Distinguishing between bipolar and unipolar disorder is essential for effective treatment, yet accurate diagnosis remains challenging despite extensive research. The MDQ (Mood Disorder Questionnaire) and BSDS (Bipolar Spectrum Diagnostic Scale) are widely used self-assessment tools, each offering unique advantages. However, these tools are typically used based on total scores, potentially overlooking valuable information within individual items.

This study aims to employ clustering analysis on the MDQ and BSDS, utilizing subscales derived from factor analysis, to better differentiate patients with bipolar and unipolar disorders.

The study included patients diagnosed with bipolar and bipolar depression, with diagnoses confirmed by a psychiatrist according to DSM-IV-TR criteria. A total of 299 patients with bipolar depression and 142 with unipolar depression completed the MDQ and BSDS. Based on prior factor analysis, the MDQ was divided into two subscales: the positive activation subscale (items 3, 4, 8, 9) and the negative activation subscale (items 1, 2, 6, 7, 12, 13). K-means clustering was performed twice: once using the total scores from the MDQ and BSDS (two scores), and using the positive activation subscale, negative activation subscale from the MDQ, and the total score from the BSDS (three scores). The analysis was iterated 1000 times to avoid overfitting.

The analysis identified an optimal solution with K=2. Cluster 1, characterized by high scores on both questionnaires, predominantly comprised bipolar patients. In contrast, Cluster 2, with lower scores, was primarily composed of unipolar patients. Using the total scores from both the MDQ and BSDS for clustering yielded an accuracy of 67.88%. In the second analysis using the MDQ subscales and the BSDS total score, the accuracy improved to 77.55%.

Clustering based on the MDQ and BSDS achieved a 77.55% accuracy in distinguishing bipolarity when using MDQ subscales alongside the BSDS score, demonstrating a promising level of precision with self-report questionnaires. Importantly, segmenting the MDQ into positive and negative activation subscales resulted in a nearly 10% increase in accuracy compared to using total scores alone. This suggests that increasing the dimensionality of the data by incorporating disorder-specific subscales can improve clustering accuracy. These findings highlight the potential of using high-dimensional psychiatric data to develop more effective classification models.

None Declared

From March 2020 to 2021, the risk of youth developing a mental health issue increased by 50% in Canada. To address the detrimental effects of the COVID-19 pandemic, this project collaborated with youth and community partners in Ontario, Canada, to co-design a Preventative Online Mental Health Program for Youth (POMHPY) focused on improving mental, physical, and social well-being.

1. (1) To co-design a preventative online mental health program tailored to the needs of Ontario youth. (2) To evaluate the program’s efficacy in improving mental well-being and health-related quality of life. (3) To engage youth in the development and continuous improvement of the program.

Initially, literature reviews were conducted to identify evidence-based programs that could be integrated into POMHPY. Surveys and focus groups were used to capture youths’ mental health concerns and program needs. The findings were presented to community partners for additional feedback and refinement of the program. A second survey and focus group explored the likelihood of program use and piloted the first session. Subsequently, 53 youths (mean age=19.15) participated in the POMHPY program during the summer of 2023. Pre-, post-, and follow-up surveys measuring mental well-being were administered. Preliminary descriptive statistics and t-test analysis were conducted to measure the program’s efficacy. A subset of participants (n = 21) attended 90-minute focus groups to discuss program perceptions, perceived benefits, impact on personal life, and areas of improvement.

Youths’ mental well-being, measured by the Warwick-Edinburgh Mental Well-being Scale, significantly improved after the completion of the program [t (24) =-2.91, p=.008]. Health-related quality of life, measured by the AqoL-6D, also significantly improved [t (6) =-3.34, p=.016]. These improvements were maintained one month after completing the program. Participants viewed the skills and strategies learned in POMHPY as beneficial in improving their stress and well-being. Peer facilitators in the same age range as participants contributed to meaningful discussions and interactive activities that contrasted with a lecture-style learning environment. Suggestions for improvement included flexible scheduling, increasing reminders, and enhancing understanding of program components.

