Artificial sweeteners have become almost impossible to avoid.

They're in diet sodas, energy drinks, protein powders, pre-workouts, flavored waters, and countless other products marketed as healthier alternatives to their sugar-containing counterparts.

And for good reason. They allow manufacturers to create sweet-tasting products with little or no added sugar and considerably fewer calories.

But there's a question that's been debated for years.

Are artificial sweeteners actually better for your long-term metabolic health, or are we simply replacing one potential problem with another?

A new 2026 systematic review and meta-analysis published in Frontiers in Nutrition examined the relationship between artificial sweetener consumption and the development of type 2 diabetes.

The researchers analyzed six prospective cohort studies involving more than 720,000 adults, with an average follow-up of approximately 10 years.

Their findings certainly caught my attention.

People consuming higher amounts of artificial sweeteners had a 31% higher relative rate of developing type 2 diabetes compared with those consuming little or none.

That's a substantial association. But before anyone starts throwing out their diet soda or replacing sucralose with regular sugar, there's quite a bit we need to unpack.

Because this study raises some important questions, but it doesn't prove that artificial sweeteners cause diabetes.

TLDR

A new 2026 meta-analysis examined six prospective cohort studies involving 721,003 adults and 38,445 new cases of type 2 diabetes.

• Higher artificial sweetener intake was associated with a 31% higher relative rate of developing diabetes. The pooled hazard ratio was 1.31, with a 95% confidence interval of 1.16 to 1.49.
• The studies followed participants for an average of 10.1 years. This provides useful long-term observational data, but it doesn't establish causation.
• The results varied considerably between studies. The researchers reported substantial heterogeneity, and the overall certainty of evidence was rated low.
• Reverse causation is a major concern. People already at greater risk of diabetes may be more likely to choose artificially sweetened products in the first place.
• Most studies examined diet beverages. The results cannot automatically be applied to every sweetener, dose, or product containing artificial sweeteners.

My takeaway: This study gives us a reason to continue investigating the long-term metabolic effects of artificial sweeteners, but it doesn't establish that they cause diabetes or that consuming sugar instead would be healthier.

Why Artificial Sweeteners Are So Popular

Let's start with why artificial sweeteners exist in the first place.

Sugar provides approximately four calories per gram. In beverages and flavored supplements, those calories can add up quickly without contributing much to satiety.

Artificial sweeteners such as sucralose, aspartame, saccharin, and acesulfame potassium provide sweetness with little or no caloric contribution at typical use levels.

That makes them appealing for people trying to reduce calorie intake, manage body weight, or limit added sugar consumption.

From a purely calorie-management perspective, replacing a sugar-sweetened beverage with a zero-calorie alternative can make sense.

A regular soda might contain 35 to 40 grams of added sugar. A diet version can provide a similar flavor without those calories.

But reducing calories and improving long-term metabolic health aren't necessarily identical outcomes.

And that's what makes this new research interesting.

The question isn't simply whether artificial sweeteners contain fewer calories than sugar. We already know they do.

The question is whether consuming them regularly over many years is associated with changes in diabetes risk.

How the Researchers Conducted the Study

The researchers performed a systematic review and meta-analysis of prospective observational studies.

They searched several scientific databases, including PubMed, Embase, Web of Science, and a Chinese research database, for studies published through March 31, 2025.

To qualify, studies needed to follow adults over time, assess artificial sweetener consumption, and examine the development of type 2 diabetes.

Six studies met the inclusion criteria.

The studies were conducted in the United States, Europe, and Japan, with follow-up periods ranging from 5 to 20 years.

Researchers used food-frequency questionnaires to estimate artificial sweetener intake and tracked new diagnoses of type 2 diabetes.

They also adjusted their statistical analyses for various factors, including age, body mass index, dietary habits, and lifestyle characteristics, although the exact adjustments differed between studies.

This is an important strength.

Unlike a cross-sectional study that simply examines people's diets and health at one point in time, these studies followed participants prospectively.

That allowed researchers to examine whether reported sweetener consumption was associated with diabetes diagnoses occurring later.

But there's a limitation.

These were observational studies, not randomized controlled trials.

Participants weren't randomly assigned to consume artificial sweeteners or avoid them.

They made their own dietary choices, which creates several challenges when trying to determine cause and effect.

The Main Finding: A 31% Higher Relative Rate of Diabetes

The researchers combined the results of all six studies using a random-effects meta-analysis.

They found that adults with higher artificial sweetener consumption had a pooled hazard ratio of 1.31 compared with those reporting lower or no consumption.

The 95% confidence interval ranged from 1.16 to 1.49.

In practical terms, that corresponds to an estimated 31% higher relative hazard of developing type 2 diabetes in the higher-consumption groups.

All six studies reported positive associations, although the size of those associations varied.

The strongest association came from the French NutriNet-Santé cohort, which reported a hazard ratio of 1.69.

The weakest came from a 2025 analysis of large U.S. cohorts, which reported a hazard ratio of 1.11.

That's a pretty wide range.

And there's another detail worth understanding.

