Identifying Suicide-Related Language in Smartphone Keyboard Entries Among High-Risk Adolescents

Adolescent suicide rates have risen over the past two decades, underscoring the need for improved risk detection strategies. This study leverages passively-collected smartphone data to identify suicide-related language in adolescents’ keyboard usage using natural language processing (NLP). We developed a suicide lexicon for adolescent language and validated it with labeled data (N=121,515 text entries), demonstrating higher performance than lexicons not designed for youth. Across two independent cohorts at elevated suicide risk (Ns=208/211; >6 million text entries), 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%). Compared with the lexicon alone, human coding of suicide-related entries, especially first-person language, showed stronger associations with STB history, highlighting the need for approaches to distinguish intent. Findings highlight both potential and limitations of NLP for suicide prevention.