Irritable bowel syndrome (IBS) affects 9.3

Schizophrenia, bipolar depression and major depressive disorders collectively affect over 10 million people across the European Union and are associated with annual healthcare and societal costs of more than 600 billion euros. When diagnosed, patients are usually prescribed psychotropic medication. A substantial proportion of patients insufficiently responds to first- and second-line treatments, which is termed as treatment resistance. While third-line treatments are relatively effective in TR conditions, it remains unclear whether providing these treatments in an earlier phase provides benefits. INTENSIFY consists of three international, multicenter, controlled, randomized, open-label trials, with a treatment duration of 6 weeks. After failure of a first-line treatment in the current episode, 1254 participants will be recruited by at least 16 sites and randomized to either a third-line treatment (early intensified pharmacological treatment) or treatment as usual (standard second-line treatment). Adult participants (age ≥ 18yrs) with a diagnosis of schizophrenia, schizoaffective disorder or schizophreniform disorder, major depressive disorder, or bipolar depression are eligible after a treatment failure in the current episode due to lack of efficacy. The primary objective is to compare change in symptom severity after 6 weeks. The secondary objective is to compare the two treatment arms on changes of severity symptoms, cognitive function, quality of life, functional measures, digital mental health markers, side effects and concomitant medication. ANCOVA analyses will be performed. Should the current study show clinical evidence that the use of third-line treatment is efficacious and safe earlier to treat severe mental illness, the results could impact treatment guidelines in due course. In addition, the study will contribute to the identification of clinical, digital and blood-based markers that are associated with treatment response to early and intensified treatments. The trials are registered in www.clinicaltrials.gov under numbers NCT05603104 (merged study protocol; study center in Israel) and under the numbers for the split protocols NCT05958875 (SZ), NCT05973786 (BD) and NCT05973851 (MDD) for the EU and UK sites. The study is registered in the Australian and New Zealand Clinical Trials Registry under Registration number ACTRN12625000327482 (SZ), ACTRN12625000328471 (BD) and ACTRN12625000329460 (MDD). Registered on April 17, 2025.

Introduction: The identification of biomarkers for treatment response in major depression is critical to the further development of personalized treatment. There is a recognized relationship between facial expression and depression of mood, and previous literature also indicates that facial expression is associated with treatment outcomes in depression. This suggests that facial expression may have use as a biomarker for treatment response. There is no previous synthesis of related research to drive the development of new digital approaches. Methods: We conducted a systematic review using three databases (MEDLINE, Scopus, and PsycINFO), identifying English-language publications (journal articles or books) that assessed either facial muscle activity or expression as predictors of treatment response or correlates of treatment outcome in depression. Risk of bias was assessed using a National Institutes of Health quality assessment tool. Results: We identified 12 studies, involving a total of 389 participants, which used a variety of different assessment methods and thus assessment outcomes, including electromyography, observer-related assessments (including Facial Action Coding System), and automated tools of facial expression assessment. Depression treatment response correlated with an increase in facial expressivity. Greater activity in the corrugator and zygomatic muscles, and lower levels of lip tightening and downward lip movement, may predict treatment response. Conclusions: Included studies were limited by heterogeneity in facial expression assessment tools and outcomes, along with demographic homogeneity. The findings of the review suggest that facial expression analysis may offer an avenue for biomarkers of depression status and treatment response prediction.

