Berberine

updated

Berberine is sold as "nature's Ozempic" for weight loss and blood glucose, and no FDA-approved berberine drug was found in searches on 11 October 2026. In type 2 diabetes, a meta-analysis of 37 randomized trials found fasting glucose 0.82 mmol/L lower and HbA1c 0.63% lower, and an umbrella review calls for better-designed trials. Average weight loss across 23 trials was 0.88 kg, the largest placebo-controlled trial found no change in visceral or liver fat, and the GLP-1 link rests on rat and cell data.

Berberine salts; promoted as a GLP-1 stimulus, not a GLP-1 receptor agonist

  • On 11 October 2026 an openFDA Drugs@FDA query for berberine returned no match, so no FDA-approved berberine drug was found.
  • In 37 randomized trials of 3,048 patients with type 2 diabetes, berberine lowered fasting glucose by 0.82 mmol/L (95% CI -0.95 to -0.70) and HbA1c by 0.63% (95% CI -0.72 to -0.53).
  • Across 23 randomized trials in adults, body weight fell by a mean 0.88 kg (95% CI -1.36 to -0.39).
  • In 337 diabetes-free adults with obesity and fatty liver disease, 6 months of berberine did not change visceral fat area or liver fat compared with placebo.
  • Berberine raised GLP-1 release in rats and an intestinal cell line; a PubMed search on 11 October 2026 found no randomized or clinical trial measuring GLP-1 response to berberine in people.
  • Dihydroberberine has one human study, an exposure study in 5 young men that found higher plasma berberine and no glucose or insulin difference.

What is berberine, and is it an approved drug?

Berberine is a quaternary cation sold as salts such as the hydrochloride, and dihydroberberine is a separate, reduced compound. On 11 October 2026 no FDA-approved berberine drug was found; the one DailyMed listing is a bulk-ingredient entry with no indication or dosing.

Berberine is PubChem CID 2353, C20H18NO4+, a quaternary cation sold as salts such as the hydrochloride, with a molecular weight of 336.4 g/mol and CAS 2086-83-1 among the listed synonyms. Dihydroberberine is PubChem CID 10217, C20H19NO4, 337.4 g/mol, CAS 483-15-8. It is a different, reduced compound, not a salt of berberine, and the one human study that compares them found different plasma berberine levels, so no berberine result transfers to dihydroberberine or the reverse. HTD1801 (berberine ursodeoxycholate) is a third, distinct entity.

On 11 October 2026 an openFDA Drugs@FDA query for berberine returned no match, so no FDA-approved berberine drug was found. A DailyMed search the same day returned one listing, berberine hydrochloride granules from K. Patel Phyto Extractions Private Limited (label effective 9 January 2026). Its readable text gives names and ingredient fields only, with no indication or dosing, so it is a bulk-ingredient listing, not a drug label. FDA dietary-supplement status, and any FDA or FTC action on berberine weight claims, are not covered here.

Product, dose and purity vary across the trials on this page, so no trial result applies to a specific retail product.

PubChem: berberine identity; no FDA-approved berberine drug found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound records CID 2353 (berberine) and CID 10217 (dihydroberberine). openFDA Drugs@FDA query and DailyMed search for berberine run on 11 October 2026.

Berberine: PubChem CID 2353, C20H18NO4+ (a quaternary cation, sold as salts such as the hydrochloride), 336.4 g/mol, with CAS 2086-83-1 among the listed synonyms. Dihydroberberine: PubChem CID 10217, C20H19NO4, 337.4 g/mol, CAS 483-15-8. Dihydroberberine is a different, reduced compound, not a salt of berberine. HTD1801 (berberine ursodeoxycholate) is a third, distinct entity. On 11 October 2026 a DailyMed search for berberine returned one listing, berberine hydrochloride granules from K. Patel Phyto Extractions Private Limited (label effective 9 January 2026), whose readable text gives names and ingredient fields only, with no indication or dosing: a bulk-ingredient listing, not a drug label. An openFDA Drugs@FDA query for berberine the same day returned no match, so no FDA-approved berberine drug was found.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined cation and its salts; dihydroberberine separate
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search
Funding
not stated in the source

No berberine result transfers to dihydroberberine or the reverse. FDA dietary-supplement status is not established here.

Read the original source
Dihydroberberine: 5 men, higher plasma berberine, no glucose or insulin difference

Clinical trial

Moon JM et al. Nutrients. 2021;14(1). PMID 35010998.

Across placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine, each given as a four-dose protocol, plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04). Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions, which the authors attribute to short duration and insulin-responsive participants, and they call for efficacy studies in overweight, glucose-intolerant people.

Participants / model
5 healthy young men (mean age 26)
Treatment
Oral berberine 500 mg or dihydroberberine 100 or 200 mg, four doses over a day, or placebo
Follow-up
One day per condition
Study design
Randomized double-blind crossover pilot
Funding
The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan

The only human dihydroberberine study found. It measured exposure in 5 people, not weight or glycemic control.

Read the original source

What did the type 2 diabetes studies find?

Blood glucose fell. A meta-analysis of 37 randomized trials in 3,048 patients with type 2 diabetes found fasting glucose 0.82 mmol/L lower and HbA1c 0.63% lower, with larger effects at higher baseline values. An umbrella review of 11 meta-analyses calls for better methodological quality and confirmation in high-quality randomized trials.

A meta-analysis of 37 randomized trials (3,048 patients with type 2 diabetes, searched to November 2021) found that oral berberine, alone or with oral glucose-lowering drugs, lowered fasting plasma glucose by 0.82 mmol/L (95% CI -0.95 to -0.70), HbA1c by 0.63% (95% CI -0.72 to -0.53) and 2-hour post-load glucose by 1.16 mmol/L (95% CI -1.36 to -0.96). Effects were larger at higher baseline glucose and HbA1c. Total adverse events had RR 0.73 (95% CI 0.55 to 0.97), and hypoglycemia RR 0.48 (95% CI 0.21 to 1.08).

An earlier pilot randomized 36 adults with newly diagnosed type 2 diabetes to berberine or metformin (0.5 g three times a day) for 3 months. The authors reported a similar hypoglycemic effect, with HbA1c in the berberine group falling from 9.5% to 7.5%.

An umbrella review of 11 meta-analyses (searched to June 2022) found berberine affected glucose, insulin resistance, lipids and body parameters, and its authors call for better methodological quality in the meta-analyses and confirmation in high-quality randomized trials. A 2025 meta-analysis of placebo-controlled trials in metabolic syndrome found fasting glucose lower by 0.515 mmol/L (95% CI -0.847 to -0.183) and BMI lower by 0.435 kg/m2 (95% CI -0.856 to -0.013), and also asked for more rigorous trials.

Doses and products vary across these trials, and some combined berberine with other glucose-lowering drugs.

Glucose: 37 trials, 3,048 patients with type 2 diabetes; HbA1c -0.63%

Systematic review and meta-analysis

Xie W et al. Front Pharmacol. 2022;13:1015045. PMID 36467075.

