Stevia

Stevia is a high-intensity, non-caloric sweetener derived from the leaves of Stevia rebaudiana Bertoni, a plant native to Paraguay and Brazil. The sweet-tasting compounds responsible for its intense sweetness are a family of diterpene glycosides collectively known as steviol glycosides. The most commercially prominent of these are stevioside and rebaudioside A (Reb A), which can be 200–400 times sweeter than sucrose by weight.

Steviol glycosides are approved for use in food and beverages by major regulatory agencies worldwide, including the U.S. Food and Drug Administration (FDA), the European Food Safety Authority (EFSA), and the Joint FAO/WHO Expert Committee on Food Additives (JECFA). They are used as a sugar substitute in a wide range of products including soft drinks, dairy products, baked goods, confectionery, and tabletop sweeteners.

From a health perspective, stevia-based sweeteners do not raise blood glucose or insulin levels in the way that sucrose does, making them of particular interest for individuals managing diabetes or caloric intake. The established Acceptable Daily Intake (ADI) is 4 mg/kg body weight per day, expressed as steviol equivalents. At typical dietary exposures, stevia is generally considered safe by authoritative scientific bodies.

Some aspects of stevia's health effects remain under active investigation, including its potential influence on the gut microbiome and long-term metabolic outcomes. Public discourse around stevia has occasionally been influenced by misinformation, conflating well-studied high-purity steviol glycoside extracts with crude, unrefined leaf preparations or other sweeteners entirely.

Steviol glycosides share a common aglycone core called steviol, a tetracyclic diterpene acid based on the ent-kaurane skeleton. The steviol backbone (molecular formula C₂₀H₃₂O₃) bears carboxyl and hydroxyl functional groups at specific positions that serve as attachment points for sugar moieties. In stevioside, two sophorose-type sugar units (glucose) are attached to the C-13 hydroxyl group via a β-glycosidic linkage, and a single glucose is esterified at the C-19 carboxyl group. In rebaudioside A (the most abundant glycoside in modern commercial extracts), an additional glucose unit is present on the C-13 sugar chain, contributing to its cleaner, less bitter taste profile compared with stevioside. Newer commercially relevant glycosides such as rebaudioside D and rebaudioside M carry still larger sugar chains and are considered to have taste profiles even more similar to sucrose. The intense sweetness arises from the interaction of these glycosides with sweet taste receptors (T1R2/T1R3 heterodimers) on human taste cells, with the sugar chain length and branching strongly influencing perceived sweetness and aftertaste.

Commercial production of steviol glycosides begins with the cultivation and harvest of Stevia rebaudiana leaves, predominantly in China, Paraguay, Kenya, and India. After harvesting, leaves are dried to reduce moisture content, then subjected to hot-water extraction to release steviol glycosides into solution. The resulting aqueous extract is clarified through filtration and ion-exchange or adsorption resin chromatography to remove pigments, chlorophylls, polyphenols, and other non-target plant compounds. The clarified extract is then concentrated and subjected to crystallisation or spray-drying to yield a high-purity white powder. Regulatory standards in the EU and USA require a minimum purity of 95% total steviol glycosides in the final product. For newer glycosides such as rebaudioside M and rebaudioside D—which occur at low natural concentrations in the leaf—an alternative biosynthetic or enzymatic bioconversion route is used: stevioside or rebaudioside A is subjected to enzymatic glycosylation using UDP-glucosyltransferases expressed in yeast or bacterial fermentation systems, enabling cost-effective production of these otherwise rare glycosides at scale. Finished product is standardized for sweetness potency and blended with carriers such as maltodextrin or erythritol for tabletop applications.