Objective Electroencephalography (EEG) signals comprise both oscillatory (periodic) and non-oscillatory (aperiodic) components. Aperiodic activity forms the 1/f-like background of the EEG power spectrum and can be characterised by parameters describing the offset and slope (1/f exponent). This study examined the effects of two antiseizure medications (ASMs), which reduce cortical excitability through different mechanisms of action, on aperiodic and periodic EEG activity. Methods Resting EEG was recorded with eyes open and closed from 13 healthy male volunteers at baseline and two hours after administration of lamotrigine (300 mg), levetiracetam (3000 mg), or placebo. Power spectra were computed using Welch’s method. Aperiodic parameters were estimated using the specparam algorithm, and periodic activity was quantified after subtraction of the aperiodic component. Results In the eyes-open condition, lamotrigine significantly reduced both aperiodic offset and exponent relative to placebo, consistent with a flattening of the aperiodic spectrum, whereas levetiracetam reduced the exponent without significantly altering the offset. Neither drug significantly altered aperiodic exponent or offset during eyes-closed. Lamotrigine reduced corrected theta power in both conditions and alpha power during eyes-open, whereas levetiracetam increased beta power in both conditions and reduced gamma power during eyes-closed. Conclusions Both lamotrigine and levetiracetam altered aperiodic and periodic EEG activity, with aperiodic effects observed only during eyes-open. Significance Aperiodic EEG measures may provide a non-invasive approach for characterising ASM-related neurophysiological effects and may help elucidate how distinct pharmacological mechanisms influence large-scale cortical activity.

Psychiatric disorders are often comorbid with metabolic syndrome (MetS), suggesting shared genetic architecture. This systematic review investigated the genetic connections between psychiatric disorders, antipsychotic use, and MetS through genome-wide association studies and related genetic approaches. We systematically searched PubMed, Scopus, and Web of Science for relevant studies published between January 2005 and August 2025. Seventy-seven studies were included in the systematic review upon meeting the inclusion criteria. Substantial shared genetic architecture was found between psychiatric disorders and metabolic traits, particularly involving major depressive disorder (MDD), schizophrenia (SCZ), and body mass index (BMI). Key implicated pathways included lipid metabolism, glucose homeostasis, and inflammatory processes. Mendelian randomization studies provided evidence for causal relationships, notably a unidirectional effect of MDD on MetS, and bidirectional relationships between SCZ and BMI. Antipsychotic-induced weight gain showed a more specific genetic basis compared to broader psychiatric disorder-related MetS. Future research should focus on diverse populations, refined phenotype definitions, and translating genetic insights into clinical practice. The integration of pharmacogenomic scores with clinical data shows promise for personalizing treatment strategies in psychiatry.

Mobile health (mHealth) apps are increasingly deployed for evidence-based mental health interventions, broadening access to care. While effective, Internet-based Cognitive Behavioural Therapy, delivered via web or app, frequently overlooks ethnic minority and migrant populations. Effective cultural adaptation of mHealth apps is critical to their impact and accessibility; however, existing frameworks often lack specific guidance for digital contexts, relying on superficial adjustments or omitting the explicit integration of religious factors. Our novel framework fundamentally departs from prior models by embedding cultural responsiveness throughout the entire digital development lifecycle of mHealth apps, rather than treating it as a peripheral concern. This comprehensive approach is structured across four interconnected layers: adapting the therapeutic foundation (explicitly incorporating religious considerations), culturally grounding features and reframing standard tools, tailoring content and messaging, and optimizing UX/UI design. Central to this framework is a participatory co-design and prototyping methodology that ensures deep cultural insights, including religious and spiritual dimensions, are profoundly integrated from the earliest stages of development. This framework thus offers a practical, inclusive, and ethical blueprint for designing culturally relevant digital health products and can be applied and tested in future work through iterative co-design workshops, user-centred prototyping, and the development of culturally tailored digital interventions, with ongoing evaluation of their effectiveness and acceptability in diverse populations.

