Sweeteners: the trials and the observations disagree

The World Health Organization recommends against using non-sugar sweeteners for weight control.

Inside the same body of evidence, randomised controlled trials found that people consuming more of them had lower body weight and lower BMI.

Both statements are true, the contradiction is the interesting part, and almost nobody who quotes the headline has read the second sentence.

What the guideline actually says

In May 2023 the WHO released a guideline on non-sugar sweeteners recommending against their use to control body weight or reduce the risk of noncommunicable disease.

The basis given was a systematic review finding no long-term benefit in reducing body fat in adults or children, and suggesting potential undesirable effects from long-term use, including increased risk of type 2 diabetes, cardiovascular disease and mortality in adults.

That is the headline and it travelled widely. The guideline document itself contains the detail that complicates it.

The split inside the evidence

Two study designs, opposite directions.

That pattern has an obvious candidate explanation, stated plainly: people who are gaining weight switch to diet drinks. The observational finding may be recording the reason people started rather than the consequence of them starting.

This is called reverse causation, it is the standard difficulty with observational nutrition data, and it leaves the WHO position standing while showing the evidence is weaker and more ambiguous than a headline recommending against them suggests.

The honest position: in controlled conditions, replacing sugar with sweeteners produces a modest weight benefit. Across populations followed for years, the association runs the other way, and nobody can currently separate cause from confounding.

And the part that matters most here

The WHO guideline addresses long-term weight control and disease risk in the general population.

That is not the question a fighter is asking.

A fighter is asking whether a zero-calorie drink is acceptable in a camp where he is running a deficit for 8 weeks, and whether it is acceptable in the last 3 days when everything is being managed.

Those are two different questions and neither is the one the guideline answers. Applying a general-population, long-horizon recommendation to a specific short-term athletic context is a category error, and it is the most common error made when this guidance gets quoted in a gym.

So: in camp

The practical case for sweeteners in a camp is behavioural rather than metabolic, and it is real.

A fighter in week six of a deficit is hungry every evening. The failure mode that ends camps is not a macronutrient error, it is a person who could not tolerate the restriction any longer.

A zero-calorie sweet drink occupies a craving at no energy cost. So does a sugar-free jelly, or sweetener in coffee replacing two spoons of sugar.

Two spoons of sugar 3 times a day is around 100 calories. Across 8 weeks that is not nothing, and it is a saving that costs no adherence.

The counterargument worth taking seriously is habituation to sweetness. If the objective is to reduce a taste for sweet food across a career, substituting sweetness for sweetness does not advance it. A legitimate long-term position, and a different project from getting through a camp.

And the reason it backfires for some people

There is a mechanism that explains why some fighters find diet drinks help and others find they make hunger worse.

If a diet drink in the evening reliably leaves you hungrier an hour later, that is not imagination and the answer is to stop using it that way.

The gut microbiome question

This is where most of the alarming coverage comes from, and the evidence is messier than either side presents.

Findings vary by which sweetener, by whether a purified compound or a commercial formulation was studied, by the carrier ingredients in that formulation, by dose, by duration and by the individual.

That last point is not a hedge. A 2022 study in Cell found personalised, microbiome-driven effects of non-nutritive sweeteners on human glucose tolerance, meaning the response differed between individuals rather than being uniform.

Meanwhile a 2019 study found that high-dose saccharin did not induce gut microbiota changes or glucose intolerance in healthy humans.

And a great deal of the frightening material comes from animal models at doses far above human intake.

The most useful framing from the literature: purified compounds and commercial formulations are biologically non-equivalent. A study on pure sucralose does not tell you about a diet drink, because the drink contains other things.

Which leaves a fighter with a genuinely uncertain picture rather than a clean answer, and the appropriate response to genuine uncertainty is moderation rather than either enthusiasm or avoidance.

And one mechanism that may work in your favour

Sweeteners activate the same sweet taste receptors in the gut, T1R2 and T1R3, that sugar does.

