miragenews.com
A new study from UCLA Health highlights a troubling gap in detecting suicide-related emergencies among children and adolescents in emergency departments, with Black, Hispanic, male, and preteen youth disproportionately underrecognized. The findings underscore the need for enhanced, equitable approaches to identify and support youth at risk of suicide.
Researchers analyzed nearly 3,000 cases of young patients, aged 6-17, who visited two emergency departments in Southern California for mental health concerns. Using machine learning, they compared standard emergency department practices—such as patient-reported concerns and clinician diagnostic codes—with advanced electronic health record phenotyping to detect suicidality. The results were clear: standard methods often missed signs of suicide, especially in specific minority and younger demographics. Dr. Juliet Edgcomb, one of the study’s authors, emphasized that “existing methods are missing kids, and not missing them at random,” pointing out the uneven and insufficient detection capabilities.
The study’s timing is crucial as suicide rates in the U.S. continue to rise, with preteen suicide deaths increasing annually by 8% between 2008 and 2022. Although standard methods for detecting suicide risks typically rely on chief concerns and diagnostic codes, these methods can fall short when suicidality is not the primary diagnosis. Clinicians may, for instance, document symptoms like depression or sadness without explicitly coding for suicide-related issues, making it challenging to recognize those at risk, particularly among young boys and minority groups.
These findings call for immediate attention from healthcare providers, as children and adolescents experiencing suicide-related crises are slipping through the cracks. As Edgcomb noted, “If you’re a Black preteen presenting with suicidality, you’re much less likely to be detected than if you are a white adolescent.” Given that men have a higher suicide mortality rate, the study also raises concerns about the lower detection rates for boys, suggesting that stigma and clinician perceptions of suicide risk may further exacerbate disparities.
The research advocates for the development of more sophisticated algorithms, supported by artificial intelligence and natural language processing, to analyze detailed clinical notes and other nuanced data. Enhanced detection tools could close these gaps, particularly in vulnerable populations, and better support emergency department clinicians in identifying and aiding youth at risk for suicide. The study’s findings indicate an urgent need for health systems to integrate these tools across multiple settings to validate and expand their use, ensuring comprehensive and equitable suicide prevention strategies.