Preliminary analysis supports the program’s efficacy in improving mental well-being and health-related quality of life. Participants also reported a positive experience with the program and suggested improvements for integration. The program will be scaled nationally in the next phase, ensuring broader access to preventative mental health care for youth across Canada.

None Declared

Older adults with treatment-resistant depression (TRD) can be treated with augmentation or switched to a different drug.

We aimed to identify factors that moderate the effectiveness of these strategies on treatment outcomes to guide the selection of the optimal strategy for each patient.

We analyzed data from 742 older adults with TRD in the Outcomes of Treatment-Resistant Depression in Older Adults (OPTIMUM) clinical trial. All participants were randomized to one of two treatment strategies, which were augmentation with aripiprazole, bupropion, or lithium; or switching to bupropion or nortriptyline. Treatment outcomes were change in MADRS scores and remission after 10 weeks. Age, burden of comorbid physical illness, number of adequate previous antidepressant trials, presence of executive cognitive impairment, and clinically relevant comorbid anxiety were examined as potential moderators of the effect of the two treatment strategies (augmentation vs. switching) on treatment outcomes.

Overall, augmentation produced more improvement in MADRS scores and produced a higher rate of remission than switching. For change in MADRS scores after 10 weeks of treatment, the number of adequate previous antidepressant trials was the only significant moderator of the superiority of augmentation over switching (b = -1.6, t = -2.1, p = 0.033, 95%CI [-3.0,-0.1]). There were no significant moderators for remission.

Older patients with TRD with less than three previous antidepressant trials benefit more from augmentation than from switching. Future studies validating this finding with different drugs in more diverse samples can facilitate their application in real world settings.

H. Kim Grant / Research support from: Dr. Kim reports grant support from the PSI foundation (R23-21). She is supported by the Canadian Institutes of Health Research (CIHR) and the Temerty Faculty of Medicine (Chisholm Memorial Fellowship)., J. Karp: None Declared, H. Lavretsky Grant / Research support from: Dr. Lavretsky received support from grants (K24 AT009198, R01 AT008383, and R01 MH114981) from the NIH., D. Blumberger Grant / Research support from: Dr. Blumberger reports grants from Canadian Institutes of Health Research (CIHR) and the Temerty family through the Centre for Addiction and Mental Health (CAMH) Foundation during the conduct of the study; nonfinancial support from Magventure (in-kind equipment support for investigator-initiated research); grants from Brainsway (principal investigator of an investigator-initiated study and site principal investigator for sponsored clinical trials), National Institutes of Health (NIH), Brain Canada Foundation, Campbell Family Research Institute, and Patient-Centered Outcomes Research Institute outside the submitted work; received medication supplies for an investigator-initiated trial from Indivior; and has participated in advisory boards for Janssen and Welcony., P. Brown Grant / Research support from: Dr. Brown received additional support from the National Institute of Mental Health OPTIMUM NEURO grant (5R01MH114980)., A. Flint Grant / Research support from: Dr. Flint has received grant support from the US National Institutes of Health, the Patient-Centered Outcomes Research Institute, the Canadian Institutes of Health Research, Brain Canada, the Ontario Brain Institute, and Alzheimer’s Association., E. Lenard: None Declared, P. Miller: None Declared, C. Reynolds Shareolder of: Dr. Reynolds receives payment from the American Association of Geriatric Psychiatry as Editor-in-Chief of the American Journal of Geriatric Psychiatry and royalty income for intellectual property as co-inventor of the Pittsburgh Sleep Quality Index., S. Roose: None Declared, E. Lenze Grant / Research support from: Dr. Lenze received additional support from the Taylor Family Institute for Innovative Psychiatric Research at Washington University School of Medicine, as well as the Washington University Institute of Clinical and Translational Sciences grant (UL1TR002345) from the National Center for Advancing Translational Sciences of the National Institutes of Health (NIH)., B. Mulsant Grant / Research support from: Dr. Mulsant received additional support from the Labatt Family Chair in Biology of Depression in Late-Life Adults at the University of Toronto. He holds and receives support from the Labatt Family Chair in Biology of Depression in Late-Life Adults at the University of Toronto. He currently receives or has received during the past three years research support from Brain Canada, the CAMH Foundation, the Canadian Institutes of Health Research, and the US National Institutes of Health (NIH); Capital Solution Design LLC (software used in a study funded by CAMH Foundation), and HAPPYneuron (software used in a study funded by Brain Canada).

