Person-centered trajectories of psychopathology from early childhood to late adolescence

Importance: The understanding of the development of psychopathology has been hampered by a reliance on cross-sectional data and symptom- or disorder-centered methods. Person-centered methods can accommodate both the problems of comorbidity and the movement between different psychopathological states at different phases of development. Objective: To examine the profiles and map the trajectories of psychopathology from early childhood to late adolescence. Design, setting, and participants: This cohort study used 2 longitudinal nationally representative community-based cohorts from the Growing Up in Ireland study covering developmental periods from early childhood to late adolescence. Data in this investigation came from children and their families who participated in all waves of cohorts recruited in 2008 (children ages 3, 5, and 9 years) and 1998 (adolescents ages 9, 13, and 17 or 18 years). Both samples were weighted to account for representation and attrition. Latent transition analyses were used to map the trajectories of psychopathology. Data were analyzed between October 2020 and September 2021. Main outcomes and measures: Psychopathology was measured using the Strengths and Difficulties Questionnaire at all ages in both samples. Results: A total of 13 546 individuals were included in the analyses. In the child cohort, mean (SD) age was 3.0 [0.01] years; 3852 (51.3%) were male participants. In the adolescent cohort, mean age was 9.0 (0.1) years; 3082 (51.0%) were male participants. Four profiles were identified in both cohorts that could be broadly labeled as no psychopathology (incidence range, 60%-70%), high psychopathology (incidence range, 3%-5%), externalizing problems (incidence range, 15%-25%), and internalizing problems (incidence range, 7%-12%). Transition between the profiles was common in both cohorts, with 3649 of 7507 participants (48.6%) in the child cohort and 2661 of 6039 participants (44.1%) in the adolescent cohort moving into 1 of the 3 psychopathology profiles at some point in development. Transition to the high psychopathology profile was most often preceded by externalizing problems. Approximately 3% to 4% of the sample had persistent psychopathology (child cohort, 203 participants [2.7%]; adolescent cohort, 216 participants [3.6%]). All psychopathology profiles were more common in boys in early life but, by late adolescence, girls were more likely to have internalizing problems. In a cross-cohort comparison at age 9, there were differences in the sex distributions of the profiles between the samples. Conclusions and relevance: Using person-centered methods, this study demonstrated that from early life young peoples' experience of psychopathology is dynamic-they can move between different mental health problems; for most children, these problems are transient, but a small proportion (fewer than 5%) have persistent difficulties. In the context of finite resources, optimizing care requires the early identification of those with persistent phenomena.