Turmeric and Curcumin: Benefits, Best Forms, Dosing, and Side Effects

Table of Contents

• Current view (reviewed 21 October 2026)
• Key safety points
• Brad's videos on this topic
• Overview
• Forms and Bioavailability
• Evidence for Benefits
• Recommended Dosing
• Safety and Side Effects
• Drug Interactions
• Dietary Sources
• References

Current view (reviewed 21 October 2026)

Brad tends to encourage his patients not to take curcumin supplements, because of problems with labelling, absorption and disappointing human trials (Brad, 9 Feb 2026). He says curcumin appears to work for specific inflammatory conditions, sometimes as well as medicines, and names ulcerative colitis, knee arthritis and chronic indigestion, along with prediabetes (Brad, 9 Feb 2026). Trials that found tentative benefit often used 500 to 2,000 mg of curcumin a day, and there is no standard guidance on the best dose for any condition (Brad, 9 Feb 2026). If you do try it, he advises checking the curcumin content rather than the turmeric root weight and choosing a product with a bioavailability enhancer; the study that cast doubt on piperine found NovaSOL the most effective at raising blood levels (Brad, 9 Feb 2026). Speak with your doctor before starting, because liver injury has been reported with curcumin supplements (Brad, 9 Feb 2026). If you take curcumin supplements, he advises regular blood tests to check your liver function (Brad, 13 Sep 2024).

Brad's videos on this topic

• 9 Feb 2026, Almost All Turmeric (Curcumin) Brands Are a Scam: most curcumin supplements are mislabelled or poorly absorbed, and Brad tends to steer patients away from them; trials suggest benefit in ulcerative colitis, knee arthritis, chronic indigestion and prediabetes; if you try it, check the curcumin content, choose a bioavailability enhancer (NovaSOL did best, and the evidence that piperine helps is weak) and talk to your doctor first because of liver injury reports.
• 13 Sep 2024, FAQ: What's the Difference Between Curcumin and Turmeric?: curcumin's effects on inflammation, weight, diabetes and dementia are small or unproven; the best evidence is for osteoarthritis pain, where a meta-analysis found benefit similar to NSAIDs; liver injury is rare but serious, so anyone taking it should have regular liver function tests. Superseded on piperine and on the overall advice: in February 2026 Brad said several analyses found piperine does not raise curcumin absorption, and that he tends to steer patients away from curcumin supplements. The advice on regular liver tests still stands.

1. Overview

Turmeric is a spice derived from the dried rhizome (underground stem) of the plant Curcuma longa, a member of the ginger family grown commercially in Southeast Asia and India ^[2]. Curcumin, a major component of turmeric, gives the spice its yellow color, and its health effects are commonly attributed to curcuminoids, which include curcumin ^[2].

Turmeric has been used for centuries in Ayurvedic, Chinese, Islamic, and Thai traditional medicine for conditions including indigestion, arthritis, abdominal pain, skin infections, liver disease, and the common cold ^[2]. Today, oral turmeric and curcumin supplements are promoted for osteoarthritis, depression, allergies, high cholesterol, inflammatory bowel disease, metabolic syndrome, and a wide range of other conditions ^[2].

However, a 2017 review by medicinal chemists raised doubts about the whole field. It argued that much of curcumin's apparent activity in laboratory tests is likely false, because curcumin interferes with the tests themselves; that curcumin is unstable, reactive and barely absorbed; and that no double-blind, placebo-controlled trial of curcumin had yet succeeded ^[3].

The NIH's National Center for Complementary and Integrative Health (NCCIH) states that conventionally formulated oral turmeric or curcumin is likely safe in recommended amounts for up to 2 to 3 months, but emphasizes that highly bioavailable formulations may harm the liver, and that higher-quality evidence is still needed to reach definitive conclusions about efficacy for most conditions ^[2].

2. Forms and Bioavailability

The Core Problem: Poor Natural Absorption

Standard curcumin has extremely low oral bioavailability. Taken with water alone, little of it reaches the blood, because the gut wall and liver break much of it down quickly ^{[4, 8]}. Unformulated curcumin also has a short half-life of approximately two hours ^[4]. Because curcumin is lipophilic (fat-soluble), taking it with foods containing fats or oils enhances absorption by stimulating bile production, which aids the absorption of fat-soluble compounds ^[4].

Turmeric Powder vs. Curcumin Extract

In a small crossover study, healthy men ate a meal of mashed potato and cream that held 400 mg of curcumin, given as curcumin powder, turmeric powder or grated fresh turmeric root. Both turmeric meals raised blood curcuminoids more than the curcumin powder did over the first 1 to 3 hours, and the turmeric powder meal gave the highest levels of two curcuminoids. The authors suggest that other compounds in turmeric, or the way the root holds curcumin, may explain the difference ^[5].

Bioavailability-Enhanced Formulations

Because plain curcumin is so poorly absorbed, manufacturers have developed many special formulations. A review of bioavailability studies in healthy volunteers found three that raised blood levels more than 100-fold compared with unformulated curcumin: NovaSol (185-fold), CurcuWin (136-fold) and Longvida (100-fold) ^[6]. In single studies, Meriva raised total curcuminoid absorption about 29-fold ^[10] and Theracurmin about 27-fold ^[13].

