Considerations for peptic ulcer trials.

A drug trial concerned with peptic ulcer treatment has to be carefully planned to make possible a valid conclusion. Calculations of the number of patients required have to be made early. This depends upon factors such as: if the comparison is to be made between two active drugs or an active drug and a placebo, if the aim is to show that two treatments are almost equal or if a certain difference is expected, the number of groups to be compared, and if intermediate checks are to be made. Thus, from a few dozen patients up to several thousand may be needed. If a multicenter dosing is adopted, which is often necessary, the minimum number of patients required from each centre should be specified and a good reporting system has to be created. When presenting the results, it is preferable to calculate a confidence interval for the difference between the treatments--especially in small studies where the risk is false conclusions would otherwise be large.