Sugar vs. Sugar Alcohols vs. Artificial Sweeteners: What the Research Actually Says

Key takeaways

• The sweeteners on store shelves split into three categories: caloric sugars (sucrose, fructose, glucose, honey, agave), sugar alcohols (erythritol, xylitol, maltitol), and non-nutritive sweeteners (stevia, monk fruit, sucralose, aspartame, saccharin, acesulfame-K).
• For weight management, replacing sugar-sweetened beverages with non-nutritive-sweetened versions reliably reduces calorie intake — a real, replicable benefit.
• For health concerns, the evidence is mixed: aspartame has been the most contested in 2023’s WHO classification (Group 2B “possibly carcinogenic,” based on limited evidence and at high intakes); stevia, monk fruit, and acesulfame-K have generally clean safety records.
• Erythritol had a 2023 cardiovascular-association study that’s been heavily discussed but doesn’t meet the standard for causal claims; it remains FDA-permitted and widely used.
• The honest synthesis: use sweeteners deliberately, prefer the most-studied options, and don’t substitute heavy daily artificial sweetener use for actually addressing sweet preferences over time.

If you’ve shopped for any “diet” or “low-sugar” product in the past five years, you’ve encountered a parade of sweeteners with names that read like a chemistry quiz: erythritol, allulose, monk fruit, sucralose, aspartame, ace-K, stevia. The marketing claims around each oscillate between “miraculously safe” and “secretly toxic” depending on which website you land on.

This article is the current honest synthesis. What each sweetener is, what it does in your body, what the evidence supports about safety and weight outcomes, and how to think about choosing.

It’s a deeper dive on the sugar half of Macronutrients Explained and complements The Sugar Detective’s Guide.

The three categories at a glance#

GRAS = “Generally Recognized as Safe” by the FDA.

The three categories work in different ways:

• Caloric sugars raise blood glucose, deliver real calories, digest fully.
• Sugar alcohols are partially absorbed and partially fermented by gut bacteria. Some calories, smaller blood-glucose effect, variable digestive tolerance.
• Non-nutritive sweeteners are largely not absorbed; they pass through the digestive tract or are metabolized minimally. Effective zero calories, no blood-glucose impact, but a more complex long-term safety conversation.

Caloric sugars#

These are the original sweeteners. The chemistry varies a bit but the metabolic impact converges:

• Sucrose (table sugar): glucose + fructose, 50/50.
• High-fructose corn syrup (HFCS): 55% fructose, 42% glucose in the most-common variant. Despite the name, the ratio is close to sucrose’s, and at population intake levels, there’s no evidence HFCS is metabolically distinct from sucrose at matched doses.
• Honey: 38% fructose, 31% glucose, plus trace minerals and some bioactive compounds. Higher GI than its sugar profile suggests because of the partial digestion-already-happened state.
• Agave nectar: ~85% fructose. Marketed as “low GI” because fructose has a lower acute glucose response — but the high fructose load is metabolically more taxing on the liver, not less. Skip.
• Maple syrup, coconut sugar, molasses: trace nutrients, but metabolically similar to sucrose at typical intake amounts.

The upshot for caloric sugars: at typical Western intake levels, the type of caloric sugar matters much less than the amount. The 2020–2025 Dietary Guidelines for Americans recommend keeping added sugars under 10% of daily calories (≤50 g/day on a 2,000-cal diet); the WHO and AHA recommend an even tighter ≤25 g/day for additional benefit.

Sugar alcohols#

Sugar alcohols (also called polyols) are partially-absorbed sweeteners with a chemistry that’s close enough to sugar to taste sweet but different enough that they’re not fully metabolized.

• Erythritol (0.2 cal/g): the most-used sugar alcohol. Almost entirely absorbed in the small intestine and excreted unchanged in urine — meaning it doesn’t ferment in the colon and rarely causes GI upset. Sweetness is ~70% of sucrose. Often combined with stevia or monk fruit in commercial blends.
• Xylitol (2.4 cal/g): roughly as sweet as sucrose. Metabolized more like a regular carb than erythritol. Notably toxic to dogs even in small amounts.
• Maltitol (2.1 cal/g): sweet, but the most likely of the common sugar alcohols to cause GI upset (gas, bloating, diarrhea) because it’s heavily fermented by gut bacteria.
• Sorbitol (2.6 cal/g): similar GI tolerance issues to maltitol.
• Allulose (0.4 cal/g): a “rare sugar” that’s chemically a sugar but absorbed and excreted nearly unchanged. Sweetness is ~70% of sucrose. FDA-approved and excluded from “Total Sugars” on labels since 2019. Generally well-tolerated.

