Emerging Diet

Green Tea Extract: The Pill Is Not a Stronger Cup, and the Liver Signal Is Real

Summary

Drinking green tea is a benign, mildly-beneficial habit, but the concentrated EGCG supplement is a different thing, not just a stronger cup: the fat-loss and metabolism claims are trivial in humans and collapse to nothing once caffeine is controlled out, while high-dose green tea extract carries a real, documented idiosyncratic liver-injury signal (case series, an RCT with transaminase elevations, a strong genetic susceptibility marker, and regulatory action), which is worst on an empty stomach, so the honest verdict is "the beverage is fine and lightly good for you, the pill is oversold for f

Why Emerging

Emerging Evidence as an overall verdict, because the affirmative case for taking the supplement is thin and the parts that are firm push against it rather than for it. The efficacy claim (fat loss) is trivial and caffeine-attributable, resting on an industry-tilted literature; the observational beverage benefit is soft, confounded, and non-transferable to the extract. The safety signal is genuinely stronger than the efficacy case (a regulatory-grade review, an RCT, a replicated genetic marker), but it is a signal about harm in susceptible people, idiosyncratic and low-incidence, not a robust dose-response, so it does not lift the entry's affirmative confidence. A supplement whose best-established property is a rare idiosyncratic liver risk and whose marketed benefit is trivial belongs at Emerging.

NOT Foundational or Strong because there is no gold-standard evidence for an affirmative benefit; the fat-loss effect is negligible and caffeine-confounded, and the observational tea-drinking data is confounded and off the supplement case.

NOT purely Experimental because the load-bearing safety claim is better than experimental: an idiosyncratic hepatotoxicity signal documented across a pharmacopeial systematic review, a double-blind RCT with transaminase elevations, and a replicated HLA association, plus concordant action from three independent regulators. That safety scaffolding is why the entry sits at Emerging rather than the floor, even though the affirmative-benefit case is weak.

The per-claim split (read this, not just the headline):
• Beverage is benign / mildly beneficial: observational, weak, confounded, kept off the supplement case.
• Fat-loss effect trivial and caffeine-driven: Emerging-to-Moderate as a debunk (meta-analysis of RCTs, but small effect, industry-tilted literature).
• Idiosyncratic hepatotoxicity of the extract: the firmest strand (USP review + RCT + HLA marker + three regulators), but a harm signal, not an affirmative benefit.
• Empty-stomach / dose-ceiling guidance: regulator-endorsed (EFSA, USP), high practical confidence.

Practical takeaway

The framing to hold: drinking green tea is a benign habit with a weak, observational upside; the concentrated EGCG supplement is a different product that does essentially nothing for fat loss beyond its caffeine and adds a rare-but-real liver risk. The pill is not a stronger cup of tea.

The drink: fine, enjoy it.
• Green tea as a beverage is safe and lightly beneficial. A few cups a day is a reasonable, low-risk habit. Do not scare people off it, and do not oversell it as heart-disease prevention: the cardiovascular association is observational and confounded.

The extract: don't take it for the marketed reason.
• Fat loss / "thermogenic fat-burner": skip it. The anthropometric effect in humans is trivial and driven by caffeine, not catechins. If you want the caffeine effect, that is a caffeine conversation, see caffeine_stimulant_guidance, not a reason to buy high-dose EGCG.
• "Antioxidant detox": not a real indication; this entry does not endorse it.

If you take a high-dose extract anyway, reduce the risk you can control.
• Never on an empty stomach. Fasting-state bolus dosing raises EGCG bioavailability and is exactly the condition the USP empty-stomach warning and the liver signal target. Take it with food if at all.
• Respect the dose ceiling. EFSA could not identify a safe supplement dose and flagged transaminase elevations at or above 800 mg EGCG/day; the RCT injury signal sits in that high-dose, multi-month range. Lower is safer, and there is no efficacy reason to push the dose.
• Stop and get liver enzymes checked if you develop symptoms. New fatigue, nausea, right-upper-quadrant discomfort, dark urine, or jaundice while taking green tea extract warrants stopping the supplement and a clinician review; do not attribute those symptoms to something else by default (see liver_health_nafld for the workup).
• Extra caution if you have known liver disease, take other hepatotoxic agents, or have had a prior idiosyncratic drug reaction. The injury is host-dependent, and stacked liver stress is the wrong context for a high-dose bolus.

