JMIR Mental Health

Remotely Supervised, Home-Based Transcranial Direct Current Stimulation for Major Depressive Disorder: Systematic Review and Meta-Analysis

2026-09-04T16:45:11-04:00

Background: Major depressive disorder affects over 280 million people worldwide, and access to effective treatment remains limited. Transcranial direct current stimulation (tDCS) is a noninvasive option, and portable devices now allow for home-based delivery under varying degrees of remote supervision. Objective: This study aimed to systematically review and meta-analyze the efficacy, safety, feasibility, and acceptability of home-based and remotely supervised tDCS for depressive disorders. Methods: Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 and PRISMA-S (Preferred Reporting Items for Systematic Reviews and Meta-Analyses literature search extension) guidelines, we searched MEDLINE, Embase, Web of Science, the Cochrane databases, ClinicalTrials.gov, and the World Health Organization International Clinical Trials Registry Platform up to July 2025, with backward and forward citation searching. Two reviewers independently screened records, extracted data, and assessed risk of bias (version 2 of the Cochrane risk-of-bias tool for randomized trials, Newcastle-Ottawa Scale for observational studies, and Critical Appraisal Skills Programme for qualitative studies) and certainty of evidence (Grading of Recommendations Assessment, Development, and Evaluation; GRADE). Results: This review included 12 distinct studies (16 reports), of which 6 (50%) were randomized sham-controlled trials forming the meta-analytic pool. Active home-based tDCS produced a small, statistically significant improvement over sham (pooled Hedges =0.36, 95% CI 0.06-0.66; =.03; =34.3%). The effect was not robust to removal of the single largest positive trial (omitting the one study from 2025: =0.39, 95% CI −0.12 to 0.91), and trial-level results were mixed: the 2 largest trials (one unsupervised [n=210] and one self-administered [n=141]) were negative on their primary depression outcomes, whereas the largest real-time supervised trial (n=174) was positive (between-group 95% CI 0.51‐4.01; =.01). This estimate was concordant in direction with an independent peer-reviewed meta-analysis of overlapping trials, which reported a pooled Montgomery-Åsberg Depression Rating Scale reduction (weighted mean difference −2.74, 95% CI −4.19 to −1.29) and Hamilton Depression Rating Scale reduction (weighted mean difference −2.24, 95% CI −4.16 to −1.49), attenuating to nonsignificance (>.05) in major depressive disorder without comorbid cognitive impairment. The pooled effect fell at or near the minimal clinically important difference. GRADE certainty was moderate. Adverse events were predominantly mild: one pilot study was terminated early for skin lesions, and one nonfatal suicide attempt occurred in an unsupervised trial. Conclusions: Home-based and remotely supervised tDCS produces a small, statistically significant but clinically modest antidepressant effect that is sensitive to the inclusion of the largest positive trial, with the 2 largest trials being negative. The available controlled evidence does not establish supervision intensity as a determinant of efficacy. Current data are insufficient to recommend routine clinical adoption; adequately powered trials with standardized supervision and longer follow-up are needed. Trial Registration: PROSPERO registration number CRD420251109275; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251109275

Medication Dispensing Patterns Among Individuals With Serious Mental Illness Using a Remote Medication Dispensing and Adherence Monitoring Platform: Cohort Study