Recently, CBT-based digital therapy has been developed and used for the treatment of various psychiatric disorders, including insomnia, depression, anxiety and panic disorders, and alcohol/drug addiction. In the United States, the first game-based digital therapy for ADHD has also received FDA approval and is being used for the treatment of children and adolescents with ADHD.

We conducted a randomized controlled study to examine the effectiveness of a digital therapeutic (model named ‘ADAM-101’) for children with ADHD in Korea, which was developed by Dragonfly GF Co., Ltd.”

Participants are 18 children with ADHD, aged 7 to 13 years, who are visiting the Department of Child and Adolescent Psychiatry at Seoul National University Children’s Hospital in Seoul, Korea. ADHD children with an IQ of 70 or above, who are currently taking stimulants and do not have other pediatric psychiatric disorders such as depression, anxiety disorders, tic disorders, ASD, were included in the study. They were randomly assigned to either the combined treatment group (medication + digital therapy, n = 9) or the medication-only group (n = 9). The digital therapy program was conducted using a tablet PC for 25 minutes a day, 5 days a week, for 4 weeks. Before starting the study, permission was obtained from the Institutional Review Board of Seoul National University Hospital. As a primary outcome measure, the Korean version of the Continuous Performance Test (KAT) was administered individually to the ADHD children by child clinical psychologists to assess inattention, impulsivity, and processing speed, after obtaining written agreement to participate in the study. Additionally, the Korean version f the ADHD Rating Scale-5 (K-ARS-5) was administered to the parents of the ADHD children.

We have not yet completed the study. Currently, out of the 18 ADHD children, 8 have completed the training and both pre- and post-assessments. All training and evaluations are expected to be completed by early October, and an analysis to verify the effectiveness of the digital therapeutic will be conducted in mid-October. Since this was not a double-blind study, we observed that, based on some children’s CPT and K-ARS-5 results, children in the combined treatment group tended to show a reduction in omission and commission errors on the CPT compared to those in the medication-only group. Additionally, there was a trend towards a reduction in inattention and hyperactivity-impulsivity scores on the K-ARS-5 in the combined treatment group.

Despite being conducted with a small sample, these results suggest the potential efficacy of the digital therapeutic (model named ‘ADAM-101’) for Korean ADHD children, indicating its potential clinical usefulness as an adjunctive treatment tool for ADHD children

None Declared

We present the case of a 61-year-old retired woman with hypothyroidism and rheumatoid arthritis who was diagnosed with bipolar disorder in 2006 after a manic episode. Her initial treatment included venlafaxine, valproic acid, quetiapine, zolpidem, and lormetazepam. She had several manic episodes over the years, some requiring hospitalization. In 2017, venlafaxine was replaced with vortioxetine, which she now introduces when detecting depressive phases, under psychiatric supervision but with some autonomy. She has remained stable, with occasional manic or hypomanic episodes triggered by stress, but none requiring hospitalization.

Figures 1 and 2 illustrate the patient’s self-perception changes with vortioxetine treatment. Figure 1 shows her as unhappy during the depressive phase (top) and happy after recovery (bottom). Figure 2 depicts her self-image during the depressive phase (left) and after recovery (right).

• - To assess the effectiveness, safety and risk of mood swings of vortioxetine in a patient with bipolar disorder during depressive phases.

