Background:While antidepressant side effects are often expected to diminish with continued treatment, empirical evidence remains limited. This study investigates co-trajectories of depressive severity with side-effect burden (SEB) and associated factors in patients with major depressive disorder (MDD). Methods:We analyzed longitudinal data from 1377 MDD outpatients across three Chinese multicenter studies (2015-2022), with assessments at baseline, weeks 2, 4, 8, 12, and 24. Depression severity was measured using the Quick Inventory of Depressive Symptomatology-Self Report (QIDS-SR16). At the same time, SEB was quantified using the frequency, intensity, and burden of side effects rating (FIBSER). Dual-trajectory modeling was used to identify distinct SEB and depression severity trajectories, with linear mixed-effects models comparing depression changes across SEB groups. Results:Four SEB trajectories were identified: no SEB (41.5%), early-onset SEB (29.1%), late-onset SEB (14.7%), and persistent SEB (13.9%). Depressive severity followed three trajectories: mild-responsive (66.6%), moderate-progressive (18.7%), and chronic-severe (14.7%). Persistent SEB was associated with higher baseline depression severity (OR = 1.13, 95% CI: 1.08-1.17), antidepressant combination (OR = 3.7, 95% CI: 1.52- 9.02), and poorer treatment outcomes. 7.3% exhibited concurrent chronic-severe depressive severity and persistent SEB. Female (OR = 1.95, 95% CI: 1.11-3.42), younger age (OR = 5.02, 95% CI: 2.63-9.55), higher education (high school: OR = 2.5, 95% CI: 1.12-5.36; bachelor and above: OR = 2.68, 95% CI: 1.24-5.82), and combination antidepressant (OR = 8.27, 95% CI: 2.69-25.45) were significant risk factors for concurrent severe symptoms and persistent SEB. Conclusion:Persistent antidepressant side effects coevolve with unfavorable depression trajectories over 6 months. Clinicians should prioritize early monitoring and tailored interventions for high-risk subgroups, particularly those with severe baseline symptoms or on combination therapy. These findings underscore the importance of continuous monitoring and personalized interventions to manage antidepressant side effects effectively.

BACKGROUND:Current treatment algorithms for major depressive disorder (MDD) lack dynamic prediction capabilities, leading to delayed therapeutic adjustments. This study sought to develop escitalopram-specific decision tree models to identify critical treatment adjustment time points and optimize personalized treatment strategies for MDD. METHODS:Using longitudinal data from two multicenter studies in China (2015-2020), we analyzed 800 patients with MDD receiving escitalopram monotherapy. Decision tree models incorporated baseline characteristics (age, BMI, disease duration, depressive symptoms) and dynamic treatment parameters (dose, 2-/4-week improvement) to predict full response (>50% symptom reduction) or non-full response (≤50% reduction) at weeks 2 and 4, and remission status (QIDS-SR16≤5 vs. >5) at week 8. Model performance was assessed by accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC). RESULTS:The week 2 model (n = 800) identified BMI, age, disease duration, course and baseline symptom severity as primary predictors (accuracy = 61.88%, NPV = 84.04%). By week 4 (n = 650), early response status (week 2) merged as a key predictor (accuracy = 69.23%, NPV = 71.62%). The week 8 model (n = 456) demonstrated enhanced predictive power, driven by life quality score, week 2/4 response status, and week 4 dosage (accuracy = 78.02%, PPV = 81.48%, NPV = 72.97%). Logistic regression confirmed week 4 response status as a significant predictor of week 8 outcome (p < 0.005). CONCLUSIONS:Week 4 emerges as a key decision point for escitalopram-treated MDD patients, where integration of baseline profiles, early response patterns, and dose parameters allows timely intervention. Our decision tree framework offers a methodological approach for dynamic decision points that warrant prospective validation and extension to other antidepressants.