The genus Stevia includes approximately 240 species native to the Americas, but Stevia rebaudiana is the only species with commercially significant sweetness. Indigenous Guaraní peoples of Paraguay and southern Brazil used leaves of the plant, which they called ka'a he'ê (sweet herb), for centuries to sweeten herbal teas and as a general tonic. The plant was first documented for Western science by Paraguayan botanist and natural historian Moisés Santiago Bertoni in 1899, who described it formally and named it in honor of Spanish botanist Pedro Esteve. The sweet-tasting constituents were isolated and partially characterised by chemists Ovidio Rebaudi and M. S. Bertoni in the early twentieth century. Full structural elucidation of stevioside was achieved in the 1950s and 1960s by Japanese, Dutch, and Paraguayan research groups. Japan became the first country to commercialise steviol glycosides as a food sweetener in the early 1970s, driven by concerns about the safety of then-common synthetic sweeteners. By the 1980s, stevia extracts held a substantial share of the Japanese sweetener market. Broad Western regulatory approval was slower: the FDA initially issued an import alert on stevia in 1991, citing insufficient safety data, but revised this position following JECFA's safety evaluation. In 2008, the FDA granted GRAS (Generally Recognized as Safe) status to high-purity rebaudioside A, and EFSA authorised steviol glycosides as E960 in the European Union in 2011. The global stevia ingredient market has grown substantially since then, driven by consumer demand for reduced-sugar and natural-positioned products.

• Intense sweetness at low doses: Steviol glycosides are 200–400 times sweeter than sucrose, allowing very small quantities to replace large amounts of sugar, reducing caloric content.
• Zero glycaemic impact: They are not metabolised to glucose and do not raise blood sugar or stimulate insulin secretion, making them suitable for diabetic and low-carbohydrate formulations.
• Thermal stability: Stability at baking temperatures allows use in cooked and baked products where some other sweeteners degrade.
• pH stability: Stable across a wide pH range (3–9), making them suitable for acidic beverages such as carbonated soft drinks and fruit juices.
• Natural positioning: Derived from a plant source, enabling 'natural sweetener' labeling in many markets, which aligns with consumer preferences for clean-label products.
• Non-cariogenic: Does not promote dental caries because oral bacteria cannot ferment steviol glycosides to produce acid.
• Regulatory acceptance: Approved in the majority of global markets, facilitating international product formulation.
• Blending synergies: Combines well with other sweeteners (e.g. erythritol, sucralose) to moderate aftertaste and achieve a more sucrose-like flavor profile.

Glycaemic control and diabetes management

Multiple randomised controlled trials and systematic reviews have demonstrated that steviol glycosides do not raise blood glucose or insulin levels in healthy adults or individuals with type 2 diabetes when consumed in place of sucrose. A 2018 systematic review published in PLOS ONE (Tey et al.) found no significant effect of stevia on postprandial blood glucose, insulin, or glucagon-like peptide-1 (GLP-1) in healthy adults. This evidence base is considered moderately strong for acute postprandial effects; long-term glycaemic outcomes are less well characterised.

Caloric reduction and weight management

Because steviol glycosides contribute negligible calories, replacing sugar with stevia reduces the caloric density of foods and beverages. Whether this translates to meaningful weight loss in real-world dietary settings is less certain; evidence for sustained body weight reduction attributable specifically to stevia use is limited and inconsistent across trials, partly because total dietary compensation behavior varies among individuals.

Blood pressure

Some early studies, particularly those using high-dose stevioside in hypertensive populations in China, reported modest antihypertensive effects. However, the doses used in these studies (750–1500 mg/day of stevioside) substantially exceed typical dietary exposure from food uses, and findings have not been consistently replicated. Regulatory bodies do not currently classify steviol glycosides as having a proven blood-pressure-lowering function at normal food-use levels. This area remains under investigation and evidence is limited.

Dental health

Stevia is non-cariogenic. It is not fermented by cariogenic oral bacteria (Streptococcus mutans), and some in-vitro studies suggest it may inhibit bacterial growth; however, clinical evidence for a direct anti-caries benefit beyond the absence of fermentable substrate is preliminary.

Established findings

At levels consistent with current ADI (4 mg/kg body weight/day as steviol equivalents), no adverse effects have been established in healthy adults by JECFA, EFSA, or FDA review processes. High-purity steviol glycoside extracts are considered safe for the general population at approved use levels.

Aftertaste and palatability

At higher concentrations, steviol glycosides—particularly stevioside—can produce a bitter or liquorice-like aftertaste. This is a sensory and formulation concern rather than a safety issue, but it limits the practical maximum concentration usable in some foods.

Gut microbiome — limited/emerging evidence

A small number of in-vitro and animal studies have suggested that steviol glycosides may alter the composition of gut microbiota. A 2020 study by Ruiz-Ojeda et al. noted some changes in microbial populations in rodent models. Human clinical data are very limited, and no adverse health outcome attributable to microbiome changes from stevia consumption has been demonstrated in humans at dietary levels. This remains an area of ongoing research and cannot currently be characterised as an established risk.