A 2021 crossover trial in 12 healthy adults compared a single 207 mg dose of curcumin in eight forms. Only micellar curcumin (NovaSol, 57-fold) and a gamma-cyclodextrin complex (30-fold) raised blood levels significantly above plain curcumin; both stayed more stable and better dissolved after digestion. Forms with turmeric oils, with piperine and other additives, as liposomes, as submicron particles or as phytosomes did not raise levels significantly in this study, and the authors judged that blocking curcumin's breakdown after absorption appears ineffective ^[7].

Comparison of Major Formulations

Fed versus fasted: A 2026 laboratory study of eight commercial turmeric supplements found that none released more than 40% of its curcuminoids. Release was lowest under fasted conditions and a little better under fed conditions, and a phytosome product released the most despite its lower dose ^[18].

Piperine (Black Pepper Extract)

Piperine blocks one of the main ways the liver and gut wall break curcumin down ^[8]. The commonly cited 20-fold increase in bioavailability comes from one early study in healthy volunteers ^[8]. Brad notes that several analyses have found piperine does not increase the amount of curcumin absorbed, and that the claim it does rests mostly on a single study from the 1990s ^[139]. Piperine might also change how the body handles some medicines (see Drug Interactions).

Cooking with Turmeric

Curcumin from turmeric cooked into everyday foods still reaches the blood, though mostly in a changed (conjugated) form. In a small study of volunteers who ate foods made with an amount of turmeric typical of daily use in some South Asian populations, curcumin conjugates were found in everyone's blood for at least 8 hours, while free curcumin showed up in only one person ^[19].

Fermented Turmeric

Fermented turmeric products are also sold. How well people absorb curcumin from them is unclear.

3. Evidence for Benefits

Osteoarthritis

Osteoarthritis is the condition with the most clinical evidence supporting curcumin use. Several meta-analyses have evaluated curcumin for knee osteoarthritis; the initial evidence for pain and function is positive, but higher-quality evidence is needed ^[2].

Curcuminoids (1,500 mg/day): In a small pilot trial in people with mild to moderate knee osteoarthritis, curcuminoids (1,500 mg a day in three doses) for 6 weeks improved pain and physical function scores more than placebo, though not stiffness ^[20].

Curcuminoid complex (500 mg three times daily): In a 12-week trial in 201 people with osteoarthritis, a curcuminoid extract (CuraMed, 333 mg curcuminoids per capsule) improved physical performance tests compared with placebo but not the joint pain index; a version combined with boswellic acid improved both ^[21].

Low-dose turmeric extract: In a 3-month trial in 150 people with knee osteoarthritis, a low and a high dose of a bio-optimized turmeric extract both cut pain more than placebo, by 29.5 mm and 36.5 mm on a pain scale, against 8 mm with placebo. The high dose caused more product-related side effects than placebo ^[22].

Curcugen (500 mg twice daily): In an 8-week trial in 101 adults with knee osteoarthritis, Curcugen reduced knee pain more than placebo and improved some walking and mobility tests, though not all. Pain medicine use fell in 37% of the curcumin group against 13% on placebo ^[24].

Added to standard treatment: In a trial in 160 people with knee osteoarthritis, turmeric extract added to their usual medicines improved pain and function scores more than placebo, and lowered blood markers of inflammation (IL-1β) and oxidative stress ^[25].

No benefit in older adults with overweight or obesity: In a 16-week trial in 152 sedentary older adults with overweight or obesity, most of whom had chronic pain, curcumin (160 mg a day), alone or with fish oil, did not reduce pain ^[26].

Turmeric extracts low in curcumin: A water extract of turmeric (NR-INF-02) contains almost no curcuminoids ^[57]. In a 42-day trial in 120 people with knee osteoarthritis, 500 mg of it twice daily improved pain and knee function more than placebo, which suggests that compounds in turmeric other than curcumin may also play a part ^[27]. A separate 12-week trial in 70 people found that a turmeric extract (2 capsules a day) improved knee pain by 9.1 mm on a visual pain scale more than placebo but did not reduce joint swelling on MRI ^[28].

After knee cartilage surgery: In 43 people followed for a year after knee cartilage repair (mosaicplasty), Theracurmin (180 mg curcumin a day) did not improve pain or knee scores more than placebo. At follow-up arthroscopy, though, the cartilage was stiffer and smoother in the Theracurmin group ^[29].

Meta-analyses: Several meta-analyses of RCTs have evaluated curcumin for knee osteoarthritis, finding generally positive initial evidence for pain relief, though the NCCIH states that higher-quality evidence is still needed to reach definitive conclusions ^{[2, 30, 31, 32]}.

Rheumatoid Arthritis

Small trials suggest curcumin may ease rheumatoid arthritis symptoms, with mixed results against anti-inflammatory painkillers, but it did not help people stay in remission while their usual medicines were tapered.

