Berberine benefits: what the human trials actually show

Blood sugar and type 2 diabetes

Berberine is an alkaloid from plants such as barberry and goldenseal, and blood sugar is where its human evidence is deepest. A 2022 meta-analysis of 37 randomized trials in 3,048 adults with type 2 diabetes found lower fasting glucose and HbA1c at 0.6 to 1.5 g a day. A 2023 meta-analysis of 20 trials in 1,761 adults, mostly with impaired glucose metabolism, found improved fasting glucose and insulin resistance (HOMA-IR) at the same doses. See our topic page on berberine for the wider picture.

The most informative single trial is PREMOTE. It enrolled 409 adults with newly diagnosed type 2 diabetes for 12 weeks. HbA1c fell by 0.99 percentage points on berberine alone and 1.04 on berberine plus probiotics. Placebo fell 0.59 and probiotics alone 0.53, so the gap over placebo was about 0.4 points. Probiotics did not beat placebo, and adding them to berberine gave no extra benefit.

A smaller three-month trial in 36 newly diagnosed adults tested 1.5 g a day against a metformin comparison arm. About a third of those on berberine reported transient gastrointestinal effects. No trial has tested berberine as a replacement for diabetes medication.

Cholesterol and blood pressure

Lipids are the second-best-supported area. A 2018 meta-analysis of 16 trials in 2,147 adults with dyslipidemia found lower total cholesterol, LDL and triglycerides at 0.6 to 1.5 g a day. A 2023 meta-analysis of 18 placebo-controlled trials in 1,788 adults found similar LDL and total cholesterol reductions. HDL did not improve and moved differently by sex: slightly higher in women, slightly lower in men.

Individual trials are less tidy. A 2021 trial in 84 men with high cholesterol gave 1,000 mg a day for 12 weeks. Total and HDL cholesterol dropped modestly versus placebo, while triglycerides, blood pressure and body measurements did not differ. An 8-week trial of a four-ingredient formula (berberine, red yeast rice, milk thistle and guggul) lowered cholesterol markers. No arm isolated berberine, so it says little about berberine alone.

On blood pressure, a 2025 meta-analysis of 12 placebo-controlled trials in 889 adults with metabolic syndrome found lower triglycerides and waist circumference at 300 mg to 1.5 g a day. Systolic and diastolic pressure and HDL did not change. Berberine is not an established blood pressure treatment, and taking it with antihypertensive drugs carries a separate interaction concern, covered below.

Weight and waist size

Berberine is often sold for weight loss, and the trials are unconvincing. A 2020 dose-response meta-analysis of 10 trials found small reductions in BMI (−0.29 kg/m²) and waist circumference (−2.75 cm) but no significant change in body weight (−0.11 kg, p=0.79). A separate meta-analysis of 12 trials in 849 adults found no significant change in weight or BMI, though waist-to-hip ratio fell slightly.

A third analysis of 12 trials did report modest reductions in weight and BMI, so the pooled results conflict. Participants generally had an existing metabolic condition, and no trial measured appetite, hunger or cravings. There is no direct evidence for healthy adults trying to lose weight. Evidence strength: weak and inconsistent.

PCOS

Two meta-analyses reach limited conclusions. A 2018 analysis of 9 trials in women with PCOS and insulin resistance judged the data insufficient to conclude anything and asked for better placebo-controlled trials. A larger 2019 analysis of 12 trials found lower total testosterone, a lower LH/FSH ratio, lower cholesterol and a smaller waist versus placebo or no treatment. BMI did not change, and there was no solid evidence of better live birth or other clinical outcomes.

One 2023 trial in 130 women with PCOS used a more absorbable phytosome formulation at 1,100 mg a day for 90 days. Acne improved in 50% of the treatment group versus 16% of controls. It was open-label, specific to PCOS-related acne and used a different formulation, so it does not generalize to standard berberine or to acne at large. A rat study of higher-dose berberine improved glucose tolerance, which cannot be turned into a human dose.

For context, an umbrella review of 28 PCOS nutrition meta-analyses found no high-certainty evidence for any supplement. Probiotics, omega-3, vitamin D, inositol and curcumin showed moderate-to-high certainty effects on specific insulin, glucose or lipid markers. Berberine was not among them. Evidence strength: weak to moderate for hormone markers, absent for clinical outcomes.

