Comments | phygelus: revolutionary indeed

phygelus

revolutionary indeed

Feb 07, 2006 16:43

ganked from prettyoctopussy
"These studies are revolutionary," said Dr. Jules Hirsch, physician in chief emeritus at Rockefeller University, who has spent a lifetime studying the effects of diets on weight and health. "They should put a stop to this era of thinking that we have all the information we need to change the whole national diet and make everybody ( Read more... )

Comments 11

salimondo February 8 2006, 01:40:59 UTC

Probably the reporter garbling the JAMA embargo release. If I had more free time tonight I'd actually dig around.

Subgroup speculation still interesting if larger studies back it up. Even a rank layman like me looks at those numbers and thinks, "my God, I wonder if those three cancers in a thousand have genetic markers in common, they can use this to identify women with elevated genetic risk of dietary-linked cancer." For the rest, blessed freedom to eat buttered chicken aplenty!

Best to all I'll miss at the con. Maybe next year.

hints phygelus February 8 2006, 01:55:40 UTC

1) They will very likely find the effect more pronounced in their follow-on study. Why? A new and larger study is unlikely, as one of their experts said the studies covered in the article were so large and expensive as to likely be the final word on the subject.

2) This will almost give it away: "Dr. Venkman was intruiged by the psi data, even though it was not statistically significant. The people who wore blue shirts had a 9 percent higher rate of success at the Zener-Rhine test... He added that the study investigators would continue to follow the people who wore blue shirts to see if the effect became more pronounced."

survivor bias salimondo February 8 2006, 02:09:11 UTC

Well, that too. I'm still stuck on the odd math of how they got a "9 percent" lower rate from 45/1000 to 42/1000.

Re: survivor bias phygelus February 8 2006, 02:17:10 UTC

It's even worse than that, it's optional stopping.

(The comment has been removed)

phygelus February 8 2006, 02:22:01 UTC

Well, you know, if you had an MD, you could keep doing studies until you found that the effect was statistically significant. That would be your god-given right. And odds are, no one would catch the problem in peer review: this stuff goes on in medical journals and the soft sciences all the time.

baal_kriah February 8 2006, 18:39:11 UTC

It's very common, and even enabled by folks like the FDA. Big-money backers just keep the "research' going until they get the results they want.

revlainiep February 8 2006, 13:42:03 UTC

Though I personally wonder if gass fed red meat would have any nutritional value whatsoever. :-)

Thread 5

treecat February 9 2006, 00:46:08 UTC

http://www.medpagetoday.com/PrimaryCare/DietNutrition/tb/2636... )

phygelus February 9 2006, 04:00:15 UTC

I'm just chortling over the pull-quotes about this being the end of the era of quacks pretending they know how to recommend the ideal diet, just because they got proved wrong in a big way this one time. Not too likely! If you notice the NYT article has lots of whining about how the study didn't test the most fashionable current reccomendations!

amberite July 6 2006, 11:01:29 UTC

Re: breast cancer, higher levels of bodyfat raise estrogen level and breast size.

Pretty much anything that raises levels of estrogen, or mimicks estrogen, will increase the rate of breast cancer by some percentage; this includes any kind of hormonal medication and the plastic by-products in the water supply. Most things that promote growth or fluctuation have some overlap with cancer risk, in general, in most areas.