Evaluating and Validating the Classification Accuracy of a Screening Instrument to Assess Risk for Commercial Sexual Exploitation of Child Welfare–Involved Children and Adolescents

Abstract

Objectives: Identifying children and adolescents within child welfare at risk for commercial sexual exploitation (CSE) can ensure referrals to appropriate services. However, screening measures to understand the prevalence of CSE are missing in child welfare. We evaluated the classification accuracy of a screener developed for the purpose of this study, guided by the Sexual Exploitation among Youth (SEY) risk assessment framework used in practice with child welfare–involved young people, (1) to identify young people at high versus low risk for experiencing CSE and (2) to estimate the prevalence of CSE risk for child welfare–involved children and adolescents. Methods: We used extant data from the National Survey of Child and Adolescent Well-being study with a nationally representative sample of children and adolescents aged 11-17 years (n = 1054) investigated by child welfare from February 2008 to April 2009. The 26-item screener showed acceptable reliability (α = .73) and test-criterion validity evidence using a CSE proxy outcome (ie, narrowly defined as being paid for sexual relations). We used the receiver-operating curve to classify risk and calculate the optimal cutoff score. Results: Higher scores on the SEY screener (range, 0-20 points) increased the odds of experiencing CSE by 34%. The screener was good at discriminating CSE risk at the 6-point cutoff, with 26.7% of child welfare–involved young people identified as being at high risk for CSE. Conclusions: Given the absence of accurate prevalence rates of CSE risk in the population, a theoretical cutoff index using an established method can provide an objective decision on how to distinguish risk levels. Prevalence estimates for CSE risk highlight the need for systematic screening in child welfare to identify and provide services for young people at risk.