Benefits
Anti-Inflammatory Effects
Curcumin lowers circulating markers of inflammation in randomized trials, with a pooled analysis of 66 trials reporting reductions in C-reactive protein, TNF-alpha and interleukin 6. The clearest symptom level evidence is in knee osteoarthritis. Lowering a blood marker is not the same as treating a disease, and curcumin has not been shown to replace prescribed anti-inflammatory or disease modifying medication.
Antioxidant Protection
In pooled randomized trials curcumin raised total antioxidant capacity and superoxide dismutase activity and lowered malondialdehyde, a marker of lipid peroxidation. These are blood markers measured over weeks to months, not health outcomes, and the phase I study on this page that measured detoxification and DNA damage markers in blood cells did not report a benefit on them.
Joint Health Improvement
This is the best supported use on this page. In knee osteoarthritis, several randomized trials and two systematic reviews point to a modest reduction in pain and better function, and one placebo controlled trial that imaged the knee found less pain but no change in effusion synovitis or cartilage. In rheumatoid arthritis a 45 patient pilot reported a benefit, but a 200 person placebo controlled trial found none in keeping remission.
Memory and Cognition
In laboratory studies curcumin protects brain cells from inflammation and oxidative stress, but human results point partly the other way. A meta-analysis of six trials found a modest memory benefit in older adults without dementia, while in people with Alzheimer's disease the curcumin groups tended to score worse than placebo on a standard cognitive test. No trial has shown that curcumin prevents or slows cognitive decline.
Cardiovascular Health Support
Pooled randomized trials show better flow mediated dilation, a measure of blood vessel function, and small drops in total cholesterol, triglycerides and LDL of about 4 to 7 mg/dL, graded low certainty, so the size of any effect is uncertain. There was no effect on arterial stiffness, and an earlier, smaller meta-analysis found no lipid change. No trial has tested whether daily curcumin lowers heart attacks or strokes.
Digestive Health Benefits
In mild to moderate ulcerative colitis, adding curcumin to mesalamine brought clinical remission at four weeks in about half of patients versus none on placebo in one trial, and a pooled analysis of eight trials in 482 patients found more clinical remission, though endoscopic remission was not significant. Curcumin was added to prescribed therapy, not used instead of it. Evidence in IBS is weaker and comes partly from mixed herbal formulas.
Mood Regulation
A meta-analysis of six trials in 377 patients reports a small but statistically significant reduction in depressive symptoms, in trials lasting only four to eight weeks. A meta-review of plant based psychiatric treatments flagged weak methodology and possible publication bias in the curcumin analyses. The serotonin and dopamine explanation comes from animal and laboratory work, and curcumin is not a treatment for depression.
Upper Digestive Comfort After Meals
In adults with functional dyspepsia (recurring upper stomach pain, burning or fullness), curcumin or turmeric capsules eased symptoms more than placebo in two of three placebo controlled trials; the third, 500 mg a day added to famotidine, found no difference. A Cochrane review rated this moderate certainty, another review found most trials at high risk of bias, and we found no placebo controlled heartburn trial.
Mechanism of action
Anti-Inflammatory Activity
In laboratory studies curcumin inhibits pro-inflammatory pathways such as NF-kB and COX-2 and reduces production of cytokines like TNF-alpha and IL-6. In pooled human trials it lowered blood levels of C-reactive protein, TNF-alpha and IL-6, which are markers of inflammation rather than disease outcomes.
Antioxidant Effects
In laboratory studies curcumin scavenges free radicals and switches on antioxidant enzymes such as superoxide dismutase. In pooled human trials it raised superoxide dismutase activity and total antioxidant capacity in blood, which are markers rather than health outcomes.
Neuroprotection
Curcumin crosses the blood brain barrier in animal models, where it reduces neuroinflammation and amyloid plaque formation and raises BDNF expression. Human brain exposure after oral dosing is very low, and the Alzheimer's disease trials that tested this idea did not show a clinical benefit, so this remains a proposed mechanism rather than a demonstrated one in people.
Cardiovascular Protection
Curcumin improves flow mediated dilation, a measure of endothelial function, in pooled randomized trials, which also show small reductions in cholesterol and triglycerides. The nitric oxide and LDL oxidation steps are inferred from laboratory work. No trial has shown that curcumin slows atherosclerosis, and the one trial that counted heart attacks gave high dose curcuminoids for about a week around bypass surgery, which says little about everyday supplement use.
Modulates Gut Inflammation
In laboratory and animal studies curcumin strengthens the gut barrier, shifts gut bacteria and dampens intestinal inflammation. In people the evidence is limited to small add-on trials in ulcerative colitis and mixed results in other digestive conditions, so these steps remain proposed rather than demonstrated.
