The Sweetener Aisle Is Lying to You (A Little)

If you want to ditch added sugar without trading it for something equally or more harmful, this piece tells you which sweeteners are genuinely low-risk, which are sugar in nicer packaging, and the exact label clues (yes — erythritol again) to watch for at the supermarket so you don’t accidentally up your cardiovascular or digestive risk.

By Dr Ferghal Armstrong · MedHEADS

Here’s a scene that might feel familiar. You’ve been trying to eat a bit better, so you swapped the sugar in your coffee for stevia, grabbed a keto-friendly bar at the servo, bought the protein powder sweetened with monk fruit. Tick, tick, tick. Problem solved.

Except maybe not. Switching to sweeteners doesn’t necessarily solve your sugar problem — you may have quietly traded it for a different one, and your body is keeping score in ways you can’t taste.

This isn’t a scare piece; most of what follows is good news. But the world of sugar replacement is confusing on purpose, and some of what’s sold as “healthy” is only telling you the flattering half of the story. So let’s walk the whole aisle together.

First, why we bother

Sugar manages to be completely natural and, in the amounts most of us eat, genuinely harmful. The average Australian gets through 14 to 17 teaspoons of added sugar a day. The World Health Organization suggests staying under six. That’s not a rounding error — it’s a chasm.

Sugar doesn’t cause obesity, type 2 diabetes, heart disease, fatty liver and chronic inflammation in a neat one-to-one way — it’s messier than that. But it’s a major contributor to the metabolic dysfunction sitting underneath most of those conditions. So wanting something sweet without the metabolic bill is completely rational. The only real question is whether the alternatives deliver.

It helps to know there are three tribes living in that aisle. The plant-derived sweeteners like stevia and monk fruit — near-zero calories, no effect on blood glucose, the cleanest safety records. The novel sugars, chiefly allulose, technically sugar but handled almost nothing like the real thing. And the artificial sweeteners and sugar alcohols — sucralose, aspartame, saccharin, erythritol, xylitol — the oldest, most widespread and most argued-over. Let’s take the reassuring ones first.

The ones holding up well

Stevia comes from a South American plant used for centuries. The sweet compounds — steviol glycosides — are 200 to 300 times sweeter than sugar, and your body simply doesn’t metabolise them: they pass straight through, no calories, no glucose spike. Its safety record is about as solid as any sweetener we have, with no tumour risk flagged in long-term animal studies and even hints of modest benefit for people with high blood pressure or diabetes. The catch is a bitter, faintly metallic aftertaste some people can’t ignore.

Here’s the nuance that matters at the shelf, though. What you buy labelled “stevia” is almost never pure stevia. It’s usually a refined extract (rebaudioside A) blended with a bulking agent to make it measure spoon-for-spoon like sugar — and that agent is very often erythritol. Hold that thought.

Monk fruit — luo han guo — is a small gourd from southern China and Thailand, long used in traditional Chinese medicine. Its sweetness comes from mogrosides, 150 to 250 times sweeter than sugar, again zero calories and no glucose effect. Most people find it cleaner-tasting than stevia. The safety research is reassuring but thinner, simply because it’s newer to Western shelves; the FDA rates it “generally recognised as safe.” A solid pick — look for single-ingredient versions, or ones cut with allulose rather than erythritol.

Allulose is the one I find genuinely fascinating. It’s technically a sugar, occurring naturally in tiny amounts in figs, dates and wheat, but your body absorbs it without really metabolising it — about 0.4 calories per gram against sugar’s four, with essentially no impact on blood glucose or insulin. And unlike stevia or monk fruit, which are fine in coffee but hopeless in the oven, allulose behaves like sugar when you cook: it browns, it caramelises, it does the structural work. There’s also early, promising evidence it may improve insulin sensitivity and even nudge GLP-1, the same satiety hormone drugs like semaglutide target. Early days, but compelling. The main downside is digestive — bloating or loose stools if you overdo it — so keep it under about 10 to 15 grams in one sitting.

The “natural sugar” trap

A quick word on something that comes up constantly. Patients tell me they’ve switched from sugar to honey, coconut sugar or maple syrup because those are natural. Metabolically, these are all still sugar. Yes, honey has some antioxidant and antimicrobial properties, coconut sugar a slightly lower glycaemic index, maple syrup some trace minerals. But “slightly lower” is not “low,” and none of it is clinically meaningful at the amounts people actually use. A teaspoon of raw honey in your tea is fine. Swapping a tablespoon of white sugar for a tablespoon of coconut sugar and calling it a health win is not — the calorie and glucose load is virtually identical. These aren’t alternatives to sugar. They’re sugar in fancier outfits.

