Background Ageism, defined as stereotypes, prejudices and discrimination based on a person's age is prevalent and negatively impacts individuals and society. In 2016, the World Health Organization (WHO) received a mandate from its Member States to lead the global campaign to combat ageism. However, success in combating ageism is critically dependent on the availability of valid and reliable measures of people's experiences of ageism that can capture its multidimensional nature and as it is directed against multiple age groups across diverse country settings. The goal of current study is to present a preliminary psychometric validation of a measure specifically designed to meet this need: the 15-item WHO Ageism Experiences Scale. The scale includes items measuring self-directed, interpersonal, and institutional ageism and spans stereotypes, prejudices, and discrimination.Method We collected data from a UK mixed-age adult sample (N = 402), and several psychometric properties were tested, including internal consistency reliability, test-retest reliability, and convergent validity.Results Results suggested that the 15-item WHO Ageism Experiences Scale has good internal consistency (where relevant) and test-retest reliability. The exploratory factor analysis supported a one-factor model for self-directed ageism items. The scale also proved to have adequate convergent validity.Discussion The results provide sufficient support for the adoption of the scale in this population and motivates its translation and validation in other settings, as well as the evaluation of a more extended set of psychometric properties in future studies.

Background Previous studies have suggested associations between stress and mental health in adolescence; however, studies are needed to illuminate the developmental relations between cortisol and mental health in longitudinal data. This study combines hair and fingernail cortisol with ecological momentary assessment (EMA) and survey data to offer a new way of examining developmental relations between cumulative stress and mental health throughout adolescent development. The objective is to assess the relations between stress biomarkers and mental health outcomes and trajectories throughout adolescent development. Methods Participants will include approximately n = 500 adolescents from the Mental Health in the Moment (MHIM) project and attending school in Scotland in years S1-S5 (aged 11–16 years) at intake. Cortisol, cortisone and gonadal hormones will be measured via either hair or fingernail samples. Samples will be collected alongside a biosampling survey at a minimum of three time-points during the 5-year data collection period. As part of the MHIM project, participants will complete a survey on demographics, mental health, and lifestyle-factors, as well as daily surveys via EMA, which will be linked to the stress biomarker data. Analysis Data will be analysed with methods such as linear regression, latent growth curve models, random intercept cross-lagged panel models, and dynamic structural equation modelling to illuminate the predictors, outcomes, and developmental trajectories of stress biomarkers over adolescence, as well as their relations with mental health. Discussion Measuring stress biomarkers across the course of adolescence has the potential to provide developmental insights into cumulative stress and its links to mental health during this critical developmental period. Insights can inform intervention and mitigation strategies throughout adolescence. However, stress biomarker measurement carries a number of challenges, including potential difficulties with acceptability. Working closely with a young person’s advisory group and offering multiple participation options (hair and fingernails) is anticipated to help mitigate these challenges.

INTRODUCTION:There is growing recognition of the methodological benefits of co-producing adolescent mental health research. Co-production supports the development of study methodologies that are more relevant and appropriate for the examination of adolescents' mental health. Within this article, insights from the Mental Health in the Moment Young Person Advisory Group (MHIM-YPAG) are presented in relation to which mental health constructs, predictive factors and measures were endorsed by young people for inclusion in a mental health study. The aim of this paper is to provide guidance to researchers, highlight the benefits of co-production and provide a template for co-production in mental health study measures. METHODS:Members (n = 18) took part in online 'focus group style' sessions covering constructs, predictive factors and measures related to adolescent mental health. They were asked to discuss and then individually rank a set of constructs, predictive factors and measures from most to least relevant. Following this, a semantic thematic analysis of session materials was performed by researchers and findings were verified by MHIM-YPAG members. RESULTS:Members emphasised the importance of general and specific constructs that reflected both positive and negative traits. In addition, they emphasised the importance of examining suicidality/suicide-ideation. Social media, academic stress and relationships were the most relevant predictive factors for this group. Language use and response scale length were important when deciding measures appropriateness. CONCLUSION:This paper highlights the importance of involving young people as co-producers within adolescent mental health research. Benefits include improving the appropriateness and relevance of the study design. Insights may act as a guide for researchers involved in adolescent mental health research. PATIENT AND PUBLIC CONTRIBUTION:The 'MHIM-YPAG' was established to enable the involvement of young people to co-produce the MHIM project. Thus, this paper details the insights made by the MHIM-YPAG regarding the inclusion of specific mental health constructs, predictive factors and measures. As detailed in the MHIM-YPAG protocol paper, young people will be involved through the 5-year project in a co-production capacity.

