In a previous post by Growacet, he linked to a paper entitled "Rationales for an accurate sample size evaluation." I haven't read the whole thing, but part of the paper is called "The ingredients for sample size calculations" and states:

"There are four essential ‘ingredients’ used to prepare a suitable sample size: (I) effect size; (II) variability; (III) significance level; (IV) power.......In addition to the four basic ‘ingredients’ above, there are also other factors (‘seasonings’) that the investigator may factor in during the sample size calculations. These may be taken into account with some of the sample size calculation methods available (‘recipes’): 1) Underlying event rate of the condition under study (prevalence rate) in the population; 2) Expected dropout rate; 3) Unequal ratio of allocation to the study arms; 4) Specific considerations related to the objective and design of the study."

Resverlogix and the Clinical Steeric Committee had to make a lot of assumptions going into writing the recipe for BETonMACE. It might taste better or worse than originally planned. The Sample Size Reestimation Analysis may allow them to tweak their recipe now that they know more about the variable listed above.

BearDownAZ