Know the honest ceiling. Outside the caffeine effect (which you can get more cheaply and predictably elsewhere), the expected benefit of the extract is essentially nil, and it comes with a rare downside the beverage does not have. For most people the correct move is: drink the tea, skip the pill.

Evidence detail

Why This Entry Exists

Green tea extract lives on a rhetorical trick: the marketing borrows the reassuring halo of a healthy beverage ("it's just green tea, and green tea is good for you") to sell a concentrated capsule that behaves nothing like the drink. The infusion and the extract share a name and a molecule, and almost nothing else that matters for either efficacy or safety. This entry exists to break that equivalence in both directions at once: to keep drinking green tea in its rightful place as a benign, weakly-beneficial habit, and to separate it cleanly from a supplement that is oversold for fat loss and carries a genuine, regulator-flagged liver signal.

The failure modes run both ways. Over-claim it and a healthy person swallows a high-dose "thermogenic fat-burner" that does essentially nothing for body composition beyond its caffeine content, while quietly running an idiosyncratic hepatotoxicity risk on an empty stomach. Over-correct and you either scare people off a perfectly benign cup of tea, or you wave away a documented liver-injury signal as supplement-industry scaremongering when the entities documenting it (a pharmacopeial standards body, a European food-safety authority, an NIH injury network) have no product to sell and no motive to invent harm. The whole point of the entry is to hold the split: beverage benign, pill oversold and non-trivially risky, and to say exactly why the concentrated form changes the risk category.

What bad advice this protects against, in all directions:
• "Green tea extract burns fat and boosts metabolism" → in humans the anthropometric effect is trivial and driven by caffeine; catechins on their own do not significantly move BMI, weight, or waist at all. The "fat-burner" pitch is not supported by the isolated compound.
• "It's just concentrated green tea, so it must be as safe as drinking it" → false. The concentrated extract has a documented idiosyncratic hepatocellular injury signal (case series, an RCT, regulatory labels) that the beverage does not carry. Concentration and empty-stomach bolus dosing, not the leaf, create the risk.
• "Green tea extract wrecks your liver" → the opposite over-claim. Injury is idiosyncratic and rare in absolute terms; most people taking high-dose extract are not harmed. The signal is real but low-incidence and host-dependent, not a routine outcome.
• "The fat-loss effect is literally zero" → don't overstate the null either. With caffeine present, catechins produced a small, statistically significant drop (about half a BMI point, roughly a kilogram of body weight). The honest read is "trivial and not from the catechins," not "nothing."
• "Drinking green tea prevents heart disease" → the cohort association (roughly 2 cups a day with lower cardiovascular mortality) is observational, confounded, and does not transfer to the isolated extract. Don't launder tea-drinking epidemiology into a case for the pill.
• "Take the extract any time, doesn't matter" → it matters. Fasting-state dosing raises EGCG bioavailability and is exactly the condition the empty-stomach warning targets; the safest use is with food, if at all.

This entry owns the beverage-versus-extract verdict (benign drink, oversold pill) and the green-tea-extract hepatotoxicity signal. It defers the caffeine mechanism and stimulant dosing to caffeine_stimulant_guidance, and any actual liver-disease workup or NAFLD context to liver_health_nafld. It states those boundaries and routes there rather than re-arguing them.

Evidence

Organised by claim, with the tier signal inline. The safety signal (concentrated extract, idiosyncratic liver injury) and the efficacy debunk (trivial, caffeine-attributable fat loss) are the load-bearing parts; the observational beverage benefit is deliberately kept weak and off the supplement-efficacy case. Read the tier signals, not just the thesis.

The efficacy claim is trivial and caffeine-driven, not a catechin effect (Efficacy debunk).

1. The fat-loss effect is small, statistically significant with caffeine, and vanishes without it. In a meta-analysis of 15 randomised controlled trials (1,243 patients), green tea catechins with caffeine lowered BMI by only about 0.55 units and body weight by about 1.38 kg; catechins without caffeine did not significantly alter anthropometric measures at all. The effect is real in the statistical sense and clinically negligible, and what little there is tracks the caffeine, not the EGCG. This is the core reason the "thermogenic fat-burner" pitch does not survive contact with the isolated compound. (Phung OJ, et al. "Effect of green tea catechins with or without caffeine on anthropometric measures: a systematic review and meta-analysis." Am J Clin Nutr 2010;91(1):73-81. Emerging-to-Moderate as an efficacy claim: a meta-analysis of RCTs, but the effect is trivial and collapses when caffeine is removed. The weight-loss/thermogenesis literature is disproportionately supplement-industry funded, with publication bias favouring positive metabolism findings, so even this small pooled effect is likely an overestimate, see publication_bias_and_evidence_distortion.)