2026-09-02T12:30:03-04:00

Background: Medication adherence is poor among individuals with serious mental illness (SMI). Few studies have demonstrated the effectiveness of remote medication dispensing and adherence monitoring interventions among individuals with SMI. Objective: This study aimed to understand medication dispensing rates for users of a remote medication dispensing and adherence monitoring device and to identify associated demographic and clinical characteristics. Methods: In this cohort study, individuals’ characteristics were measured at baseline, and dispensing records were followed from their enrollment and subsequent device installation as early as January 2019 until June 2023. Individuals were eligible to participate if they had an SMI diagnosis, were aged 18 to 64 years, were currently being prescribed psychiatric medications, and were receiving mental health services from a participating community mental health agency. Participants were recruited through a combination of self-selection and referrals from agency staff. Our intervention involved using a remote medication dispensing and adherence monitoring device to measure participants’ daily medication dispensing. Results: The final sample consisted of 99 participants. The mean age of the participants was 49 (SD 12.08) years; 64% (n=63) of the participants identified as men and 41% (n=41) as Black or African American. The overall dispensing rate was 92.9%, with 90 (91%) individuals having dispensing rates >80%. The results of the hierarchical Bayesian logistic regression model showed that participants adhered better to evening doses than morning doses (incidence rate ratio [IRR] 1.11, 95% credible interval [CrI] 1.06-1.16). Dispensing adherence was poorer on weekends than on weekdays (IRR 0.87, 95% CrI 0.83-0.91). For every additional year of using the device, the rate of adherence increased by 1% (IRR 1.01, 95% CrI 1.00-1.01). The rate of dispensing dropped by 22% after the onset of the COVID-19 pandemic (IRR 0.78, 95% CrI 0.71-0.86), and African American participants had a 29% lower rate of dispensing than White participants (IRR 0.71, 95% CrI 0.55-0.90). The rate of dispensing did not differ by age; sex; educational attainment; or the level of sadness, emotional and behavioral dyscontrol, cognitive function, or psychotic symptoms at baseline. Conclusions: The high adherence rate observed, regardless of baseline psychopathology levels, highlights the potential of remote medication dispensing and adherence monitoring devices to address adherence challenges in people with SMI. Observed variation in dispensing behavior by dose timing and contextual factors suggests opportunities for intervention, including aligning dosing schedules with patient routines, providing additional support during periods of disruption (eg, weekends or major life events), and tailoring strategies to address disparities across patient groups. These findings highlight the role of targeted, context-aware approaches to improve adherence in community-based SMI care. These findings support the integration of digital adherence monitoring within mental health services, especially in settings where traditional adherence support may be challenging. Clinical Trial: ClinicalTrials.gov NCT03775044; https://clinicaltrials.gov/study/NCT03775044

Large Language Model–Based Behavioral Activation Chatbot for Young People With Depression Using Artificial Users and Clinical Experts: Mixed Methods Evaluation

2026-09-01T15:00:18-04:00

Background: Mental health chatbots are increasingly used to support people with depressive symptoms, and large language models make these systems more flexible than rule-based chatbots. However, it remains unclear how well large language model–based chatbots deliver structured psychological interventions. Objective: This study examined how well a GPT-4o–based chatbot delivered a behavioral activation intervention for young people with depression using sessions with artificial users and clinical expert assessment. It also identified limitations and potential refinements. Methods: We implemented a GPT-4o (gpt-4o-2024-08-06; OpenAI)–based chatbot using a structured system prompt to deliver a single-session behavioral activation intervention for people with depression aged 14 to 29 years. We generated 48 sessions with GPT-4o–based artificial users derived from clinical vignettes varying across 7 characteristics. Ten clinical experts, either licensed psychotherapists or advanced psychotherapy trainees, independently assessed the sessions using the 14-item Quality of Behavioral Activation Scale (Q-BAS), rated from 0 to 6, supplemented by rating therapeutic capabilities, artificial user authenticity and difficulty, and qualitative feedback. Results: The chatbot completed all 7 intervention phases in every session. The mean holistic session quality rating was 3.94 (SD 1.23), and the mean Q-BAS rating was 4.03 (SD 1.18). Thirteen of 14 Q-BAS components exceeded the satisfactory threshold of 3. Ratings were highest for mood assessment (mean 5.42, SD 1.09) and activity planning (mean 4.98, SD 1.41) and lowest for explaining positive reinforcement (mean 2.92, SD 2.30) and supporting activity-mood monitoring (mean 3.02, SD 2.04). Therapeutic capability ratings were highest for message safety (mean 5.90, SD 0.37), message clarity (mean 5.56, SD 0.77), and objective, nonjudgmental communication (mean 5.17, SD 1.04) and lowest for therapeutic rapport (mean 4.12, SD 1.45) and natural conversation flow (mean 4.25, SD 1.42). Artificial users were rated below the scale midpoint for authenticity (mean 2.75, SD 1.41) and difficulty (mean 1.23, SD 1.46). Clinical experts described the chatbot as structured, clear, and safe but identified insufficient clinical reasoning as the main limitation, particularly in evaluating the therapeutic suitability and feasibility of activities, barriers, solution strategies, and rewards. Artificial users were often highly compliant, especially when identifying positive activities. Conclusions: In expert-rated sessions with artificial users, the chatbot delivered the behavioral activation intervention as intended and performed strongest on procedural components. It performed less well on positive reinforcement and activity-mood monitoring, indicating refinement needs in clinical reasoning, follow-up questioning, and evaluating whether proposed activities, plans, barriers, solution strategies, and rewards are therapeutically appropriate and feasible. The findings identify targets for improvement before testing with human users, while the artificial user design and expert ratings limit conclusions about real therapeutic interactions.

