By SciRouter Editorial Team · Last reviewed: 9 October 2026
In small, short trials from China, berberine lowered blood sugar in adults with type 2 diabetes by amounts the researchers called similar to metformin[1, 2]. Pooled analyses found no statistically significant difference but rated the trials low quality[3]. Metformin also has a decade-long health-outcome trial; the berberine trials lasted weeks to months[3, 4]. Berberine is not a substitute for prescribed metformin.
Important: Don't stop or change a prescribed medication without talking to your prescriber.
Berberine is a plant compound found in herbs such as goldenseal and sold in supplements[5, 6]. Metformin is an oral diabetes drug[3]. If you take metformin, talk to your prescriber before adding berberine, because berberine can change how the body handles other medicines[7]. For the wider picture, see our evidence guide to berberine.
How we researched this. We searched PubMed through NCBI E-utilities, read Europe PMC full text where open, and read agency pages from NIH (NCCIH, LactMed) and France's food-safety agency, ANSES. Searches ran on 8 October 2026. Human head-to-head trials and meta-analyses came first; animal and cell studies are labelled by species. We cite only sources we opened and checked against PubMed or the agency page. This page cites 30 sources. SciRouter sells no berberine product. This page is not medical advice.
How did berberine compare with metformin in head-to-head trials?
In two small randomized trials (chance decided who got which option) of adults with type 2 diabetes, berberine and metformin lowered blood sugar by amounts the researchers called similar over 8 to 13 weeks[1–3]. A few trials in women with PCOS also compared them, with mixed and less reliable results[8–11].
In the table, HbA1c is a blood test that reflects average blood sugar over the past few months.
| Study | Who, how many, how long | Daily doses | Results and quality notes |
|---|---|---|---|
| Yin 2008, study A[1] | Adults newly diagnosed with type 2 diabetes, Shanghai; 36 randomized, 31 analyzed; 13 weeks | Berberine 0.5 g vs metformin 0.5 g, each three times daily | HbA1c: berberine 9.47% to 7.48%, metformin 9.15% to 7.72%. Fasting glucose: berberine 10.63 to 6.85 mmol/L, metformin 9.96 to 7.16 mmol/L. Triglycerides and total cholesterol were lower with berberine than metformin at week 13. Quality: pilot at one centre; blinding not described; only completers analyzed. |
| Zhang 2010[2, 3] | Adults with type 2 diabetes, China; 50 berberine, 26 metformin, 21 rosiglitazone; 8 weeks | Not given in the abstract | Drops in fasting glucose and HbA1c were reported as similar to metformin and rosiglitazone; no numbers in the abstract. Quality: abstract only; group sizes and length from a 2012 meta-analysis. |
| Wei 2012[8] | Women with PCOS and insulin resistance, China, all also on a hormonal drug (cyproterone acetate); 89 (31 berberine, 30 metformin, 28 placebo); 3 months | Not given in the abstract | Versus metformin, larger drops in waist-to-hip ratio, total cholesterol, triglycerides and LDL cholesterol. Glucose and insulin results were reported against placebo only. Quality: abstract only; small. |
| An 2014[9] | Women with PCOS preparing for IVF, China; 150 across berberine, metformin and placebo groups; 3 months | Not given in the abstract | Both groups had larger drops in fasting glucose, fasting insulin and insulin resistance than placebo. Fewer digestive side effects with berberine than metformin. Quality: abstract only. |
| Mishra 2022[10] | Women with PCOS, India; 129 analyzed (a third group took myo-inositol); 3 months | Berberine 500 mg vs metformin 500 mg, each twice daily | No significant difference in body weight or BMI. Quality: open-label (not blinded); no registration reported; abstract's lipid claims conflict with its own tables. |
In the Yin trial, the headline HbA1c drop, from 9.5% to 7.5%, was a before-and-after change within the berberine group[1]. For blood sugar, the paper compared each group with its own starting point and called the changes similar[1]. The same paper's study B added berberine to existing diabetes drugs in 48 adults with no comparison group, so its HbA1c fall from 8.1% to 7.3% cannot be credited to berberine[1].
