A Doubly Robust Machine Learning Procedure to Estimate Disparities by Race/Ethnicity and Sex in the Relationship...

Suicide is a leading cause of death among youth, and adverse childhood experiences (ACEs) are established risk factors for suicidality. This study is the first to investigate disparities in the cumulative impact of ACEs in individuals with suicidal ideation only (SI+) and suicide attempt (SA+) in a current, diverse, and nationally representative sample using a causal inference analysis framework.

Using the 2023 National Youth Risk Behavior Survey (YRBS) (N = 20,103), adolescent demographics, suicidal thoughts and behaviors (STBs), ACEs, and confounders (eg, bullied at school) were examined. First, bivariate differences were analyzed by race/ethnicity and sex for ACE prevalence. Next, the dose-response relationship of ACEs on suicidality was estimated with a doubly robust estimation process and tested for moderation by race/ethnicity and sex.

There were significant disparities by race/ethnicity in the types of ACEs that participants endorsed. Female adolescents saw higher prevalences across most ACEs compared to their male counterparts. For SI+ and SA+ outcomes, a dose-response effect of ACEs on suicidality was observed, with higher effects for SI+ compared to SA+. For both outcomes, disparities in this effect across race/ethnicity and sex were not observed.

These results suggest that the number of ACEs has a similar incremental impact on the risk of youth suicidality across races/ethnicities and sex despite differences in the types of ACEs experienced (ie, emotional abuse vs physical abuse). It is critical that suicide intervention and prevention strategies better support youth experiencing ACEs; in addition, they should consider differences in prevalences of ACE types by demographics for effective risk mitigation.

We worked to ensure sex and gender balance in the recruitment of human participants. We worked to ensure race, ethnic, and/or other types of diversity in the recruitment of human participants. We worked to ensure that the study questionnaires were prepared in an inclusive way. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented sexual and/or gender groups in science. We actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our author group.