Study regarding stevia and insulin release

(Siobhan) #1

Can someone tell me if I’m interpreting this study correctly? Basically the study was done comparing aspartame, stevia, and sugar (sucrose, or table sugar).
The results were that none of the subjects ate more or less based on the sweetener (sugar or not) that they ingested prior to the meal. Aspartame and stevia did not spike blood sugar, while sucrose obviously did.
The weird portion is the part about insulin wherein the study seems to show that both stevia and aspartame spiked insulin more, and over a longer amount of time compared to sugar. I’m not sure if I’m reading the data correctly here or not.

Interestingly enough a separate study on erythritol, and chicory root both found no impact on insulin or blood glucose. I’d have to dig those up though.

EDIT: Erythritol study:

EDIT2: I can’t find the chicory root study, though I’d just seen it a few nights ago :confused:

What's worse for you occasionally: artificial sweetener or sugar?
Studies showing Sweeteners that Raise Insulin Levels?
Is stevia safe to use?
June 2018 ZornFast
No sweeteners at all, anybody else?
Insulin spike with sweeteners
(Bart) #2

Very interesting indeed. What I found interesting was how they listed the foods the participants where allowed to eat for their meals. Did they all eat the same thing? Where they able to choose what they ate. All they talked about was total calories consumed, I may have missed it but it did not break down the macros. What would the responses have looked like for those who ate keto? Very interesting and cool stuff though. Makes me want to look into this even more.


There’s a more recent but similar study by a different team that seems to contradict the findings of the first study, here:

Ad libitum lunch intake was significantly higher for the NNS treatments compared with sucrose (P=0.010). The energy ‘saved’ from replacing sucrose with NNS was fully compensated for at subsequent meals; hence, no difference in total daily energy intake was found between the treatments (P=0.831).

As for interpreting the data… maybe @erdoke can help out with that.

Here’s the two relevant charts from the first study.(


Insulin INDEX (defined as Δ insulin/Δ glucose)


Thanks for the study.

“Can someone tell me if I’m interpreting this study correctly?”

No, you didn’t interpret it correctly. The study shows that stevia and aspartame spike insulin less. See here:

“Consumption of stevia in preloads significantly lowered postprandial insulin levels compared to both aspartame and sucrose, as well as postprandial glucose levels compared to sucrose.”

And earlier:

"Postprandial Insulin Levels

AUC analyses also indicated that there was a significant main effect for type of sweetener consumed on postprandial blood insulin levels, F(2, 60) = 6.48, p = .003. Post-hoc comparisons revealed that postprandial insulin levels were significantly lower in the stevia condition compared to both the aspartame (p = .04) and sucrose conditions (p = .003; see Figure 3). Specifically, postprandial insulin levels were significantly reduced at 30 and 60 minutes after the test lunch meal in the stevia condition compared to the aspartame condition (all ps < .05). Postprandial insulin levels were also significantly lower at 20 minutes following consumption of the preload, as well as 30 and 60 minutes after the test lunch meal, in the stevia condition compared to the sucrose condition (all ps < .05). Postprandial insulin levels at 20 minutes after consumption of the preload were significantly lower in the aspartame condition compared to the sucrose condition (p < .01)."

What you may have been confused about was the IGI (insulinogenic index) which is a ratio of insulin over glucose. If both are high numbers you could end up with a ratio nearing 1 (i.e sucrose) whereas if the insulin response was low and the glucose even lower you could end up with a high ratio (i.e stevia). So while the second ratio shows up as a higher IGI it isn’t related to whether or not the actual insulin level was high.

So, in short, you misinterpreted. The study shows that the insulin response to stevia is low and the glucose response was even lower. It also shows that it was a significant difference.

It is a very supportive study if you’re intending to consume stevia. Probably the best possible results you could want from a non-nutritive sweetener.

No sweeteners at all, anybody else?
(Siobhan) #5

Thanks so much! I knew I wasn’t 100% on my interpretation of the study and it turns out I’m completely wrong. Ha.
Your explanation now makes it way easier to understand, so I guess that means Stevia is up there with Erythritol with ones to choose if you’ve got a need. You are why this forum exists :slight_smile:


You’re welcome. Glad I could help. It’s an awesome forum.


So, something I don’t see mentioned: how much of each were they putting in the food/pre-load? Were they putting in the same amount for each, or were they adjusting for relative sweetness (in which case, significantly less stevia should be consumed than sucrose) or doing something else to negate the higher sweetness at equal amounts?

That’s just to answer the question as to whether the better results are due to less being used, for instance. Effectively, that still shows well for the sweetener that can sweeten with less. Or, on the other hand, was a preload made super sweet (or have that stevia weird after-taste) to the extent that it made subjects not want to eat as much 20-mins later? It’s also notable (as they do mention) that no control was used, which would give more information.