What did pooled analyses of the trials find?
Meta-analyses, which combine results from many trials, found no statistically significant difference in blood-sugar measures between berberine and oral diabetes drugs, including metformin[3, 12, 13]. The same reviews rated most of the trials as low or limited quality[3, 12].
| Analysis and size | Comparison | Result | Caveats |
|---|---|---|---|
| Dong 2012[3]: 7 trials, 448 adults with type 2 diabetes | Berberine vs metformin, glipizide or rosiglitazone | HbA1c difference −0.11 points (95% CI −0.32 to 0.09); fasting glucose 0.20 mmol/L (95% CI −0.11 to 0.51); neither significant | Low quality; all from China; no metformin-only estimate |
| Lan 2015[12]: 27 trials, 2,569 patients with type 2 diabetes, high blood lipids or high blood pressure | Berberine vs oral glucose-lowering drugs (a subset of the trials) | No significant difference in blood-sugar measures | Limited quality; no pooled numbers in the abstract |
| Liang 2019[13]: 28 trials, 2,313 adults with type 2 diabetes; 7 to 12 comparisons against drugs | Berberine vs oral glucose-lowering drugs | Fasting glucose, post-meal glucose and HbA1c differences not significant | All from China; metformin not analysed alone |
| Li 2018[14]: 9 trials, women with PCOS and insulin resistance | Berberine vs metformin, and the two combined vs metformin | No significant difference; adding berberine was not better than metformin alone | Authors said data were too limited for conclusions |
| Xie 2019[15]: 12 trials in women with PCOS, 3 to 4 vs metformin | Berberine vs metformin | No BMI difference (4 trials, 262 women); lower total cholesterol and waist size with berberine | Few small trials per outcome |
| Perera 2026[11]: 4 eligible trials, 452 women with PCOS | Integrity check | All four at high risk of compromised integrity; none pooled | No registration, implausible recruitment timelines, possible duplicated data |
The 2026 integrity review concluded that reliable conclusions about berberine in PCOS cannot be drawn from current trials, and its abstract does not name which trials it flagged[11].
Does "no significant difference" mean berberine equals metformin?
No. These trials were small, short and mostly low quality, so they could not show that the two are equal[3].
- Few people. The Yin trial analyzed 31 people[1]. In a trial that small, a real difference can go undetected.
- Short follow-up. Head-to-head trials lasted 8 to 13 weeks[3]. In a 2019 meta-analysis of 28 trials, only 1 to 3 comparisons per blood-sugar measure lasted longer than 90 days[13].
- Weak design. Most trials had unclear randomization or blinding, and most analyzed only people who finished[3].
- One country. All trials in two of the meta-analyses came from China[3, 13].
- Lab numbers only. The trials tracked blood tests such as HbA1c, not long-term health events[3].
Do berberine and metformin work the same way?
Partly, in lab studies. In muscle cells the two share one effect on cell energy, but berberine also has other proposed actions, and none of this shows they match in people[2, 16, 17].
AMPK (AMP-activated protein kinase) is an energy sensor inside cells that helps control how they use glucose and fat[17]. In cultured fat and muscle cells, berberine switched AMPK on, and in insulin-resistant mice and rats it improved glucose handling, which the authors linked partly to AMPK[17]. In rat muscle cells and isolated mitochondria, the cell's power plants, berberine blocked part of the energy chain called complex I[16]. The researchers said this was similar to an effect seen with metformin, and proposed it as a way berberine switches on AMPK[16].
Berberine also has other proposed routes:
- A Chinese study reported that berberine raised insulin-receptor levels in human cells, a route the authors called different from metformin's[2].
- In mice, berberine raised insulin release through a potassium channel called KCNH6[18]. In 15 healthy men given one dose, it raised insulin release when blood sugar was raised on purpose, but not at baseline[18].
