EMR-Based Alert System Can Identify Possible Child Abuse Victims

Sometimes, it’s pretty easy for a physician to tell that a child might be experiencing physical abuse at home. However, sometimes physicians are rushed and may not do an adequate screening for abuse, the signs of which aren’t always available at first glance.

However, a group of researchers has developed an algorithm, drawing on patient records in the EMR, which it says can improve screening rates for physical abuse and identify such cases earlier.

The project, the write-up of which appears in the Journal of the American Medical Informatics Association, coded triggers to identify children less than two years old at risk for physical abuse into the EMR using a freestanding pediatric hospital with a level 1 trauma center. The researchers embedded 30 age-specific triggers in the EMR.

During the test, the system ran a “silent mode,” in which study personnel saw data on children whose clinical symptoms triggered the alert system but physicians did not. During the period between October 21, 2014 through April 6, 2015, 226 children triggered the alert, the mean age of whom was 6.5 months.

During the pilot the system detected 98.5% of children less than two years of age with signs of probable or definite child abuse, according to the study authors.

If these algorithms are that successful in identifying at-risk children, one would hope that the system moves from pilot to widespread rollout fairly soon. In theory, the system should help clinicians who encounter children in potential danger, especially ones presenting with serious injuries in the emergency department, be better prepared to identify these children and take appropriate action.

Ultimately, this study suggests that even if such clinicians are alert and careful, triggers generated by an EMR might be more effective at detecting these cases. After all, while clinicians must juggle multiple patients in an extremely hectic environment, especially in the ED, EMRs don’t get tired and they don’t need to check a signs and symptoms list manually to detect signs of trouble.

Of course, while these triggers can be very helpful in investigating signs of abuse, clinicians would be ill-advised to rely on them entirely, as there’s no substitute for experience and medical judgment. Also, there’s always a risk that adding another alert to the cacophony of existing alerts could lead to it going unnoticed. Still, it seems certain that if nothing else, this is a promising approach to protecting children from harm.

About the author

Anne Zieger

Anne Zieger

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.