In 37 randomized trials (3,048 patients with type 2 diabetes, searched to November 2021), berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI -0.95 to -0.70), HbA1c by 0.63% (95% CI -0.72 to -0.53) and 2-hour post-load glucose by 1.16 mmol/L (95% CI -1.36 to -0.96), with larger effects at higher baseline glucose and HbA1c. Total adverse events had RR 0.73 (95% CI 0.55 to 0.97) and hypoglycemia RR 0.48 (95% CI 0.21 to 1.08). An earlier pilot randomized 36 adults with newly diagnosed type 2 diabetes to berberine or metformin (0.5 g three times a day) for 3 months and reported a similar hypoglycemic effect, with HbA1c on berberine falling from 9.5% to 7.5%; in that paper 20 patients (34.5%) had transient gastrointestinal adverse effects and no functional liver or kidney damage was seen. An umbrella review of 11 meta-analyses (searched to June 2022) found effects on glucose, insulin resistance, lipids and body parameters, lists constipation and diarrhea as common side effects, and calls for better methodological quality and confirmation in high-quality randomized trials. A 2025 meta-analysis of placebo-controlled trials in metabolic syndrome found fasting glucose lower by 0.515 mmol/L (95% CI -0.847 to -0.183) and BMI by 0.435 kg/m2 (95% CI -0.856 to -0.013), and also asked for more rigorous trials.

Participants / model
3,048 patients with type 2 diabetes in 37 trials; 36 newly diagnosed adults in the pilot; adults with metabolic syndrome in the 2025 review
Treatment
Oral berberine, alone or with oral glucose-lowering drugs; doses vary
Follow-up
Trials searched to November 2021; 3 months in the pilot
Study design
Meta-analyses, an umbrella review and a small pilot randomized trial
Funding
not stated in the source

Products and doses vary, and some trials combined berberine with other glucose-lowering drugs. The reviews ask for better-designed trials.

Read the original source

How much weight did people lose on berberine?

Less than 1 kg on average. Across 23 randomized trials in adults, body weight fell by a mean 0.88 kg and BMI by 0.48 kg/m2. In the largest placebo-controlled trial found, 337 diabetes-free adults with obesity and fatty liver disease, 6 months of berberine did not change visceral fat area or liver fat.

A systematic review and meta-analysis of 23 randomized trials in adults found berberine lowered body weight by a mean difference of -0.88 kg (95% CI -1.36 to -0.39) and BMI by -0.48 kg/m2 (95% CI -0.89 to -0.07). Waist circumference fell by -1.32 (95% CI -2.24 to -0.41); the abstract prints a unit for this figure that does not fit a waist measurement, so no unit is given here. Waist-to-hip ratio did not change significantly. The authors say future trials should address lack of blinding and randomization and report purity, potency and gram amounts. No competing interests were declared.

The largest placebo-controlled trial found randomized 337 diabetes-free adults with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD; mean age 41.8, 65.6% male) to berberine 1 g daily or placebo for 6 months at 11 Chinese hospitals. Visceral adipose tissue area (difference 1.4%, 97.5% CI -2.4% to 5.2%) and liver fat (0.9%, 97.5% CI -0.4% to 2.1%) did not differ. LDL cholesterol was lower by 7.72 mg/dL (95% CI -13.13 to -1.93), adverse events were similar, and no conflict of interest was reported. The abstract does not state body weight change. The trial is registered as NCT05647915.

For scale, the WEGOVY (semaglutide) label reports a mean body weight change at week 68 of -14.9% versus -2.4% on placebo in Study 2 (1,306 and 655 adults with obesity or overweight and a comorbidity), and -9.6% versus -3.4% in Study 3 (404 and 403 adults with type 2 diabetes). The berberine meta-analysis reports kilograms, so the two are not subtracted here.

Weight: 23 trials, -0.88 kg; largest placebo trial (337 adults) found no change in visceral or liver fat

Systematic review and meta-analysis

Elahi Vahed I et al. Int J Obes (Lond). 2026;50(1):53-73. PMID 41310257.

Across 23 randomized trials in adults, berberine lowered body weight by a mean difference of -0.88 kg (95% CI -1.36 to -0.39), BMI by -0.48 kg/m2 (95% CI -0.89 to -0.07) and waist circumference by -1.32 (95% CI -2.24 to -0.41), with no significant change in waist-to-hip ratio. The authors say future trials should address lack of blinding and randomization and report purity, potency and gram amounts. In the largest placebo-controlled trial found, 337 diabetes-free adults with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD) took berberine 1 g daily or placebo for 6 months at 11 Chinese hospitals: visceral adipose tissue area (difference 1.4%, 97.5% CI -2.4% to 5.2%) and liver fat (0.9%, 97.5% CI -0.4% to 2.1%) did not differ, LDL cholesterol was lower by 7.72 mg/dL (95% CI -13.13 to -1.93), and adverse events were similar. That abstract does not state body weight change.

Participants / model
Adults of mixed health status in 23 randomized trials; 337 diabetes-free adults with obesity and MASLD (mean age 41.8, 65.6% male)
Treatment
Oral berberine; doses and products vary across trials; 1 g daily in the large trial
Follow-up
Varied across trials; 6 months in the large trial
Study design
Systematic review and meta-analysis, plus a randomized placebo-controlled trial
Funding
No competing interests were declared for the meta-analysis; no conflict of interest was reported for the 337-adult trial

The meta-analysis abstract prints a unit for the waist figure that does not fit a waist measurement, so no unit is given here.

Read the original source
WEGOVY label, version effective 18 June 2026: the drug the nickname invokes

FDA prescribing information

DailyMed, U.S. National Library of Medicine. WEGOVY (semaglutide) injection, prescribing information, label version effective 18 June 2026. Set ID ee06186f-2aa3-4990-a760-757579d8f77b.

WEGOVY is FDA-labeled, with a reduced calorie diet and more physical activity, to reduce excess body weight and maintain the reduction long term in adults and pediatric patients 12 and older with obesity and in adults with overweight plus at least one weight-related comorbidity, and to reduce the risk of major adverse cardiovascular events in adults with established cardiovascular disease and obesity or overweight. In label Study 2 (obesity or overweight with comorbidity, intent-to-treat, 1,306 on WEGOVY and 655 on placebo), mean body weight change at week 68 was -14.9% versus -2.4%, a difference of -12.4% (95% CI -13.3 to -11.6). In Study 3 (type 2 diabetes with obesity or overweight, 404 and 403) it was -9.6% versus -3.4%.

Participants / model
Adults with obesity or overweight in the label trials
Treatment
WEGOVY subcutaneous injection (the label also covers tablets)
Follow-up
Label version effective 18 June 2026; label trials of 68 weeks
Study design
FDA prescribing information
Funding
not stated in the source

This is semaglutide, not berberine. The berberine weight meta-analysis reports kilograms, so the two results are not subtracted here.

Read the original source

Is dihydroberberine better absorbed, and does it do more?

It raised plasma berberine more than berberine itself in one exposure study of 5 young men, and glucose and insulin did not differ. No weight or glycemic control data for dihydroberberine were found.

In a randomized double-blind crossover pilot in 5 young men (mean age 26), placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine were each given as a four-dose protocol over a day. Plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04).

Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions. The authors attribute this to the short duration and to insulin-responsive participants, and call for efficacy studies in overweight, glucose-intolerant people. The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan.

This is the only human dihydroberberine study found, and it measured exposure in 5 people, not weight or glycemic control. Higher plasma berberine is not the same as a better outcome.

Dihydroberberine: 5 men, higher plasma berberine, no glucose or insulin difference

Clinical trial

Moon JM et al. Nutrients. 2021;14(1). PMID 35010998.

Across placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine, each given as a four-dose protocol, plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04). Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions, which the authors attribute to short duration and insulin-responsive participants, and they call for efficacy studies in overweight, glucose-intolerant people.

Participants / model
5 healthy young men (mean age 26)
Treatment
Oral berberine 500 mg or dihydroberberine 100 or 200 mg, four doses over a day, or placebo
Follow-up
One day per condition
Study design
Randomized double-blind crossover pilot
Funding
The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan

The only human dihydroberberine study found. It measured exposure in 5 people, not weight or glycemic control.

Read the original source

Is HTD1801 the same as a berberine supplement?

