The point is that you have to evaluate the significance of the numbers against the background data, rather than in isolation. RRD encourages you to ignore the background data.

Why do they say “meteoric rise. meteors fall, don’t they?

Appealing to statistics like “numbers needed to treat” is potentially very misleading as well. Both heart attack and stroke often have very severe consequences for quality of life in survivors. But certainly “saving” one life in 100 is a proportion that is either big or small, depending on the costs (money and side effects) of the treatment to the other 199 folks living on your envelope.

Anyway, playing a little fast and loose, let’s imagine that your risk of heart attack in your lifetime is 30% on placebo, 15% on the drug, and that the same numbers apply to stroke. Let’s also fantasize that they’re independent. That would mean you have a 51% chance of experiencing either a stroke or a heart attack or both on placebo, a 28% chance on the drug. Is that a big enough effect to be meaningful to you? Ultimately the answer doesn’t depend on what statistic you use to express exactly the same data. It depends on whether the cost (money, side effects) is worth the benefit. Which is the part of this question you sidestepped.

At the end of the day, these statistics are somewhat hard to interpret. I’d rather die from stroke or heart attack than in a skydiving accident, because the latter would be more likely to afflict me earlier in life. But if there’s a drug that will reduce my risk of heart attack and stroke, it probably won’t cause a skydiving accident. Unless it encourages me to continue skydiving later in life. At any rate, it’s never a great idea to read too much into absolute values, even if they’re small. No one dies from highly likely events.

More men +0.6%
More smokers +0.3%
More metabolic syndrome +0.8%
Family history of premature CHD +0.6%

AZ were clearly paying attention when you lectured them on tricks big pharma employ.

This could explain the greater than expected benefit…..

You seem to have forgotten people don’t go into journalism because they’re able to understand statistics. If they were good with math or science, they’d have chosen a different career.

But selective by whom? When making these judgments people rely on those we expect to know the overall risks such as A & E departments or organisations which compile accident statistics. It’s not just “imagination”.

That Uni. of Chicago study is behind a paywall, but one of the authors has it available at http://faculty.chicagogsb.edu/christopher.hsee/vita/Papers/SpecificationSeeking.pdf

A large proportion of the ‘news’ is simply press releases – sometimes paraphrased or rewritten, but largely unresearched – and this is no exception. I’ve no doubt that the press release had these ‘facts’ in it, and of course the aim of the release is to promote their drug so they want to make the effectiveness look as large as possible. This is particularly the case where they’re proposing that people self-medicate with this drug every day.

I surprised my GP at discussing with him how the outcome might be presented in terms of relative and absolute risk statistics!

From reading Ben’s column and book I look forward to discussing the marvels of the placebo effect with him, as well as considering trial protocols. Should be fun!

http://www.colbertnation.com/the-colbert-report-videos/210357/november-12-2008/cheating-death—women-s-health

At about 2:10 is when he talks about rosuvastatin and comes to about the same conclusion.