Childhood Anxiety Symptoms as a Predictor of Psychotic Experiences in Adolescence in a High-Risk Cohort for Psychiatric Disorders

Background and hypothesis: When occurring in adolescence, psychotic experiences (PE), subclinical psychotic symptoms, can be an early marker of mental illnesses. Studies with high-risk populations for psychosis show that anxiety symptoms often precede the onset of psychosis. Although anxiety symptoms are frequently experienced across the continuum of psychosis, no previous study has analyzed this association using a cross-lagged panel model (CLPM) longitudinally to identify if anxiety can be a predictor of PE over time or vice versa. The aim of the current study was to investigate whether one symptom domain predicts the other over time.

Study design: 2194 children from the Brazilian High-Risk Cohort (BHRC) were evaluated at baseline (T ₀), and 76.5% completed a 3-year follow-up (T ₁) interview. Childhood anxiety symptoms and PE were assessed using a standardized self-report questionnaire at both time points. Cross-lagged panel models evaluated time-lagged associations between PE and anxiety longitudinally.

Study results: Higher levels of anxiety in childhood predicted an increase in PE levels in adolescence. The cross-lagged effect of anxiety scores at T ₀ on PE scores at T ₁ was significant (β = .03, SE = 0.01, P ≤ .001) and PE in childhood did not increase levels of anxiety in adolescence, when controlling for sociodemographic and clinical characteristics.

Conclusions: Our findings reinforce that anxiety may represent an early marker of psychosis proneness, not a consequence of already presenting PE, which can help to develop better screening approaches. Therefore, future studies should focus on identifying biological or other clinical markers to increase prediction accuracy.