Kidney stones represent a common condition characterized by significant morbidity and economic costs. The epidemiology of kidney stones is not completely understood and may vary substantially based on geographic, socioeconomic and clinical factors; the present study aims at defining the prevalence and diagnostic patterns of kidney stones in a cohort representative of the general population in Florence, Italy. A sample of 1,543 adult subjects, all Caucasians, was randomly selected from a population of over 25,000 subjects followed by 22 general practitioners (GPs). Subjects were administered a questionnaire requesting the patient's age and sex, any history of kidney stones and/or colics and the prescription of kidney ultrasound (US) examination. GPs data-bases were also interrogated. Crude and adjusted prevalence proportions and ratios (PRs) with corresponding 95% confidence intervals (CIs) were computed. Furthermore, the association between the practice pattern of each physician with respect to US prescription and the prevalence of kidney stones was investigated. The overall prevalence of kidney stones was 7.5% (95% confidence interval 6.2, 8.9%), increasing with age until 55-60 years and then decreasing. About 50% reported recurrent disease. There were no significant differences in prevalence among males and females. GPs who tended to prescribe more US examinations were more likely to have more patients with kidney stones (adjusted PR 1.80, 95% CI 1.11, 2.94; p = 0.020). The present study confirms both the high prevalence and the regional variability of kidney stones. Practice patterns may be involved in such variability.