Health-Behaviors Associated With the Growing Risk of Adolescent Suicide Attempts: A Data-Driven Cross-Sectional Study

Area of Science:

• Public Health
• Adolescent Psychology
• Data Science

Background:

• Adolescent suicide attempts represent a critical public health issue, particularly in vulnerable states like Montana.
• Understanding the complex interplay of health behaviors and socio-demographic factors is crucial for effective prevention strategies.

Purpose of the Study:

• To identify and examine associations between specific health behaviors and the risk of adolescent suicide attempts.
• To control for socio-economic and demographic variables in analyzing suicide attempt risk.
• To leverage machine learning for predicting adolescent suicide risk based on health behaviors.

Main Methods:

• Cross-sectional data analysis of 22,447 adolescents in Montana (1999-2017).
• Utilized machine learning algorithms and logistic regression to model suicide attempt risk.
• Evaluated 29 variables including psychological behaviors, substance use, school activities, and demographics.

Main Results:

• A Bayesian tree ensemble model achieved 80.0% goodness-of-fit and 78.2% predictive accuracy.
• Key predictors identified: sadness/hopelessness, school safety concerns, physical fighting, inhalant/drug/cigarette use, and early sexual initiation.
• Minority groups (American Indian/Alaska Natives, Hispanics/Latinos) and females showed higher vulnerability.

Conclusions:

• This study elucidates critical health behaviors and disparities linked to adolescent suicide attempts.
• Identified health behaviors serve as potential early warning signs for suicide risk.
• The findings underscore the importance of addressing complex interactions in adolescent mental health research.