Allergy — limited evidence

Stevia rebaudiana belongs to the Asteraceae (Compositae) family, which includes plants that cause allergic reactions in some individuals (e.g. ragweed, chrysanthemums). Isolated case reports of allergic contact dermatitis and hypersensitivity have been published, but the frequency of clinically significant allergy to high-purity steviol glycoside extracts appears to be very low. Those with known Asteraceae hypersensitivity are sometimes advised to exercise caution, though regulatory bodies have not issued formal warnings.

Drug interactions — limited evidence

High pharmacological doses of stevioside have shown interactions with antihypertensive and antidiabetic medications in animal and some small human studies. These interactions are unlikely to be clinically significant at normal dietary exposure levels, but individuals on relevant medications who consume very large amounts of stevia-sweetened products may wish to consult a healthcare provider. Evidence is limited.

Pregnancy and special populations

EFSA and JECFA have not identified specific risks for pregnant or breastfeeding women consuming steviol glycosides within the ADI. However, robust prospective data in pregnant populations are limited, and the standard precautionary note of remaining within established ADI levels applies.

The Acceptable Daily Intake (ADI) established by the Joint FAO/WHO Expert Committee on Food Additives (JECFA) and adopted by EFSA is 4 mg/kg body weight per day, expressed as steviol equivalents. To convert from specific glycosides: 1 mg of stevioside corresponds to approximately 0.40 mg steviol equivalents; 1 mg of rebaudioside A corresponds to approximately 0.33 mg steviol equivalents.

For a 70 kg adult: The ADI equates to approximately 280 mg steviol equivalents per day, corresponding to roughly 700–840 mg of rebaudioside A or stevioside. A standard can of a stevia-sweetened beverage typically contains 90–130 mg of steviol glycosides, placing typical single-serving consumption well below the ADI.

For children: Because children have lower body weights, the ADI in absolute terms is proportionally lower (e.g. approximately 80 mg steviol equivalents/day for a 20 kg child). At reported levels of consumption among children in most markets, dietary exposure is estimated to remain below the ADI, but regular high consumption of multiple stevia-sweetened products could bring sensitive subgroups (particularly low-body-weight children with high soft-drink consumption) closer to the ADI boundary. Monitoring by EFSA and national agencies is ongoing.

Pregnancy and breastfeeding: No specific lower ADI has been set for pregnant or lactating women. As a general precautionary principle, remaining well within established ADI limits is advised. Individuals with specific concerns should consult a registered dietitian or physician.

The evidence base for steviol glycosides spans several decades and covers areas including toxicology, metabolism, glycaemic response, and emerging questions about the gut microbiome. Metabolic and safety studies established that steviol glycosides are not absorbed intact from the upper gastrointestinal tract; they are cleaved by colonic bacteria to steviol, which is then absorbed, conjugated to glucuronide in the liver, and excreted in urine. This metabolic fate underpins the lack of caloric contribution and was a key factor in regulatory safety assessments. Glycaemic response research is considered moderately robust: a 2019 systematic review and meta-analysis by Samuel et al. (European Journal of Clinical Nutrition) covering 12 randomised trials found that stevia consumption did not significantly affect fasting or postprandial glucose or insulin levels compared with sugar or other non-caloric sweeteners; however, most trials were short-term and used single-serving designs. Gut microbiome studies are at an early stage. In-vitro fermentation studies (e.g. Kosti et al., 2021, Nutrients) show that steviol glycosides can be partially metabolised by colonic bacteria, and some animal studies have reported shifts in microbial community composition, but whether these changes are adverse or neutral in humans is unresolved. Antihypertensive research produced intriguing early results from trials by Ferri et al. (2006) and others using high-dose stevioside, but methodological concerns—including small sample sizes and doses far exceeding typical dietary exposure—limit generalisability. Genotoxicity and carcinogenicity have been extensively studied; multiple in-vitro and in-vivo tests have not produced evidence of genotoxicity or carcinogenicity for purified steviol glycosides at relevant doses, forming a key pillar of regulatory approvals. Ongoing research areas include long-term epidemiological studies, effects on appetite regulation, and the metabolic impact of newer glycoside variants (Reb M, Reb D).

