Yes that might be interesting. I suspect the numbers he is quoting are correct. Seems like I heard about that study some time ago where they actually got a positive result on all cause mortality. But I bet if you look at the actual numbers it will be underwhelming. Big group of people with very few deaths. But one or two deaths difference makes up the relative improvement difference.

Then the question is raised if this only has a one in a thousand (or whatever it turns out to be for the absolute risk instead of the relative risk they quote) chance of helping me, do I want to take it for the next 5 or 10 years for those poor odds? When quoting a relative risk percentage improvement number of 33% improvement it makes it sound like one in three people are helped. Yet when the absolute risk is figured it likely will be one in several hundred at best and more likely one in a few thousand. The relative risk percentage can almost become meaningless if the study group is large but very few people in either the control or the treatment group have a problem. That is the way most of the statin trials turn out. Show a modest improvement relative risk but actual risk very low because there weren't many in either group that had whatever the problem the study was looking at.

The drug companies use the lack of knowledge the public has in understanding the difference so they can fool them with advertising and make their product look lots better than it actually is.

Figures lie and liars figure. Or in the case of prescription drug advertisements they just shade the truth because they use figures the public at large do not fully understand or contemplate.