BCM-95 vs. diclofenac: In a small pilot trial in people with active rheumatoid arthritis, curcumin (500 mg) improved disease activity scores more than diclofenac (50 mg), and the combination did no better than curcumin alone. The trial was small and had no placebo group ^[33].

Curcumin as an add-on: In women with rheumatoid arthritis, taking 500 mg of curcumin a day for 8 weeks lowered tender and swollen joint counts, pain and disease activity scores compared with placebo ^[34].

High-dose curcumin (1,200 mg/day): An older study found 1,200 mg curcumin reduced symptoms, but not as effectively as the NSAID phenylbutazone ^[35].

For maintaining remission: In 200 adults with rheumatoid arthritis in lasting remission whose usual DMARDs were tapered and stopped, curcumin (1 g with 5 mg piperine twice a day) did not keep more people free of flares. At 52 weeks, 60% on curcumin and 64% on placebo had stayed flare-free, and the median time to a flare was alike in both groups. The per-protocol analysis included 92 people on curcumin and 93 on placebo ^[36].

Back Pain

A few short trials suggest products that combine turmeric and Boswellia may ease low back pain. Because the two herbs were given together, these trials cannot show which one helped.

Posture-related pain: In 52 adults with low back discomfort from poor posture, 500 mg a day of a turmeric-Boswellia product (Rhuleave-K) for 15 days cut pain intensity by 56% at day 7 and by 98.5% at day 15, and disability scores by 98%, against 3% with placebo ^[38].

Chronic low back pain: In a trial in adults with chronic low back pain (45 per group), a 300 mg capsule of a different Boswellia and turmeric product once daily for 90 days reduced pain, unpleasantness and disability more than placebo ^[39].

Gastrointestinal Conditions

Indigestion (Functional Dyspepsia)

A few small, short trials suggest curcumin may ease indigestion, and two found it worked about as well as omeprazole:

• Turmeric vs. placebo: In 116 people with indigestion, turmeric capsules (2 capsules four times a day for 7 days) helped 87% of those taking them, against 53% with placebo ^[40].
• Curcumin vs. omeprazole: In 132 people with functional dyspepsia, curcumin and omeprazole gave similar pain reductions after 4 weeks (16.98 and 14.69 points on a symptom scale), both more than placebo (10.53 points); curcumin also eased other symptoms more than placebo ^[41].
• A second curcumin vs. omeprazole trial: In 206 people with functional dyspepsia, curcumin (two 250 mg capsules four times a day), omeprazole (20 mg once a day) and the two together, each taken for 28 days, gave similar improvements in pain and other symptoms; combining them added nothing ^[42].
• Curcugen for digestive symptoms: In 79 adults with digestive complaints, 500 mg once daily for 8 weeks reduced overall symptom scores and anxiety more than placebo, but no other self-reported measure differed ^[43].

Ulcerative Colitis

Curcumin may be helpful as an adjunct to standard medication for ulcerative colitis, though results vary by formulation and dose:

• High-dose curcumin (3 g/day) for active UC: In 50 people whose mild to moderate ulcerative colitis had not responded to mesalamine alone, adding curcumin capsules for one month brought 53.8% into remission, against none on placebo ^[44].
• Nanomicelle curcumin (240 mg/day): In 56 people with mild or moderate ulcerative colitis, 80 mg three times daily added to mesalamine for 4 weeks reduced urgency, improved well-being and lowered overall disease activity scores compared with placebo ^[45].
• Maintenance of remission: In people with ulcerative colitis in remission, 1 g of curcumin after breakfast and 1 g after the evening meal, added to sulfasalazine or mesalamine for 6 months, led to fewer relapses than placebo (2 of 43 against 8 of 39) ^[46].
• Low dose for active colitis: At 450 mg a day for 8 weeks, curcumin brought no more people with mild or moderate ulcerative colitis into remission than placebo did ^[47].
• Crohn's disease: In 62 people who had bowel surgery for Crohn's disease and were taking azathioprine, 3 g of curcumin a day for 6 months did not prevent recurrence better than placebo. Severe recurrence was more common with curcumin (55%) than with placebo (26%) ^[48].

Familial Adenomatous Polyposis

Early small reports suggested curcumin might shrink intestinal polyps in people with familial adenomatous polyposis. But in a 12-month trial in 44 people, pure curcumin taken at 1,500 mg twice a day did not reduce polyp number or size compared with placebo ^[49].

Other GI Conditions

For other digestive conditions, such as stomach or duodenal ulcers, there isn't enough evidence to conclude whether curcumin helps ^[2].

Muscle Soreness and Recovery After Exercise

Results are mixed. Curcumin taken before hard exercise may dampen inflammation, and taken afterwards may ease soreness and muscle damage, but several trials found no effect on soreness.