I don’t know but my educated guess is they always compare based on standardised sweetness factors. Stevia is 200 times sweeter than sugar. It would make no sense to compare gram to gram unless you were researching the effects of accidental ingestion.


Exactly, but that leaves open the question as to whether the insulin levels were significantly lower with Stevia simply because the dose was significantly lower. Insuline is spiked less because there is less triggering insulin, just like if you used 200 times less sugar (there’s not as much sugar to deal with, so less insulin is used to deal with it).

I’m guessing this means that the food was adjusted for sweetness, rather than equal amounts, unless the researchers added something to curve sweetness though:

There were no differences, however, in the hedonic ratings of the stevia and sucrose preloads in terms of appearance, aroma, sweetness, or texture.

It’s also interesting to note that the chart showed the increase in total calorie consumption for the day was almost entirely in the pre-load meals, not in the amounts of the other meals. Which makes me wonder if the extra mean 300ish calories from those meals were actually from the sucrose itself (which required about 200 times as much and is caloric).

For many purposes, this would practically still favor stevia and aspartame, of course, since one way or the other, if less is required to get the same effect on satiety and taste, and it has less of an impact otherwise, then it’s the better option. It may also just indicate that what we want is simply things with higher sweetness ratings, which would mean, for instance, that High Fructose Corn Syrup may really be better than sucrose due to it’s higher sweetness rating making it so less is required to achieve the same sweetness when replacing one with the other (this seems reasonable for other reasons as well, though I recall Dr. Lustig’s presentations showing that when sucrose was replaced with hfcs in many foods and drinks, more was put in for some reason, and the foods simply became sweeter). It would also say the key is simply to look for higher sweetness rated products, and make it a good idea to avoid those alternative sweeteners that have less sweetness. But that would require further study.

(Sylvia Sage) #10

I saw that study, and I have been obsessing on it ever since. Years ago I tried to get into a ketogenic state. I spent five days eating nothing but meat and diet soda. My ketone strips never turned pink. I broke the diet because it made me sleep most of the time. I’m wondering if the diet soda could have impaired the fat burning.

Tomorrow I will start a fast and attempt to get into a ketogenic state, consuming nothing but water. If I succeed in that and have enough will power, I will reintroduce small quantities of single foods to see what allows me to stay in a ketogenic state, and what disrupts the ketogenic state. If a small quantity of food disrupts the ketogenic state, I want to see how long without food I will have to go to get back into that ketone if state. Eventually I want to test the artificial sweeteners. Of course, all this will depend on my will power.


Wow. Very interesting

(Sylvia Sage) #12

Another studyhat was kind of intriguing showed that rats produce insulin in response to the sweet taste but not in response to the taste of starch. Sorry I don’t know how to post links. The insulin was in response to the taste of fructose and artificial sweeteners.

(Sylvia Sage) #13

I may have to review those graphs again, but I remember at the two hour mark, glucose levels were lower in the sucrose group. Don’t be hypnotized by what you read. Just look at the graphs. The study was obviously conducted with a confirmation bias in favor of artificial sweeteners. Also, in terms of satiety, 290 calories of sucrose preloading might have been just as satisfying as 493 calories of sucrose. They used different caloric amounts to prove their points. 493 calories of sucrose was less insulinogenic than 290 calories of aspartame.

(Sylvia Sage) #14

I shared this information with my diabetic brother. He already knew this because he has to monitor his insulin. Aspartame makes him sick and is worse for him than sugar.

(Sylvia Sage) #15

Look again at the two hour mark. The sucrose group had lower levels of glucose. I believe you interpreted it correctly. I would love to see a similar study comparing starch to sucrose. Artificial sweeteners may be harmful to diabetics. Also sucrose is often blamed for diabetes, but I don’t believe sugar causes diabetes.


Good point, but I don’t think they gave subjects 290 calories of sweetener. That would be a lot of non-caloric sweetener! Rather, the base they used for the preload was 290 cals, then they added the sucrose to one which bumped up the total calories to 493, and sweetener to the other 2. Granted, that doesn’t say what lesser amounts of sugar would have done. Perhaps they used a sweetness index of some kind to determine the amounts of each.

I believe they are comparing AUC values for the entire time frame, not just a one point comparison at 120 min. whic they deem not “significant”.


Here’s the graph from the other study I mentioned above. Done in 2016.

The non-sucrose sweeteners show an increase over sucrose when a meal is given at 60 minutes after the drink.


There’s a good explanation about the effects of refined starches here:

Overall, lower glucose and insulin was seen in the SNTT (starch) interventions than in the OGTT ones. SNTT AUCs accounted for 82.7% of glucose AUCs. … SNTTs could be used to replace OGTTs in certain populations, such as T2DMs, since it amplified the incretin effect.

(Sylvia Sage) #19

Did they all have the same lunch?

(Sylvia Sage) #20

What is AUC and GLP?