No. HTD1801 is berberine ursodeoxycholate, a single drug entity with its own dosing. In a 12-week phase 2 trial in 113 patients with type 2 diabetes it lowered HbA1c by 0.4% and 0.7% more than placebo, and its phase 3 trials are listed completed with no posted results.

HTD1801 was tested in a phase 2 double-blind placebo-controlled 12-week trial in China in 113 patients with type 2 diabetes and HbA1c of 7.0% to 10.5% (placebo 38, 500 mg twice daily 37, 1000 mg twice daily 38). The least-squares mean HbA1c difference versus placebo was -0.4% (95% CI -0.79 to -0.03; P = .04) at 500 mg and -0.7% (95% CI -1.10 to -0.35; P < .001) at 1000 mg. Treatment-emergent adverse events occurred in 52.2% overall, generally mild, with one serious event (retinal hemorrhage) judged unlikely related. Authors report stock options in the developer, Shenzhen HighTide Biopharmaceutical.

On 11 October 2026 ClinicalTrials.gov listed phase 3 HTD1801 trials in type 2 diabetes, including NCT06350890 (n=408), NCT06353347 (n=551, add-on to metformin) and NCT06415773 (n=367, versus dapagliflozin), all completed with no posted results. HTD1801 results do not transfer to over-the-counter berberine.

HTD1801, a different entity: 113 patients, 12 weeks, HbA1c -0.4% and -0.7% versus placebo

Randomized controlled trial

Ji L et al. JAMA Netw Open. 2025;8(3):e2462185. PMID 40029660.

In 113 patients with type 2 diabetes (placebo 38, 500 mg twice daily 37, 1000 mg twice daily 38), the least-squares mean HbA1c difference versus placebo at week 12 was -0.4% (95% CI -0.79 to -0.03; P = .04) at 500 mg and -0.7% (95% CI -1.10 to -0.35; P < .001) at 1000 mg. Treatment-emergent adverse events occurred in 52.2% overall, generally mild, with one serious event (retinal hemorrhage) judged unlikely related. On 11 October 2026 ClinicalTrials.gov listed phase 3 HTD1801 trials in type 2 diabetes, including NCT06350890 (n=408), NCT06353347 (n=551, add-on to metformin) and NCT06415773 (n=367, versus dapagliflozin), all completed with no posted results.

Participants / model
113 patients with type 2 diabetes (HbA1c 7.0% to 10.5%)
Treatment
Oral HTD1801 (berberine ursodeoxycholate) 500 or 1000 mg twice daily, or placebo
Follow-up
12 weeks
Study design
Randomized double-blind placebo-controlled phase 2 trial in China
Funding
Authors report stock options in the developer, Shenzhen HighTide Biopharmaceutical

HTD1801 is a single drug entity with its own dosing, so its results do not transfer to over-the-counter berberine.

Read the original source

Is berberine "nature's Ozempic"?

No. The GLP-1 link rests on rat and cell-line data, no trial comparing berberine with a GLP-1 drug was found, and a PubMed search on 11 October 2026 for randomized or clinical trials of berberine and GLP-1 returned 0 records. Average weight loss across 23 trials was 0.88 kg.

Sellers say berberine, and its dihydroberberine form, is "nature's Ozempic" for weight loss and blood glucose, acting like a GLP-1 drug. The drug the nickname invokes is semaglutide, which as WEGOVY is FDA-labeled, with a reduced calorie diet and more physical activity, to reduce excess body weight and maintain the reduction long term in adults and pediatric patients 12 and older with obesity, and in adults with overweight plus at least one weight-related comorbidity.

The GLP-1 link comes from rats and cells. In normal rats, 5 weeks of berberine enhanced glucose-load-induced GLP-1 secretion and increased proglucagon mRNA and intestinal L cell proliferation. In the NCI-H716 intestinal cell line, berberine stimulated GLP-1 release in a concentration-dependent way, and inhibitors of PKC and AMPK reduced the release. That shows berberine can raise the body's own GLP-1 release in rodents and cells. It is not a measurement in people, and it does not make berberine a GLP-1 receptor agonist: semaglutide is the drug acting at the receptor, not a stimulus for the body's own release.

A PubMed search on 11 October 2026 for berberine with semaglutide, liraglutide, tirzepatide, exenatide or dulaglutide returned 8 records, and all 8 titles were reviewed. None describes a trial comparing berberine with a GLP-1 receptor agonist: they were reviews, a schizophrenia network meta-analysis, a rat liraglutide study, an in vitro steatosis comparison and a berberine cell study. A second search for berberine with GLP-1 or glucagon-like peptide, limited to randomized or clinical trial publication types, returned 0 records.

One of the 8 records, a 2026 pharmacy review, identified 24 transdermal patch dietary supplement products and 1 gel marketed as "natural GLP-1", with berberine, glutamine or glutamate, cinnamon and pomegranate most often listed. It found no certificates of analysis posted and many products with deceptive advertising, and its authors argue the patches are illegal because dietary supplements by law must be swallowed.

Dihydroberberine has one human study, an exposure study in 5 young men with no weight or glycemic outcome.

Searches, 11 October 2026: no berberine versus GLP-1 drug trial; 0 human GLP-1 trials

PubMed search record

PubMed search run on 11 October 2026: berberine[tiab] AND (semaglutide[tiab] OR liraglutide[tiab] OR tirzepatide[tiab] OR exenatide[tiab] OR dulaglutide[tiab]), 8 records. Second search: berberine[tiab] AND (GLP-1[tiab] OR glucagon-like peptide[tiab]) AND (randomized controlled trial[pt] OR clinical trial[pt]), 0 records. The first of the 8 records is White 2026, PMID 41480958.

The first search returned 8 records, and all 8 titles were reviewed. None describes a trial comparing berberine with a GLP-1 receptor agonist; they covered a review of "natural GLP-1" transdermal supplements, an epigenetics review, a kinase review, a microbiome review, a schizophrenia network meta-analysis, a rat liraglutide study, an in vitro steatosis comparison and a berberine cell study on cardiac and skeletal muscle. The second search returned 0 records, so no trial indexed under those publication types measured GLP-1 response to berberine in people. The first of the 8 is a 2026 pharmacy review that identified 24 transdermal patch dietary supplement products and 1 gel marketed as "natural GLP-1" (berberine, glutamine or glutamate, cinnamon and pomegranate most often listed); it reports no certificates of analysis posted and many products with deceptive advertising, and argues the patches are illegal because dietary supplements by law must be swallowed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 11 October 2026
Study design
Literature searches
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed study exists.

Read the original source
Origin of the GLP-1 claim: normal rats and an intestinal cell line, no living humans

Animal study

Yu Y et al. Biochem Pharmacol. 2010;79(7):1000-6. PMID 19945441.

In normal rats, 5 weeks of berberine enhanced glucose-load-induced GLP-1 secretion and increased proglucagon mRNA expression and intestinal L cell proliferation. In the NCI-H716 intestinal cell line, berberine concentration-dependently stimulated GLP-1 release and raised prohormone convertase 3 and proglucagon mRNA, and inhibitors of PKC and AMPK reduced the release. The abstract says earlier work from the same group found increased GLP-1 secretion in streptozotocin diabetic rats.

Participants / model
Normal rats; NCI-H716 intestinal cell line. No living humans
Treatment
Berberine given to rats for 5 weeks; cell exposure
Follow-up
5 weeks in rats
Study design
Animal and cell study
Funding
not stated in the source

This shows berberine can raise the body's own GLP-1 release in rodents and cells. It is not a measurement in people, and it does not make berberine a GLP-1 receptor agonist like semaglutide.

