OBJECTIVE: To investigate the associations of demographic variables, childhood adversities (CAs), and mental disorders (MDx) with onset, transition, and persistence of suicidal thoughts and behaviors (STB) among first-year university students. METHOD: Poisson regression models within a discrete-time survival framework were constructed using web-based self-report survey data from 72,288 incoming university students across 18 countries (response rate=20.9%; median age=19 years, 57.9% female, 1.4% transgender, 21.0% non-heterosexual). These models examined the associations of four demographic variables, five CAs, and eight MDx with STB outcomes. RESULTS: Lifetime prevalence of suicidal ideation, plans, and attempts was 47.0%, 26.0%, and 9.6%, respectively; 12-month estimates were 30.6%, 14.0%, and 2.3%. In unadjusted analyses, associations were strongest between lifetime onset of suicidal ideation and CAs (RR range 4.4-7.0), particularly parental psychopathology (relative risk [RR]=7.0 [95% CI 6.5-7.7]), followed by MDx (RR range 1.3-3.0). Of the demographic subgroups, transgender students had highest risk of STB (lifetime ideation onset RR=2.4 [2.3-2.6]; ideation-to-attempt transition RR=1.5 [1.3-1.8]). In fully adjusted models, strongest predictors of lifetime ideation onset were emotional abuse (RR=2.1 [1.9-2.2]), major depressive disorder (RR=2.0 [1.9-2.1]), and bipolar disorder (RR=1.8 [1.6-2.0]). Ideation-to-attempt transition remained most strongly associated with panic disorder (RR=1.5 [1.3-1.7]), bipolar disorder (RR=1.4 [1.2-1.7]), and sexual abuse (RR=1.4 [1.2-1.7]). Most predictors were significantly but weakly associated with persistence of ideation and plan, while only physical abuse remained associated with repeated suicide attempts (RR=1.3 [1.0-1.8]). CONCLUSION: CAs and MDx are strong predictors of both onset of and transition within the STB spectrum, underscoring the importance of implementing early-life prevention interventions.< Réduire