Understanding the dynamics of suicidal ideation and behaviour in youth and the factors associated with transitions from thoughts to behaviours is critical for early identification, monitoring, and prevention. Using longitudinal self-report data from the Adolescent Brain Cognitive Development (ABCD) Study (n = 11,864) spanning ages 9 to 13 years, we developed a time-inhomogeneous discrete-time Markov chain framework to model transitions across eight states defined by suicidal ideation and behaviour and the co-reported presence or absence of non-suicidal self-injury (NSSI). The framework enables inference of year-to-year and multi-year transition probabilities, quantification of transition stability and uncertainty, and statistical comparison of transition likelihoods. We identified structured but developmentally changing transition patterns, including a generally high probability of recovery to a no-report state following reports of suicidal ideation or behaviour. Co-reported lifetime NSSI marked a distinct trajectory profile characterized by both greater risk and lower predictability: children with NSSI were more likely to transition to or persist in suicidal behaviour, were less likely to recover to a no-report state, and exhibited greater uncertainty in their transition likelihoods compared to those reporting suicidal ideation and/or behaviour alone. These findings suggest that NSSI marks not only elevated risk but also greater trajectory instability during pre- to early adolescence. By estimating probabilities of escalation, remission, persistence, and fluctuation in longitudinal cohort data, the framework provides a systematic, interpretable, and scalable approach to characterising suicidal trajectory dynamics in large, sparse mental health datasets, with potential to inform future research on monitoring, early prevention, and time-varying risk in youth mental health.

Objectives To identify research priorities for improving the mental health of neurodivergent higher education (HE) students by exploring the perspectives of individuals with lived experience. Design Qualitative study using an online survey. Data was analysed using a deductive-inductive, hybrid semantic thematic analysis. Setting UK higher education institutions. Participants 104 current and former neurodivergent HE students with diverse neurodivergent profiles and intersecting identities. Main outcome measures Participant recommendations regarding priorities for future research on neurodivergent student mental health. Results Six themes were identified and were grouped into (1) general recommendations for research and (2) recommendations specific to neurodivergence within a HE context. Participants prioritised a shift away from medical model approaches towards research informed by social and strengths-based perspectives. Key priorities included improving understanding of diagnostic barriers and misdiagnosis, reducing stigma, investigating institutional barriers within HE, evaluating the effectiveness of support and accommodations and examining the experiences of underrepresented and intersectional groups. Participants emphasised the need for research on more flexible teaching practices, sensory-friendly learning environments, integrated mental health and educational support and alternatives to diagnosis-dependent access to services. Conclusions Future research should move beyond descriptive accounts towards evaluating interventions and current support provision to understand if they improve the mental health of neurodivergent students. Adopting intersectional approaches, moving beyond binary deficit- or strengths-based frameworks and focusing on inclusive, needs-based support rather than diagnosis-led systems are likely to produce more equitable and effective outcomes for neurodivergent students in higher education.

Background: The rise of online data collection in psychological and health research has been accompanied by an increase in fraudulent participation, threatening data quality and research integrity. Method: We conducted semi-structured interviews and an open-ended questionnaire with six research teams across the UK who encountered fraudulent participants while conducting online studies across diverse populations (children aged 10+, adolescents, mid- older-aged adults aged 55 – 75, and general adult populations). Studies employed multiple methodologies, including online surveys, ecological momentary assessment (EMA), qualitative online interviews, and online diaries. Teams were selected to represent a range of methodological contexts; findings are intended as an exploratory contribution to awareness and discussion rather than a comprehensive account of fraud experiences across online research. Results: Reported fraudulent participation rates ranged from 1% to 89%, with social media recruitment platforms (particularly Facebook and X/Twitter) being most vulnerable. Analysis revealed that the manifestation of fraud varied by study design and population, with compensation-offering studies experiencing higher rates. Standard detection methods employed by studies included IP address inspection, address verification, demographic consistency checks, and interviewer intuition for qualitative studies. Teams reported significant impacts on research resources, study timelines (including a 2-month study pause in one case), and the decision-making process in ethics applications. In addition, teams also developed preventive strategies through experience. Conclusion: This paper presents an exploratory, experience-based framework intended to raise awareness of online research fraud and support researchers in thinking through context-appropriate detection and prevention strategies. Findings suggest that one-size-fits-all solutions are inadequate and that approaches need to be tailored to study design, recruitment methods, and resource constraints. We recommend that fraud mitigation planning be considered a standard requirement in research funding applications.

