Do higher standards of proof for child abuse and neglect impact rates of substantiation?



Do higher standards of proof for child abuse and neglect impact rates of substantiation?

Kahn, N. E., Gupta-Kagan, J., & Eschelbach Hansen, M. (2017). The standard of proof in the substantiation of child abuse and neglect. Journal of Empirical Legal Studies, 14(2), 333-369.

What can we learn from this study?

Child welfare policy and practice vary across the country, including the standard of proof for substantiating reports of child abuse or neglect. While some states require “clear or convincing,” or a “preponderance” of evidence before substantiating a report, others have a much lower threshold, with some requiring only “reasonable” evidence. This study analyzes the extent to which thresholds influence the disposition of child maltreatment reports, and ultimately the number of children involved in a substantiated report of maltreatment or placed into foster care. The study also looks at the impact of recent changes to these thresholds on overall trends. 

Study details:

  • Population: 7.9 million reports of abuse and neglect with a focus on five states that increased their standard of proof
  • Data source: National Child Abuse and Neglect Data System (NCANDS) state summary file and child-level file
  • Methodology: Logistic regression, inverse-probability weighting estimator, fixed effects framework
  • Dates: 2000-2012

What are the critical findings?

States with a high standard of proof (or a high threshold/burden for child protection agencies to provide evidence of child abuse or neglect) tend to have lower rates of substantiated maltreatment. 

  • An increase in the standard of proof resulted in a 14% decrease in the likelihood of substantiation, primarily among those cases that are harder to prove. (There was no change among cases that are easier to prove.)
  • There are costs associated with errors in decision-making. An increase in the standard of proof increases protections to parents against a Type 1 error (finding abuse and neglect when there is none), which also increases the possibility of a Type 2 error (failure to intervene when child abuse or neglect has occurred). The consequences of a Type 1 error include undue trauma for families, as well as forced participation in unneeded and unnecessary services. The consequences of a Type 2 error include the potential that abuse or neglect will continue. 
  • While an increase in the standard of proof is only weakly associated with a decline in foster care placement, it importantly is associated with an increase of almost 30% in the provision of housing, family preservation, and other services to support families. 

Why is this important for our work?

Decision-making at the front door of the child protection agency has a profound impact on the 7.8 million children and their families involved in a report of maltreatment each year, and is costly for child protection agencies. Given that increases in the standard of proof result in more upstream services being offered to families, additional study is worthwhile to further explore the benefits of modifying state policies and to fully understand the associated impact on child safety and well-being. 

This summary synthesizes the findings from a single research study. To learn more, please review the following resources: Can decreasing unwarranted reports to child protection agencies improve outcomes for children and families? and How can helplines serve as a better pathway for families to access support?

For additional information, access the article directly or email