Positive findings:

• Theracurmin (180 mg/day): In healthy men, curcumin taken for 7 days before an arm exercise did not reduce soreness, though it blunted one inflammation marker 12 hours later. In other men who took it for 7 days after the exercise, soreness and creatine kinase, a sign of muscle damage, were lower, and strength and range of motion recovered faster ^[50].
• Meriva (1 g twice daily, giving 200 mg curcumin twice daily): In 20 healthy men, Meriva taken from 48 hours before to 24 hours after a downhill running test (4 days in all) eased leg pain compared with placebo, though the difference was significant only in the front of the thighs. Fewer men on curcumin had MRI signs of muscle injury at the back or inner side of the thighs ^[51].
• High-dose curcumin (5,000 mg/day): In a crossover trial in men, 2.5 g twice daily from 2 days before to 3 days after hard leg exercise caused moderate to large reductions in leg pain at 24 and 48 hours, with small falls in creatine kinase and mixed effects on the inflammation marker IL-6 ^[52].
• Standard curcumin (500 mg twice daily): In 22 adults, curcumin taken for 6 days before, on the day of and 3 days after jumping exercise reduced soreness at 48 and 72 hours and preserved jump height, but did not change creatine kinase or the inflammation marker ESR compared with placebo ^[54].

Negative findings:

• Longvida (400 mg/day): In 28 adults, Longvida taken from 2 days before to 4 days after leg-press exercise did not reduce muscle soreness compared with placebo, but blunted rises in creatine kinase and two inflammation markers ^[55].
• Curcumin plus piperine: In elite rugby players, curcumin with piperine taken before and after damaging exercise lessened the drop in sprint power the next day but had no other effect, including on soreness ^[56].

Turmeric polysaccharides (without curcumin): In an open study of 15 healthy adults with no placebo group, 2 g a day of a turmeric extract with almost no curcuminoids for 7 days lengthened the time before knee discomfort on a stair machine by 32 seconds and slightly lowered peak pain scores ^[57].

Memory, Cognition, and Alzheimer's Disease

Some trials found small cognitive benefits ^{[12, 58, 59]}, while others found none ^{[60, 61]}.

Positive short-term findings:

• Longvida (400 mg): In 60 healthy adults aged 60 to 85, attention and working memory were better than with placebo one hour after a first dose. After 4 weeks, working memory and mood, including fatigue, were also better ^[12].
• Turmeric powder (1 g) and pre-diabetes: In people aged 60 or over with untreated pre-diabetes, 1 g of turmeric taken with a white-bread breakfast raised working memory scores from 2.6 to 2.9 out of 3.0 over 6 hours. The gain did not depend on blood sugar or insulin levels ^[58].
• Theracurmin (180 mg/day): In 40 adults aged 51-84 years without dementia, 90 mg twice daily for 18 months improved verbal and visual memory and attention compared with placebo. Brain scans suggested less build-up of amyloid and tau in some regions with curcumin, though what this means clinically is unclear ^[59].

Negative longer-term findings:

• Longvida (2,000 mg/day): In 39 healthy adults aged 45 to 74, 12 weeks of Longvida did not improve motor or cognitive function compared with placebo ^[60].
• Curcumin (160 mg/day): In 152 sedentary older adults with overweight or obesity, 16 weeks of curcumin did not improve blood vessel responses in the brain ^[26].
• BCM-95 (1,500 mg/day): In 96 older adults followed for 12 months, BCM-95 (Biocurcumax) made no difference on most memory and thinking tests. Scores on one screening test fell at 6 months in the placebo group but not in the curcumin group ^[61].

Alzheimer's disease trials: Results have been disappointing. In trials in people with Alzheimer's disease, curcumin did not improve memory or thinking more than placebo ^{[62, 63]}. A case report described three patients with severe Alzheimer's whose agitation, anxiety and irritability eased after 12 weeks of turmeric capsules ^[64]. A 2016 review said clinical studies have shown limited effects so far, most likely because curcumin is poorly absorbed and because trials enrolled people whose disease was already advanced ^[65].

Depression

Curcumin may help treat depression; in one trial its benefits appeared only after 4 weeks and were greater in atypical depression ^[66].

Curcumin for major depression: In 56 people with major depression, 500 mg of curcumin twice daily did no better than placebo for the first 4 weeks but worked better from weeks 4 to 8, and helped most in atypical depression ^[66]. A later 12-week trial in 123 people found that the curcumin groups together improved more than placebo, with no difference between 250 mg and 500 mg twice daily ^[67].

Curcuminoid escalation with antidepressants: In adults already taking antidepressants, curcuminoids starting at 500 mg/day and escalating to 1,500 mg/day over 3 weeks, then maintained for 8 more weeks, significantly improved depression symptoms at 12 weeks compared to placebo ^[68].

Long-term in diabetic adults with depression: In 227 adults with obesity and type 2 diabetes, 12 months of curcumin improved depression severity more than placebo ^[69].

No effect in postmenopausal women: In 74 postmenopausal women, 500 mg of curcumin twice a day for 8 weeks did no better than placebo on depression scores ^[70].

Insulin Resistance and Diabetes

In healthy people: In a crossover study in 16 healthy volunteers, 180 mg of curcumin taken before a standard meal reduced the rise in blood sugar after the meal by 36% and the insulin response by 26% compared with placebo ^[71].

Insulin resistance: In adults at high risk of type 2 diabetes, curcumin at 180 mg/day for 12 weeks reduced insulin resistance (HOMA-IR fell by 0.3, against a 0.01 rise with placebo) and lowered blood levels of GSK-3β and IAPP, two proteins involved in insulin resistance ^[72].

