Turmeric and Curcumin: Benefits, Evidence, Doses and Safety

Turmeric is a yellow spice from the root of a ginger-family plant, and curcumin is its main active compound. Human studies, mostly on curcumin supplements, give the most promising signal for knee osteoarthritis pain. But the US National Center for Complementary and Integrative Health (NCCIH) says there isn’t enough evidence to conclude definitively that turmeric or curcumin helps any health condition. Most other claims you see online rest on small, short or mixed trials, or on lab and animal work.

Quick facts

Is there any scientific evidence that turmeric works?

There is a lot of research, but it is uneven. A 2023 review of clinical trials says almost 20,000 papers on curcumin and almost 7,000 on turmeric have been published. The same review points out that cell and animal results are often not reproducible in humans, which is why human trials are what count.

NCCIH, part of the US National Institutes of Health, says many studies have looked at turmeric and curcumin (NCCIH) for fatty liver disease, osteoarthritis and high cholesterol or other blood-fat problems. Its conclusion is that we don’t know enough to say definitively that either one helps any health purpose. It also notes that the evidence is hard to compare because oral curcumin products vary in how much curcumin they contain and often include other plant ingredients, such as piperine from black pepper.

What the human evidence shows, use by use

The uses below run from the most-studied to the least. Several reviews cited here are summaries by authors who also describe the products’ benefits in warm terms. The trial details are more useful than the conclusions. If terms like meta-analysis or hs-CRP are new, our guides on how meta-analyses combine studies and surrogate outcomes in trials explain them.

Knee osteoarthritis pain (strongest signal, still preliminary)

NCCIH says several meta-analyses have looked at oral turmeric or curcumin for knee pain, stiffness, joint strength and mobility. It calls the initial evidence positive and says higher-quality evidence is needed before firm conclusions. Individual trials described in a 2017 review in the journal Foods include:

• 40 people, 6 weeks, placebo-controlled: 500 mg a day of curcuminoids with 5 mg of piperine in each 500 mg dose. The treatment group had bigger drops in pain and function scores than placebo, but no significant change in the stiffness score.
• 50 people, 8 months: standard care plus a curcuminoid-phospholipid tablet (two 500 mg tablets daily) versus standard care alone. Pain, function and stiffness scores improved more with the add-on. Because both groups kept their usual care, this trial tells you about adding curcumin, not replacing anything.
• 367 people, 4 weeks, compared with a drug: 1,500 mg a day of a Curcuma domestica extract versus 1,200 mg a day of ibuprofen. Scores improved in both groups, and the extract met the study’s test for being no worse than ibuprofen. The ibuprofen group had more stomach and gut complaints. The limits: one trial, four weeks, and a no-worse-than test. It does not show the extract can stand in for a prescribed or recommended pain medicine, and it is not a reason to change any medicine. That decision belongs with your clinician.

The topical form is less clear. NCCIH says it’s unclear whether curcumin ointment helps knee osteoarthritis pain. Johns Hopkins Medicine’s dietitian notes that some people with osteoarthritis reported less joint pain when eating turmeric in recipes, but her page does not give trial details.

Fatty liver disease (NAFLD)

NCCIH says initial research suggests oral turmeric or curcumin might improve some fatty liver measures, but it is unclear which measures improve consistently. The 2023 review’s tables list small and mid-size trials reporting changes in liver enzymes, blood sugar and blood fats. One trial of 80 obese people with fatty liver used two 40 mg nanocurcumin capsules a day for 3 months and found lower inflammation markers. These are lab measures, not proof the liver disease itself improved.

Mouth sores from cancer treatment

NCCIH says initial research suggests oral or mouthwash forms might ease oral mucositis (painful swelling and irritation in the mouth) linked to cancer treatment. The trials are small, including pilot trials, so the finding is early.

Blood sugar, cholesterol, weight and inflammation markers

• Type 2 diabetes, inflammation markers: a 2026 meta-analysis in Nutrients pooled six randomized trials (472 people). Curcumin was linked to lower hs-CRP (an inflammation blood marker) by about 1.06 mg/L, but the result got less certain when some data had to be reconstructed. TNF-α, another marker, also dropped in five trials (475 people), though the trials differed widely from one another. The authors say the evidence base is small and call for larger, standardized trials.
• Weight, lipids and glucose: a 2026 systematic review of 105 studies covering several antioxidants linked curcumin to small drops: up to 0.82 kg in body weight and up to 0.30 kg/m2 in BMI, plus lower triglycerides, total cholesterol and fasting glucose. Those changes are small, and the review does not show better health outcomes such as fewer heart attacks.
• Metabolic syndrome: in the Foods review, a placebo-controlled trial of 117 people gave 1 g of curcumin plus 10 mg of piperine for 8 weeks and reported larger drops in several inflammation markers than placebo. In a crossover trial of 36 obese adults (1 g curcumin plus 10 mg piperine for 30 days), only triglycerides fell. Cholesterol, hs-CRP, BMI and body fat did not change significantly.
• Diabetes prevention: Johns Hopkins Medicine cites a 9-month study of people with prediabetes in which 16.4% of the placebo group developed type 2 diabetes and none of the curcumin group did. The Hopkins page doesn’t give the study’s size, and NCCIH’s overall view is that evidence is insufficient, so treat this as one early result.

