This article is more than 5 years old. Information may no longer be current.
Prediction tool increases accuracy of abusive head trauma diagnosis
Click Here to Manage Email Alerts
Researchers have validated the effectiveness of a prediction tool that can determine the probability of abusive head trauma in children when specific symptoms are present in a recent study.
“In the era when the very diagnosis of [abusive head trauma (AHT)] continues to be contested in both the public and legal domain, adequately validated [clinical decision rules (CDR)] provide more scientific evidence to support clinical decision-making,” Laura Elizabeth Cowley, MSc, MBPsS, of the department of primary care and public health, School of Medicine at Cardiff University, United Kingdom, and colleagues wrote. “This tool has the potential to contribute to decision-making in these challenging cases.”
The researchers used the Predicting Abusive Head Trauma tool, designed to detect the likelihood of AHT in children diagnosed with intracranial injuries. To test the tool’s sensitivity, Cowley and colleagues studied a cohort of 198 children aged younger than 36 months admitted to two centers with intracranial injuries. Participants were screened for six features associated with AHT: retinal hemorrhage, rib and long-bone fractures, apnea, seizures, and head or neck bruising.
When at least three of the screened injury features were present, the tool estimated a greater than 81.5% likelihood that the injuries in patients aged younger than 36 months were caused by AHT (95% CI, 63.3-91.8). The sensitivity for the tool was validated to be 72.3% (95% CI, 60.4 -81.7), and the specificity was 85.7% (95% CI, 78.8-90.7).
The researchers suggested that using this tool, in lieu of CDR, can aid in diagnosing pediatric AHT. The tool uses algorithms to give clinicians the information required to determine the best course of action, rather than dictating a specific diagnosis trajectory.
“Given the rarity of AHT for clinicians who are not child abuse specialists, it is important to be able to explicitly define which combination of clinical features carries a high probability of abuse when a clinical workup has been conducted,” Cowley and colleagues wrote. – by David Costill
Disclosure: The researchers report no relevant financial disclosures.
Perspective
Back to Top
Leticia Castillo, MD
This is a serious attempt by a well-established group of investigators from Cardiff University, Cardiff, United Kingdom to validate a predictive instrument on the likelihood of abusive head trauma. Differentiating between inflicted non-accidental trauma and accidental trauma is challenging and its diagnosis often leads to serious ethical and social considerations.
Mistakenly allowing a child to return to an abusive environment may result in further injury and death of a child; conversely, wrongly removing a child from his/her family environment will lead to significant emotional and psychological trauma for the child and the family. Thus, the development of a predictive tool to help improve the accuracy of the diagnosis of abusive head trauma is a welcoming development. However, no tool is accurate 100% of the time every time.
Although this development is a laudable effort to improve decision-making on the problem of child abuse diagnosis, caution should be exercised when applying this and any other developed tools, until replication of findings is confirmed, particularly when the stakes of erring on one side or the other are so high. Predictive analytics have yet to come of age when applied to the diagnosis of abusive head trauma.
Click Here to Manage Email Alerts