How Herbs Lower Blood Sugar

Research describes five ways a plant can lower blood sugar: slowing the digestion of starch, slowing absorption with a gel of fibre, raising insulin release, improving the body’s response to insulin, and reducing sugar uptake or the taste for sweetness. This page is a map of the whole subject. It reports what pooled analyses of randomized controlled trials found for each herb, in the reviews’ own numbers, with each review’s own judgement of how far its result can be trusted.

Evidence base: meta-analyses and systematic reviews of randomized trials in people, mostly adults with type 2 diabetes. The pooled falls are modest, most trials are short and small, reviews of the same herb often disagree, and several reviewers rate their own evidence as low quality. What the reviews pooled are blood-test results, not heart, kidney, eye or nerve outcomes. Two reviews include trials that set a plant preparation against a diabetes drug; both judged that evidence limited or insufficient.


Table of Contents

  1. The Five Ways a Plant Can Lower Blood Sugar
  2. What the Pooled Trials Found, Herb by Herb
  3. How Herbs Compare With Each Other and With Medication
  4. Herbs With Little or No Human Evidence
  5. Safety Findings That Apply Across Herbs
  6. Limits of the Evidence
  7. Where to Read More on This Site
  8. Key Research Papers
  9. Connections
  10. Featured Videos

The Five Ways a Plant Can Lower Blood Sugar

After a meal, starch and sugars are broken down in the small intestine into glucose, which passes into the blood. The pancreas answers by releasing insulin, the hormone that moves glucose out of the blood into muscle, fat and liver cells. A plant compound can act at each step. Most of what is known about which plant acts where comes from laboratory tests on enzymes and cells and from animal studies; a mechanism shown in a test tube is not a result in people. The trial results are in the next section.

Five points where a plant can act on blood sugar WHERE A PLANT CAN ACT ON BLOOD SUGAR Five mechanisms described for herbs and plant fibres insulin PANCREAS MOUTH SMALL INTESTINE BLOOD GLUCOSE MUSCLE, FAT AND LIVER 3 5 1 2 5 4 1 2 3 4 5 Blocks the enzymes that split starch into glucose Fibre gel slows the passage of glucose into the blood The pancreas releases more insulin Cells respond better to insulin Sweet taste is dulled, or less sugar crosses the gut wall

1. Slowing Carbohydrate Digestion by Blocking Alpha-Glucosidase Enzymes

Alpha-glucosidases are enzymes on the lining of the small intestine. They carry out the last step of starch digestion, cutting short chains of sugar into single glucose units. When they are blocked, glucose enters the blood more slowly and the rise after a meal is lower. The prescription drug acarbose works this way. Compounds that block the same enzymes in laboratory tests are reported for many plants; two plants studied for this action, salacia and mulberry leaf, have their own pages.

2. Slowing Absorption With Viscous Fibre

Some plant fibres take up water and form a thick gel. In the gut, the gel slows the mixing of food with digestive enzymes and the movement of glucose toward the gut wall, so glucose reaches the blood over a longer time. Psyllium husk is the best-known gel-forming fibre; fenugreek seed is also rich in soluble fibre.

3. Raising Insulin Release

Some plant compounds act on the beta cells of the pancreas, which make insulin. Berberine, a yellow compound found in barberry and goldthread, is the example in the reviews used here. A 2022 meta-analysis states as background that berberine blocks a potassium channel on the beta cell and so promotes insulin release, and that it does this only when blood glucose is high. The same abstract notes that sulfonylurea drugs, which also raise insulin release, have low blood sugar as a side effect (PMID 36467075).

4. Improving the Body’s Response to Insulin

In type 2 diabetes, cells respond weakly to insulin; this is insulin resistance. Trials estimate it with the homeostatic model assessment of insulin resistance (HOMA-IR), a figure calculated from fasting glucose and fasting insulin. Pooled results differ from herb to herb. A 2023 meta-analysis of turmeric or curcumin reported an improvement in HOMA-IR (PMID 37748368), while the meta-analyses of black seed, ginger, ginseng and green tea described in the next section found no significant change in it. A 2025 review of reviews, in contrast, listed ginger and black seed among the herbs that improved measures of insulin resistance (PMID 41029669), so the reviews do not agree.