Background The Global Bipolar Cohort (GBC) was established to identify existing bipolar disorder (BD) cohorts worldwide and foster collaborations focused on descriptive and analytic outcomes relevant to BD. A distributed analytic framework has been implemented to engage multiple sites without the need for central data pooling. This report describes the GBC endeavor and global functional impairment patterns. Cross-cohort comparisons of functional correlates are limited by heterogeneous measures and data-sharing constraints. Large, culturally diverse comparisons are needed to distinguish broadly reproducible correlates from cohort-specific effects. Participating sites completed a 28-item descriptive survey covering diagnostic methods, cognition, genetics, treatment, functioning, and follow-up strategies. We implemented a harmonized local logistic regression model of dichotomized functional outcome and shared summary statistics only. Results We identified 69 cohorts across five continents. Thirty-seven cohorts contributed functional outcome analyses from 17,130 participants. Outcome measures included clinician-rated disability scales and social indicators such as employment and marital status. The proportion classified with poor functioning ranged from 16% to 77% (mean 50%). In 32 of 37 cohorts, the overall regression model significantly explained variance in functioning. Current depressive symptoms were the most robust and reproducible correlate of poor functional outcome: they were assessed in 29 cohorts, significant in 22 (75.8%), ranked among the top three correlates in 22, and were the top-ranked correlates in 19. Associations between depressive burden and poor functioning were observed across clinician-rated disability scales and work or social indicators, and across geographically diverse cohorts. Comorbid substance use disorder and medication-related variables were associated with poorer functioning in subsets of cohorts, whereas sex, ancestry, bipolar subtype, psychosis history, and premorbid IQ showed weak or inconsistent associations. Cognitive measures, available in a minority of regression models, showed modest and non-uniform effects. Conclusions Across heterogeneous international cohorts, current depressive symptom burden emerged as the most consistent correlate of poor functioning in bipolar disorder. These findings replicate earlier multisite work at a larger scale, show that protocol-based distributed analyses can identify reproducible clinical signals without sharing individual-level data, and support prioritizing detection and treatment of depressive symptoms when aiming to improve real-world functioning. Future work should expand longitudinal harmonization and representation of under-studied populations.

Adolescence marks a critical window for the emergence of mental health problems and reproductive maturation. Menstrual cycle characteristics - including pubertal timing, cycle regularity, menstrual pain, and premenstrual symptoms -may capture biological and experiential processes relevant to psychopathology. However, evidence linking menstrual features to a broad range of mental health outcomes, including ADHD symptoms, remains limited. Using data from the Adolescent Brain Cognitive Development (ABCD) Study (v6.1 N = 5,678 biological females; 17,567 post-menarche observations across 7 waves), we examined associations between reproductive timing, menstrual characteristics and symptoms of attention-deficit/hyperactivity disorder (ADHD), depression, and anxiety symptoms across repeated assessments in early adolescence using linear mixed-effects models with crossed random intercepts for site, family, and participant. Two co-primary model specifications were used: without BMI (M1, full sample) and with BMI (M2). False discovery rate correction was applied across outcomes. Earlier age at menarche was specifically associated with higher depression symptom scores (B = -0.087, pFDR < 0.001), with attenuation after BMI adjustment. Greater gynaecologic age (time since menarche TSM) also associated with higher depression scores (B = + 0.118, pFDR < 0.001) independent of chronological age and pubertal timing. Among girls with established cycles (TSM ≥ 2 years), irregular cycles were associated with higher depression and anxiety, whereas associations with ADHD did not remain after BMI adjustment. Long cycles were associated with anxiety only. Severe menstrual pain and premenstrual symptom severity showed dose-response associations with depression, anxiety, and ADHD that were robust to BMI adjustment. The association between irregular cycles and depression, and between long cycles and depression and anxiety, strengthened significantly with increasing gynaecologic age, while symptom burden associations were consistent throughout the post-menarcheal period. Menstrual symptom burden particularly severe pain and premenstrual distress and earlier menarche were consistently associated with higher levels of psychiatric symptoms in adolescent girls. These findings highlight menstrual health as a potentially informative dimension of adolescent mental health assessment. When girls present with significant menstrual symptoms, assessment of emotional wellbeing may be warranted.

