Non-Inferiority Trials

Sven: Non-inferiority trials have a number of subtle differences that I think we need to take into account. You really need a strong rationale for why you'd undertake a non-Inferiority trial. And often it's around economics and can you get away with using one treatment without detriment. Marianne: In standard superiority analysis intention to treat is the most conservative approach. But if you flip, you have to flip that on its head when you think about non- inferiority because actually an intention to treat pools people towards the null. High crossover can be really problematic in a non-in feriority design.