Background:COVID-19 public health and social measures (PHSM) may have affected children's development, for example by reducing their interaction with others. We examined associations between PHSM and developmental concerns among young children in Scotland. Methods:We utilised data from routine 13-15 month and 27-30 month child health reviews, covering 257,532 children in Scotland who took part in reviews between January 2019 and August 2023. Interrupted time-series analysis assessed slope changes in the weekly proportion of children with health visitor-identified developmental concerns following the March 2020 introduction of, and August 2021 removal of, PHSM. Outcomes were any 13-15 month and 27-30 month developmental concerns, and domain-specific concerns regarding speech-language-communication, problem-solving, gross-motor, personal-social, emotional-behavioural and fine-motor development. Findings:The March 2020 PHSM introduction was associated with a slope change increase in the proportion of children with any developmental concerns (+0.091 percentage points per week [95% CI: 0.065, 0.116] at 13-15 months and +0.076 percentage points per week [95% CI: 0.048, 0.104] at 27-30 months. The August 2021 PHSM removal was associated with a slope change decrease in the proportion of children with any developmental concerns at 27-30 months (-0.067 percentage points per week [95% CI: -0.088, -0.046]), but not 13-15 months, in the main analysis. Results were broadly consistent across developmental domains. Interpretation:COVID-19 PHSM were associated with increased developmental concerns among young children in Scotland. While leveraging interrupted time-series analysis yields findings consistent with a causal impact of PHSM, the influence of potential time-varying confounders cannot be ruled out.

Emotion malleability beliefs have been demonstrated to influence emotion regulation at the trait level. Contemporary theories propose that emotion regulation involves several stages: identifying the need to regulate, selecting strategies, and implementing those strategies. It remains unclear how emotion malleability beliefs relate to these stages in everyday life. This study examined the relationships between emotion malleability beliefs and the three stages of emotion regulation using a dynamic structural equation modelling (DSEM). Data from three studies (total N = 390) from the EMOTE database employing daily diary and ecological momentary assessment (EMA) were analysed. Emotion malleability beliefs were assessed at baseline, while emotion regulation strategies and negative emotions were assessed in daily lives. Results showed that emotion malleability beliefs were associated with the selection of cognitive reappraisal in the daily diary study, but not with any stage of emotion regulation in the EMA studies. Exploratory analyses further demonstrated a significant association between emotion malleability beliefs and the selection of situation modification in the daily diary study. These findings highlight the potential of EMA and daily diary studies to explore emotion regulation theories, while emphasising the need for further research into the dynamic relationships between emotion beliefs and regulation in daily life.

Recent theory predicts that young people's emotional reactions are important in the impact of social media use on their mental health. Here we explore this using ecological momentary assessment (EMA) data measuring social media use and emotion within the 2-week 'Decades-to-minutes' (D2M) study (n = 255). Dynamic structural equation modelling (DSEM) was used to examine the within-person relations between social media use and negative emotion (and vice versa), as well as the relations between negative emotions following social media use (defined as the within-person cross-lagged effect of social media use on negative affect) and depression. Results suggested a significant concurrent relation between social media use and negative emotion but no significant association between social media use and later negative emotion, or vice versa. They also suggested no significant relation between negative emotion following social media use and depression but a significant relation between social media inertia (stability of social media use from timepoint to timepoint) and depression. Our study illustrates the feasibility of capturing emotional dynamics in relation to social media use 'in ecological context' and their relations to mental health risk. It suggests that patterns of social media use (e.g., sustained versus intermittent use over and above levels of engagement) merit further research attention in terms of their association with mental health. Recommendations for future research include using shorter inter-prompt intervals to capture more immediate emotional reactions to social media use, and the development of more comprehensive EMA measures of social media use.

