Everstack Evidence

CC Small or few human trials. Promising, not proven. How we grade →Evidence grade · Last reviewed

Berberine dose, side effects and drug interactions: what the trials used and who should skip it

Bottom line: Berberine has Grade C evidence for modest reductions in fasting glucose, triglycerides and weight; trials are mostly small, short and low quality, digestive complaints are common, and it can alter how the liver processes medications.

The clearest findings are practical ones. Trials mostly used 600 to 1,500 mg a day split across meals, digestive complaints such as constipation were the most common side effect, and in healthy men two weeks of berberine slowed three liver enzymes that process many common drugs (Guo 2012). Pooled trials show modest drops in fasting glucose and triglycerides and a weight change of under 1 kg. Most efficacy trials were small, short and of low quality, so grade C.

Published Sources checked 12 sourcesEverstack research teamHow we work

Reviewed . We sell some of what we grade.

A bowl of lentil salad with greens
A bowl of lentil salad with greens. Photo: James Sutton, CC0 1.0

Editorial section. Research summaries are not product claims and contain no product links or offers. A summary is never proof that any product works.

The question
What doses did berberine trials use, how often did people get side effects, which medicines can it interact with, and does it really help with weight?
Study design
Dose and effect: a pilot trial in adults with type 2 diabetes at 500 mg three times a day for 3 months, including a head-to-head comparison with metformin in 36 adults (Yin 2008); a placebo-controlled trial of 1.0 g a day for 3 months in 116 adults with type 2 diabetes and high blood lipids (Zhang 2008); a meta-analysis of 28 randomized trials with 2,313 people with type 2 diabetes (Liang 2019); a meta-analysis of placebo-controlled trials on the parts of metabolic syndrome (Liu 2025); and a meta-analysis of 23 randomized trials on weight and waist size (Elahi Vahed 2026). Long-term safety: the largest placebo-controlled trial, 1,108 adults in China given 300 mg twice a day for up to 2 years after colon polyp removal (Chen 2020). Interactions: a crossover study in healthy men taking 300 mg three times a day for 2 weeks with standard probe drugs (Guo 2012), a randomized trial in 104 kidney transplant recipients taking cyclosporine (Wu 2005), and a cell and rat study of berberine and metformin (Kwon 2015). Newborns: a lab and rat study of bilirubin binding (Chan 1993) and NIH NCCIH guidance. Absorption: a review of berberine's oral bioavailability (Liu 2016). [1][2][3][4][5][6][7][8][9][10][11][12]
What they found
Dose: trials mostly used 600 to 1,500 mg a day, split into two or three doses with meals: 300 mg twice a day (Chen 2020), 1.0 g a day (Zhang 2008) and 500 mg three times a day (Yin 2008). Berberine is poorly absorbed when swallowed (Liu 2016); we found no trial comparing one daily dose with split doses. More was not better: in Liang 2019 the effect on blood glucose was no longer clear in trials lasting more than 90 days, at doses above 2 g a day, or in people over 60. Glucose and lipids: Liang 2019 pooled a fasting glucose about 0.5 mmol/L lower than control, and berberine added to glucose-lowering drugs lowered glucose more than either alone. Liu 2025 found lower triglycerides (about 0.37 mmol/L), fasting glucose (about 0.52 mmol/L), LDL cholesterol (about 0.50 mmol/L) and waist size (about 3.3 cm), with no effect on HDL cholesterol or blood pressure and no difference in side effects versus placebo. Weight: Elahi Vahed 2026 pooled a weight change of 0.88 kg and a BMI change of 0.48 kg/m2, with no change in waist-to-hip ratio. That is far smaller than the 'nature's Ozempic' label on social media suggests. Side effects: in Yin 2008, 20 of 58 people (34.5%) had short-lived digestive side effects and no liver or kidney harm was seen; in Zhang 2008, 5 people had mild to moderate constipation; in Chen 2020, constipation was the most common side effect (1% vs under 0.5% on placebo) and no serious adverse events were reported over up to 2 years. Interactions: after 2 weeks at 300 mg three times a day in healthy men, a marker of CYP2D6 activity rose ninefold, a CYP2C9 marker doubled, and exposure to midazolam, a CYP3A4 test drug, rose about 40% with 27% slower clearance; CYP2C19 and CYP1A2 did not change (Guo 2012). In kidney transplant recipients, cyclosporine blood levels ended 29.3% higher with berberine than without it (Wu 2005). In cells and rats, berberine blocked the transporters (OCT1 and OCT2) that clear metformin and raised metformin levels (Kwon 2015); we found no human study confirming this. Newborns: in lab tests berberine pushed bilirubin off its carrier protein about ten times more strongly than phenylbutazone, a known displacer, and it raised unbound bilirubin in rats; the authors advised avoiding it in jaundiced newborns and in pregnancy (Chan 1993). NIH NCCIH says the berberine in goldenseal can be harmful to newborns and that goldenseal should not be used in pregnancy or breastfeeding. [1][2][3][4][5][6][7][8][9][10][11][12]