Read the original source
WEGOVY label, version effective 18 June 2026: the drug the nickname invokes

FDA prescribing information

DailyMed, U.S. National Library of Medicine. WEGOVY (semaglutide) injection, prescribing information, label version effective 18 June 2026. Set ID ee06186f-2aa3-4990-a760-757579d8f77b.

WEGOVY is FDA-labeled, with a reduced calorie diet and more physical activity, to reduce excess body weight and maintain the reduction long term in adults and pediatric patients 12 and older with obesity and in adults with overweight plus at least one weight-related comorbidity, and to reduce the risk of major adverse cardiovascular events in adults with established cardiovascular disease and obesity or overweight. In label Study 2 (obesity or overweight with comorbidity, intent-to-treat, 1,306 on WEGOVY and 655 on placebo), mean body weight change at week 68 was -14.9% versus -2.4%, a difference of -12.4% (95% CI -13.3 to -11.6). In Study 3 (type 2 diabetes with obesity or overweight, 404 and 403) it was -9.6% versus -3.4%.

Participants / model
Adults with obesity or overweight in the label trials
Treatment
WEGOVY subcutaneous injection (the label also covers tablets)
Follow-up
Label version effective 18 June 2026; label trials of 68 weeks
Study design
FDA prescribing information
Funding
not stated in the source

This is semaglutide, not berberine. The berberine weight meta-analysis reports kilograms, so the two results are not subtracted here.

Read the original source
Dihydroberberine: 5 men, higher plasma berberine, no glucose or insulin difference

Clinical trial

Moon JM et al. Nutrients. 2021;14(1). PMID 35010998.

Across placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine, each given as a four-dose protocol, plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04). Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions, which the authors attribute to short duration and insulin-responsive participants, and they call for efficacy studies in overweight, glucose-intolerant people.

Participants / model
5 healthy young men (mean age 26)
Treatment
Oral berberine 500 mg or dihydroberberine 100 or 200 mg, four doses over a day, or placebo
Follow-up
One day per condition
Study design
Randomized double-blind crossover pilot
Funding
The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan

The only human dihydroberberine study found. It measured exposure in 5 people, not weight or glycemic control.

Read the original source

How is berberine supposed to work, and how much reaches the blood?

In rats and an intestinal cell line, berberine raised the body's own GLP-1 release. Systemic exposure is very low: a 2026 review reports oral bioavailability of 0.68% in rats and low ng/mL plasma levels in humans, with high concentrations in the gut and liver.

The mechanism sellers cite is GLP-1. In normal rats and in NCI-H716 intestinal cells, berberine raised GLP-1 release, and inhibitors of PKC and AMPK reduced that release in cells. Semaglutide and other GLP-1 drugs act directly at the receptor. Raising the body's own GLP-1 release in rodents is a different action, and it was not measured in people in the records found.

A 2026 structured narrative review reports very low systemic exposure, with oral bioavailability of 0.68% in rats and low ng/mL plasma levels in humans, alongside high gut and liver concentrations. Those high local concentrations are what the review links to drug interactions. In the only human comparison found, 500 mg of berberine gave a plasma berberine Cmax of 0.4 ng/mL in 5 young men. No half-life or Tmax appears in the sources on this page.

Origin of the GLP-1 claim: normal rats and an intestinal cell line, no living humans

Animal study

Yu Y et al. Biochem Pharmacol. 2010;79(7):1000-6. PMID 19945441.

In normal rats, 5 weeks of berberine enhanced glucose-load-induced GLP-1 secretion and increased proglucagon mRNA expression and intestinal L cell proliferation. In the NCI-H716 intestinal cell line, berberine concentration-dependently stimulated GLP-1 release and raised prohormone convertase 3 and proglucagon mRNA, and inhibitors of PKC and AMPK reduced the release. The abstract says earlier work from the same group found increased GLP-1 secretion in streptozotocin diabetic rats.

Participants / model
Normal rats; NCI-H716 intestinal cell line. No living humans
Treatment
Berberine given to rats for 5 weeks; cell exposure
Follow-up
5 weeks in rats
Study design
Animal and cell study
Funding
not stated in the source

This shows berberine can raise the body's own GLP-1 release in rodents and cells. It is not a measurement in people, and it does not make berberine a GLP-1 receptor agonist like semaglutide.

Read the original source
Interactions review, 2026: CYP and transporter effects; cyclosporine AUC up 34.5%

Narrative review

Dumitru CN et al. Pharmaceuticals (Basel). 2026;19(8). PMID 42653808.

The review (PubMed, Scopus, Web of Science and Embase to May 2026) reports very low systemic exposure (oral bioavailability 0.68% in rats, low ng/mL plasma levels in humans) alongside high gut and liver concentrations that create interaction liability: CYP3A4 inhibition and induction, quasi-irreversible CYP2D6 and CYP2C9 inhibition, transporter effects (P-glycoprotein, OCT1, OCT2, MATE1), and additive hypoglycemia, hypotension and QT effects. Interactions are documented in humans for a few pairs and mechanistic for most. The clinical anchor is cyclosporine in renal-transplant recipients, with AUC up 34.5% and trough levels 29.3% above control. The review also notes unstandardized product quality and European national limits that differ by more than an order of magnitude, and it flags metabolic self-medication in the GLP-1 era.

Participants / model
Mostly mechanistic and review-level; renal-transplant recipients for the cyclosporine finding
Treatment
Oral berberine
Follow-up
Literature to May 2026
Study design
Structured narrative review
Funding
No conflicts declared

The cyclosporine figures are reported by this review. The primary cyclosporine trial is not among the sources on this page.

Read the original source
Dihydroberberine: 5 men, higher plasma berberine, no glucose or insulin difference

Clinical trial

Moon JM et al. Nutrients. 2021;14(1). PMID 35010998.

Across placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine, each given as a four-dose protocol, plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04). Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions, which the authors attribute to short duration and insulin-responsive participants, and they call for efficacy studies in overweight, glucose-intolerant people.

Participants / model
5 healthy young men (mean age 26)
Treatment
Oral berberine 500 mg or dihydroberberine 100 or 200 mg, four doses over a day, or placebo
Follow-up
One day per condition
Study design
Randomized double-blind crossover pilot
Funding
The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan

The only human dihydroberberine study found. It measured exposure in 5 people, not weight or glycemic control.

Read the original source

What are the safety concerns, including drug interactions?

Drug interactions are the main concern. A 2026 review reports CYP3A4, CYP2D6 and CYP2C9 effects and transporter effects, and cites cyclosporine exposure 34.5% higher in renal-transplant recipients, a figure from the review rather than the primary trial. Gastrointestinal adverse effects were common in trials, and the newborn jaundice risk is unresolved.

A structured narrative review (PubMed, Scopus, Web of Science and Embase to May 2026; no conflicts declared) reports CYP3A4 inhibition and induction, quasi-irreversible CYP2D6 and CYP2C9 inhibition, transporter effects (P-glycoprotein, OCT1, OCT2, MATE1), and additive hypoglycemia, hypotension and QT effects. Interactions are documented in humans for a few pairs and mechanistic for most. The review's clinical anchor is cyclosporine in renal-transplant recipients, with AUC up 34.5% and trough levels 29.3% above control; those numbers are as reported in the 2026 review, and the primary cyclosporine trial is not among the sources on this page. The review also notes unstandardized product quality and European national limits that differ by more than an order of magnitude.

