Empirically Derived Subtypes of Youth Withdrawn Behavior Across Eight Years: A Latent Class and Latent Transition Analysis

Withdrawn behavior is present in many psychological problems and it is often used interchangeably with other psychological constructs, making accurate classification difficult. In an effort to better understand the classification and developmental course of withdrawn behavior in youth, the current study used parent-report to assess 649 Dutch youth at three time points across eight years, using the Withdrawn/Depressed (W/D) scale of the Child Behavior Checklist (CBCL). Mean age of participants at each time point are 7.52 (Wave 1), 11.77 (Wave 2), and 15.53 (Wave 3) years. Latent class analysis and latent transition analysis was employed to empirically derive distinct subclasses of withdrawn youth and to examine how these youth transition between classes over time. Most youth fell within a low symptom class (Time 1 = 79.4%, Time 2 = 79.2%, Time 3 = 68.5%). A shy/secretive class also emerged (Time 1 = 20.6%, Time 2 = 20.8%, Time 3 = 31.5%). For girls, the interpretation of the shy/secretive class changed at Time 3 (e.g., ages 14-17 years), such that the majority of girls in this class also exhibited depressed mood. The majority of youth remained in the same class over time. This study used a model-based person-centered approach to derive classes of withdrawn behavior over time. The W/D subscale could be used as a brief screening tool for early detection of the shy/secretive behavioral phenotype, which may be a risk factor for later psychopathology. It is recommended that future studies identify predictors of class transition and investigate whether withdrawn classes predict diagnostic outcome.