Turmeric’s golden color comes from curcuminoids, a family of yellow compounds, and curcumin is the best known. It is a small part of the spice. Researchers tested 28 products sold as turmeric or curry powder. Pure turmeric powder averaged 3.14% curcumin by weight, and the curry powders held far less.
So a level teaspoon (3 g) of turmeric holds roughly 95 mg of curcumin. The trials below used hundreds to thousands of milligrams.
Curcumin from plain turmeric is poorly absorbed, and that shapes all the research. In volunteers who took 2 g of curcumin alone, blood levels were undetectable or very low. Adding 20 mg of piperine, the compound in black pepper, raised absorption by 2000%. Curcumin also breaks down fast at body pH (the body’s acidity level) in the lab.
Eating it inside turmeric helps. In a meal study, turmeric powder and fresh root gave far higher blood curcumin than pure curcumin powder with the same curcumin dose. Even so, levels still stayed low.
A teaspoon of turmeric (3 g) has 9 calories, 0.7 g of fiber and 0.3 g of protein. It also gives 26% of the Daily Value (the daily target on US food labels) for manganese. But you eat it for color and flavor, not nutrients.
Almost all the health research tests curcumin, turmeric extracts or supplements, often with piperine or special formulations to aid absorption. That is the evidence below, and it says little about a pinch in your curry.
Health benefits of turmeric
The top findings from 28 studies, strongest evidence first
What the stars mean
- Strong. Several randomized trials, pooled in a meta-analysis, point the same way, or regulators (FDA or EFSA) have approved the claim.
- Moderate. Good randomized trials support it, but there are fewer of them, or their results vary.
- Limited. Small, short or single trials, or mixed results. Promising, but it needs confirmation.
Population and lab studies get no stars: they show a link or a mechanism, not proof.
How we rate evidence →- ★★Less osteoarthritis pain than with placeboPooled trials found turmeric extract or curcumin eased osteoarthritis pain, stiffness and function more than placebo (15 trials, 1,621 people). Against painkillers such as ibuprofen, two reviews found similar relief, one found less and one found more.
- ★★Slightly lower blood sugarIn pooled trials of supplements, fasting blood sugar went down a little (4.6 mg/dL) and HbA1c, average blood sugar, fell by 0.32%. This came from turmeric and curcumin supplements, not cooking with the spice.
- ★★Slightly lower LDL cholesterolLDL cholesterol fell a little (by about 5 mg/dL) across 64 trials of turmeric and curcumin supplements. The trials differed a lot and the evidence was low or very low in certainty.
- ★★Lower blood marker of inflammation (CRP)The inflammation marker CRP fell (by 3.67 mg/L) with curcumin in people with inflammatory conditions. A lower marker is not the same as less disease.
- ★★Better chance of remission in ulcerative colitisAdding curcumin to usual treatment made clinical remission about 2.3 times as likely, across 8 trials in 482 patients with ulcerative colitis. Results varied a lot between trials.
All 28 studies, strongest evidence first
Across 103 trials, curcumin supplements changed 23 of 42 health measures
This meta-analysis (a study that pools earlier trials) combined 103 randomized trials in 7,216 people. They tested curcumin and Curcuma longa (turmeric) products against a placebo (dummy pill). Certainty was rated high for fasting blood sugar, C-reactive protein (CRP), HDL (“good”) cholesterol and weight. It was moderate, low or very low for the rest, and the authors point to weak methods in many of the trials.
Turmeric extract and curcumin beat placebo for osteoarthritis pain, stiffness and function
This meta-analysis pooled 15 randomized trials in 1,621 people with osteoarthritis. Against anti-inflammatory painkillers (NSAIDs, such as ibuprofen), the effects were similar and side effects were fewer. The authors suggest using the supplement for more than 12 weeks.
Turmeric extract beat placebo for knee osteoarthritis pain, by a noticeable margin
This meta-analysis pooled 10 randomized trials in 783 patients, covering all types of Curcuma longa extract. Even the smallest estimated effects on pain and on the WOMAC total score (a knee arthritis scale) were larger than the smallest change patients notice. The trials differed from each other, and the authors call for larger, high-quality trials.
Curcuminoids beat placebo for knee osteoarthritis, but the trials were weak
This meta-analysis pooled 11 randomized trials in 1,009 people, of curcuminoid or Boswellia products, for knee pain and function. Quality was low overall, and most trials had fewer than 100 people. Curcuminoids did not differ from NSAIDs in effect and caused fewer stomach side effects. The authors say the evidence is too small and weak for firm clinical advice.
