Assessing the factor structure of posttraumatic stress disorder symptoms in war-exposed youths with and without Criterion A2 endorsement

Posttraumatic stress disorder's (PTSD) tripartite factor structure proposed by the DSM-IV is rarely empirically supported. Other four-factor models ( King et al., 1998 and Simms et al., 2002) have proven to better account for PTSD's latent structure; however, results regarding model superiority are conflicting. The current study assessed whether endorsement of PTSD's Criterion A2 would impact on the factorial invariance of the King et al. (1998) model. Participants were 1572 war-exposed Bosnian secondary students who were assessed two years following the 1992–1995 Bosnian conflict. The sample was grouped by those endorsing both parts of the DSM-IV Criterion A (A2 Group) and those endorsing only A1 (Non-A2 Group). The factorial invariance of the King et al. (1998) model was not supported between the A2 vs. Non-A2 Groups; rather, the groups significantly differed on all model parameters. The impact of removing A2 on the factor structure of King et al. (1998) PTSD model is discussed in light of the proposed removal of Criterion A2 for the DSM-V.

Research highlights▶ Both the King and the Simms model fit adequately in the A2 Group. ▶ Neither the King or Simms model fit adequately in the Non-A2 Group. ▶ Invariance testing revealed that the A2 Group significantly differed from the Non-A2 Group in all factor structure parameters. ▶ The A2 Group generally had higher intercepts, variances and factor means compared to the Non-A2 Group.