Dee Wilson’s Sounding Board: Do Child Fatality Reviews Prevent Child Maltreatment Deaths?  

Dee Wilson’s Sounding Board: Do Child Fatality Reviews Prevent Child Maltreatment Deaths?  

first published July 2023   at      Dee Wilson

Child welfare systems, multi-disciplinary child fatality review committees child welfare scholars and child advocates have invested enormous time and resources for decades in attempts to learn from suspected child maltreatment (CM) fatalities how to prevent (or greatly reduce) deaths of children with open child welfare cases and among children without CPS reports prior to death. Nationally, these efforts have failed according National Child Abuse and Neglect Data System (NCANDS) statistics: 1570  child maltreatment fatalities (2.1 per 100,000) in  2011 vs.1820 (2.47 per 100,000) during 2021. NCANDS statistics are imperfect. A few states may not report fatalities during a fiscal year, and even when all states submit CM fatality data, it’s likely that official data is a large undercount of maltreatment fatalities for various reasons, e.g., classifying neglect related deaths as accidental. 

There is a possibility that the steady increase in child maltreatment fatalities reflected in NCANDS data is the result of improved detection and more  accurate reporting of child maltreatment deaths rather than increased incidence, if so, states that have invested in better detection and reporting of CM fatalities should show significant progress in reducing suspected CM deaths if their reform initiatives are working.

“I find no reason to believe DCYF is applying lessons from CFRs.”

There may be some states that have made notable progress in reducing CM fatalities.  If so, I am not aware of them. In Washington State, there has been little or no change in the number of annual child fatality reviews (CFR) during the past decade, an average of 15 per year. In addition, there has not been a noticeable change in the themes, findings, and recommendations of the CFRs from year to year. I find no reason to believe DCYF is applying lessons from CFRs. This is not the  fault of the reviews, most of which are balanced and insightful, though limited by the agency’s policy framework.  It appears that DCYF, like other U.S. child welfare systems, has difficulty learning from its vast experience with child protection even when the lives of children are at stake. 


Child welfare reform following child maltreatment deaths    

Until recent years, child welfare reform initiatives in most states and large county administered systems have typically followed a high-profile child maltreatment death or multiple deaths as outlined in news stories. Agency leaders have then convened multidisciplinary committees of prestigious professionals to review the circumstances of suspected child maltreatment deaths in order to increase public confidence that the CFR is objective, well informed and not influenced by pressure from the public agency to cover up performance problems or questionable decisions.  CFR’s usually contain recommendations for changes in practice or policy which the review team believes may have prevented a child death and similar deaths. Top managers commonly endorse the committee’s findings and proceed to develop a reform initiative which contains most of the recommendations, and usually other reforms as well.       

 Two or three decades ago, this process was repeated every few years and  then more frequently in many states where a new reform initiative was often initiated before the prior initiative had been completed. In some instances, policymakers enacted law named after a deceased child, e.g., Sirita’s law in Washington. It soon became apparent to some scholars, advocates, and managers that developing policy based on the circumstances of a specificchild death was not a sound way to improve practice. Some states instituted annual reviews of all child deaths, or child deaths viewed as preventable, while in other states (including Washington), the legislature mandated a review of all suspected child maltreatment deaths known to the public child welfare agency within a specific time frame, e.g., an open case within the the 12 months preceding the child’s death. 


More is not better 

In medium to large state systems or large county administered systems, the number of recommendations from CFR’s quickly accumulated, and external organizations such as a Children’s Ombuds office, or a child welfare advisory committee, tracked agency implementation of dozens or even hundreds of recommendations from past CFRs. A former director of one of the largest county operated child welfare agencies in the U.S. once told me that, in the agency he managed, there were several hundred recommendations from CFRs within a decade, most of which recommended new training programs, and that most of these recommendations were ignored and then quickly forgotten by agency staff.   

Child welfare directors and regional administrators are under persistent pressure from the federal Children’s Bureau, legislators, stakeholder and child advocates to support and implement multiple reforms, many of which do not come from CFRs. One of the challenges all child welfare managers face is managing the pace of change in understaffed agencies subject to demands for reform from multiple directions. Recommendations from large numbers of CFRs add to these pressures and, as a result, may be resented by managers savvy enough to refrain from saying publicly about a reform initiative: “This is too much.” 