Gut health, infectious diarrhea and colon polyps

The best gut result is in diarrhea-predominant IBS. A 2015 placebo-controlled trial in 132 adults gave 400 mg a day of berberine hydrochloride for 8 weeks. Diarrhea, abdominal pain and urgency fell significantly versus placebo, and treatment was well tolerated. It is one trial in one IBS subtype. A 12-person safety trial in ulcerative colitis, at 900 mg a day for three months, is too small to judge benefit and recorded one grade 3 liver-enzyme elevation.

Older trials tested berberine sulfate in hospitalized adults with acute diarrhea. A single 400 mg dose reduced stool volume and shortened diarrhea caused by E. coli, with a smaller effect against cholera. A separate trial found no clear antisecretory effect against cholera on its main measure and no benefit in diarrhea not caused by cholera. This is a different salt, dose and population than a daily supplement, and says nothing about yeast infections, urinary infections or viral illness.

The most notable finding involves colon polyps. In a trial of 1,108 adults who had precancerous colorectal adenomas removed, 600 mg a day for up to two years lowered recurrence to 36% versus 47% on placebo (relative risk 0.77). No cancers occurred in either group. A follow-up six years after treatment ended still showed lower recurrence: 34.7% versus 52.1%. This is polyp recurrence in a specific post-removal population, not cancer prevention or treatment. Evidence strength: moderate for that one outcome, from a single trial program.

Liver and inflammation

The liver picture depends on who is studied. A 2020 meta-analysis of 12 trials found no significant change in ALT or AST, even as weight, BMI and CRP dropped modestly. A 2024 meta-analysis of 10 trials in 811 people with diagnosed fatty liver disease found significant reductions in ALT, AST and GGT, with improvements in lipids and insulin resistance and only mild gastrointestinal side effects. These are enzyme markers, not clinical outcomes.

Safety signals are mixed. The LiverTox reference states that berberine has not been linked to enzyme elevations or clinically apparent liver injury in typical use. Against that, the ulcerative colitis trial recorded one grade 3 transaminase elevation among 12 people, and a 2024 mouse study found 30 days of berberine worsened liver inflammation and fibrosis in mice carrying the alpha-1 antitrypsin deficiency mutation, though not in normal mice.

For inflammation, a 2023 meta-analysis pooled 18 trials (1,600 participants). It found small reductions in IL-6 (−1.18 pg/mL), TNF-alpha (−3.72 pg/mL) and CRP (−1.33 mg/L), with signs of an effect at under 1,000 mg a day. Participants mostly had metabolic conditions, study quality was low and results varied widely. No trial tested arthritis, gout or lupus, or measured pain or swelling.

Claims with little or no human evidence

Many popular claims have no human trial behind them. The table summarizes what exists.

Animal findings often fail to carry over to people, and several of these studies did not report a dose or tested berberine only in combination. The menopause trials are a good example: a soy isoflavone-and-berberine product improved cholesterol and reported better symptoms versus calcium and vitamin D3, but berberine's dose was not stated. A 14-week four-ingredient trial with about 200 mg a day of berberine measured a bone-turnover marker, not hot flashes or mood.

The GLP-1 label deserves particular skepticism. Berberine raised GLP-1 secretion in rats and a human intestinal cell line, but no human trial has measured it in a person, and weight outcomes in people were small or null. Related fiber and probiotic evidence is also small and inconsistent.

Safety, interactions and dose realities

Gastrointestinal effects are the most consistently reported. At 1.5 g a day, about a third of participants in one trial had transient diarrhea, constipation, nausea or cramping. A 2008 trial of 1.0 g a day in 116 adults noted mild to moderate constipation in five people. There is no evidence it acts as a stimulant.

• Heart rhythm: a 2026 systematic review of 22 studies documented QT prolongation, slowed heart rate and low blood pressure in some cases, and a 2026 case report describes torsades de pointes in a 92-year-old man two weeks after he started a berberine supplement. His rhythm normalized once he stopped.
• Drug interactions: a 2026 review found berberine can raise blood levels of some other drugs and add to the effect of blood-sugar-lowering medications, even though little berberine itself reaches the bloodstream. See the berberine interactions guide for more.
• Pregnancy and newborns: laboratory and animal work found berberine displaces bilirubin from blood proteins, which is why it is advised against in pregnancy and in newborns.

Absorption also complicates dosing. A phytosome formulation produced about ten times higher blood berberine levels than unformulated berberine in healthy volunteers, and 100 to 200 mg of dihydroberberine gave higher blood levels than 500 mg of plain berberine in five men, with no difference in blood sugar or insulin. Both findings are small and do not show that better absorption means a better clinical result. For the trial dose ranges in detail, see the berberine dosage overview.