Clinical predictors of at-risk mental states in general adolescent population: a 2-year follow-up study

Adolescence is a critical period of neurodevelopment during which psychopathological trajectories associated with the risk of psychosis may emerge. The construct of the At-Risk Mental State (ARMS) enables the identification of subclinical presentations preceding psychosis; however, its course and predictors in the general adolescent population remain poorly understood. Most available studies have focused on clinical samples or cross-sectional designs, limiting our understanding of the longitudinal and transdiagnostic nature of psychosis risk in the general adolescent population, where psychotic-like experiences are often transient. A longitudinal observational study was conducted with two assessments separated by 24 months in a community sample of school-attending adolescents (T1: n = 697; T2: n = 575; ages 11-17 years). Psychosis risk screening was conducted using standardized screening questionnaires, followed by the Comprehensive Assessment of At-Risk Mental States (CAARMS) clinical interview for diagnostic confirmation. Psychopathological variables (depression, anxiety, attention-deficit/hyperactivity disorder [ADHD], trauma, suicidal behavior, and emotional and behavioral problems), addictions, and parental socialization were assessed. Descriptive analyses, intergroup comparisons, ordinal logistic regression analysis, and structural equation modeling were applied to identify predictors of ARMS. At the 2-year assessment, 3% of participants were classified as ARMS. Only 0.9% of those initially classified as healthy controls developed an ARMS during follow-up, and diagnostic stability from baseline to 2 years was limited. Adolescents with ARMS already showed, at T1, higher levels of depression, anxiety, trauma, ADHD symptoms, suicidal ideation and planning, as well as poorer academic performance and greater exposure to bullying. In adjusted regression models, in addition to ARMS status at T1, only depression, hyperactive-impulsive ADHD and suicide planning emerged as independent predictors of ARMS at T2. These findings suggest that ARMS in the general adolescent population is embedded within dynamic, transdiagnostic psychopathological trajectories. Depression, hyperactive-impulsive ADHD symptoms, and suicidal behavior emerge as key clinical signals for early risk identification. Together, these results underscore the potential value of preventive strategies in adolescence that prioritize the early detection and targeted management of affective distress, neurodevelopmental difficulties, and suicidality within community and educational settings.