Validity and clinical interpretability of Overall Anxiety Severity And Impairment Scale (OASIS)

Abstract Background The Overall Anxiety Severity and Impairment Scale is a brief generic measure for anxiety that encompasses frequency and intensity as well as behavioral and functional aspects of anxiety. This study was conducted to elucidate aspects of reliability, validity, and interpretability, such as equivalence of factor loadings across non-clinical and clinical populations, convergence and discriminance of related variables, and performance of detecting diagnostic and medical status of anxiety disorders. Methods Non-clinical and clinical Japanese populations were taken from a panelist pool registered with an internet survey company (total n =2830; 619 panic disorder, 576 for social anxiety disorder, 645 for obsessive-compulsive disorder, a 619 for major depressive disorder, and 371 for non-disorder panelists). Conventional measures of anxiety, depression, mental health and measures for discriminant validity were administered in addition to OASIS. Results Exploratory and confirmatory factor analyses indicated good fit to data for the one-factor model of OASIS. Multi-group confirmatory factor analysis showed the equivalence of the factor loadings between those of non-clinical and clinical subsamples. The OASIS reliability was confirmed by internal consistency and test–retest coefficients. Receiver operating characteristic curve analyses showed that OASIS and conventional anxiety measures have fair performance for detecting diagnostic and medical status as anxiety disorders. Limitations Participants were limited to a Japanese population of people who had registered themselves at an internet survey company. Conclusions Along with useful information to interpret OASIS, the results suggest the reliability and validity of OASIS in Japanese populations. These results also suggest cross-cultural validity.