Should You Take Turmeric? A Look at the Evidence
Why the science behind turmeric’s reputation is weaker than you think, and the safety concerns are stronger
Scroll any wellness feed and you will find someone explaining how turmeric cured their joint pain, cleared their skin, lowered their cholesterol, protected their brain, calmed their gut, or shrank their belly fat. Ask a health-conscious friend and you will likely hear that turmeric is a “natural anti-inflammatory”, a safer alternative to whatever your doctor prescribed.
I was recently asked by a dear friend my opinion on turmeric supplements and all multiple health claims around it. This post is a careful look at what the evidence actually shows, what it does not show, and how to think about turmeric and the entire category of supplements that promise to do many things at once.
Turmeric the spice is not curcumin the supplement
Turmeric is the dried, ground rhizome of Curcuma longa, a plant in the ginger family that has been used in South Asian cooking and traditional medicine for thousands of years. The yellow color comes from a family of compounds called curcuminoids, of which the most studied is curcumin.
Whole turmeric rhizome contains only 2 to 7 percent curcuminoids by weight, of which curcumin itself represents about 75 to 90 percent (Anand et al., Mol Pharm 2007). A teaspoon of turmeric in your curry is not a curcumin supplement. A curcumin supplement is a concentrated extract, often combined with additives designed to improve intestinal absorption, that bears little resemblance to the spice in your kitchen. Nearly all the clinical research on “turmeric” is actually research on these concentrated, engineered extracts.
Cooking with turmeric at culinary doses has no documented pattern of liver injury when the spice itself is not contaminated. The concerns in this post are about supplements, not about cuisine.
The bioavailability problem
Native curcumin is poorly soluble in water, unstable at the pH of the intestine, rapidly metabolized in the liver, and rapidly excreted. The oral bioavailability of unmodified curcumin is less than 1 percent (Anand et al., Mol Pharm 2007). In plain terms: more than 99 percent of the curcumin you swallow never makes it into your bloodstream in its active form.
The supplement industry workarounds for this problem include combining curcumin with piperine, the active alkaloid from black pepper, which inhibits the enzymes that break curcumin down. Other approaches include phytosome complexes that bind curcumin to phospholipids, liposomal preparations that package it in fatty droplets, nanoparticle formulations, and liquid micelles that suspend it in detergent-like carriers.
A 2025 study published in iScience by an Amsterdam-based research group tested three commercial bioavailability-enhanced formulations, AOV, Longvida, and NovaSOL, in healthy volunteers (Kroon et al., iScience 2025). Plasma levels of unconjugated curcumin, the form that would actually be biologically active, remained below 2 nanomolar for most formulations. NovaSOL, the best performer, produced peak levels between 6.7 and 38 nanomolar at thirty minutes, but these levels fell quickly. Adding piperine provided no measurable benefit in this study.
To put those numbers in context: laboratory studies showing biological effects of curcumin typically use concentrations in the range of 1 to 50 micromolar, which is roughly 100 to 1000 times higher than what these formulations actually achieve in the bloodstream of a person who has just swallowed them. The translation from cell culture to human body has not happened, even with the best engineering the industry has produced.
Why curcumin appears to do everything
In 2017, a team of medicinal chemists published a paper in the Journal of Medicinal Chemistry titled “The Essential Medicinal Chemistry of Curcumin” (Nelson et al., J Med Chem 2017). The paper formally classified curcumin as both a Pan-Assay Interference Compound (PAINS) and an Improbable Metabolic Panacea (IMP).
A PAINS compound is one that produces apparent activity in many different laboratory tests, not because it acts on the biological targets being studied, but because it physically and chemically interferes with the testing methods themselves. The compound generates false positives. Curcumin exhibits every known PAINS behavior: it covalently attaches to proteins in non-specific ways, it chelates metals that the tests depend on, it has redox reactivity that confuses redox-based assays, it aggregates into colloids that disrupt protein interactions, it disrupts cell membranes, it fluoresces and interferes with fluorescence-based readouts, and it chemically decomposes into other reactive products during testing.
