Clinimetric properties of the 21-item Depression, Anxiety and Stress Scales (DASS-21)

Background: The Depression, Anxiety and Stress Scale-21 (DASS-21) is a widely used patient-reported outcome measure. While psychometric properties of the DASS-21 have been studied, insufficient attention has been devoted to the assessment of its clinimetric properties. This study verified the clinimetric properties of the Italian version of the DASS-21 according to Clinimetric Patient-Reported Outcome Measures (CLIPROM) criteria. Methods: This is a cross-sectional study involving 951 university students from April to September 2020. Participants were asked to fill in the DASS-21 via an online survey. Participation was voluntary. Item Response Theory (IRT) models were used to test dimensionality, scalability, and sensitivity of DASS-21. Results: IRT analyses showed that the DASS-21 total score was a multidimensional measure of psychological distress. Fit to the Rasch model was achieved after excluding five misfitting items and adjusting the sample size, resulting in a 16-item version of the DASS-21. The 16-item version entailed the clinimetric property of sensitivity but included inter-correlated items. Brief versions of the DASS-21 subscales of depression, anxiety, and stress, which did not include locally dependent items, fitted the Rasch model expectations, and had an acceptable unidimensionality and scalability, were identified. Conclusion: The 16-item version of the DASS-21 may be used as an overall indicator of dysthymia and should be supplemented with the brief versions of the depression, anxiety, and stress subscales, which were found to be valid clinimetric indices.