Demographic and mental health assessments in the adolescent brain and cognitive development study: Updates and age-related trajectories

Abstract

The Adolescent Brain Cognitive Development (ABCD) Study of 11,880 youth incorporates a comprehensive range of measures assessing predictors and outcomes related to mental health across childhood and adolescence in participating youth, as well as information about family mental health history. We have previously described the logic and content of the mental health assessment battery at Baseline and 1-year follow-up. Here, we describe changes to that battery and issues and clarifications that have emerged, as well as additions to the mental health battery at the 2-, 3-, 4-, and 5-year follow-ups. We capitalize on the recent release of longitudinal data for caregiver and youth report of mental health data to evaluate trajectories of dimensions of psychopathology as a function of demographic factors. For both caregiver and self-reported mental health symptoms, males showed age-related decreases in internalizing and externalizing symptoms, while females showed an increase in internalizing symptoms with age. Multiple indicators of socioeconomic status (caregiver education, family income, financial adversity, neighborhood poverty) accounted for unique variance in both caregiver and youth-reported externalizing and internalizing symptoms. These data highlight the importance of examining developmental trajectories of mental health as a function of key factors such as sex and socioeconomic environment.

Keywords: Assessment; Longitudinal assessment; Mental health; Psychopathology.

Fig. 1

Sex differences in trajectories of youth reported mental health using raw data: graphs illustrating sex differences in youth-reported Total Problem, Internalizing, Externalizing and Attention Problems on the Brief Problem Monitor. The shaded areas indicate the 99th percentile confidence intervals around the estimated linear slope. Graphs were created in R using ggplot2 version 3.3.2 using one of the five imputed datasets and present the “raw” data. The analogous graph created using ggpredict plotting the relationships to sex using estimated marginal means controlling for all other factors in the model is shown in Fig. S1.

Fig. 2

Sex differences in trajectories of caregiver reported mental health summary scores using raw data: graphs illustrating sex differences in caregiver-reported Total Problem, Internalizing, Externalizing on the Child Behavior Checklist. The shaded areas indicate the 99th percentile confidence intervals around the estimated linear slope. Graphs were created in R using ggplot2 version 3.3.2 using one of the five imputed datasets and present the “raw” data. The analogous graph created using ggpredict plotting the relationships to sex using estimated marginal means controlling for all other factors in the model is shown in Fig. S6.

Fig. 3

Caretaker education differences in trajectories of caregiver reported mental health summary scores using raw data: graphs illustrating differences in caregiver-reported Total Problem, Internalizing, Externalizing on the Child Behavior Checklist as a function of caretaker education. The shaded areas indicate the 99th percentile confidence intervals around the estimated linear slope. Graphs were created in R using ggplot2 version 3.3.2 using one of the five imputed datasets and present the “raw” data. The analogous graph created using ggpredict plotting the relationships to caretaker education using estimated marginal means controlling for all other factors in the model is shown in Fig. S9.