5. Reducing Sugar Uptake or Sweet Taste

Glucose crosses the gut wall on transport proteins, and some plant compounds are reported in laboratory studies to slow that transport. Gymnema leaf has a different effect: its gymnemic acids dull the taste of sweetness for a short time after the leaf touches the tongue. None of the reviews used for this page pooled trials of sweet-taste suppression.

Back to Table of Contents

What the Pooled Trials Found, Herb by Herb

How to Read the Numbers

A meta-analysis pools several trials into one average; an umbrella review summarizes earlier meta-analyses. Fasting glucose is reported in milligrams per decilitre (mg/dL) or millimoles per litre (mmol/L); 18 mg/dL equals 1 mmol/L, and each number below is in the unit its review used. Hemoglobin A1C (HbA1c) reflects average blood sugar over about three months, in percent. A mean difference is the average gap between the herb group and the comparison group; a minus sign means the herb group was lower. The 95% confidence interval (CI) is the range in which the true value probably lies; if it includes zero, the result is not statistically significant. Heterogeneity (I²) shows how much the pooled trials disagree with one another, from 0% to 100%.

The herbs are not ranked by this site. They are grouped by what the reviewers themselves wrote about the certainty of their result: herbs a review graded moderate to high; herbs with pooled falls whose reviewers asked for caution over trial quality; herbs with a fall in fasting glucose but no clear fall in HbA1c, or with reviews that disagree; herbs whose reviewers called the trials or the evidence low quality or insufficient; and a herb whose pooled result was no effect. Within a group the order is alphabetical. The grouping is not a judgement about which herb works best.

Herbs a Review Graded as Moderate to High Certainty

Ginger. A 2025 umbrella review covered eight meta-analyses and two systematic reviews published between August 2018 and August 2023. It reported that ginger lowered HbA1c compared with placebo by a mean difference of −0.467% and graded the evidence for ginger and turmeric as moderate to high certainty. The same authors wrote that, because of high clinical heterogeneity and the low quality of the included reviews, “our confidence in this finding is somewhat limited” (PMID 41029669).

Two meta-analyses of ginger trials in type 2 diabetes reached different results. A 2024 meta-analysis of five randomized trials (ginger powder capsules, 1.2 to 2 g a day, for 4 to 12 weeks) found no significant effect on fasting glucose or HbA1c when the trials were pooled, although two of the trials on their own showed a significant fall in fasting glucose (PMID 39053695). A 2026 meta-analysis of 13 articles comparing ginger with placebo found a mean difference of −16.27 mg/dL in fasting glucose (95% CI −25.75 to −6.80) and −0.41% in HbA1c (95% CI −0.63 to −0.20), with high heterogeneity (I² 86.39% and 92.09%); HOMA-IR did not change significantly. Its authors wrote that “the magnitude of these effects was of limited clinical importance” (PMID 42337792).

Turmeric and curcumin. The 2025 umbrella review reported an HbA1c mean difference of −0.486% for turmeric against placebo, under the same grading and the same stated limits (PMID 41029669). A 2023 meta-analysis pooled 59 randomized trials in adults, not only people with diabetes, and reported that turmeric or curcumin improved fasting blood sugar by a weighted mean difference of 4.60 mg/dL and HbA1c by 0.32%. Its title describes it as grade-assessed; its abstract does not state the grade (PMID 37748368). A 2024 meta-analysis of curcumin taken with piperine, a compound from black pepper, found three randomized trials in people with prediabetes or type 2 diabetes. The pooled fall in fasting glucose (mean difference −7.61, 95% CI −15.26 to 0.03, p = 0.05) was not statistically significant (PMID 38863920).

Pooled Falls in Fasting Glucose and Hemoglobin A1C, With Caution Over Trial Quality

Berberine. A 2022 meta-analysis pooled 37 randomized trials with 3,048 patients with type 2 diabetes who took berberine alone or with diabetes tablets. The weighted mean difference was −0.82 mmol/L for fasting glucose (95% CI −0.95 to −0.70), −0.63% for HbA1c (95% CI −0.72 to −0.53) and −1.16 mmol/L for two-hour plasma glucose (95% CI −1.36 to −0.96). The size of the fall was related to how high fasting glucose and HbA1c were at the start (PMID 36467075).