Exercise soreness and recovery

A 2026 systematic review of 15 randomized trials found favorable effects in 6 of 7 studies on oxidative stress, 4 of 6 on muscle damage, 3 of 8 on inflammation, 3 of 7 on soreness or fatigue and 4 of 8 on performance. The reviewers rated certainty of evidence as low for oxidative stress and very low for the rest, because trials were small and differed in people, doses and forms. In one small trial from the Foods review, 28 untrained people took 400 mg a day for 2 days before and 4 days after hard exercise and had smaller rises in some blood markers but no less muscle soreness.

Enlarged prostate (BPH) symptoms

A 2026 meta-analysis pooled six studies (697 men) in which curcumin or placebo was added to standard BPH drugs. Symptom scores improved more with curcumin (a 4.11-point difference on the IPSS scale). The studies varied a lot, and the authors call for large randomized trials. This looked at adding curcumin to existing treatment, not replacing it.

Mood, anxiety and thinking

Evidence here is thin. In the Foods review, a 30-person crossover trial of obese adults (1 g a day of curcuminoids with piperine for 30 days) found lower anxiety scores but no change in depression scores. A 60-person trial of healthy adults aged 60 to 85 using about 80 mg of curcumin in a lipid formulation reported better attention and working memory after a single dose, and better working memory and mood after four weeks. Johns Hopkins Medicine says studies on mood disorders, depression and dementia are small.

Skin, women’s health and everything else

The 2023 review’s tables list trials in eczema, psoriasis, facial redness, period pain, hot flashes, irritable bowel symptoms and more, mostly small, and many using multi-ingredient or branded formulas. The tables report reductions in symptom scores but don’t let you judge how reliable each result is. NCCIH says there isn’t enough evidence to conclude whether oral turmeric or curcumin helps conditions beyond those above. Healthline’s claims about cancer prevention, Alzheimer’s disease and slowing aging rest largely on lab research and are not shown by human trials.

Tooth decay

The oral-health trials listed in the 2023 review looked at plaque, gum inflammation and gum pocket depth with turmeric mouthwash, gel and chips. Tooth decay itself is not among the listed outcomes, so no source here shows turmeric stops cavities.

Curcumin vs turmeric: what’s the difference?

Turmeric is the whole root and spice. It contains more than 200 compounds, per the 2023 review. Curcumin is the best-known one and gives turmeric its yellow color, and NCCIH says turmeric’s health activities are commonly attributed to the group of curcuminoids that includes curcumin. Spice powder is only a few percent curcumin (Harvard Health says 2% to 9%; the Ohio State dietitian says about 2%), while extract supplements can be about 95%. Most trials used curcumin or extract supplements rather than kitchen turmeric. The Ohio State dietitian says it’s hard to know whether you would get the same benefit from spice alone.

Doses used in studies

These are amounts researchers used, not recommendations.

Safety studies went much higher. The Foods review says trials reported good tolerability at 4,000 to 8,000 mg a day and up to 12,000 mg a day of a 95% curcuminoid product. Harvard Health says turmeric and curcumin appear well tolerated up to about 8 grams a day. Even so, side effects occurred at high doses (below), and NCCIH’s safety statement covers conventional (not absorption-enhanced) products for up to 2 or 3 months.

Safety, side effects and interactions

Side effects

Johns Hopkins Medicine says food use is considered the safest route and most side effects are tied to the very high curcumin levels in pills, capsules and gummies. Its dietitian says supplements are probably not a good idea. Harvard Health and Ohio State are more open to supplements with a doctor’s go-ahead. That is where the sources split, so it’s a conversation to have with your own clinician.

