Survival with Rare Outcomes: Examples from Diagnostic Device Trials
View Presentation View Presentation
*Estelle Russek-Cohen, U.S. Food and Drug Administration 

Keywords: survival, sample size

In survival analysis, when event rates are low, the effective sample size is often considerably less than the number of patients. I would like to discuss two examples from diagnostic device trials. The first example has come up in the evaluation of prognostic markers for patients diagnosed at an early stage of a chronic disease such as cancer. The second example has come up in a trial in which “adverse events” are the primary endpoint.