Enhancing inpatient services for children and adolescent mental illness is a primary goal of medical systems, yet knowledge of child and adolescent psychiatric hospitalization utilization is limited. This study aimed to characterize hospitalizations and rehospitalizations over a decade and to identify factors associated with rehospitalizations. In this retrospective analysis, we obtained data from the Beijing Hospital Electronic Record Database from January 1, 2013, to December 31, 2022. We analyzed an admission-level dataset (n = 20215 admissions) to describe overall trends and established a patient-level cohort (n = 5048 unique patients) to evaluate rehospitalization risk. Modified Poisson regression models, with variables selected using the Andersen Behavioral Model, were used to identify factors associated with psychiatric rehospitalizations at 30-day, 180-day, and 365-day intervals. From 2013 to 2022, 20215 hospitalizations (involving 17175 unique patients) were analyzed. Mood disorders (50.6

Mental disorders affect nearly one billion individuals worldwide, yet professional psychiatric care remains constrained by workforce shortages and experience-dependent decision-making. Despite recent advances in large language models (LLMs), current applications in mental health are primarily patient-oriented and lack alignment with real-world psychiatric clinical workflows. Here we present PsychFound, a domain-adapted and clinician-oriented LLM developed to support psychiatric clinical practice. Developed through a three-phase framework using expert-curated psychiatric corpora and 64,588 Chinese real-world electronic health records, PsychFound integrates psychiatric professional knowledge, clinical reasoning capabilities and adaptation to the full spectrum of psychiatric clinical tasks across diagnosis, treatment planning and longitudinal management in Chinese clinical settings. In retrospective evaluations spanning three professional knowledge assessments and five clinical task benchmarks, the 7B-parameter PsychFound delivered the top overall performance among 22 LLMs. In a real-world, two-arm prospective study, resident psychiatrists assisted by PsychFound demonstrated higher consultation quality, higher diagnostic accuracy, more appropriate medication selection and reduced documentation time (all P < 0.01). A reader study with 60 psychiatrists (20 residents, 20 attendings and 20 seniors) showed that PsychFound's clinical reasoning performance matched that of attending psychiatrists. These findings demonstrated that PsychFound provides an interpretable, expert-level decision support tool capable of improving consistency, efficiency and standardization in psychiatric clinical care.

Background:Elevated body mass index (BMI) predicts poor treatment response in major depressive disorder (MDD), yet the neurobiological mechanisms underlying this association remain poorly understood. Cortical thickness asymmetry has been proposed as a structural correlate of affective regulation and may represent a pathway through which metabolic factors influence treatment outcomes. We investigated whether BMI-related alterations in prefrontal cortical asymmetry mediate antidepressant treatment resistance in MDD. Methods:We analyzed baseline structural MRI (sMRI) and 12-week clinical outcome data from 312 adults with MDD across a discovery cohort (n = 107) and an independent replication cohort (n = 205). Treatment response was operationalized as absolute reduction in Hamilton depression rating scale (HAMD-17) scores. Cortical thickness asymmetry indices were derived from regional parcellations. Associations between BMI and asymmetry were examined using linear regression models; mediation analyses tested whether BMI-related asymmetry statistically mediates the link between higher BMI and reduced treatment response. Sex-stratified analyses were conducted to identify divergent pathways. Results:Across both cohorts, higher BMI was consistently associated with greater leftward cortical thickness asymmetry in the prefrontal cortex (PFC). This structural asymmetry significantly mediated the relationship between BMI and poorer treatment response. Sex-stratified analyses revealed additional female-specific mediation through prefrontal opercular regions, with no corresponding effect in males. Transcriptomic annotation of implicated regions identified enrichment for genes involved in metabolic and cytoplasmic signaling pathways. Conclusions:BMI-associated leftward prefrontal asymmetry statistically mediates antidepressant resistance in MDD via both a sex-shared structural pathway and a female-specific opercular circuit. These findings suggest that metabolic factors may influence treatment outcomes partly through hemispheric structural imbalance in prefrontal regions and position cortical thickness asymmetry as a candidate neuroimaging biomarker for patient stratification in precision psychiatry.