A 31% higher relative hazard does not mean that 31% of people consuming artificial sweeteners developed diabetes.

It also doesn't mean that someone's personal probability of developing diabetes automatically increases by 31 percentage points.

Relative measures need to be interpreted alongside baseline risk.

For example, if a hypothetical group had a 10-year diabetes risk of 5%, a 31% relative increase would correspond to approximately 6.55%, assuming the relative effect could be applied directly to that risk.

That's an illustrative calculation, not an absolute-risk estimate from this meta-analysis.

The study did not establish how much artificial sweetener consumption changes an individual's absolute diabetes risk.

The Results Weren't Nearly as Consistent as They Might Appear

One of the first things I look at when reviewing a meta-analysis is how consistent the included studies are.

In this case, the researchers reported an I² value of approximately 86%.

That's substantial heterogeneity.

In simpler terms, the study results varied considerably more than we'd expect from random sampling differences alone.

The researchers used a random-effects model to account for differences between studies, but that doesn't make the differences disappear.

The paper also reported a 95% prediction interval of 0.97 to 1.78.

That's important because a prediction interval estimates the range of effects that might reasonably be observed in a comparable future study.

Unlike the confidence interval around the pooled estimate, this prediction interval included 1.0, the value corresponding to no association.

So although the pooled result was statistically significant, the evidence wasn't consistent enough to assume that every population or exposure would show the same relationship.

The researchers also rated the overall certainty of evidence as low using the GRADE framework.

That doesn't mean the findings should be ignored.

It means we should be cautious about treating the 31% figure as a precise estimate of a causal effect.

The Biggest Problem: Reverse Causation

This is probably my biggest concern with interpreting the results.

Imagine two people.

One is relatively healthy, physically active, and has no history of blood sugar problems. They drink regular soda occasionally.

The other is overweight, has a family history of diabetes, and recently learned that their blood glucose is elevated.

Their doctor recommends reducing added sugar, so they switch from regular soda to diet soda.

Ten years later, the second person develops type 2 diabetes.

If we simply compare beverage consumption, we might observe that the person drinking diet soda developed diabetes while the person drinking regular soda didn't.

But did the artificial sweeteners cause diabetes?

Not necessarily.

The person drinking diet soda may have been at greater risk long before they started consuming it.

This is called reverse causation, and it's a major challenge in observational nutrition research.

The authors acknowledged this possibility.

They noted that people with obesity, prediabetes, or other metabolic risk factors may be more likely to consume artificial sweeteners as part of an effort to improve their diets.

The included studies adjusted for factors such as body weight and lifestyle, but statistical adjustments cannot eliminate every difference between people who choose diet products and those who don't.

There's also a subtle issue in the paper's discussion.

The authors suggested that reverse causation was not directly present because the studies excluded people who already had diabetes at baseline.

I don't think that completely resolves the concern.

Excluding diagnosed diabetes doesn't exclude prediabetes, insulin resistance, obesity, or other risk factors that could influence someone's decision to use artificial sweeteners.

Prospective follow-up establishes that the diabetes diagnosis occurred after the dietary assessment. It doesn't establish that the sweetener exposure preceded the underlying metabolic risk.

That's a distinction I wouldn't overlook.

Could Artificial Sweeteners Affect the Gut Microbiome?

The researchers discussed several possible mechanisms that might explain an association between artificial sweetener consumption and diabetes.

One involves the gut microbiome.

Your gastrointestinal tract contains trillions of microorganisms that influence digestion, microbial metabolite production, and various aspects of metabolic regulation.

Some experimental research has suggested that certain artificial sweeteners may alter gut microbial composition or activity.

Those changes could potentially influence glucose tolerance and insulin sensitivity.

For example, the paper referenced research examining personalized microbiome-driven responses to non-nutritive sweeteners.

That's interesting because it raises the possibility that different people may respond differently to the same sweetener.

But there are two important limitations.

First, not all artificial sweeteners have identical chemical structures, absorption patterns, or metabolic effects.

Second, this meta-analysis did not measure gut microbiome changes.

The proposed mechanism is biologically plausible, but the researchers didn't establish that microbiome alterations explained the observed diabetes association.

I wouldn't take these findings as proof that artificial sweeteners damage your gut bacteria.

That's a much stronger claim than the evidence supports.

What About Insulin and Blood Sugar?

Another common claim is that artificial sweeteners increase insulin secretion despite containing no sugar.

The researchers discussed the possibility that sweet taste receptors in the mouth and gastrointestinal tract could influence hormonal pathways involved in glucose regulation.

That includes incretin hormones such as GLP-1.

But this is an area where it's easy to confuse proposed mechanisms with demonstrated effects.

Different sweeteners may produce different physiological responses, and results can depend on the dose, whether the sweetener is consumed alone or with food, and the characteristics of the person consuming it.

This particular study didn't measure insulin, GLP-1, blood glucose responses, or insulin sensitivity.

It examined the long-term association between reported sweetener consumption and diabetes diagnoses.

We cannot conclude from this meta-analysis that artificial sweeteners cause insulin spikes or directly impair glucose regulation.

