Disability in bipolar I disorder: Application of Mokken scaling analysis and the graded response model to the World Health Organization Disability Assessment Schedule 2.0

Background The World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) evaluates an individual's functioning and disability within the conceptual framework of the ICF. The present study examines the measurement properties of the WHODAS 2.0 in patients with bipolar disorder using Mokken scaling analysis (i.e., monotone homogeneity and double monotonicity models) and the graded response model. No previous studies applying these models to this instrument were found. Methods A sample of 291 patients with bipolar disorder (42.6% males) was tested. Results The WHODAS 2.0 domains showed strong unidimensionality, with no items being omitted. In addition, the analysis of invariant item ordering showed that the items of each domain formed a hierarchical scale, with the exception of the ‘Life activities’ items for employed persons or students and item D4.5 ‘Sexual activities’ in the ‘Getting along’ domain. The WHODAS 2.0 domains and the whole scale also showed excellent reliability in bipolar disorder. Limitations Although the study was limited to patients in Spain, the use of non-sample dependent procedures minimizes this limitation since the results are independent of the sample used. Conclusions The WHODAS 2.0 contains six strong unidimensional domains that are hierarchical and reliable for detecting disability in bipolar disorder, although caution should be exercised with regard to some items.