Social Determinants of Health and Suicidal Behaviors among Children: U.S. Longitudinal Adolescent Brain Cognitive Development (ABCD) Study

ImportanceSocial determinants of health (SDoH) have well-characterized associations with child mental health outcomes. Their complex, multilevel influences on child mental health are less well understood. ObjectiveTo identify patterns across multiple domains of SDoH and estimate their associations with child mental health and suicidality outcomes over time. DesignPanel study of 11 810 children aged 9 to 10 years who were enrolled in the Adolescent Brain Cognitive Development (ABCD) study and followed from September 1, 2016, and April 24, 2021. SettingNationally-representative, population-based study across 21 sites in the U.S. ParticipantsABCD participants and their participating parents/guardians. Exposures84 structural SDoH factors at baseline across 9 domains (discrimination, crime and drug use, education, health and environment, family type and disability, housing and transportation, minority status and language, socioeconomic status, and urbanization). We used unsupervised machine learning to identify patterns of clustering underlying the SDoH data. Main Outcomes and MeasuresChild mental health was measured with the parent-report Child Behavior Checklist. Suicidal ideation and suicide attempts were measured with child- and parent-report computerized versions of the Kiddie Schedule for Affective Disorders and Schizophrenia. ResultsOf 10 504 children included at baseline (median [SD] age, 9.9 [0.6] years), 5510 [52.5%] were boys and 4994 [47.5%] were girls; 229 (2.2%) were Asian, 1468 (14.0%) were Black, 2128 (20.3%) were Hispanic, 5565 (53.0%) were White, and 1108 (10.5%) were multiracial. Four SDoH patterns were identified: affluence (SDoH Pattern I); structural racism and discrimination (SDoH Pattern II); socioeconomic deprivation (SDoH Pattern III); and high crime, low education, and populated (SDoH Pattern IV) areas. Children with High Socioeconomic Deprivation at baseline (SDoH Pattern III) reported higher rates of externalizing ({beta}, 1.43, 95% CI, 0.83, 2.02), internalizing ({beta}, 0.75, 95% CI, 0.14, 1.37), and total ({beta}, 1.16, 95% CI, 0.50, 1.81) problems, but these trajectories decreased over time. Children with High Structural Racism and Discrimination at baseline were the only group showing increasing trajectories of suicide attempts (OR, 1.42, 95% CI, [1.04, 1.93]) and depression ({beta}, 0.19, 95% CI, 0.08, 0.29) over time. In contrast, living in affluent communities (SDoH Pattern I) was associated with lower rates of all internalizing and externalizing problems at baseline, but increasing trajectories of depression ({beta}, 0.17, 95% CI, 0.09, 0.25), anxiety ({beta}, 0.10, 95% CI, 0.02, 0.18), and withdrawal ({beta}, 0.09, 95% CI, 0.01, 0.17) over time. Conclusions and RelevanceMultiple domains of SDoH are associated with child mental health outcomes in cross section and over time. Targeted structural interventions may improve mental health outcomes and reduce suicide attempts among children. Key PointsO_ST_ABSQuestionC_ST_ABSWhat are the social determinants of mental health, suicidal ideation, and suicidal behavior among children in the U.S.? FindingsIn this cohort of 10 504 children, we used machine learning to identify four patterns of social determinants of health (SDoH). At baseline, socioeconomic deprivation was associated with internalizing and externalizing problems. Over follow-up, structural racism and discrimination were associated with suicide attempts. MeaningMultiple dimensions of structural interventions targeting different SDoH are needed to improve child mental health outcomes.