Here, in one study, is what burns me about NUTRITIONAL EPIDEMIOLOGY. The BLUEBERRY people fund a study to see if blueberries improve INSULIN RESISTANCE. But they might not! So there's also a bunch of pre-registered SECONDARY OUTCOMES. Study here:https://academic.oup.com/ajcn/article/109/6/1535/5499342 …
-
Show this thread
-
Turns out, blueberries have NO EFFECT on insulin resistance. But WAIT! They have a small effect on ARTERIAL STIFFNESS! Let's run with that. Oh, and all those OTHER outcomes (lung & cognitive function, liver fat), we just won't mention those. Pre-reg here:https://www.clinicaltrials.gov/ct2/show/NCT02035592?term=NCT02035592&rank=1 …
3 replies 2 retweets 22 likesShow this thread -
I'm no STATISTICIAN but if your signficance test is p < .05 then on average ONE IN TWENTY measurements will meet it by chance BY DEFINITION. So all you have to do is measure 20 things and odds are SUCCESS! 40 things is better. Make it stop. OK done.
11 replies 4 retweets 49 likesShow this thread
The blueberry people have been running studies that are somehow always positive for years. Every time I've seen a study in the news about blueberries, it's funded by the same group!
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.