The potential benefits and risks of prophylactic treatment for stress ulcer haemorrhage remain controversial. Analysis of previous studies indicates that 1. the incidence of haemorrhage and the mortality from haemorrhage have declined over the last two decades even in the absence of prophylactic treatment; 2. all drugs used for prophylaxis reduce the incidence of haemorrhage, and all drugs are comparably effective. 3. There is, however, no evidence that prophylactic drug therapy reduces the mortality or that the drugs used for prophylaxis differ in their effect on mortality. 4. A higher rate of pneumonia is seen only with antacid treatment.