The concentrated extract carries a real, documented liver signal (Safety signal, load-bearing).

2. There is a documented idiosyncratic hepatotoxicity signal from high-dose green tea extract. In the NIH Drug-Induced Liver Injury Network, 40 of 1,414 causality-assessed cases (about 3%) were attributed to green tea (Hoofnagle 2021); the United States Pharmacopeia's comprehensive review documented a hepatocellular injury pattern, with case reports linking injury to EGCG intakes of roughly 140 to 1,000 mg/day and large inter-individual variability. This is not a beverage phenomenon; it is the concentrated extract, and the USP review was substantial enough to drive a cautionary label. (Oketch-Rabah HA, et al. "United States Pharmacopeia (USP) comprehensive review of the hepatotoxicity of green tea extracts." Toxicology Reports 2020;7:386-402; PMID 32140423. Strong as a safety signal: a regulatory-grade systematic review of adverse-event case series plus toxicology. USP is a non-profit standards body with no product to sell, so cui bono runs against the supplement industry here, which strengthens the harm signal rather than weakening it.)

3. The injury has a strong genetic susceptibility marker, explaining why it is idiosyncratic. Among DILIN patients with definite or probable green-tea-induced liver injury, 72% (26 of 36) carried at least one HLA-B\35:01 allele, far above the roughly 10 to 15% frequency in the general population. That points to an immune-mediated, host-dependent mechanism, which is why the injury is rare and unpredictable rather than a smooth dose-response everyone slides down. (Hoofnagle JH, et al. "HLA-B\35:01 and Green Tea Induced Liver Injury." Hepatology 2021;73(6):2484-2493; PMID 32892374. A case-control genetic association nested in a prospective registry; NIH/DILIN funded, independent of manufacturers. It explains the idiosyncrasy, it does not make the risk universal.)

4. The liver signal reproduces in a controlled trial, not just case reports. In the Minnesota Green Tea Trial, postmenopausal women were dosed with about 843 (+/- 44) mg EGCG/day for 12 months; ALT rose by roughly 5.4 U/L versus placebo, 8 participants developed ALT above five times the upper limit of normal, and 5.1% showed moderate-to-severe liver-function abnormalities, with the serious adverse events predominantly liver-enzyme elevations. Risk tracked COMT and UGT1A4 genotype, consistent with the host-dependent picture. This upgrades the signal from anecdote to interventional evidence, with an approximate threshold: sustained high-dose (over ~800 mg EGCG) over months. (Dostal AM, et al. Food Chem Toxicol 2015 [Minnesota Green Tea Trial safety]; Yu Z, et al. "Effect of Green Tea Supplements on Liver Enzyme Elevation." Cancer Prev Res 2017;10(10):571-579. Prospective double-blind RCT; NIH/NCCIH-funded cancer-prevention trial with no supplement-industry stake in finding harm.)

5. Regulators have formally separated the drink from the pill. The European Food Safety Authority (2018) concluded that EGCG at or above 800 mg/day taken as a supplement significantly raises serum transaminases and could not identify a safe supplement dose, while catechins from green tea infusions were judged generally safe. The USP added "Do not take on an empty stomach" to its extract monograph, and the EU subsequently restricted EGCG levels in food. This is the strongest institutional receipt for the both-ways framing: the beverage is safe, the concentrated extract is dose-capped and labelled. (EFSA Panel. "Scientific opinion on the safety of green tea catechins." EFSA Journal 2018;16(4):5239; EU Commission Regulation 2022/2340; USP extract monograph. Formal regulatory consensus from government safety bodies. Mechanistically, fasting-state bolus dosing saturates first-pass elimination and raises EGCG bioavailability, which is the basis for the empty-stomach warning.)

The beverage is the benign side, on weak evidence (Observational, kept off the supplement case).