Mood Regulation
In rodent studies curcumin raises brain serotonin and dopamine and inhibits monoamine oxidase enzymes. These measurements come from animal brain tissue and have not been reproduced in people, so this is a proposed mechanism rather than a demonstrated one.
Anti-Cancer Effects
In laboratory and animal models curcumin slows cancer cell growth, triggers cell death through pathways such as PI3K/Akt and suppresses new blood vessel growth. These effects have not been shown in people.
Enhances Detoxification
Curcumin upregulates phase II detoxification enzymes such as glutathione S-transferase in laboratory models. The one human trial cited on this page that measured glutathione S-transferase activity in blood cells did not report an increase, so this remains a laboratory finding.
Clinical trials
Multicenter randomized trial in 367 adults with knee osteoarthritis comparing Curcuma domestica extract 1,500 mg/day with ibuprofen 1,200 mg/day for 4 weeks; outcomes were WOMAC total, pain, stiffness and function scores and adverse events. (Kuptniratsaikul et al. 2014, Clin Interv Aging)
367 adults with knee osteoarthritis and a pain score of 5 or more out of 10; 4 weeks.
WOMAC total, pain and function scores with the extract were noninferior to ibuprofen at week 4, and stiffness showed only a trend. The share of patients with any adverse event was similar (29.7 percent vs 35.7 percent); only abdominal pain or distension was significantly less frequent with the extract (10.8 vs 18.1 percent), while indigestion (11.4 vs 15.9 percent) did not differ significantly. There was no placebo group, so this shows similarity to ibuprofen over four weeks, not proof that either beat no treatment, and it is not a reason to stop or replace a prescribed medicine.
Systematic review of 10 randomized trials in knee osteoarthritis, 1,287 participants in total, comparing turmeric or curcumin therapy with NSAIDs or no therapy. The authors synthesized the trials narratively and did not pool them into a single combined effect estimate. (Paultre et al. 2021, BMJ Open Sport Exerc Med)
10 randomized trials, 1,287 participants with knee osteoarthritis.
All 10 studies showed improvement in pain and function from baseline with turmeric therapy, and in the 3 studies comparing turmeric with anti-inflammatory drugs there was no difference in outcome scores. The authors concluded there appears to be a benefit compared with placebo, but stated that optimal dosing, frequency and formulation remain unclear. This review includes the 367 patient trial listed above, so the two entries are not independent evidence and should not be counted twice.
Meta-analysis of 8 randomized trials in knee osteoarthritis or rheumatoid arthritis, roughly 900 participants in total, evaluating turmeric or curcumin at about 1,000 mg/day of curcumin. The pooled pain estimate came from only 3 of the 8 trials and the pooled WOMAC estimate from only 4. (Daily et al. 2016, J Med Food)
8 randomized trials, roughly 900 participants with knee osteoarthritis or rheumatoid arthritis.
Turmeric or curcumin lowered pain scores versus placebo (mean difference -2.04 on a visual analogue scale, 3 trials) and WOMAC scores (mean difference -15.36, 4 trials), with no significant difference from pain medication in 5 trials. The authors concluded that the number of trials, the total sample size and the methodological quality were not sufficient to draw definitive conclusions. The lead author is president of a dietary supplement manufacturer, which the paper discloses. This analysis also contains both the 367 patient osteoarthritis trial and the 45 patient rheumatoid arthritis pilot listed on this page, so those trials are counted more than once across the page.
Randomized pilot trial in 45 patients with active rheumatoid arthritis comparing curcumin 500 mg, diclofenac sodium 50 mg, or both, twice daily for 8 weeks; primary outcome DAS28, secondary ACR response criteria. (Chandran et al. 2012, Phytother Res)
45 adults with active rheumatoid arthritis; 8 weeks.
All three groups improved on DAS28. The curcumin group showed the largest percentage improvement in DAS28 and ACR 20, 50 and 70 scores, significantly better than diclofenac alone, and curcumin was not linked to any adverse events. Cautions: 45 patients, no placebo group, a high absorption curcumin product, and a result that has not been repeated; a later 200 person placebo controlled trial in people whose rheumatoid arthritis was in remission found curcumin did not keep them flare free as their drugs were tapered. Curcumin is not a substitute for disease modifying antirheumatic drugs.
Scoping review of 389 published clinical trial reports of oral curcumin across metabolic, musculoskeletal, neurocognitive, gastrointestinal and cancer conditions. (Panknin et al. 2023, Int J Mol Sci)
389 clinical trial reports, most of them double blind placebo controlled.