• - To determine if effective psychoeducation allows patients to manage some of their medications safely under specialist supervision.

The patient mantains treatment with vortioxetine 20 mg daily, valproic acid 500 mg every 12 hours, quetiapine extended release 400 mg at dinner, zolpidem 15 mg, and lormetazepam 2 mg at bedtime during depressive phases. In manic episodes, quetiapine 300 mg is added, vortioxetine is discontinued, and the dosage of hypnotics is doubled. Intensive psychological support has improved her disease awareness and treatment adherence, allowing her to adjust vortioxetine (but not other medications) as needed under medical supervision. She has not reported any adverse effects from vortioxetine.

Vortioxetine 20 mg could effectively treat depressive phases without causing mania. Though research on its use in bipolar disorder is limited, it shows potential when combined with a mood stabilizer, with a response time of about nine weeks and a low risk of hypomania/mania. It may also help with cognitive decline and brain inflammation related to bipolar disorder, improving cognitive performance and reducing inflammatory markers. Vortioxetine is noted for its effectiveness, tolerability, and low dropout rate.

Image 1:

Image 2:

Vortioxetine 20 mg may be effective for treating depressive phases in bipolar disorder with a lower risk of manic episodes compared to other antidepressants. Its procognitive and potentially anti-inflammatory effects could also support stability in non-psychiatric comorbidities. For this patient, good psychoeducation has facilitated a degree of independence in managing the medication, which is aided by vortioxetine’s safety and ease of use for both professionals and patients.

None Declared

Suicide is a major public health concern, especially among older adults. Early identification of individuals at risk of suicide is crucial for early intervention, which significantly improves prevention efforts. Early identification of individuals at risk of suicide is crucial for prevention.

This study aimed to develop a model for predicting suicidal ideation in community-based older adults using deep phenotype data with machine learning classifiers.

A study investigating suicidal ideation in community-based older adults utilized a mobile assessment bus to collect data from 358 participants. Deep phenotype data, including Patient Health Questionnaire-9 (PHQ), Generalized Anxiety Disorder-7 (GAD), World Health Organization Quality of Life (WHOQOL), Perceived Stress Scale-10 (PSS) questionnaires, and 32-channel EEG recordings using the 10/20 system, were acquired. Of these participants, 238 completed all assessments. Suicidal ideation was defined by a score of 1 or higher on the ninth question of the PHQ-9. Data from both groups were compared, and features with an effect size of 1 or greater (Cohen’s D) were selected for further analysis. Cohen’s D. Machine-learning classifiers, including Support Vector Machine (SVM), Random Forest (RF), and Linear Discriminant Analysis (LDA) were employed to predict suicidal ideation using a 7:3 training-test split repeated 100 times to obtain performance metrics.

Scores on the PHQ, GAD, and WHOQOL scales differed significantly, while the PSS data showed variations in all items except one between the group with suicidal ideation and the group without. Notably, analysis of the EEG data from eight brain regions identified disparities in 108 out of 248 features. Among all data, ten features with Cohen’s D values exceeding 1 were identified, primarily consisting of questions directly related to themes of negative emotions. Using these features, the classification model achieved an AUC of 0.8913, demonstrating strong predictive performance for suicidal ideation.

Our findings demonstrate the potential of deep phenotyping, even in community-based settings, to predict suicidal ideation in older adults. These insights can inform the development of suicide intervention systems. Additionally, refining predictive models to encompass broader mental health symptoms could solidify deep phenotyping as a crucial tool for early intervention in public healthcare.

None Declared

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by difficulties in social communication and interaction, as well as restrictive and repetitive stereotyped behaviors, with considerable variation in cognitive and adaptive functioning. Accordingly, behavioral symptoms observed in daily life are likely to vary across individuals.

The present study aimed to explore the heterogeneity in intellectual abilities and behavioral issues among individuals with ASD using Latent Profile Analysis (LPA). This study was conducted between 2020 and 2021, with approval from the Institutional Review Board (IRB) of SNUH.

