Rasch analysis of the Depression, Anxiety, and Stress Scale in patients with chronic pain

Abstract

ObjectiveQuestionnaires are commonly used to assess and screen pain-related, psychological symptoms in patients with chronic pain. In Germany, the “German pain questionnaire” provided by the German Pain Society used for this purpose includes the Depression, Anxiety, and Stress Scale (DASS). This study aims to analyze the DASS by fitting its data to the Rasch model to test the psychometric quality. Study design and settingIn this cross-sectional study, 932 patients with chronic pain answered the DASS. The 21-item short version was tested via Rasch analysis using the parameters threshold order, fit to the model, Differential Item Functioning, unidimensionality and reliability. ResultsThe Rasch analysis results showed a low reliability, misfitting items, Differential Item Functioning or multidimensionality. It was necessary to remove items from the subscales to improve fit to the Rasch model. A revised depression subscale of the DASS-21 was the only scale that achieved all the required psychometric parameters. The summation of all items to a total scale was not supported. ConclusionMore research is required on somatic free measurement of psychological symptoms in patients with chronic pain. The results demonstrate that the development of a new instrument or a revision of existing instruments for screening of psychological symptoms in chronic pain are needed.