Anxiety onset in adolescents: a machine-learning prediction.

Recent longitudinal studies in youth have reported MRI correlates of prospective anxiety symptoms during adolescence, a vulnerable period for the onset of anxiety disorders. However, their predictive value has not been established. Individual prediction through machine-learning algorithms might help bridge the gap to clinical relevance. A voting classifier with Random Forest, Support Vector Machine and Logistic Regression algorithms was used to evaluate the predictive pertinence of gray matter volumes of interest and psychometric scores in the detection of prospective clinical anxiety. Participants with clinical anxiety at age 18-23 (N = 156) were investigated at age 14 along with healthy controls (N = 424). Shapley values were extracted for in-depth interpretation of feature importance. Prospective prediction of pooled anxiety disorders relied mostly on psychometric features and achieved moderate performance (area under the receiver operating curve = 0.68), while generalized anxiety disorder (GAD) prediction achieved similar performance. MRI regional volumes did not improve the prediction performance of prospective pooled anxiety disorders with respect to psychometric features alone, but they improved the prediction performance of GAD, with the caudate and pallidum volumes being among the most contributing features. To conclude, in non-anxious 14 year old adolescents, future clinical anxiety onset 4-8 years later could be individually predicted. Psychometric features such as neuroticism, hopelessness and emotional symptoms were the main contributors to pooled anxiety disorders prediction. Neuroanatomical data, such as caudate and pallidum volume, proved valuable for GAD and should be included in prospective clinical anxiety prediction in adolescents.

Practitioner Review: Assessment and treatment of body dysmorphic disorder in young people.

Body dysmorphic disorder (BDD) is a relatively common and highly impairing mental disorder that is strikingly underdiagnosed and undertreated in Child and Adolescent Mental Health Services (CAMHS). The only clinical guidelines for the management of BDD in youth were published nearly 20 years ago, when empirical knowledge was sparse. Fortunately, there has been a surge in research into BDD over the last 10 years, shedding important insights into the phenomenology, epidemiology, assessment and treatment of the disorder in young people. This review aimed to provide an overview of recent research developments of relevance to clinicians and healthcare policymakers. We summarise key findings regarding the epidemiology of BDD in youth, which indicate that the disorder usually develops during teenage years and affects approximately 2% of adolescents at any one point in time. We provide an overview of aetiological research, highlighting that BDD arises from an interplay between genetic and environmental influences. We then focus on screening and assessment strategies, arguing that these are crucial to promote detection and diagnosis of this under-recognised condition. Additionally, we summarise the recommended treatment approaches for BDD in youth, namely cognitive behaviour therapy with or without selective serotonin reuptake inhibitors. The review concludes by highlighting key knowledge gaps and priorities for future research including, but not limited to, better understanding aetiological factors, long-term consequences and treatment.

Explainable machine learning approach to predict and explain the relationship between task-based fMRI and individual differences in cognition.

Despite decades of costly research, we still cannot accurately predict individual differences in cognition from task-based functional magnetic resonance imaging (fMRI). Moreover, aiming for methods with higher prediction is not sufficient. To understand brain-cognition relationships, we need to explain how these methods draw brain information to make the prediction. Here we applied an explainable machine-learning (ML) framework to predict cognition from task-based fMRI during the n-back working-memory task, using data from the Adolescent Brain Cognitive Development (n = 3,989). We compared 9 predictive algorithms in their ability to predict 12 cognitive abilities. We found better out-of-sample prediction from ML algorithms over the mass-univariate and ordinary least squares (OLS) multiple regression. Among ML algorithms, Elastic Net, a linear and additive algorithm, performed either similar to or better than nonlinear and interactive algorithms. We explained how these algorithms drew information, using SHapley Additive explanation, eNetXplorer, Accumulated Local Effects, and Friedman's H-statistic. These explainers demonstrated benefits of ML over the OLS multiple regression. For example, ML provided some consistency in variable importance with a previous study and consistency with the mass-univariate approach in the directionality of brain-cognition relationships at different regions. Accordingly, our explainable-ML framework predicted cognition from task-based fMRI with boosted prediction and explainability over standard methodologies.

