Classes of Maltreatment in a Sample of Foster Youth

Purpose: Studies examining the child welfare population have found that the vast majority of system-involved children experience more than one form of parental maltreatment in their homes of origin (Lau et al, 2005; Pears, Kim & Fisher, 2008). Despite this, the majority of studies concerning the well being of this population focus on evaluating the unique effects of one form of maltreatment without considering how those effects may be moderated by other forms of co-occurring maltreatment. This can lead to inaccurate conclusions about the relationship between early maltreatment experiences and subsequent developmental outcomes. Effective intervention and program development depends on a thorough evaluation of multiple dimensions of maltreatment. A person-oriented approach allows for the interaction of various maltreatment types. In this study, we employ latent class analysis (LCA) in an effort to explore the co-occurring victimization of foster youth who participated in the Midwest Evaluation of the Adult Functioning of Former Foster Youth (Midwest Study).

Methods: Our sample includes 732 17-year-old foster youth living in three midwestern states. LCA models were fit using Mplus Version 7. The program evaluated indicators of maltreatment to establish the maximum likelihood of each case falling into a particular subgroup or class. The indicators of class were: (1) seven questions relating to physical abuse before care, (2) nine questions relating to neglect before care; and (3) two questions relating to sexual abuse before care. The questions relating to maltreatment varied in severity.

Results: A three-class solution was selected as the best fitting model based on BIC, entropy and LMRT. 58% of the sample was placed in Class 1. This class showed the least frequent and severe maltreatment experiences, with the majority indicating one or two types of neglect. 12% of the sample was placed in Class 2. This class showed the most extreme experience of co-occurring maltreatment; these youths were most likely to report being seriously physically maltreated and neglected by caregivers. Specifically, they were most likely to report that their caretaker/s threw, shoved, hit or kicked them, that they went without things they needed and that they missed school to care for someone else. 30% of the sample was placed in Class 3. This class was middle range: slightly more likely to indicate physical abuse and neglect than those in Class 1 but not as likely as those in Class 2. Sex abuse was consistent among all 3 classes. Boys and girls were equally likely to be placed in any class.

Implications: Considering children who experience multiple or severe forms of maltreatment tend to fare the worst (Finkelhor et al., 2007; Ney, Fung & Wickett, 1994), it is useful to classify the experiences of youth in foster care based on type of maltreatment and severity of their maltreatment experiences.  The classification of these experiences provides those working in the field of child welfare with greater information about the likelihood of co-occurring maltreatment. This knowledge enables the provision of more tailored and effective interventions.