Prediabetes prevention: In a 9-month trial in 240 people with prediabetes, none of those taking curcumin developed type 2 diabetes, against 16.4% of those on placebo. Curcumin also appeared to improve the function of the insulin-making beta cells ^[73].

Type 2 diabetes — mixed results:

• In 33 people with impaired glucose tolerance or type 2 diabetes, Theracurmin 180 mg/day for 6 months left HbA1c no lower than placebo ^[74].
• Curcuma longa 500 mg/day with 5 mg piperine for 120 days lowered blood glucose, HbA1c, insulin resistance (HOMA) and triglycerides in the curcuma group of a 71-person trial in type 2 diabetes ^[75].
• Curcuminoids 1,500 mg/day for 12 months in 272 adults with type 2 diabetes lowered fasting blood sugar and HbA1c and reduced insulin resistance compared with placebo ^[76].

Cardiovascular risk in diabetics: In 72 adults with type 2 diabetes, 500 mg of curcumin three times daily added to usual treatment lowered systolic and diastolic blood pressure and LDL cholesterol, and raised HDL cholesterol, compared with usual treatment alone. It did not change HbA1c or fasting blood sugar, and the control group took no placebo ^[77].

Fatty Liver Disease (MASLD/NAFLD)

Early research suggests curcumin might improve some measures of fatty liver disease (MASLD, formerly called NAFLD), but which ones improve consistently is unclear ^[2].

Positive findings:

• Curcumin 500 mg/day for 24 weeks: In 80 people with simple fatty liver, curcumin lowered a scan measure of liver fat (controlled attenuation parameter) by 17.5 dB/m more than placebo, along with body weight (by 2.6 kg), triglycerides, fasting blood sugar, HbA1c and insulin ^[78].
• Curcuminoids 500 mg/day plus piperine for 12 weeks: In 70 people with fatty liver, disease severity on ultrasound improved more than with placebo. Liver enzymes fell within the curcumin group, but the change did not differ significantly from placebo ^[79].
• Curcumin 500 mg/day plus piperine for 12 weeks: In people with moderate to high levels of liver fat (30 per group), curcumin lowered ALT, AST, total and LDL cholesterol, fasting blood sugar, waist size and systolic blood pressure compared with placebo, but liver stiffness on FibroScan did not change ^[80].
• Meriva 2,000 mg/day for 72 weeks (NASH): In 52 people with biopsy-proven NASH, the disease resolved without worse scarring in 62% on Meriva against 12% on placebo, and chronic kidney disease regressed in 50% against 0% ^[81].

Negative finding: In 37 adults with obesity, lecithin-formulated curcumin (200 mg twice daily) for 6 weeks did not lower liver fat compared with placebo ^[82].

Multiple meta-analyses of RCTs support modest benefits for liver enzymes and some metabolic markers in NAFLD, though conclusions are limited by heterogeneous study designs ^{[83, 84, 85, 86]}.

Cholesterol

Evidence is mixed. A 2017 meta-analysis found modest falls in LDL cholesterol and triglycerides ^[89], but turmeric did not lower LDL cholesterol in a trial that compared it with a low-dose statin ^[91].

Regular curcumin showed no benefit: A 2014 meta-analysis of five trials found no significant effect of curcumin on total, LDL or HDL cholesterol or triglycerides ^[87]. An 8-week study of 1 g curcumin (with or without probiotics) in overweight adults with metabolic syndrome on a calorie-restricted diet also showed no benefit ^[88].

Modest benefit in people at heart risk: A 2017 meta-analysis of 7 trials (649 patients) found that turmeric and curcumin lowered LDL cholesterol and triglycerides in people with cardiovascular risk factors ^[89]. Adding 200 mg of curcumin a day to phytosterols for 4 weeks lowered total and LDL cholesterol by 11.0% and 14.4% (against 4.8% and 8.1% with phytosterols alone); curcumin alone gave small, non-significant falls ^[90].

Comparison with statin: In a 28-day trial in 190 adults at raised heart risk, rosuvastatin 5 mg lowered LDL cholesterol by 35.2% more than placebo, while turmeric, like the other supplements tested, did not lower it significantly compared with placebo ^[91].

Blood Pressure

Postmenopausal women: In a pilot trial in postmenopausal women, 150 mg of curcumin a day for 8 weeks did not lower arm blood pressure on its own; pressure in the aorta fell only in women who combined curcumin with endurance exercise ^[141]. A related trial in 32 postmenopausal women found that 8 weeks of curcumin improved how well arteries widen (flow-mediated dilation) as much as aerobic exercise did ^[92].

Lupus nephritis: In 24 people with hard-to-treat lupus kidney disease, a 500 mg turmeric capsule (22.1 mg curcumin) with each meal for 3 months lowered systolic blood pressure, as well as protein and blood in the urine; placebo had no significant effect ^[93].

Weight Loss

A meta-analysis of 11 trials (876 people) found curcumin lowered body weight by an average of 1.14 kg and BMI by 0.48 kg/m2. Waist size did not change overall, but did fall in trials of overweight people taking at least 1,000 mg a day for at least 8 weeks ^[94].

