Child Maltreatment and Abuse – A Predictor Agencies are Using Prediction to Prevent

According to 2011 data,  12.5 percent of children under the age of 18 are abused or neglected in the United States each year.  A Facts on Youth brief from the Center for Health and Justice at TASC cites a study published in JAMA Pediatrics that found confirmed maltreatment for 1 in 8 youth, with nearly 6 percent of cases  (just less than half of confirmed reports) involving children ages 5 and under.  The brief also notes that studies of child abuse and maltreatment that rely on self-reports rather than substantiated reports indicate a rate of up to 40 percent.

The Child Trends brief Preventing Violence: Understanding and addressing determinants of youth violence in the United States reviewed relevant research on interventions and policy approaches to reducing youth violence, with an emphasis on individual, family and school/community factors.  This review identified several predictors of violence, including domestic violence, dysfunctional parenting, gun availability, low self-control, and lack of connectedness to school.  Child maltreatment, however, was a strong predictor of nearly every type of violence.  The prevention of child abuse and provision of interventions to address the impact of such trauma appear to be critical actions in reducing the potential of future violence.  That said, although child maltreatment is a risk factor for criminal behavior, the longer term negative effects of that experience may be offset or amplified by other life events. Completing high school/getting a GED and getting married were two factors identified by a research team at the Social Development Research Group at the University of Washington as having a positive impact on a person’s life, thus reducing the power of the relationship between the maltreatment and future high risk behaviors.  A history of maltreatment combined with additional risk factors, such as poverty, increases the likelihood of criminal behavior.

As safety and health are essential factors in optimal child development, and may affect a multitude of life outcomes, new strategies have emerged to better identify and “triage” high-risk situations.  States are turning to the big data playbook to assist in investigations of abuse and maltreatment, using predictive analysis to help prioritize reports and better provide preventive services. Information such as family history, school reports and other administrative data, plus case officer knowledge, gives child welfare decision-makers more (if not necessarily better) data to guide the use of resources for the protection of children. Along with Connecticut, Florida, and Los Angeles County, Allegheny County here in western Pennsylvania is utilizing predictive analytics in an effort to reduce child maltreatment, abuse, and fatalities.  For more information on how predictive analysis is being used in child welfare, see Who will Seize the Child Abuse Prediction Market by Darian Woods and Checklists, Big Data and the Virtues of Human Judgement by Holden Slattery, both in The Chronicle of Social Change.