6. Drinking green tea is associated with modestly lower cardiovascular mortality, but this is soft and does not transfer to the extract. Observational cohorts, largely Japanese, associate roughly 2 cups a day with about a 22 to 33% lower cardiovascular mortality. This is confounded by healthy-user patterns, is not replicated in supplement RCTs of hard outcomes, and says nothing about the isolated extract. It earns green tea its "benign, mildly beneficial habit" status and nothing more. (EFSA 2018 opinion [infusions generally safe]; representative cohort evidence summarised in a Frontiers in Nutrition 2022 meta-analysis of green tea and cardiovascular risk factors, PMC9871939. Observational, soft-endpoint evidence only, appropriately kept out of the supplement-efficacy claim, see healthy_user_bias reasoning via publication_bias_and_evidence_distortion.)

Mechanism

This entry owns the beverage-versus-extract distinction and the hepatotoxicity mechanism, not the general caffeine pharmacology (deferred to caffeine_stimulant_guidance) or liver-disease workup (deferred to liver_health_nafld). What follows is only enough mechanism to make the split intelligible.

Why the fat-loss claim is biochemically plausible but functionally empty. Catechins, EGCG chief among them, can inhibit catechol-O-methyltransferase and modestly extend noradrenergic tone, which is the sliver of thermogenic rationale the marketing rests on. But in humans that translates into a negligible anthropometric effect, and the meta-analytic signal that does exist rides on co-ingested caffeine rather than the catechins themselves. "Plausibly thermogenic in a cell assay" is not "burns fat in a person," and the gap between those two statements is where the fat-burner pitch lives.

Why concentration, not the leaf, creates the liver risk. A cup of brewed green tea delivers a modest catechin dose, diluted, sipped over time, usually with food or between meals, and heavily first-pass metabolised. A high-dose extract capsule delivers a large EGCG bolus, and taken on an empty stomach that bolus saturates first-pass elimination and raises systemic EGCG bioavailability sharply. The same molecule at the same total daily amount is not the same exposure once you concentrate it and remove the food matrix, which is precisely why regulators wrote an empty-stomach warning for the pill and cleared the drink.

Why the injury is idiosyncratic rather than dose-linear. If green tea extract were a simple dose-dependent hepatotoxin, everyone at high doses would show injury proportional to intake. Instead the injury is rare, unpredictable, and concentrated in people carrying HLA-B\*35:01, with severity tracking COMT and UGT1A4 genotype. That signature (immune-associated HLA allele, host-dependent, low incidence) is the fingerprint of idiosyncratic drug-induced liver injury: a susceptibility that most people do not have, triggered in the few who do. It is why the honest framing is "real but rare and host-dependent," not "destroys livers."

Why the beverage benefit does not carry over. The cohort cardiovascular association attaches to a lifestyle pattern (people who drink green tea daily), soaked in healthy-user confounding, and to a low, food-integrated catechin exposure. None of that machinery survives isolation into a capsule, so a soft observational benefit for the drink is not evidence for the supplement, and treating it as such is the category error the entry exists to block.

Risks And Contraindications

• The load-bearing risk is idiosyncratic hepatotoxicity from the concentrated extract, worst on an empty stomach. High-dose green tea extract has a documented hepatocellular liver-injury signal: case series in the NIH DILIN (about 3% of causality-assessed cases attributed to green tea), transaminase elevations in a controlled 12-month RCT (about 843 mg EGCG/day, ALT up ~5.4 U/L, 8 participants with ALT above five times the upper limit of normal, 5.1% with moderate-to-severe liver-function abnormalities, the serious adverse events predominantly ALT rises), and regulatory action (EFSA flagged transaminase elevation at or above 800 mg/day and could not set a safe supplement dose; USP added an empty-stomach warning; the EU restricted EGCG in food). Fasting-state bolus dosing raises bioavailability and is the highest-risk condition. This is the safety fact the entry exists to surface.
• But do not overstate it: the injury is idiosyncratic and rare in absolute terms. Most people taking high-dose extract are not harmed. Susceptibility is host-dependent, concentrated in carriers of HLA-B\*35:01 (72% of definite/probable cases versus 10 to 15% of the general population) with severity tracking COMT and UGT1A4 genotype. The correct message is "real but low-incidence and immune-mediated," not "green tea extract routinely destroys livers." Overstating the harm is its own failure mode.
• The beverage does not carry this risk. Brewed green tea is judged generally safe and is not the concern. Do not let the extract signal frighten anyone off drinking tea; the distinction between the drink and the concentrated pill is the entire point and is regulator-endorsed, not activist framing.
• Do not overstate the fat-loss null as "zero." There is a small, statistically significant, caffeine-mediated anthropometric effect (about half a BMI point, roughly a kilogram). The honest line is "trivial and not from the catechins," not "nothing." Overstating the null invites a fair rebuttal that undercuts the credible safety point.
• Keep caffeine and NAFLD out of scope. Stimulant dose, tolerance, and timing belong in caffeine_stimulant_guidance; actual liver-disease evaluation and NAFLD management belong in liver_health_nafld. This entry owns the extract's fat-loss verdict and its hepatotoxicity signal, and defers those specifics rather than re-arguing them.
• This entry does not issue a blanket verdict on all EGCG indications. Green tea extract is also studied, weakly, for cardiovascular and cancer endpoints. This entry owns fat-loss efficacy and liver safety, not a global judgement on every possible use; do not read the fat-loss debunk as a claim that EGCG can do nothing anywhere.