Those questions require controlled intervention studies.

Does This Mean Sugar Is the Better Choice?

Absolutely not based on these findings.

And I think this is where some interpretations of this research could become genuinely misleading.

If a study finds that people consuming artificial sweeteners have a higher rate of diabetes than people consuming fewer artificial sweeteners, that doesn't establish that sugar is healthier.

The appropriate comparison matters.

Are we comparing diet soda with regular soda?

Diet soda with water?

Artificially sweetened yogurt with unsweetened yogurt?

A pre-workout containing sucralose with an equivalent product containing 30 grams of added sugar?

Those aren't interchangeable comparisons.

This meta-analysis primarily examined artificial sweetener intake or artificially sweetened beverages compared with lower or no intake.

It did not randomly assign people to replace sugar with artificial sweeteners over a decade.

It also didn't establish that switching from a diet beverage to its sugar-sweetened counterpart would reduce diabetes risk.

A higher observed risk among artificial sweetener consumers is not evidence that adding sugar back into the diet would improve metabolic health.

If someone is consuming substantial amounts of added sugar, reducing that intake can still be a reasonable dietary goal.

Whether artificial sweeteners are the best long-term replacement is a separate question.

What Does This Mean for Sports Nutrition Supplements?

This study is particularly relevant to the supplement industry because artificial sweeteners are widely used in flavored powders.

Pre-workouts, hydration products, amino acid supplements, and protein powders often contain sucralose or acesulfame potassium.

Manufacturers use them because they provide sweetness without requiring large amounts of sugar.

But here's an important limitation in applying this study to sports nutrition.

Most of the included cohorts examined artificially sweetened beverages, particularly diet soda.

The researchers didn't specifically investigate pre-workout supplements, protein powders, or electrolyte products.

They also didn't establish a universal dose-response relationship that could tell us whether consuming a small amount of sucralose in a supplement has the same association as drinking several diet sodas every day.

And the included studies generally didn't separate every individual sweetener into distinct analyses.

So I wouldn't use this paper to claim that a particular pre-workout containing sucralose increases diabetes risk by 31%.

The research simply doesn't support that level of specificity.

At Simply Good Supplements, we've chosen to formulate without artificial sweeteners. That's a product philosophy I believe in, particularly when there are other ways to make a supplement taste good.

But I also think it's important to distinguish a formulation preference from a scientifically proven health advantage.

This study doesn't demonstrate that our products prevent diabetes or that avoiding artificial sweeteners necessarily reduces diabetes risk.

What it does reinforce is the value of ingredient transparency and continued research into substances people may consume regularly for years.

What I Think This Study Actually Tells Us

I think the biggest takeaway is that we shouldn't automatically assume something is metabolically harmless simply because it contains zero calories.

Artificial sweeteners have clear practical advantages when it comes to reducing added sugar and calories.

But the long-term health effects of regular consumption deserve continued investigation.

This meta-analysis found a statistically significant association between higher artificial sweetener intake and type 2 diabetes across six prospective cohorts.

That's worth paying attention to.

At the same time, the evidence had substantial heterogeneity, was rated low certainty, and couldn't fully address reverse causation or residual confounding.

Those are not minor limitations.

And I think there's another point worth emphasizing.

The research question we should really be asking isn't whether artificial sweetener consumers are healthier than nonconsumers. It's what happens when people replace a specific amount of sugar with a specific artificial sweetener over time.

That's a much more useful question for making dietary recommendations.

Unfortunately, it's also considerably more difficult to answer with long-term randomized research.

The Bottom Line

A new 2026 systematic review and meta-analysis involving 721,003 adults found that higher artificial sweetener consumption was associated with a 31% higher relative hazard of developing type 2 diabetes.

The association remained positive across all six included cohort studies and was stable when the researchers removed individual studies from the analysis.

However, the studies differed substantially, and the overall certainty of evidence was rated low.

Most importantly, the research was observational.

It cannot establish that artificial sweeteners caused the increased diabetes risk.

My takeaway is that artificial sweeteners shouldn't automatically be considered beneficial simply because they're calorie-free, but this study also doesn't justify claiming that they cause diabetes.

If you're consuming large amounts of artificially sweetened beverages every day, I think it's reasonable to consider reducing your reliance on them.

Water, unsweetened beverages, and a generally balanced diet remain sensible options.

But if you're using a diet beverage to replace several hundred calories of added sugar each day, I wouldn't interpret this study as a reason to switch back to regular soda.

The science isn't that simple.

About the Author

Matt Mosman, MS, CISSN, CSCS, is the founder and Chief Science Officer of Simply Good Supplements. With a background in exercise science, sports nutrition, and strength and conditioning, Matt focuses on turning research into practical advice that athletes, parents, and everyday active people can actually use.

References

Lu S, Lu Y, Xu R, Hou W, Xu T, Tang W. The longitudinal association between artificial sweetener intake and the risk of type 2 diabetes among adults aged 18 years and older: a systematic review and meta-analysis. Frontiers in Nutrition. 2026;13:1886458. https://doi.org/10.3389/fnut.2026.1886458