Objective Veterans Health Administration Clinical Resource Hubs are regional telehealth hubs that provide Veterans access needed to geriatric mental health (GMH) services without long drive times to urban centers to find specialists. This analysis characterized the GMH telehealth services to examine differences between rural and non‐rural Veterans. Methods Electronic health record data were extracted (October 2021 to September 2023) for encounters delivered by telehealth GMH teams staffed by geriatric psychiatrists, pharmacists, geropsychologists, and neuropsychologists. Variables included Veteran demographics, diagnoses, severity of care needs, services delivered and visit modality. Descriptive statistics, chi‐square and t ‐tests summarized the data and examined rural and non‐rural differences. Results GMH clinicians completed 9282 encounters with 2013 Veterans (Mean age = 75.8). Most services were video to home (48.2%) followed by video to clinic (25.4%), and phone (22%) with no differences in service delivery modality among rural and non‐rural Veterans. Rural Veterans had a higher proportion of psychotherapy and dementia‐related services, and lower proportions of medication management and services for depression, anxiety, and serious mental illness compared with non‐rural Veterans. Conclusions Older Veterans access virtual GMH care for a wide range of mental health and cognitive disorders. The apparent higher utilization of psychotherapy and dementia services for rural compared with non‐rural Veterans may relate to barriers to obtaining these services locally, although it may be partially explained by variations in virtual team compositions and settings served. Relevance to Clinical Practice This evaluation suggests high acceptability of video‐delivered care for older, rural Veterans with high care needs.
Background:Patients with schizophrenia experience gaps in the human immunodeficiency virus (HIV) care continuum. We examined the impact of dual enrollment in Medicare upon HIV testing and retention among Medicaid-enrolled patients with schizophrenia. Methods:Using Medicaid Analytic eXtract and Medicare files from 2008 to 2012, we examined adult participants with schizophrenia enrolled in Medicaid or dually enrolled in Medicare and Medicaid. Our outcomes were (a) HIV testing probability among participants without HIV and (b) retention in HIV care among people living with HIV (PLWH) (two HIV viral loads or CD4+ counts >90 days apart and within 1 year). Results:The average annual HIV testing probability was 4.9% among dually enrolled and 7.4% among Medicaid-enrolled patients (p < 0.001). Compared to Medicaid-only, HIV testing for dually enrolled patients were highest among those with opioid use disorder or sexually transmitted infection (STI); and lowest among study sample aged 50-59 years, Asian or White race. Among PLWH, the average retention probability in care was 54.7% for dually enrolled and 44.7% for Medicaid-enrolled patients (p < 0.001). Compared to Medicaid-only, retention probability for dually enrolled patients were highest among those with an STI or dyslipidemia, and lowest among younger sample. Conclusions and Relevance:Dual Medicare enrollment for Medicaid-enrolled patients benefitted HIV retention efforts more than HIV testing efforts. Differences in public insurance coverage between HIV care retention and testing among patients with schizophrenia, emphasizes the need for ongoing evaluation of the HIV care continuum as payor coverage and policies continue to evolve.
Objective:This review examined the historical development of artificial intelligence (AI) in psychiatry from 1972 to 2025 and identified persistent barriers to clinical translation. Method:The review used a structured source-identification approach across PubMed/MEDLINE, Web of Science, Google Scholar, ScienceDirect, citation tracking, and targeted historical searches. It synthesized sources qualitatively and organized them chronologically. A structured evidence table summarized representative milestones by era, AI paradigm, clinical task, data modality, validation approach, implementation status, and translational limitation. Results:Psychiatric AI evolved from symbolic simulation and rule-based expert systems to connectionist models, supervised machine learning, computational psychiatry, digital phenotyping, multimodal monitoring, digital mental health tools, and large language models. Despite increasing computational sophistication, recurring barriers persisted, including uncertain target validity, diagnostic heterogeneity, limited external validation, poor transportability, interpretability challenges, workflow integration, equity, patient trust, and governance. Across eras, technical progress was cyclical rather than linear, with successive waves reproducing unresolved clinical and implementation challenges. Conclusions:The clinical impact of psychiatric AI will likely depend less on algorithmic novelty alone than on clearer clinical targets, prospective validation, implementation trials, patient-centered evaluation, equity-sensitive generalizability, and mental health-specific governance. Relevance to Clinical Practice:For routine psychiatric care, AI tools require evidence of clinical validity, transportability, workflow compatibility, patient acceptability, equity, and appropriate governance rather than technical performance alone.