In a 12-week trial in 58 people with inflammatory bowel disease, curcumin (1,000 mg a day) with piperine (10 mg) increased lean (fat-free) mass compared with placebo ^[95].

Cancer

Human evidence is limited to small trials with mixed results ^{[96, 100, 101]}.

Colon cancer prevention: In an open-label trial in 44 smokers, 4 g of curcumin a day for 30 days cut the number of aberrant crypt foci (possible early precancerous changes in the colon) by 40%; the 2 g dose did not ^[96].

Myeloma precursors (MGUS/SMM): In 36 people with these precursor conditions, curcumin (a 4 g dose, then an 8 g dose in an open extension) lowered the free light-chain ratio, a marker linked to disease progression ^[97]. Another study gave 4-8 g daily for as long as a year; it lowered some inflammatory markers and a protein that drives myeloma growth, but clinical outcomes did not improve ^[98]. A case report described a woman with myeloma who began daily curcumin as she approached her third relapse; without further myeloma treatment, her disease stayed stable for 5 years ^[99].

Pancreatic cancer: In 21 patients taking 8 g curcumin daily, only 2 showed a positive biological response ^[100].

Prostate cancer: In men with metastatic castration-resistant prostate cancer on docetaxel, adding 6 g of curcumin a day for 7 days of each 3-week cycle did not lengthen time to progression or survival, or improve PSA response or quality of life, compared with placebo. The trial stopped early for futility ^[101].

Oral mucositis (cancer treatment side effect): Initial research suggests turmeric or curcumin in oral or mouthwash formulations may improve symptoms of oral mucositis related to cancer treatment ^{[2, 102, 103, 104]}.

Eye Disease

Dry eye: In a trial in adults with dry eye (77 on the supplement, 78 on placebo), a daily capsule with curcumin (200 mg curcuminoids), lutein (20 mg), zeaxanthin (4 mg) and vitamin D3 (600 IU) for 8 weeks improved tear production and symptoms compared with placebo, but did not reduce use of artificial tears ^[105].

Age-related macular degeneration (AMD): A study of electronic health records compared adults prescribed curcuma-based supplements with matched adults who were not. Among those aged 50 or older without AMD, supplement use was linked to a lower risk of nonexudative AMD (RR 0.23, 95% CI 0.21 to 0.26), exudative AMD (RR 0.28, 95% CI 0.24 to 0.32) and blindness (RR 0.46, 95% CI 0.36 to 0.59). This kind of study cannot prove that the supplements caused the difference ^[106].

Chronic anterior uveitis: In 106 people with recurring anterior uveitis, Meriva tablets twice a day added to usual treatment for 12 months reduced how often relapses occurred, and eased eye discomfort in more than 80% of patients. There was no placebo group ^[107].

Allergies (Perennial Allergic Rhinitis)

In a trial in China among 241 people with perennial (year-round) allergic rhinitis, curcumin for 2 months improved sneezing, runny nose, and nasal congestion compared with placebo ^[108]. Curcumin also decreased inflammatory mediators including IL-4 and TNF-alpha ^[108].

Common Cold

In a 12-week trial in 99 healthy Japanese adults, 150 mg a day of Theracurmin or Theracurmin Super curcumin cut the total number of days with cold symptoms compared with placebo ^[109].

Postmenopausal Health

In 74 postmenopausal women, curcumin 500 mg twice daily for 8 weeks improved musculoskeletal health scores by 5.3 points and fatigue scores by 2.6 points compared with placebo; the first gain fell just short of the smallest change counted as clinically important ^[70].

Asthma

In children and adolescents with persistent asthma on standard medicines, powdered turmeric root (30 mg/kg/day) for 6 months led to fewer night-time awakenings, less use of reliever inhalers and better asthma control than placebo ^[110].

Gout

Curcumin may block xanthine oxidase, the enzyme that makes uric acid. Yet in 39 people with persistently high uric acid, 500 mg twice daily for 8 weeks lowered blood uric acid no more than placebo and did not raise the amount cleared in urine ^[111].

Parkinson's Disease

In people with Parkinson's disease (30 per group) whose symptoms were controlled on standard treatment, 80 mg a day of nanomicelle curcumin for 9 months gave no greater gain in symptoms or quality of life than placebo ^[112].

Psoriasis

In a small open-label trial without a placebo group, 4.5 g a day of curcuminoids helped few people with plaque psoriasis (a 16.7% response rate), and the authors said even this might reflect a placebo effect or the natural course of the disease ^[113].

Reducing Salt Intake

In a taste study, the aroma of turmeric made low-salt mashed potato taste saltier, about as much as bay leaf did. Eugenol, an aroma compound, was the only one whose levels tracked the effect ^[114].

4. Recommended Dosing

Brad tends to encourage his patients not to take curcumin supplements, so the doses below describe what trials used, not a recommendation ^[139].

