Machine Learning Uses Predictive Analytics For Suicide Prevention

Johns Hopkins University researchers developed a machine-learning algorithm that uses predictive analytics to identify adolescents experiencing suicidal thoughts and behavior.

After decades of analysis, researchers discovered specific risk factors associated with suicidal thought and behavior among adolescents, helping to improve suicide prevention efforts. However, few studies have examined these risk factors in combination with each other, especially in a large adolescent population.

According to researchers, machine learning can now provide new opportunities to study risk factors in adolescents, improving suicide prevention efforts.

The research team used machine learning to analyze data from a survey of high school students in Utah. The survey is routinely conducted to monitor issues, including drug abuse and mental health.

The data featured responses from more than 300 questions each from over 179,000 high school students who took the survey between 2011 and 2017. Additionally, the researchers studied demographic data provided by the United States census.

The team discovered they could use the survey data to predict with 91 percent accuracy which students’ answers indicated suicidal thoughts or behavior.

“In doing so, they were able to identify which survey questions had the most predictive power; these included 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,” the press release stated.

The new algorithm’s accuracy is greater than previously developed predictive analytic methods, suggesting that machine learning could potentially improve the understand of adolescent suicidal thoughts and behavior. According to researchers, this study could play a crucial role in enhancing suicide prevention programs and policies.

Additionally, future research could expand the researcher’s findings by incorporating data from other states as well as data on suicide rates in adolescents.

“Our paper examines machine learning approaches applied to a large dataset of adolescent questionnaires, in order to predict suicidal thoughts and behaviors from their answers. We find strong predictive accuracy in identifying those at risk and analyze our model with recent advances in ML interpretability,” the study authors wrote.

“We found that factors that strongly influence the model include bullying and harassment, as expected, but also aspects of their family life, such as being in a family with yelling and/or serious arguments. We hope that this study can provide insight to inform early prevention efforts.”

The study results were presented by Orion Weller of Johns Hopkins University and colleagues in the journal PLOS ONE on November 3, 2021.