The 2023 erythritol study — Witkowski et al., published in Nature Medicine — observed an association between higher blood erythritol levels and cardiovascular events. The study had real limitations (the people with elevated erythritol were on a sicker baseline, making causality hard to attribute) and the FDA continues to classify erythritol as GRAS. The current honest read: this is a hypothesis-generating finding, not a definitive harm signal, and the matter is being actively studied.

For most people, sugar alcohols are a reasonable tool for reducing sugar intake. The main practical issue is GI tolerance — start with smaller amounts (10–15 g/day) and increase as tolerated.

Non-nutritive sweeteners#

These sweeteners are 200–13,000× sweeter than sucrose by weight, so the amounts used to sweeten a product are tiny.

Stevia (Reb A, Reb M)#

Source: leaves of the Stevia rebaudiana plant.

The sweetness is from steviol glycosides — the most common are rebaudioside A and rebaudioside M.

• FDA status: GRAS (since 2008).
• Metabolism: Steviol glycosides are not absorbed. Gut bacteria cleave them to steviol, which is absorbed, processed by the liver, and excreted in urine.
• Safety: No reliable evidence of harm at typical intake. The acceptable daily intake (ADI) is 4 mg/kg of body weight per day of steviol equivalents — roughly 12 packets for a 70 kg adult.
• Aftertaste: the usual complaint. Reb M has less bitter aftertaste than Reb A; modern stevia products often blend the two.

Monk fruit#

Source: extract of the Siraitia grosvenorii fruit.

The sweetness is from mogrosides, especially mogroside V.

• FDA status: GRAS (since 2010).
• Metabolism: Largely not absorbed.
• Safety: Clean track record. No specific concerns.
• Aftertaste: Cleaner than stevia for most people; often the preferred natural-source non-nutritive sweetener.

Sucralose#

Source: synthetic, derived from sucrose with three chlorine substitutions. About 600× sweeter than sugar.

• FDA status: Approved 1998.
• Brand: Splenda.
• Metabolism: ~85% passes through unabsorbed; ~15% absorbed and excreted unchanged in urine.
• Safety: Long history of use. The most-studied non-nutritive sweetener. The 2023 WHO advisory on non-sugar sweeteners noted observational associations between high sucralose intake and cardiovascular events; the same advisory acknowledged that the observational data is confounded by reverse causation (people switching to NNS because they have metabolic issues).
• Concerns: A 2016 study suggested sucralose could break down to chloropropanols at very high baking temperatures. For typical use (in beverages, sweetening foods at typical cooking temps), this isn’t a meaningful concern.

Aspartame#

Source: synthetic, made from two amino acids (phenylalanine and aspartic acid) plus a methyl ester. ~200× sweeter than sugar.

• FDA status: Approved 1981.
• Brand: NutraSweet, Equal.
• Metabolism: Broken down in the digestive tract into phenylalanine, aspartic acid, and methanol — all components found naturally in many foods.
• PKU warning: People with phenylketonuria (PKU) cannot metabolize phenylalanine and must avoid aspartame. The label declaration “phenylketonurics: contains phenylalanine” exists for this reason.
• 2023 WHO classification: The IARC (the WHO’s cancer agency) classified aspartame as Group 2B — “possibly carcinogenic to humans” based on limited evidence in humans and inadequate evidence in animals. Group 2B is the same category as gasoline exhaust and coconut oil; it does not mean the substance is definitively carcinogenic. The simultaneous JECFA review (the WHO’s food-safety committee) maintained the existing acceptable daily intake of 40 mg/kg/day, equivalent to about 14 cans of diet soda per day for a 70 kg adult.
• Practical translation: the 2023 review didn’t change FDA approval or recommended intake. People who are anxious about aspartame can substitute stevia, monk fruit, or sucralose; people comfortable with the IARC Group 2B classification can continue using it within ADI.

Acesulfame-K (Ace-K)#

Source: synthetic, ~200× sweeter than sugar.

• FDA status: Approved 1988.
• Metabolism: Not metabolized; excreted unchanged in urine.
• Safety: Long use, no significant concerns. Often blended with aspartame or sucralose because it provides a slightly different sweetness profile that masks aftertaste.

