treatment outcomes to responsibly embed LLMs into mental health care ecosystems.

BACKGROUND: Citizens affected by substance use disorders are high-risk populations for both SARS-CoV-2 infection and COVID-19-related mortality. Relevant vulnerabilities to COVID-19 in people who suffer substance use disorders are described in previous communications. The COVID-19 pandemic offers a unique opportunity to reshape and update addiction treatment networks. MAIN BODY: Renewed treatment systems should be based on these seven pillars: (1) telemedicine and digital solutions, (2) hospitalization at home, (3) consultation-liaison psychiatric and addiction services, (4) harm-reduction facilities, (5) person-centered care, (6) promote paid work to improve quality of life in people with substance use disorders, and (7) integrated addiction care. The three "best buys" of the World Health Organization (reduce availability, increase prices, and a ban on advertising) are still valid. Additionally, new strategies must be implemented to systematically deal with (a) fake news concerning legal and illegal drugs and (b) controversial scientific information. CONCLUSION: The heroin pandemic four decades ago was the last time that addiction treatment systems were updated in many western countries. A revised and modernized addiction treatment network must include improved access to care, facilitated where appropriate by technology; more integrated care with addiction specialists supporting non-specialists; and reducing the stigma experienced by people with SUDs.

IMPORTANCE: More than two-thirds of US adults who screen positive for depression in the primary care setting do not receive treatment. These adults need evidence-based and scalable interventions. OBJECTIVE: To determine the effectiveness of Moodivate, a self-directed digital intervention for mental health, in treating depression symptoms among patients in the primary care setting. DESIGN, SETTING, AND PARTICIPANTS: This 3-group decentralized randomized clinical trial recruited participants from September 22, 2021, to December 27, 2023, and completed data collection on March 29, 2024. Adult patients with at least moderate symptoms of depression on the Patient Health Questionnaire-9 (score ≥10) were enrolled from 22 primary care clinics in South Carolina. INTERVENTIONS: Participants received Moodivate (a digital behavioral activation intervention), Moodivate with health care provider access to information on patient use of the digital behavioral activation intervention in the electronic health record (EHR), or usual care for depression. MAIN OUTCOMES AND MEASURES: The primary outcome was a change in depression symptoms on the Beck Depression Inventory-II (BDI-II) over 12 weeks. Secondary outcomes included a clinically significant improvement in depression symptoms on the BDI-II (10-point decrease in score), depression remission on the BDI-II (score ≤13), digital behavioral activation intervention engagement, and primary care provider (a physician or other health care professional who is responsible for a patient's primary care) use of the EHR features. RESULTS: Among 649 participants, 495 (76%) were female and the mean (SD) age was 44.68 (15.22) years. Participants who received the digital behavioral activation intervention, with and without EHR integration, had significantly improved depression symptoms vs those who received usual care over 12 weeks (least squares mean change from baseline for Moodivate: -10.34; SE = 0.82; d = 0.98; Moodivate with EHR: -9.88; SE = 0.81; d = 0.93; usual care: -5.94; SE = -0.80; d = 0.54). Participants in the Moodivate groups had 2.5 to 3.0 times higher odds of having a clinically significant improvement in depression symptoms (Moodivate: OR, 2.98 [97.5% CI, 1.69-5.27]; P < .001; Moodivate with EHR: OR, 2.53 [97.5% CI, 1.45-4.41]; P < .001) and 2.3 to 2.6 times higher odds of experiencing depression remission (Moodivate: OR, 2.27 [97.5% CI, 1.16-4.44; P = .006; Moodivate with EHR: OR, 2.63 [97.5% CI, 1.38-5.04]; P < .001) than participants who received usual care. Participant engagement with Moodivate was high in the first month (68% to 100% weekly retention), and 33% of patients continued to use the digital behavioral activation intervention after 12 weeks. Fourteen percent of primary care providers who received access used the EHR functionality. CONCLUSIONS AND RELEVANCE: This randomized clinical trial found that a digital behavioral activation intervention is effective for treating adults with at least moderate symptoms of depression in the primary care setting. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04463914.