Non Inferiority Trial

Hi there,

I have a trial in which one group is receiving a drug and the other group is getting placebo. We want to try and show that the placebo is non inferior to the drug group. So we decided that non inferiority will be established if the lower bound of the 95% confidence interval does not fall more than 10% below that of the drug group. We have calculated the mean and confidence intervals for each group on a cognitive test. The problem is that the placebo group is not normally distributed. Is there a way to look at non inferiority non parametrically? We have tried log transforming the data and it still is non parametric. Any suggestions would be greatly appreciated.



Omega Contributor
Tell us a little bit more about how these data look. Perhaps you can get a little more creative in your transformations. However you always run the risk of not being able to readily interpret the results if you get too far removed.