This result was mentioned previously on HN: http://news.ycombinator.com/item?id=783912 http://news.ycombinator.com/item?id=783912 Apparently, not only are some

This result was mentioned previously on HN: http://news.ycombinator.com/item?id=783912 http://news.ycombinator.com/item?id=783912 Apparently, not only are some new drugs failing to pass a test showing that they are more effective than placebos, but old drugs that previously passed a placebo test might fail if they were tested today. They give a number of possible reasons in the article why this might occur. One hypothesis was that suitable trial patients are not as sick as they used to be, since the really bad cases today would be on drugs that didn't exist back then (which makes them ineligible for the trial). Another hypothesis was that trials today are performed in more third-world locations, where the quality-of-life difference between standard medical care and a drug trial is more pronounced, amplifying the placebo effect. But no one knows for sure why it's happening.