One of the positive features of DCYF’s CFRs is that review committees usually attempt to keep recommendations for practice changes to a minimum, and sometimes make no recommendations. Experienced child welfare managers gradually learn that it is more effective to do a few things well rather than many things poorly. A child welfare director who promises many reforms to multiple groups without considering their workload impacts is in danger of losing support at all levels of the public agency, and among advocates and stakeholders who expect promises to be kept. 


Reform initiatives lose momentum 

Child welfare reform agendas developed largely from CFR recommendations are usually top-down initiatives whose goal is to accommodate and damp down public outrage that may lead to demands for a change in leadership. For this reason, strong motivation for implementation of reform initiatives tends to come from managers who feel threatened or have recently been appointed to make big changes in child protection. Given these pressures, agency leaders may not take the time needed to generate strong internal support for proposed reforms. However, without “buy-in” from agency staff at  throughout the chain of command, it is impossible to effectively implement major changes in policy and practice. Experienced child welfare staff become adept at playing the “change game” in which they appear to comply with top-down directives until an agency director and her/his direct reports begin to lose interest in reforms after major media have moved on to other stories, or perhaps due to another CM death that has restarted the whole reform process, beginning with a fatality review of a high-profile child death. 

Non-stop reform initiatives wear out middle managers and line units who learn how to go through the motions of adherence to new policies while “reading” the motivation of top managers to persist (or not) in a reform initiative. Child welfare leaders who pride themselves on embracing change and attempt to lead their agency through a forced march of overlapping reform initiatives demanded by advocates and policymakers may find themselves in an implementation morass of failed reforms with dwindling support both inside and outside the public agency.    


Practice changes require investments in the workforce

DCYF fatality reviews frequently call attention to workload pressures and vacancies which make it difficult for caseworkers to comply with assessment protocols, maintain frequent contact with families and follow up on safety plans.  Reviews of high-profile child deaths often recommend increases in casework positions which sometimes occurs to a modest extent, but without a legislative commitment to reasonable workload standards. Procedural requirements in bureaucracies tend to steadily expand regardless of staffing resources, leaving line units unable to fully comply with policy requirements, but without permission to waive any of them. 

During recent years, some child welfare managers have attempted to force compliance with policy frameworks regardless of staffing shortages through accountability measures and bullying, managerial practices that have a ruinous effect on staff morale and lead to widespread “gaming” of performance indicators. In these circumstances, reform initiatives invariably fail unless they reduce work requirements, which is rare. After decades of failed reform initiatives, one might expect that child welfare leaders and policymakers would have learned that unmanageable workload pressures will eventually undermine any and all efforts to improve child welfare practice that add to workload requirements; and that in understaffed agencies it is impossible to improve practice without major investments in the workforce. 

In most states, it has become increasingly difficult to fill vacant positions, even after salary increases or to reduce staff turnover. As a result, inexperienced caseworkers with minimal skills and little knowledge are being assigned overwhelming responsibilities, often without the mentoring of experienced peers. This is a formula for egregious failures of child protection in which children are left in danger from unrecognized safety threats or overreactions to situational maltreatment resulting from environmental stressors. 


Limitation of training programs

CFRs often identify knowledge deficits that contributed to a child’s death, and recommend more or better training programs for all caseworkers and  supervisors. However, one-off training programs have little or no effect on practice skills absent follow-up coaching and supervisory support. Training programs as delivered by contracted or professional trainers are a necessary but not sufficient way of transmitting practical skills and knowledge which requires repetition, emotional engagement, trial and error and understanding of exceptions to standard guidelines.  As a rule, too much is expected of training programs, and there is too little investment in practice improvements. As a result, training is usually an inadequate way of disseminating knowledge in child welfare agencies and the end of the road for many reform initiatives. 

Child welfare agencies strongly committed to reform initiatives do periodic trainings on specific subjects over a period of years, develop or hire specialists or consultants whom caseworkers can access as needed, seek out partnerships with child welfare scholars to research the outcomes of reforms, fund certification programs and provide timely feedback on performance, with goal of actually improving practice rather than looking good on performance indicators. 


Distinguishing between inadequate implementation and knowledge Deficits

CFRs have had little or no apparent effect on the number of CM fatalities nationally and in Washington in part due to the mismanagement of reform initiatives with: 

  1. too many recommendations generated top-down with
  2. too little support for reforms at the unit level and among middle managers reduced motivation from leadership teams as the interest of media, child advocates and policymakers wanes, while
  3. expecting sustained practice improvements without a commitment to reducing workloads and managing to reasonable workload standards,
  4. attempting to disseminate knowledge and skills through one-off training programs in response to CFR recommendations.