Can you take berberine and metformin together?
That is a decision for your prescriber, because the blood-sugar effects may add up and berberine can alter how the body handles some medicines[3, 7]. The human data we found on metformin levels come from goldenseal, a berberine-containing herb, not from berberine supplements[19, 20].
Blood-sugar effects can add up
In a 2012 meta-analysis, adding berberine to an oral diabetes drug lowered HbA1c by 0.53 percentage points more than the drug alone, across 6 trials of 8 to 24 weeks[3]. A 2015 meta-analysis found the same pattern, while noting limited trial quality[12]. In women with PCOS, though, adding berberine to metformin was not better than metformin alone[14].
The largest add-on trial we found tested a drug, not a supplement. HTD1801 is berberine ursodeoxycholate, an investigational drug that pairs berberine with ursodeoxycholic acid[21]. In a 24-week double-blind trial of 549 adults in China with type 2 diabetes already taking metformin, HTD1801 1,000 mg twice daily lowered HbA1c by 0.5 percentage points more than placebo[22]. Diarrhea affected 23.8% of the HTD1801 group versus 1.1% on placebo, and the drug's developer funded the trial[22]. Its results should not be read as results for berberine supplements.
Low blood sugar
People who take blood-sugar-lowering drugs risk low blood sugar if berberine's effect adds to theirs, so ask your prescriber how to watch for it[3, 6]. In UKPDS 34, a long-term UK trial in overweight adults with type 2 diabetes, insulin and sulfonylurea pills such as glibenclamide were associated with more low-blood-sugar attacks than metformin[4].
ANSES lists low blood sugar among the risks of berberine supplements[6]. In a 2022 meta-analysis of 37 trials, berberine alone or with oral diabetes drugs did not significantly raise the risk of low blood sugar (relative risk 0.48, where 1 means no difference; 95% confidence interval 0.21 to 1.08)[23]. That range is wide, so a higher risk cannot be ruled out[23].
Metformin levels: the goldenseal data
In 16 healthy adults, a goldenseal product standardized to berberine lowered metformin exposure by about 23%, which the researchers said may have unfavorable effects on glucose control[19]. In 22 adults with type 2 diabetes on metformin, goldenseal caused no clinically meaningful change on average, but metformin levels fell about 20% at low metformin doses[20]. That goldenseal dose held only about 95 mg of berberine a day[20]. In cell and rat studies, berberine blocked OCT1 and OCT2, transporters that move metformin into cells, and raised metformin levels in rats[24]. That is the opposite direction to the human data, so the interaction is poorly understood[19, 24].
Digestive side effects
Across the 2008 paper's two pilot studies, 34.5% of people taking berberine had temporary digestive side effects, most often gas, diarrhea and constipation[1]. The authors said these were often seen when berberine was combined with metformin or acarbose[1]. In a 2023 review of 18 placebo-controlled trials, the 12 that reported digestive side effects found them in 2% to 23% of people on berberine versus 2% to 15% on placebo[25].
If you also take a GLP-1 drug, our page on how berberine compares with GLP-1 drugs covers what is and is not known.
What is missing from the berberine evidence?
Long-term data and consistent products are the main gaps. Trials in two meta-analyses lasted 4 to 24 weeks[3, 25], and reviewers have asked for better reporting of what the products contained[12, 26].
- Health outcomes. The authors of a 2023 meta-analysis called berberine's blood-lipid effects small and asked for large trials of clinical outcomes[25].
- Product content. Meta-analysts have asked trials to report product purity and potency, and to use standardized preparations[12, 26]. When a US lab tested 15 berberine supplements, the berberine it found ranged from 33% to 100% of the label amount, averaging 75%[27]. Our checklist for choosing a berberine supplement covers label claims and third-party testing.
- Regulation and safety review. ANSES concluded in 2019 that the safety of berberine supplements cannot currently be guaranteed, and reported that Belgium caps berberine in supplements at 10 mg a day while France requires a pregnancy warning on labels[6].
