In plain terms: Berberine HCl has some human trial evidence supporting it, but the evidence is not yet strong enough for EBL to call it established. It is worth knowing about and worth discussing with your clinician.
Berberine HCl
The evidence as the record states it
Zhang et al. (J Clin Endocrinol Metab, 2008; PMID 18397984): placebo-controlled randomized trial, 116 adults with type 2 diabetes and dyslipidemia, berberine 1 g/day for 3 months; HbA1c fell from 7.5 to 6.6 percent and LDL cholesterol from 3.23 to 2.55 mmol/L within the berberine arm, each significantly different from placebo; clamp-measured glucose disposal did not differ from placebo (P = 0.063). The comparison with metformin is a different trial, Yin et al. (Metabolism, 2008; PMID 18442638), a 36-person pilot cited on the berberine protocol record. Replication: PREMOTE (Zhang et al., Nat Commun, 2020; PMID 33024120), 409 newly diagnosed patients, 20 centers, HbA1c the primary outcome, 0.99 points down on berberine alone versus 0.59 on placebo, with more gastrointestinal side effects. Pooled, placebo-controlled only: Zhao et al. (J Nutr, 2023; PMID 37598753), 20 trials, fasting glucose 0.52 mmol/L lower, HbA1c about 0.4 percent lower, HOMA-IR 0.85 lower; Blais et al. (Drugs, 2023; PMID 36941490), 18 trials, 1,788 adults, LDL cholesterol 0.46 mmol/L lower, triglycerides 0.34 lower, apolipoprotein B lower, gastrointestinal adverse events more frequent than on placebo. Lan et al. (J Ethnopharmacol, 2015; PMID 25498346): 27 trials; effects versus oral hypoglycemic drugs not significant. Lei et al. (JAMA Netw Open, 2026; PMID 41543854): 337 diabetes-free adults with obesity and fatty liver, 1 g/day for 6 months; LDL and apolipoprotein B lower, but null on the primary endpoints, visceral fat and liver fat. Interactions: Guo et al. (Eur J Clin Pharmacol, 2012; PMID 21870106), berberine 300 mg three times daily for 2 weeks inhibited CYP2D6, CYP2C9 and CYP3A4 in healthy men. Form: the dihydroberberine absorption claim rests on Moon et al. (Nutrients, 2021; PMID 35010998), a 5-man crossover funded by a dihydroberberine supplier, with no difference in glucose or insulin (see form note). Grade B: replicated placebo-controlled trials on glycemic and lipid surrogates.
Why Grade B and not Grade A
Grade B reflects moderate human evidence -- typically a single strong trial, multiple smaller trials, or surrogate endpoint data without hard clinical outcome replication. To reach Grade A, Berberine HCl would require at least one additional independent randomized trial demonstrating a hard clinical endpoint or objectively measured functional outcome in a comparable population.
What would move this grade
An independent randomized trial demonstrating a hard clinical endpoint -- mortality, major adverse event, or objectively confirmed functional improvement -- in an adequately powered sample, with results independently replicated, would support Grade A.
How this was studied
Form: Berberine taken orally as a capsule or tablet; the salt form (hydrochloride or other) is not named in any cited abstract (Zhang 2008, PMID 18397984; Yin 2008, PMID 18442638; Lei 2026, PMID 41543854). The single dihydroberberine comparison was a pharmacokinetic pilot, not an outcome trial (Moon 2021, PMID 35010998). · Dose: 1.0 g per day against placebo in the two largest single trials (Zhang 2008, PMID 18397984; Lei 2026, PMID 41543854) and 0.5 g three times daily (1.5 g per day) against metformin (Yin 2008, PMID 18442638); the placebo-controlled lipid trials pooled used 900 to 1,500 mg per day (Blais 2023 meta-analysis, PMID 36941490). The PREMOTE dose is not given in its abstract (Zhang 2020, PMID 33024120). · Population: Adults with type 2 diabetes and dyslipidemia (116 in Zhang 2008, PMID 18397984), newly diagnosed type 2 diabetes (36 plus 48 poorly controlled in Yin 2008, PMID 18442638; 409 at 20 Chinese centers in Zhang 2020, PMID 33024120), and adults with dyslipidemia, mainly in mainland China and Hong Kong (18 trials, 1,788 participants; Blais 2023 meta-analysis, PMID 36941490; 20 trials, 1,761 participants, Zhao 2023 meta-analysis, PMID 37598753). One trial enrolled 337 diabetes-free adults with obesity and fatty liver, mean age 41.8, 65.6 percent male (Lei 2026, PMID 41543854). · Duration: 12 weeks to 3 months in the diabetes trials (Zhang 2008, PMID 18397984; Yin 2008, PMID 18442638; Zhang 2020, PMID 33024120), 4 to 24 weeks across the pooled lipid trials (Blais 2023 meta-analysis, PMID 36941490), and 6 months in the fatty liver trial (Lei 2026, PMID 41543854).