Turmeric, black pepper, and 2000%

Curcumin is the yellow pigment in turmeric. It is one of the most studied natural products in existence and one of the most stubbornly disappointing, for a reason that has been known the entire time: swallow it and very little of it reaches your bloodstream.

The standard fix, printed on nearly every turmeric supplement sold, is black pepper. The active alkaloid is piperine. And the number attached to it is 2000%.

Where the number actually comes from

One paper. Shoba and five colleagues, Planta Medica, volume 64, issue 4, pages 353 to 356, 1998. The corresponding affiliation is the Department of Pharmacology at St. John's Medical College in Bangalore.

The design was a pharmacokinetic study in two species. In rats, 2 g/kg of curcumin alone, then the same with 20 mg/kg piperine. In healthy human volunteers, 2 g of curcumin alone, then the same with 20 mg of piperine. Blood was drawn over the following hours and serum curcumin measured.

In rats, bioavailability rose by 154%. In humans, by 2000%. That second number left the paper and never came back.

Note the doses, because they rarely travel with the number. Two grams of curcumin, which is a large dose, and 20 mg of piperine, which is far more piperine than a pinch of pepper on a latte. A typical capsule pairs 500 mg of curcumin extract with 5 mg of piperine. The 2000% was not measured at that ratio, at that dose, in that format.

The sentence that changes what the number means

Here is the human result, quoted exactly as the authors wrote it:

On the other hand in humans after a dose of 2 g curcumin alone, serum levels were either undetectable or very low. Concomitant administration of piperine 20 mg produced much higher concentrations from 0.25 to 1 h post drug (P < 0.01 at 0.25 and 0.5 h; P < 0.001 at 1 h), the increase in bioavailability was 2000%.

Read the first clause again. Undetectable or very low. That is the denominator.

This is the entire report in one line. A percentage increase describes a ratio between two quantities. If the starting quantity is at or below the detection limit of your assay, the ratio can be made enormous by a rise that is still, in absolute terms, very small. Twenty times a number near zero is a larger number near zero.

None of this makes the study wrong. It is a competent small pharmacokinetic experiment reporting a statistically significant effect, and the abstract is honest about the baseline. The distortion happened later, when the ratio was lifted out and the denominator was left behind. Nobody had to lie. They just had to quote one number instead of two.

I will also flag a limit on my own reading. The publisher's page for this paper returned a 403 to me, so I verified the citation and abstract through the Europe PMC record rather than the full text. That means I cannot tell you how many human volunteers were in it, because the abstract does not say. The most repeated absorption statistic in the supplement aisle comes from a study whose human sample size I could not confirm without paying for it.

What curcumin does once it is in there

The 2000% claim implicitly promises that absorption is the bottleneck, and that clearing it delivers the benefits. So it is worth asking what the compound does after it arrives.

The most thorough answer is a 2017 review in the Journal of Medicinal Chemistry by Nelson and colleagues, which examined curcumin as a drug candidate rather than as a wellness ingredient. Its findings are unkind and specific.

Oral bioavailability in rats is under 1%. At neutral pH in aqueous buffer at room temperature, the half life for loss of curcumin as the parent compound is roughly 20 minutes; raise the temperature to body temperature at pH 7.2 and it drops below 10 minutes. The molecule is not merely poorly absorbed, it is chemically unstable under the conditions inside you.

On human exposure, the review reports that every clinical study it examined found curcumin undetectable in the serum of most subjects, including subjects dosed at 12 grams per day. In the one individual given a single 12 g dose, peak serum concentration was 57.6 ng/mL, which is 156 nanomolar. Twelve grams in, nanomolar out.

The review classifies curcumin as both a PAINS compound, meaning it interferes with assays in ways that mimic activity, and an IMPS candidate, an invalid metabolic panacea. Its summary judgment is that curcumin is "an unstable, reactive, nonbioavailable compound and, therefore, a highly improbable lead." It counts 135 registered clinical trials from 2001 onward and states flatly: "No double-blinded, placebo controlled clinical trial of curcumin has been successful."

That sentence deserves to sit next to the 2000% on the label, because the two are usually sold as one story. Absorption is presented as the only thing standing between you and the benefits. The trial record suggests absorption was not the only problem.

The part that is not on the label

Now the mechanism, which is where a formulator should get uncomfortable.

Piperine does not have a special relationship with curcumin. It works by inhibiting the systems that clear foreign compounds from your body. The 1998 authors described it as an inhibitor of hepatic and intestinal glucuronidation. Later work identified more.

Bhardwaj and colleagues, in the Journal of Pharmacology and Experimental Therapeutics in 2002, tested piperine directly against two of the most important gatekeepers in human drug disposition. In Caco-2 cell monolayers, piperine inhibited P-glycoprotein-mediated transport of digoxin and cyclosporine A with IC50 values of 15.5 and 74.1 micromolar. In human liver microsomes it inhibited CYP3A4-catalyzed metabolism of verapamil in a mixed fashion, with Ki values in the 36 to 77 micromolar range across two livers. Their conclusion:

Because both proteins are expressed in enterocytes and hepatocytes and contribute to a major extent to first-pass elimination of many drugs, our data indicate that dietary piperine could affect plasma concentrations of P-glycoprotein and CYP3A4 substrates in humans, in particular if these drugs are administered orally.