Editorial Perspective: When is a 'small effect' actually large and impactful?

Reporting of effect sizes is standard practice in psychology and psychiatry research. However, interpretation of these effect sizes can be meaningless or misleading - in particular, the evaluation of specific effect sizes as 'small', 'medium' and 'large' can be inaccurate depending on the research context. A real-world example of this is research into the mental health of children and young people during the COVID-19 pandemic. Evidence suggests that clinicians and services are struggling with increased demand, yet population studies looking at the difference in mental health before and during the pandemic report effect sizes that are deemed 'small'. In this short review, we utilise simulations to demonstrate that a relatively small shift in mean scores on mental health measures can indicate a large shift in the number of cases of anxiety and depression when scaled up to an entire population. This shows that 'small' effect sizes can in some contexts be large and impactful.

Magnetoencephalographic correlates of mood and reward dynamics in human adolescents.

Despite its omnipresence in everyday interactions and its importance for mental health, mood and its neuronal underpinnings are poorly understood. Computational models can help identify parameters affecting self-reported mood during mood induction tasks. Here, we test if computationally modeled dynamics of self-reported mood during monetary gambling can be used to identify trial-by-trial variations in neuronal activity. To this end, we shifted mood in healthy (N = 24) and depressed (N = 30) adolescents by delivering individually tailored reward prediction errors while recording magnetoencephalography (MEG) data. Following a pre-registered analysis, we hypothesize that the expectation component of mood would be predictive of beta-gamma oscillatory power (25-40 Hz). We also hypothesize that trial variations in the source localized responses to reward feedback would be predicted by mood and by its reward prediction error component. Through our multilevel statistical analysis, we found confirmatory evidence that beta-gamma power is positively related to reward expectation during mood shifts, with localized sources in the posterior cingulate cortex. We also confirmed reward prediction error to be predictive of trial-level variations in the response of the paracentral lobule. To our knowledge, this is the first study to harness computational models of mood to relate mood fluctuations to variations in neural oscillations with MEG.

Assessing the feasibility of a web-based outcome measurement system in child and adolescent mental health services - myHealthE a randomised controlled feasibility pilot study.

BACKGROUND: Interest in internet-based patient reported outcome measure (PROM) collection is increasing. The NHS myHealthE (MHE) web-based monitoring system was developed to address the limitations of paper-based PROM completion. MHE provides a simple and secure way for families accessing Child and Adolescent Mental Health Services to report clinical information and track their child's progress. This study aimed to assess whether MHE improves the completion of the Strengths and Difficulties Questionnaire (SDQ) compared with paper collection. Secondary objectives were to explore caregiver satisfaction and application acceptability. METHODS: A 12-week single-blinded randomised controlled feasibility pilot trial of MHE was conducted with 196 families accessing neurodevelopmental services in south London to examine whether electronic questionnaires are completed more readily than paper-based questionnaires over a 3-month period. Follow up process evaluation phone calls with a subset (n = 8) of caregivers explored system satisfaction and usability. RESULTS: MHE group assignment was significantly associated with an increased probability of completing an SDQ-P in the study period (adjusted hazard ratio (HR) 12.1, 95% CI 4.7-31.0; p = <.001). Of those caregivers' who received the MHE invitation (n = 68) 69.1% completed an SDQ using the platform compared to 8.8% in the control group (n = 68). The system was well received by caregivers, who cited numerous benefits of using MHE, for example, real-time feedback and ease of completion. CONCLUSIONS: MHE holds promise for improving PROM completion rates. Research is needed to refine MHE, evaluate large-scale MHE implementation, cost effectiveness and explore factors associated with differences in electronic questionnaire uptake.

Capturing brain-cognition relationship: Integrating task-based fMRI across tasks markedly boosts prediction and test-retest reliability.