Ageism can be defined as stereotypes, prejudices and discrimination based on a person’s age and has been shown in past research to be both common and harmful. In 2016 WHO received a mandate from its Member States to lead a major global campaign to address ageism against all ages. Evidence synthesised as part of the campaign brought into focus the need for new measures of ageism experiences that can capture its multidimensional nature, and that can be used to measure the experiences of individuals of a wide range of ages and living in diverse country settings. Previous research has documented the development and validation of a 15-item version of the newly developed 15-item scale; however, as space in large-scale surveys is at a premium, there is a strong need for even briefer versions. The current article proposes and evaluates a 5-item brief WHO ageism experiences scale that measures self-directed, interpersonal, and institutional ageism and covers stereotypes, prejudices, and discrimination. Psychometric properties, including test-retest reliability and concurrent validity in a UK mixed-age adult sample (N= 402) were tested. Findings supported the convergent and divergent validity of the scale, and its test-retest reliability. It also showed patterns of associations with health and age that would be expected based on previous research. These results provide preliminary for support the use of the brief scale in contexts where survey space is highly constrained. It lays the foundation for future work to extend the testing of the 5-item scale in other languages, populations, and settings and to a wider range of relevant psychometric properties.

Sleep is essential for mental and physical health, and research in the field has substantially expanded over the past 50 years. Co-production methodology has been increasingly used within health and social care research, and refers to collaboration between researchers, policy makers, community partners and wider stakeholders. The aim of this scoping review was to detail the use of co-production methods within sleep research. A review of the existing literature was conducted using seven databases following PRISMA-ScR guidelines. Search terms included objective and subjective sleep outcomes, and the use of co-production research methodologies. Sixteen studies were included in the final review: 10 studies used solely qualitative co-production methods to inform intervention design and development (sleep as a primary outcome [n = 5] and as a secondary outcome [n = 5]), and six studies used co-production methodologies to establish sleep as a priority outcome for future research. Most studies used consultation approaches to design interventions (n = 8), instead of using co-design teams (n = 2). Two studies focusing on intervention development recruiting participants from clinical populations with poor sleep, other studies recruited from those with other underlying conditions or a healthy population. The most common limitations of the included studies were small sample size, researcher driven topics/domains for the PAR components, under-representative samples and COVID-19 pressures. Future sleep research should consider the use of co-production methodologies from the study conceptualisation, through to the design, development and implementation of research to further benefit the intended research population.

OBJECTIVES:We created the Digital Activity and Feelings Inventory (DAFI) to measure youth digital activities and the psychological reactions they evoke, established its psychometric properties and tested its validity in predicting mental health relative to screen time estimates. METHODS:An initial pool of items was generated using the existing research on youth digital activity and mental health and further refined via consultations with experts and young people (online youth panel sessions, n = 14). The participants (n = 383, mean age = 19 years) completed the resulting DAFI alongside established measures of depression, anxiety, wellbeing, and screen time. The DAFI factor structure, reliability and predictive validity were tested. RESULTS:Exploratory factor analyses identified five digital activity subscales: Risky Content, Risky Interactions, Social Comparison, Leisure Activities and Social Engagement and three psychological reactions subscales: Negative Self-Reactions, Negative Stress Reactions, and Positive Reactions. Internal consistency and test-retest reliability were high. Social Comparison and Negative Self-Reactions, but not screen time, independently predicted depression and anxiety symptoms. Positive Reactions, lack of Negative Self-Reactions, lower screen time and Social Engagement predicted wellbeing. CONCLUSION:The DAFI is a reliable measure of digital activities and associated psychological reactions and predicts youth mental health better than screen time.

INTRODUCTION:Adolescence is a key period of development for mental health; however, little is known about how (cumulative) daily life experiences impact long-term mental health development in this period, and vice versa. 'Mental health in the moment' (MHIM) is an accelerated cohort measurement burst study designed to illuminate these links. METHODS AND ANALYSIS:The current protocol describes the rationale and design for MHIM, which aims to recruit and follow up approximately 500 adolescents across five age cohorts (in secondary school years S1-S5, aged 11-16 at baseline) and follow them over a 5-year data collection period. Data collection will include online surveys and ecological momentary assessments bursts every 6 months, annual caregiver surveys, the collection of stress biomarker data at three key measurement points and continuous radar-based sleep measurement for a subsample of participants. The study is informed by a young person advisory group input throughout its lifecycle. Data will be analysed using techniques such as dynamic structural equation modelling. The study can provide insights into mental health development from a multitimeframe developmental perspective, including insights into 'daily life' intervention targets for improving adolescent mental health. ETHICS AND DISSEMINATION:The study received ethical approval from the philosophy, psychology and language science ethics committee at the University of Edinburgh (404-2425/3) and the findings will be published in a series of peer-reviewed publications.