In the 37-trial type 2 diabetes meta-analysis, total adverse events had RR 0.73 (95% CI 0.55 to 0.97) and hypoglycemia RR 0.48 (95% CI 0.21 to 1.08). In the paper reporting the metformin comparison, 20 patients (34.5%) had transient gastrointestinal adverse effects and no functional liver or kidney damage was seen. The umbrella review lists constipation and diarrhea as common side effects. In the 337-adult placebo-controlled trial, adverse events were similar between groups.

In vitro, berberine displaced bilirubin from albumin about tenfold more potently on a molar basis than phenylbutazone, and in adult rats given berberine intraperitoneally for a week (10 and 20 micrograms/g daily) it raised unbound and total bilirubin. The author advises avoiding berberine-rich herbs in jaundiced newborns and pregnant women. A Singapore group reviewed the topic after the herbs were banned there for three decades over concern about jaundice and kernicterus in G6PD-deficient newborns. They found the literature conflicting, and in a single-arm study of 20 patients with chronic hematologic disease given berberine-containing herbs they saw no organ toxicity, with a transient bilirubin rise in 3 patients with thalassemia intermedia. The human newborn risk is unresolved, and the animal and in vitro findings are a reason for caution, not a measured human harm rate.

The longest placebo-controlled trial on this page ran 6 months, so safety beyond that is not covered by these sources.

Interactions review, 2026: CYP and transporter effects; cyclosporine AUC up 34.5%

Narrative review

Dumitru CN et al. Pharmaceuticals (Basel). 2026;19(8). PMID 42653808.

The review (PubMed, Scopus, Web of Science and Embase to May 2026) reports very low systemic exposure (oral bioavailability 0.68% in rats, low ng/mL plasma levels in humans) alongside high gut and liver concentrations that create interaction liability: CYP3A4 inhibition and induction, quasi-irreversible CYP2D6 and CYP2C9 inhibition, transporter effects (P-glycoprotein, OCT1, OCT2, MATE1), and additive hypoglycemia, hypotension and QT effects. Interactions are documented in humans for a few pairs and mechanistic for most. The clinical anchor is cyclosporine in renal-transplant recipients, with AUC up 34.5% and trough levels 29.3% above control. The review also notes unstandardized product quality and European national limits that differ by more than an order of magnitude, and it flags metabolic self-medication in the GLP-1 era.

Participants / model
Mostly mechanistic and review-level; renal-transplant recipients for the cyclosporine finding
Treatment
Oral berberine
Follow-up
Literature to May 2026
Study design
Structured narrative review
Funding
No conflicts declared

The cyclosporine figures are reported by this review. The primary cyclosporine trial is not among the sources on this page.

Read the original source
Glucose: 37 trials, 3,048 patients with type 2 diabetes; HbA1c -0.63%

Systematic review and meta-analysis

Xie W et al. Front Pharmacol. 2022;13:1015045. PMID 36467075.

In 37 randomized trials (3,048 patients with type 2 diabetes, searched to November 2021), berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI -0.95 to -0.70), HbA1c by 0.63% (95% CI -0.72 to -0.53) and 2-hour post-load glucose by 1.16 mmol/L (95% CI -1.36 to -0.96), with larger effects at higher baseline glucose and HbA1c. Total adverse events had RR 0.73 (95% CI 0.55 to 0.97) and hypoglycemia RR 0.48 (95% CI 0.21 to 1.08). An earlier pilot randomized 36 adults with newly diagnosed type 2 diabetes to berberine or metformin (0.5 g three times a day) for 3 months and reported a similar hypoglycemic effect, with HbA1c on berberine falling from 9.5% to 7.5%; in that paper 20 patients (34.5%) had transient gastrointestinal adverse effects and no functional liver or kidney damage was seen. An umbrella review of 11 meta-analyses (searched to June 2022) found effects on glucose, insulin resistance, lipids and body parameters, lists constipation and diarrhea as common side effects, and calls for better methodological quality and confirmation in high-quality randomized trials. A 2025 meta-analysis of placebo-controlled trials in metabolic syndrome found fasting glucose lower by 0.515 mmol/L (95% CI -0.847 to -0.183) and BMI by 0.435 kg/m2 (95% CI -0.856 to -0.013), and also asked for more rigorous trials.

Participants / model
3,048 patients with type 2 diabetes in 37 trials; 36 newly diagnosed adults in the pilot; adults with metabolic syndrome in the 2025 review
Treatment
Oral berberine, alone or with oral glucose-lowering drugs; doses vary
Follow-up
Trials searched to November 2021; 3 months in the pilot
Study design
Meta-analyses, an umbrella review and a small pilot randomized trial
Funding
not stated in the source

Products and doses vary, and some trials combined berberine with other glucose-lowering drugs. The reviews ask for better-designed trials.

Read the original source
Weight: 23 trials, -0.88 kg; largest placebo trial (337 adults) found no change in visceral or liver fat

Systematic review and meta-analysis

Elahi Vahed I et al. Int J Obes (Lond). 2026;50(1):53-73. PMID 41310257.

Across 23 randomized trials in adults, berberine lowered body weight by a mean difference of -0.88 kg (95% CI -1.36 to -0.39), BMI by -0.48 kg/m2 (95% CI -0.89 to -0.07) and waist circumference by -1.32 (95% CI -2.24 to -0.41), with no significant change in waist-to-hip ratio. The authors say future trials should address lack of blinding and randomization and report purity, potency and gram amounts. In the largest placebo-controlled trial found, 337 diabetes-free adults with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD) took berberine 1 g daily or placebo for 6 months at 11 Chinese hospitals: visceral adipose tissue area (difference 1.4%, 97.5% CI -2.4% to 5.2%) and liver fat (0.9%, 97.5% CI -0.4% to 2.1%) did not differ, LDL cholesterol was lower by 7.72 mg/dL (95% CI -13.13 to -1.93), and adverse events were similar. That abstract does not state body weight change.

Participants / model
Adults of mixed health status in 23 randomized trials; 337 diabetes-free adults with obesity and MASLD (mean age 41.8, 65.6% male)
Treatment
Oral berberine; doses and products vary across trials; 1 g daily in the large trial
Follow-up
Varied across trials; 6 months in the large trial
Study design
Systematic review and meta-analysis, plus a randomized placebo-controlled trial
Funding
No competing interests were declared for the meta-analysis; no conflict of interest was reported for the 337-adult trial

The meta-analysis abstract prints a unit for the waist figure that does not fit a waist measurement, so no unit is given here.

Read the original source
Newborn jaundice signal: in vitro and rats, plus a 20-patient human study; human risk unresolved

Animal study

Chan E. Biol Neonate. 1993;63(4):201-8. PMID 8513024.

In vitro, berberine displaced bilirubin from albumin about tenfold more potently on a molar basis than phenylbutazone. In adult rats given berberine intraperitoneally for a week (10 and 20 micrograms/g daily), it lowered bilirubin protein binding and raised unbound and total bilirubin. The author advises avoiding berberine-rich herbs in jaundiced neonates and pregnant women. A Singapore group reviewed the topic after the herbs were banned there for three decades over concern about jaundice and kernicterus in G6PD-deficient neonates: the literature evidence was conflicting, and in a single-arm study of 20 patients with chronic hematologic disease given berberine-containing herbs they saw no organ toxicity, with a transient bilirubin rise in 3 patients with thalassemia intermedia.

Participants / model
In vitro albumin; adult rats; 20 adults with chronic cytopenic hematologic conditions
Treatment
Berberine in vitro and intraperitoneally in rats; oral herbal preparations containing berberine herbs in adults
Follow-up
1 week in rats
Study design
Animal and in vitro study; small single-arm human study with a literature review
Funding
not stated in the source

The human neonatal risk is unresolved. The animal and in vitro findings are a reason for caution, not a measured human harm rate.

Read the original source

Why does berberine sit in C tier?