Exploring Real-World Use of AI Chatbots for Mental Health Support: Cross-Sectional Survey Study

2026-09-01T15:00:03-04:00

Background: In recent years, innovations in generative AI, in particular large language models (LLMs) in the form of AI chatbots, have found their way to the general public. First studies indicate a growing prevalence of individuals talking to AI chatbots about mental health–related topics; yet, knowledge of how, why, and which individuals are using these AI chatbots for their mental health is limited. Objective: This study aimed to provide insights into the use of AI chatbot–delivered mental health support. Methods: To do so, an online survey in 2 Belgian samples was conducted. A student sample was collected, and individuals who used AI chatbots for mental health support were included (approximately 40% of the student sample were eligible users). Second, a recruitment call for users of AI chatbots for mental health support was launched in the general public. Data from 349 participants, 276 members of the general public, and 73 students were included in the analyses. Results: Descriptive analyses were used to report the use of AI chatbot–delivered mental health support. Most respondents in the present samples were women and indicated having received or receiving professional mental health support. The most common conversation topics across both samples focused on personal and interpersonal issues. Freely accessible AI chatbots, mainly ChatGPT, were the predominant choice in both samples and were mainly chosen for their constant availability and accessibility. Respondents in the student sample also preferred their anonymity, and those in the general sample also used them because they felt supported by them. The preliminary associations found between digital working alliance and engagement factors in the student sample and in the general sample suggest that relational factors might play a role in sustained AI chatbot use but warrant further investigation. Conclusions: To conclude, this study confirms that general-purpose AI chatbots that are not designed or regulated for mental health support, specifically ChatGPT, are being used to obtain social and emotional support, often by individuals already familiar with professional mental health support.

Experiences and Perceptions of Crisis Text Services: Interview Study Among Young Adults With Suicidal Ideation

2026-08-26T17:00:09-04:00

Background: Suicide remains a leading cause of death among young adults aged 18 to 25 years. Young adults experiencing suicidal ideation (SI) are increasingly using crisis text services (CTSs), a free and accessible option for crisis intervention. Little is known about CTSs from the young adult perspective. Objective: This study aimed to characterize young adults’ experiences with and perceptions of CTSs for SI. Methods: We conducted in-depth interviews, by phone, Zoom, or text, with young adults (n=39) in the United States who had a lifetime history of SI. Participants included those who had or had not engaged with CTSs for SI. Semistructured interviews were conducted from January to July 2024. The data were analyzed using a modified grounded theory approach. Results: We constructed 5 key themes to characterize young adults’ perceptions of and experiences with CTSs for SI. Young adults perceived CTSs as a unique component of their mental health crisis management. They appreciated CTSs’ technological features, particularly the privacy they provided and the ability to reflect on and edit responses. However, they expressed dissatisfaction with the nonspecific nature of many CTS interactions. The perceived anonymity of CTSs served multiple functions, both as a motivator for CTS use and as a potential point of vulnerability, should it be lost during a CTS interaction. Participants’ perceptions of CTSs’ impact varied; some viewed them as beneficial, whereas others reported neutral or inconsistent effects over time. Conclusions: Among young adults with SI, CTSs are a key yet imperfect resource. Quality improvement and evaluation efforts may be needed to understand how responders can better tailor responses to improve conversational quality and consistency for young adult texters.

