Has the p value needed to stop a trial based on placebo patients being harmed been disclosed by the fda? Is p <.05 the value they use for the end of a study and interim recommendation, or is that a presumed value?