C fits a supplement with a large but heterogeneous glucose literature in type 2 diabetes, an average weight effect under 1 kg, poor absorption, variable products and a real drug-interaction record. B is arguable for glucose in type 2 diabetes alone, and D is arguable for the weight marketing.

A meta-analysis of 37 trials shows lower fasting glucose and HbA1c in type 2 diabetes, but the trials are heterogeneous, and both the umbrella review and the weight meta-analysis call for better-designed studies. The weight effect averages 0.88 kg, and the largest placebo-controlled trial in diabetes-free adults with obesity found no change in visceral or liver fat. Absorption is poor, product and dose vary widely, and the interaction record (cyclosporine, CYP3A4, CYP2D6) matters for anyone taking other drugs.

"Nature's Ozempic" is not supported. The GLP-1 link rests on rat and cell data, no human GLP-1 trial or head-to-head comparison with a GLP-1 drug was found, and dihydroberberine has one 5-person exposure study and no efficacy data in the records found.

Glucose: 37 trials, 3,048 patients with type 2 diabetes; HbA1c -0.63%

Systematic review and meta-analysis

Xie W et al. Front Pharmacol. 2022;13:1015045. PMID 36467075.

In 37 randomized trials (3,048 patients with type 2 diabetes, searched to November 2021), berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI -0.95 to -0.70), HbA1c by 0.63% (95% CI -0.72 to -0.53) and 2-hour post-load glucose by 1.16 mmol/L (95% CI -1.36 to -0.96), with larger effects at higher baseline glucose and HbA1c. Total adverse events had RR 0.73 (95% CI 0.55 to 0.97) and hypoglycemia RR 0.48 (95% CI 0.21 to 1.08). An earlier pilot randomized 36 adults with newly diagnosed type 2 diabetes to berberine or metformin (0.5 g three times a day) for 3 months and reported a similar hypoglycemic effect, with HbA1c on berberine falling from 9.5% to 7.5%; in that paper 20 patients (34.5%) had transient gastrointestinal adverse effects and no functional liver or kidney damage was seen. An umbrella review of 11 meta-analyses (searched to June 2022) found effects on glucose, insulin resistance, lipids and body parameters, lists constipation and diarrhea as common side effects, and calls for better methodological quality and confirmation in high-quality randomized trials. A 2025 meta-analysis of placebo-controlled trials in metabolic syndrome found fasting glucose lower by 0.515 mmol/L (95% CI -0.847 to -0.183) and BMI by 0.435 kg/m2 (95% CI -0.856 to -0.013), and also asked for more rigorous trials.

Participants / model
3,048 patients with type 2 diabetes in 37 trials; 36 newly diagnosed adults in the pilot; adults with metabolic syndrome in the 2025 review
Treatment
Oral berberine, alone or with oral glucose-lowering drugs; doses vary
Follow-up
Trials searched to November 2021; 3 months in the pilot
Study design
Meta-analyses, an umbrella review and a small pilot randomized trial
Funding
not stated in the source

Products and doses vary, and some trials combined berberine with other glucose-lowering drugs. The reviews ask for better-designed trials.

Read the original source
Weight: 23 trials, -0.88 kg; largest placebo trial (337 adults) found no change in visceral or liver fat

Systematic review and meta-analysis

Elahi Vahed I et al. Int J Obes (Lond). 2026;50(1):53-73. PMID 41310257.

Across 23 randomized trials in adults, berberine lowered body weight by a mean difference of -0.88 kg (95% CI -1.36 to -0.39), BMI by -0.48 kg/m2 (95% CI -0.89 to -0.07) and waist circumference by -1.32 (95% CI -2.24 to -0.41), with no significant change in waist-to-hip ratio. The authors say future trials should address lack of blinding and randomization and report purity, potency and gram amounts. In the largest placebo-controlled trial found, 337 diabetes-free adults with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD) took berberine 1 g daily or placebo for 6 months at 11 Chinese hospitals: visceral adipose tissue area (difference 1.4%, 97.5% CI -2.4% to 5.2%) and liver fat (0.9%, 97.5% CI -0.4% to 2.1%) did not differ, LDL cholesterol was lower by 7.72 mg/dL (95% CI -13.13 to -1.93), and adverse events were similar. That abstract does not state body weight change.

Participants / model
Adults of mixed health status in 23 randomized trials; 337 diabetes-free adults with obesity and MASLD (mean age 41.8, 65.6% male)
Treatment
Oral berberine; doses and products vary across trials; 1 g daily in the large trial
Follow-up
Varied across trials; 6 months in the large trial
Study design
Systematic review and meta-analysis, plus a randomized placebo-controlled trial
Funding
No competing interests were declared for the meta-analysis; no conflict of interest was reported for the 337-adult trial

The meta-analysis abstract prints a unit for the waist figure that does not fit a waist measurement, so no unit is given here.

Read the original source
Origin of the GLP-1 claim: normal rats and an intestinal cell line, no living humans

Animal study

Yu Y et al. Biochem Pharmacol. 2010;79(7):1000-6. PMID 19945441.

In normal rats, 5 weeks of berberine enhanced glucose-load-induced GLP-1 secretion and increased proglucagon mRNA expression and intestinal L cell proliferation. In the NCI-H716 intestinal cell line, berberine concentration-dependently stimulated GLP-1 release and raised prohormone convertase 3 and proglucagon mRNA, and inhibitors of PKC and AMPK reduced the release. The abstract says earlier work from the same group found increased GLP-1 secretion in streptozotocin diabetic rats.

Participants / model
Normal rats; NCI-H716 intestinal cell line. No living humans
Treatment
Berberine given to rats for 5 weeks; cell exposure
Follow-up
5 weeks in rats
Study design
Animal and cell study
Funding
not stated in the source

This shows berberine can raise the body's own GLP-1 release in rodents and cells. It is not a measurement in people, and it does not make berberine a GLP-1 receptor agonist like semaglutide.

Read the original source
Searches, 11 October 2026: no berberine versus GLP-1 drug trial; 0 human GLP-1 trials

PubMed search record

PubMed search run on 11 October 2026: berberine[tiab] AND (semaglutide[tiab] OR liraglutide[tiab] OR tirzepatide[tiab] OR exenatide[tiab] OR dulaglutide[tiab]), 8 records. Second search: berberine[tiab] AND (GLP-1[tiab] OR glucagon-like peptide[tiab]) AND (randomized controlled trial[pt] OR clinical trial[pt]), 0 records. The first of the 8 records is White 2026, PMID 41480958.

The first search returned 8 records, and all 8 titles were reviewed. None describes a trial comparing berberine with a GLP-1 receptor agonist; they covered a review of "natural GLP-1" transdermal supplements, an epigenetics review, a kinase review, a microbiome review, a schizophrenia network meta-analysis, a rat liraglutide study, an in vitro steatosis comparison and a berberine cell study on cardiac and skeletal muscle. The second search returned 0 records, so no trial indexed under those publication types measured GLP-1 response to berberine in people. The first of the 8 is a 2026 pharmacy review that identified 24 transdermal patch dietary supplement products and 1 gel marketed as "natural GLP-1" (berberine, glutamine or glutamate, cinnamon and pomegranate most often listed); it reports no certificates of analysis posted and many products with deceptive advertising, and argues the patches are illegal because dietary supplements by law must be swallowed.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Searches run on 11 October 2026
Study design
Literature searches
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed study exists.

Read the original source
Interactions review, 2026: CYP and transporter effects; cyclosporine AUC up 34.5%

Narrative review

Dumitru CN et al. Pharmaceuticals (Basel). 2026;19(8). PMID 42653808.