BackgroundContinuous follow-up for patients with major depressive disorder (MDD) is essential for treatment decisions and a better prognosis. There remains limited evidence regarding the critical issue of depression variation trajectory prediction using mobile health (mHealth) measures. Moreover, the temporal dynamics of mHealth measures have not been fully modeled in previous studies, and the poor patient adherence to mHealth records poses great challenges to the dynamic feature modeling. ObjectiveThis study aimed to examine the contribution of mHealth measures in predicting depression variation trajectory for patients with MDD, with full consideration of the temporal dynamics of mHealth measures. MethodsA total of 229 patients with MDD from a multiple-center, prospective cohort were included. A 12-week follow-up was conducted involving the collection of the Hamilton Depression Rating Scale (HAMD-17), along with patient-reported outcomes (Immediate Mood Scaler and Altman Self-Rating Mania Scale) via mobile devices and sleep duration through wearable wristbands. We used functional data analysis to extract dynamic features from the sparse mHealth records, rather than aggregating the data to a single scalar summary measure through collapsing over time. Subsequently, 3 machine learning models were applied to predict the depression variation trajectory classes based on the baseline characteristics and these extracted dynamic features. ResultsBased on the variation of HAMD-17 scores within 12 weeks, the participants were labeled into 4 classes through the k-means algorithm. The classes included stable decline (n=93), fluctuate decline (n=44), fast decline (n=60), and delayed and fluctuate (n=32), in light of the shape of depression trajectories. With both baseline features and dynamic features of the mHealth measures, accuracy rates for the overall data were 54.35%, 60.87%, and 56.52%, for the stable decline patients were 78.95%, 84.21%, and 73.68%, for the nonstable decline patients were 59.26%, 62.96%, and 70.37% based on the 3 machine learning models, respectively. The results were significantly superior to the prediction obtained without mHealth measures (with an overall accuracy below 50%) and only showed a marginal reduction in accuracy relative to the ideal prediction with assessment obtained from clinical visits. Moreover, in the construction of the most accurate prediction model, dynamic features of the Immediate Mood Scaler, the Altman Self-Rating Mania Scale, and sleep duration emerged as the most influential predictors, ranking first, third, and fourth, respectively, in terms of their relative importance. ConclusionsLongitudinal mHealth measures show potential in depression variation trajectory monitoring for patients with MDD even under poor patient adherence. Our work provides practical help in alleviating the follow-up burden for patients with MDD and validates the effectiveness of mHealth measures in clinical applications.

Gut microbiota may influence antidepressant treatment outcomes, yet whether targeted modulation can enhance efficacy remains unclear. We conducted a randomized, double-blind, placebo-controlled trial in patients with major depressive disorder, administering a 2-week course of fecal microbiota transplantation (FMT) capsules or placebo as an adjunct to escitalopram (ChiCTR2300071421). Remission rates at week 8 did not differ significantly between groups, but FMT produced greater reductions in Hamilton Depression Rating Scale, 17-item version (HAMD-17) scores at weeks 2 and 8. FMT was well-tolerated with a safety profile comparable to placebo. Multi-omics analyses show durable donor microbial engraftment and enrichment of beneficial Lachnospiraceae and Oscillospiraceae taxa. Microbial remodeling is accompanied by an increase in serum bile acids that correlate with the alleviation of depressive symptoms. Mediation analysis supports a bile-acid-mediated suppression of inflammatory pathways linking microbial changes to antidepressant effects. Overall, FMT may provide a safe avenue to enhance escitalopram efficacy through microbiota-directed regulation of bile-acid metabolism and inflammation.

