DSM-5 and ICD-11 as competing models of PTSD in preadolescent children exposed to a natural disaster: assessing validity and co-occurring symptomatology

Background: Major revisions have been made to the DSM and ICD models of post-traumatic stress disorder (PTSD). However, it is not known whether these models fit children's post-trauma responses, even though children are a vulnerable population following disasters. Objective: Using data from Hurricane Ike, we examined how well trauma-exposed children's symptoms fit the DSM-IV, DSM-5 and ICD-11 models, and whether the models varied by gender. We also evaluated whether elevated symptoms of depression and anxiety characterized children meeting PTSD criteria based on DSM-5 and ICD-11. Method: Eight-months post-disaster, children (N = 327, 7-11 years) affected by Hurricane Ike completed measures of PTSD, anxiety and depression. Algorithms approximated a PTSD diagnosis based on DSM-5 and ICD-11 models. Results: Using confirmatory factor analysis, ICD-11 had the best-fitting model, followed by DSM-IV and DSM-5. The ICD-11 model also demonstrated strong measurement invariance across gender. Analyses revealed poor overlap between DSM-5 and ICD-11, although children meeting either set of criteria reported severe PTSD symptoms. Further, children who met PTSD criteria for DSM-5, but not for ICD-11, reported significantly higher levels of depression and general anxiety than children not meeting DSM-5 criteria. Conclusions: Findings support the parsimonious ICD-11 model of PTSD for trauma-exposed children, although adequate fit also was obtained for DSM-5. Use of only one model of PTSD, be it DSM-5 or ICD-11, will likely miss children with significant post-traumatic stress. DSM-5 may identify children with high levels of comorbid symptomatology, which may require additional clinical intervention.