Controversy

Nature: a concentrated supplement sold under the reassuring halo of a healthy beverage, where the two are conflated on purpose, and where error is possible at both poles: over-claiming the extract as a safe fat-burner on one side, and over-claiming it as a routine liver poison (or scaring people off the drink) on the other.

Position A — "Drinking green tea is benign and mildly good for you; the extract is oversold." The beverage-and-debunk take.
• Best evidence: EFSA judged catechins from infusions generally safe; observational cohorts associate about 2 cups a day with lower cardiovascular mortality; and the fat-loss effect of the isolated catechins is trivial and caffeine-attributable, collapsing to non-significant without caffeine. The "thermogenic fat-burner" pitch is a marketing construct.
• Where it goes wrong if overstated: it can tip into laundering the observational tea-drinking benefit into a case for the pill, or into "the fat-loss effect is literally zero," when a small caffeine-mediated effect does exist.

Position B — "High-dose green tea extract is a genuinely different risk profile from the drink." The safety-signal take.
• Best evidence: an idiosyncratic hepatocellular injury documented in DILIN case series, reproduced as transaminase elevations in a controlled RCT, tied to a strong genetic susceptibility marker (HLA-B\*35:01), and acted on by three independent regulators (EFSA dose flag, USP empty-stomach label, EU food restriction). The concentrated extract is not just a stronger cup of tea.
• Where it goes wrong if overstated: it can slide into "green tea extract wrecks your liver" or frighten people off the beverage, when the injury is idiosyncratic, rare in absolute terms, and host-dependent, and the drink is safe.

The funding/bias dimension — cui bono, both ways. Toward over-claiming benefit: the supplement industry markets green tea extract as a thermogenic fat-burner and antioxidant detox, and the weight-loss RCT base is disproportionately industry-sponsored with a publication-bias tilt toward positive metabolism results, so even the small pooled effect is likely inflated and is anyway caffeine-attributable. Toward the safety pole, cui bono runs the other way and is conflict-clean: the bodies documenting the liver signal (USP, EFSA, the EU Commission, NIH-DILIN, the NCCIH-funded Minnesota Green Tea Trial) sell no product and have no financial motive to invent risk. The marketing points one way; the independent safety science points the other.

Realised Position: substantiate both ways, and reject the equivalence. Drinking green tea is a benign, mildly-beneficial habit; enjoy it, and defer the caffeine specifics to caffeine_stimulant_guidance. The isolated EGCG supplement delivers at best a trivial, caffeine-confounded fat-loss effect (catechins alone do not move anthropometrics) while adding a real idiosyncratic hepatotoxicity risk that is worst on an empty stomach and concentrated in genetically susceptible people (HLA-B\*35:01). Concentration and delivery, not dose alone, change the risk category. The pill is not a stronger cup of tea: drink the tea, and if you take the extract, do it with food, at a modest dose, and stop at any sign of liver trouble.