A 2015 meta-analysis of 27 randomized trials with 2,569 patients found that berberine with lifestyle measures tended to lower fasting glucose, post-meal glucose and HbA1c more than lifestyle measures alone or placebo, and that berberine added to diabetes tablets did so more than the same tablets alone. Its authors wrote that, because of the overall limited quality of the included studies, the benefit “can be substantiated to a limited degree” (PMID 25498346).

Black seed (Nigella sativa). A 2026 meta-analysis, described in its title as grade-assessed, pooled 31 trials with 2,145 participants who had metabolic diseases, type 2 diabetes and obesity among them. The weighted mean difference was −18.62 mg/dL for fasting blood sugar (95% CI −23.30 to −13.95) and −0.56 for HbA1c (95% CI −0.79 to −0.33). Black seed had no effect on insulin or HOMA-IR. The authors called the improvements modest and wrote that the findings are to be “interpreted cautiously given variability in study quality and outcomes” (PMID 41858302).

A Fall in Fasting Glucose Without a Clear Fall in Hemoglobin A1C, or Reviews That Disagree

Cinnamon. Reviews of cinnamon disagree. A 2023 umbrella meta-analysis, which opens by noting that earlier meta-analyses had reported conflicting findings, combined 11 meta-analyses of randomized trials in people with type 2 diabetes or polycystic ovary syndrome. It reported a weighted mean difference of −10.93 mg/dL for fasting plasma glucose (95% CI −16.22 to −5.65) and −0.10% for HbA1c (95% CI −0.17 to −0.03) (PMID 37316893). A dose-response meta-analysis of randomized trials in type 2 diabetes, published the same year, found no significant effect of cinnamon on fasting glucose or HbA1c overall; a fall in glucose appeared only in the subgroup of trials that used 1,200 mg or less (PMID 37447309). A 2011 meta-analysis limited to placebo-controlled trials of single herbs that lasted at least eight weeks and reported HbA1c found that cassia cinnamon did not improve glycemic control (PMID 21843614).

Ginseng. A 2014 meta-analysis pooled 16 randomized trials of at least 30 days of ginseng (the genus Panax) in people with and without diabetes. Fasting glucose fell by a mean difference of −0.31 mmol/L (95% CI −0.59 to −0.03), from comparisons with 770 participants. There was no significant overall effect on HbA1c, fasting insulin or HOMA-IR. Most trials were short (67% lasted under 12 weeks) and enrolled people whose blood sugar was already fairly well controlled. The authors described the effect as modest and wrote that larger, longer trials with standardized preparations are needed to settle the uncertainty (PMID 25265315). See American Ginseng for Blood Sugar.

Trials or Evidence the Reviewers Called Low Quality or Insufficient

Bitter melon (Momordica charantia). Reviews of bitter melon disagree. A 2012 Cochrane systematic review found four randomized trials, none longer than three months, with 479 participants and a risk of bias that was “generally high”. It did not pool them, because no preparation had been tested twice. In the two trials that used a placebo there was no statistically significant difference in glycemic control. The review concluded that there is insufficient evidence on the effects of bitter melon in type 2 diabetes (PMID 22895968).

A 2019 meta-analysis pooled 10 studies with 1,045 people with type 2 diabetes, followed for 4 to 16 weeks, at an overall moderate to high risk of bias. Compared with placebo, single-herb bitter melon preparations gave a mean difference of −0.72 mmol/L in fasting glucose (95% CI −1.33 to −0.12) and −0.26% in HbA1c (95% CI −0.49 to −0.03). The authors wrote that this conclusion rests on “low to very low quality” evidence and sparse safety data (PMID 30385422).