Stevia has been subject to several waves of controversy and misinformation since its introduction to Western markets. The early FDA import alert (1991) was widely interpreted in popular media as evidence that stevia was dangerous; in fact, the alert reflected insufficient safety data at the time rather than evidence of harm, and it did not prevent stevia's sale as a dietary supplement in the USA. This history has been selectively cited by critics to suggest that regulators 'suppressed' stevia, a narrative that does not accurately reflect the regulatory record. The 'natural vs synthetic' debate has generated ongoing discussion: some consumer advocates argue that enzymatically produced glycosides (Reb M, Reb D) should not be labeled 'natural' or 'stevia-derived' because their production involves fermentation and enzymatic modification rather than direct leaf extraction. Regulatory bodies have taken varying positions on this; in the EU, these compounds are labeled E960c to distinguish them. Claims of suppression by the sugar industry circulate on social media and in alternative-health publications, alleging that sugar industry lobbying drove early regulatory resistance to stevia. While the sugar industry did historically file objections to stevia petitions, the primary reason for delayed FDA approval was a genuine absence of the safety data required under applicable regulatory standards. Exaggerated health benefit claims—particularly that stevia 'cures' diabetes, lowers blood pressure, or has anti-cancer properties—are common in alternative-health media. These claims exceed what current evidence supports. Conversely, some anti-sweetener campaigns have conflated stevia with synthetic sweeteners such as aspartame, attributing adverse effects that are not supported by the specific evidence base for steviol glycosides. The scientific and regulatory consensus is that high-purity steviol glycosides are safe at approved use levels, and this consensus is not seriously contested within the mainstream scientific community.

Stevia rebaudiana is a perennial plant cultivated primarily in China (Shandong, Fujian provinces), as well as Paraguay, Kenya, and India. Compared with sugar cane and sugar beet per equivalent unit of sweetness delivered, stevia cultivation requires substantially less land and water—estimates suggest stevia uses approximately 10–15% of the land area needed to produce an equivalent quantity of sweetness from sugar cane, owing to its high sweetness intensity. Life-cycle assessment (LCA) studies, including an industry-sponsored LCA published by PureCircle (2014), indicate a lower carbon footprint per unit of sweetness compared with sugar, but these assessments should be interpreted with awareness of their sponsorship. Water use varies significantly by cultivation region and farming practice; drip-irrigation systems substantially reduce water footprint. The extraction and purification process is water- and solvent-intensive, which partially offsets raw material efficiency gains. Stevia cultivation does not require unique pesticides beyond standard agricultural practice; however, as with any monoculture, large-scale stevia farming raises generic concerns about biodiversity, soil health, and agrochemical use that have not been specifically documented as exceptional compared with other specialty crops. Enzymatic fermentation production of rare glycosides (Reb M, Reb D) may have a different environmental profile, potentially reducing crop land requirements further, but detailed independent LCAs of this production route are limited in the published literature.

Occupational exposure to stevia in the food and beverage manufacturing sector is primarily through inhalation of powdered steviol glycoside dust during handling, blending, and packaging operations. Because steviol glycosides are derived from a plant in the Asteraceae family, there is a theoretical risk of occupational sensitisation and respiratory allergic responses (occupational asthma or rhinitis) in workers exposed to plant-derived dust, analogous to known occupational allergens from other Asteraceae species. Case reports in the peer-reviewed literature on this specific issue are very limited. Standard industrial hygiene practices—including enclosed handling systems, appropriate respiratory protective equipment, and workplace dust monitoring—are recommended as precautionary measures consistent with good manufacturing practice for any fine organic powder. There are no specific occupational exposure limits (OELs) established by national agencies specifically for steviol glycosides; general organic dust OELs apply in most jurisdictions.