Abstract Transcranial magnetic stimulation combined with electroencephalography (TMS-EEG) enables direct measurement of cortical reactivity via TMS-evoked potentials (TEPs). Interpretation of early TEP components however, is highly sensitive to stimulation and hardware-related artifacts. We identified and characterised a persistent, non-neural ‘step-drift’ artifact unexpectedly present in recent TMS-EEG recordings from our group. We show that the artifact is distinct from previously described TMS pulse and discharge/decay artifacts and likely reflects a hardware interaction phenomenon. We demonstrated that amplifier settings, but not TMS pulse shape, substantially influenced artifact expression, with DC-coupled recordings with no online high-pass filter reducing step amplitude compared with AC-coupled recordings with a high-pass filter. Simulations additionally revealed that filtering over the step-drift artifact introduced pronounced ringing and edge artifacts, highlighting the need to address this artifact prior to data processing. We propose a processing pipeline incorporating robust polynomial detrending and a modified Butterworth filter with autoregressive extrapolation that minimised TEP distortion in both simulated and real data containing the step-drift artifact. Together, these findings provide practical recommendations for both preventing and correcting step-drift artifacts and underscore the need for formal definition and routine recognition of this artifact to improve reproducibility and data quality in TMS-EEG research.

OBJECTIVE:Identifying those at highest risk for making a first suicide attempt during adolescence is crucial to inform early suicide prevention. Our study aimed to predict the first ideation-to-attempt transition during adolescence among children with suicidal ideation at baseline using 187 sociodemographic, clinical, neurocognitive, functional, and structural brain predictors. METHOD:Data were obtained from the multisite, longitudinal Adolescent Brain Cognitive Development℠ (ABCD) study, conducted in 21 US sites among 11,864 children 9 to 10 years of age at baseline, with 4 follow-up waves measured between 2018 and 2022. The primary outcome was suicide attempt reported at any of the follow-up waves among children with suicidal ideation at baseline. Machine learning models were trained using 70% of the sample from 14 sites, and were validated in participants from 7 holdout sites. RESULTS:The final sample included 660 children with suicidal ideation at baseline (no previous suicide attempt; mean age = 9.91 years, SD = 0.63 years; 42% female at baseline), of whom 83 children had a first suicide attempt within 4-year follow-up. The final model, which excluded the brain imaging feature as its inclusion did not improve performance, generalized well to the external holdout sites (area under the receiver operating characteristic curve [95% CI] = 0.75 [0.68, 0.83], sensitivity = 0.65 [0.61, 0.75], specificity = 0.69 [0.50, 0.80], positive predictive value = 0.23 [0.15, 0.34], negative predictive value = 0.94 [0.88, 0.97]), p <. 01) with good expected calibration error of 0.03. The model was unbiased across race and sex subgroups. The top contributing features included female sex, presence of self-harm, access to means, generalized anxiety disorder, social anxiety, impulsivity, severity of suicidal ideation, parental income, and clinical treatment history. CONCLUSION:Our model using clinically accessible features predicts the first-onset suicide attempt in children. Most predictors (eg, suicidal ideation severity, impulsivity, anxiety symptoms) are modifiable, highlighting the potential intervention targets. Findings provide longitudinal evidence for key risk factors for the ideation-to-attempt transition in current suicide theories.

Background:Culturally and religiously responsive mental mobile health (mHealth) apps may improve access to and acceptability of mental health support among migrant communities; however, evidence to inform their design remains limited. Objective:This formative study investigated mental health perceptions, digital health information-seeking, and mental mHealth app use among first-generation Arabic-speaking migrants in Australia, with the aim of informing culturally adapted mental mHealth app design. Methods:An online survey was conducted among 219 first-generation Arabic-speaking migrants in Australia (aged 18-75 years), recruited from non-clinical community settings. The survey assessed attitudes toward mental health, awareness and use of mental mHealth apps, acceptance of app-based support, and desired features. Open-ended questions provided qualitative insights into cultural and religious preferences. Results:Strong cultural and religious influences on mental health perceptions were observed, including high agreement regarding the role of divine will and religious practices. While most participants (76.3%) used the internet to seek mental health information, awareness (45.7%) and use (6.4%) of mental mHealth apps were low. Participants expressed high acceptance of mental mHealth apps that are free, user-friendly, confidential, and professionally developed. Highly valued features included culturally informed behavioural activation, mindfulness and religious practices (such as Dua'a and Tadabbur), and educational content incorporating Quranic verses and prophetic narratives. Information on crisis services and local multicultural mental health providers was also considered essential. Qualitative findings supported the inclusion of faith-based community features and religious motivational content, with several participants emphasising the importance of optional rather than mandatory religious elements. Conclusions:There is a clear demand for mental mHealth apps tailored to the cultural and religious needs of first-generation Arabic-speaking migrants in Australia. Formative evidence from this study highlights the importance of culturally and religiously congruent design, practical support features, confidentiality, and flexibility to accommodate individual preferences when developing mental mHealth interventions.