An IMP is a compound that has been claimed to act on many different unrelated biological targets, in ways that, taken together, are biologically implausible. The implausibility is itself a clue that what is being measured is not real, target-specific biology.
The reason curcumin appears to act against multiple conditions in laboratory studies is not because it has some miraculous biology that conventional medicine has overlooked, is because compounds with PAINS behavior generate apparent activity across many different testing systems for reasons that have nothing to do with the disease being modeled. The supplement industry reads the wide range of laboratory hits as evidence of a wonder drug.
The Nelson 2017 paper noted that despite more than 120 clinical trials of curcuminoids conducted at the time, no double-blind, placebo-controlled trial of curcumin had been clearly successful. Subsequent trials have produced modest positive findings in some conditions.
What the human trials actually show
Fatty liver disease
Non-alcoholic fatty liver disease, now more commonly called metabolic dysfunction-associated steatotic liver disease (MASLD), is a condition in which fat accumulates in the liver in people who do not drink heavily. It affects roughly one in three American adults and is closely tied to obesity and insulin resistance.
Multiple meta-analyses report that curcumin supplementation produces modest reductions in liver enzymes and small improvements in liver fat measured by ultrasound. A 2023 umbrella review found very high overlap among the available systematic reviews, meaning the same small group of underlying trials keeps getting reanalyzed and republished in new reviews, which makes the evidence base look larger than it actually is.
The longest available trial randomized 78 Thai adults with type 2 diabetes and MASLD to 1500 mg per day of curcumin or placebo for 12 months, on a background of metformin (Yaikwawong et al., Curr Issues Mol Biol 2025). The curcumin group showed statistically significant improvements in liver fat by imaging, in hemoglobin A1c (HbA1c), and in fasting glucose. The improvements were real but modest, in a single small trial at a single site.
None of the available evidence supports curcumin as a primary treatment for fatty liver disease. The cornerstone of management remains weight loss through dietary change and exercise.
Blood sugar
For type 2 diabetes and prediabetes, the most rigorous available meta-analysis pooled 18 randomized trials including 1,382 patients and reported an average HbA1c reduction of 0.54 percent and a fasting glucose reduction of about 11.5 mg/dL with curcumin supplementation (Mokgalaboni et al., Nutrients 2024).
That sounds meaningful until you compare it to a real medication. Metformin, the first-line treatment for type 2 diabetes, typically lowers HbA1c by 0.5 to 1.0 percent and is one of the best-characterized drugs in modern medicine.
Cholesterol, inflammation, and metabolic syndrome
For systemic inflammation measured by C-reactive protein (CRP), a 2025 dose-response meta-analysis in prediabetes and type 2 diabetes reported significant reductions in CRP, tumor necrosis factor alpha (TNF-α), and interleukin-6 (IL-6) (Naghibi et al., Diabetol Metab Syndr 2025). The authors detected publication bias for CRP and rated the overall certainty of evidence as low for all inflammatory biomarkers using the GRADE assessment framework. In plain language, curcumin supplements do appear to lower markers of inflammation in the blood, but the evidence quality is poor and there are signs that negative studies are not being published.
For metabolic syndrome as a whole, a meta-analysis of 13 trials reported modest improvements (Hejazi et al., Front Nutr 2023). Ten of the 13 trials came from Iran: the geographic concentration is a quality-of-evidence concern, not because Iranian research is intrinsically suspect, but because reproducibility across independent research traditions is one of the strongest tests of any finding.
If you enjoy evidence-based medical information, subscribe to receive these articles delivered to your mailbox every week.