Limitations
Most efficacy trials were small, lasted about 3 months, were done in China and had a high or unclear risk of bias; Elahi Vahed 2026 noted poor reporting of purity, potency and amounts and a lack of blinding. Glucose and lipid numbers are lab markers, not health outcomes. Chen 2020 studied polyp recurrence, not metabolic health, but it gives the best long-term safety data. The interaction studies were small and done in healthy men or transplant patients, and the metformin interaction has only been shown in cells and rats. Supplement products vary in how much berberine they contain. Funding: Chen 2020 was funded by Chinese government science grants, Liu 2025 by the China Academy of Traditional Chinese Medicine, and Yin 2008 received US NIH grants (NIDDK and NCCIH) according to Europe PMC; funding for the other studies was not stated in the abstracts we read.
What this does not mean
Berberine is not a diabetes treatment, not a weight-loss drug and not a substitute for any prescribed medicine. Our only claim is structure/function: supports healthy glucose metabolism. Skip it if you are pregnant, breastfeeding or buying for a child. If you take cyclosporine, diabetes medicines such as metformin or insulin, or any other prescription drug, ask your pharmacist or clinician first; they can check your list against CYP3A4, CYP2D6 and CYP2C9. Follow the label, take it with meals, and do not stack it with other berberine or goldenseal products. If you want to know whether it does anything for you, get a fasting glucose or HbA1c test before you start and again after about 3 months.
Sources
  1. Yin J, Xing H, Ye J. Metabolism. 2008;57(5):712-717. Pilot trials incl. vs metformin. doi.org/10.1016/j.metabol.2008.01.013 Our summary: Efficacy of berberine in patients with type 2 diabetes mellitus
  2. Zhang Y et al. J Clin Endocrinol Metab. 2008;93(7):2559-2565. 1.0 g a day vs placebo, 116 adults. doi.org/10.1210/jc.2007-2404 Our summary: Treatment of Type 2 Diabetes and Dyslipidemia with the Natural Plant Alkaloid Berberine
  3. Liang Y et al. Endocr J. 2019;66(1):51-63. Meta-analysis, 28 trials. doi.org/10.1507/endocrj.EJ18-0109 Our summary: Effects of berberine on blood glucose in patients with type 2 diabetes mellitus: a systematic literature review and a meta-analysis
  4. Liu D et al. Front Pharmacol. 2025;16:1572197. Meta-analysis, metabolic syndrome components. doi.org/10.3389/fphar.2025.1572197 Our summary: Efficacy and safety of berberine on the components of metabolic syndrome: a systematic review and meta-analysis of randomized placebo-controlled trials
  5. Elahi Vahed I et al. Int J Obes. 2026;50(1):53-73. Meta-analysis, 23 trials, weight and waist. doi.org/10.1038/s41366-025-01943-x Our summary: The effect of berberine on obesity indices: a systematic review and meta-analysis
  6. Chen YX et al. Lancet Gastroenterol Hepatol. 2020;5(3):267-275. 2-year placebo-controlled trial, 1,108 adults. doi.org/10.1016/S2468-1253(19)30409-1 Our summary: Berberine versus placebo for the prevention of recurrence of colorectal adenoma: a multicentre, double-blinded, randomised controlled study
  7. Guo Y et al. Eur J Clin Pharmacol. 2012;68(2):213-217. CYP inhibition in healthy men. doi.org/10.1007/s00228-011-1108-2 Our summary: Repeated administration of berberine inhibits cytochromes P450 in humans
  8. Wu X et al. Eur J Clin Pharmacol. 2005;61(8):567-572. Cyclosporine levels in kidney transplant recipients. doi.org/10.1007/s00228-005-0952-3 Our summary: Effects of berberine on the blood concentration of cyclosporin A in renal transplanted recipients: clinical and pharmacokinetic study
  9. Kwon M et al. Arch Pharm Res. 2015;38(5):849-856. Berberine and metformin, cells and rats. doi.org/10.1007/s12272-014-0510-6 Our summary: Organic cation transporter-mediated drug-drug interaction potential between berberine and metformin
  10. Chan E. Biol Neonate. 1993;63(4):201-208. Bilirubin displacement, lab and rats. doi.org/10.1159/000243932 Our summary: Displacement of Bilirubin from Albumin by Berberine
  11. Liu CS et al. Fitoterapia. 2016;109:274-282. Oral bioavailability review. doi.org/10.1016/j.fitote.2016.02.001 Our summary: Research progress on berberine with a special focus on its oral bioavailability
  12. NIH NCCIH: Goldenseal (berberine and newborns, pregnancy, breastfeeding). www.nccih.nih.gov/health/goldenseal

Related summaries

Spot an error? Email hello@everstackhealth.com and we will correct it within 48 hours.