Continuous glucose monitors reveal variable glucose responses to the same meals
35 comments
·May 10, 2025csours
blitzar
It is wonderful that we have things like CGM's and patients can establish their own baselines and their own evidence profiles on a high(ish) frequency basis.
Such tools also enable studies that can be done at a scale and cost level that is reasonable and can push forward the communal knowledge base.
csours
Yup, and this was the original dream of Theranos - lots of measurements, lots of data. It's a nice dream, shame about the lies though.
mjburgess
A counter-point, in a certain sense: when the conclusions of scientific papers (in these softer science fields), contradict common sense, they tend to be unreproducible; the ones which don't, are.
The problem with studying humans is, roughly, the central limit theorem doesnt work: properties of biological and social systems do not have well-behaved statistics. So all this t-test pseudoscience can be a great misdirection, and common sense more reliable.
In the case where effect sizes are small and the data generating process "chaotic", assumptions of the opposite can be more dangerous than giving up on science and adopting "circumstantial humility". (Consider eg., that common sense is very weakly correlated across its practicioners, but "science" forces often pathological correlations on how people are treated -- which can signficantly mangify the harm).
crazygringo
> when the conclusions of scientific papers (in these softer science fields), contradict common sense, they tend to be unreproducible; the ones which don't, are.
Citation needed?
I don't know what would lead to that conclusion. And it would seem to run counter to the entire history of the field of psychology, for example.
blindriver
I've been diagnosed with Type 2 diabetes and wear a CGM. I find that my blood glucose spikes much later than what is suggested. For example, I might eat a meal and my blood glucose doesn't spike until after 2 hours. When it spikes, I see the body react with insulin and it drives my blood sugar levels down, so I'm not sure if I actually do have T2D or if this is just how my body works.
Another thing I've noticed is that if I eat a very rice-heavy meal, my blood sugar levels may rise throughout the night. I don't think this is insulin-resistance but rather my body digesting the rice. You can't expect the body to digest all the rice in 2 hours, can you, there certainly must be parts that are protected from the stomach acid until much later in the digestion process. So that feeds into the high blood sugar levels overnight in my opinion.
I think if anything, CGMs have opened up the idea of what diabetes really is and how different bodies handle blood sugar. I think I'm borderline T2D, not full-on T2D despite what my doctor says, and I've started wondering if my blood sugar has always been high, but normal for me. On average it's about 120 mg/dL, but I do see my body react properly to new sources of blood sugar and drive it back to "normal" levels, so the idea that I have insulin resistance doesn't make sense to me.
bilsbie
Interesting. What’s your a1c level ?
I think you could have a point at around 100-105 baseline but 120 seems too high.
blindriver
It's high, like 6.7-6.9. It doesn't make sense to me because that would suggest average blood glucose of over 150, but it's objectively around 120.
taeric
Odd, I would have expected this to be somewhat the case? Specifically, I would expect your activity leading into the meals, along with your expectations of what you will be doing after, would have some impact? Probably more impact from how hungry you were going into a meal. If you were already sated, for example, I'd expect your body to largely try and push the meal through as fast as it can.
Neat to see what other people's priors are, on this.
coldpie
> Odd, I would have expected this to be somewhat the case?
I don't think anybody was expecting to be surprised by this study. In practice, most science is pretty boring and rarely breaks expectations. But being unsurprising does not mean it's not worth doing. A lot of studies are simply validating expected outcomes and providing foundational data points for future studies to refer back to. For example, a future study might use this study to justify funding ("as shown in Study 2025.abcd, glucose is highly variable... we propose to further study this by controlling for ... which will help us understand the influence of ..." etc etc).
taeric
Ah, great point! I definitely got the impression that this was a surprise, but I can't say why I got that impression. That said, I did not intend this as a criticism of running the study! Quite the contrary, glad they did it. I was only meaning my comment to be contra the sense of surprise I had.
perrygeo
I'm very sensitive to sugar and starch when "at rest". If I wake up first thing in the morning and drink an orange juice or have a bowl of white rice before bed, my blood sugar and mood are out of control. But the same foods mid-day before/after/during intense mental or physical work are very well tolerated if not beneficial. So my priors agree with yours, timing and context absolutely matters.
layer8
The study used 7-day rotating meals, so one would expect some consistency on average in a weekly rhythm, but it also only covered 14 days, so only two data points per meal and participant, if I read this correctly. In that case, I’m not surprised either that there is a lot of variation, also considering that this is a new meal regimen for the participants.
blitzar
Non expert here ... My understanding was sleep, stress and many other variables all impact these things significantly (before we even consider food). Having a different context when you then add the same food thus should not result in the same outcome.
pfortuny
People think medicine is Physics. They really do.
taeric
I think I agree with the idea you are saying. That people think you can formulate our body to ignore a lot of the state that it holds.
