Profiles of Maltreatment Among Foster Youth Making the Transition to Adulthood

Background and Purpose:     Little is known about the maltreatment experiences of youth who age out of substitute care.   Their late age at entry and failed attempts at reunification may nevertheless mean prolonged exposure to child maltreatment.   Evidence from longitudinal investigations measuring outcomes in adulthood find they are twice as likely as same aged peers in the general population to have a lifetime diagnosis of PTSD (Keller et al., 2010; McMillen et al., 2005; Pecora et al., 2005) even though they are provided with mental health services at rates greater than other maltreated children (Garland et al., 1996; McMillen et al., 2004).  For these reasons, greater understanding of maltreatment experiences may offer critical information.  This study applies person-oriented methods to a large representative sample of youth making the transition to adulthood in one state. 

Methods:            This study involves a secondary data analysis of state child welfare administrative data.  A latent variable modeling approach is used to generate profiles of maltreatment experiences based on maltreatment reports (McCutcheon, 1987).  A cohort of foster youth who reached the age of majority is selected from state administrative data based upon the following criteria: 1) In out-of-home care at age 17; 2) In care for at least 1-year; 3) Turned 18-years in out-of-home care; and 4) Entered due to child abuse/neglect.  There are 801 foster youth that meet sample selection criteria, and the observation period is from FY1989-2011.  All unduplicated maltreatment records are included in the analysis, including substantiated and unsubstantiated findings, and records occurring prior to and during any placement spell.  Four categorical indicators of maltreatment are included in the LCA: # of different maltreatment types (1-7); Predominant type (1-3); Chronicity (1-5 developmental periods); and Number of different perpetrators (1-9).  Classes are differentiated on variables not used in the classification process.

Results:                The average age at first allegation is 4.4 years.  In total, the young people in this sample were the subject of 11 allegations perpetrated by 3 different persons, on average.  Analysis of the model fit statistics and visual inspection identify a 5-class model as the best fitting solution.  Class 1 (30%) is differentiated from the other classes by chronic maltreatment involving multiple types, predominantly neglect.  Class 2 (19%) is differentiated from the other classes by high and approximately equivalent types of maltreatment, with moderate chronicity.  Class 3 (28%) is differentiated by predominant combination of neglect and physical abuse.   Class 4 (12%) is differentiated by chronic neglect.  Class 5 (10%) is differentiated by predominant neglect in combination with sexual abuse with low chronicity.  

Conclusions and Implications:     Study findings highlight the heterogeneity of maltreatment experiences in the lives of foster youth emancipating to adulthood from one state child welfare system.  Study findings draw attention to the need to raise awareness among service providers about childhood maltreatment in older youth; to enhance research on the multiple dimensionality of maltreatment; and to develop best practices for promoting the safety of older adolescents who are without permanence.