Curcuminoids eased knee pain more than placebo, less than ibuprofen
This meta-analysis pooled 7 randomized trials in 797 people, all run in Asia. Quality of life improved and people used less rescue medication (extra painkillers), with no serious side effects. The trials were small, varied widely and were reported unevenly, so the authors call for more trials.
Low and high doses of curcuminoids eased knee osteoarthritis pain about equally
This meta-analysis pooled 11 randomized trials in 1,258 people. Doses of 1,000 mg a day or more gave no more pain relief than lower doses. Unlike the 2017 review above, this one found curcuminoids relieved pain better than NSAIDs, and the NSAIDs had higher rates of side effects.
Turmeric and curcumin supplements slightly lowered blood sugar and HbA1c
This meta-analysis pooled 59 randomized trials. Fasting blood sugar fell by 4.6 mg/dL (milligrams per deciliter of blood) and HbA1c (long-term average blood sugar) by 0.32%. Fasting insulin fell by 0.87 µIU/mL and insulin resistance (HOMA-IR) by 0.33. These are small drops, and they come from supplements, not from cooking with the spice.
Curcumin lowered CRP, a blood marker of inflammation, more than placebo
This meta-analysis pooled randomized trials in people with inflammatory conditions. CRP fell by 3.67 mg/L (9 trials measured it), and high-sensitivity CRP was also lower (23 trials). The effect showed at doses of 1,000 mg a day or less. A lower marker is not the same as less disease.
Turmeric and curcumin supplements slightly lowered LDL cholesterol
This meta-analysis pooled 64 randomized trials. LDL cholesterol fell by about 5 mg/dL and total cholesterol by 3.99 mg/dL. Triglycerides (blood fats) fell 6.69 mg/dL, while HDL rose 1.80 mg/dL. The trials differed a lot and the evidence was low or very low in certainty, so the authors urge caution.
Curcumin with colitis treatment made clinical remission about 2.3 times as likely
This meta-analysis pooled 8 randomized trials in 482 patients with ulcerative colitis, who added curcumin to their usual treatment. Remission seen on a camera exam of the gut leaned the same way, but that result could have been chance. Results varied a lot between trials, and no serious side effects were reported.
Curcumin eased depression scores modestly in 6 small trials
This meta-analysis pooled 6 trials in 377 patients, against placebo. Hamilton depression scores fell by a standardized 0.34, and 3 of the trials also found less anxiety. The trials lasted 4 to 8 weeks, so long-term effects are unknown.
Curcumin lowered the liver enzymes ALT and AST in fatty liver disease
This meta-analysis pooled 9 randomized trials in non-alcoholic fatty liver disease. Total and LDL cholesterol, fasting blood sugar and waist size also fell. Triglycerides, body weight and BMI (body mass index) did not change.
Turmeric in a rice meal did not block iron absorption; chili did
Each crossover experiment had 10 healthy young women, who ate every test meal; the iron in it was labeled so it could be traced. The turmeric dose was 0.5 g of powder, about a sixth of a teaspoon. Chili (4.2 g) cut iron absorption by 38%, from 9.7% to 6.0%. Each test was one meal at a time, in small groups.
Turmeric in a meal gave far more blood curcumin than pure curcumin
In this crossover study, healthy men ate meals of mashed potato and cream. Each meal held 400 mg of curcumin, as pure powder, turmeric powder or grated fresh turmeric. Peak blood curcumin was 8.4 ng/ml with turmeric powder, 4.9 ng/ml with fresh turmeric and 0.19 ng/ml with pure curcumin. At the average curcumin content of store powder, 400 mg is about four teaspoons.
Turmeric powder capsules lowered blood glucose, insulin and leptin in fatty liver
In this double-blind trial, 46 people with non-alcoholic fatty liver disease took part for 12 weeks. Half took six 500 mg capsules of turmeric powder a day, 3 g or about a teaspoon, and half took a placebo. Against placebo, glucose fell 1.22%, insulin 17.69%, HOMA-IR (insulin resistance) 19.48% and leptin (an appetite hormone) 21.33%. Weight, BMI and liver enzymes did not clearly change.
Fermented turmeric powder lowered liver enzyme ALT in people with raised levels
This double-blind trial gave 3 g of fermented turmeric powder a day for 12 weeks. It started with 60 people, and 48 were analyzed (26 on turmeric, 22 on placebo). AST fell too, and no severe side effects were reported. This was fermented turmeric in capsules, not the plain kitchen spice.