The review (PubMed, Scopus, Web of Science and Embase to May 2026) reports very low systemic exposure (oral bioavailability 0.68% in rats, low ng/mL plasma levels in humans) alongside high gut and liver concentrations that create interaction liability: CYP3A4 inhibition and induction, quasi-irreversible CYP2D6 and CYP2C9 inhibition, transporter effects (P-glycoprotein, OCT1, OCT2, MATE1), and additive hypoglycemia, hypotension and QT effects. Interactions are documented in humans for a few pairs and mechanistic for most. The clinical anchor is cyclosporine in renal-transplant recipients, with AUC up 34.5% and trough levels 29.3% above control. The review also notes unstandardized product quality and European national limits that differ by more than an order of magnitude, and it flags metabolic self-medication in the GLP-1 era.

Participants / model
Mostly mechanistic and review-level; renal-transplant recipients for the cyclosporine finding
Treatment
Oral berberine
Follow-up
Literature to May 2026
Study design
Structured narrative review
Funding
No conflicts declared

The cyclosporine figures are reported by this review. The primary cyclosporine trial is not among the sources on this page.

Read the original source
Dihydroberberine: 5 men, higher plasma berberine, no glucose or insulin difference

Clinical trial

Moon JM et al. Nutrients. 2021;14(1). PMID 35010998.

Across placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine, each given as a four-dose protocol, plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04). Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions, which the authors attribute to short duration and insulin-responsive participants, and they call for efficacy studies in overweight, glucose-intolerant people.

Participants / model
5 healthy young men (mean age 26)
Treatment
Oral berberine 500 mg or dihydroberberine 100 or 200 mg, four doses over a day, or placebo
Follow-up
One day per condition
Study design
Randomized double-blind crossover pilot
Funding
The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan

The only human dihydroberberine study found. It measured exposure in 5 people, not weight or glycemic control.

Read the original source

Studies and sources

PubChem: berberine identity; no FDA-approved berberine drug found, 11 October 2026

Chemical identity record

National Library of Medicine. PubChem compound records CID 2353 (berberine) and CID 10217 (dihydroberberine). openFDA Drugs@FDA query and DailyMed search for berberine run on 11 October 2026.

Berberine: PubChem CID 2353, C20H18NO4+ (a quaternary cation, sold as salts such as the hydrochloride), 336.4 g/mol, with CAS 2086-83-1 among the listed synonyms. Dihydroberberine: PubChem CID 10217, C20H19NO4, 337.4 g/mol, CAS 483-15-8. Dihydroberberine is a different, reduced compound, not a salt of berberine. HTD1801 (berberine ursodeoxycholate) is a third, distinct entity. On 11 October 2026 a DailyMed search for berberine returned one listing, berberine hydrochloride granules from K. Patel Phyto Extractions Private Limited (label effective 9 January 2026), whose readable text gives names and ingredient fields only, with no indication or dosing: a bulk-ingredient listing, not a drug label. An openFDA Drugs@FDA query for berberine the same day returned no match, so no FDA-approved berberine drug was found.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined cation and its salts; dihydroberberine separate
Follow-up
Searches run on 11 October 2026
Study design
NLM chemical record plus dated US approval-record search
Funding
not stated in the source

No berberine result transfers to dihydroberberine or the reverse. FDA dietary-supplement status is not established here.

Read the original source
Origin of the GLP-1 claim: normal rats and an intestinal cell line, no living humans

Animal study

Yu Y et al. Biochem Pharmacol. 2010;79(7):1000-6. PMID 19945441.

In normal rats, 5 weeks of berberine enhanced glucose-load-induced GLP-1 secretion and increased proglucagon mRNA expression and intestinal L cell proliferation. In the NCI-H716 intestinal cell line, berberine concentration-dependently stimulated GLP-1 release and raised prohormone convertase 3 and proglucagon mRNA, and inhibitors of PKC and AMPK reduced the release. The abstract says earlier work from the same group found increased GLP-1 secretion in streptozotocin diabetic rats.

Participants / model
Normal rats; NCI-H716 intestinal cell line. No living humans
Treatment
Berberine given to rats for 5 weeks; cell exposure
Follow-up
5 weeks in rats
Study design
Animal and cell study
Funding
not stated in the source

This shows berberine can raise the body's own GLP-1 release in rodents and cells. It is not a measurement in people, and it does not make berberine a GLP-1 receptor agonist like semaglutide.

Read the original source
WEGOVY label, version effective 18 June 2026: the drug the nickname invokes

FDA prescribing information

DailyMed, U.S. National Library of Medicine. WEGOVY (semaglutide) injection, prescribing information, label version effective 18 June 2026. Set ID ee06186f-2aa3-4990-a760-757579d8f77b.

WEGOVY is FDA-labeled, with a reduced calorie diet and more physical activity, to reduce excess body weight and maintain the reduction long term in adults and pediatric patients 12 and older with obesity and in adults with overweight plus at least one weight-related comorbidity, and to reduce the risk of major adverse cardiovascular events in adults with established cardiovascular disease and obesity or overweight. In label Study 2 (obesity or overweight with comorbidity, intent-to-treat, 1,306 on WEGOVY and 655 on placebo), mean body weight change at week 68 was -14.9% versus -2.4%, a difference of -12.4% (95% CI -13.3 to -11.6). In Study 3 (type 2 diabetes with obesity or overweight, 404 and 403) it was -9.6% versus -3.4%.

Participants / model
Adults with obesity or overweight in the label trials
Treatment
WEGOVY subcutaneous injection (the label also covers tablets)
Follow-up
Label version effective 18 June 2026; label trials of 68 weeks
Study design
FDA prescribing information
Funding
not stated in the source

This is semaglutide, not berberine. The berberine weight meta-analysis reports kilograms, so the two results are not subtracted here.

Read the original source
Weight: 23 trials, -0.88 kg; largest placebo trial (337 adults) found no change in visceral or liver fat

Systematic review and meta-analysis

Elahi Vahed I et al. Int J Obes (Lond). 2026;50(1):53-73. PMID 41310257.

Across 23 randomized trials in adults, berberine lowered body weight by a mean difference of -0.88 kg (95% CI -1.36 to -0.39), BMI by -0.48 kg/m2 (95% CI -0.89 to -0.07) and waist circumference by -1.32 (95% CI -2.24 to -0.41), with no significant change in waist-to-hip ratio. The authors say future trials should address lack of blinding and randomization and report purity, potency and gram amounts. In the largest placebo-controlled trial found, 337 diabetes-free adults with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD) took berberine 1 g daily or placebo for 6 months at 11 Chinese hospitals: visceral adipose tissue area (difference 1.4%, 97.5% CI -2.4% to 5.2%) and liver fat (0.9%, 97.5% CI -0.4% to 2.1%) did not differ, LDL cholesterol was lower by 7.72 mg/dL (95% CI -13.13 to -1.93), and adverse events were similar. That abstract does not state body weight change.

Participants / model
Adults of mixed health status in 23 randomized trials; 337 diabetes-free adults with obesity and MASLD (mean age 41.8, 65.6% male)
Treatment
Oral berberine; doses and products vary across trials; 1 g daily in the large trial
Follow-up
Varied across trials; 6 months in the large trial
Study design
Systematic review and meta-analysis, plus a randomized placebo-controlled trial
Funding
No competing interests were declared for the meta-analysis; no conflict of interest was reported for the 337-adult trial

The meta-analysis abstract prints a unit for the waist figure that does not fit a waist measurement, so no unit is given here.

Read the original source
Glucose: 37 trials, 3,048 patients with type 2 diabetes; HbA1c -0.63%

Systematic review and meta-analysis

Xie W et al. Front Pharmacol. 2022;13:1015045. PMID 36467075.