Capturing individual differences in cognition is central to human neuroscience. Yet our ability to estimate cognitive abilities via brain MRI is still poor in both prediction and reliability. Our study tested if this inability can be improved by integrating MRI signals across the whole brain and across modalities, including task-based functional MRI (tfMRI) of different tasks along with other non-task MRI modalities, such as structural MRI, resting-state functional connectivity. Using the Human Connectome Project (n = 873, 473 females, after quality control), we directly compared predictive models comprising different sets of MRI modalities (e.g., seven tasks vs. non-task modalities). We applied two approaches to integrate multimodal MRI, stacked vs. flat models, and implemented 16 combinations of machine-learning algorithms. The stacked model integrating all modalities via stacking Elastic Net provided the best prediction (r = 0.57), relatively to other models tested, as well as excellent test-retest reliability (ICC=∼.85) in capturing general cognitive abilities. Importantly, compared to the stacked model integrating across non-task modalities (r = 0.27), the stacked model integrating tfMRI across tasks led to significantly higher prediction (r = 0.56) while still providing excellent test-retest reliability (ICC=∼.83). The stacked model integrating tfMRI across tasks was driven by frontal and parietal areas and by tasks that are cognition-related (working-memory, relational processing, and language). This result is consistent with the parieto-frontal integration theory of intelligence. Accordingly, our results contradict the recently popular notion that tfMRI is not reliable enough to capture individual differences in cognition. Instead, our study suggests that tfMRI, when used appropriately (i.e., by drawing information across the whole brain and across tasks and by integrating with other modalities), provides predictive and reliable sources of information for individual differences in cognitive abilities, more so than non-task modalities.

Longitudinally stable, brain-based predictive models mediate the relationships between childhood cognition and socio-demographic, psychological and genetic factors.

Cognitive abilities are one of the major transdiagnostic domains in the National Institute of Mental Health's Research Domain Criteria (RDoC). Following RDoC's integrative approach, we aimed to develop brain-based predictive models for cognitive abilities that (a) are developmentally stable over years during adolescence and (b) account for the relationships between cognitive abilities and socio-demographic, psychological and genetic factors. For this, we leveraged the unique power of the large-scale, longitudinal data from the Adolescent Brain Cognitive Development (ABCD) study (n ~ 11 k) and combined MRI data across modalities (task-fMRI from three tasks: resting-state fMRI, structural MRI and DTI) using machine-learning. Our brain-based, predictive models for cognitive abilities were stable across 2 years during young adolescence and generalisable to different sites, partially predicting childhood cognition at around 20% of the variance. Moreover, our use of 'opportunistic stacking' allowed the model to handle missing values, reducing the exclusion from around 80% to around 5% of the data. We found fronto-parietal networks during a working-memory task to drive childhood-cognition prediction. The brain-based, predictive models significantly, albeit partially, accounted for variance in childhood cognition due to (1) key socio-demographic and psychological factors (proportion mediated = 18.65% [17.29%-20.12%]) and (2) genetic variation, as reflected by the polygenic score of cognition (proportion mediated = 15.6% [11%-20.7%]). Thus, our brain-based predictive models for cognitive abilities facilitate the development of a robust, transdiagnostic research tool for cognition at the neural level in keeping with the RDoC's integrative framework.

Linked patterns of biological and environmental covariation with brain structure in adolescence: a population-based longitudinal study.

Adolescence is a period of major brain reorganization shaped by biologically timed and by environmental factors. We sought to discover linked patterns of covariation between brain structural development and a wide array of these factors by leveraging data from the IMAGEN study, a longitudinal population-based cohort of adolescents. Brain structural measures and a comprehensive array of non-imaging features (relating to demographic, anthropometric, and psychosocial characteristics) were available on 1476 IMAGEN participants aged 14 years and from a subsample reassessed at age 19 years (n = 714). We applied sparse canonical correlation analyses (sCCA) to the cross-sectional and longitudinal data to extract modes with maximum covariation between neuroimaging and non-imaging measures. Separate sCCAs for cortical thickness, cortical surface area and subcortical volumes confirmed that each imaging phenotype was correlated with non-imaging features (sCCA r range: 0.30-0.65, all PFDR < 0.001). Total intracranial volume and global measures of cortical thickness and surface area had the highest canonical cross-loadings (|ρ| = 0.31-0.61). Age, physical growth and sex had the highest association with adolescent brain structure (|ρ| = 0.24-0.62); at baseline, further significant positive associations were noted for cognitive measures while negative associations were observed at both time points for prenatal parental smoking, life events, and negative affect and substance use in youth (|ρ| = 0.10-0.23). Sex, physical growth and age are the dominant influences on adolescent brain development. We highlight the persistent negative influences of prenatal parental smoking and youth substance use as they are modifiable and of relevance for public health initiatives.

