Skip to main content
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

Childhood Predictors and Moderators of Lifetime Risk of Self-Harm in Girls with and without Attention Deficit Hyperactivity Disorder


Objective: Attention-deficit/hyperactivity disorder (ADHD)—especially the underlying symptom dimension of hyperactivity/impulsivity—is associated with self-harm, including both non-suicidal self-injury and suicidal behavior, during adolescence and young adulthood. This link is especially strong for females. Yet little is known about the developmental trajectories, or childhood predictors and moderators, of self-harm in women with and without childhood histories of ADHD.

Method: Utilizing data from the Berkeley Girls with ADHD Longitudinal Study (BGALS; N = 228), I first characterized lifetime risk of self-harm, including non-suicidal self-injury (NSSI), suicidal ideation (SI), and suicide attempts (SA), in adolescent and young adult women, comparing those with (n = 140) and without (n = 88) childhood histories of ADHD. At Wave 3 (M age = 19.6, range 17-24), self-harm was assessed via two self-reported measures; at Wave 4 (M age = 25.56, range 21-29), it was assessed via a clinician-administered structured interview. I then examined childhood predictors and moderators of lifetime risk by utilizing measures ascertained at baseline assessment in childhood (Wave 1; M age = 9.6, range 6-12). The focus was on six predictor domains of interest: early psychopathology, adverse experiences (retrospectively recalled), peer rejection, measure of overall executive functioning, negative parenting practices (especially related to fathers), and child self-esteem, each of which has been found to be (a) impaired in samples with ADHD and (b) a risk factor for self-harm.

Results: First, regarding developmental trajectories, most participants with positive histories of lifetime NSSI engaged in such behaviors in adolescence (W3) yet desisted in adulthood (W4). Furthermore, nearly 60% of the girls endorsing a positive history of NSSI or SI by late adolescence were in the ADHD-C group. Interestingly, the vast majority of SA occurred during adolescence, with few new attempts during young adulthood; Attempters into the period of young adulthood (i.e., “persisters”) all emanated from the ADHD-C subgroup, revealing the importance of impulsivity for repeated SA. Second, using binary logistic regressions, I found that early externalizing symptoms, overall executive functioning, and father’s negative parenting (reported by the participant) predicted lifetime risk of NSSI, over and above sociodemographic covariates. Adverse childhood experiences and low self-esteem predicted lifetime SI; early externalizing behaviors, adverse childhood experiences, and low self-esteem predicted lifetime SA. Finally, to determine interactions between individual risk factors, I conducted receiver operator characteristics (ROC) analyses. Results revealed that parent rated externalizing symptoms predicted NSSI, qualified by internalizing symptoms and executive functioning measures, such that girls with high childhood externalizing and internalizing symptoms, plus poor executive functioning, had a lifetime NSSI history of 80%. For SI, self-esteem was a predictor, moderated by poor executive functioning, such that girls with low self-esteem and poor executive functioning had an 83% lifetime history of SI. Finally, moderate externalizing symptoms predicted lifetime history of SA and this was moderated by low childhood self-esteem.

Conclusions: Findings indicate that pathways to self-harm are multifaceted for females with ADHD. Understanding early childhood predictors and moderators of self-harm can inform both intervention strategies and help clinicians understand which domains to consider during risk assessment. Crucially, utilizing a multidimensional approach to understanding risk should be prioritized in suicide prevention efforts globally.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View