TL54: Analysis of Safety Events of Interest in Placebo-Controlled Clinical Trials
*Bradley McEvoy, CDER, FDA  *Elena Polverejan, Janssen R&D, Johnson & Johnson 


Safety events of interest could be specific to a drug or a class of drugs. In placebo-controlled clinical trials, the comparison of the occurrence of events of interest in the active arms versus placebo is of particular interest. While standard statistical comparisons versus placebo are based on methods such as odds ratio, relative risk or risk difference, they don’t take into account differential treatment duration or observation time between treatment groups. Methods such as incidence rate computed per a certain person-year observation time for each treatment group as well as incidence rate ratio or hazard ratio versus placebo remediate this problem but they rely on some particular assumptions that might not hold. Graphs are also heavily used as important visual aids for this type of analysis. The participants will share their views and experience on these topics. Key Discussion Questions: -What alternative methods could be applied when there is differential observation time between treatment groups and basic assumptions such as constant hazard or proportional hazard versus placebo don’t hold? -What are the most useful types of graphs used in this particular setting?