This study used latent profile analysis (LPA) to identify distinct psychological profiles among 7,879 officially reported maltreated children (ages 9–17) in South Korea (2023–2024), then examined how sex, school level, re-abuse history, and maltreatment type predicted profile membership via multinomial logistic regression.
Four profiles emerged: **Severe Symptoms**, **Elevated Symptoms**, **Stable**, and **Elevated Symptoms with Mild Self-Harm**. Female sex and higher school level predicted membership in the Severe Symptoms group. Physical abuse, sexual abuse, and re-abuse history were more common in the Severe Symptoms group compared with the Stable and Elevated Symptoms with Mild Self-Harm groups.
- Data are confidential and cross-sectional, preventing causal inference or replication by outside researchers. - The sample is limited to *officially reported* maltreatment cases in South Korea, likely underrepresenting milder or unreported cases and limiting generalizability. - Profile-level psychological instruments were developed and standardized specifically for this South Korean maltreated-child context, so cross-cultural applicability is uncertain.
Maltreated children are not a uniform group—screening should aim to detect which of several distinct symptom clusters a child falls into, particularly flagging girls and older students for more severe symptom profiles. Tailored, trauma-informed interventions—rather than one-size-fits-all approaches—appear warranted given the clear heterogeneity in risk.