Abstract BackgroundIndividuals with schizophrenia or bipolar disorder face a significantly elevated risk of obesity, primarily due to weight gain associated with psychiatric medications and lifestyle factors. While digital self-monitoring tools offer scalable solutions, their application remains underexplored in psychiatric populations. To address these gaps, this type 1 hybrid effectiveness-implementation study investigates the preliminary effectiveness and implementation feasibility of a mobile health-assisted weight management intervention for patients with severe mental illness. ObjectiveThis study aims to evaluate the preliminary effectiveness and implementation feasibility of a mobile health–assisted weight management intervention for patients with severe mental illness transitioning from inpatient care to community-based recovery. MethodsThis single-center, open-cohort stepped-wedge cluster randomized trial with a 2-month step duration will recruit 204 patients from 6 clinical units. Clusters are randomized into 2 waves, with staggered transitions to a digital intervention, including smart scales, health apps, and biweekly educational modules, over a 6-month observation period. The design evaluates the intervention across the transition from inpatient care to community-based recovery. The primary outcome is the proportion of participants achieving ≥5% weight loss at month 6. Implementation feasibility is assessed through device technical success and intervention adherence (defined as ≥50% completion of weekly weigh-ins and daily dietary logs). ResultsParticipant data collection began in May 2023 and was completed by June 2025 with a total of 204 participants. The publication of key findings and results is anticipated in late 2026. ConclusionsThis protocol describes a pragmatic, technology-supported intervention designed to address metabolic side effects in a tertiary psychiatric setting. By bridging the gap between acute hospitalization and community recovery, this hybrid stepped-wedge cluster randomized trial provides a crucial framework for integrating digital metabolic monitoring into routine clinical workflows for vulnerable populations.

The diagnosis of bipolar disorder (BD) often faces delays in Chinese patients, affecting treatment outcomes. This study examines the factors associated with the interval from first psychiatric consultation to BD diagnosis and its association with 9-month depressive and manic symptom change. Data from a multicenter study in China was analyzed, involving 520 BD patients (399 type I BD [BD-I] and 121 type II BD [BD-II]) from seven medical institutions. We assessed sociodemographic and clinical features contributing to delayed BD diagnosis and the consequences of long DUBD on manic and depressive symptoms over a 9-month follow-up period. The median DUBD was 0.78 years; 0.51 years for BD-I and 1.59 years for BD-II. Younger age at first psychiatric consultation and older age at the first visit for the BD event were associated with long DUBD in both BD-I and BD-II. In BD-I, higher education level and manic/hypomanic/mixed polarity of first episode were associated with lower odds of long DUBD. Over the 9-month follow-up, longer DUBD was associated with less improvement in depressive symptoms among patients with BD-II. DUBD and its associated factors differed between BD subtypes. Longer DUBD was associated with less improvement in depressive symptoms among patients with BD-II over 9 months.

Computerized, self-adaptive cognitive training has emerged as a promising digital therapeutic for cognitive impairment; however, its underlying training mechanisms in schizophrenia remain poorly understood. This double-blind, randomized trial compared self-adaptive difficulty (SAD) with fixed-difficulty (FD) computerized cognitive training program, evaluating both domain-specific cognitive and clinical outcomes, and exploring difficulty-performance coupling as a potential active therapeutic mechanism. Seventy-two patients with schizophrenia (SAD = 41, FD = 31) completed 4 weeks of training, and 40 patients (n = 20 per group) were assessed at week 8. Cognitive outcomes were measured using the Cognitive Index (CI) and MATRICS Consensus Cognitive Battery (MCCB), while clinical symptoms were also assessed. Linear mixed-effects models (LMMs) revealed significantly greater and faster cognitive improvements in the SAD compared with the FD group in attention and agility, with faster improvement also observed in thinking. Robust difficulty-performance coupling was observed across all cognitive domains. Additionally, positive symptoms improved significantly in the SAD group compared to the FD group at week 4. Self-adaptive cognitive training enhances and accelerates cognitive improvements in schizophrenia, particularly in attention and processing speed, and also benefits positive symptoms. Difficulty-performance coupling may reflect dynamic load calibration as a key therapeutic mechanism in digital cognitive training. Trial registration: Chictr.org.cn, number, ChiCTR2000040326, registered 2020-11-27.