Metformin's evidence is different in kind: in the UK's UKPDS 34 trial, 753 overweight adults with newly diagnosed type 2 diabetes were randomly assigned to metformin or to a mainly diet-based approach and followed for a median of 10.7 years[4]. The metformin group had a 32% lower risk of any diabetes-related clinical endpoint and a 36% lower risk of death from any cause[4]. In a smaller trial in the same paper, adding metformin early for people already on maximum sulfonylurea therapy was associated with more diabetes-related deaths, though an observational analysis of 4,416 UKPDS patients did not show that increase[4]. Your prescriber can explain what this evidence means for you.
What other safety cautions apply?
Berberine is not recommended in pregnancy or breastfeeding, and because it slowed drug-clearing liver enzymes in a human study, anyone on other prescription medicines should check with a prescriber or pharmacist first[5–7]. People on glucose-lowering drugs also risk low blood sugar if berberine's effect adds to theirs, and France's food-safety agency advised people with diabetes to avoid berberine supplements[3, 6].
In a 3-month trial of 104 kidney-transplant recipients, cyclosporine levels ended about 29% higher with berberine than without it[28]. A 2026 case report described torsades de pointes, a dangerous heart rhythm, in a 63-year-old woman after several months of berberine, though one case cannot prove cause[29, 30]. Our page on who should not take berberine has the details and the full interaction table.
Questions to ask your prescriber
Take these questions to your prescriber or pharmacist before deciding whether to add berberine to metformin. Bring a list of everything you take, including supplements.
- Is berberine reasonable for me, given my blood sugar results and my other medicines?
- Could berberine plus my diabetes medicines push my blood sugar too low? Which warning signs should I watch for?
- Would you want to check my HbA1c sooner if I start or stop berberine?
- Do any of my medicines depend on CYP3A4, CYP2D6 or CYP2C9, or need blood-level checks like cyclosporine[7, 28]?
- What should I do if I get diarrhea or other stomach upset?
- Is berberine off-limits if I am pregnant, trying to conceive or breastfeeding[6]?
For the evidence on glucose, cholesterol, weight and safety in one place, read our berberine evidence scorecard.
Get our plain-English longevity research digest
Short, sourced summaries of new longevity research, written in plain English. No product pitches. Unsubscribe anytime.
Frequently asked questions
Is it safe to take berberine instead of metformin?
Berberine has not been shown to be a safe replacement for prescribed metformin. The head-to-head evidence in diabetes is a handful of small trials lasting 8 to 13 weeks, all from China, which a 2012 meta-analysis rated as generally low quality[3]. The best-known trial analyzed only 31 people[1]. Don't stop or change a prescribed medication without talking to your prescriber, and ask whether berberine has any place in your plan.
Can I take berberine with metformin?
Whether to take berberine with metformin is your prescriber's call. In a 2012 meta-analysis, adding berberine to an oral diabetes drug lowered HbA1c 0.53 percentage points more than the drug alone over 8 to 24 weeks[3]. People on glucose-lowering drugs risk low blood sugar if berberine's effect adds to theirs[3, 6]. Goldenseal, a berberine-containing herb, lowered metformin levels by about 20% in adults with type 2 diabetes on low metformin doses[20]. Berberine slowed drug-clearing liver enzymes in healthy men[7].
Is berberine or metformin better for blood sugar?
No one can say from current evidence. Pooled analyses found no statistically significant difference in HbA1c or fasting glucose between berberine and oral diabetes drugs, including metformin[3, 13]. But the trials were small, short and mostly rated low quality, and all came from China[3, 13]. In trials that small, a finding of "no significant difference" does not prove that berberine and metformin are equal.
How long did the berberine vs metformin trials last?