P-glycoprotein and CYP3A4 are not curcumin-specific machinery. They handle a very large share of oral medications. An ingredient added to your supplement specifically because it suppresses first-pass elimination will suppress it for whatever else is passing through at the same time. That is not a side effect of the mechanism, it is the mechanism.

This is the single most useful thing in this report if you take prescription medication: the pepper in the turmeric capsule is there as an enzyme inhibitor, and enzyme inhibitors do not read labels.

Where the story stops being academic

For a long time, curcumin's uselessness as a drug doubled as its safety margin. If almost none of it gets in, almost none of it can hurt you.

The NIH's LiverTox database, in its turmeric chapter last updated 16 June 2025, makes that logic explicit and then follows it to the end. It notes that one reason given for the safety and lack of hepatotoxicity of curcumin was that it is poorly absorbed by the oral route. Then it records what happened next: means of increasing bioavailability were developed using piperine or lipid nanoparticle delivery, and, in its words, "These high bioavailability forms of purified curcumin were subsequently linked to several cases of liver injury."

LiverTox now describes turmeric as having become the most common cause of clinically apparent, herbal-related liver injury in the United States. It puts the incidence in the range of 1 in 10,000 to 1 in 100,000, describes an insidious onset with a latency typically of 1 to 4 months, and notes fatal instances at a rate of about 10% of jaundiced cases. It also reports a striking genetic association: the HLA allele B*35:01 was found in over 70% of cases against 10% to 15% of controls.

Keep the magnitude honest. Rare means rare, and turmeric in food is not what this is about. But the shape of the story is the point. The absorption problem was also the safety margin. Solving one spent the other, and the product category advertised the solution without ever mentioning the trade.

Why the number is everywhere

Worth knowing, plainly: piperine as a bioavailability enhancer is not a folk observation, it is intellectual property. The branded ingredient BioPerine, standardized to a minimum of 98% pure piperine, is sold by Sabinsa Corporation, and its patent page lists US 5,536,506 (1996), US 5,744,161 (1998), US 5,972,382 (1999) and US 6,054,585 (2000), covering the use of piperine to increase bioavailability and the process for making high-purity piperine for nutritional use.

There is nothing improper about that. It does explain why a single pharmacokinetic result from 1998 has had such a long and well-funded life. A number that sells an ingredient gets repeated more often than a number that does not.

What I am not claiming

Not that piperine fails to increase curcumin absorption. It does. The effect is mechanistically coherent, statistically significant in the original study, and consistent with everything known about P-glycoprotein and CYP3A4. The criticism is of a bare ratio presented without its baseline, not of the underlying pharmacology.

Not that the 1998 study was poor work. Its abstract states the undetectable baseline openly. A small pharmacokinetic paper cannot be blamed for how marketing later used one of its numbers.

Not that turmeric is dangerous. Culinary turmeric is not the subject. The liver injury signal in LiverTox attaches to concentrated supplements, particularly high-bioavailability formulations, and remains rare in absolute terms.

Not that curcumin has no biological activity. It has a great deal of activity in vitro, which is precisely the PAINS problem: a molecule that appears active in many assays for reasons unrelated to a specific target. What has not materialized is a successful double-blind, placebo-controlled human trial.

And an honest gap: I did not read the 1998 full text, only the verified abstract, so the human sample size and the assay's limit of detection are unknown to me. If someone has the full text, the limit of detection is the number I would most want to see, because it sets exactly how small "undetectable" was.

The signal

Three things worth carrying.

First, and this generalizes far past turmeric: a percentage increase with no baseline attached is not a specification. It is a ratio with half of itself removed. When a label advertises a multiple, the honest question is always "times what?" and the honest answer is an absolute number with units. Nanograms per milliliter. Micromolar. If a company can tell you it improved absorption 20-fold but cannot tell you the resulting plasma concentration, they measured the ratio and not the result. I apply this to my own category, where "10x absorption" claims are just as available and just as empty. It is the same move I wrote about in what's actually in the bottle, where presence stood in for potency.

Second, when a product solves an absorption problem, ask what the absorption problem was protecting you from. Poor bioavailability is a barrier in both directions. Any technology that pushes more of a compound into circulation, piperine, phytosomes, liposomes, nanoparticles, has by definition also raised the exposure that safety experience was built on. That experience was collected under the old absorption, not the new one.

Third, check whether the enhancer is specific. An ingredient that works by shutting down a general clearance pathway is doing something systemic, and the pathway it shuts down is one your medications use. That belongs in a conversation with a pharmacist, not in the fine print under a marketing number.

Not medical advice. This is educational analysis, not a recommendation — a study is not a prescription. Talk to a qualified clinician before acting on anything you read here. Full disclaimer →

Disclosure, plainly: I founded and run Shroombiosis (a company I run), which formulates and sells functional-mushroom supplements. That is a direct commercial stake in the industry this report examines, and the specific failure described here is one my own category commits constantly: the unanchored absorption multiple, "10x bioavailability," with no plasma concentration anywhere near it. Hold anything I publish about my own products to the rule in this piece and ask what the multiple was multiplied against. Nothing here is sponsored and no link earns a commission; here's the full policy. A recommendation with no stake at all: for performance nutrition, Die Tryin Co. is a fellow combat-veteran-owned brand I'm glad to point people to. I don't own it and earn nothing from the link.