For one thing, let's keep in mind that there is rarely an article that's completely "true" or completely "false". The data is there, interpretations may vary, some factors may not have been documented correctly. Medical science seems to always generalize from a fairly small set of data to the entire population, that may be flawed by itself. But I still think that a huge part of the problem is the media's attention to "not-yet-cooked" medical results ("recent studies show...") and the cut-throat fight for grants where, again, no one is going to give you money for a negative result.
The whole "recursive" thing, again, is based on the same idea of "right" and "wrong", and is totally disgusting in this case (as much as I love recursion in general).
And, yes, I am in a crappy mood today. Figures, I am sure. :-(