A cross-sectional analysis was conducted on 66 children (ages 6-18) diagnosed with ASD. The following psychometric instruments were used: Autism Diagnostic Observation Schedule-2 (ADOS-2), Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV), Child Behavior Checklist 6-18 (CBCL 6-18), Social Responsiveness Scale (SRS).

As the 3-profile solution returned a significant BLRT value and model entropy was estimated to be 0.93, the LPA of the WISC-IV indices and ADOS-2 comparison scores revealed three profiles that were clinically meaningful and balanced in group sizes: Profile 1 (“Higher ASD score with intellectual disability”; ADOS = 6.25, VCI = 55.16, PRI = 52.34, WMI = 53.47, PSI = 50.31), Profile 2 (“Higher ASD score with borderline intelligence”; ADOS = 5.76, VCI = 78.59, PRI = 80.65, WMI = 80.82, PSI = 67.12), and Profile 3 (“Lower ASD score with above-average intelligence”; ADOS = 4.41, VCI = 105.53, PRI = 108.41, WMI = 106.35, PSI = 89.41). On the SRS subscales, Profile 3 showed significantly lower scores in Social Cognition, Social Communication, and Social Motivation compared to Profile 1 (F(2, 63) = 10.45, p <.001; F(2, 63) = 5.24, p < .01; F(2, 63) = 8.75, p < .001). Additionally, on the CBCL syndrome subscales, Profile 3 showed significantly lower problem behaviors in Withdrawal/Depression, Social Immaturity, and Attention Problems (F(2, 63) = 4.57, p <.05; F(2, 63) = 5.07, p < .01; F(2, 63) = 4.19, p < .01).

In the present study, Latent Profile Analysis (LPA) using IQ and ADOS scores identified three distinct profiles within children with ASD. The findings suggest that while high-functioning ASD has traditionally been defined by an IQ threshold of 70–75, further distinction between borderline and above-average intelligence ASD groups may be warranted. Furthermore, targeted interventions addressing negative emotions, such as depression, may be indicated for the ASD group with intellectual disability.

Y. K. Lim Financial Support for Research from:, Financial Support for Research from: This research was supported by the National Research Foundation (NRF) funded by the Korean Government (MSIT) (RS-2024-00397737), and co-funded by the National IT Industry Promotion Agency(NIPA), an agency under the MSIT and with the support of the Daegu Digital Innovation Promotion Agency (DIP), the organization under the Daegu Metropolitan Government., E. Chung: None Declared, B. N. Kim: None Declared

Comprehensive cognitive remediation improves cognitive and functional outcomes in people with serious mental illness, but the specific components required for effective programs are uncertain. The most common methods to improve cognition are facilitated computerized cognitive training with coaching and teaching cognitive self-management strategies. We compared these methods by dismantling the Thinking Skills for Work program, a comprehensive, validated cognitive remediation program that incorporates both strategies.

In a randomized controlled trial we assigned 203 unemployed people with serious mental illness in supported employment programs at two mental health agencies to receive either the full Thinking Skills for Work (TSW) program, which included computerized cognitive training (based on Cogpack software), or the program with cognitive self-management (CSM) but no computer training. Outcomes included employment, cognition, and mental health over 2 years. To benchmark outcomes, we also examined competitive work outcomes in a similar prior trial comparing the TSW program with supported employment only.

The TSW and CSM groups improved significantly on all outcomes, but there were no differences between the groups. Competitive work outcomes for both groups resembled those of the TSW program in a prior trial and were better than the supported employment-only group in that study, suggesting that participants in both groups benefited from cognitive remediation.

Providing facilitated computerized cognitive training improved neither employment nor cognitive outcomes beyond teaching cognitive self-management strategies in people receiving supported employment. Computerized cognitive training may not be necessary for cognitive remediation programs to improve cognitive and functional outcomes.