In 37 randomized trials (3,048 patients with type 2 diabetes, searched to November 2021), berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI -0.95 to -0.70), HbA1c by 0.63% (95% CI -0.72 to -0.53) and 2-hour post-load glucose by 1.16 mmol/L (95% CI -1.36 to -0.96), with larger effects at higher baseline glucose and HbA1c. Total adverse events had RR 0.73 (95% CI 0.55 to 0.97) and hypoglycemia RR 0.48 (95% CI 0.21 to 1.08). An earlier pilot randomized 36 adults with newly diagnosed type 2 diabetes to berberine or metformin (0.5 g three times a day) for 3 months and reported a similar hypoglycemic effect, with HbA1c on berberine falling from 9.5% to 7.5%; in that paper 20 patients (34.5%) had transient gastrointestinal adverse effects and no functional liver or kidney damage was seen. An umbrella review of 11 meta-analyses (searched to June 2022) found effects on glucose, insulin resistance, lipids and body parameters, lists constipation and diarrhea as common side effects, and calls for better methodological quality and confirmation in high-quality randomized trials. A 2025 meta-analysis of placebo-controlled trials in metabolic syndrome found fasting glucose lower by 0.515 mmol/L (95% CI -0.847 to -0.183) and BMI by 0.435 kg/m2 (95% CI -0.856 to -0.013), and also asked for more rigorous trials.

Participants / model
3,048 patients with type 2 diabetes in 37 trials; 36 newly diagnosed adults in the pilot; adults with metabolic syndrome in the 2025 review
Treatment
Oral berberine, alone or with oral glucose-lowering drugs; doses vary
Follow-up
Trials searched to November 2021; 3 months in the pilot
Study design
Meta-analyses, an umbrella review and a small pilot randomized trial
Funding
not stated in the source

Products and doses vary, and some trials combined berberine with other glucose-lowering drugs. The reviews ask for better-designed trials.

Read the original source
Dihydroberberine: 5 men, higher plasma berberine, no glucose or insulin difference

Clinical trial

Moon JM et al. Nutrients. 2021;14(1). PMID 35010998.

Across placebo, 500 mg berberine, 100 mg dihydroberberine and 200 mg dihydroberberine, each given as a four-dose protocol, plasma berberine Cmax was 3.76 ng/mL on 100 mg dihydroberberine against 0.4 ng/mL on 500 mg berberine (p = 0.005), and AUC was 284.4 against 42.3 ng/mL x 120 min (p = 0.04). Glucose (p = 0.97) and insulin (p = 0.24) did not differ across conditions, which the authors attribute to short duration and insulin-responsive participants, and they call for efficacy studies in overweight, glucose-intolerant people.

Participants / model
5 healthy young men (mean age 26)
Treatment
Oral berberine 500 mg or dihydroberberine 100 or 200 mg, four doses over a day, or placebo
Follow-up
One day per condition
Study design
Randomized double-blind crossover pilot
Funding
The last author is a paid advisor for NNB Nutrition under a conflict-of-interest management plan

The only human dihydroberberine study found. It measured exposure in 5 people, not weight or glycemic control.

Read the original source
HTD1801, a different entity: 113 patients, 12 weeks, HbA1c -0.4% and -0.7% versus placebo

Randomized controlled trial

Ji L et al. JAMA Netw Open. 2025;8(3):e2462185. PMID 40029660.

In 113 patients with type 2 diabetes (placebo 38, 500 mg twice daily 37, 1000 mg twice daily 38), the least-squares mean HbA1c difference versus placebo at week 12 was -0.4% (95% CI -0.79 to -0.03; P = .04) at 500 mg and -0.7% (95% CI -1.10 to -0.35; P < .001) at 1000 mg. Treatment-emergent adverse events occurred in 52.2% overall, generally mild, with one serious event (retinal hemorrhage) judged unlikely related. On 11 October 2026 ClinicalTrials.gov listed phase 3 HTD1801 trials in type 2 diabetes, including NCT06350890 (n=408), NCT06353347 (n=551, add-on to metformin) and NCT06415773 (n=367, versus dapagliflozin), all completed with no posted results.

Participants / model
113 patients with type 2 diabetes (HbA1c 7.0% to 10.5%)
Treatment
Oral HTD1801 (berberine ursodeoxycholate) 500 or 1000 mg twice daily, or placebo
Follow-up
12 weeks
Study design
Randomized double-blind placebo-controlled phase 2 trial in China
Funding
Authors report stock options in the developer, Shenzhen HighTide Biopharmaceutical

HTD1801 is a single drug entity with its own dosing, so its results do not transfer to over-the-counter berberine.

Read the original source
Interactions review, 2026: CYP and transporter effects; cyclosporine AUC up 34.5%

Narrative review

Dumitru CN et al. Pharmaceuticals (Basel). 2026;19(8). PMID 42653808.

The review (PubMed, Scopus, Web of Science and Embase to May 2026) reports very low systemic exposure (oral bioavailability 0.68% in rats, low ng/mL plasma levels in humans) alongside high gut and liver concentrations that create interaction liability: CYP3A4 inhibition and induction, quasi-irreversible CYP2D6 and CYP2C9 inhibition, transporter effects (P-glycoprotein, OCT1, OCT2, MATE1), and additive hypoglycemia, hypotension and QT effects. Interactions are documented in humans for a few pairs and mechanistic for most. The clinical anchor is cyclosporine in renal-transplant recipients, with AUC up 34.5% and trough levels 29.3% above control. The review also notes unstandardized product quality and European national limits that differ by more than an order of magnitude, and it flags metabolic self-medication in the GLP-1 era.

Participants / model
Mostly mechanistic and review-level; renal-transplant recipients for the cyclosporine finding
Treatment
Oral berberine
Follow-up
Literature to May 2026
Study design
Structured narrative review
Funding
No conflicts declared

The cyclosporine figures are reported by this review. The primary cyclosporine trial is not among the sources on this page.

Read the original source
Newborn jaundice signal: in vitro and rats, plus a 20-patient human study; human risk unresolved

Animal study

Chan E. Biol Neonate. 1993;63(4):201-8. PMID 8513024.

In vitro, berberine displaced bilirubin from albumin about tenfold more potently on a molar basis than phenylbutazone. In adult rats given berberine intraperitoneally for a week (10 and 20 micrograms/g daily), it lowered bilirubin protein binding and raised unbound and total bilirubin. The author advises avoiding berberine-rich herbs in jaundiced neonates and pregnant women. A Singapore group reviewed the topic after the herbs were banned there for three decades over concern about jaundice and kernicterus in G6PD-deficient neonates: the literature evidence was conflicting, and in a single-arm study of 20 patients with chronic hematologic disease given berberine-containing herbs they saw no organ toxicity, with a transient bilirubin rise in 3 patients with thalassemia intermedia.

Participants / model
In vitro albumin; adult rats; 20 adults with chronic cytopenic hematologic conditions
Treatment
Berberine in vitro and intraperitoneally in rats; oral herbal preparations containing berberine herbs in adults
Follow-up
1 week in rats
Study design
Animal and in vitro study; small single-arm human study with a literature review
Funding
not stated in the source

The human neonatal risk is unresolved. The animal and in vitro findings are a reason for caution, not a measured human harm rate.

Read the original source

Why is Berberine in C tier?

C fits a supplement with a large but low-quality trial literature for glucose in type 2 diabetes, an average weight effect under 1 kg, poor absorption and a real drug-interaction record. The "nature's Ozempic" claim is not supported: no human GLP-1 trial or comparison with a GLP-1 drug was found.