I'll note that even basic physics has that problem. Try explaining to a 5th grader why a feather would fly in the same arc that a rock will take, if there was no air.
pfortuny
Well, I did not want to say "Mathematics" because it sounded a bit exaggerated. Of course you are right.
jzacharia
Shouldn't come as a surprise - there are so many factors involved in glucose response to food that it's almost impossible to replicate a glucose response even with a controlled intake. Sleep, activity levels (before, after, during, even days before, etc), stress, hormones, all of these are major factors involved in how glucose is metabolized.
francisofascii
Sounds expected? If you are glycogen depleted, it will cause less of an insulin spike. This sounds like if you fill up the same type of car with 5 gallons of gas. Sometimes it reports full, and sometimes it doesn't. That sounds odd, until you accept that you are never fully sure the gas level you started at.
siliconc0w
Is this already pretty expected? We already knew that glucose response was highly dependent on time of day, pre or post meal movement, and whether the meal is moderated by other components like fats that can slow down the immediate impact.
stranded22
Yep.
My wife is T1D and this is infuriating for her.
She’ll think that she’s cracked it, and then the next day, with the same meal at the same time, her sugar levels go high.
Her words: there is nothing else where you have studied for over 30 years and STILL feel like you know nothing.
It is incredibly demoralising for her sometimes - especially when she’s suffering also from a high/low sugar level. I have the upmost respect for anyone having to do the amount of work, to get to zero (sometimes).
coldpie
I'm also T1D and yep. That's how it goes. Despite having one of the same two breakfasts pretty much every workday for a decade, it's a crapshoot whether I'll be 300 (very high) when I get to work or 50 (very low) or 110 (good). I just have to adjust when I get in to the office. Most days I deal with it fine, but every once in a while I get fed up with it and want to throw something.
je42
More data can help inform semi closed loop systems to deal with this variability.
For example, AAPS has since version 3.2 dynamic IFS. ( https://androidaps.readthedocs.io/en/latest/DailyLifeWithAap... )
For me this works quite well
null
ddorian43
How about just not eat carbs at all and have more consistent & lower blood sugar? (this was an interesting case https://www.youtube.com/watch?v=CG8UU7P8FBU)
I do keto diet long term but for other reasons, often the epilepsy version where it's more strict and higher fat.
jzacharia
Keto works well, but I'm concerned about the extreme fat intake. I did manage to lower my A1C from 9s to mid 5s using Keto as a T1D, but eventually settled into a high protein, moderate fat, low-ish carb diet and that has worked pretty well without being hyper-restrictive. Heavily inspired by the late Dr. Richard K. Bernstein.
ddorian43
I target for high ketones (1.5+) & low glucose 4-5mmol (blood test 1 hour after wake up) and higher protein interferes with both of them.
msarrel
Fascinating. I've always felt that this was true with my own body, yet every medical professional I've spoken to said that I was wrong.
jzacharia
Hate to sound like "that guy" but a majority of medical professionals are running their practice on outdated data or false studies funded by pharma.
ggandv
80% of the variation due to individual differences OR measurement error.
zingababba
"Additionally, the summarized study did not record the timing of snack and water consumption. This is relevant because the sequence and timing of subsequent meals (or snacks), as well as the food processing and variability, can influence postmeal glucose responses." - probably a factor, hydration status is huge. This result is kind of like 'duh' though.
endoblast
One factor that scientific food experiments don't seem to include is gluttony. Presumably because it is subjective (though nonetheless real for that). Yet if people are motivated by the amount of comfort and/or pleasure they obtain from eating their favourite foods this may have an effect on their physiological response.
Some assumptions are so obvious no one bothers to state them, or even remember them.
Assumption: Medical professionals are trained to use Evidence Based Medicine (EBM).
One might assume that EBM means something specific, and I'm sure it does, but that specific thing is different for different people.
One thing that EBM sometimes means is: common sense is no substitute for evidence. There are uncountable times that common sense has been wrong in the medical context.
So, you have a lot of people commenting here that this is obvious common sense, but many medical professionals will pull out a reference chart of caloric content and glycemic index and say "look at the evidence".
So, it is very useful to do studies that bring evidence to common sense.
---
My other pet peeve with EBM is that it does a poor job of understanding that different interventions work for different people - it is time consuming and expensive to do that kind of investigation, thus some/many medical professionals do not understand or believe it.
This comment is intended as a critique, not a dismissal.