Testing the Utility of Different Conceptualizations of Childhood Adversity to Predict Mental Health Symptoms Among Child Welfare-Involved Adolescents: A Latent Trajectory Model

Background and Purpose: Child welfare-involved youth are at increased risk for exposure to childhood adversity and subsequent mental health problems. Few studies have compared different conceptualizations of childhood adversity among child welfare-involved youth, and few studies have examined whether the association between adversity and mental health varies across racial/ethnic groups. The current study had two aims: (a) to compare the individual risk, cumulative risk, and dimensional model of adversity and psychopathology (DMAP) approaches based on their ability to predict mental health symptom trajectories and (b) to examine whether those associations vary across racial/ethnic groups.

Methods: This study included a nationally representative sample of child welfare-involved adolescents (n = 1,011; ages 11-17 years) from the National Survey of Child and Adolescent Well-Being study. Data was collected over three time-points from youth, caregivers, and caseworkers. Childhood adversity was assessed using nine-items that corresponded to the Adverse Childhood Experiences (ACEs) Study measure, such as physical abuse and household substance abuse. A total score was created for the cumulative risk approach and DMAP dimensions of threat (i.e., exposure to violence) and deprivation (i.e., lack of physical, emotional, cognitive stimuli). Internalizing, externalizing, and post-traumatic stress symptoms were assessed using t-scores from the Youth Self-Report and the Trauma Symptom Checklist for Children. Separate latent growth curve models were used to examine associations between childhood adversity and mental health symptom trajectories, adjusting for covariates. Differences across racial/ethnic groups were examined using multiple group models.

Results: Across all three mental health symptom trajectories, intercept values (i.e., means) were below the clinical cut-off (<65) and symptoms decreased over time in the full sample. The individual risk approach did not perform significantly better than the cumulative risk approach (X² ∆ p-values >.05). The cumulative risk score was significantly associated with the intercept factor of internalizing (β=0.18, p=.022) and post-traumatic stress symptoms (β=0.19, p=.012). Including dimensions of threat and deprivation in the model provided more nuanced findings, in that exposure to deprivation, but not threat, was associated with the intercept factor of post-traumatic stress symptoms (β=0.17, p=.019). Neither the cumulative risk score or threat and deprivation were associated with the rate of change in mental health symptoms. Additionally, there were no significant differences in associations across racial/ethnic groups.

Conclusions and Implications: The findings highlight the importance of critically examining how childhood adversity is assessed and conceptualized in research studies as this may have significant implications for how findings are interpreted and translated into practice and policy. This study found that accounting for individual types of adversity did not perform better than a cumulative score, and the cumulative score may have masked unique associations between childhood adversity and mental health symptom trajectories. Findings also suggest that there were not significant racial/ethnic differences in the association between childhood adversity and mental health symptom trajectories. This study emphasizes the need for future research that continues to examine the best ways to assess the impact of childhood adversity on mental health to inform child welfare practice and interventions to promote positive developmental outcomes for youth.