Even before the spike in attempted suicide due to the challenges of COVID-19, suicide was the 2nd leading cause of death among adolescents in the U.S. But researchers have recently developed a new, machine learning-based algorithm that could be a significant help to mental health providers working to identify adolescents who are experiencing suicidal thoughts and behavior. 

While suicide prevention efforts have long been aided by research identifying specific risk factors associated with suicidal thoughts and behavior among adolescents, stakeholders have noted few studies that have explored these risk factors in combination with each other across large groups of adolescents.

To that end, researchers from Johns Hopkins University applied machine-learning analysis to data from a survey of high school students in Utah that is routinely conducted to monitor issues such as drug abuse and mental health. The data included responses to more than 300 questions each for more than 179,000 high school students who took the survey between 2011 to 2017, as well as demographic data from the U.S. census.

According to a statement, the researchers found that they could use the survey data to predict with 91 percent accuracy which individual adolescents’ answers indicated suicidal thoughts or behavior. Moreover, they were able to identify which survey questions had the most predictive power, including questions about digital media harassment or threats, at-school bullying, serious arguments at home, gender, alcohol use, feelings of safety at school, age, and attitudes about marijuana.

In their report, the researchers identified the top ten factors that were most predictive of adolescent suicidal thoughts and behavior (STB), including, among other things, being threatened or harassed over the internet, picked on or bullied by a student at school, or involved in serious family arguments.

“The implications of this research are important for prevention programming and policies related to adolescent STB,” the researchers wrote. “Prevention program specialists and other policymakers can use the STB risk profile and its associated rankings to prepare services, resources, and assessments aimed at school, community, and family settings. As such, these findings also have important implications for prevention policy and resource allocation. For example, continuing and growing a focus on bullying on social media and bullying or intimidation at school or other community settings are prime areas for policy response.”

The new algorithm’s accuracy is higher than that of previously developed predictive approaches, suggesting that machine-learning could indeed improve understanding of adolescent suicidal thoughts and behavior.

Photo by phonlamai/Getty Images