Importance: The suicidal risk of psychopathology in adolescence is suggested to differ based on its longitudinal trajectory, but the comorbidity of these symptom trajectories has not been well examined. This study comprehensively clustered trajectories of multiple psychopathological and behavioral symptoms and examined their associations with suicidal thoughts in adolescence.
Objective: To determine which categories and trajectories of psychopathological and behavioral symptoms are associated with suicidal thoughts in adolescence, accounting for comorbid symptoms.
Design, setting, and participants: This population-based cohort study in Japan used data from the Tokyo Teen Cohort (TTC) study, which was established in 2012 and is currently ongoing. Data from 3 waves of surveys conducted at ages 10, 12, and 16 years from October 2012 to September 2021 were used. Of the adolescents in the cohort, participants with at least 2 evaluations of psychopathological and behavioral symptoms were included. Data were analyzed from December 2022 to March 2023.
Exposure: Latent class growth analysis was used to cluster the trajectory of each psychopathological and behavioral symptom.
Main outcomes and measures: The associations between symptom trajectories and suicidal thoughts at age 16 were examined. Suicidal thoughts were assessed using a self-report questionnaire. Psychopathological and behavioral symptoms were assessed using the 8 subscale scores of the caregiver-report Child Behavior Checklist.
Results: This study included 2780 adolescents (1306 female participants [47.0%]). Of the 1920 adolescents with data on suicidal thoughts, 158 (8.2%) had suicidal thoughts. The median (IQR) age was 10.2 (10.0-10.3) years at the first evaluation, 11.9 (11.8-12.1) years at the second evaluation, and 16.3 (16.1-16.5) years at the last evaluation. The clustering pattern of trajectories varied depending on symptom categories. After adjusting for each symptom trajectory and confounders, adolescents with persistent high withdrawn symptoms (odds ratio [OR], 1.88; 95% CI, 1.10-3.21) and those with increasing somatic symptoms (OR, 1.97; 95% CI, 1.16-3.34) had a significantly higher risk of suicidal thoughts than adolescents without these symptoms. There was no interaction between these symptom trajectories and the risk of suicidal thoughts.
Conclusions and relevance: This cohort study found that persistent withdrawn symptoms and increasing somatic symptoms during early to midadolescence were associated with an increased risk of suicidal thoughts in midadolescence, even after accounting for comorbid symptoms and confounders. Attention should be paid to the suicidal risk associated with these symptoms, particularly when they persist or increase in the longitudinal follow-up.