Comparison
Berberine vs. Metformin
Berberine or metformin – what really works for blood sugar?
The short answer
Both lower blood sugar in type 2 diabetes, but on very different foundations. Metformin has been approved for type 2 diabetes for decades, with outcome data and comprehensive pharmacovigilance; in the Diabetes Prevention Program it reduced diabetes incidence by a relative 31 percent in 3,234 people with elevated glucose levels. Berberine improves surrogate markers, pooled across more than 30 studies by 0.19 percentage points of HbA1c, with low study quality and not a single metabolic outcome trial. The head-to-head comparison rests on a pilot study with 36 participants and no non-inferiority test; meta-analyses find no significant difference from oral antidiabetic drugs, which is not proof of equivalence. The EFSA NDA Panel also could not derive a safe intake level for berberine. Metformin is prescription-only; the decision belongs with a doctor.
The numbers side by side
| Axis | Berberine | Metformin | Difference |
|---|---|---|---|
| Human evidenceHow much and how good the data in humans are | 6 | 8 | Metformin better studied |
| MechanismHow well the chain of action is understood | 6 | 8 | Metformin better studied |
| Safety dataHow well safety has been studied – not how safe it is | 5 | 9 | Metformin better studied |
| Hype gap10 means: the advertising stays level with the data | 3 | 5 | Metformin closer to the data |
| Track record of useHow much documented use in humans exists – does not measure whether it works | 6 | 10 | Metformin in use longer |
| Rating | Supported, with caveats · Well supported, heavily overhyped | ||
“Track record of use” measures how long and how widely something has been used in humans – not whether it works. The BK-Score is a subjective assessment by Biohacking Kompakt based on published scoring rules (German). It is not a scientific rating, not the consensus of a professional society and not a medical recommendation.
Why these numbers
Berberine
The effect on surrogate markers is real and has been replicated several times, but the framework around it no longer supports the previous rating. Evidence stays at 6: there are many randomized trials (Lan et al., J Ethnopharmacol 2015, PMID 25498346: 27 RCTs, 2,569 patients; Ju et al., Phytomedicine 2018, PMID 30466986: 16 studies, 2,147 participants; Shadin et al., PLoS One 2026, PMID 42213651: more than 30 studies, more than 2,000 participants), but the meta-analyses by Lan and Ju rate the quality of the included studies as low, Shadin 2026 reports considerable heterogeneity, the pooled effects are small (HbA1c −0.19 percentage points, fasting glucose −0.71 mmol/l) and outcome data are missing entirely (Seo & Woo, Int J Mol Sci 2026, PMID 42653115: “Berberine improves glycemic surrogates but lacks cardiovascular or renal outcome evidence”). The methodologically best study, Chen et al., Lancet Gastroenterol Hepatol 2020, PMID 31926918, met its primary endpoint (adenoma recurrence 36 percent versus 47 percent, relative risk 0.77; 1,108 randomized) — but it tests a use for which berberine is not marketed. Mechanism down from 7 to 6: the target has not been confirmed in humans. Oral bioavailability is 0.68 percent (rat), only low nanogram-per-millilitre plasma levels are reached in humans (Dumitru et al., Pharmaceuticals 2026, PMID 42653808), and the glucose disposal rate measured in the euglycaemic clamp did not differ significantly from placebo in the group comparison (Zhang et al., J Clin Endocrinol Metab 2008, PMID 18397984, P = 0.063). The lipid mechanism via stabilisation of LDL receptor mRNA, by contrast, is cleanly described (Kong et al., Nature Medicine 2004, PMID 15531889). Safety down from 6 to 5: Chen 2020 does provide controlled data over up to two years in more than 1,000 people, but in the draft endorsed on 29 January 2026 the EFSA NDA Panel concluded that the data do not allow a safe intake level for any of the 13 berberine-containing botanical preparations assessed, and it considers the genotoxicity question open. Hype down from 4 to 3: the nickname “herbal Ozempic” has no comparative data behind it — no randomized trial against a GLP-1 receptor agonist could be found — while the market is growing strongly (96 million US dollars in US sales over 12 months, up 39 percent; Holland & Barrett up 167 percent) and, according to the Verbraucherzentrale (German consumer advice centre), the substance may currently not be used in food supplements as an unauthorised novel food. Application down from 7 to 6: berberine-containing plants have been used for centuries, but in the EU without a regulatory framework, with national upper limits that differ by more than an order of magnitude, and without an authorised medicinal product in Germany. Direction from positive to mixed: a demonstrated effect on surrogate markers with consistently low study quality, almost complete regional concentration (Yin 2008, Zhang 2008 and Chen 2020 all in China; Lan and Dong additionally draw on CBM, CNKI and Wanfang), missing outcome data and a safety assessment that did not yield a safe intake level. The existing statement “plant compound, almost as effective as metformin for blood sugar” rests solely on a pilot study with 36 participants (Yin et al., Metabolism 2008, PMID 18442638) without a non-inferiority test; the two meta-analyses on the head-to-head comparison find no statistically significant difference from oral antidiabetic drugs, which is an absence of evidence of a difference and not evidence of equivalence.
Metformin
Approved for type 2 diabetes for decades, with outcome data and comprehensive pharmacovigilance. The longevity claim is something else: the trial that is decisive for it has not been completed, and there are indications that metformin blunts training adaptations.
Frequently asked questions
Is berberine as effective as metformin?
That has not been shown. Its reputation as “nature's metformin” goes back to Yin 2008: 36 adults with newly diagnosed type 2 diabetes were randomized to berberine or metformin, and HbA1c fell from 9.5 to 7.5 percent on berberine. The paper calls itself a pilot study, and there is no non-inferiority test. Two meta-analyses with 27 and 14 studies found no significant difference from oral antidiabetic drugs. That is an absence of evidence of a difference, not evidence of equivalence.
Are there outcome trials for berberine like there are for metformin?
Not for blood sugar and blood lipids. There, only surrogate markers such as HbA1c, fasting glucose and LDL are documented; a 2026 review notes that cardiovascular and renal outcome data are missing. The only large berberine trial with a clinical endpoint concerns colorectal adenomas: recurrence in 36 versus 47 percent, 1,108 randomized. Metformin has decades of outcome data for type 2 diabetes. For the longevity claim, however, metformin lacks proof too: the TAME trial has no results yet.
Are berberine and metformin legal in Germany?
Metformin is prescription-only and approved for type 2 diabetes; any use to slow ageing is off-label. There is no authorised berberine medicinal product in Germany. According to the Verbraucherzentrale, the German consumer advice centre, isolated berberine may not be used in food supplements because it counts as an unauthorised novel food. In January 2026, the EFSA NDA Panel could not derive a safe intake level for any of the 13 berberine-containing botanical preparations it assessed.
How do berberine and metformin differ in safety?
Metformin's safety has been well studied over decades: gastrointestinal complaints are common at the start, vitamin B12 status needs watching with long-term use, and in two trials training adaptation was weaker. With berberine, gastrointestinal complaints are also common, but interactions via CYP3A4 and P-glycoprotein matter more; in kidney transplant recipients, ciclosporin exposure rose by 34.5 percent. The EFSA NDA Panel considers the genotoxicity question open. For anyone on regular medication, this belongs in a doctor's hands.
See all ratings in the ranking (German app)
Information only, not medical advice and not a usage or dosage recommendation. Prescription-only and unapproved substances belong in a doctor's hands.