Machine learning models for diagnosis and risk prediction in eating disorders, depression, and alcohol use disorder.

This study uses machine learning models to uncover diagnostic and risk prediction markers for eating disorders (EDs), major depressive disorder (MDD), and alcohol use disorder (AUD). Utilizing case-control samples (ages 18-25 years) and a longitudinal population-based sample (n=1,851), the models, incorporating diverse data domains, achieved high accuracy in classifying EDs, MDD, and AUD from healthy controls. The area under the receiver operating characteristic curves (AUC-ROC [95% CI]) reached 0.92 [0.86-0.97] for AN and 0.91 [0.85-0.96] for BN, without relying on body mass index as a predictor. The classification accuracies for MDD (0.91 [0.88-0.94]) and AUD (0.80 [0.74-0.85]) were also high. Each data domain emerged as accurate classifiers individually, with personality distinguishing AN, BN, and their controls with AUC-ROCs ranging from 0.77 to 0.89. The models demonstrated high transdiagnostic potential, as those trained for EDs were also accurate in classifying AUD and MDD from healthy controls, and vice versa (AUC-ROCs, 0.75-0.93). Shared predictors, such as neuroticism, hopelessness, and symptoms of attention-deficit/hyperactivity disorder, were identified as reliable classifiers. For risk prediction in the longitudinal population sample, the models exhibited moderate performance (AUC-ROCs, 0.64-0.71), highlighting the potential of combining multi-domain data for precise diagnostic and risk prediction applications in psychiatry.

Predicting Depression Onset in Young People Based on Clinical, Cognitive, Environmental, and Neurobiological Data.

BACKGROUND: Adolescent onset of depression is associated with long-lasting negative consequences. Identifying adolescents at risk for developing depression would enable the monitoring of risk factors and the development of early intervention strategies. Using machine learning to combine several risk factors from multiple modalities might allow prediction of depression onset at the individual level. METHODS: A subsample of a multisite longitudinal study in adolescents, the IMAGEN study, was used to predict future (subthreshold) major depressive disorder onset in healthy adolescents. Based on 2-year and 5-year follow-up data, participants were grouped into the following: 1) those developing a diagnosis of major depressive disorder or subthreshold major depressive disorder and 2) healthy control subjects. Baseline measurements of 145 variables from different modalities (clinical, cognitive, environmental, and structural magnetic resonance imaging) at age 14 years were used as input to penalized logistic regression (with different levels of penalization) to predict depression onset in a training dataset (n = 407). The features contributing the highest to the prediction were validated in an independent hold-out sample (three independent IMAGEN sites; n = 137). RESULTS: The area under the receiver operating characteristic curve for predicting depression onset ranged between 0.70 and 0.72 in the training dataset. Baseline severity of depressive symptoms, female sex, neuroticism, stressful life events, and surface area of the supramarginal gyrus contributed most to the predictive model and predicted onset of depression, with an area under the receiver operating characteristic curve between 0.68 and 0.72 in the independent validation sample. CONCLUSIONS: This study showed that depression onset in adolescents can be predicted based on a combination multimodal data of clinical characteristics, life events, personality traits, and brain structure variables.

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.

Association of Genetic and Phenotypic Assessments With Onset of Disordered Eating Behaviors and Comorbid Mental Health Problems Among Adolescents.