However, there are also knowledge deficits which reflect the limitations of research, scholarly insight, and current practice wisdom. Almost a third of CM fatalities in Washington are due to unsafe sleep practices that result in a child death from suffocation. Neither DCYF or CFR review teams have an effective strategy for preventing these deaths. There is currently no DCYF protocol for preventing child deaths from fentanyl ingestion, drunk driving, or drowning. As a result, in-home safety plans for infants and toddlers are often little more than an excuse for leaving infants and toddlers in dangerous circumstances with parents who have co-occurring substance abuse and mental health disorders, and sometimes a pattern of domestic violence as well. 

Some scholars and child advocates believe that child protection programs can prevent CM fatalities by applying lessons from airline safety, i.e., 

viewing child protection as an engineering problem in which well understood elements of child safety are consistently applied.  A Casey Family Program (CFP) policy brief asserts the following regarding Safety Science: 

“In the context of child protection, this means using an evidence-based lens applied to inform preventive and responsive actions, rather than basing policy and practice decisions on emotions and assumptions.” 

The CFP policy brief does not concretely identify screening guidelines, investigative and assessment practices, in-home safety plans or out of home placement guidelines that are elements of Safety Science, arguably because there is no agreement among experts that any widely used child protection framework, or the elements of these frameworks, are evidence based. In fact, just about every element of child protection frameworks from mandatory reporting to use of risk and safety tools, differential response, guidelines for in-home safety plans and the threshold for foster placement is hotly debated. Furthermore, child welfare systems are not even using cogent measures of child safety. 

A more defensible use of Safety Science following CM deaths has been to learn from CM by focusing on system issues, without blaming child welfare staff and dispelling fear from the workplace. The emphasis in Safety Science is on vigilance in identifying organizational weaknesses that lead to failures of child protection, and then altering unsafe practices before children are severely harmed. 

With a few exceptions, DCYF fatality reviews in recent years have followed the principles of Safety Science, i.e., with a focus on system issues rather than laying blame on caseworkers. In practice, most CFRs have utilized a bureaucratic version of safety science, i.e., identifying the many ways in which casework practice departs from DCYF policy and the agency’s safety framework. CFRs often question screening decisions, point out the lack of comprehensive assessment, or inaccurate information on the agency’s risk assessment tool, or the failure to observe an infant’s sleeping arrangements, or the lack of collateral contacts or “curiosity” in assessing child safety, or a lack of follow up on in-home safety plans.

The implicit assumption in many DCYF CFRs is that if caseworkers had the time to comply with the agency’s safety framework and had adequate knowledge of substance abuse, chronic mental illness, domestic violence,  and a better understanding of indicators of child maltreatment, a child’s death could have been prevented. Given this assumption, it is unusual for CFRs to recommend changes in the DCYF child safety framework or new  investments in specific types of services. 

Improved assessment and investigative practices combined with increased understanding of substance abuse, chronic mental illness, and domestic violence would surely save some children’s lives, but does not address the main challenge to child protection in troubled families with preschool children: ineffective in-home safety plans and the lack of a service array required to reduce the childcare burden on low-income  parents with functional impairments due to substance abuse, mental illness, and interpersonal violence and a history of trauma. 

Child welfare systems which seek to safely reduce the use of foster care should be steadily developing alternatives to foster care that include residential  treatment programs for mothers and infants, therapeutic childcare programs, and public health services for children with disabilities and chronic illness, along with greatly enhanced poverty related services. 

States that reduce use of foster care for young children through law and by persistent attacks on foster care, without making large investments in safety-oriented services are taking a reckless gamble with children’s lives.   ©



  • Child Fatalities and Near Fatalities in Washington State, Washington State Office of Family and Children’s Ombuds, August 2022. 
  • “How can child protection agencies use safety science to promote a safety culture?” Casey Family Programs, October 2020, available online.

Sounding Board commentaries available at

[email protected]



Conference Sponsors

Previous slide
Next slide
24/7 support hotline
Foster Families
Family Time Network
Visitation Services

All Family Time related inquiries

Upcoming Events

june, 2024

Small Donations, Bigger Impact

Support Families Helping Families

become a Friend of FPAWS