Fenugreek. A 2014 meta-analysis pooled 10 clinical trials of fenugreek seed. Fasting glucose changed by −0.96 mmol/L (95% CI −1.52 to −0.40; 10 trials; I² 80%), glucose two hours after a glucose drink by −2.19 mmol/L (7 trials) and HbA1c by −0.85% (95% CI −1.49% to −0.22%; 3 trials). Significant effects on fasting and two-hour glucose were found only in trials that gave medium or high doses to people with diabetes. The authors wrote that “most of the trials were of low methodological quality” (PMID 24438170). The 2011 meta-analysis of single herbs reported an HbA1c mean difference of −1.13% for fenugreek (95% CI −0.11% to −2.14%), with high heterogeneity (PMID 21843614).

Gymnema. A 2023 meta-analysis pooled six randomized trials in adults and found a significant fall in fasting blood sugar (p < .001); its abstract does not give the size of the fall. The authors listed “notable heterogeneity, low quality of studies, and lack of diversity among research participants” as limits on the result (PMID 36580574).

Milk thistle and sweet potato. The 2011 meta-analysis of single herbs covered nine randomized, placebo-controlled trials with 487 patients in all. The HbA1c mean difference was −1.92% for milk thistle (95% CI −0.51% to −3.32%) and −0.30% for a sweet potato (Ipomoea batatas) preparation (95% CI −0.04% to −0.57%). Fasting glucose differed by −38.05 mg/dL and −10.20 mg/dL respectively. The authors pointed to the limitations of the available studies and to high heterogeneity in the milk thistle and fenugreek results (PMID 21843614).

A Pooled Result of No Effect

Green tea. A 2021 meta-analysis of 14 articles reporting randomized trials in people with type 2 diabetes found that green tea had no significant effect on fasting plasma glucose, fasting insulin, HbA1c or HOMA-IR (PMID 33285391).

Back to Table of Contents

How Herbs Compare With Each Other and With Medication

Herbs Compared With Each Other

None of the reviews used here rests on trials that tested one herb directly against another. Pooled numbers from separate reviews come from different patients, doses and trial lengths, so setting one review’s figure beside another’s does not show which herb is stronger. Three reviews looked across herbs with a single method.

Herbs Compared With Medication

Most pooled trials tested a herb against placebo or added a herb to the medication people were already taking. Two reviews include trials that set a plant preparation against a diabetes drug.

The 2025 umbrella review concluded that herbal medicines belong beside conventional treatment, as an adjunct, and not in place of it (PMID 41029669). The 2019 bitter melon meta-analysis likewise described the preparations it pooled as adjuncts (PMID 30385422).

Back to Table of Contents

Herbs With Little or No Human Evidence

For the herbs below, the reviews gathered for this page give no pooled figure. That is not a finding that a herb does nothing. It means there are too few comparable trials in people for a review to average, or none at all. Each herb’s own page reports the individual trials, animal studies and safety reports that exist.

Named in the 2003 systematic review, with no pooled analysis: ivy gourd and nopal (prickly pear cactus).

No pooled analysis in the reviews gathered for this page: banaba, coriander seeds, holy basil, salacia, stevia, mulberry leaf, neem, and three herbs whose pages carry findings a reader of this section needs:

Back to Table of Contents

Safety Findings That Apply Across Herbs

Low Blood Sugar When Herbs Are Combined With Glucose-Lowering Drugs

Herbs that lower blood sugar add to the effect of insulin and sulfonylurea drugs; trials and case reports describe low blood sugar (hypoglycemia) when they are combined. The signs of low blood sugar are shakiness, sweating, a fast heartbeat and confusion, and at lower levels seizure or loss of consciousness. See Hypoglycemia Awareness and Prevention.

A hospital computer system in Taiwan flagged drug–herb interactions in herbal prescriptions from 2011 to 2013. Of the 50 patients who took ginseng and antidiabetic drugs at the same time, three had hypoglycemia, defined as a fasting blood sugar of 70 mg/dL or lower (PMID 25733840). This is a review of hospital records, not a trial. For berberine, the 2022 meta-analysis found that hypoglycemia was not significantly more frequent with berberine, alone or with diabetes tablets, than in the comparison groups (relative risk 0.48, 95% CI 0.21 to 1.08, p = 0.08) (PMID 36467075).