Objective:Prescription stimulants are commonly used to manage attention deficit hyperactivity disorder among women of reproductive age, yet their safety during pregnancy remains unclear. This study leveraged large-scale electronic health record data to assess whether use of these medications during pregnancy was associated with adverse pregnancy outcomes and to support clinicians in making informed treatment decisions. Methods:The study included adult women with deliveries from 2010 to 2022 who had two or more stimulant prescriptions before pregnancy. Participants were categorized into three mutually exclusive groups: stimulant, atomoxetine, and control (neither medication during pregnancy). The outcomes of interest included prenatal maternal, placental, fetal growth, labor-delivery, and birth-related complications occurring during pregnancy. Propensity score matching analysis was applied to reduce confounding effects in patient baseline characteristics, followed by logistic regression to assess the likelihood of each adverse pregnancy outcome type. Results:There were 32,152 pregnant women in the analysis, including 8394 (26.1%) prescribed stimulants, 405 (1.3%) prescribed atomoxetine, and 23,353 (72.7%) controls during pregnancy. The stimulant group was 1.2-3.1 times more likely to develop prenatal maternal, placental, and birth-related complications than the atomoxetine and control groups, while the atomoxetine group was 1.1-2.3 times more likely to experience fetal growth abnormality and labor-delivery issues than the control groups. Interestingly, the stimulant group showed 54% lower odds of fetal growth restriction than the atomoxetine group. Conclusions:This study highlighted differential risk profiles between stimulant and non-stimulant therapies, contributing evidence that informs future research on perinatal safety. Relevance to Clinical Practice:The distinct risk profiles identified for stimulant and atomoxetine therapies underscore the need for personalized treatment decisions during pregnancy.
Objective:Measurement-based care (MBC) is a vital tool in outpatient psychiatric care, enhancing treatment outcomes through systematic symptom monitoring. This study aimed to identify factors associated with remission and response in depression treatment using MBC, focusing on longitudinal changes in Patient Health Questionnaire-9 (PHQ-9) scores. Methods:Electronic health records of 4778 patients aged 12+ from an academic outpatient psychiatry clinic were analyzed. Inclusion criteria required a baseline PHQ-9 score > 9 and at least one follow-up PHQ-9 measurement. Factors associated with treatment response and remission, using different definitions, were evaluated using Cox regression models, adjusting for demographic and clinical covariates. Results:Response and remission rates were 46.3% and 28.2%, respectively. For response, significant associated factors were combined treatment (HR = 1.189, p = 0.014), follow-up PHQ-9 within 4-9 months (HR = 1.303, p < 0.001), and absence of post-traumatic stress disorder (PTSD) (HR = 0.817, p = 0.018). Key factors associated with faster remission (HR > 1) included combined psychiatry and psychotherapy treatment (HR = 1.189, p = 0.014), timely follow-up PHQ-9 administration within 4-9 months (HR = 1.255, p = 0.002), lower baseline PHQ-9 scores (HR = 0.931, p < 0.001), and absence of PTSD (HR = 0.751, p = 0.010). Alternative remission definitions (PHQ-9 < 12 or < 9) yielded similar findings. Conclusion:Integrated psychiatry-psychotherapy approaches, regular symptom monitoring, and addressing comorbid conditions like PTSD are critical for improving depression outcomes. The findings shed light on how MBC can be used to evaluate treatment effectiveness, inform clinical practice, and guide optimized treatment strategies. Relevance to Clinical Practice:Integrating MBC into a combined psychiatry-psychotherapy framework and systematically addressing comorbidities provides clinicians with powerful strategies to significantly improve depression remission rates.
