Everyone is reporting this study as if it shows efficacy for a specific diet. Nothing could be further from the truth
-
-
Show this thread
-
Firstly, everything here is self-report. It's people telling the researchers on their online survey what they think is true I hate to break it to you, but people lie
Show this thread -
People don't always lie out of malice or spite, but because we are generally shit at remembering things
Show this thread -
So, for example, if you ask people who are using a specific diet whether they think that diet works, they're more likely to say yes EVEN IF IT DOESN'T
Show this thread -
This study could easily be just people misreporting symptoms on an online survey, which isn't unlikely because look at where they were recruited from A Facebook Grouppic.twitter.com/3fprKqEFPj
Show this thread -
If you recruit people who are in a Facebook group promoting a methodology, and then ask them if it's a good idea, what do you expect them to say????
Show this thread -
It's also worth noting that fully 1/3 of their survey responses were ineligible or incomplete Perhaps people who had bad experiences and didn't want to talk about them?pic.twitter.com/0H5kKiUPOT
Show this thread -
This is borne out pretty well by their reported adverse events. They are very low. Does that mean that the diet works or just that people who do badly on the diet don't respond to online surveys? Who knows!pic.twitter.com/zScYBY19G1
Show this thread -
This could have all been fixed with a control group. It was an online survey, so presumably not hard to include a control Weird that they didn't, isn't it?
Show this thread -
A control group - say, T1 diabetes by on a different diet - would give us some information about whether it's the diet itself making these changes or just people's bad reporting
Show this thread -
This statement of limitations is ~woefully inadequate~ for a biased online survey of 300 people...pic.twitter.com/KLq7aLKy5A
Show this thread -
And this conclusion is...stunning. To paraphrase: "If a big number of future studies are done that agree with ours, we can conclude that this study was right"pic.twitter.com/VrBeEuOegv
Show this thread -
That's like me saying that you should use a pill because eventually we'll find that it works As I said, stunning
Show this thread -
-
If THE AUTHORS OF THE STUDY didn't draw a single firm conclusion from their research - which they didn't - why would anyone else?
Show this thread -
And this quote from the
@nytimes sums it up for me Maybe the control seems to good to be true...because it is?pic.twitter.com/50sGkrgXNH
Show this thread -
Essentially, this was an online survey that proved nothing whatsoever, and had results that were almost certainly too good to be true
Show this thread -
And yet, every single news source quoted the authors who said that this diet demonstrated "exceptional" glucose control
Show this thread -
The only thing "exceptional" about this research is that people have read more into it than "interesting but needs more research"
Show this thread
End of conversation
New conversation -
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.