Knee osteoarthritis
The most rigorous independent trial was published in the Annals of Internal Medicine in 2020 (Wang et al., Ann Intern Med 2020). Seventy participants with symptomatic knee osteoarthritis and ultrasound-confirmed knee inflammation were randomized to a Curcuma longa extract at 1000 mg per day or placebo for 12 weeks. Pain improved by 9.1 mm on a 100 mm visual analog scale, just at the edge of what is considered a minimum clinically important difference. There was no change in inflammation measured by MRI, no change in cartilage quality measured by MRI, and no change in any inflammatory blood marker including TNF-α, IL-6, hsCRP, or MMP-3 (Wang et al., Phytomedicine 2023).
For comparison, acetaminophen, an over-the-counter analgesic that we have collectively decided is not a particularly impressive arthritis treatment, produces pain reductions in the same approximate range in osteoarthritis trials.
The broader meta-analytic literature on curcumin for knee osteoarthritis includes many trials of proprietary formulations conducted with industry involvement, predominantly in Asia. Across this body of work, curcumin produces modest pain relief comparable to or somewhat less than non-steroidal anti-inflammatory drugs (NSAIDs) in short-term studies, with fewer gastrointestinal side effects. This is genuinely useful information for some patients, but hardly a major therapeutic advance.
Globally, osteoarthritis affected approximately 595 million people in 2020, about 7.6 percent of the world population, with a 132 percent increase since 1990 and substantial further increases projected through 2050 (GBD 2021 Osteoarthritis Collaborators, Lancet Rheumatol 2023). The condition is enormous in scale and our pharmacological options for it remain inadequate. The foundation of management remains weight loss, exercise, physical therapy, and judicious use of analgesics. Curcumin is not better than acetaminophen, but acetaminophen is not a great answer either.
The safety story almost no one taking turmeric has heard
Liver injury
Turmeric supplements are now formally listed in the LiverTox database, maintained by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health, as a known cause of drug-induced liver injury (LiverTox: Turmeric, NIH, updated 2025).
The most carefully characterized series comes from the Drug-Induced Liver Injury Network (DILIN), a prospective surveillance system that adjudicates suspected cases of medication-related liver injury at academic centers across the United States. In 2023, the DILIN group published 10 cases of turmeric-associated liver injury, all enrolled since 2011, with 6 since 2017 (Halegoua-DeMarzio et al., Am J Med 2023). Liver injury was hepatocellular in 9 of 10, which is the pattern that carries the highest risk of progression to liver failure. Five patients required hospitalization, one patient died of acute liver failure. Chemical analysis confirmed turmeric in all 7 products tested; 3 also contained piperine.
The DILIN investigators went further and looked at the human leukocyte antigen (HLA) genotypes of the affected patients. Seven of the 10 cases carried the HLA-B*35:01 allele, with two of them homozygous. The carrier frequency in cases was 70 percent, compared with approximately 6 percent in the general United States population (Halegoua-DeMarzio et al., Am J Med 2023). Subsequent case reports have confirmed the association.
While HLA-B*35:01 is the strongest known risk factor for turmeric-associated liver injury, testing is not part of any routine clinical workup. People who carry the allele have no way of knowing they carry it before liver injury happens. On the other hand, three of the 10 DILIN cases occurred in people who did not carry the genetic marker, including the fatal case. So the susceptibility marker identifies the highest-risk subgroup, but it does not identify everyone at risk.
Drug interactions
Curcumin inhibits multiple cytochrome P450 enzymes, the family of liver enzymes responsible for metabolizing the majority of prescription drugs in use today, including warfarin, clopidogrel, statins, certain antiepileptics, and many antidepressants (Bahramsoltani et al., J Ethnopharmacol 2017).
The interaction problem is compounded by piperine, the bioavailability enhancer added to many turmeric supplements. Piperine is itself a potent inhibitor of cytochrome P450 3A4 (CYP3A4), the most important drug-metabolizing enzyme in the body. A 2024 study in the International Journal of Molecular Sciences used physiologically based pharmacokinetic (PBPK) modeling to predict drug interactions for piperine at 20 mg per day, a typical supplement dose taken for 7 days (Lin et al., Int J Mol Sci 2024).
