of a new study. But it turns out it really doesn't measure blood sugar.

So how could this conclusion be that measuring blood sugar doesn't help with weight loss if the study doesn't actually measure blood sugar?

It shows that measuring blood sugar doesn't help you lose weight. And I'm not here to say it definitely does. It does a lot of good things that I'm going to get to at the end.

But the important thing is, what does this study actually show? So the study is called effect of a personalized diet to reduce postprandial glycemic response versus a low fat diet on weight loss in adults with abnormal glucose metabolism and obesity. A randomized controlled trial, and it was published in JAMA Network Open. And so what they did was they took people with prediabetes or type two diabetes on metformin only.

So sort of like mild diabetes, I guess you could say. And they randomized them to either a standard low fat diet or this personalized response to glucose. So a personalized diet recommendation that's supposed to control their glucose.

And then they followed them over the course of six months. Now, the other important thing about this is that they both were calorie restricted. They both were trying to lose 7% of their body weight, and both groups were restricting their calories by about 500 calories per day.

Both groups had exercise recommendation. Both groups had behavioral cognitive therapy and regular meetings. So it was an intensive intervention in both groups, but the diet was different.

Now, in the end, there was no statistical difference. If you look at the charts and you look at the graphs, it looks like there was quite a big difference. But based on the cohort size and the power, there was no statistical difference.

But here's the other thing that there was no difference in. There was no difference in their macronutrients. If you look again, there looked like there's some difference, but statistically, there was no difference in carbohydrates, fats, proteins, fiber between the two groups.

So that's also very interesting. But here's the most important point. How did they determine this personalized dietary recommendation to control glucose? Well, it was based off of this protocol designed by David Zivi and colleagues in a paper titled Personalized Nutrition by Prediction of Glycemic Responses, published in Cell in 2015.

And basically, it's an algorithm that uses anthropometrics, the A one C, a questionnaire you fill out and your microbiome. So what's missing there? Your blood sugar response to food. So this is supposed to predict your blood sugar response to food, but there's no actual measurement of your blood sugar response to food.

So if we go back to that 2015 study, that's where they did wear a CGM, and they used this algorithm, and they showed that the algorithm was pretty good at determining blood sugar response to food, but it was short term and pretty good is not 100% by any stretch of the imagination. So it assumes, one, there's going to be no change over time, and it assumes that it's just going to work for everybody all the time, with precision. Well, why don't you just measure their blood sugar? That's what I find so interesting, and that's what I think this study really shows.

This study shows it was basically an attempt to revalidate this protocol, this algorithm, by Zivi and colleagues, and it showed that this algorithm itself, in this small cohort over six months, doesn't lead to excess weight loss. But that's not saying that measuring blood sugar doesn't help with weight loss. Those are two totally different things.

I'm a little baffled by the study. I think you just need more studies showing people wearing CGMs for longer period of time, changing their diets Based on certain criteria of how they respond to foods, and then deciding if that helps them lose weight or not, versus a standard low fat diet or some other controlled diet. And that's the data.

We're getting tons of data from companies like levels or nutrisense or other companies using CGMs. And there are some studies that have looked at this pilot studies. We need bigger studies, longer studies to look at it.

But look, are CGMs helpful? Is using blood sugar to guide your dietary choices helpful? I mean, absolutely. Clinically, in my opinion, it absolutely is. One, it's an accountability partner, just the behavioral feedback mechanism of wearing it.

Two, you learn a ton about how different foods affect your blood sugar. And we know, based on studies like the ZV study, like a study out of Stanford, that there's a tremendous individual variability in responses to this exact same meals. So how do you know how you respond? You test it.

You test it with a CGM? It's likely the best way to do it. Does that mean CGMs are the best tool to help you lose weight? Well, no, that doesn't prove that, but for some people, they're absolutely beneficial. And the other thing is, what criteria do you use? Do you want just the flat line, in which case you just eat fat and nothing else, or do you accept a rise up to 120 or 130, as long as it comes back down within an hour? You have to set the criteria.

So what I usually recommend for healthy people without glucose metabolism issues is you don't want your peaks to go above 120, if you can, and you want them to come down within an hour. But that's very general guidance. What I really like to do is give individual guidance if I can, but of course, that's much harder to do.

And then someone who has type two diabetes, maybe you let that go up to 140 and coming down within an hour. There are different criteria to use, and that's what needs to be tested. What's the ideal criteria to show that you're having the best metabolic response? And there are critics, right? There are definitely critics who say, look, CGM has not been validated in healthy people for weight loss, or even people in type two diabetes for weight loss.

It's just for blood sugar control shouldn't be used for weight loss. And they certainly have a point. I think the enthusiasm for it is far greater than the data.

But that doesn't mean it still can't be helpful. It means we can't say CGMs are the tool, the evidence based tool to help you with weight loss. That we clearly cannot say.

But we can say CGMs have a lot of benefit in people clinically. People see this in their practice. People see it individually.

So if your patient or you individually happen to be one of those people who get value from CGM, then by all means, you should use it. But it doesn't mean it's an evidence based tool that should be used in all people. And that's going to be the limitation for insurance coverage, to make it widely acceptable to everybody to use as a weight loss tool, that's going to need the evidence.

But one, if you can afford it, and two, if you find you're getting benefit from it, then it's probably worth trying. But a study like this doesn't show that there's no benefit to tailoring your food based on your glucose variability. For that, you got to measure your glucose variability.