BackgroundThe primary health care version of the ICD-11 is currently being revised. AimTo test two brief sets of symptoms for depression and anxiety in primary care settings, and validate them against diagnoses of major depression and current generalised anxiety made by the CIDI. MethodThe study took place in general medical or primary care clinics in 14 different countries, using the Composite International Diagnostic Interview adapted for primary care (CIDI-PC) in 5,438 patients. The latent structure of common symptoms was explored, and two symptom scales were derived from item response theory (IRT), these were then investigated against research diagnoses. ResultsCorrelations between dimensions of anxious, depressive and somatic symptoms were found to be high. For major depression the 5 item depression scale has marked superiority over the usual 2 item scales used by both the ICD and DSM systems, and for anxiety there is some superiority. If the questions are used with patients that the clinician suspects may have a psychological disorder, the positive predictive value of the scale is between 78 and 90%. ConclusionThe two scales allow clinicians to make diagnostic assessments of depression and anxiety with a high positive predictive value, provided they use them only when they suspect that a psychological disorder is present.