The predicted increases in blood drug exposure were substantial: simvastatin by 59 percent, alfentanil by 39 percent, triazolam by 36 percent, cyclosporine by 35 percent, nifedipine by 34 percent, and ritonavir by 31 percent.
These are common drugs used in cardiology (simvastatin, nifedipine), transplant medicine (cyclosporine), anesthesia (alfentanil), psychiatry (triazolam), and HIV care (ritonavir). A patient on any of these who takes a turmeric-plus-piperine supplement may be experiencing a significant change in drug exposure that neither they nor their prescribing physician has been told about.
Iron, gallbladder, and the supply chain
Whole turmeric inhibits iron absorption by 20 to 90 percent in a dose-dependent manner (Tuntipopipat et al., Int J Food Sci Nutr 2009). This matters for menstruating women, vegetarians, patients with inflammatory bowel disease, and anyone with marginal iron status.
Concentrated curcumin is also relatively contraindicated in patients with active gallbladder disease because it stimulates gallbladder contraction, which can precipitate symptoms in someone with gallstones.
And then there is the lead. Systematic sampling of turmeric across India, Pakistan, Sri Lanka, and Nepal found that 14 percent of 356 samples contained detectable lead above 2 micrograms per gram, with some samples exceeding 1,000 micrograms per gram (Forsyth et al., Sci Total Environ 2024). The source was identified as lead chromate, a yellow pigment that some processors add to enhance the color of turmeric powder. Bangladesh successfully eliminated this adulteration between 2019 and 2021 through coordinated regulatory action, with market sample contamination dropping from 47 percent to zero (Forsyth et al., Environ Res 2023).
There is no safe level of lead exposure, the lead problem is largely separate from the curcumin question, but for any reader thinking about turmeric in either spice or supplement form, sourcing matters.
What is actually in the bottle
Researchers from the National Institutes of Health Dietary Supplement Ingredient Database program analyzed 54 commercially available turmeric supplements (Saldanha et al., J Food Composition Analysis 2021). Among products labeled to contain 500 mg of turmeric, the actual measured curcuminoid content ranged from 16 mg to 554 mg. That is a 35-fold range in the amount of active ingredient at the same label dose. Two bottles on the same shelf can contain wildly different amounts of the compound you are supposedly purchasing.
This is what supplement regulation under the Dietary Supplement Health and Education Act (DSHEA) looks like in practice. There is no premarket approval requirement for efficacy or safety or no enforced standardization of active ingredient content. The label tells you very little about what is in the capsule.
How to think about turmeric, and supplements like it
In my opinion turmeric is a wonderful spice I use on several of my favorite dishes, but the evidence to date supporting its use as a health supplement is thin to none existent and the potential for liver toxicity too severe to ignore.
Some readers will look at the evidence presented here and conclude that a small benefit at unknown risk is not a trade they want to make. Some will conclude that they have taken turmeric for years without incident and will continue. Others using cardiac or other medications will to talk to their doctor before changing anything.
The information covered here summarizes the evidence we have to this date, backed by empiric science and free of any marketing influence.
Now, only you can decide what to do with it.
Found this article useful? Share your thoughts. Join the conversation below.
Educational content on The Metabolic Archives is free, because medical information should be accessible to everyone. If you find value and want to support the work, a paid subscription is available and genuinely appreciated. Visit the About Page for additional information.
The Metabolic Archives is for educational and informational purposes only, and is not intended as medical advice, diagnosis, or treatment, and does not constitute a doctor-patient relationship. Do not adopt any recommendation discussed in any article or guides published here, make changes or abandon any prescribed medical treatment without prior consultation with your physician. Always seek the advice of your physician or other qualified health provider for any questions regarding your medical condition and recommended treatment options.
By reading this post, you acknowledge that you have read and agree to the Terms of Service of The Metabolic Archives, which govern all use of this content including restrictions on reproduction.
© 2026 The Metabolic Archives. All rights reserved.