BACKGROUND:Bipolar depression (BD-D) is associated with high morbidity and suicide risk, yet effective non-pharmacological interventions remain limited. Low-field magnetic stimulation (LFMS), an ultra-low intensity whole-brain neuromodulation, offers potential for home-based therapy. This first randomized controlled trial (RCT) evaluated repeated home-based LFMS efficacy and safety in BD-D. METHODS:Sixty patients with ICD-10-defined bipolar depression (current depressive episode) were randomized to adjunctive active LFMS or sham stimulation (1:1), delivered twice daily for 2 weeks (28 sessions) using a chip-card blinded device. All participants maintained stable lithium. Primary outcome was change in clinician-rated Hamilton Depression Rating Scale-17 (HAMD-17) from baseline to Week 6. Secondary outcomes included self-reported Quick Inventory of Depressive Symptomatology (QIDS-SR16), and safety (affective switches). RESULTS:In the full analysis set (FAS, n = 51), HAMD-17 improvements did not differ between LFMS and sham at any timepoint (Week 6: Δ - 0.27, 95% CI: -1.93 to 1.39; P = 0.7427). However, LFMS significantly improved self-reported depression (QIDS-SR16) at Week 4 (Δ1.79, 95% CI: 0.16-3.43; P = 0.0319) and Week 6 (Δ1.83, 95% CI: 0.30-3.36; P = 0.0200). Baseline anxiety (GAD-7) was lower in the LFMS group (P < 0.05), though adjusted in analyses. Mood switch events occurred in both groups (one each). Safety analysis revealed no severe treatment-related adverse events. CONCLUSIONS:Although home-based LFMS failed to outperform sham on clinician-rated depression, it proved highly feasible and safe while suggesting possible subjective symptom benefits. This divergence emphasizes the need to include subjective endpoints in neuromodulation trials. Future studies should optimize LFMS parameters and target subjective symptom domains in BD-D. TRIAL REGISTRATION:This study was registered at a Chinese drug clinical trial institution under the registration number ChiCTRUE-INR-17013338, Date: 2017-11-10.

Abstract The subgenual anterior cingulate cortex (sgACC) is a key node in treatment-resistant depression (TRD), but precise non-invasive neuromodulation of this target is challenging. Preclinical studies of non-ablative stereotactic radiosurgery (SRS) have shown neuromodulatory (“radiomodulation”) effects. In this single-center, double-masked, randomized, dose-seeking pilot trial, nine adults with TRD were randomly assigned to bilateral sgACC radiomodulation at a dose of either 15, 20, or 25 Gy per hemispheric target. Primary endpoints were safety and feasibility; the efficacy endpoint was week-4 change in the Montgomery–Åsberg Depression Rating Scale (MADRS). Both primary endpoints were met: the only treatment-related adverse event was transient grade 1 dizziness, with no structural MRI abnormality through week 12. Mean MADRS fell from 33.0 to 17.0 (48.5% reduction); 67% responded and 44% remitted, with benefit sustained to week 12. Resting-state fMRI revealed regional connectivity changes correlating with clinical improvement, with tractography showing streamline counts differing by response status. These first-in-human findings support a larger randomized controlled trial of sgACC radiomodulation for TRD. ClinicalTrial.gov registration: NCT07274917 .