There is no standard guidance on the best dose of curcumin for any condition ^[139]. The following table summarizes dosages used in clinical studies:

Dosage by Condition

General Dosing Principles

• Take with a fat-containing meal, because fat aids curcumin absorption ^{[4, 139]}.
• Split doses may help, because standard curcumin has a short half-life (~2 hours) ^[4].
• Standard curcumin extract (95%): Trials that found tentative benefit often used 500-2,000 mg of curcumin a day ^[139].
• Start with a lower dose to reduce the risk of GI side effects, particularly with higher-potency formulations ^[22].

How to Read a Supplement Label

When comparing products, pay attention to:

• The weight of the total formulation vs. the weight of curcuminoids.
• Whether the product lists total curcuminoids or just curcumin.
• Whether a bioavailability enhancer is included (piperine/Bioperine, phospholipids, etc.).

5. Safety and Side Effects

General Safety

In short studies, high doses have generally been well tolerated: single doses from 500 to 12,000 mg caused only minimal side effects in healthy volunteers ^[115], and people with advanced pancreatic cancer took 8 g a day, one of them for more than 18 months, with no toxicity reported ^[100].

The NCCIH states that conventionally formulated oral turmeric or curcumin (not modified to enhance bioavailability) is "likely safe in the recommended amounts for up to 2 or 3 months" ^[2].

Common Side Effects

• Nausea, diarrhea, and stomach upset ^{[2, 115]}
• Yellow stool ^[77]
• Headache ^[77]
• Allergic skin reactions (oral use); hives or itching (topical use) ^[2]

GI side effects may be dose-dependent. In a 24-week Alzheimer's trial of curcumin, 5 of 24 people on curcumin withdrew, 3 of them because of stomach or bowel symptoms ^[63]. In a knee trial, a higher dose of a turmeric extract caused more product-related side effects than placebo ^[22]. Starting with a lower dose may reduce this risk ^[22].

Liver Injury

This is the most serious safety concern. Even though curcumin is studied for fatty liver disease treatment, it can paradoxically cause liver injury in some individuals ^[2].

Case report: One report described autoimmune hepatitis in an older woman taking turmeric supplements, which resolved quickly after she stopped them ^[116].

DILIN cases: The U.S. Drug-Induced Liver Injury Network (DILIN) found 10 cases of turmeric-linked liver injury among the cases it enrolled from 2004 to 2022 ^[117]:

• 8 of the 10 were women; median age 56 years
• Liver injury typically appeared 1 to 4 months after starting turmeric
• 3 of the 7 products tested also contained piperine (black pepper)
• In 9 of the 10, the injury mainly affected liver cells (hepatocellular)
• Five people needed hospital care, and one died of acute liver failure
• 7 of the 10 carried the HLA-B*35:01 gene variant

Additional case review: In a pathology series of 11 further cases, 10 (91%) were women, and the median age was 58 years. More than half had no symptoms and were found only through abnormal liver tests. After they stopped the supplements (four also received steroids), liver tests returned to normal in all of them ^[118].

LiverTox: The NIH's LiverTox database says turmeric and curcumin have a long history of safety and are linked to a low rate of brief rises in liver enzymes, but some turmeric products have recently been implicated in several dozen cases of clinically apparent acute liver injury ^[142].

Population exposure: A study estimated the exposure to potentially hepatotoxic botanicals in U.S. adults, identifying turmeric/curcumin as one of the most commonly used supplements with liver injury reports ^[119].

NCCIH warning: "Many curcumin products with increased bioavailability are on the market, and liver damage has been reported in some people who have consumed these bioavailable formulations" ^[2].

Liver tests: Brad advises anyone taking curcumin supplements to have regular blood tests of liver function, because liver injury from curcumin is rare but serious ^[140]. In a series of 11 people with liver injury linked to turmeric supplements, more than half (six of the 11) had no symptoms and were found only through abnormal liver tests ^[118].

Warning signs: The NCCIH lists fatigue, nausea, poor appetite, dark urine and jaundice (yellow skin or eyes) as signs of liver damage. If any occur, stop turmeric or curcumin products straight away and talk with your doctor ^[2].

Methotrexate: One report in the WHO VigiBase database described raised liver enzymes in a person taking methotrexate with curcuminoids ^[132].

Drug interaction risk: Use of GLP-1 agonists (e.g., semaglutide/Ozempic) may increase liver injury risk from turmeric, as GLP-1 drugs slow intestinal transit and could increase curcumin absorption ^[120].

Gallbladder Effects

In 12 healthy volunteers, single doses of 20, 40 and 80 mg of curcumin made the gallbladder contract within 2 hours, and 40 mg was enough to shrink it by half ^[121]. In people with gallstones, curcumin might therefore cause pain, so check with your doctor before taking it.

Kidney Stones

Turmeric is rich in oxalate, and most of it dissolves in water. In a small crossover study, supplemental doses of turmeric raised urinary oxalate more than cinnamon or water did, which the authors said can increase kidney stone risk in susceptible people ^[122]. People prone to calcium oxalate kidney stones should limit turmeric intake.

Yellowing of Skin

Rarely, turmeric consumption can cause yellow skin discoloration without liver injury. This has been reported with high dietary turmeric intake and with standard supplement doses, resolving after discontinuation ^{[123, 124]}.

