Measurement and Factor Structure of Developmental Trauma Disorder Symptoms in Children Involved in Child Welfare

Abstract

Childhood trauma is associated with a wide array of neurodevelopmental, physiological, psychosocial, and emotional challenges beyond those captured by posttraumatic stress disorder (PTSD)—especially in instances of multiple and/or repeated traumas and traumas that occur in the context of a caregiving relationship. As a result, children who have experienced complex developmental trauma often receive multiple diagnoses concurrently and across their lifespan. Indeed, childhood trauma has been identified as a central transdiagnostic risk factor in the etiology of numerous mental disorders and in research examining the existence of a general psychopathology factor (p-factor) (Caspi et al., 2014). However, recent criticisms of p-factor modelling have questioned the interpretation and cross-study comparability of work in this area, calling for a more theory-driven approach to defining the general factor. Emotion dysregulation has been identified as a potential mediator in the relationship between childhood trauma and the transdiagnostic risk of psychopathology, and some researchers interpret the general factor as emotion dysregulation. However, researchers have yet to test emotion dysregulation as a reference domain for the p-factor or the structure of psychopathology within a Developmental Trauma Disorder (DTD) framework. This study attempted to address these gaps in a sample of (N = 555) children involved in the Ontario child welfare system who have experienced maltreatment. In the first part of the study, I assessed the degree to which the Assessment Checklist for Children (ACC) captures the proposed DTD diagnostic criteria. In the second part, I tested the factorial structure of DTD symptoms using Confirmatory Factor Analysis, including a single factor, correlated factors, second-order, fully symmetrical bifactor, and bifactor(s-1) model with emotion dysregulation as the general factor reference domain. The results identify gaps in the ACC when applied to the DTD framework. Further, the results suggest that the bifactor(s-1) model fits the data best and provides the most interpretable results with meaningful clinical practice and research implications.