BACKGROUND:There is now widespread recognition within adolescence mental health research of the ethical imperative and benefits of coproduction. This has led to the development of best practice guidelines and the routine reporting of coproduction methods. However, there are unique considerations associated with involving young people in research with varying designs, e.g., longitudinal. Ideally, these decisions are made in consultation with young people themselves. Thus, the objective and novel contribution of this paper is (1) the involvement of young people in the development of the MHIM protocol, including an evaluative framework, and (2) a study protcol that utilises coproduction in the context of a longitudinal mental health project. METHODS:We present a coproduction protocol (MHIM-YPAG), developed with young people (pre-study YPAG), that describes the planned approach to involving young people in the MHIM study. MHIM will use an accelerated longitudinal-cohort design that will combine ecological momentary assessment, bio-sampling, radar-based sleep measurement and online questionnaires to examine the relationship between daily-life-experiences and adolescence mental health. The protocol describes how young people will be involved in MHIM and how impact will be evaluated. Within this context, coproduction relates to (1) the codevelopment, with young people, of the study's structure and (2) their involvement as researchers, e.g., recruitment, data interpretation and knowledge-exchange. PATIENT OR PUBLIC CONTRIBUTION:As one of the first protocols of its kind, our coproduction protocol, informed by literature and consultations with young people (pre-study YPAG), can provide a template for planning and reporting coproduction in mental health studies of adolescence.

Bullying, cyberbullying and substance use are prevalent problem behaviors in adolescents, with adverse consequences. Although previous research has provided important insights into their connections, it remains to be established if there are directional relations in which bullying and cyberbullying are predictors of substance use or whether their relation primarily reflects a common dependence on individual difference factors that tend to be stable within people but vary across people. The current study thus delineated between-individual and withinindividual longitudinal relations between bullying, cyberbullying, and substance use in adolescents using a prospective longitudinal study with three data collection points. The sample included 1,481 students enrolled in primary and secondary schools in Spain. Data were collected through a survey with validated questionnaires. Analyses were performed using ALT-SRs to examine the within-person reciprocal longitudinal relations between bullying/cyberbullying and substance use. Regarding the within-person relations between substance use, bullying, and cyberbullying, only bullying perpetration in 10-17-years olds predicted substance use in 11-17years olds. We also found a between-person relation between cybervictimization and substance use, and significant cross-sectional and longitudinal correlations among bullying, cyberbullying and substance use in most waves. These results should be taken into account for policy and practice against bullying, cyberbullying, and substance use. Comprehensive interventions that target all these problem behaviors should be designed, implemented and evaluated. More longitudinal research on the relation between these and other problem behaviors should be conducted.

BackgroundLived experience (LE) expertise is increasingly recognised as a vital component in mental health research. In our project to develop a digital tool for counterfactual analysis (DigiCAT), with an emphasis on researching active ingredients for adolescent mental health, we incorporated LE expertise across the lifecycle of tool development and dissemination.MethodsWe consulted young person advisory groups (YPAGs; aged 11-18 years old) across three project phases-Discovery, Prototyping, and Dissemination-using structured discussions, ranking exercises, and iterative feedback loops to shape research priorities, tool design and user tutorials, and dissemination strategies of the tool.ResultsThe YPAGs advised on active ingredients and features of such ingredients that existing research has not taken into account. Examples include young person advisory group (YPAG) suggestions to prioritise research in social media, peer relationships, and teacher-student relationships. We incorporated these suggestions as illustrative examples in our tutorial paper introducing DigiCAT to our target audience, to demonstrate how insights from YPAGs can inform the use of the tool in research, by guiding areas of study. These areas were also prioritised in our own empirical analyses using the tool. Additionally, the YPAG contributed practical guidance on how to effectively involve youth LE experts in both research and digital tool development processes.ConclusionsThe integration of LE expertise fundamentally shaped the development of our tutorial paper, influencing both its instructional components and its broader discussion of applications in mental health research. This project highlights the value of embedding LE perspectives into research guidance for software use, offering a model for future mental health and digital innovation initiatives.