The berberine vs metformin trials were short. In adults with type 2 diabetes, the head-to-head trials in a 2012 meta-analysis lasted 8 to 13 weeks[3]. Three trials comparing the two in women with PCOS each lasted 3 months[8–10]. By contrast, the UKPDS 34 trial followed overweight adults with type 2 diabetes assigned to metformin for a median of 10.7 years, tracking diabetes-related health events and deaths[4].
What is the best natural alternative to metformin?
We found no evidence that any supplement can safely replace prescribed metformin. A 2015 meta-analysis of 27 trials in people with type 2 diabetes, high blood lipids or high blood pressure found no significant difference in blood-sugar measures between berberine and oral diabetes drugs, but judged the trials to be of limited quality[12]. Don't stop or change a prescribed medication without talking to your prescriber. If cost or side effects are a problem, raise that with your prescriber.
Does berberine cause the same stomach upset as metformin?
Digestive side effects were reported with both berberine and metformin, and direct comparisons are limited[1, 9]. In a 2008 pilot paper, 34.5% of people taking berberine had temporary digestive side effects, mostly gas, diarrhea or constipation, more often alongside metformin or acarbose[1]. In a 3-month trial of 150 Chinese women with PCOS preparing for IVF, the berberine group reported fewer digestive side effects than the metformin group[9]. A 2026 review questioned the reliability of PCOS berberine trials[11].
How much berberine was used in the metformin trials?
The main head-to-head trial gave berberine 0.5 g three times a day against the same metformin dose for 13 weeks[1]. Across trials in a 2012 meta-analysis, berberine doses were generally 0.5 to 1.5 g a day[3]. These are research doses, not a recommendation.
References
- Yin J, Xing H, Ye J. Efficacy of berberine in patients with type 2 diabetes mellitus. Metabolism. 2008. PMID: 18442638
- Zhang H, Wei J, Xue R, et al. Berberine lowers blood glucose in type 2 diabetes mellitus patients through increasing insulin receptor expression. Metabolism. 2010. PMID: 19800084
- Dong H, Wang N, Zhao L, et al. Berberine in the treatment of type 2 diabetes mellitus: a systemic review and meta-analysis. Evid Based Complement Alternat Med. 2012. PMID: 23118793
- UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998. PMID: 9742977
- National Center for Complementary and Integrative Health (NIH). Goldenseal: Usefulness and Safety. 2025.
- ANSES (French Agency for Food, Environmental and Occupational Health & Safety). Use of berberine-containing plants in food supplements. 2019.
- Guo Y, Chen Y, Tan ZR, et al. Repeated administration of berberine inhibits cytochromes P450 in humans. Eur J Clin Pharmacol. 2012. PMID: 21870106
- Wei W, Zhao H, Wang A, et al. A clinical study on the short-term effect of berberine in comparison to metformin on the metabolic characteristics of women with polycystic ovary syndrome. Eur J Endocrinol. 2012. PMID: 22019891
- An Y, Sun Z, Zhang Y, et al. The use of berberine for women with polycystic ovary syndrome undergoing IVF treatment. Clin Endocrinol (Oxf). 2014. PMID: 23869585
- Mishra N, Verma R, Jadaun P. Study on the Effect of Berberine, Myoinositol, and Metformin in Women with Polycystic Ovary Syndrome: A Prospective Randomised Study. Cureus. 2022. PMID: 35251851
- Perera E, Mantzioris E, Moran L, et al. Questionable Evidence for Berberine in Polyendocrine Metabolic Ovarian Syndrome: An Integrity-Based Reappraisal. Adv Ther. 2026. PMID: 42836929
- Lan J, Zhao Y, Dong F, 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. PMID: 25498346