Clinical utility of family history of depression for prognosis of adolescent depression severity and duration assessed with predictive modeling.

BACKGROUND: Family history of depression (FHD) is a known risk factor for the new onset of depression. However, it is unclear if FHD is clinically useful for prognosis in adolescents with current, ongoing, or past depression. This preregistered study uses a longitudinal, multi-informant design to examine whether a child's FHD adds information about future depressive episodes and depression severity applying state-of-the-art predictive out-of-sample methodology. METHODS: We examined data in adolescents with current or past depression (age 11-17 years) from the National Institute of Mental Health Characterization and Treatment of Adolescent Depression (CAT-D) study. We asked whether a history of depression in a first-degree relative was predictive of depressive episode duration (72 participants) and future depressive symptom severity in probands (129 participants, 1,439 total assessments). RESULTS: Family history of depression, while statistically associated with time spent depressed, did not improve predictions of time spent depressed, nor did it improve models of change in depression severity measured by self- or parent-report. CONCLUSIONS: Family history of depression does not improve the prediction of the course of depression in adolescents already diagnosed with depression. The difference between statistical association and predictive models highlights the importance of assessing predictive performance when evaluating questions of clinical utility.

Validation of an irritability measure in preschoolers in school-based and clinical Brazilian samples.

The Affective Reactivity Index (ARI) is an irritability measure with good psychometric properties. However, there are no published studies in preschool children, an important population in which to differentiate normative from non-normative irritability. The goal of this study was to validate the ARI in preschoolers. Two samples were included: a school-based sample (N = 487, mean age = 57.80 ± 7.23 months, 52.8% male) and a clinical sample of children with Attention Deficit Hyperactivity Disorder (ADHD; N = 153, mean age = 60.5 ± 7.6 months, 83.7% males). Confirmatory factor analysis assessed ARI unidimensionality. ARI criterion validity was tested through comparison to other scales measuring irritability, related constructs, and other aspects of psychopathology. Test-retest reliability was assessed in the school-based sample. Analyses confirmed a single-factor structure and good internal consistency. The ARI showed stronger correlations with irritability measures than with measures of other constructs. In the clinical sample, ADHD children with comorbid disruptive behavior disorders had higher ARI scores than those without this comorbidity. In the school-based sample, test-retest reliability was moderate. This is the first study to demonstrate ARI validity and reliability in preschoolers. The scale performed well in both school-based and clinical samples. Having a concise and validated irritability measure for preschoolers may facilitate both clinical assessment and research on early irritability.

Genetic Correlates of Psychological Responses to the COVID-19 Crisis in Young Adult Twins in Great Britain.

We investigated how the COVID-19 crisis and the extraordinary experience of lockdown affected young adults in England and Wales psychologically. One month after lockdown commenced (T2), we assessed 30 psychological and behavioural traits in more than 4000 twins in their mid-twenties and compared their responses to the same traits assessed in 2018 (T1). Mean changes from T1 to T2 were modest and inconsistent. Contrary to the hypothesis that major environmental changes related to COVID-19 would result in increased variance in psychological and behavioural traits, we found that the magnitude of individual differences did not change from T1 to T2. Twin analyses revealed that while genetic factors accounted for about half of the reliable variance at T1 and T2, they only accounted for ~ 15% of individual differences in change from T1 to T2, and that nonshared environmental factors played a major role in psychological and behavioural changes. Shared environmental influences had negligible impact on T1, T2 or T2 change. Genetic factors correlated on average .86 between T1 and T2 and accounted for over half of the phenotypic stability, as would be expected for a 2-year interval even without the major disruption of lockdown. We conclude that the first month of lockdown has not resulted in major psychological or attitudinal shifts in young adults, nor in major changes in the genetic and environmental origins of these traits. Genetic influences on the modest psychological and behavioural changes are likely to be the result of gene-environment correlation not interaction.

