Interesting info about pharaceutical trials

I just learned this in my medical bioethics class. Pills are compared to placebo to test their effectiveness. So if a pill helps 10% of people and the sugar pill helps 5%, it is considered "effective". Then it is marketed to, and prescribed to, 100% of people. I'm not sure what would be a better system, but I think we should be aware that when a company says that their product was found to be effective, that it could be found effective as little as 10% of the time.