Saccharin#

The oldest non-nutritive sweetener (discovered 1879, approved in the U.S. in 1958).

• Brand: Sweet’N Low.
• Metabolism: Not metabolized; excreted unchanged in urine.
• History: A 1970s rat study suggested bladder cancer risk; follow-up research established the mechanism was rat-specific and doesn’t apply to humans. The “saccharin causes cancer” warning was removed from U.S. labels in 2000.
• Current status: Generally considered safe at typical intake. Less popular than newer sweeteners because of bitter aftertaste.

Neotame and advantame#

Newer high-intensity sweeteners (~7,000× and ~20,000× sweeter than sugar, respectively). Used in industrial food products at very low concentrations. Limited consumer use; not commonly seen in tabletop products.

Effects on weight and metabolism#

The major reason most people use non-nutritive sweeteners is weight management. The evidence:

• Replacing sugar-sweetened beverages with non-nutritive-sweetened versions reliably reduces caloric intake — multiple controlled trials show 100–250 calorie/day reductions, translating to modest weight loss over months. This is the strongest, most consistent finding.
• Effects on appetite and subsequent intake are mixed. Some studies suggest non-nutritive sweeteners increase craving for sweet foods later; other studies don’t replicate this. The current consensus is that the effect is small if it exists.
• The 2023 WHO advisory on non-sugar sweeteners recommended against using them for long-term weight control, citing observational associations with increased cardiovascular events and Type 2 diabetes. The advisory has been criticized for overweighting reverse-causation-prone observational data; the AND, AHA, and ADA have not changed their guidance based on it.

The honest practical synthesis: non-nutritive sweeteners are a useful transition tool away from heavy sugar consumption. Sustained heavy daily use is more uncertain. Treat them as bridges, not destinations.

How to choose#

A reasonable hierarchy for someone trying to reduce added sugar:

1. Use less sweetener overall. Train your palate down. Most people who reduce sweetness for 2–4 weeks find their preferred sweetness level lowers permanently.
2. For coffee and tea, try unsweetened first. Most people who stop sweetening discover they prefer it within a few weeks.
3. For occasional sweet treats, real sugar is fine in moderation. The “added sugars under 25 g/day” target is achievable without non-nutritive substitutions if your defaults are unsweetened.
4. For diet sodas and “diet” products, the most-studied options are stevia, monk fruit, sucralose, and acesulfame-K. Preferences vary; pick what tastes best to you.
5. For baking, erythritol or allulose blends usually work without significant texture issues. Stevia and monk fruit alone sometimes leave aftertaste in baked goods.
6. For people with diabetes or carb-sensitive metabolic patterns, non-nutritive sweeteners are a useful tool to reduce glycemic load while keeping food palatable. Work with a clinician.

A note on “natural” vs. “artificial”#

The “natural sweeteners” framing — usually meaning stevia, monk fruit, allulose, and erythritol — vs. “artificial” — meaning the synthetic ones — is a marketing distinction more than a scientific one. Synthetic doesn’t mean harmful, and natural doesn’t mean safe. The actual question is the evidence on each specific compound: typical intake, safety data, decades of use.

That said, if you’d rather use sweeteners derived from plant sources, stevia and monk fruit are the well-studied options.

Frequently asked questions#

Where to go next#

• Macronutrients Explained — the broader framework
• Carbohydrates Decoded — the carb side
• The Sugar Detective’s Guide — added sugar in disguise
• The Complete Guide to Reading the Nutrition Facts Label — finding sweeteners on labels
• Drink Smarter — beverages with and without sugar

Sources#

1. World Health Organization. Use of non-sugar sweeteners: WHO guideline. 2023. who.int
2. International Agency for Research on Cancer (WHO). Aspartame hazard and risk assessment. 2023. iarc.who.int
3. Witkowski M, Nemet I, Alamri H, et al. The artificial sweetener erythritol and cardiovascular event risk. Nature Medicine, 2023. PubMed
4. Magnuson BA, Carakostas MC, Moore NH, et al. Biological fate of low-calorie sweeteners. Nutrition Reviews, 2016. PubMed
5. U.S. Food and Drug Administration. High-Intensity Sweeteners. fda.gov
6. American Heart Association. Position statement on non-nutritive sweeteners. heart.org
7. Pearlman M, Obert J, Casey L. The Association Between Artificial Sweeteners and Obesity. Current Gastroenterology Reports, 2017. PubMed