Statistical Analysis of Ulcer Trials with Intermittent Treatment

Abstract

In studies of intermittent (episodic) treatment, patients receive treatment only during acute episodes, not at other times. The number of recurrences and total time spent receiving treatment are often used as measures of the efficacy of treatment. Low values for these variables, however, may indicate either a successful treatment during the preplanned study period or a failure resulting in a discontinuation of treatment at an early stage of the study period. Therefore, these variables should not be used to evaluate the outcome of treatment strategies. Other variables, such as the proportion of patients cured during the initial episode, time to first recurrence, time to final treatment failure, total time in remission, and ranked outcomes from best to worst, are more meaningful and describe different aspects of the outcome of treatment.