“ACEs”-Informed Subgroups Among Youth in the “Pipeline to Crime”: A Latent Class Analysis

Background and Purpose: Juvenile crime and delinquency remains a serious problem in the United States, with approximately 1.5 million youth arrested annually (OJJDP, 2013). The juvenile justice system that responds to these youth is currently undergoing a sea change, in part because of emerging research about neurobiological development limitations of adolescents. Among current reforms is a pushback against using “one-size fits all” approaches to delinquency (Springer & Roberts, 2010); however there is little guidance to practitioners about the diverse needs and backgrounds of youth (Reid & Sullivan, 2009). This study uses the adverse childhood experiences (“ACEs”) framework as a means of understanding delinquent and maladaptive behaviors within the context of the neurophysiological effects of accumulated trauma and adversity (Duke et al., 2010). We hypothesize the existence of subgroups among these youth on the basis of recognized ACE factors in conjunction with social disadvantage. Furthermore, these groups are hypothesized to differ on the basis of risk and resilience factors that may provide guidance for tailored interventions within the courts.

Methods: Psychosocial assessment data from the Washington State Juvenile Court Assessment (used in 16 states) was linked with court records for moderate- to high-risk youth (n=4290) adjudicated to probation in an urban, diverse Western region during an 8-year period. Latent Class Analysis was run with sixteen indicators of childhood adversity: victimization (neglect, emotional, sexual, physical abuse), non-violent adversity (parental substance use, parental health problems, family incarceration, out-of-home placements), and social disadvantage (family poverty, parental unemployment, lack of health insurance, homelessness). Resulting classes were then tested for differences in 1) distribution of social demographics characteristics and 2) additional psychosocial assessment factors germane to interventions.

Results: Examination of multiple fit statistics indicated that a six-class model best fit the data. A 1) Low-All group (35.6% of the sample) reported comparatively lower levels of assessed adversities. Other groups were characterized by varying levels of exposure to different types of ACEs: 2) high Parental Substance Use and Incarceration (8.6%), 3) Family Poverty and Parental Ill Health (17.4%), 4) high Family Violence Witnessing with Lower levels of Poverty (17.6%), 5) extremely high Maltreatment and Out-of-Home placements (10.2%), and 6) very high levels of All Adversities (10.6%). Between group tests provided insights regarding race/ethnicity and gender distributions as well as differences at baseline in risk and protective factors that bear upon intervention planning (e.g., delinquency histories, aggression, mental health, and protective skills).  

Conclusion/Implications: These results provide powerful evidence of the value of assessing multi-form adversity and trauma for court-involved youth (Greenwald, 2006). The six groups represent distinct histories of adversity and family context, evincing widely varying needs for interventions. Whereas some youth needs indicate targeted interventions (e.g., extreme family poverty and parental illness versus familial patterns of substance abuse), others will require multi-faceted interventions to address traumatic backgrounds, low resources, and patterned behavioral problems. Effective interventions that address the needs of youth exposed to multiple forms of adversities as well as trauma-informed probationary practices are critical in providing resilience resources and preventing “pipeline to prison” pathway.