Although child deaths are fortunately rare events in any system, the approach systems take to reviewing and learning from them can have an outsized impact on improvement and reliability. For example, when a child welfare system’s response to a high-profile death results in blame, as is commonly seen, professionals in that system can become more conservative and less likely to trust reunification and preservation attempts. This can result in an increased number of kids in care as rates of removal and length of stay increase. By contrast, a child death review process that focuses on building up teams, identifying systems concerns, and promoting continuous learning can be an organization’s most effective improvement tool. To address these concerns, many have called for child death review innovations that apply safety science and support a safety culture (Commission on the Elimination of Child Abuse and Neglect Fatalities, 2016; Children’s Bureau Capacity Building Center: Child Welfare Workforce Safety, 2018; David Sanders, Casey Family Programs, in testimony to U.S. House Committee on Ways and Means, 2012).


In 2013, Tennessee’s Department of Children’s Services (DCS or the Department) successfully developed and implemented a systems focused, non-punitive, critical incident review process. However, reviewers quickly identified that the extensive narrative content derived from reviews could be better organized through use of a standardized, systematic assessment. To that end, The Department collaborated with The Praed Foundation and Chapin Hall Center for Children at the University of Chicago to develop the Safe Systems Improvement Tool (SSIT) (Appendix A) for use in its critical incident reviews (i.e., Child Death Reviews). It is designed to support efficient and reliable event analysis and to foster data-driven recommendations. Similar to how a barometer measures pressures in the atmosphere, the SSIT measures pressure existing within organizations and provides a frame for targeted quality improvement work.

The SSIT is a first of its kind, multi-purpose, information integration tool designed to be the output of a systems analysis process. It is also a performance and quality improvement system that supports a culture of safety, improvement, and resilience. As such, completion of this instrument is accomplished in order to allow for effective communication at all levels of the system. Since its primary purpose is communication, DCS leveraged its experience and expertise with the Child Adolescent Needs and Strengths to develop an approach based on communication theory rather than the psychometric theories that have influenced most measurement development. The SSIT’s thirteen systems items are organized into three domains: Professional (cognitive fixation, fatigue, stress, knowledge deficit, documentation and evidence); Teamwork (teamwork/coordination, supervisory support, production pressure); and Environment (demand/resource mismatch, equipment/technology, policies, service array) (Appendix B).

Use of the SSIT facilitates a review process that: 1) Understands the inherently complex nature of child welfare work and the factors that influence decision-making; 2) Acknowledges decisions alone are rarely direct causal factors in a child’s death, but these decisions may affect the overall trajectory of well-being for a child or family and be an influence, among many influences, of poor outcomes; and 3) Provides a safe and supportive environment for staff to process, share, and learn from child deaths in an effort to best support quality case management practices and influence increasingly safe outcomes for children.

The SSIT provides both a guide in facilitating these debriefings (i.e., items to be assessed) and an efficient means to capture the rich information provided. After debriefings, critical incident reviews are presented to a multi-disciplinary team who dissects the case and relevant findings from a systems perspective. Once again, the SSIT guides these conversations and serves as systematic documentation of the review’s results. The SSIT is completed once, at the closing of every case review.

Using the SSIT

There are four levels of rating for each item with specific anchored definitions and assessment prompts (Appendix C). These item level definitions are designed to translate into the following action levels: “0” = no evidence, no need for action; “1” = latent factor; “2” = action needed to mitigate risk and avoid recurrence of non-proximal actions/decisions; “3” = immediate or intensive action required to prevent recurrence of proximal actions/decisions. Systems analysts involved in reviews are trained to reliably rate each item.

Items scored at a ‘2’ or ‘3’ are actionable and require the rater to enter narrative justifying and explaining the score. For example: A child dies from medical neglect during an open Child Protective Service case. Upon review, it is discovered the assigned case manager relied on the mother and in-home nurse to explain how the child’s durable medical equipment was used. The information provided was against the doctor’s and manufacturer’s orders, but the case manager did not know this. In this situation, the item “knowledge deficit” is rated a ‘3’ with an explanatory narrative entered alongside the score. Problem statements (i.e., findings) also receive a recurrence score. It is important for a system to understand the likelihood of recurrence when making decisions about when and how to apply finite quality improvement resources.

On a quarterly basis, the SSIT’s scores are aggregated and analyzed to review findings and discuss trends. The SSIT score can be aggregated and analyzed in a number of ways (e.g., looking only at: unsafe sleep-related infant deaths, custodial youth who died from medical complications) to identify problem areas and track change over time.
Analysis of SSIT data has resulted in several quality improvements including: the creation of a statewide protocol for educating caregivers on infant safe sleep and same-day distribution of pack-n-plays to families, increased provision of health services for recovered runaway youth at risk of exploitation, improved supports for frontline professionals traveling long distances with children/youth, a pilot of new equipment (e.g., satellite phones) to aid child and employee safety through improved communication in rural areas, and Spaced Education curriculums. The SSIT data is integral to content delivered in child protective service trainings as well.

Case Study: SSIT Scoring ExampleEach SSIT rating suggests a different systems influence on casework.


In summary, the SSIT is an innovative tool that supports organizational learning and an improvement approach focused on human interaction in complex systems. The SSIT helps the Department identify and prioritize systems improvement opportunities. The structure of the SSIT allows a system to uncover those threats/opportunities most proximal and likely to recur. SSIT data contributes to professional learning at the individual case level and can be aggregated at any level of the system to support improvement and evaluate change over time. Classifying complex systems findings into a common language supports improvement discussions at all levels of the organization. Use of the SSIT in critical incident reviews reinforces important organizational values and shifts focus away from discussions of blame-worthy acts and simple cause and effect relationships. It supports efforts to create a culture of safety by increasing understanding of complex interactions in tightly-coupled systems.


Find additional resources at: