Latent structure and measurement invariance of the post-traumatic stress Scale-10 in cancer patients

OBJECTIVE: The aim of this study was to evaluate the factor structure and measurement invariance of the German version of the Post-Traumatic Stress Scale-10 (PTSS-10) in a sample of patients with solid cancer or hematopoietic neoplasms.

METHOD: A sample of 3841 cancer outpatients was included in this study. The factor structure was analyzed using confirmatory factor analysis (CFA). Multi-group CFA was employed to assess the measurement invariance of the scale concerning age, sex, cancer type, and socio-economic factors.

RESULTS: The results indicate that a one-factor structure is appropriate in this population only when excluding two items, namely item 7 (frequent swings in mood) and 9 (fear when approaching the place of the event). This model achieved only partial scalar measurement invariance. Item 4 (being easily startled) was identified as a source of non-invariance between women and men.

CONCLUSIONS: This study recommends excluding items 7 and 9 when using the PTSS-10 in cancer patients, introducing the PTSS-8-CANCER score for the evaluation of PTSS-10 data. For group comparisons, item 4 should also be excluded from scoring. Overall, the study indicates that the PTSS-10 questionnaire, in its original form, lacks the necessary psychometric strength to serve as a reliable screening instrument for post-traumatic stress disorder in cancer patients.