AI Tool Flags At-Risk Teens Before Mental Health Crisis

More than 40% of students in high school report feeling consistently sad or hopeless, and nearly 20% have seriously considered suicide, according to the CDC. With too few child mental health specialists in the country to screen every teen who needs it, researchers at Duke University are testing whether artificial intelligence can help doctors identify at-risk teens before a crisis develops.

Mental Health & Addictions

Teen mental health struggles like persistent sadness, anxiety, and mood instability rarely exist in isolation. Left untreated, these conditions grow into risk factors for substance use later on, as teens sometimes turn to alcohol or drugs to manage symptoms they couldn’t otherwise address. Comprehensive treatment for young people increasingly means screening for both mental health symptoms and early signs of substance use side by side, rather than treating them as separate problems.

The AI model being tested at Duke draws on data from more than 11,000 children nationwide, including family conflict, social factors, and health history. The team used the data to flag which teens may be at elevated risk for self-harm, including suicide. Duke child psychiatrist Dr. Jonathan Posner noted that his team don’t want to replace doctors but to help clinicians see risk sooner because early warning signs can be subtle and easy to miss in routine visits. 

Researchers expressed surprise by how strongly sleep patterns predicted later risk, according to Duke AI researcher Dr. Matthew Engelhard. The model’s early tests could estimate whether a teen might develop a mental health condition up to a year before symptoms appeared, with roughly 75% accuracy.

Treating Dual Diagnosis in Adolescents

When a teen has both mental health and substance use concerns, clinicians call this a co-occurring disorder, or dual diagnosis. Treating only one side, for example addressing depression without ever asking about potential opioid use, tends to produce worse long-term outcomes than integrated care that addresses both simultaneously. Tools like Duke’s AI model can eventually help primary care providers flag teens for this kind of comprehensive dual diagnosis evaluation earlier than current screening allows.

Residential and outpatient adolescent programs increasingly combine evidence-based therapies such as behavior counseling with substance use interventions and family involvement. Many teens already use chatbots for mental health advice, and teens showing signs of both a mental health condition and substance use, an integrated program is generally preferable to separate, uncoordinated care.

Comprehensive Care for All

Parents and caregivers noticing early warning signs, persistent sadness, withdrawal, changes in sleep, or any substance use, don’t need to wait for a crisis to seek an evaluation. Dual diagnosis treatment programs and mental health treatment facilities that screen for co-occurring conditions can address both issues together. 

Call 800-908-4823 (Sponsored) today to chat with an expert about mental health and addiction treatment options. Or browse our directory for verified rehabs located anywhere in the United States. If anyone shows signs of personal distress, call 988.

Author

Eric Owens

Read Bio

Eric Owens has been a writer and editor for various businesses as well as his own successful websites. He has extensive experience creating content in the health and wellness space and the sustainability space. He holds a bachelor degree in Philosophy which helped him with presenting complex information in a simple way that all audiences can understand.

Editor

Peter Lee, PhD

Read Bio

Peter W.Y. Lee is a writer and historian of American history during the Cold War. His primary focus is the relationship between youth and popular culture and its impact on U.S. society during the twentieth century. He has published widely on how the public has used popular culture as a mechanism to address political and social shifts throughout time