Background:Mental disorders represent a growing public health priority in China. Existing studies have largely relied on superseded data iterations and focused on specific population subsets or disease categories, leaving a gap in understanding the full epidemiological landscape of mental health across the country. This study provides the most up-to-date and comprehensive assessment of the burden of mental disorders in China and its provinces, utilizing the latest estimates from the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2023. Methods:Following the GBD 2023 methodology, we systematically analyzed epidemiological and demographic data to assess the prevalence, disability-adjusted life year (DALY), age-standardized prevalence rate (ASPR) and age-standardized DALY rate (ASDR) for mental disorders and ten subtypes in China and its 33 provincial-level administrative units from 1990 to 2023. Findings:In 2023, the estimated ASPR of mental disorders in China was 12,706 (95% uncertainty interval: 11,438, 14,573) per 100,000 population, reflecting a 14.6% (3.5, 30.7) increase since 1990. The ASDR reached 1718 (1272, 2307) per 100,000, rising by 18.4% (5.3, 36.9) over the past three decades. Depressive and anxiety disorders remained the dominant contributors, accounting for 35.6% and 26.3% of overall DALYs attributable to mental disorders, respectively. Furthermore, eating disorders exhibited the largest growth in ASPR (63.4% [58.2, 68.2]). Substantial subnational variations were observed, with the highest ASPRs in Hong Kong, Gansu, Macao, and Shandong, and the largest percentage increases in Macao, Hubei, Hong Kong, Xizang, and Ningxia. Females experienced a higher ASPR from depressive disorders, while males bore a heavier burden of autism spectrum disorders, attention-deficit/hyperactivity disorder, and conduct disorder. Across the lifespan, the burden of DALYs was relatively higher in the 5-19, 30-39, and 50-59 age groups. Interpretation:China has witnessed a substantial escalation in the burden of mental disorders over the past three decades. This complex epidemiological landscape, characterized by distinct regional, sex, and age-specific vulnerabilities, underscores the necessity for structural balancing and targeted life-course interventions. These up-to-date estimates provide a valuable evidence base for policymakers to further optimize resource allocation and implement tailored life-course interventions within the "Healthy China 2030" framework. Funding:Sci-Tech Innovation 2030-Major Project of Brain Science and Brain-Inspired Intelligence Technology (No. 2021ZD0200600); 2025 China Association for Science and Technology (CAST) Young Science and Technology Talents Cultivation Project-Special Program for Doctoral Students.

Despite the crucial distinction between insomnia symptoms and a diagnosed disorder, population-level studies based on contemporary criteria and clinical interviews are scarce. This study therefore examined the insomnia spectrum by assessing the prevalence, identifying subtypes, and exploring associations with sociodemographic factors and comorbid mental disorders for both conditions. This large-scale, community-based cross-sectional study was conducted in Beijing from October to December 2021. A sample of 10,778 adults was recruited via multistage stratified random sampling. Trained psychiatrists conducted standardized diagnostic interviews based on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) to collect data on insomnia disorder and mental disorders. Descriptive analysis and weighted Rao-Scott chi-square tests, were performed using the Statistical Analysis System (SAS, version 9.4). The weighted prevalence rates were 17.7

Objectives Antipsychotic-induced hyperprolactinemia is a relevant clinical issue. In this study, we aimed to evaluate the comparative efficacy of adjunctive aripiprazole and metformin in treating antipsychotic-induced hyperprolactinemia among female patients with schizophrenia. Methods A chart review of females diagnosed with schizophrenia and hospitalized from 2010 to 2020, all with antipsychotic-induced hyperprolactinemia and elevated serum prolactin levels. Data included antipsychotic types, baseline and post-intervention prolactin levels. Remission was defined as prolactin levels below 25 ng/dL in females. Cox regression and instrumental variables were used to assess remission hazard ratios at 30, 60, and 180 days. Results Among 652 female inpatients (mean age = 38.8 ± 12.7 years; 53.2 % on haloperidol, 17.8 % on risperidone) with hyperprolactinemia (mean baseline prolactin: 69.9 ± 47.8 ng/dL), 366 (56.1 %) received add-on aripiprazole (mean baseline prolactin: 76.5 ± 51.3 ng/dL) and 286 (43.9 %) received metformin (mean baseline prolactin: 61.5 ± 41.6 ng/dL). Aripiprazole was associated with decreased prolactin levels on the 30th day with a remission rate of 73.6 % compared to a 15.0 % remission rate in the metformin group. The effect was significantly greater in the low-dose group (aripiprazole ≤5 mg). Throughout the 180-day follow-up period, the final remission rate was substantially higher in the aripiprazole group than in the metformin group (77.6 % vs 23.1 %). Aripiprazole outperformed metformin in treating hyperprolactinemia induced primarily by haloperidol (remission rate 79.9 %), quetiapine (72.7 %), olanzapine (68.8 %) and risperidone (65.2 %) (all p < 0.01). Conclusions This real-world study suggests that adjunctive aripiprazole therapy effectively reduces prolactin levels in females with antipsychotic-induced hyperprolactinemia. Maximum efficacy is achieved at no >5 mg/day and within 30 days.