Lowering Iron Levels

Curcumin may bind iron and reduce its availability. A case report described iron deficiency anemia in a man taking high-dose turmeric extract, which resolved within 2 weeks of stopping it ^[125]. However, in a controlled trial in 155 healthy adults, 500 mg of bioavailability-enhanced curcumin taken alongside an iron supplement did not impair acute iron absorption ^[126]. As a precaution, people with anemia or low iron stores may want to avoid taking curcumin at the same time as iron.

Tooth Staining

Turmeric may stain teeth. A laboratory study suggests that freshly bleached teeth can take up the stain ^[127].

Pregnancy and Nursing

Turmeric supplements may be unsafe during pregnancy, and little is known about their safety while breastfeeding in amounts greater than those found in food ^[2].

6. Drug Interactions

Anticoagulants and Antiplatelet Drugs

Avoid combining turmeric or curcumin supplements with blood thinners, including aspirin, clopidogrel (Plavix), warfarin (Coumadin) and rivaroxaban (Xarelto), unless your doctor agrees, because curcumin might add to their effect. A case report linked turmeric to a raised INR, a sign of slower blood clotting, in a person taking the vitamin K antagonist fluindione ^[128]. Turmeric might also raise bleeding risk during surgery or other procedures, so tell your doctor you take it beforehand; they may advise stopping it for a time. Reports in the WHO VigiBase database also describe bleeding in people taking the leukemia drug ibrutinib with curcumin or curcuminoids; the authors suggested curcumin might raise ibrutinib levels ^[132].

Blood Sugar-Lowering Medications

Curcumin may add to the glucose-lowering effect of diabetes medicines. In a small study of people with type 2 diabetes taking glyburide, adding curcumin lowered blood glucose further, though none had low blood sugar ^[130]. Use caution with insulin, glyburide (DiaBeta), pioglitazone (Actos), and rosiglitazone (Avandia).

CYP3A4 Enzyme Interactions

Curcumin may interfere with the CYP3A4 enzyme system that metabolizes many drugs, potentially increasing or decreasing their blood levels:

Drugs with increased levels (potentially dangerous):

• Tacrolimus. A case report described a liver transplant patient whose tacrolimus level rose sharply, with kidney damage, after eating large amounts of turmeric with food; kidney function improved after tacrolimus was withheld ^[131].
• Bortezomib (Velcade). One report in VigiBase, the WHO pharmacovigilance database, described constipation, fewer red blood cells, night sweats, nerve damage in the hands or feet (peripheral neuropathy) and a rash in a person with myeloma taking bortezomib with curcuminoids. The authors suggested curcuminoids might raise bortezomib levels ^[132].
• Curcumin might, in theory, affect other drugs broken down by CYP3A4. If you take prescription medicines, check with your doctor or pharmacist before starting.

Drugs with decreased levels:

• Tamoxifen. In breast cancer patients, high-dose curcumin (1,200 mg three times daily) lowered exposure to endoxifen, tamoxifen's active form, by 7.7% (not significant) and by 12.4% with piperine; the authors estimate this could push endoxifen below the effective level in 20-40% of patients ^[134].
• Talinolol — Some evidence curcumin may decrease bioavailability of this beta-blocker ^[135].

Drugs with levels that might go up or down:

• Everolimus. Curcumin supplements might change blood levels of everolimus, a cancer drug ^{[132, 133]}.

Sulfasalazine

Curcumin may increase the effects of this anti-inflammatory drug ^[136].

Testosterone (Oral)

Curcumin can raise blood levels of oral testosterone (testosterone undecanoate) by blocking the gut enzyme UGT2B17. In a small crossover study in men, taking it with curcumin raised testosterone exposure by 50% in the first 6 hours and the peak level by 80% ^[137].

Piperine Interactions

Piperine, the black pepper extract added to some curcumin supplements, might also raise blood levels of some medicines. If you take prescription medicines, check with your doctor or pharmacist before choosing a product that contains it.

Chemotherapy

In a study of 202 older adults having cancer treatment, pharmacists judged 6 possible herb-drug interactions clinically relevant, and turmeric was one of the four herbs involved ^[138]. If you are having cancer treatment, tell your cancer team before taking turmeric or curcumin.

7. Dietary Sources

Turmeric is the primary dietary source of curcumin.

Turmeric Content

The curcuminoids make up about 3 to 5% of turmeric, though products vary considerably in their make-up ^[132].

Practical Notes

• Turmeric powder or fresh root beats curcumin powder: In one small study, a meal made with turmeric powder or grated fresh root raised blood curcuminoids more than one made with curcumin powder, and the powder gave the highest levels of two curcuminoids ^[5].
• Cooked turmeric is still absorbed: Curcumin from turmeric cooked into food reaches the blood, mostly in conjugated form ^[19].
• Fat may help absorption: Some evidence shows that eating curcumin with foods that contain fat can help absorption ^[139].
• Black pepper: Piperine, from black pepper, raised curcumin absorption in one early study ^[8], though Brad notes that several later analyses found it does not ^[139].