SUMMARYAims— To assess in a national sample the ability of GPs to detect psychiatric disorders using a clinical vs. a standardized interview and to characterize the patients that were falsely diagnosed with an anxiety or affective disorder.Methods— This is a national, cross—sectional, epidemiological survey, carried out by GPs on a random sample of their patients. The GPs were randomly divided into two groups. Apart from the routine clinical interview, the experimental group (group A) had to administer the Mini—International Neuropsychiatric Interview (MINI).Results— Data was collected by 143 GPs. 17.2% of all patients had a clinical diagnosis of an affective disorder, and 25.4% a clinical diagnosis of an anxiety disorder. In group A, the number of clinical diagnoses was about twice that of MINI diagnoses for affective disorders and one and a half times that for anxiety disorders. The majority of clinical diagnoses were represented by MINI subsyndromal cases (52.3%). Females showed a higher OR of being over—detected by GPs with anxiety disorders or of not being diagnosed with an affective disorder. Being divorced/separated/widowed increased the OR of over—detection of affective and anxiety disorders. The OR of over—detection of an affective or an anxiety disorder was higher for individuals with a moderate to poor quality of life.Conclusions— In the primary care a gap exists between clinical and standardized interviews in the detection of affective and anxiety disorders. Some experiential and social factors can increase this tendency. The use of a psycho.Declaration of Interest: GlaxoSmithKline provided unrestricted economic and organizational support to the study. No further declarations on other form of financing or any other involvement that might be considered a conflict of interest in connection with the submitted article.