Herb and Drug Interactions

A herb can also change how the body absorbs, breaks down or clears a drug. The reviews used here hold almost no human data on this. The 2025 umbrella review states that there is “a current lack of data on herb-drug interactions”, and for that reason its authors described careful monitoring for hypoglycemia as essential when herbal medicines are taken together with antidiabetic drugs (PMID 41029669). Interaction studies of single herbs are reported on their own pages: the Berberine page covers its effect on the liver enzymes that process many drugs, and the American Ginseng page covers a randomized trial of its effect on the blood thinner warfarin.

Products Found to Contain Undeclared Drugs

Some products sold as herbal remedies for diabetes have been found to contain prescription drugs that the label does not declare. A toxicology referral centre in Hong Kong reviewed every such case sent to it from 2005 to 2010: 27 cases involving 29 adulterated herbal antidiabetic products. Seventeen of the patients (63%) had toxic effects; hypoglycemia was the most common, followed by lactic acidosis, a dangerous build-up of acid in the blood.

Laboratory analysis found eight undeclared antidiabetic drugs, registered and banned. Glibenclamide (a sulfonylurea) was in 22 of the products, phenformin in 18, metformin in 6 and rosiglitazone in 6, and up to four drugs were found in a single product. The authors wrote that people taking these products “could be at risk of potentially fatal adverse effects” (PMID 22034838). A hidden sulfonylurea lowers blood sugar strongly, so such a product can appear to work as a herb while acting as a drug of unknown dose.