About 29 percent of the reports covered obesity related metabolic disorders and 17 percent musculoskeletal disorders, and most of these described a benefit on clinical outcomes or biomarkers. Reports in neurocognitive disorders (11 percent), gastrointestinal disorders (10 percent) and cancer (9 percent) were fewer and gave mixed results. A scoping review maps what has been published rather than pooling results, so it gives no effect estimate, and the share of reports describing a benefit is not a measure of how well curcumin works. The authors call for systematic testing of formulations and doses in larger trials.
Phase I dose escalation trial in 15 patients with advanced colorectal cancer taking curcumin 0.45 to 3.6 g/day for up to 4 months; outcomes were pharmacokinetics, blood biomarkers and tolerability. (Sharma et al. 2004, Clin Cancer Res)
15 patients with advanced colorectal cancer that no longer responded to standard chemotherapy.
No dose limiting toxicity was seen up to 3.6 g/day. Curcumin and its metabolites reached plasma only in the 10 nanomolar range. At 3.6 g/day, inducible prostaglandin E2 in blood fell by 62 and 57 percent one hour after dosing on days 1 and 29. This was a dose finding and safety study in patients with advanced cancer, with no control group, and it does not show that curcumin treats anything; its main lesson is how little oral curcumin reaches the blood.
Randomized, double blind, placebo and omeprazole controlled trial in 132 adults with functional dyspepsia (Rome IV) for 4 weeks; outcome was the Severity of Dyspepsia Assessment (SODA). The abstract gives no dose; a later systematic review lists Curcuma longa capsules 500 mg four times daily. (Yongwatana et al. 2022, J Gastroenterol Hepatol)
132 adults with functional dyspepsia in Thailand (45 Curcuma longa, 43 omeprazole, 44 placebo); 4 weeks.
At 4 weeks, SODA pain and non pain scores fell more with Curcuma longa than with placebo (pain -16.98 vs -10.53, P<0.001; non pain -7.96 vs -6.05, P<0.008). Scores did not differ significantly from omeprazole. One 4 week trial; a non significant difference from a drug does not show that the two work the same, and this is not a reason to stop prescribed acid reducing medicine.
Multicenter randomized double blind trial in 116 adults with acid, flatulent or atonic dyspepsia comparing Curcuma domestica rhizome capsules, a herbal flatulence remedy and placebo, 2 capsules four times daily for 7 days. (Thamlikitkul et al. 1989, J Med Assoc Thai)
116 adults with dyspepsia at one provincial and five community hospitals in Thailand; 7 days; 10 did not return for follow up.
87 percent responded on Curcuma domestica and 83 percent on the flatulence remedy, versus 53 percent on placebo; the differences between placebo and the active treatments were statistically significant. Mild, self limited side effects occurred at similar rates in all three groups. Limitations: one week only, response judged by the patients, no formal diagnostic criteria for dyspepsia, and a later review rated the trial at high risk of bias.
Triple blinded randomized trial in 75 adults with functional dyspepsia (Rome III): 500 mg curcumin or placebo daily, each with famotidine 40 mg daily, for 1 month plus 1 month follow up; outcome was dyspepsia severity on the Hong Kong questionnaire. (Panahi et al. 2021, Adv Exp Med Biol)
75 adults with functional dyspepsia in Iran (39 curcumin, 36 placebo).
No significant difference was found between the curcumin and placebo groups in dyspepsia severity, H. pylori rate or blood measures, so this add-on trial was null. Severity fell over time within the curcumin group, which does not show an effect of curcumin when the comparison with placebo was not different. One author is from Sabinsa Corporation, and a later review identifies the product as C3 Complex curcuminoids.
Randomized double blind trial with no placebo arm in 206 adults with functional dyspepsia at Thai hospitals: curcumin, omeprazole 20 mg daily, or both, for 28 days, with curcumin as two 250 mg capsules four times daily (2,000 mg/day; the abstract states this for the combination arm); outcome was SODA at days 28 and 56. (Kongkam et al. 2023, BMJ Evid Based Med)
206 adults with functional dyspepsia (mean age about 50, 73 percent women); 151 completed.
All three groups improved on SODA pain, non pain and satisfaction scores at days 28 and 56, with no significant differences among them and no serious adverse events. With no placebo group and about a quarter of participants not completing, this cannot show that curcumin beat no treatment. A published correction states the trial was registered retrospectively, fixes dropout numbers, and removes the word equivalence because the trial was not designed to test it, so it is not evidence that curcumin works like a proton pump inhibitor.