Correction: Using Psychological Artificial Intelligence (Tess) to Relieve Symptoms of Depression and Anxiety: Randomized Controlled Trial

2026-08-26T13:00:03-04:00

Large Language Models as a New Tool for Therapists in Internet-Based Cognitive Behavioral Therapy: Blinded Clinician Rating Pilot Experiment

2026-08-26T11:30:36-04:00

Background: Internet-based cognitive behavioral therapy (iCBT) is an effective and scalable alternative to face-to-face psychotherapy, but its reach is constrained by the time therapists spend reviewing patient input and manually drafting written responses. Studies suggest that large language models (LLMs) may be capable of generating high-quality therapeutic text, with the potential to support therapists in delivering treatment. Their suitability as therapist-support tools in structured iCBT, however, remains insufficiently studied. Objective: This study aims to assess the quality of LLM-generated iCBT responses to patient messages by comparing them to the quality of responses produced by humans. Methods: In a preregistered blinded clinician rating experiment, experienced clinicians assessed the quality of human-produced vs LLM-generated therapist responses within a simulated iCBT treatment for functional somatic disorder. Raters were exposed to a stimulus material consisting of 5 fictitious patient messages, each paired with 1 human and 1 LLM-generated response. Raters assessed message/response pairs on 5 quality dimensions (overall quality, helpfulness, empathy, professionalism, and protocol adherence) and were asked to indicate the source of the response (human/LLM). Analyses were primarily descriptive, supplemented by exploratory statistical tests and descriptive thematic content analysis of open-ended text fields. The full preregistered study protocol is available at Open Science Framework. Results: A total of 61 raters provided data, while 54 were eligible and included for analysis. Human- and LLM-generated responses were rated similarly across quality dimensions on a 1-5 scale: overall quality (LLM: mean 4.00, SD 0.54 vs human: mean 3.96, SD 0.53; d=0.06), helpfulness (LLM: mean 3.85, SD 0.57 vs human: mean 3.93, SD 0.49; d=0.13), professionalism (LLM: mean 4.25, SD 0.53 vs human: mean 4.11, SD 0.53; d=0.24), protocol adherence (LLM: mean 4.13, SD 0.52 vs human: mean 4.13, SD 0.54; d=0.03) and empathy (LLM: mean 4.31, SD 0.47 vs human: mean 4.08, SD 0.50; d=0.42). Raters correctly identified the source of human-generated responses (mean 79%, SD 19.65%) more accurately than LLM-generated responses (mean 63%, SD 21.30%). In all, 30/54 (55%) raters responded to one or more open text fields. Qualitative analysis indicated that LLM-generated responses were perceived as polished but also generic and at times excessively empathetic. Conclusions: LLM-generated responses were judged to be of comparable quality to those written by human therapists, though qualitative feedback indicated they were at times generic and insufficiently challenging. These findings provide initial support for the feasibility of using LLMs as therapist-support tools in iCBT, but further research is needed to determine whether their integration yields tangible clinical and organizational benefits.

When Markets Shape AI Mental Health Self-Management Tools: Consequences for Serious Mental Illness

2026-08-25T16:00:04-04:00

AI-enabled self-management health tools are increasingly promoted within health care policy as part of digital self-management models for mental health care. However, development is concentrated on scalable, low-intensity interventions for common conditions, such as anxiety and depression, rather than on populations with the greatest clinical need, such as those with serious mental illness (SMI). This includes schizophrenia-spectrum and bipolar disorders, which remain comparatively underserved, despite experiencing a disproportionate burden of morbidity and service use. However, the clinical features of SMI—comprising fluctuating symptoms, multimorbidity, and elevated risk—may limit the suitability of low-intensity AI-driven self-management tools designed for mild-to-moderate conditions. In this Viewpoint, we argue that the scarcity of AI-enabled self-management tools for SMI reflects a structural feature of current innovation systems rather than one of technical infeasibility alone. Market incentives, regulatory pathways, and fragmented research pipelines favor low-risk, high-volume populations, thereby limiting development for clinically complex groups. Addressing this unevenness is essential and will require upstream intervention, including targeted public funding, improved data infrastructure, and administrative frameworks that support safe innovation in high-risk populations. Embedding equity for SMI as a primary design requirement will be necessary to ensure that AI-driven mental health tools do not reinforce current inequalities.

