I previously went into some depth about the JUPITER trial of the statin Crestor, in patients with elevated c-reactive protein, because of the extensive and largely one-sided press coverage:
Donald Light has just added a nice commentary to the skeptical side of the equation, which you can find at:
Among several good points Don makes is that the study was stopped early because of the supposed success of Crestor in preventing cardiovascular events--despite the very low absolute size of that "success." The early stoppage of trials is becoming a staple in industry-sponsored research, notes Light. The result is that the trial may not have gone on long enough to show the true side effect profile of this drug in an elderly population. If the Kentucky Derby were run like a modern clinical drug trial (if Light is correct about this), it would no longer be a mile-and-a-quarter race. Rather, the race would be declared over at whatever point the drug company's favored horse was ahead.