<i>Background:</i> This paper presents lifetime prevalences and estimated risks of DSM-IV psychiatric disorders from a community survey conducted in Sesto Fiorentino, Italy, using psychiatric interviewers with clinical experience and clinical instruments. <i>Methods:</i> Two thousand five hundred subjects aged 14 or more were randomly selected from the lists of 15 general practitioners (GPs) regardless of wheter or not they had consulted the GP. A three-phase design was adopted, with the GPs using the Mini International Neuropsychiatric Interview (MINI) for the first stage. All positive cases at the MINI and a probability sample of 123 negative cases were re-interviewed by psychiatrists or trained residents in psychiatry using the Florence Psychiatric Interview (FPI) at the second stage. During phase III, the subjects were administered the rating scales specific to the pathology detected by the FPI. <i>Results:</i> Two thousand three hundred and sixty-three subjects were interviewed (response rate 94.5%) by their own GP; 623 were found positive for any psychiatric disorder. The psychiatrists could re-interview 605 of these, along with a random sample of 123 negatives. Almost twenty-five percent (24.4%; 15.7% males, 31.7% females) of the population was found positive for any DSM-IV disorder during their lives. The most common diagnosis was major depressive episode, followed by anxiety not otherwise specified. Women had higher rates for most disorders. <i>Conclusions:</i> The prevalence rates for most of the disorders considered are generally comparable with the range identified by previous studies conducted in other Western countries, even though they were using different methodologies. Exceptions are represented by the high prevalence of residual categories and the lower prevalence of phobias.