Posttraumatic stress disorder (PTSD) factor structure among survivors of the 2010 Haitian Earthquake: Adjudicating heterogeneous post-disaster PTSD dimensionality

Background: There is inconsistent empirical evidence supporting a universal underlying factor structure of posttraumatic stress disorder (PTSD).

Method: We assessed model fit of six theoretical factor structures of PTSD: 1-factor Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV; 3-factor DSM-IV (arousal, avoidance, and intrusion); 3-main factor (arousal, avoidance, and intrusion) and 1-hierarchical factor DSM-IV; 4-factor King 1998 (avoidance, hypervigilance, emotional numbing, re-experiencing); 4-factor Simms 2002 (avoidance, dysphoria, hyperarousal, intrusion); and 4-factor DSM-5 (arousal, avoidance, intrusion, negative mood and cognition) models. Confirmatory factor analyses were conducted on PTSD symptoms identified through the Posttraumatic Checklist – Civilian Version in a population-based sample of 1,302 survivors of the 2010 Haitian Earthquake.

Results: All models adequately fit the data, indexes ranged from: root mean square error of approximation (RMSEA) 0.056–0.069; comparative fit index (CFI) 0.885-0.927; Tucker Lewis Index (TLI) 0.865-0.915; weighted root mean square residual (WRMR) 1.768-2.148; Akaike information criterion (AIC) 24,768.459-29,346.352; and Bayesian information criterion (BIC) 24,952.178-29,584.705. The King 1998 and Simms 2002 models fit slightly better than the DSM-IV 1-factor and DSM-5 models. However, the 3-factor DSM-IV (arousal, avoidance, and intrusion) model fits the sample best (x²=593.257, 116 degrees of freedom; RMSEA=0.056; CFI=0.927; TLI=0.915, WRMR=1.769; AIC=24,760.459; and BIC=24,952.178).

Implications: The tight range in model fit statistics, consistent with published heterogeneous factor structures, suggests that empirical-based model selection is insufficient to universally characterize PTSD in this context. Adjudication of PTSD model selection should be theoretically driven.