Background:The predictive power of polygenic scores (PGSs) for lithium treatment response in bipolar disorder (BD) remains limited. Aim:To enhance prediction of lithium responsiveness by developing a multi-trait PGS (mt-PGS) combining genetic information from multiple phenotypes implicated in lithium response and/or BD aetiology. Methods:We analysed data collected from BD patients who had received lithium treatment for at least six months and participated in the International Consortium on Lithium Genetics (ConLi+Gen, N=2,367) study. The ALDA scale was used to assess lithium responsiveness, and treatment outcome was defined as continuous ALDA score (0-10) and categorical outcome (favourable ≥7 vs unfavourable response). PGSs were calculated for 59 phenotypes grouped into five clinical-biological clusters: clinical lithium exemplar (#22 phenotypes), cardiometabolic (#17), autoimmune/inflammatory (#5), neurocognitive (#8) and renal function (#7). We applied cross-validated machine learning regression approaches in both outcomes within each cluster, and the selected features from each cluster were subsequently combined to construct the final mt-PGS models. Model performance was assessed using explained variance (R2) for the continuous outcome, and McFadden's pseudo-R2 as well as standard classification model parameters for the categorical outcomes. Results:The mt-PGS explained 5.07% (continuous outcome) to 9.02% (categorical outcome) of the interindividual variability in lithium responsiveness. Classification accuracy (AUC) for the categorical outcome was 68.13% (95% CI: 64.86, 71.77). Of the five clusters, the PGSs for clinical lithium exemplar phenotypes were most strongly associated with lithium responsiveness, accounting for 2.97%-6.20% of its variability. Conclusions:By integrating polygenic scores for multiple relevant phenotypes, predictive accuracy for lithium response improved up to nine-fold compared to single-trait methods. Future research incorporating larger, more diverse populations and combining genetic scores with clinical data holds promise for further enhancing prediction and advancing clinical implementation.

OBJECTIVES:Clozapine, the gold-standard antipsychotic for treatment-resistant schizophrenia causes severe metabolic complications, including metabolic syndrome and increased type 2 diabetes (T2D) risk. A better understanding of the genetic factors influencing clozapine pharmacokinetics and the associated metabolic risk could inform precision medicine approaches to clozapine prescribing. METHODS:Using a series of genetic-epidemiological approaches, we aimed to identify candidate biomarkers associated with clozapine-induced metabolic dysfunction. Mendelian randomisation (MR) was employed to investigate evidence of causal relationships between clozapine metabolism and cardiometabolic traits. RESULTS:Higher plasma clozapine and clozapine-norclozapine ratio were associated with a higher risk of T2D and blood pressure. The phenome-scan-colocalization-MR pipeline identified traits influenced by clozapine-metabolism loci that might serve as markers of cardiometabolic risk. This pipeline identified 28 colocalizing markers associated with clozapine pharmacokinetic loci. Subsequent MR highlighted associations for 16 of these 28 biomarker candidates with cardiometabolic outcomes, which included haematological markers and excretory traits. CONCLUSIONS:These findings are hypothesis-generating and do not, in the absence of prospective clinical validation, establish causal relationships between clozapine and the identified cardiometabolic traits. They may inform the development of biomarker-guided monitoring approaches for risk stratification and early intervention, enabling a shift from reactive monitoring to predictive approaches in managing clozapine-induced metabolic dysfunction with appropriate clinical validation.

