In the wonderful age of the recursive, self amplifying, instant news Internet few people have yet to comment on the Enhance trial which failed to show enhanced reduction in vessel wall thickness by the addition of Zetia to simvastatin in patients with heterozygous hypercholestrolemia.
The media shy Steven Nissen tells us what the study "showed". "The study showed it matters how you lower cholesterol -not just how much you lower cholesterol." As indicated below I am not sure what it showed.
In record time, the American College of Cardiology (ACC) has issued reassurance that we should not panic. Since apparently no one was harmed in the trial,it is not clear to me who would panic anyway except the drug company that makes Vytorin. See here for the ACC summary of the trial and here for the ACC 's comments on the trial and the non-panic advisory.
This was a trial that looked at a surrogate maker-what the bottom line is regarding the drugs that lower cholesterol is the reduction of heart attacks or cardiac events. Enhance looked at vessel wall thickness and there was no shrinkage in the Vytorin group over that seen in the simavastatin alone group. Since we have been warned to not take trials that look at surrogate markers too seriously we should not maybe take any of this too seriously. Had a marked advantage accrued to the Vytorin group we would have been appropriately warned by whomever to not go wild and prescribe Vyotrin to everyone we see on the basis of a surrogate marker trial. Similarly with a negative trial we perhaps should restrain ourselves from calling all the patients on Vytorin and urging them to stop the pills.
All of which gets us to Wittgensteine's ruler.
Unless you have confidence in the ruler's reliability, if you use a ruler to measure a table you may also be using the table to measure the ruler."
Note: I cannot find that Wittgenstein actually said that.I am quoting from a delightful book by Nassim N. Taleb, entitled "Fooled by Randomness" onto whose trail I was put by the blog Healthcare Epistemocrat.