- Liang Y, Xu X, Yin M, et al. Effects of berberine on blood glucose in patients with type 2 diabetes mellitus: a systematic literature review and a meta-analysis. Endocr J. 2019. PMID: 30393248
- Li MF, Zhou XM, Li XL. The Effect of Berberine on Polycystic Ovary Syndrome Patients with Insulin Resistance (PCOS-IR): A Meta-Analysis and Systematic Review. Evid Based Complement Alternat Med. 2018. PMID: 30538756
- Xie L, Zhang D, Ma H, et al. The Effect of Berberine on Reproduction and Metabolism in Women with Polycystic Ovary Syndrome: A Systematic Review and Meta-Analysis of Randomized Control Trials. Evid Based Complement Alternat Med. 2019. PMID: 31915452
- Turner N, Li JY, Gosby A, et al. Berberine and its more biologically available derivative, dihydroberberine, inhibit mitochondrial respiratory complex I: a mechanism for the action of berberine to activate AMP-activated protein kinase and improve insulin action. Diabetes. 2008. PMID: 18285556
- Lee YS, Kim WS, Kim KH, et al. Berberine, a natural plant product, activates AMP-activated protein kinase with beneficial metabolic effects in diabetic and insulin-resistant states. Diabetes. 2006. PMID: 16873688
- Zhao MM, Lu J, Li S, et al. Berberine is an insulin secretagogue targeting the KCNH6 potassium channel. Nat Commun. 2021. PMID: 34556670
- Nguyen JT, Tian DD, Tanna RS, et al. Assessing Transporter-Mediated Natural Product-Drug Interactions Via In vitro-In Vivo Extrapolation: Clinical Evaluation With a Probe Cocktail. Clin Pharmacol Ther. 2021. PMID: 33174626
- Nguyen JT, Arian CM, Tanna RS, et al. The Pharmacokinetic Interaction Between Metformin and the Natural Product Goldenseal Is Metformin Dose-Dependent: A Three-Arm Crossover Study in Adults With Type 2 Diabetes. Clin Transl Sci. 2025. PMID: 39943692
- Ji L, Ma J, Ma Y, et al. Berberine Ursodeoxycholate for the Treatment of Type 2 Diabetes: A Randomized Clinical Trial. JAMA Netw Open. 2025. PMID: 40029660
- Ji L, Cheng Z, Ma J, et al. HTD1801 in Combination with Metformin for Type 2 Diabetes. NEJM Evid. 2026. PMID: 42251702
- Xie W, Su F, Wang G, et al. Glucose-lowering effect of berberine on type 2 diabetes: A systematic review and meta-analysis. Front Pharmacol. 2022. PMID: 36467075
- Kwon M, Choi YA, Choi MK, et al. Organic cation transporter-mediated drug-drug interaction potential between berberine and metformin. Arch Pharm Res. 2015. PMID: 25359200
- Blais JE, Huang X, Zhao JV. Overall and Sex-Specific Effect of Berberine for the Treatment of Dyslipidemia in Adults: A Systematic Review and Meta-Analysis of Randomized Placebo-Controlled Trials. Drugs. 2023. PMID: 36941490
- Elahi Vahed I, Shahir-Roudi E, Nojumi S, et al. The effect of berberine on obesity indices: a systematic review and meta-analysis. Int J Obes (Lond). 2026. PMID: 41310257
- Funk RS, Singh RK, Winefield RD, et al. Variability in Potency Among Commercial Preparations of Berberine. J Diet Suppl. 2018. PMID: 28792254
- Wu X, Li Q, Xin H, et al. Effects of berberine on the blood concentration of cyclosporin A in renal transplanted recipients: clinical and pharmacokinetic study. Eur J Clin Pharmacol. 2005. PMID: 16133554
- Williams JM, Yadlapalli N, Franchi F. Natural Is Not Always Safe: A Case of Torsades De Pointes Induced by Berberine Supplementation. Cureus. 2026. PMID: 42488242
- Williams JM, Yadlapalli N, Franchi F. Correction: Natural Is Not Always Safe: A Case of Torsades De Pointes Induced by Berberine Supplementation. Cureus. 2026. PMID: 42698826
This content is for education only and is not medical advice. Talk to your healthcare provider before starting any supplement, especially if you are pregnant, nursing, take medication, or have a medical condition.