Other Safety Facts

Back to Table of Contents

Limits of the Evidence

Back to Table of Contents

Where to Read More on This Site

Herbs and Plant Compounds

Herb Pages Elsewhere on the Site That Cover Blood Sugar

Minerals and Nutrients

Food and Practice

Back to Table of Contents

Key Research Papers

  1. Xie W, Su F, Wang G, Peng Z, Xu Y, Zhang Y, et al. Glucose-lowering effect of berberine on type 2 diabetes: A systematic review and meta-analysis. Front Pharmacol. 2022;13:1015045. PubMed PMID: 36467075
  2. Lan J, Zhao Y, Dong F, Yan Z, Zheng W, Fan J, et al. Meta-analysis of the effect and safety of berberine in the treatment of type 2 diabetes mellitus, hyperlipemia and hypertension. J Ethnopharmacol. 2015;161:69-81. PubMed PMID: 25498346
  3. Li E, Maunder A, Liu J, Pandey C, Cave A, O'Fee A, et al. The efficacy and safety of herbal medicines for glycaemic control and insulin resistance in individuals with type 2 diabetes: an umbrella review. BMC Complement Med Ther. 2025;25(1):341. PubMed PMID: 41029669
  4. Schumacher JC, Mueller V, Sousa C, Peres KK, da Mata IR, Menezes RCR, et al. The effect of oral supplementation of ginger on glycemic control of patients with type 2 diabetes mellitus - A systematic review and meta-analysis. Clin Nutr ESPEN. 2024;63:615-622. PubMed PMID: 39053695
  5. Chen X, Cheng S, Zhang D. Meta-analysis of the effects of ginger supplementation on glycemic control, blood pressure and lipid profile in patients with type 2 diabetes. Syst Rev. 2026;15(1):231. PubMed PMID: 42337792
  6. Dehzad MJ, Ghalandari H, Nouri M, Askarpour M. Effects of curcumin/turmeric supplementation on glycemic indices in adults: A grade-assessed systematic review and dose-response meta-analysis of randomized controlled trials. Diabetes Metab Syndr. 2023;17(10):102855. PubMed PMID: 37748368
  7. Widjanarko ND, Tamio E, Jusni LFJ, Alvianto S, Arifin ES, Iryaningrum MR. Effects of Combination of Curcumin and Piperine Supplementation on Glycemic Profile in Patients with Prediabetes and Type 2 Diabetes Mellitus: A Systematic Review and Meta-analysis. J ASEAN Fed Endocr Soc. 2024;39(1):106-114. PubMed PMID: 38863920
  8. Musazadeh V, Jahangir M, Mahmoudinezhad M, Faghfouri AH, Teymoori F, Anoushirvani AA, et al. The Effect of Nigella sativa Supplementation on Cardiometabolic Health in Patients With Metabolic Diseases: A GRADE-Assessed Systematic Review and Meta-Analysis. Endocrinol Diabetes Metab. 2026;9(2):e70207. PubMed PMID: 41858302
  9. Zarezadeh M, Musazadeh V, Foroumandi E, Keramati M, Ostadrahimi A, Mekary RA. The effect of cinnamon supplementation on glycemic control in patients with type 2 diabetes or with polycystic ovary syndrome: an umbrella meta-analysis on interventional meta-analyses. Diabetol Metab Syndr. 2023;15(1):127. PubMed PMID: 37316893
  10. Yu T, Lu K, Cao X, Xia H, Wang S, Sun G, et al. The Effect of Cinnamon on Glycolipid Metabolism: A Dose-Response Meta-Analysis of Randomized Controlled Trials. Nutrients. 2023;15(13):2983. PubMed PMID: 37447309
  11. Suksomboon N, Poolsup N, Boonkaew S, Suthisisang CC. Meta-analysis of the effect of herbal supplement on glycemic control in type 2 diabetes. J Ethnopharmacol. 2011;137(3):1328-33. PubMed PMID: 21843614
  12. Shishtar E, Sievenpiper JL, Djedovic V, Cozma AI, Ha V, Jayalath VH, et al. The effect of ginseng (the genus panax) on glycemic control: a systematic review and meta-analysis of randomized controlled clinical trials. PLoS One. 2014;9(9):e107391. PubMed PMID: 25265315
  13. Ooi CP, Yassin Z, Hamid TA. Momordica charantia for type 2 diabetes mellitus. Cochrane Database Syst Rev. 2012;2012(8):CD007845. PubMed PMID: 22895968
  14. Peter EL, Kasali FM, Deyno S, Mtewa A, Nagendrappa PB, Tolo CU, et al. Momordica charantia L. lowers elevated glycaemia in type 2 diabetes mellitus patients: Systematic review and meta-analysis. J Ethnopharmacol. 2019;231:311-324. PubMed PMID: 30385422
  15. Neelakantan N, Narayanan M, de Souza RJ, van Dam RM. Effect of fenugreek (Trigonella foenum-graecum L.) intake on glycemia: a meta-analysis of clinical trials. Nutr J. 2014;13:7. PubMed PMID: 24438170
  16. Zamani M, Ashtary-Larky D, Nosratabadi S, Bagheri R, Wong A, Rafiei MM, et al. The effects of Gymnema Sylvestre supplementation on lipid profile, glycemic control, blood pressure, and anthropometric indices in adults: A systematic review and meta-analysis. Phytother Res. 2023;37(3):949-964. PubMed PMID: 36580574
  17. Asbaghi O, Fouladvand F, Gonzalez MJ, Ashtary-Larky D, Choghakhori R, Abbasnezhad A. Effect of green tea on glycemic control in patients with type 2 diabetes mellitus: A systematic review and meta-analysis. Diabetes Metab Syndr. 2021;15(1):23-31. PubMed PMID: 33285391
  18. Yeh GY, Eisenberg DM, Kaptchuk TJ, Phillips RS. Systematic review of herbs and dietary supplements for glycemic control in diabetes. Diabetes Care. 2003;26(4):1277-94. PubMed PMID: 12663610
  19. Lin SS, Tsai CL, Tu CY, Hsieh CL. Reducing drug-herb interaction risk with a computerized reminder system. Ther Clin Risk Manag. 2015;11:247-53. PubMed PMID: 25733840
  20. Ching CK, Lam YH, Chan AY, Mak TW. Adulteration of herbal antidiabetic products with undeclared pharmaceuticals: a case series in Hong Kong. Br J Clin Pharmacol. 2012;73(5):795-800. PubMed PMID: 22034838

PubMed Topic Searches

  1. PubMed: systematic reviews of herbal medicine for glycemic control in type 2 diabetes
  2. PubMed: medicinal plants as alpha-glucosidase inhibitors
  3. PubMed: herb and drug interactions with antidiabetic drugs
  4. PubMed: herbal antidiabetic products adulterated with undeclared drugs

Back to Table of Contents

Connections

Back to Table of Contents