In the trial, 20 adults were fed one diet (low-carb or low-fat) for 2 wk, then changed to the other diet for 2 wk. But there was no “wash-out” period to allow effects of the first diet to fade. The authors just averaged the data to show the low-carb diet ⬆️calorie intake 2/10

HOWEVER, they did not recognize at the time a massive carry-over effect – about 2000 calories a day! According to Stats 101, carry-over effects invalidate cross-over trials, as we consider in our paper (also see quotes at the bottom of my blog post) 3/10 @davidludwigmd/designed-to-fail-44d278e591a6 class="text-blue-500 hover:underline" target="_blank" rel="noopener noreferrer">https://medium.com/@davidludwi...

Trial Design 101: “The recommendation ... is to avoid the problems caused by differential carryover effects at all costs by employing lengthy washout periods." But here, there was no washout period ... 4/10 #main-content class="text-blue-500 hover:underline" target="_blank" rel="noopener noreferrer">https://online.stat.psu.edu/st...

How did the Author Reply address this fatal design flaw? Mostly misdirection. TECHNICAL NOTE: There are two ways to address carry-over effects. One is a 2-step process. Test for carry-over. If present, then throw out the biased second period. 5/10 https://www.nature.com/article...

But this test is frowned upon by statisticians, as we explicitly stated (The Author Reply misrepresents us as saying this approach is "widely accepted," see highlighted excerpt from our paper) The issue is what to do with a massive, visually evident carry-over (see below) 6/10

The other approach, per the statistician cited (& acknowledged) in the Author Reply, is that validity: "must depend on medical opinion as to whether the wash-out period can be regarded as achieving its aim" A CRITICAL point omitted from the Reply 7/10 https://onlinelibrary.wiley.co...

Anyone who, based on medical opinion, thinks that the (nonexistent) washout period achieved its aim here (i.e., to prevent the first diet from affecting the second) should not be practicing medicine. 8/10

Because of a critical study design flaw, we’re forced to choose between two bad options: 👉 Discard the second period, resulting in relatively subtle bias 👉 Keep the second period, resulting in massive bias 9/10

But all roads lead to Rome: 👉2-step method, study is invalid 👉Medical opinion criterion, also invalid Thus, the trial conclusions are simply wrong So why does it remain uncorrected, with ongoing bias to the evidence base on which patient care and public health depend? fin

PS, here is the Share Link for full free access to our paper https://rdcu.be/eBSS6

@davidludwigmd @NatureMedicine @NIH @grok podsumoj to w 3 zdaniach

@davidludwigmd @NatureMedicine @NIH Hall & his team should republish with just the first period showing a slight advantage to low carb. Publishing results that don't confirm one's bias is nothing to be ashamed about; instead, it's a badge of honor. Only true scientific thinkers are capable of that

@davidludwigmd @NatureMedicine @NIH Anybody that has tried low carb knows.

1/9) In the most gentlemanly manner possible, @realDaveFeldman just released the lipid-nerd equivalent of “Here, hold my beer”on the KETO-CTA study. Drawing from his slide deck and some new reveals, I’m going to summarize what you need to know. First big question: Why have we

One man, Hal Cranmer (@HalCranmer), is totally revolutionizing assisted living! He shares the story of his first obese resident’s amazing weight loss following a program of fasting, exercise, and a keto/carnivore diet.

Misinformation sometimes spread by healthcare professionals: • fiber prevents/fixes constipation (it doesn't) • excess protein damages kidneys (it doesn't) • fat makes you fat (it doesn't) • low carb is dangerous (it isn't) • red meat is bad for you (it's not) • all

When you buy into the fake idea that it’s “all downhill” after 40, 50, 60 or 70, you’re setting yourself up for all kinds of bad outcomes. It’s not too late to improve your life. It’s not too late to get lean and fit.