Internalising problems such as anxiety and depression affect a sizeable proportion of adolescents and show a peak age of onset around the transition to adolescence. Emotion regulation issues have been associated with internalising problems; however, it is less clear whether emotion regulation issues in childhood are a causal precursor to internalising problems in adolescence. The present study uses a counterfactual analysis approach in a large-scale UK representative study to assess whether emotion regulation issues at age 7 appear to have a causal impact on internalising problems at age 11,14, and 17, when accounting for confounding factors, including pre-existing internalising problems. Results suggested a modest but significant effect of emotion dysregulation in childhood on adolescent mental health. These results suggest that childhood emotion regulation problems impact mental health later in development and potentially point to childhood emotion dysregulation as a promising intervention target for reducing adolescent mental health issues.

Despite growing evidence on the influence of neighborhood characteristics on maternal well-being, there is a critical gap in the availability of validated instruments for measuring these constructs across different cultural settings. Existing neighborhood-related scales often lack cross-cultural validation, limiting their applicability in low- and middle-income countries. Understanding the impact of neighborhood characteristics is crucial given its multigenerational impact. We used data from the Evidence for Better Lives dataset to assess the conceptual and measurement equivalence of the community scales of neighborhood cohesion, intergenerational closure, and neighborhood and social disorder, testing for measurement invariance across eight low- and middle-income countries. Secondly, we examined patterns of associations with prenatal maternal stress, well-being, and depressive symptoms through the use of nomological networks. We found that the conceptual and measurement equivalence of the neighborhood domains were comparable across the eight studied countries. Additionally, our results suggest that higher levels of neighborhood and social disorder and lower levels of cohesion and intergenerational closure in the community are associated with adverse maternal outcomes across the included sites. The results of this study stress the importance of exploring the community context when assessing maternal well-being and supports the need to advocate for community-based interventions that promote safer physical and social environments within maternal programs and urban planning.

Research suggests a strong, dose-response relationship between adverse childhood experiences (ACEs) and poor adult mental health outcomes. This systematic review and meta-analysis aimed to systematically investigate the existence and strength of prospective associations between ACEs and adult mental health outcomes: anxiety, depression, psychotic-like experiences, post-traumatic stress disorder (PTSD), self-harm, and suicidality. We searched 12 electronic databases for publications after 1990. A narrative synthesis of included studies and random-effects meta-analyses with moderation were completed for all outcomes, excluding self-harm. In total, 62 studies from 15 countries were included. Most studies were from the United States; 95% of publications (N = 59) came from high-income countries (HICs) and 5% (N = 3) from upper-middle-income countries. Pooled associations between ACEs and adult mental illness were strongest for PTSD (OR = 2.26; 95% CI [1.75, 2.77]), followed by anxiety (OR = 1.78; 95% CI [1.45, 2.11]), depression (OR = 1.61; 95% CI [1.45, 1.76]), psychotic-like experiences (OR = 1.34; 95% CI [1.13, 1.54]), and suicidality (OR = 1.28; 95% CI [1.13, 1.43]). Moderation analyses suggested ACEs with a violence or maltreatment component were significant risk factors for adult depression compared to household ACEs, and that study location was a significant moderator in the depression, anxiety, and PTSD models. Further moderation effects will be discussed. Findings confirm ACEs are a significant risk factor for mental ill-health in adulthood. Our review highlights the urgent need for research exploring associations between ACEs measured in childhood and adult mental illness outside of HIC settings.

Early identification of risk for attention-deficit hyperactivity disorder (ADHD) symptoms can enable more timely interventions and improve long-term outcomes. While previous research has linked various maternal and perinatal factors to ADHD, few studies have examined these predictors collectively in a single comprehensive analysis. This study aimed to assess whether later ADHD symptoms can be predicted from information available at birth, specifically ethnicity, maternal metabolic markers, mental health, and socioeconomic status. It additionally aimed to identify the most influential predictors. Using data from the Born in Bradford (BiB) study, we applied multiple linear regression (LR) and machine learning techniques to predict ADHD symptoms as measured by the Hyperactivity/Inattention subscale of the Strengths and Difficulties Questionnaire (SDQ). A 10-fold cross-validated LR model explained 6.97% of the variance in SDQ scores. In the random forest model, infant male sex and maternal smoking during pregnancy emerged as the top predictors. These findings provide proof of principle for early identification of children at risk of ADHD. Future models may benefit from incorporating additional perinatal data to improve predictive accuracy.