The safety of steviol glycosides has been extensively evaluated in animal studies as part of the regulatory dossiers reviewed by JECFA and EFSA. Sub-chronic and chronic toxicity studies in rats and mice administered high doses of stevioside and rebaudioside A (up to 2,500 mg/kg bw/day in some studies) did not produce treatment-related adverse effects at doses relevant to human exposure. The No-Observed-Adverse-Effect Level (NOAEL) used by JECFA to derive the ADI was 970 mg/kg bw/day in rats, providing a substantial safety margin. Genotoxicity studies, including Ames tests, chromosomal aberration assays, and in-vivo micronucleus tests, consistently returned negative results for purified steviol glycosides; early concerns about genotoxicity of steviol (the aglycone) in some in-vitro studies were not replicated in more definitive assays. Reproductive and developmental toxicity studies in rats and rabbits found no evidence of teratogenicity or adverse reproductive outcomes at high dietary doses. Carcinogenicity bioassays in rodents did not demonstrate increased tumour incidence. Regarding the gut microbiome, several rodent studies (notably Abou-Donia et al., 2008, and subsequent replications) reported alterations in faecal microbiota composition with high-dose stevioside or rebaudioside A; interpretation is complicated by the very high doses used, which do not reflect human dietary exposure, and by significant differences between rodent and human gut microbiome composition.

Human clinical data on steviol glycosides cover glycaemic response, blood pressure, safety, tolerability, and—more recently—gut microbiota effects. Pharmacokinetic studies consistently show that intact steviol glycosides are not absorbed in the small intestine; they are hydrolysed by colonic microbiota to steviol, absorbed and conjugated hepatically, and excreted in urine as steviol glucuronide, with no accumulation. Acute glycaemic response trials: Multiple randomised crossover trials, including work by Geuns et al. (2003), Anton et al. (2010), and Tey et al. (2017), have confirmed that stevia-sweetened preloads do not significantly raise postprandial glucose or insulin compared with water, sucrose, or other non-caloric sweeteners in healthy adults. A 2017 randomised trial by Sylvetsky et al. (Nutrients) found no acute insulin or glucose effect in adolescents. Type 2 diabetes: A 12-week double-blind RCT by Barriocanal et al. (2008) in patients with type 2 diabetes found no significant changes in glucose, insulin, or blood pressure compared with placebo when consuming 1000 mg stevioside/day—a dose substantially above typical dietary exposure. Appetite and energy intake: Studies examining compensatory caloric intake following stevia-sweetened meals have produced mixed results; some (e.g. Anton et al., 2010) reported lower total energy intake, while others found compensation. Meta-analytic data from the 2019 Samuel et al. review did not support a significant effect on body weight over short-term intervention periods. Gut microbiome: A 2020 double-blind RCT by Lobach et al. (Nutrients) found no significant changes in faecal microbiota diversity after four weeks of steviol glycoside consumption in healthy adults, though power was limited. Overall, the human evidence base is considered adequate for regulatory purposes at current approved use levels but is acknowledged by agencies including EFSA to have gaps in long-term exposure data and data in specific vulnerable populations.

In the European Union, steviol glycosides must be declared in the ingredients list by their functional class followed by specific name or E number: for example, 'sweetener: steviol glycosides (E960)' or 'sweetener: E960'. Products labeled 'with sweetener(s)' must carry an adjacent indication on the label. Enzymatically produced rebaudioside D and M are listed under E960c.

In the United States, stevia-derived sweeteners may appear on ingredients lists under a range of names including: stevia extract, stevia leaf extract, steviol glycosides, rebaudioside A, Reb A, rebaudioside M, or specific trade names such as Truvia, PureVia, or Enliten. The term 'stevia' alone is commonly used in consumer-facing product descriptions and front-of-pack claims.

In Australia and New Zealand, declaration as 'steviol glycosides' or by E number is standard in the ingredients list under FSANZ requirements.

In Canada, the ingredient is declared as 'steviol glycosides' or 'stevia extract' in the ingredient list.

Consumers seeking to identify stevia on a label should look for: steviol glycosides, stevia, stevia extract, stevia leaf extract, rebaudioside A (or B, C, D, M), Reb A, E960, E960a, E960b, E960c.

Steviol glycosides occur naturally and almost exclusively in significant quantities in the leaves of Stevia rebaudiana Bertoni. Other species within the genus Stevia contain related compounds at low concentrations, but none are commercially relevant. Trace levels of some structurally related diterpene glycosides have been identified in a small number of other plant species, but none constitute a meaningful dietary source outside of S. rebaudiana leaves. There are no commonly consumed foods that naturally contain steviol glycosides at appreciable levels other than products derived from stevia leaf itself. Fresh or dried stevia leaves, available in some markets and home gardens, naturally contain stevioside (typically 5–10% dry weight) and rebaudioside A (typically 2–4% dry weight), along with smaller amounts of rebaudioside B, C, D, F, and other minor glycosides.