Activating them increases expression of the intestinal sugar transporters SGLT-1 and GLUT2, which are the routes carbohydrate takes across the intestinal wall.

Which raises a possibility that runs opposite to the usual concern: a sweetener consumed alongside carbohydrate may alter how quickly that carbohydrate is absorbed.

For a fighter that is potentially relevant in exactly one window. In the hours after a weigh-in, faster carbohydrate absorption is what you are trying to achieve, and SGLT-1 is the same transporter that sodium-glucose cotransport depends on for fluid uptake.

I would not build a protocol on this. The mechanism is established and the practical magnitude in a refuelling athlete is not. The sweetener question has a plausible upside as well as a plausible downside, which is rarely how it gets presented.

The sweeteners themselves

They are not one category and treating them as one is part of the confusion.

Sugar alcohols, which are a different problem entirely

Erythritol, xylitol, maltitol, sorbitol and the rest are not zero-calorie and not non-absorbed.

They carry some energy, and more importantly they are osmotically active in the gut. Poorly absorbed in the small intestine, they draw water in and are fermented in the large intestine.

The consequences are bloating, gas and, at higher doses, diarrhoea. Products containing them frequently carry a warning on the packet.

Maltitol and sorbitol are the worst offenders for gut symptoms. Erythritol is the best tolerated of the group and it carries a separate question that the others do not, which the next section covers.

Sorbitol is the same compound described under dried fruit, where it is the reason prunes work.

Erythritol needs its own section

Erythritol behaves least like the others in the gut, and it is the one with the most serious question attached elsewhere.

On digestion it is the best of the group. Around 90 percent is absorbed in the small intestine and excreted unchanged in urine rather than being fermented in the colon, which is why it produces far less bloating and gas than maltitol or sorbitol.

That is why it appears in almost every sugar-free and low-carbohydrate product on the shelf.

The cardiovascular finding

A 2023 study in Nature Medicine examined circulating erythritol and cardiovascular event risk.

The outcome measured was 3-year major adverse cardiovascular events: death, non-fatal heart attack or non-fatal stroke.

Mechanistic work accompanied the epidemiology. Erythritol at physiological concentrations enhanced platelet reactivity in vitro, and increased the rate of clot formation in a mouse model of arterial injury.

A subsequent intervention study gave healthy volunteers 30 grams of erythritol against 30 grams of glucose. Erythritol produced an approximately 1,000-fold rise in plasma concentration and increased platelet reactivity. Glucose did neither.

And the serious counter-argument

This is contested, and the objections are substantive rather than industry noise.

Erythritol is produced endogenously in humans through the pentose phosphate pathway, and endogenous production increases under metabolic and oxidative stress. Which raises the possibility that high plasma erythritol in cardiac patients is a marker of their disease rather than a consequence of their diet.

That is the same reverse causation problem described earlier on this page, and it applies with equal force here.

It also occurs naturally in grapes, melons, pears and in fermented products including wine and soy sauce, so circulating levels are not purely a function of sweetener intake. And the mechanism is unclear: erythritol is a small, relatively unreactive molecule, and long-term animal toxicology found nothing remarkable.

The intervention studies answer part of the objection, because they are experimental rather than observational. People drank erythritol and platelet reactivity rose. That does not establish clinical outcomes, and no trial has followed erythritol consumers for cardiovascular endpoints.

What I think a fighter should do with this

The honest position: unresolved, a more concerning signal than for the non-nutritive sweeteners above, and the regulators have not changed their classification.

For a healthy athlete eating erythritol occasionally, I would not act on it.

Aspartame, IARC and what actually changed

In 2023 the International Agency for Research on Cancer classified aspartame as possibly carcinogenic to humans, Group 2B.

That headline travelled a long way, and two things about it are worth knowing.

Group 2B is the category for agents with limited evidence in humans. It is a hazard classification, meaning it describes whether something could cause harm under some conditions, not how likely harm is at real intakes.

And the body that assesses risk rather than hazard, JECFA, reviewed aspartame at the same time and did not change the acceptable daily intake.

