Fatality-Driven Child Protection Systems: Helpful or Harmful?



Fatality-Driven Child Protection Systems: Helpful or Harmful?

Raelene Freitag, PhD, Director of Social Service Practice

When parents kill their own children, we can scarcely imagine it. It’s no less shocking when a child is murdered by an intimate partner of a parent. The sad fact that this ever happens is an outrage to our fundamental commitment to keep children safe from harm. So when it does happen, we want something to be done. We think, someone should have prevented this! Someone should be fired! Something should be fixed! We need a new law, a new regulation! We need new technology to spot these frightening situations before they happen again! We need to remove children from dangerous parents! 

All of these prescriptions for solutions are understandable. Sometimes any or all of them are needed. However, if we reflexively respond to a child fatality or near-fatality by introducing laws, regulations, technologies, or personnel actions without careful and balanced review, we may do little to prevent a future fatality and may, in fact, cause immeasurable non-fatal harm to thousands of children and their families. 

Deciding how the child protection system (CPS) should change (or not) in response to child fatalities requires understanding what the CPS system does most of the time. In 2013 (the last full year official statistics for the nation are available), nearly 3.5 million children were referred to the nation’s CPS systems. That same year, the estimated number of child fatalities from abuse or neglect was 1,520. Thus, fatalities represent 0.0004% of the referrals for which CPS systems are responsible. This percentage would be even lower if limited to families who were known to CPS systems before the referral related to the fatality.1 

When a system that spends 99.999% of its time working with families where there are no fatalities gets overly focused on preventing fatalities, can that system do the work it does most of the time well? A fatality-driven system can become so fear-based and risk-averse that it relies too heavily on removing children; creates complex, picayune, and even conflicting policies and procedures; and charges workers with viewing all families through a lens of preventing a very rare event.  

While many pressures can push systems toward being fatality driven, recent conversations around predictive analytics may tip the balance. While there is great potential for predictive analytics to add value to child protection work, the conversation seems focused on the potential in using predictive analytics to find the families most likely to have fatal or near-fatal events.

In context, this idea has great promise and should be pursued. But we should feel worried when policymakers begin to view the potential to identify families most likely to have fatal or near-fatal events as THE solution for child protection. The gravitational pull on child protection work toward preventing fatalities may have the unintended consequence of undermining the work with the vast majority of families who are never in danger of killing their children, but for whom a lack of skilled and resourced intervention will cause slow but predictable life-long harm and adverse outcomes. 

By all means, we should explore whether we can improve our ability to predict families most likely to have the most serious outcomes. Existing actuarial risk assessments do about as well at this task as tools to predict who will have a heart attack.2 That’s pretty good information to help workers focus their attention on the highest-risk families.  

The biggest need in child protection systems, however, is not marginally improving predictive ability. Rather, child protection systems need to be properly resourced to effectively intervene with the families we already know are at high risk. Systems should be designed so workers can spend the time they need with these families, and have the skills to engage these families in a change process so that the parents can keep their children safe. Systems need to form a culture in which workers are engaged in this process as change agents. Systems need to be designed around equipping workers with the tools, time, and resources to help the highest risk families become safe, so that the 99.999% of children in the CPS system get the help they need to thrive.

1 US Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth, and Families, Children's Bureau. (2015). Child maltreatment 2013. Available from http://www.acf.hhs.gov/programs/cb/research-data-technology/statistics-…

2 Johnson, W., Clancy, T., & Bastian, P. (2015). Child abuse/neglect risk assessment under field practice conditions: Tests of external and temporal validity and comparison with heart disease prediction. Children and Youth Services Review, 56, 76–85.


Submitted by Visitor on September 21, 2015 - 7:05am.

Brilliantly put!

Submitted by Visitor on September 22, 2015 - 6:54pm.

Timely advice as we here in Massachusetts respond to several recent fatalities and near-fatalities of children and reconsider the value of several recent policy and practice initiatives such as differential response and use of structured decision-making tools for assessing safety/danger and risk. Resources are key!

Submitted by Visitor on September 24, 2015 - 5:04pm.

I agree that the linkage s from Child Welfare and child fatalities is an adverse relationship. I think that another approach is twofold, child fatalities in care and child fatalities within proximity (investigations within 30 days, alleging physical abuse and prior dependencies, should be assessed as to any association with child fatalities An interesting side analysis could be assessing the relationship between.the perpetrator and the child, as I believe that males in the home are the leading cause of fatalities in child homicide. Of course, fatalities associated with neglect are a different matter and the analysis would have to involve many more factors, such as mental status of the parents,physical condition of the child , per-exixsting medical conditions. etc.

Add new comment

Filtered HTML

  • Allowed HTML tags: <a href hreflang> <img src alt height width> <em> <strong> <span> <cite> <code> <ul type> <ol start type> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Enter the characters shown in the image.