The factor structure of the International Trauma Questionnaire – Heywood cases in confirmatory factor analysis

Background: A number of studies have tested the factor structure of the suggested ICD-11 symptom criteria for PTSD and complex PTSD (CPTSD) across various trauma populations, finding support for two different models in line with the ICD-11 theoretical rationale.

Objective: Here, we aim to explore the factor structure of the Danish version of the International Trauma Questionnaire (ITQ) by testing two alternative factor models that have previously gained support in a large sample of treatment-seeking veterans.

Method: Treatment-seeking Danish soldiers and veterans (N = 599) recruited from the Military Psychology Department in the Danish Defence completed the International Trauma Questionnaire (ITQ). Confirmatory factor analysis (CFA) was used to assess fit of a first-order and a second-order model.

Results: Both models fit the data well but displayed latent variable correlations above 1 and negative variances (known as Heywood cases), indicating model misspecification or other problems. The specification problems included the latent variable Affect Dysregulation.

Conclusions: Based on our results and results reported in previous CFAs of the ITQ, we suggest consideration of the proposed models. While many previous studies did find support for the models with no indications of misfit, others find Heywood cases concerning the same items and latent variables as our analysis. Hence, models of (C)PTSD based on the ITQ should be carefully evaluated and interpreted.