Predictors of Identity Fraud Victimization Among a Population-Based Cohort of Adolescent Youth in Foster Care

Background and Purpose: Identity fraud—or the misuse of a victim’s personal data by perpetrators seeking financial gain—poses serious risks and challenges for youth in foster care. As young people move through various out-of-home placements and schools, their sensitive personal data are often widely disseminated and difficult to secure. When they transition from foster care to adulthood, identity fraud victims with compromised credit histories may be denied loans, employment, and/or stable and affordable housing, adding barriers to the achievement of key developmental tasks during the already-precarious period of emerging adulthood.

To discover and resolve cases of child identity fraud, state child welfare agencies are required by federal law to conduct annual credit checks for 14- to 17-year-olds in foster care. Although the federal credit check mandate has been in force for more than a decade, little is known about the factors contributing to risk of identity fraud victimization among foster care-involved youth, either before or after their initial placement in out-of-home care.

Methods: To address this knowledge gap, we analyze linked child welfare and consumer credit records for a population-based cohort of 4,670 adolescent youth in foster care in a mid-Atlantic state who received one or more annual credit checks from November 2015 through December 2021. We use gradient boosted machine learning methods to identify factors associated with identity fraud victimization that occurred at any point during cohort youths’ time in foster care. Our machine learning model incorporated 61 independent variables, including measures of youth and caregiver demographics, history of prior (i.e., pre-foster care placement) identity fraud victimization, home removal reasons (e.g., physical abuse, neglect), and foster care placement history (e.g., number of placements, total time in care).

Results: Annual credit checks revealed some indication of identity fraud victimization for approximately one in eight cohort youth (n = 604). Among these fraud victims, 51% had some history of victimization prior to their first foster care placement, and more than three-quarters experienced victimization at some point during their time in foster care.

Of all the predictors in our machine learning model, history of prior fraud victimization had the greatest relative influence on the likelihood of in-care victimization. Youth with no history of fraud victimization prior to their earliest foster care placement had a 6.8% chance of being victimized while in care. In contrast, youth with history of prior victimization had a 53.0% chance of in-care victimization. Other factors, such as youth and caregiver demographics and measures of foster care involvement, had considerably less relative influence. Overall, the predictors in our model explained 19% of variance in the outcome.

Conclusions and Implications: Underscoring the importance of identity protection services for foster care youth, findings also highlight the need for a reevaluation of current policies and practices. Given that more than half of foster youth fraud victims had some history of victimization prior to their first foster care placement, results support an expansion of the federal credit check mandate to include all minor children in foster care, regardless of age.