A turmeric extract cut knee pain more than placebo at both doses
In this double-blind trial, 150 people with knee osteoarthritis took a bio-optimized turmeric extract or a placebo for 3 months. Pain fell by 29.5 and 36.5 mm at the two doses, against 8 mm on placebo, scored on a pain line in millimeters. Patients rated their disease activity better on both doses, but a blood marker of cartilage breakdown did not differ between groups. Product-related side effects were more common at the high dose than on placebo.
A turmeric preparation eased indigestion about as well as omeprazole
In this randomized, double-blind trial, 132 people had indigestion with no clear cause. In all, 45 took Curcuma longa, 43 took omeprazole (an acid-lowering drug) and 44 took a placebo, for 4 weeks. Pain scores fell more than on placebo and did not differ from omeprazole. Quality-of-life scores did not clearly differ between groups.
Prediabetes: none on curcumin got type 2 diabetes, against 16.4% on placebo
In this double-blind trial, 240 people with prediabetes took curcumin capsules or a placebo for 9 months. Beta-cell (insulin-making cell) function, insulin resistance (HOMA-IR 3.22 against 4.04) and adiponectin also looked better on curcumin. Side effects were very minor. It is a single trial and needs repeating.
Theracurmin, a well-absorbed curcumin, improved memory and attention over 18 months
In this double-blind trial, 40 adults aged 51 to 84 without dementia took part. In all, 21 took 90 mg of curcumin twice a day and 19 took a placebo. Brain scans in 15 people in each group suggested less amyloid and tau signal (protein deposits linked to Alzheimer’s disease) in two regions. But in one region the difference between groups could have been chance, and the trial is small.
Curcumin did not slow cognitive decline in older adults, except one test
In all, 96 older adults living at home took 1,500 mg a day of a curcumin formulation or a placebo for 12 months. Neither they nor the researchers knew who got which. Only the Montreal Cognitive Assessment differed: the placebo group declined at 6 months while the curcumin group did not. All other clinical and cognitive (thinking) measures were the same.
Curcumin did not improve Alzheimer’s symptoms or markers in 24 weeks
In this double-blind trial, 36 people with mild to moderate Alzheimer’s disease took 2 or 4 g of curcumin a day, and 30 finished. Blood curcumin was low (7.32 ng/mL), and 3 people on curcumin stopped because of stomach symptoms. The authors point to limited absorption.
Pooled curcumin trials helped thinking in older adults, not in Alzheimer’s
This meta-analysis pooled 6 trials in 289 people. Thinking scores improved in older adults (a standardized difference of 0.33). People with Alzheimer’s disease tended to do worse than on placebo. There were too few trials to check for publication bias (missing studies that found nothing).
Black pepper’s piperine raised curcumin absorption by 2000% in people
This study was in healthy volunteers and rats. After 2 g of curcumin alone, blood levels in people were undetectable or very low. With 20 mg of pure piperine they were much higher, with no adverse effects. It did not test a pinch of ground pepper.
Liquid micelle curcumin was absorbed about 185 times better than plain powder
In this crossover study, 13 women and 10 men took 500 mg of curcuminoids in each form. Micronized (finely ground) powder was absorbed about 9 times better than plain curcumin powder. Women absorbed curcumin better than men, and all safety measures stayed within the reference ranges.
Asians eating curry sometimes or often scored better on a thinking test
This population study asked 1,010 people aged 60 to 93 without dementia about curry. Those who ate it now and then or often did better on a memory and thinking test than those who rarely did. Curry is a spice mix, not turmeric alone. The authors call the evidence tentative, and people who eat curry may differ in other habits.
Frequent curry eaters in Singapore kept slightly better attention and language scores
This cohort study (it follows a group over time) tracked 2,734 adults aged 55 and older living at home for 4.5 years. Those who ate curry often or very often kept slightly better attention, language and spatial scores. The differences were small (effect sizes 0.130 to 0.186) and showed on only some tests. It covers curry, a spice mix, and links are not proof of cause.
Lab test: about 90% of curcumin broke down in 30 minutes
This test-tube study used a buffer at pH 7.2, close to body pH, and 37°C. Breakdown was faster at neutral to basic pH. Curcumin was steadier in blood and serum-containing medium: less than 20% broke down in 1 hour, and about half remained after 8 hours. This is chemistry, not a study in people.
Who should be careful with turmeric
Most people can eat turmeric freely in food. Take a little care if one of these applies:
- Turmeric and curcumin supplements. LiverTox, the liver-injury database of the NIH (the US National Institutes of Health), rates turmeric a well-documented cause of liver injury. This happens mostly with highly absorbed curcumin products, probably in 1 in 10,000 to 100,000 people exposed. In one series, liver tests returned to normal in all 11 patients after they stopped. See a clinician for dark urine, deep fatigue or yellow skin.
