“Houston, we have a problem.”
Apparently you can’t believe everything you read. Whether it is a problem of vanishing results, reliance on unrepeatable or outdated information, or the fact that not all publications are created equally, there seems to be a serious problem with relying on medical publications. While research papers provide valuable clues, only after those results are repeatedly verified can they reach the status of fact. And even that could change over time as the sample set grows beyond what is practical for a single study.
It seems that even rigorous science can fall prey to the human failing that people see what they want to see. This results in selective reporting as well as selective publication (rarely do negative results get submitted). And in the pharmaceutical industry there is the ever-present threat of monetary corruption. Note that this should not create a panic of Nihilism, but rather instill healthy skepticism and diminish false hope.