Uh, what? At worst, that's indicative of nothing, not affirmation of the negative hypothesis. Think of it this way: if you are healthy, you are more likely to

Uh, what? At worst, that's indicative of nothing, not affirmation of the negative hypothesis. Think of it this way: if you are healthy, you are more likely to get the flu vaccine and less likely to die of any reason. If you are unhealthy, you are less likely to get the flu vaccine and more likely to die of any reason. Notice that neither of these statements say the following, "Patients who receive the flu vaccine are [more/less] likely to die of the flu then they would have been." Instead, it simply makes the statement that you cannot treat statistics as affirmation of flu effectiveness.