Predicting childhood and adolescent attention-deficit/hyperactivity disorder onset: a nationwide deep learning approach

Abstract

Attention-deficit/hyperactivity disorder (ADHD) is a heterogeneous disorder with a high degree of psychiatric and physical comorbidity, which complicates its diagnosis in childhood and adolescence. We analyzed registry data from 238,696 persons born and living in Sweden between 1995 and 1999. Several machine learning techniques were used to assess the ability of registry data to inform the diagnosis of ADHD in childhood and adolescence: logistic regression, random Forest, gradient boosting, XGBoost, penalized logistic regression, deep neural network (DNN), and ensemble models. The best fitting model was the DNN, achieving an area under the receiver operating characteristic curve of 0.75, 95% CI (0.74–0.76) and balanced accuracy of 0.69. At the 0.45 probability threshold, sensitivity was 71.66% and specificity was 65.0%. There was an overall agreement in the feature importance among all models (τ > .5). The top 5 features contributing to classification were having a parent with criminal convictions, male sex, having a relative with ADHD, number of academic subjects failed, and speech/learning disabilities. A DNN model predicting childhood and adolescent ADHD trained exclusively on Swedish register data achieved good discrimination. If replicated and validated in an external sample, and proven to be cost-effective, this model could be used to alert clinicians to individuals who ought to be screened for ADHD and to aid clinicians’ decision-making with the goal of decreasing misdiagnoses. Further research is needed to validate results in different populations and to incorporate new predictors.

The experience of Thai adolescents with depression: A qualitative study

“…Unfortunately, approximately 60% of adolescents with major depressive disorder have at least one comorbid mental health diagnosis, most commonly anxiety, attention‐deficit hyperactivity disorder (ADHD) (Garcia‐Argibay et al., 2024), conduct disorder, substance use disorders and somatic disorders (Mullen, 2018). Adolescents with chronic medical conditions (e.g., chronic pain, neurological disorders, and autoimmune or inflammatory diseases) also have higher levels of comorbid depression than healthy adolescents (Garcia‐Argibay et al., 2024; Korczak et al., 2023). Among these adolescents, symptoms such as fatigue, decreased concentration, sleep problem and appetite disturbance may overlap with feature of depression, making the diagnosis challenging (Korczak et al., 2023).…”

Predicting new onset thought disorder in early adolescence with optimized deep learning implicates environmental-putamen interactions

“…Outside the US, national registries or school system data may be available offering large sample sizes (n>10,000) but these typically lack physiologic, psychometric and TD information. (60)(61)(62)(63) However, in recent years data has started to appear from the landmark Adolescent Brain Cognitive Development (ABCD) study. Sponsored by NIH, this is the largest long-term study of child and adolescent brain development and health in the US.…”

The ADHD Phenotype in Black and White Girls From Childhood to Adolescence: Results From the Community-Based Pittsburgh Girls Study

“…There are three subtypes of ADHD: predominantly inattentive, predominantly hyperactive/impulsive, and a combined inattention and hyperactivity/impulsivity subtype. ADHD typically has an onset in childhood, is persistent (Van Meter et al, 2023), and is associated with an increased risk for depression (Garcia-Argibay et al, 2023), suicidality (Austgulen et al, 2023), and substance use (Mustonen et al, 2023). Thus, accurate diagnostic phenotyping of ADHD may reduce the risk of persistent behavioral and emotional problems across the lifespan via early identification, deployment of preventive interventions, and surveillance regarding symptom exacerbation.…”