Treatment-resistant depression is a condition with significant morbidity, despite the current standard of treatment, including traditional pharmacotherapy, psychotherapy, augmentation strategies and electroconvulsive therapy. While novel therapies have emerged, such as ketamine/esketamine, transcranial magnetic stimulation and psilocybin-assisted therapy, the uptake of these treatments is relatively low, particularly within public mental health services. Commercial clinics across Australia offer these treatments, but can only be accessed by those able to pay, leaving those unable to pay with limited or no access. We compare novel treatments for treatment-resistant depression and examine barriers to their implementation within the Australian mental health system. We also propose specialist treatment-resistant depression clinics as a potential model to improve access and build clinical expertise. We identified challenges in identifying treatment-resistant depression, lack of training and expertise, and regulatory and economic barriers. We propose that the lack of public access to these novel treatments be initially addressed by the establishment of specialist treatment-resistant depression clinics in Australia, including in the public sector, which will drive training and the development of expertise within public mental health.

Modern research management, particularly for publicly funded studies, assumes a data governance model in which grantees are considered stewards rather than owners of important data sets. Thus, there is an expectation that collected data are shared as widely as possible with the general research community. This presents problems in complex studies that involve sensitive health information. The latter requires balancing participant privacy with the needs of the research community. Here, we report on the data operation ecosystem crafted for the Accelerating Medicines Partnership® Schizophrenia project, an international observational study of young individuals at clinical high risk for developing a psychotic disorder. We review data capture systems, data dictionaries, organization principles, data flow, security, quality control protocols, data visualization, monitoring, and dissemination through the NIMH Data Archive platform. We focus on the interconnectedness of these steps, where our goal is to design a seamless data flow and an alignment with the FAIR (Findability, Accessibility, Interoperability, and Reusability) principles while balancing local regulatory and ethical considerations. This process-oriented approach leverages automated pipelines for data flow to enhance data quality, speed, and collaboration, underscoring the project’s contribution to advancing research practices involving multisite studies of sensitive mental health conditions. An important feature is the data’s close-to-real-time quality assessment (QA) and quality control (QC). The focus on close-to-real-time QA/QC makes it possible for a subject to redo a testing session, as well as facilitate course corrections to prevent repeating errors in future data acquisition. Watch Dr. Sylvain Bouix discuss his work and this article: https://vimeo.com/1025555648 .

AIMS:The value of assessing basic symptoms in clinical-high-risk for psychosis (CHR) is becoming increasingly apparent. Greater recognition of subjective experience in neuroscience and psychiatry has renewed research interest in electrophysiological biomarkers of basic symptoms. This study aims to investigate whether cognitive basic symptoms (COGDIS), which capture a subset of basic symptoms, are associated with P3b attenuation and the modulation of brain connectivity in a large sample of CHR. METHODS:Data from the North American Prodromal Longitudinal Study- 3 (NAPLS-3) comprised 440 male and female CHR individuals who completed both the COGDIS items of the schizophrenia proneness instrument as well as a two-tone auditory oddball task. P3b amplitude was measured at the central (Cz) as well as left (P3) and right (P4) parietal electrodes. Brain connectivity was calculated across 300 ms windows before (-300 ms to 0 ms) and after (100 ms to 400 ms) onset of target stimuli. Brain connectivity modulation was calculated as the difference between pre-stimulus and post-stimulus windows. RESULTS:Multiple linear regression analysis indicated that COGDIS was associated with reduced P3b amplitude at the P4 electrode. This effect was not associated with the severity of positive or negative symptoms. No differences in connectivity strength or modulation were found between the groups. CONCLUSIONS:In a large sample of CHR/UHR individuals, cognitive basic symptoms criteria was associated with reduced P3b amplitude at the P4 electrode, approximating the right temporo-parietal area. Parietal P3b attenuation may reflect greater preoccupation towards sensory data, which could play a role in cognitive basic symptom pathogenesis.

