E960 / Glazing agents, gases & sweeteners
Steviol Glycosides (Stevia)
Safety in brief
Safe in moderation
Recorded EatMatrix rating, not a new clinical review. The amount, your sensitivities and country rules still matter.
- Watch for
- No allergen or sensitivity is recorded for this ingredient here. Check the whole product label; this is not an allergen-free guarantee.
- Our advice
- “Plant-derived” does not establish a health benefit. The 2026 result applies to the tested rebaudioside-A preparation, while the earlier trial used stevia sachets. Neither gives every stevia extract or leaf product the same evidence.
What it is
Natural non-caloric sweeteners extracted from the leaves of the Stevia rebaudiana plant, roughly 200-400 times sweeter than sugar. Rebaudioside A (Reb A) is one of the most common high-purity components used commercially. EFSA and the FDA (as a GRAS-notified ingredient) consider steviol glycosides safe within established acceptable daily intake levels.
Also appears on labels as: Steviol Glycosides (Stevia), Steviol Glycosides, Stevia, Stevia extract, Rebaudioside A, Rebaudioside D, Rebaudioside M, Stevia sweetener, Stevia leaf extract, Stevia rebaudiana extract, E960, E 960.
Dietary information
- Vegan
- Yes
- Halal
- Yes
- Gluten-free
- Yes
- Allergens
- None recorded
Recorded ingredient information. Check the product label or certification: sources, other ingredients and cross-contact can vary.
Research & body effects
What human studies mean for E960
Blood sugar and gut bacteria: a short experiment
A 2022 randomized study assigned 120 healthy adults to two weeks of saccharin, sucralose, aspartame or stevia sachets, a glucose-vehicle control, or no supplement. Saccharin and sucralose impaired measured glucose responses compared with controls; this finding was not uniform across all four sweeteners.
This ingredient was studied, but it was not one of the two identified in the abstract as significantly impairing glucose responses. That does not prove long-term benefit or safety for every preparation.
Changes in gut bacteria are not automatically gut damage. Transferring participants’ microbes into mice supported a possible mechanism; it was not a second human trial or a validated test of your personal sensitivity.
Four sweeteners, gut microbes and glucose responses. Indexed abstract and selected publisher text read; complete methods and supplements pending. Indexed comments were identified, not fully appraised. Some authors disclose microbiome-company interests.
A July 2026 trial: compare the treatments, not just the starting value
Thirty-nine healthy adults completed three 14-day periods: 66 g/day sucrose, 220 mg/day saccharin, or 220 mg/day steviol glycosides containing more than 97% rebaudioside A. The study found no significant between-treatment differences in glucose or insulin regulation. Sucrose produced higher triglycerides than saccharin.
Some measurements changed from the initial baseline. There was no unsweetened control period, and that baseline was not a randomized treatment. Those changes alone cannot show that a sweetener caused harm compared with having none. Short follow-up and healthy volunteers limit the conclusion.
The tested stevia preparation was chiefly rebaudioside A. Do not transfer its result to whole stevia leaves, every extract or every E960 subform.
Saccharin, rebaudioside A and sucrose: 14-day crossover trial. Selected main-text methods, results, limitations and funding read; supplements not fully appraised. Funded through the VINNOVA Antidiabetic Food Centre.
Related whole-diet evidence: a benefit and a trade-off
The 2025 SWEET trial compared low-sugar diets that allowed or excluded products with many sweeteners after an initial weight-loss phase. The analysis of 277 adults who reached the initial loss target favoured allowing these products by about 1.6 kg at one year. The difference was not statistically significant when only completers were analysed.
About 40% of recruited adults did not complete the study. The sweetener group reported more abdominal pain, loose stools and gas. Many products and ingredients varied together: neither the weight result nor the symptoms can be attributed to this one additive.
Related whole-diet SWEET trial: weight maintenance and gut symptoms. Selected main-text methods, results, limitations and declarations read; supplements not fully appraised. EU Horizon 2020 funding; several authors disclose food-industry relationships.
EatMatrix’s reading
“Plant-derived” does not establish a health benefit. The 2026 result applies to the tested rebaudioside-A preparation, while the earlier trial used stevia sachets. Neither gives every stevia extract or leaf product the same evidence.
A practical choice
Compare the whole food and your usual alternative, including free sugars, fibre and how often you eat it. Water or an unsweetened drink is a straightforward option. A sweetener name alone cannot tell you the dose in a serving or predict your personal response.
WHO’s 2023 recommendation against using non-sugar sweeteners for long-term weight control is conditional and excludes people with pre-existing diabetes. It is nutrition guidance, not a country ban or a replacement for an additive safety assessment. Read WHO’s scope and limits.
Selected sources checked . The reading scope is stated beside each source. These are selected findings, not an exhaustive clinical review or a personal diet prescription. Indexed notices were checked where accessible; missing notices do not prove that no correction or later version exists. The research does not by itself change recorded ratings or country permissions. Use the country section for the separate legal question.
Selected research on E960
How to read study types- Randomized trialSaccharin, rebaudioside A and sucrose: 14-day crossover trial
Thirty-nine adults completed three 14-day periods. No significant between-treatment glucose/insulin differences were found. Baseline changes are not a randomized comparison with no sweetener. The stevia preparation was mainly rebaudioside A.
- Randomized trialRelated whole-diet SWEET trial: weight maintenance and gut symptoms
Related whole-diet trial favoured sweetener-containing products for weight maintenance in one analysis, but not the completer analysis. Dropout was high and gut symptoms were more frequent. Many ingredients varied together; no individual-additive effect was isolated.
