This report summarises the literature on geographical variations in hospital admission rates in the UK and internationally. It is restricted to studies of inpatient admissions. It concludes that the examination of the variations provides invaluable insights into the nature and extent of medical uncertainty, and that this information is essential to proper decision making. The authors discuss methodological issues to do with the measurement of admission rates. They suggest that future work should concentrate on the feedback of information decision makers and the setting up of studies to examine the outcome associated with different levels of intervention.