BACKGROUND: The purpose of this study was to better understand students’ experiences of readiness for their psychiatry clinical placement and eventual clinical practice in mental health units. METHODS: Interviews with fourth year undergraduate medical students were conducted and analysed using reflexive thematic analysis. Themes were identified that describe participant experiences of preparation and readiness for placements in mental health units. RESULTS: Eight participants provided detailed accounts of experiences prior to medical school, during medical school and during placements in mental health units that influenced their perceptions and attitudes towards people who experience mental illness and mental distress. Themes identified in this study included: Conceptual readiness (and unreadiness), Procedural readiness (and unreadiness) and, Dispositional readiness (and unreadiness). Findings indicate that students do not approach their mental health placements in the same way as they do for other placements. They fear many components of mental health environments, including their own capacity to engage with people without causing further harm. Students had varied attitudes and experiences that shaped their conceptual ideas relating to psychiatry and their procedural readiness to undertaken tasks such as communicating with, and assessment of people experiencing mental distress and mental illnesses. CONCLUSION: Medical students will encounter people who experience mental illness and mental distress regardless of which discipline they choose to specialise. We have an obligation to support students to feel conceptually, procedurally and dispositionally ready to learn and practice when they prepare for clinical placement in mental health units, and indeed for their future practice. TRIAL REGISTRATION: Not applicable.

OBJECTIVE:Suicide is one of the leading causes of death among youth worldwide, yet existing studies that aimed to predict the first onset of suicidal thoughts and behaviors (STB) included a limited number of data modalities and/or focused on adult populations. This study aimed to prospectively predict first-onset STB across 4-year follow-ups in adolescents using an existing STB history classification model that was previously applied to baseline data and a new machine learning model with 195 biopsychosocial features. METHOD:Participants were 7,503 unrelated adolescents (54.5% female, ages 9-11 years at baseline) from the multisite, longitudinal Adolescent Brain Cognitive Development (ABCD) Study. An existing baseline STB history classification model was applied to predict longitudinal first-onset STB in adolescents compared with healthy controls and clinical controls (individuals with a mental health disorder but no STB). A new elastic net logistic regression model with 195 features was trained on data from 14 sites (n = 5,220), and the resulting top 15 features were validated at 7 independent sites (n = 2,283). RESULTS:The previously developed model to classify STB lifetime history also prospectively predicted first-onset STB in adolescents with an area under the curve (AUC) [95% CI] of 0.73 [0.70, 0.75], p < .001, compared with healthy controls and AUC [95% CI] of 0.63 [0.60, 0.66], p < .001, compared with clinical controls. The newly trained model with top 15 features performed similarly with AUC [95% CI] of 0.73 [0.71, 0.76], p < .001, and AUC [95% CI] of 0.64 [0.60, 0.66], p < .001, for the same comparison groups. The most consistent predictors across models included female sex, sleep disturbances, and maladaptive home and school environments. CONCLUSION:The models predicted first-onset STB in adolescents with moderate accuracy. This study also confirmed the roles of well-established psychological risk factors for STB and identified several novel neurocognitive and brain imaging risk factors. Future studies should validate these models in large-scale diverse samples before clinical translation. PLAIN LANGUAGE SUMMARY:This study followed over 7,500 adolescents for 4 years and tested 2 machine learning models using psychological, social, and brain data to identify those at risk of experiencing suicidal thoughts or behaviors. Both models predicted first-time suicidal thoughts or behaviors with moderate accuracy. Key risk factors that were identified included being female, experiencing sleep problems, and negative home and school environments. DIVERSITY & INCLUSION STATEMENT:We worked to ensure sex and gender balance in the recruitment of human participants. We worked to ensure race, ethnic, and/or other types of diversity in the recruitment of human participants. We worked to ensure that the study questionnaires were prepared in an inclusive way. Diverse cell lines and/or genomic datasets were not available. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented sexual and/or gender groups in science. We actively worked to promote sex and gender balance in our author group. One or more of the authors of this paper received support from a program designed to increase minority representation in science. We actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our author group. While citing references scientifically relevant for this work, we also actively worked to promote sex and gender balance in our reference list. While citing references scientifically relevant for this work, we also actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our reference list. The author list of this paper includes contributors from the location and/or community where the research was conducted who participated in the data collection, design, analysis, and/or interpretation of the work.