Test accuracy of the Overall Anxiety Severity and Impairment Scale (OASIS) for screening anxiety in adults – systematic review and multiple thresholds meta-analysis

BACKGROUND: Given the high prevalence of anxiety disorders and their frequent under-recognition, the use of questionnaires for screening anxiety is under debate. OBJECTIVE: To evaluate the test accuracy across cut-offs of the Overall Anxiety Severity and Impairment Scale (OASIS) against (semi-)structured interviews for screening any anxiety disorder (AAD) and generalised anxiety disorder (GAD) in adults. METHODS: We searched Embase, MEDLINE, PubMed-not-MEDLINE subset and PsycINFO up to 15 July 2025. We assessed the internal and external validity using the QUADAS-2-tool. We used the multiple thresholds model to estimate summary sensitivity and specificity across cut-offs, the area under the curve (AUC), and the cut-offs that maximised the Youden-Index. RESULTS: We included four studies with 736 participants, all having specific chronic somatic conditions. For the recommended cut-off ≥8, we estimated a summary sensitivity of 0.44 (95% CI 0.27 to 0.63) and a summary specificity of 0.89 (95% CI 0.81 to 0.94) for AAD and a summary sensitivity of 0.58 (95% CI 0.40 to 0.74) and a summary specificity of 0.87 (95% CI 0.79 to 0.93) for GAD. The Youden-Index was maximised at cut-off ≥5 and ≥ 6 in terms of AAD and GAD, respectively. We estimated an AUC of 0.77 (95% CI 0.74 to 0.79) for AAD and 0.81 (95% CI 0.77 to 0.84) for GAD. CONCLUSIONS: Only a few studies have examined the accuracy of the OASIS in a screening approach, all conducted in populations with somatic diseases. The AUC is comparable to other anxiety questionnaires. The recommended cut-off needs reconsideration. REGISTRATION: CRD42023485827.