(PS3-83) Assessing Consistency of Longitudinal Self-report of Traumatic Events Among Children in Acute Psychiatric Care

Background: By the age of 16, more than two-thirds of children and adolescents in the United States have experienced a traumatic event, such as abuse, neglect, bullying, natural disasters, or traumatic loss.  Around 16% of children and adolescents who have experienced trauma go on to develop PTSD, and trauma is a risk factor for many other mental health challenges including anxiety and depression. As the rate of traumatic events is particularly high in clinical samples, routine screening for exposure to traumatic events as part of initial mental health assessment is recommended as part of clinical care. Repeated screening can be helpful, as self-reports of childhood traumatic experiences often change over time. The purpose of this study was to understand the consistency of self-reports of traumatic events among children in an acute level of psychiatric care.

Method: We assessed self-report data of 69 children ages 7-13 (M =10.51, SD =1.53) admitted to a mental health partial hospitalization program. Data were collected at admission and discharge from the program. The primary measure was the Child and Adolescent Trauma Screen 2 (CATS-2), which asks if they have experienced a list of 15 events. The patients were 6% Asian, 12% Black, 65% Non-Hispanic White, and 20% Hispanic. Forty-one percent identified as female, 39% as male, and 20% as non-binary/trans.

Results: Fourteen percent of patients reported experiencing the same potentially traumatic events on the CATS-2 at admission and discharge. Seventy-one percent of patients endorsed at least one event at discharge that they did not endorse at admission; 70% percent endorsed at least one event at admission that they did not endorse at discharge. Forty-nine percent of patients reported at least one event at discharge that they did not report at admission and at least one event at admission that they did not report at discharge. On average, patients endorsed 4.32 events at admission (SD=2.42) and 4.64 events at discharge (SD=2.78). The correlation between the number of events endorsed at admission and discharge was r(67)=.70, p< .001. Stronger correlations were observed between time points among the younger children (7-10-year-olds; r(30)=.79, p< .001) compared to the older children (11-13-year-olds; r(35)=.60, p< .001).

Discussion: These data highlight the prevalence of inconsistent self-reporting of traumatic events among children with acute mental health problems and suggest that child self-reports of their experiences of specific traumatic events may have low test-retest reliability. While the events being endorsed at discharge and not admission can potentially be explained by an event occurring during the course of treatment, there is no clear explanation for why some events may be endorsed at admission but not discharge. Future research could explore if this elevated endorsement at admission correlates with a general elevated endorsement of distress at that time point. Increased consistency in the reporting by younger children may be explained by a higher frequency of assistance in filling out measures, i.e. questions being read aloud.