Race and Bias in Child Maltreatment Diagnosis and Reporting

Pediatricians have implicit and explicit racial biases that impact the health and well-being of children and their families.^1,2  Similarly, effects of racism on diagnosis and reporting of suspected child abuse and neglect to child protective services (CPS) can have serious consequences. Although we and others are mandated to report suspected child abuse or neglect in all US states and territories, the threshold for reporting requires only “reasonable suspicion” of abuse or neglect.³  Pediatricians may also report families that they perceive need additional resources.

Research on racial bias and physician reporting is limited.⁴  Disproportionality in reporting is widespread on the basis of race, culture, and ethnicity. Asian American and Hispanic children are underrepresented, but overrepresentation of Black children throughout the child protection system has been most widely described.⁵  Black children are reported at approximately twice the rate of white children, and the complex relationship of reporting with poverty and race has yet to be fully understood.^4–6  Once reported, cases with Black children are more likely to be accepted for investigation, be confirmed, be brought to court, result in removal of the children from their families for longer periods of time, and take longer to be closed, possibly related to surveillance bias.^5,7  Multiple points in this process are subject to bias, but the process begins with reporting.^8,9 

The concept of bias has been recognized and discussed within the medical community for at least half a century. Diagnostic errors due to bias can lead to errors in reporting.¹⁰  Although physicians have historically been key players in the development of US reporting systems, they and other medical sources accounted for only 10.5% of the 4.3 million reports to CPS in 2018.¹¹  Even so, CPS and other agencies rely heavily on medical diagnosis from health care providers, so biases can be magnified.^12,13  In 1985, Hampton and Newberger¹⁴  showed hospitals failed to report almost one-half of cases meeting their study’s definition of abuse. White children and those with higher family income and employment had fewer reports.

Studies have highlighted bias in the medical evaluation of child abuse and neglect. In a classic report, Jenny et al¹⁵  found 31.2% of 173 children with abusive head trauma had been previously seen by physicians. The diagnosis was more likely to be missed in young children from white, “intact” families. Worse, 27.8% were reinjured after the missed diagnosis, and 4 of 5 deaths might have been prevented. Lane et al¹⁶  later found that underrepresented minority children seen for fractures were more likely to be evaluated with a skeletal survey and reported, even after controlling for insurance status, independent expert determination, and likelihood of abusive injury. They concluded, “It is possible biases on the part of mandated reporters may contribute to these differences.”

Others identified similar trends. Wood et al¹⁷  noted fewer skeletal surveys among white infants evaluated for traumatic brain injury. Hymel et al¹⁸  observed significant race- and ethnicity-based disparities in abusive head trauma evaluation and reporting, concluding, “in the absence of local confounders, these disparities likely represent the impact of local physicians’ implicit bias.” Reports from medical providers have also been found to be affected by socioeconomic status, ethnicity, and culture.^19,20 

Diagnostic errors in medicine are influenced by knowledge gaps, communication skills, and access to resources, as well as cognitive and implicit biases.^21,22  An Institute of Medicine report²³  describes clinical reasoning as “the cognitive process that is necessary to evaluate and manage a patient’s medical problems” and, in addition, describes a dual process theory, one analytical and one nonanalytical. Systems I thinking is described as intuitive, automatic, and based on pattern recognition by using rules of thumb and mental shortcuts. In comparison, systems II thinking uses metacognition and is deliberate and analytical, with hypothetical and/or deductive reasoning and logic.

Skellern²⁴  describes examples of errors in both types of reasoning applied to child abuse. Systems I errors in diagnosis may be difficult to overcome because these are unconscious processes.²²  Mandated reporting requires only suspicion, and pediatricians may rely on their “gut” and systems I thinking, potentially relying on implicit biases. Stolper et al²⁵  report on the use of “gut feelings,” the further steps needed to analyze the source of suspicions of abuse, and their need for more training. Gut feelings regarding abuse, potentially triggered by systems I thinking and potentially biased, may initiate the pediatrician’s diagnosis and suspicion for reporting. These thoughts and feelings should be followed by analytical reasoning using systems II thinking.²⁶ 

It is important for pediatricians in all practice settings to be able to make appropriate mandated reports for suspected child maltreatment. We offer some suggestions for minimizing the effects of bias in reporting, with the goal of equitably improving the safety of all children.

First, pediatricians need ongoing education with comprehensive information about definitions, examples, and indicators of the major types of childhood maltreatment, racial disproportionality, and systemic and implicit biases. A recent study highlighted gaps in existing training curricula regarding identifying child maltreatment and when and why reports should be made.²⁷ 

Second, standard tools, such as screening protocols, clinical guidelines, and electronic health records that trigger reportable concerns independent of clinician judgment may help to objectively inform decisions.^28,29  However, as our experience with neonatal drug testing protocols teaches us, these tools have to be nonbiased and instituted with adequate training and compliance coupled with attitudinal change to be effective.^30,31  Paper checklists and published screening tools can also provide support for pediatricians’ systems II thinking.^26,32 

Third, we should strive to build on efforts to develop strong multidisciplinary teams that use diverse cultural and racial perspectives.³³  Interprofessional teamwork, recommended in hospitals, must be intentionally used for the purpose of reflecting on bias.³⁴  There are many hospital child protection teams and community-based child advocacy centers available to help. In addition, physicians practicing in office settings without easy access to teams can also use American Academy of Pediatrics resources to support their work, such as resources from the Council on Child Abuse and Neglect,³⁵  the Resilience Project,³⁶  and chapter child abuse committees. Taking a moment to review for bias in each case with others can be a powerful tool.

Fourth, it is important to recognize and reflect on personal knowledge and biases, specifically regarding child abuse and neglect. For example, our interpretation of parenting skills may rely on expectations based on our own experience and cultural and socioeconomic background. Perceptions of family situations, such as when there is use of cultural medicines³⁷  or teenager parenting,³⁸  play a role in influencing reporting. In addition, our recognition of bruising and other skin findings may be based on experiences and training in children who have different skin tones.

When considering a mandated report for suspected child abuse or neglect, we recommend considering the following questions to help differentiate systems I from systems II thinking:

• Why do I suspect maltreatment?

• What is the objective evidence?

• If the family does not look like me, share my values, or lives on the “other” side of town, is that affecting my thinking?

• If my gut is telling me to report, why is that?

Similarly, when making a decision not to report, pediatricians should check if they are thinking, “they are such a good family,” “I have known them since they were children,” “my patients would never do that,” or similar emotional reasoning.

Although self-reflection and mitigation of interpersonal bias are important, other factors should be considered, such as the many systemic and structural forces at play. Child abuse reporting, diagnosis, management, and treatment necessarily involve multiple community agencies whose work must also be free from bias. The American Academy of Pediatrics¹  has additional suggestions that address some of these forces, including the following: (1) creating a culturally safe clinical environment, (2) training staff in culturally competent care, (3) using strategies that counter or replace negative racial messages and experiences with positive ones, and others.

We all strive for culturally competent care with awareness of bias. Implicit and explicit biases are drivers of disproportionate reporting and investigation. Racial bias can be checked through personal reflection and training, input from a diverse multidisciplinary team, assistance from an objective screening process, and case review by using deductive reasoning methods, such as those provided by child abuse pediatric consultation.