Screening for bipolar spectrum disorders : a comprehensive meta-analysis of accuracy studies

Abstract

Background: Bipolar spectrum disorders are frequently under-recognized and/or misdiagnosed in various settings. Several in fl uential publications recommend the routine screening of bipolar disorder. A systematic review and meta-analysis of accuracy studies for the bipolar spectrum diagnostic scale (BSDS), the hypomania checklist (HCL-32) and the mood disorder questionnaire (MDQ) were performed. Methods: The Pubmed, EMBASE, Cochrane, PsycINFO and SCOPUS databases were searched. Studies were included if the accuracy properties of the screening measures were determined against a DSM or ICD-10 structured diagnostic interview. The QUADAS-2 tool was used to rate bias. Results: Fifty three original studies met inclusion criteria ( N ¼ 21,542). At recommended cutoffs, summary sensitivities were 81%, 66% and 69%, while speci fi cities were 67%, 79% and 86% for the HCL-32, MDQ, and BSDS in psychiatric services, respectively. The HCL-32 was more accurate than the MDQ for the detection of type II bipolar disorder in mental health care centers ( P ¼ 0.018). At a cutoff of 7, the MDQ had a summary sensitivity of 43% and a summary speci fi city of 95% for detection of bipolar disorder in primary care or general population settings. Limitations: Most studies were performed in mental health care settings. Several included studies had a high risk of bias. Conclusions: Although accuracy properties of the three screening instruments did not consistently differ in mentalhealthcareservices,theHCL-32wasmoreaccuratethantheMDQforthedetectionoftypeIIBD. More studies in other settings (for example, in primary care) are necessary.