Main :
News :
Researchers develop machine learning models that could improve suicide-risk prediction among children

Researchers develop machine learning models that could improve suicide-risk prediction among children

Archive

It is a really exciting time for clinical research informatics research in mental health, especially child mental health care, where there are many unanswered questions and substantial clinical need. The convergent expertise of the CTSI community gives me hope that it is possible to efficiently and creatively refine precision of prevention efforts and make a real difference in lives of children and families.

- Juliet Edgcomb, MD, PhD

Juliet Edgcomb, MD, PhD

A new study from UCLA Health researchers finds that the typical ways health systems store and track data on children receiving emergency care miss a sizable portion of those who are having self-injurious thoughts or behaviors. The researchers also found that several machine learning models they designed were significantly better at identifying those children at risk of self-harm.

Amid a nationwide youth mental health crisis, mental health providers are trying to improve their understanding of which children are at-risk of suicide or self-harm so providers can intervene earlier. However, health systems often do not have a full understanding of who is coming through their doors for self-injurious thoughts or behaviors, meaning that many risk-prediction models designed to flag children at future risk are based on incomplete data, limiting prediction accuracy.

“Our ability to anticipate which children may have suicidal thoughts or behaviors in the future is not great – a key reason is our field jumped to prediction rather than pausing to figure out if we are actually systematically detecting everyone who is coming in for suicide-related care,” said Juliet Edgcomb, MD, PhD, the study’s lead author and associate director of UCLA’s Mental Health Informatics and Data Science (MINDS) Hub. “We sought to understand if we can first get better at detection.”

UCLA CTSI provided support for this research through the expertise of the Biomedical Informatics Program (Amanda Do, MPH and Theona Tacorda, MS), which delivered the medical record data used in the study: "The services of the CTSI Biomedical Informatics Program and their ability to customize EHR data queries allowed us to consider and incorporate previously unexplored potential indicators of suicide-related visits among children, such as the presence of a mental health detainment order for danger to self," said Dr. Edgcomb.

As for what's next on her research agenda, Dr. Edgcomb looks forward to using her recently awarded Core Voucher award to further her work in child mental health care. "We are developing new models to detect elementary school-aged children (6-to-12-year-olds) with suicide-related emergency department visits, a population who is at high risk of being missed when relying on diagnostic codes alone," she said. "Our team also recently received a CTSI Core Voucher to leverage clinical text mining with weak supervision to detect childhood-onset suicide-related behavior and mental illnesses in a increasingly automated manner."

This story was adapted from the press release originally featured in the UCLA Health Newsroom. Click here to read the rest of the story.

Featured Projects

Clinical Trials

Newsletter
Sign Up