Distinct brain structure and behavior related to ADHD and conduct disorder traits.

Attention-Deficit/Hyperactivity Disorder (ADHD) and conduct disorder (CD) exemplify top-down dysregulation conditions that show a large comorbidity and shared genetics. At the same time, they entail two different types of symptomology involving mainly non-emotional or emotional dysregulation. Few studies have tried to separate the specific biology underlying these two dimensions. It has also been suggested that both types of conditions consist of extreme cases in the general population where the symptoms are widely distributed. Here we test whether brain structure is specifically associated to ADHD or CD symptoms in a general population of adolescents (n = 1093) being part of the IMAGEN project. Both ADHD symptoms and CD symptoms were related to similar and overlapping MRI findings of a smaller structure in prefrontal and anterior cingulate cortex. However, our regions of interest (ROI) approach indicated that gray matter volume (GMV) and surface area (SA) in dorsolateral/dorsomedial prefrontal cortex and caudal anterior cingulate cortex were negatively associated to ADHD symptoms when controlling for CD symptoms while rostral anterior cingulate cortex GMV was negatively associated to CD symptoms when controlling for ADHD symptoms. The structural findings were mirrored in performance of neuropsychological tests dependent on prefrontal and anterior cingulate regions, showing that while performance on the Stop Signal test was specifically related to the ADHD trait, delayed discounting and working memory were related to both ADHD and CD traits. These results point towards a partially domain specific and dimensional capacity in different top-down regulatory systems associated with ADHD and CD symptoms.

Sociodemographic factors associated with routine outcome monitoring: a historical cohort study of 28,382 young people accessing child and adolescent mental health services.

BACKGROUND: Patient-reported outcome measures (PROMs) are important tools to inform patients, clinicians and policy-makers about clinical need and the effectiveness of any given treatment. Consistent PROM use can promote early symptom detection, help identify unexpected treatment responses and improve therapeutic engagement. Very few studies have examined associations between patient characteristics and PROM data collection. METHODS: We used the electronic mental health records for 28,382 children and young people (aged 4-17 years) accessing Child and Adolescent Mental Health Services (CAMHS) across four South London boroughs between the 1st of January 2008 to the 1st of October 2017. We examined the completion rates of the caregiver Strengths and Difficulties Questionnaire (SDQ), a ubiquitous PROM for CAMHS at baseline and 6-month follow-up. RESULTS AND CONCLUSIONS: SDQs were present for approximately 40% (n = 11,212) of the sample at baseline, and from these, only 8% (n = 928) had a follow-up SDQ. Patterns of unequal PROM collection by sociodemographic factors were identified: males were more likely (aOR 1.07, 95% CI 1.01-1.13), whilst older age (aOR 0.87, 95% CI 0.87-0.88), Black (aOR 0.79 95% CI 0.74-0.84) and Asian ethnicity (aOR 0.75 95% CI 0.66-0.86) relative to White ethnicity, and residence within the most deprived neighbourhood (aOR 0.87 95% CI 0.80-0.94) were less likely to have a record of baseline SDQ. Similar results were found in the sub-group (n = 11,212) with follow-up SDQ collection. Our findings indicate systematic differences in the currently available PROMS data and highlights which groups require increased focus if we are to gain equitable PROM collection. We need to ensure representative PROM collection for all individuals accessing treatment, regardless of ethnic or socioeconomic background; biased data have adverse ramifications for policy and service level decision-making.

Sex effects on structural maturation of the limbic system and outcomes on emotional regulation during adolescence.

