OBJECTIVES: Substantial numbers of patients presenting to primary care suffer unrecognized disorders of mood or anxiety, potentially complicating treatment and outcome.  The objective of this study was to evaluate the validity of an electronic screening instrument based upon the World Health Organization’s Composite International Diagnostic Interview (WHO CIDI) and DSM-IV-TR designed for use in primary care.  This is a fully-structured computerized instrument designed to screen for bipolar disorder (BPD), generalized anxiety disorder (GAD), panic disorder (PD) and major depressive episodes (MDE) in  primary care patients. METHODS: A preliminary version of the instrument was piloted in individuals with known disease.    Following cognitive interviews with subjects, it was refined and tested in 3058 respondents from 29 primary care physician offices across the US.    Sub-samples were selected to receive a reappraisal interview (n=206), over-sampling on those screening positive for either the disorders. To assess validity each completed a “gold-standard” Structured Clinical Interview for DSM (SCID) administered via phone by trained clinicians.  RESULTS: Individual-level concordance was good between the screener diagnoses and the SCID assessments.   Area under the receiver operating characteristic curve (AUC), a measure of classification accuracy not influenced by disorder prevalence) demonstrated substantial agreement for MDE (AUC = 0.85).  BPD initially demonstrated fair-moderate agreement (AUC=.78), but this was improved to the “substantial” range (AUC = 0.86) with the enhancement of history of lifetime mania in the SCID interview. PD and GAD both demonstrated fair-moderate agreement (AUC = 0.79 and AUC = 0.67; respectively). CONCLUSIONS: The results demonstrate that the CIDI-based computerized screening instrument can be used to identify the vast majority of patients with a high likelihood of mood and anxiety disorders treated in the primary care setting.