We compare two scoring systems PRISM1 and PIM2 in ability to predict mortality of newly admitted patients in emergency department. Both scores had good discrimination but both had poor calibration and poor performance in almost all subgroups. PIM2 is better than PRISM1 because it is easy to collect, and because the data are collected at “point-of-care”, however, further studies are required to validate the PIM2 score to our environment