Now the complicated half

The artificial sweeteners have been the default for anyone managing weight or diabetes since the 1950s: sucralose (Splenda), aspartame (Equal, and most diet drinks), saccharin (Sweet’N Low). All FDA approved and extensively reviewed — and worth defending against the wilder claims. Saccharin’s 1970s cancer scare came from rat studies using doses equivalent to a human drinking hundreds of cans of diet drink a day; the warning was pulled in 2000. Aspartame was labelled “possibly carcinogenic” by the WHO’s IARC in 2023 — the same category as pickled vegetables and working as a hairdresser. The FDA reviewed the identical data and disagreed. So let’s not be hysterical.

But let’s be honest about the more credible concerns. Large cohort studies have linked heavy artificial-sweetener use with cardiovascular events, including stroke. These are observational — they can’t prove cause — but the signal is consistent enough to notice. And a growing body of work suggests sucralose, saccharin and aspartame may reshape the gut microbiome and, through it, impair glucose tolerance. The irony is sharp: sweeteners taken to avoid a blood-sugar spike may, in some people, quietly worsen blood-sugar control. Notably, a 2024 study found 12 weeks of stevia did not alter the microbiome — another point in its favour.

One sucralose footnote: it’s heat-stable, which is why it’s popular in baking, and exactly the problem. At high temperatures it can form compounds called chloropropanols, some flagged as potentially harmful. The research is still developing, but enough to say: keep sucralose for cold uses — coffee, smoothies, cold drinks — and don’t bake with it.

The one to actually watch: erythritol

This is the sweetener I’d pay the most attention to, precisely because most people have no idea they’re eating it. Erythritol is a sugar alcohol, about 70% as sweet as sugar, near-zero calories, no glucose spike. On paper, a winner — which is why it became the go-to bulking agent in nearly every stevia, monk fruit and keto-branded product going. If your sweetener comes in a packet, there’s a good chance it’s in there.

Then the story turned. In February 2023, a study in Nature Medicine found that higher blood levels of erythritol were associated with a significantly raised risk of heart attack and stroke in cardiac patients — those in the highest quartile were roughly twice as likely to have a major cardiac event over three years. In 2024, the same group (led by Dr Stanley Hazen at the Cleveland Clinic) showed that a single erythritol-sweetened drink spiked blood levels more than a thousandfold above baseline, and that spike tracked with increased platelet activity — blood becoming measurably more prone to clotting.

The caveats are real: the first study was observational in a high-risk group, the platelet study was small, and the body makes some erythritol on its own, so elevated levels might partly reflect existing metabolic trouble rather than diet. Legitimate, unsettled debates. But given how much erythritol is in “healthy” products at meaningful doses, and given a biologically plausible clotting mechanism, I think the precautionary principle applies — especially for anyone with cardiovascular risk factors. Keep your intake low, and when you can, choose products blended with allulose instead. The label matters.

Its cousin xylitol — familiar from toothpaste and sugar-free gum — was flagged in the same research program, with a 2024 European Heart Journal paper linking higher levels to cardiovascular risk through similar mechanisms. Toothpaste quantities are not a concern. Xylitol by the gram as a daily cooking sweetener is worth being cautious about.

Where that leaves you

If you want the short version to take to the shops: stevia and monk fruit, especially pure or blended with allulose, are your safest, best-evidenced options. Allulose is the standout if you’re actually cooking or baking, with some encouraging metabolic upside. Erythritol and xylitol — use sparingly, and know they’re hiding in more products than you’d guess. Sucralose and aspartame — fine in moderation, not consequence-free; mind the quantities and don’t bake with sucralose.

But the real point sits underneath all of it. No sweetener — natural or artificial — is a health food. The goal was never to find a perfect sugar stand-in and pour it on freely. It’s to slowly turn down the expectation that everything needs to taste sweet at all. That’s where the actual metabolic benefit lives.

So next time you’re standing in the sweetener aisle wondering what on earth to buy, you’ll at least know which half of the story the packaging isn’t telling you.\


References

Note: references cover the specific, named study claims in the article. General statements (e.g. sweetness ratios, calorie content, FDA “generally recognised as safe” status, the saccharin cancer-warning removal in 2000) reflect established regulatory and nutritional consensus.

The Nature Medicine observational analysis that linked higher blood erythritol with increased heart attack and stroke risk provides the primary data behind these concerns. erythritol and cardiovascular risk.

The Cleveland Clinic trial measuring the acute platelet effects after an erythritol drink shows a plausible clotting mechanism to explain the observational signal. erythritol ingestion and platelet activity.

For the aspartame hazard discussion, reference the IARC summary that led to the ‘possibly carcinogenic’ classification in 2023. IARC aspartame assessment.

Balance the IARC view with the FDA’s assessment and guidance on aspartame and other sweeteners. FDA on aspartame.

The 12-week trial showing no meaningful change in gut microbiota after stevia consumption supports the point that stevia appears microbiome-neutral. stevia 12-week microbiome trial.

Advertisement
Advertisement