We examined the factors influencing various subtypes of subjective cognitive change in patients who shared similar objective cognitive trajectories within 6 months. We used data from an observational, prospective, cohort study, including 598 patients with major depressive disorder (MDD) in latent class mixed models based on the digit symbol substitution test performance. Participants were stratified into four distinct objective cognitive layers: “low cognitive performance,” “lower-middle cognitive performance,” “upper-middle cognitive performance,” and “high cognitive performance.” Within each of the four layers, the trajectories of subjective cognitive complaints were identified. Multinomial regression was employed, with cognitive complaint trajectories as the outcome, and depressive symptoms, clinical features, and other covariates as predictors. The factors influencing the subjective trajectories varied among the different objective layers. Patients with comorbid anxiety disorders or functional syndromes had more prominent self-reported cognitive symptoms and a slower rate of improvement. Younger age and lower education level were also influential factors for delayed remission of subjective cognitive function. Disease severity and antidepressant type did not contribute to dedifferentiating subjective cognitive trajectory subtypes within different subjective cognitive trajectories. Despite similar objective cognitive trajectories, subjective perceptions of these cognitive changes are heterogeneous. These findings deepen our understanding of the multifaceted nature of cognitive change in individuals with MDD and underscore the importance of considering a range of factors when interpreting and treating cognitive impairment at an early stage.

Mania and depression are the predominant mood episodes in bipolar disorder (BD), and their frequency significantly affects the long-term prognosis of patients. This is a multicenter, longitudinal cohort study in China. Sociodemographic and clinical characteristics of patients were statistically analyzed. Poisson regression analyses were performed to identify factors associated with the frequency of manic and depressive episodes. A total of 520 BD patients were enrolled in this study. Poisson regression model analysis showed that shorter years of education (OR = 1.03, P = 0.03), mixed polarity of the first episode compared to mania (OR = 2.33, P < 0.01) or depression (OR = 1.79, P = 0.01), earlier age at diagnosis (OR = 1.03, P = 0.01), comorbid substance use disorder (OR = 1.41, P = 0.02), presence of psychotic symptoms (OR = 1.18, P = 0.04), use of antidepressant medication (OR = 1.52, P = 0.01), and non-use of mood stabilizers (OR = 1.57, P<0.01) are positively associated with the frequency of manic episodes. Being male (OR = 1.22, P = 0.01), the use of mood stabilizers (OR = 1.47, P<0.01) and a diagnosis of bipolar II disorder (BD-II) compared to bipolar I disorder (BD-I) (OR = 1.27, P = 0.01) are positively associated with the frequency of depressive episodes. The study highlights the critical association of clinical and sociodemographic factors with the frequency of manic and depressive episodes in BD patients. Addressing these factors may improve long-term outcomes for individuals with bipolar disorder.

Background:Depressive episodes in adolescents and young adults are a significant global health concern, marked by high prevalence, cognitive impairments, and elevated suicide risk. Despite their clinical importance, remission trajectories and cognitive function in hospitalized youth remain understudied, particularly in transdiagnostic contexts. Methods:This retrospective cohort study analyzed electronic health records from 792 hospitalized patients (aged 13-22) with depressive episodes, using the Hamilton Depression Rating Scale (HDRS-17) and the Primary Cognitive Ability Test (PCAT III) to assess symptom trajectories and cognitive function. Gaussian Mixture Models identified distinct remission patterns, while linear mixed-effects models evaluated associations between depression severity, cognitive domains, and clinical factors. Results:Three trajectory groups emerged: Severe-Rapid Remission (7.7%), Moderate-Rapid Remission (15.3%), and Moderate-Slow Remission (77.0%). Working memory was related to depression severity, and anxiety symptoms were associated with cognitive performance. Additionally, patients diagnosed with bipolar depression showed reduced performance in both language comprehension and working memory at baseline. Intensive treatments (e.g., electroconvulsive therapy) showed efficacy but highlighted variability in response. Conclusion:The findings suggest that tailored interventions addressing baseline severity, anxiety, and cognitive support may be beneficial in hospitalized youth, with possible diagnostic relevance for bipolar depression.