- Randomized trialFour sweeteners, gut microbes and glucose responses
Two-week trial in 120 healthy adults: saccharin and sucralose impaired measured glucose responses; this was not uniform across all four sweeteners. Microbiome changes and mouse transfers do not establish future diabetes or a personal sensitivity test.
Earlier source summaries checked ; latest additions checked . See the detailed research section for what was read and the remaining limits. This is a selection of sources, not a complete literature review. Authority assessments apply to the uses and places they cover.
Health notes from our ingredient record
- Some people report a bitter or licorice-like aftertaste, which is a sensory rather than safety concern.
- Limited reports of mild digestive upset (bloating, nausea) at very high intake, though this is not well established in controlled studies.
Country rules
Country checks for E960
Permission depends on the market, type of food, amount and conditions. A scientific opinion or an E-number alone does not establish a legal use. These checks are separate from the stored EatMatrix rating.
Compare all eight markets and their sources
| Market | What is checked |
|---|---|
| European Union | No exact code match in the Annex II Part B list (version 2026-08-18, read 2026-10-07). Identity, subform and legal-use review are pending. This does not establish a permission or a ban. |
| Great Britain | No verified exact register record in this review. Identity and current use rules need checking; missing a record does not establish a ban. |
| United States | Complete current-use review pending. Any scoped provision in the label guide above applies only to the stated use. FDA explains the applicable food-additive and GRAS framework. |
| Northern Ireland | Review pending. Check the applicable EU rules and territory arrangements; the GB register alone does not establish this status. FSA guidance is a starting source, not a verified permission for this entry. |
| Canada | Named food-use records located. Health Canada lists the preparations below for specified foods and purposes. This is conditional permission for those uses, not approval of every product or a personal safety verdict. The source conditions apply to specified steviol glycosides and their stated sources, not every stevia leaf or extract. Source tables captured . List publication dates are shown separately. Good Manufacturing Practice (GMP) is a condition of use, not an unlimited amount or a recommended intake. Food and Drug Regulations B.16.003 requires the listed source, purpose, amount and other conditions to match. The official consolidation read was current to 21 September 2026; product specifications and applicable later changes still need checking. Read 11 listed food-use conditionsSteviol GlycosidesHealth Canada: List of Permitted Sweeteners. List published ; item S.5.
EatMatrix reading: match the exact preparation and food first. A manufacturing limit is not an acceptable daily intake (ADI). A label without the amount cannot establish exposure, and an unlisted name here does not establish a ban. |
| Australia | Review pending. Food Standards Code permissions and conditions need an additive-specific check. FSANZ food-additive rules is a starting source, not a verified permission for this entry. |
| New Zealand | Review pending. Food Standards Code permissions and conditions need an additive-specific check. FSANZ food-additive rules is a starting source, not a verified permission for this entry. |
| Japan | Review pending. The relevant list, identity and use standards need an additive-specific check. CAA food-additive lists is a starting source, not a verified permission for this entry. |
Sources & review status
Check what each source covers: an authority’s conclusion, a research finding and a country’s legal permission answer different questions.
What this guide covers
- Selected research
- 3 source summaries, with publication dates and study types. Several publications may describe the same study population.
- Original ingredient record
- No research links in the original record. Sources added separately to this guide are not included in this count.
This guide brings together our ingredient record and the dated sources shown here. Missing information describes a gap in this guide; it does not mean an ingredient is untested, safe or harmful. This is not a complete review of every study, preparation or current country rule. The amount in a product and how it is used still matter.
How to read the ratings, research and country checksAgency labels recorded by EatMatrix
These are imported database classifications, not verified current statements from each authority. Use the dated official sources below for their actual scope and conclusions.
| Authority | Database label |
|---|---|
| FDA | Safe |
| WHO | Safe in moderation |
| EFSA | Safe in moderation |
| FSANZ | Safe in moderation |
| Japan | Safe |
| UK FSA | Safe in moderation |
| Health Canada | Safe in moderation |
Database updated: . What does an E number mean on a food label?
Common questions
Frequently asked questions
What is EatMatrix's recorded rating for E960?
The recorded EatMatrix rating for E960 (Steviol Glycosides (Stevia)) is “Safe in moderation”. It is calculated from labels stored in our ingredient database. These labels are not verified current regulator decisions or country permissions. A database update date is not a scientific review date.
Is E960 vegan, halal or gluten-free?
According to our ingredient data for Steviol Glycosides (Stevia) — vegan: yes, vegetarian: yes, halal: yes, kosher: yes, gluten-free: yes, dairy-free: yes. Manufacturing sources can vary between products; check the label if it matters to you.
What is E960 and why is it used?
Natural non-caloric sweeteners extracted from the leaves of the Stevia rebaudiana plant, roughly 200-400 times sweeter than sugar. Rebaudioside A (Reb A) is one of the most common high-purity components used commercially. EFSA and the FDA (as a GRAS-notified ingredient) consider steviol glycosides safe within established acceptable daily intake levels.
Does E960 contain allergens?
No known allergens are recorded for Steviol Glycosides (Stevia).
Read the whole label.
Scan a food label with EatMatrix on iPhone to look up its ingredients in the app.
Ingredient research, by email
Join the list for food additive research updates. Unsubscribe any time.
This page shows information and classifications recorded in our ingredient database. Current scientific and market-specific verification remains pending as described above. It is general information, not medical advice.