Cross-Pillar Connections

Green tea extract is a diet/supplement topic with a stimulant tail and a liver tail, so its connections span the supplement-literacy, caffeine, and liver lines.
• Supplements (caffeine_stimulant_guidance): owns caffeine pharmacology, dose, tolerance, and timing; this entry holds only that the extract's trivial fat-loss effect is caffeine-attributable and routes the stimulant conversation there.
• Conditions (liver_health_nafld): owns liver-disease evaluation and NAFLD context; this entry holds only that a documented idiosyncratic hepatotoxicity signal exists for the concentrated extract and defers any actual liver workup there.
• Supplements (supplement_form_elemental_dose_and_bioavailability): the general principle that form, concentration, and delivery change exposure; the green-tea case (bolus extract on an empty stomach behaves nothing like brewed tea) is a concrete instance of it.
• Foundations (publication_bias_and_evidence_distortion): the mechanism behind the industry-tilted, selectively-cited weight-loss literature that inflates the extract's apparent efficacy.
• Supplements (universal_nearuniversal_supplementation): the reference point for which supplements clear the bar for near-universal use — green tea extract conspicuously does not, which is the contrast this entry draws.

What would change our mind

Falsifiability: explicit upgrade/downgrade criteria from source

• We'd upgrade the efficacy read if a large, independent (non-industry) RCT showed a clinically meaningful fat-loss or metabolic effect from catechins with caffeine controlled out. Right now the effect is trivial and caffeine-attributable, and the literature tilts industry-positive.
• We'd soften the hepatotoxicity claim toward Experimental if a well-powered pharmacovigilance re-analysis showed the DILIN/EFSA transaminase signal is confounded (for example by co-ingested contaminants or misattribution) and HLA-B\*35:01 failed to replicate as a susceptibility marker. Two independent pillars would have to fall.
• We'd harden the safety framing if population-level analysis showed green-tea-attributable injury materially higher than the roughly 3% of DILI cases seen in DILIN, or a lower dose threshold than the EFSA 800 mg signal.
• We'd revise the dose guidance if a body like EFSA could establish a demonstrably safe upper dose for the extract; at present none has, which is itself part of the signal.
• What would NOT move us: the observational cardiovascular benefit of drinking tea (confounded, does not transfer to the pill), the caffeine-mediated sliver of fat-loss effect (real but trivial), or supplement-industry meta-analyses that omit the null and safety literature. The bias vector on efficacy runs toward over-claiming, and the safety signal comes from sellers of nothing.

Industry bias note

Structural incentives the evidence base may reflect

Cui bono runs both ways here, and they point in opposite directions — the marketing pushes toward over-claiming benefit, the independent safety science pushes toward the harm signal, and Realised should side with the independent bodies on both efficacy (trivial) and harm (real but idiosyncratic).
• The efficacy literature is industry-tilted toward over-claiming. The supplement industry markets green tea extract as a "thermogenic fat-burner" and "antioxidant detox," and the weight-loss/thermogenesis RCT base is disproportionately supplement-sponsored, with publication bias favouring positive metabolism findings. So even the small pooled BMI/weight effect is likely inflated, and it is anyway caffeine-attributable rather than a catechin effect (see publication_bias_and_evidence_distortion).
• The harm signal is conflict-clean, which strengthens it. The entities documenting hepatotoxicity — USP (a non-profit standards body), EFSA and the EU Commission (government safety bodies), and the NIH DILIN and NCCIH-funded Minnesota Green Tea Trial — sell no product and have no financial motive to manufacture risk. When the party with nothing to sell is the one raising the alarm, the alarm is more credible, not less.
• The beverage-versus-extract distinction is regulator-endorsed, not activist framing. It is EFSA that cleared the infusion and flagged the ≥800 mg supplement dose; it is USP that wrote the empty-stomach warning; it is the EU that restricted EGCG in food. The split this entry draws is the institutional consensus, not a Realised editorial line.
• Net vector: the marketing points toward "safe fat-burner," the independent science points toward "trivial benefit, rare-but-real liver risk." The bias structure is unusual and worth naming: over-claiming pressure sits on the efficacy side, and the safety side is exactly where the disinterested parties are, which is why the entry weights the safety signal as credible while marking the fat-loss claim as inflated.

Sources (7)

Related entries

Caffeine and Stimulant Management: Timing, Dose, Anxiety, Sleep, and PerformanceStrongLiver Health and Metabolic-Dysfunction-Associated Steatotic Liver Disease (MASLD / formerly NAFLD): Prevention, Reversal, and Metabolic ContextStrongPublication Bias: Why the Published Evidence Overstates Almost EverythingStrongSupplement Form Elemental Dose And BioavailabilityStrong
HIGHLast updated June 2026
Open in the Library: search, filter, every entry →

We set no cookies and run no ad trackers. We count visits with Cloudflare's cookieless, privacy-first analytics. The only thing stored on your device is which example you last viewed.