Medicines that can interact

• Pain relievers: Johns Hopkins Medicine says turmeric supplements can lessen the effects of indomethacin, aspirin, ibuprofen or acetaminophen.
• Blood thinners and antiplatelets: Hopkins says curcumin supplements can raise bleeding risk with warfarin. Ohio State names aspirin, clopidogrel, enoxaparin and heparin as drugs curcumin can make more potent.
• Chemotherapy: Hopkins says to talk to your doctor first and especially avoid supplements with camptothecin, mechlorethamine, doxorubicin or cyclophosphamide.
• Tacrolimus (an immune-suppressing drug): Hopkins says high amounts of curcumin may increase its side effects.
• Diabetes medicines: Harvard Health says turmeric can lower blood sugar.

Vitamins: none of the pages names a specific vitamin that should not be taken with turmeric. Healthline lists iron deficiency among reasons to limit turmeric and speak with a doctor first. NCCIH’s general advice is that if you take any medicine, talk with your provider before using any herbal product, because some herbs and medicines interact in harmful ways.

Who should talk to a clinician first?

• Anyone taking blood thinners, antiplatelets, immune-suppressing drugs, diabetes medicines or chemotherapy.
• People with liver problems, gallbladder disease, kidney stones or a family history of them, or bleeding disorders or iron deficiency (Healthline, Hopkins, Harvard Health).
• People with hormone-sensitive cancers such as breast, ovarian or prostate cancer. Ohio State says a curcumin supplement could worsen the cancer or interfere with cancer medicines.
• Anyone pregnant or breastfeeding. NCCIH says turmeric supplements during pregnancy may be unsafe and little is known about amounts above food levels while breastfeeding.
• Anyone with surgery scheduled (Harvard Health).

What happens if you take turmeric every day?

As a food spice, the sources describe daily turmeric as low risk. Ohio State says a few teaspoons a day in food carry no risk for people without an allergy or severe food intolerance. For supplements, NCCIH’s safety statement for conventional products covers up to 2 or 3 months, and it does not describe longer use. Healthline says 10 g or less a day is unlikely to cause diarrhea, constipation or vomiting, but that page doesn’t distinguish spice from extract. Whether long-term daily supplements are beneficial or harmful is still open.

Is it better to take turmeric in the morning or at night?

The sources give no studies comparing times of day. They do say curcumin is fat-soluble. Harvard Health says pairing it with a meal that includes fats can help absorption, and Hopkins and Ohio State say black pepper helps too. A trial in the exercise review compared timing relative to workouts, but the reviewers found no clear pattern.

Why don’t doctors recommend turmeric more?

The sources point to a few reasons. NCCIH says the evidence isn’t strong enough to conclude benefit for any condition. Products differ in actual curcumin content, and supplement contents aren’t checked by the FDA before sale (NCCIH and Harvard Health). Absorption is poor unless the product is altered, and the altered, high-absorption forms have been linked to liver injury (NCCIH). There are also real drug interactions (Hopkins, Ohio State). None of this means turmeric as a spice is a problem; it means supplements deserve a clinician’s input.

What is still unknown

• Whether benefits seen in trials hold up in larger, longer, higher-quality studies (NCCIH says they need to).
• How much absorption matters. The Foods review says piperine raised curcumin bioavailability by 2,000%, yet NCCIH warns absorption-enhanced forms may harm the liver. More on this in our curcumin bioavailability and study limits guide.
• Which fatty liver measures truly improve, and whether lab changes mean fewer symptoms or complications.
• Long-term safety of daily supplements, and safety in pregnancy and breastfeeding.
• Whether kitchen turmeric gives the effects seen with concentrated extracts.

How to read a label that lists turmeric or curcumin

• Turmeric versus curcumin: ‘500 mg turmeric’ may hold only a small share of curcumin, since powder is a few percent. Look for the milligrams of curcumin or curcuminoids per serving, and the percentage standardized.
• Form: check for piperine/black pepper extract, phospholipid or lipid, or nano forms. These are the absorption-enhanced types NCCIH flags for liver reports, and trials of one form don’t carry over to another.
• Proprietary blends: if curcumin is lumped into a blend with a total weight only, you can’t tell how much you are getting, and you can’t compare it with trial doses. Ask the maker for the per-serving curcumin amount.
• Other ingredients: several trials combined curcumin with other plants (for example Boswellia), so results can’t be credited to curcumin alone.
• Accuracy: Harvard Health says supplement contents may be more or less than labeled. Our guide on ingredient evidence vs marketing claims covers how to weigh label promises against trial results.

Before you order

• Ask the maker how many milligrams of curcumin, not just turmeric, each serving holds.
• Ask whether the product uses an absorption-enhanced form, and whether it matches any form used in the trials above.
• Bring the full ingredient list to your clinician or pharmacist if you take any medicine.
• Know the liver warning signs NCCIH lists.

By ClinicalStudyConnect.com Research Desk