A quantitative, comparative analysis of length of stay at four different Norwegian University Hospitals In this paper we are exploring the differences in length of stay between some of the biggest hospitals in Norway, and the cost reductions which could be achieved if other hospitals reduced their length of stay to the level of the best hospital for each diagnosis-related group. The data used in the analyses are available to the general public through the Norwegian Patient Registry. We have compared datas from Ullevål universitetssykehus, Haukeland universitetssykehus, St. Olavs hospital and UNN Tromsø, focusing on the 30 largest medical and surgical diagnosis-related groups. There is a substantial variation between length of stay among the hospitals. When looking at costs, we find that it would be possible to reduce costs and save 356 million NOK when using estimated marginal costs, if all the four hospitals performed as well as the best hospital in each diagnosis-related group. On the basis of the cost of a hospital bed we find that the total potential cost reduction for the four hospitals all together, is 688 million NOK. This is equal to 71.138 laying days or 229 beds. This does not necessarily mean that the number of beds in the hospitals should be reduced, but may instead be used as a tool for reducing the waiting lists for medical treatment and intervention, thus reducing the waiting time for each patient.