Association of Supportive Text Messaging With Health Service Utilization Costs Following Psychiatric Admission: Secondary Economic Analysis of a Stepped-Wedge Cluster Randomized Trial

2026-08-21T11:45:12-04:00

Background: Psychiatric hospitalizations are a major driver of mental health–related health care costs, with readmission risk and service utilization being highest in the months following discharge. Scalable, low-cost postdischarge interventions that reduce inpatient utilization are therefore of high policy relevance. Objective: This secondary exploratory economic study used linked administrative health care utilization data to evaluate the economic impact of supportive text messaging (SMS), alone and in combination with peer support services (PSS), following discharge from acute psychiatric care. Methods: This secondary exploratory economic evaluation was conducted alongside a pragmatic stepped-wedge cluster randomized trial in Alberta, Canada. Adults discharged from acute psychiatric inpatient care received treatment as usual (TAU), SMS, or SMS with or without PSS. Direct health system costs, including hospital care and physician costs, were assessed using administrative data for 6 months and 12 months before and after the index admission. Difference-in-difference analyses were used to estimate cost changes between groups. Results: At 6 months post discharge, the period of greatest new costs, participants receiving SMS experienced a significant reduction in hospital care costs compared with TAU (mean difference-in-differences −$7147; 95% CI −$12,388 to −$1906; =.01). Total health care costs were also significantly lower in the SMS group at 6 months (−$8160) relative to TAU. Sensitivity analyses demonstrated that the direction and statistical significance of the primary findings remained stable across alternative intervention costing assumptions and analytic specifications. No significant cost differences were observed at 12 months. The addition of peer support did not result in incremental cost savings at either time point. Conclusions: SMS was associated with significant reductions in hospital care and total health care utilization costs during the 6-month postdischarge period compared with TAU. These findings support the integration of cost-effective and scalable digital interventions into routine psychiatric discharge pathways to help improve system efficiency in publicly funded health systems. Trial Registration: ClinicalTrials.gov NCT05133726; https://clinicaltrials.gov/study/NCT05133726

Parasocial Engagement With Social Media Influencers and Mental Health Outcomes: Systematic Review and Meta-Analysis

2026-08-20T14:45:11-04:00

Background: Social media influencers occupy a pervasive role in billions of users’ daily digital lives, particularly among adolescents and young adults. Audiences develop parasocial engagement with these figures, including parasocial relationships (PSRs) and parasocial interactions (PSIs). Despite growing concern about their mental health implications, no prior meta-analysis has quantitatively synthesized this evidence. Objective: This systematic review and meta-analysis aimed to estimate the associations between influencer-directed parasocial engagement and mental health outcomes, examine prespecified moderators, and evaluate the quality of the existing evidence base. Methods: Seven databases (PsycINFO, Embase, MEDLINE, ERIC, PubMed, Web of Science, and Scopus) were searched from inception. Studies quantitatively assessing PSRs or PSIs with social media influencers and reporting mental health outcomes were eligible. A 3-level random-effects meta-analysis was conducted using Pearson . Primary pooled estimates were calculated separately for positive or adaptive outcomes, and negative or maladaptive outcomes. Moderators examined included outcome domain, parasocial construct type, age group, gender, cultural region, and platform. Results: Seventeen studies (52 effect sizes) were included. Parasocial engagement was positively associated with both positive (=28; =0.38, 95% CI 0.18-0.55) and negative outcomes (=24; =0.24, 95% CI 0.08-0.38), with positive effects significantly stronger. Well-being showed the largest effects (=0.42), followed by social media addiction (=0.33). PSI demonstrated stronger associations than PSR (=0.56 vs 0.22). Effects were largest among adolescents (=0.44) and in Eastern samples (=0.61 vs 0.29). No evidence of publication bias was detected (Egger =.94; fail-safe N=18,643). Conclusions: Parasocial engagement functions as a psychological double-edged sword, reliably linked to both enhanced well-being and problematic engagement. Positive mental health outcomes, particularly those involving well-being, were generally stronger and more consistent than negative ones. These findings suggest that parasocial engagement should not be understood as uniformly beneficial or harmful; rather, its psychological meaning depends on the outcome domain, parasocial construct type, developmental stage, and cultural context. Future longitudinal, mechanism-based research with diverse samples is needed to clarify when parasocial engagement is most beneficial or harmful. Trial Registration: PROSPERO CRD420261291637; https://www.crd.york.ac.uk/PROSPERO/view/CRD420261291637