- Lead in cheap or imported powder. The FDA names turmeric among spices where lead-based dyes have been found. A 2024 study of 356 samples from South Asia found 14% with lead above 2 µg/g (micrograms per gram). Buy from brands that test for lead.
- Pregnancy and medicines. The NIH says turmeric supplements during pregnancy may be unsafe. Little is known about amounts above food levels while breastfeeding. If you take any medicine, ask your clinician before you start a turmeric supplement.
How to choose, store and eat turmeric
- In a meal: turmeric powder gave peak blood curcumin of 8.4 ng/ml and fresh root 4.9 ng/ml. Pure curcumin powder in the same meal of potato and cream gave 0.19 ng/ml.
- Turmeric, not curry: pure turmeric powder averaged 3.14% curcumin by weight across store brands, and curry powders held far less. A teaspoon (3 g) holds roughly 95 mg.
- Supplements: forms differ a great deal. Liquid micelle curcumin was absorbed about 185 times better than plain curcumin powder, and the liver cases mostly involved such highly absorbed products.
Do not choose turmeric by color alone. A study in Bangladesh found that polishers added a lead-based yellow pigment to dried roots. They did it to meet wholesalers’ demand for bright yellow roots.
So a vivid shade can be a warning sign, not a mark of quality. Look for a brand that says it tests for lead.
Sources
- Turmeric. LiverTox, NIDDK, NIH.
- Turmeric. National Center for Complementary and Integrative Health, NIH.
- Lombardi N, Crescioli G, Maggini V, et al. Acute liver injury following turmeric use in Tuscany: An analysis of the Italian Phytovigilance database and systematic review of case reports. Br J Clin Pharmacol. 2021.
- Papke DJ Jr, Viveiros K, Zota V, et al. Turmeric supplement-associated hepatitis: a clinicopathological series of 11 cases highlighting pan-lobular and zone 3 injury. Histopathology. 2025.
- Economically motivated adulteration (food fraud). FDA.
- Import Alert 99-42: detention without physical examination of foods due to heavy metal (toxic element) contamination. FDA.
- Forsyth JE, Mistree D, Nash E, et al. Evidence of turmeric adulteration with lead chromate across South Asia. Sci Total Environ. 2024.
- Forsyth JE, Nurunnahar S, Islam SS, et al. Turmeric means “yellow” in Bengali: Lead chromate pigments added to turmeric threaten public health across Bangladesh. Environ Res. 2019.
- Ahmed Nasef N, Loveday SM, Golding M, et al. Food matrix and co-presence of turmeric compounds influence bioavailability of curcumin in healthy humans. Food Funct. 2019.
- Schiborr C, Kocher A, Behnam D, et al. The oral bioavailability of curcumin from micronized powder and liquid micelles is significantly increased in healthy humans and differs between sexes. Mol Nutr Food Res. 2014.
- Tayyem RF, Heath DD, Al-Delaimy WK, Rock CL. Curcumin content of turmeric and curry powders. Nutr Cancer. 2006.
- Nelson KM, Dahlin JL, Bisson J, et al. The Essential Medicinal Chemistry of Curcumin. J Med Chem. 2017.
History of turmeric
Turmeric comes from the rhizome (underground stem) of Curcuma longa, a perennial in the ginger family that is native to tropical South Asia. Growers dig the rhizomes each year, boil or steam them, dry them and grind them to a yellow powder.
The name is traced to the Latin terra merita, “meritorious earth”, for the color of the ground spice. In North India it is haldi, from the Sanskrit haridra. In the south it is manjal, a word found in ancient Tamil literature.
One of the oldest traces of turmeric in cooking is more than 4,000 years old. Archaeologists studied starch grains at Farmana, a Harappan site in Haryana dated to about 2600 to 2200 BC. In a deep, narrow-mouthed cooking pot, they found starch matching cooked turmeric, and ginger starch.
A 2011 chapter on turmeric says its use goes back nearly 4,000 years to Vedic culture in India. It was a culinary spice with religious meaning.
It then traveled. The same chapter says turmeric probably reached China by 700 AD and East Africa by 800 AD. It probably reached West Africa by 1200 AD and Jamaica in the 18th century.
In 1280, Marco Polo wrote about a vegetable with qualities much like saffron. In medieval Europe it was called “Indian saffron”.
Wikipedia also lists earlier West Asian traces. A dye plant is named in 7th-century BC cuneiform medical texts from Ashurbanipal’s library at Nineveh. And turmeric was found in a merchant’s tomb at Megiddo that dates to the second millennium BC.