So the position after 2023 is the same as before it. A hazard classification was issued, the risk assessment was reviewed and left unchanged, and the intake required to approach the limit is far beyond anything a person drinks.

Anybody who stopped using aspartame because of that headline made a decision based on a document that was not about the amounts they were consuming.

Diet drinks in fight week

Here the sweetener is almost beside the point, and the thing that matters gets forgotten.

A diet drink is fluid. Fluid is weight. A litre of anything is a kilogram on the scale.

A fighter drinking 2 litres of diet cola on Thursday because it has no calories has drunk 2 kilograms of liquid, and the absence of sugar changes nothing about that.

That is the single most important sentence on this page for anybody in the last 3 days, and it is the one nobody says because the conversation gets absorbed by aspartame.

Three further points for that week.

• Carbonation produces gas in the stomach, unwelcome when abdominal volume is being managed, and it slows how fast you can drink something.
• Diet colas carry caffeine, frequently in amounts people do not count because they are counting coffee. A can carries roughly a third to a half of a cup, and four cans is a dose.
• They carry no sodium, which places them among the drinks that performed no better than water on fluid retention.

Chewing gum, which is a fight week item

Gum belongs here because almost every sugar-free gum is sweetened with sugar alcohols, usually sorbitol, xylitol or maltitol, and because fighters chew it in exactly the week those compounds cause problems.

The uses are real. Gum occupies the mouth when hunger is constant, it promotes salivation, and some fighters use it to produce saliva for spitting during a cut.

Three problems.

• Sugar alcohols draw water into the intestine, working against the objective in the final days. Several pieces across a day is not a trivial dose, and packets carry a laxative warning for that reason.
• Swallowed air. Chewing increases air swallowing, producing gastric gas and abdominal distension when abdominal volume is being managed.
• Gastric acid on an empty stomach. Chewing triggers the cephalic phase response, stimulating hydrochloric acid and gastrin secretion into a stomach with nothing in it. In a severely hungry fighter that frequently produces heartburn and a sharp hunger spasm rather than relief.

Xylitol carries an additional warning that has nothing to do with the athlete: it is seriously toxic to dogs at small quantities, so a dropped packet is a genuine hazard in a house with one.

Reasonable in ordinary camp. In the final 3 days it is one more source of gut water, gas and acid that nobody counted.

And after the scale

Not useful, and this deserves stating because it catches people.

In the refuelling window you want fluid with sodium, and carbohydrate. A zero-calorie drink delivers fluid without either.

It does no harm there, and it still occupies stomach space that should be occupied by something doing a job, at a point where stomach space is finite and the clock is running.

The performance question

There is no meaningful evidence that non-sugar sweeteners impair athletic performance at normal intakes, and none that they improve it.

The one performance-adjacent consideration is the gut. An athlete who reacts to a particular sweetener or to sugar alcohols and does not know it may be attributing digestive symptoms to training load or to nerves.

If you have unexplained gut trouble in camp, the sugar-free products are worth removing for a fortnight as a test, in the structured way described under gluten and lactose. Cheap to test, and it occasionally explains something.

For a tested athlete

Sweeteners are not prohibited substances and this is not an anti-doping question.

The relevant point is the one that appears throughout this site. A diet drink from a shop is food. A sugar-free protein or pre-workout product carrying sweeteners is a supplement, with the batch-testing requirement attached, for reasons that have nothing to do with the sweetener in it.

Nobody has failed a test because of aspartame. People have failed tests because of what else was in the tub.

Where it sits by phase

The argument is always about the wrong thing

The WHO recommendation concerns long-term weight control in the general population, and the evidence behind it splits by study design in a way the headline does not convey.

For a fighter, sweeteners in a camp are a behavioural tool with a real calorie saving and an uncertain long-term picture nobody can currently resolve.

For a fighter in fight week, the sweetener is not the issue at all. The fluid is the issue, and the sugar alcohols are the issue, and both get missed because the argument is always about aspartame.