Importance: Eating disorders are serious mental disorders with increasing prevalence. Without early identification and treatment, eating disorders may run a long-term course. Objective: To characterize any associations among disordered eating behaviors (DEBs) and other mental health disorders and to identify early associations with the development of symptoms over time. Design, Setting, and Participants: This multicenter, population-based, longitudinal cohort study used data from baseline (collected in 2010), follow-up 1 (collected in 2012), and follow-up 2 (collected in 2015) of the IMAGEN Study, which included adolescents recruited from 8 European sites. The present study assessed data from 1623 healthy adolescents, aged 14 years at baseline, recruited from high schools. Data analyses were performed from January 2018 to September 2019. Main Outcomes and Measures: Body mass index (BMI), mental health symptoms, substance use behaviors, and personality variables were investigated as time-varying associations of DEBs (dieting, binge eating, and purging) or change in BMI over time. Polygenic risk scores were calculated to investigate genetic contributions associated with BMI, attention-deficit/hyperactivity disorder (ADHD) and neuroticism to DEBs. Results: In this cohort study of 1623 adolescents (829 girls [51.1%]) recruited at a mean (SD) age of 14.5 (0.4) years and followed up at ages 16 and 19 years, 278 adolescents (17.1%) reported binge eating, 334 adolescents (20.6%) reported purging, and 356 adolescents (21.9%) reported dieting at 14, 16, or 19 years. Among the precursors of DEBs, high BMI was associated with future dieting (OR, 3.44; 95% CI, 2.09-5.65). High levels of neuroticism (OR, 1.04; 95% CI, 1.01-1.06), conduct problems (OR, 1.41; 95% CI, 1.17-1.69), and deliberate self-harm (OR, 2.18; 95% CI, 1.37-3.45) were associated with future binge eating. Low agreeableness (OR, 0.95; 95% CI, 0.92-0.97), deliberate self-harm (OR, 2.59; 95% CI, 1.69-3.95), conduct problems (OR, 1.42; 95% CI, 1.20-1.68), alcohol misuse (OR, 1.31; 95% CI, 1.10-1.54), and drug abuse (OR, 2.91; 95% CI, 1.78-4.74) were associated with future purging. Polygenetic risk scores for BMI were associated with dieting (at 14 years: OR, 1.27; lower bound 95% CI, 1.08; at 16 years: OR, 1.38; lower bound 95% CI, 1.17); ADHD, with purging (at 16 years: OR, 1.25; lower bound 95% CI, 1.08; at 19 years, OR, 1.23; lower bound 95% CI, 1.06); and neuroticism, with binge eating (at 14 years: OR, 1.32; lower bound 95% CI, 1.11; at 16 years: OR, 1.24; lower bound 95% CI, 1.06), highlighting distinct etiologic overlaps between these traits. The DEBs predated other mental health problems, with dieting at 14 years associated with future symptoms of depression (OR, 2.53; 95% CI, 1.56-4.10), generalized anxiety (OR, 2.27; 95% CI, 1.14-4.51), deliberate self-harm (OR, 2.10; 95% CI, 1.51-4.24), emotional problems (OR, 1.24; 95% CI, 1.08-1.43), and smoking (OR, 2.16; 95% CI, 1.36-3.48). Purging at 14 years was also associated with future depression (OR, 2.87; 95% CI, 1.69-5.01) and anxiety (OR, 2.48; 95% CI, 1.49-4.12) symptoms. Conclusions and Relevance: The findings of this study delineate temporal associations and shared etiologies among DEBs and other mental health disorders and emphasize the potential of genetic and phenotypical assessments of obesity, behavioral disorders, and neuroticism to improve early and differential diagnosis of eating disorders.

Cognitive and brain development is independently influenced by socioeconomic status and polygenic scores for educational attainment.

Genetic factors and socioeconomic status (SES) inequalities play a large role in educational attainment, and both have been associated with variations in brain structure and cognition. However, genetics and SES are correlated, and no prior study has assessed their neural associations independently. Here we used a polygenic score for educational attainment (EduYears-PGS), as well as SES, in a longitudinal study of 551 adolescents to tease apart genetic and environmental associations with brain development and cognition. Subjects received a structural MRI scan at ages 14 and 19. At both time points, they performed three working memory (WM) tasks. SES and EduYears-PGS were correlated (r = 0.27) and had both common and independent associations with brain structure and cognition. Specifically, lower SES was related to less total cortical surface area and lower WM. EduYears-PGS was also related to total cortical surface area, but in addition had a regional association with surface area in the right parietal lobe, a region related to nonverbal cognitive functions, including mathematics, spatial cognition, and WM. SES, but not EduYears-PGS, was related to a change in total cortical surface area from age 14 to 19. This study demonstrates a regional association of EduYears-PGS and the independent prediction of SES with cognitive function and brain development. It suggests that the SES inequalities, in particular parental education, are related to global aspects of cortical development, and exert a persistent influence on brain development during adolescence.