Turmeric also served as medicine and dye. Susruta’s Ayurvedic compendium, dated to 250 BC, recommended a turmeric ointment for the effects of poisoned food. Turmeric paste is part of the Hindu pre-wedding haldi ceremony, and the yellow dye colored the robes of Buddhist monks. The EU allows turmeric as the food color E100, often as a cheaper stand-in for saffron.
Turmeric paper, a strip dipped in the spice, turns from yellow to reddish brown in alkaline solutions. So it works as a simple pH test. A 2011 chapter said India grows nearly all of the world’s crop and eats 80% of it. The city of Erode in Tamil Nadu is known as “Turmeric City”.
Sources
- Prasad S, Aggarwal BB. Turmeric, the golden spice: from traditional medicine to modern medicine. In: Benzie IFF, Wachtel-Galor S, eds. Herbal Medicine: Biomolecular and Clinical Aspects. 2nd ed. CRC Press. 2011.
- Kashyap A, Weber S. Harappan plant use revealed by starch grains from Farmana, India. Antiquity. 2010.
- Turmeric. Wikipedia.
- Tayyem RF, Heath DD, Al-Delaimy WK, Rock CL. Curcumin content of turmeric and curry powders. Nutr Cancer. 2006.
- Shoba G, Joy D, Joseph T, et al. Influence of piperine on the pharmacokinetics of curcumin in animals and human volunteers. Planta Med. 1998.
- Ahmed Nasef N, Loveday SM, Golding M, et al. Food matrix and co-presence of turmeric compounds influence bioavailability of curcumin in healthy humans. Food Funct. 2019.
Frequently asked questions about turmeric
Is turmeric good for you?
As a spice, it is a safe flavoring that adds little nutrition. A teaspoon (3 g) has 9 calories, 0.7 g of fiber and 26% of the Daily Value for manganese. The health findings come from curcumin and extract supplements.
A pooled analysis of 103 trials found changes in 23 of 42 outcomes. But no finding here reaches strong evidence, and the supplements carry a rare risk of liver injury.
Is turmeric the same as curcumin?
No. Curcumin is the main yellow compound in turmeric. Pure turmeric powder averaged 3.14% curcumin by weight across 28 products, with curry powders far lower.
A teaspoon holds roughly 95 mg of curcumin. Most trials gave 500 mg to 4 g of curcumin, and one gave 90 mg twice a day in a better-absorbed form. So trial results with curcumin do not directly describe the spice in your kitchen.
How much turmeric should you take a day?
No study here sets an ideal amount. A meal study used 0.5 g of powder (about a sixth of a teaspoon). A fatty liver trial used 3 g of powder in capsules for 12 weeks. Curcumin trials ranged from 90 mg twice a day to 4 g a day.
The US National Institutes of Health says conventional turmeric supplements are likely safe in recommended amounts for 2 to 3 months. For supplements, ask your clinician first.
Should you take turmeric with black pepper?
It can help absorption of pure curcumin, but the evidence is for a drug-like dose. In volunteers who took 2 g of curcumin, 20 mg of pure piperine raised absorption by 2000%. No study here tested a pinch of ground pepper. And eating turmeric in a meal already gave much higher blood curcumin than pure curcumin powder.
Does turmeric help with inflammation and joint pain?
Supplements may ease knee osteoarthritis pain, but the trials are small. Pooled results found less pain than placebo in 10 trials of Curcuma longa extract. Curcumin also lowered the inflammation marker CRP by 3.67 mg/L.
Reviews disagree on whether curcuminoids beat anti-inflammatory painkillers or fall short of ibuprofen. The NIH calls the evidence initial.
Can turmeric cause liver damage?
Rarely, mainly with supplements. The NIH’s LiverTox database says turmeric is a well-documented cause of clinically apparent liver injury. It probably happens in 1 in 10,000 to 1 in 100,000 people exposed.
In a series of 11 patients, liver tests returned to normal after they stopped the supplement. Stop and see a clinician if you get fatigue, nausea, dark urine or jaundice (yellow skin or eyes).
Does turmeric lower cholesterol?
A little, according to pooled trials of supplements. In 64 trials, LDL cholesterol fell by about 5 mg/dL and total cholesterol by 3.99 mg/dL. The authors rated the evidence low or very low in certainty, so these are small and uncertain changes.
Can turmeric cause diarrhea or an upset stomach?
It can. The NIH lists nausea, vomiting, acid reflux, stomach upset, diarrhea and constipation as possible effects of oral turmeric. In an Alzheimer’s trial of 2 and 4 g of curcumin a day, 3 people stopped because of stomach symptoms. The doses in those trials were far above the roughly 95 mg of curcumin in a teaspoon of turmeric.