The 17-Person Study That Ruined Splenda

One study, 17 people, and somehow it rewired what millions of people believe about artificial sweeteners.

The study was published in 2013 and it tested sucralose, which is the sweetener sold as Splenda, on obese individuals who had not previously used artificial sweeteners regularly. Researchers gave them sucralose and then 75 grams of glucose, which is a significant sugar load, and measured the insulin response. What they found was a 20 percent increase in insulin compared to water. That number got picked up, stripped of context, and turned into a rule.

But before that result means anything, you need to understand how insulin actually works in this context.

Your pancreas releases insulin in response to rising blood glucose, and the amount it releases is calibrated to how much glucose it senses. That is the main loop. But there is a secondary system built into that loop, something called cephalic phase insulin release, which is a small anticipatory insulin spike that your body triggers in response to sweet taste before glucose even reaches your blood. Your tongue sends a signal, your pancreas primes itself, and a small amount of insulin goes out ahead of the sugar. This is not a malfunction. It is your body getting ready.

The mechanism that makes this work involves sweet taste receptors, and those same receptors exist not just on your tongue but in your gut and in your pancreas directly. When those receptors detect something sweet, whether it is sugar or a non-caloric sweetener, they can amplify the insulin response that glucose is already driving. The word there is amplify. They do not generate an insulin response on their own. They turn up the volume on a signal that already exists.

That distinction is the entire story.

When researchers tested sucralose by itself, without any carbohydrates present, there was no meaningful effect on blood glucose or insulin. The receptor activation had nothing to amplify, so nothing happened. The 20 percent increase in the 2013 study only appeared because sucralose was given alongside 75 grams of pure glucose. Remove the glucose, and the effect disappears. Which means the headline, "sucralose spikes insulin," is technically describing something real while completely misrepresenting what is actually happening.

There is also the dose problem. The sucralose amount used in that study was approximately five times what you would find in a typical diet soda. So even within the specific condition where the effect exists, the study was not modeling normal consumption.

A 2020 study published in Cell Metabolism added more detail here. Researchers found that short-term sucralose consumption paired with carbohydrates impaired neural and metabolic sensitivity to sugar, but that effect required the combination. Sucralose without carbohydrate did not produce the same result. The researchers described this as a learned association your body builds when sweet taste and glucose arrive together repeatedly, which has more to do with the pairing than with the sweetener itself.

Now look at what the broader research base actually says.

A systematic review published in 2025 that covered 101 articles and 100 experiments found no meaningful effect of non-nutritive sweeteners on blood glucose or insulin as a category. A network meta-analysis of 36 trials covering 472 participants reached a similar conclusion for sweetener-containing beverages. These are not one-off studies. These are researchers pulling together everything available and asking what pattern emerges, and the pattern does not support the idea that artificial sweeteners meaningfully disrupt insulin or blood sugar under normal conditions.

Aspartame, which is used in Diet Coke and many sugar-free products, has been studied more extensively than almost any food additive in existence, and the evidence consistently shows no effect on glucose or insulin. Stevia, which is a plant-derived sweetener, has data suggesting it may actually reduce insulin compared to both sugar and aspartame in some contexts. A 2010 study that compared stevia, aspartame, and sucrose directly found that postprandial insulin levels were lower in the stevia group than in either of the other conditions.

So the landscape across sweeteners is not uniform, and that is worth knowing.

Sucralose is the one with a conditional caveat. The condition is specific: you would need to be consuming sucralose alongside a high-carbohydrate meal, at a meaningful dose, and based on the 2013 study, the effect was observed in obese individuals who did not regularly use artificial sweeteners. Whether that result generalizes to someone who uses Splenda regularly and has a different metabolic baseline is not established. The 2022 Cell study also showed that microbiome responses to non-nutritive sweeteners vary significantly between individuals, which means personalized response is a real variable rather than a minor footnote.

If you want a practical takeaway: stevia appears to have the cleanest profile across the full body of research, aspartame has the most accumulated evidence behind it and consistently shows no effect on glucose or insulin, and sucralose is fine in isolation but if you are drinking something sweetened with it alongside a large carbohydrate meal, switching to stevia in that specific situation is a reasonable and low-effort adjustment.

The larger point is about how health information actually spreads. A study with 17 subjects, a supraphysiological dose, and a result that only appeared in one specific condition became the foundation for a belief that millions of people now hold with complete confidence. The fitness industry did not lie exactly, because the 20 percent increase was real. It just never told you the conditions required to produce it, because conditions make for bad headlines.

What you should be afraid of is not Splenda. It is the version of nutrition advice that takes one data point and removes everything that made that data point specific, because that is how ordinary things become threats and the actual variables never get discussed at all.

---

References

1. Pepino MY, Tiemann CD, Patterson BW, Wice BM, Klein S. Sucralose affects glycemic and hormonal responses to an oral glucose load. Diabetes Care. 2013;369:2530-2535. PMID: 23633524. Source
2. Dalenberg JR, Patel BP, Denis R, et al. Short-Term Consumption of Sucralose with, but Not without, Carbohydrate Impairs Neural and Metabolic Sensitivity to Sugar in Humans. Cell Metabolism. 2020;313:493-502. PMID: 32130881. Source
3. Anton SD, Martin CK, Han H, et al. Effects of stevia, aspartame, and sucrose on food intake, satiety, and postprandial glucose and insulin levels. Appetite. 2010;551:37-43. PMID: 20303371. Source
4. Network meta-analysis of 36 trials (472 participants). Non-nutritive sweetener beverages. Nutrients. 2023. PMC9965414.
5. Systematic review and meta-analyses: 101 articles, 100 experiments. Advances in Nutrition. 2025. PMC12205327.
6. Romo-Romo A, et al. Effects of the non-nutritive sweeteners on glucose metabolism and appetite regulating hormones. Clinical Nutrition. 2019. PMID: 31697573. Source
7. Suez J, et al. Personalized microbiome-driven effects of non-nutritive sweeteners on human glucose tolerance. Cell. 2022;18518:3307-3328. PMID: 35987213. Source

---

Join the free community:
Men: Iron Forge Brotherhood
Women: Powerhouse Fitness