Though adolescence is a time of emerging sex differences in emotions, sex-related differences in the anatomy of the maturing brain has been under-explored over this period. The aim of this study was to investigate whether puberty and sexual differentiation in brain maturation could explain emotional differences between girls and boys during adolescence. We adapted a dedicated longitudinal pipeline to process structural and diffusion images from 335 typically developing adolescents between 14 and 16 years. We used voxel-based and Regions of Interest approaches to explore sex and puberty effects on brain and behavioral changes during adolescence. Sexual differences in brain maturation were characterized by amygdala and hippocampal volume increase in boys and decrease in girls. These changes were mediating the sexual differences in positive emotional regulation as illustrated by positive attributes increase in boys and decrease in girls. Moreover, the differential maturation rates between the limbic system and the prefrontal cortex highlighted the delayed maturation in boys compared to girls. This is the first study to show the sex effects on the differential cortico/subcortical maturation rates and the interaction between sex and puberty in the limbic system maturation related to positive attributes, reported as being protective from emotional disorders.

Annual Research Review: Defining and treating pediatric treatment-resistant depression.

BACKGROUND: Adolescent major depressive disorder (MDD) is a significant health problem, associated with substantial morbidity, cost, and mortality. Depression is a significant risk factor for suicide, which is now the second leading cause of death in young people. Up to twenty per cent of adolescents will experience MDD before adulthood, and while a substantial proportion will improve with standard-of-care treatments (psychotherapy and medication), roughly one third will not. METHODS: Here, we have reviewed the literature in order to discuss the concept of treatment-resistant depression (TRD) in adolescence, examine risk factors, diagnostic difficulties, and challenges in evaluating symptom improvement, and providing guidance on how to define adequate medication and psychotherapy treatment trials. RESULTS: We propose a staging model for adolescent TRD and review the treatment literature. The evidence base for first- and second-line treatments primarily derives from four large pediatric clinical trials (TADS, TORDIA, ADAPT, and IMPACT). After two medications and a trial of evidence-based psychotherapy have failed to alleviate depressive symptoms, the evidence becomes quite thin for subsequent treatments. Here, we review the evidence for the effectiveness of medication switches, medication augmentation, psychotherapy augmentation, and interventional treatments (i.e., transcranial magnetic stimulation, electroconvulsive therapy, and ketamine) for adolescent TRD. Comparisons are drawn to the adult TRD literature, and areas for future pediatric depression research are highlighted. CONCLUSIONS: As evidence is limited for treatments in this population, a careful consideration of the known risks and side effects of escalated treatments (e.g., mood stabilizers and atypical antipsychotics) is warranted and weighed against potential, but often untested, benefits.

Heavy drinking in adolescents is associated with change in brainstem microstructure and reward sensitivity.

Heavy drinker adolescents: altered brainstem microstructure.

The cortical-cerebellar circuit is vulnerable to heavy drinking (HD) in adults. We hypothesized early microstructural modifications of the pons/midbrain region, containing core structures of the reward system, in HD adolescents. Thirty-two otherwise symptom-free HDs at age 14 (HD14) and 24 abstainers becoming HDs at age 16 (HD16) were identified in the community with the Alcohol Use Disorders Identification Test (AUDIT) and compared with abstainers. The monetary incentive delay (MID) task assessed reward-sensitive performance. Voxelwise statistics of diffusion tensor imaging (DTI) values in the thalamo-ponto-mesencephalic region were obtained using tract-based spatial statistics. Projections between the ventral tegmental area (VTA) and the nucleus accumbens (NAcc) were identified by probabilistic tractography. Lower fraction of anisotropy and higher radial diffusivity (RD) values were detected in the upper dorsal pons of HD14 adolescents, and a trend for higher RD in HD16, compared with abstainers. When expecting reward, HD14 had higher MID task success scores than abstainers, and success scores were higher with a lower number of tracts in all adolescents. In symptom-free community adolescents, a region of lower white matter (WM) integrity in the pons at age 14 was associated with current HD and predicted HD at age 16. HD was related to reward sensitivity.

