Adolescent Suicide Rates Rise, Highlighting Need for Improved Risk Detection Strategies

Adolescent suicide rates have experienced a steady increase over the past two decades, underscoring the imperative for enhanced risk detection methods. A recent study has leveraged smartphone data and natural language processing to identify patterns in youth language associated with suicidal ideation. The research demonstrates the potential for this approach in suicide prevention, but also highlights its limitations.

Key Takeaways:

• The study leveraged a dataset of 121,515 text entries and identified a suicide lexicon specifically designed for adolescent language, which demonstrated higher performance than lexicons not designed for youth.
• Across two independent cohorts at elevated suicide risk, lifetime suicidal thoughts and behaviors (STB) and current suicidal ideation were associated with increased frequency of smartphone suicide-related language.
• Human coding indicated varied language, including authentic first-person current suicidal ideation (14.5%) and jokes or hyperbole (20.2%).
• Comparing the lexicon alone to human coding of suicide-related entries, especially first-person language, showed stronger associations with STB history.
• The findings highlight both the potential and limitations of natural language processing (NLP) for suicide prevention.

Statistics:

• The study analyzed a dataset of 121,515 text entries and two independent cohorts at elevated suicide risk (Ns=208/211; 6 million text entries).
• 14.5% of human-coded entries indicated authentic first-person current suicidal ideation.
• 20.2% of human-coded entries represented jokes or hyperbole.
• The lexicon alone demonstrated higher performance than lexicons not designed for youth.

Sources:

• osf.io/preprints/psyarxiv/gfa7h_v2/