Silly people. You see...the way a signicance test works is it looks at both the sample size and variance among patients. So you don't have to argue about whether the differences are meaningful. The test does the math and TELLS you. If the sample is low, but variance is low, then you can bank on the results. There were only 7 patients on drug. But ALL 7 had dramatic changes in triglyceride, HDL (!), and the rest. The significance test tells you to bank on it.