Modulation of anterior cingulate cortex reward and penalty signalling in medication-naive young-adult subjects with depressive symptoms following acute dose lurasidone.

BACKGROUND: Aberrations in reward and penalty processing are implicated in depression and putatively reflect altered dopamine signalling. This study exploits the advantages of a placebo-controlled design to examine how a novel D2 antagonist with adjunctive antidepressant properties modifies activity in the brain's reward network in depression. METHODS: We recruited 43 medication-naïve subjects across the range of depression severity (Beck's Depression Inventory-II score range: 0-43), including healthy volunteers, as well as people meeting full-criteria for major depressive disorder. In a double-blind placebo-controlled cross-over design, all subjects received either placebo or lurasidone (20 mg) across two visits separated by 1 week. Functional magnetic resonance imaging with the Monetary Incentive Delay (MID) task assessed reward functions via neural responses during anticipation and receipt of gains and losses. Arterial spin labelling measured cerebral blood flow (CBF) at rest. RESULTS: Lurasidone altered fronto-striatal activity during anticipation and outcome phases of the MID task. A significant three-way Medication-by-Depression severity-by-Outcome interaction emerged in the anterior cingulate cortex (ACC) after correction for multiple comparisons. Follow-up analyses revealed significantly higher ACC activation to losses in high- v. low depression participants in the placebo condition, with a normalisation by lurasidone. This effect could not be accounted for by shifts in resting CBF. CONCLUSIONS: Lurasidone acutely normalises reward processing signals in individuals with depressive symptoms. Lurasidone's antidepressant effects may arise from reducing responses to penalty outcomes in individuals with depressive symptoms.

Editorial: Are computers going to take over: implications of machine learning and computational psychiatry for trainees and practising clinicians.

Are computers going to take over? In some ways they already have, and they are bound to take over more. As clinicians and researchers, what do we need to do in order to handle, trust, and evaluate intelligent machines?

Childhood neurodevelopmental difficulties and risk of adolescent depression: the role of irritability.

BACKGROUND: Children with neurodevelopmental disorders are at increased risk of developing depression. Irritability predicts depression in the general population and is common in children with neurodevelopmental disorders. Thus, it is possible that irritability in children with neurodevelopmental disorders contributes to the link with later depression. This study aimed to (a) examine the association between childhood neurodevelopmental difficulties and adolescent depression and (b) test whether irritability explains this association. METHODS: Children with any neurodevelopmental difficulty at the age of 7-9 (n = 1,697) and a selected, comparison group without any neurodevelopmental difficulty (n = 3,177) were identified from a prospective, UK population-based cohort, the Avon Longitudinal Study of Parents and Children. Neurodevelopmental difficulties were defined as a score in the bottom 5% of the sample on at least one measure of cognitive ability, communication, autism spectrum symptoms, attention-deficit/hyperactivity symptoms, reading or motor coordination. The Development and Well-Being Assessment measured parent-reported child irritability at the age of 7, parent-reported adolescent depression at the age of 10 and 13, and self-reported depression at the age of 15. Depression measures were combined, deriving an outcome of major depressive disorder (MDD) in adolescence. Logistic regression examined the association between childhood neurodevelopmental difficulties and adolescent MDD, controlling for gender. Path analysis estimated the proportion of this association explained by irritability. Analyses were repeated for individual neurodevelopmental problems. RESULTS: Childhood neurodevelopmental difficulties were associated with adolescent MDD (OR = 2.11, 95% CI = 1.24, 3.60, p = .006). Childhood irritability statistically accounted for 42% of this association. On examining each neurodevelopmental difficulty separately, autistic, communication and ADHD problems were each associated with depression, with irritability explaining 29%-51% of these links. CONCLUSIONS: Childhood irritability appears to be a key contributor to the link between childhood neurodevelopmental difficulties and adolescent MDD. High rates of irritability in children with autistic and ADHD difficulties may explain elevated rates of depression in the neurodevelopmental group.