Biostats question help! Thanks.

As I am very new to Biostats and research papers, I have a few questions in regards to a paper that I just read (noninferority of medicine A over D ), please explained the bolded and underscored sections:

#1. noninferiority margin of 2.00 for the upper limit of the two-sided 95% confidence interval of the hazard ratio
#2. We determined that the enrollment of 1000 patients would provide a power of 80% to show the noninferiority of A at a one-sided alpha level of 0.025, assuming an estimated 1-month incidence of the primary efficacy outcome of 7% with D.
#3 Proportional-hazards assumptions were tested with the use of a generalized Wald test for a joint hypothesis on all interaction terms specified in the statistical models

Thank you so much, please explain in layman terms, thanks!!


Less is more. Stay pure. Stay poor.
1.) the comparator treatment would have to be greater than 2 units worse than the reference drug for it to be inferior. In particular, the top bound of its 95% confidence interval on the hazard would have to be below greater than the 2 unit threshold.

2.) I don't want to just give you all the answers, so I will let you look up power, since it is pretty straightforward. However, I will note it is calculated based on the investigator's presumed treatment effectiveness. One-sided alpha 0.05 would be the same thing as the 0.05 two-sided version. It just focuses on one of the tails.