Statistics endpoint choice

The primary endpoint is that endpoint which provides the best measure of treatment response/success and such choices are based on clinical arguments rather than statistical ones. It seems unlikely that a change in this could be justified. 

Finally most therapeutic specific guidelines contain recommendations that directly impact on statistical considerations, for example in terms of the definition of endpoints, the requirement for more than one primary endpoint, the definition of analysis sets and the choice of A. In a particular therapeutic setting it is self-evident that the requisite guidelines should be studied carefully in order to extract relevant information for statistical aspects of design and analysis. 

Bonferroni keeps the rate of type I errors down to 5 percent... In many instances, omnibus tests are not possible. If there are several comparisons to be made, each involving different groups of individuals and different endpoints, there may be no choice but to use a series of discrete statistical tests. In these circumstances, the Bonferroni correction may be applied to maintain the overall risk of false positives at the standard level of 5 per cent. 

The occurrence of individual cardiovascular events of interest in cardiovascular safety outcome trials is often low. An extremely large number of participants are therefore needed to provide the statistical power necessary to identify a statistically significant difference between tfeatment groups in the rates of occurrence of a given event of interest. One way to increase the likelihood of finding a difference between the treatment groups if one truly exists is to use a composite endpoint. A commonly employed choice is the major adverse cardiovascular events (MACE) composite endpoint, comprising of nonfatal myocardial infarction, nonfatal stroke, and cardiovascular death. An alternative is called MACE-plus. In this case, various other events such as unscheduled hospitalization for revascularization are included, each of which must also be specified in the study s statistical plan. 

---