- Curcumin is the yellow compound in turmeric (Curcuma longa). The US National Center for Complementary and Integrative Health (NCCIH) says "we don't know enough" to conclude that turmeric or curcumin benefits any health purpose, and it gives no evidence that it helps depression (NCCIH, April 2025). The EU herbal monograph covers only mild digestive complaints (EMA).
- Meta-analyses of small trials report that curcumin lowers depression scores more than placebo. The first, 6 trials with 377 people, found a standardised mean difference (SMD) of −0.34 (95% CI −0.56 to −0.13) (Ng 2017). A later one of 10 studies (531 people) found −0.75 (−1.11 to −0.39) (Fusar-Poli 2020). A 2026 review called its own positive result "fragile" (Musazadeh 2026).
- A University of Minnesota–led review funded partly by NIH argued that curcumin is "unstable, reactive, nonbioavailable", interferes with laboratory assays (a "PAINS" compound), and that "no double-blinded, placebo controlled clinical trial of curcumin has been successful" for any disease (Nelson 2017).
- Concentrated turmeric and curcumin supplements can rarely cause liver injury. Australia's regulator made a liver warning compulsory, phased in from 1 March 2024 to 1 March 2025, citing at least three deaths (TGA). Health Canada followed in 2025 (Health Canada). The US liver-injury network found most cases carried the immune gene HLA-B*35:01 (DILIN 2023).
- Curcumin supplements can raise bleeding risk with warfarin and other blood thinners, and New Zealand's regulator also named SSRI antidepressants (Medsafe 2018). Some loose turmeric spice in South Asia has been coloured with lead chromate (Forsyth 2019).
- Money trail: The best-known depression trials tested BCM-95, a branded extract from India's Arjuna Natural, which publicised the results (Arjuna press release). Their lead author declares grants or honoraria from Arjuna, Dolcas-Biotech, Verdure Sciences and Sabinsa, and co-wrote the guideline that "provisionally" recommends curcumin (WFSBP/CANMAT 2022). The Longvida mood trials were funded by its maker, Verdure Sciences (Cox 2020).
- Evidence grade: Weak for depressive symptoms, mainly as an add-on to antidepressants. Insufficient for anxiety disorders. Risk for the liver (rare), bleeding with blood thinners, and lead in some loose spice.
Independent evidence review · Supplement
Curcumin might slightly ease depressive symptoms over 4 to 12 weeks, especially alongside an antidepressant, but the evidence comes from a small number of short trials, several of them run with or supplied by extract makers, and the pooled effects swing widely from review to review. No regulator approves curcumin for depression, and NCCIH says the evidence is not enough to draw conclusions. Chemists have raised basic doubts about whether curcumin reaches the brain at all. Meanwhile, high-absorption supplements have been linked to rare but serious liver injury, prompting label warnings in Australia and Canada. Our confidence: low that curcumin has a real antidepressant effect; moderate that concentrated products carry a small liver risk.
Table of contents
- Evidence summary
- What curcumin is, and its legal status
- How it might work, and the chemistry critique
- What works and what does not
- Benefits by claim
- Risks and all side effects
- All interactions
- Who should avoid curcumin supplements
- Doses studied and product quality
- Follow the money: who makes it and who funded the evidence
- Related research
- Frequently asked questions
- Sources and funding notes
Evidence summary
| Claim | Evidence | Source | Funding / conflict | Strength |
|---|---|---|---|---|
| Curcumin reduces depressive symptoms vs placebo | 6 trials, 377 people: SMD −0.34 (−0.56 to −0.13). 10 studies, 531 people: Hedges' g −0.75 (−1.11 to −0.39). 2026: SMD −0.76 (−1.22 to −0.30; I² 94.5%), judged "fragile" on sensitivity analysis. | Ng 2017; Fusar-Poli 2020; Musazadeh 2026 | Academic teams (Singapore, Italy, Iran). Funding statements for Ng and Fusar-Poli could not be accessed; Musazadeh declares no competing interests. Pooled trials include maker-linked BCM-95 studies. | Weak |
| Curcumin works as an add-on to antidepressants | Network meta-analysis of 192 trials: curcumin + antidepressant vs antidepressant alone SMD 1.03 (0.55 to 1.51). Curcumin alone vs antidepressant: 0.62 (−0.17 to 1.40), not significant. | Cheng 2025 | Taiwan Ministry of Science and Technology grants; authors declare none | Weak |
| Curcumin matches an antidepressant | 60 people, 6 weeks, observer-masked: response 62.5% curcumin, 64.7% fluoxetine, 77.8% both; no significant difference, but no placebo arm and too small to show equivalence. | Sanmukhani 2014 | Indian academic hospital; funding statement not accessed. Described by an independent reviewer as "industry-independent". | Insufficient |
| Curcumin reduces anxiety | 8 trials, 567 people: SMD −1.56 (−2.48 to −0.64; I² 95.6%). Restricted to low-risk-of-bias trials: −0.58 (−0.99 to −0.18; 2 trials, 89 people). | Fathi 2024; Fornaro 2026 | Fathi: declares none. Fornaro: several authors declare extensive pharmaceutical payments; one declares supplement-company research support. | Insufficient |
| Branded "high-absorption" extracts improve mood | BCM-95: 56 people with major depression, benefit only from weeks 4 to 8. Longvida: lower fatigue and some mood scores in healthy older adults. | Lopresti 2014; Cox 2020 | BCM-95 trial: "industry support" per independent reviewer; extract from Arjuna. Longvida: funded by maker Verdure Sciences. | Weak |
| Curcumin supplements can injure the liver | US network: 10 cases, 1 death, 7 of 10 carrying HLA-B*35:01. Canada: 12 Canadian and more than 60 international reports, 2 deaths attributed. Australia: label warning compulsory since 2024. | DILIN 2023; Health Canada; TGA | NIH (NIDDK) cooperative agreement; government regulators | Risk |
| Curcumin raises bleeding risk with blood thinners | A patient on warfarin with a stable INR rose to over 10 within weeks of starting a turmeric product. | Medsafe 2018 | New Zealand government regulator | Risk |
| Some turmeric spice contains lead | Lead chromate pigment found in turmeric from 7 of 9 major producing districts of Bangladesh; up to 1,152 µg/g lead in powder. Lead above 1,000 µg/g also found in parts of India and Pakistan. | Forsyth 2019; Forsyth 2024 | Stanford-led academic team; no competing interests declared | Risk |
Independent evidence and credibility scorecard
Independence tiers: 1 = regulator or government body; 2 = academic or Cochrane team with no declared commercial funding for the work; 3 = independent authors analysing mostly manufacturer-run trials, or minor declared industry ties; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D.
| Source | Who funds it | Country | Independence tier (1–4) | Credibility (A–D) | Why it would tell the truth / residual bias |
|---|---|---|---|---|---|
| EMA herbal committee monograph and 2025 addendum | EU agency (public body) | EU | 1 | A | No product to sell; reviewed 7,771 new references. Assesses traditional low-dose turmeric, not high-absorption extracts, so it says little about mood. |
| Health Canada summary safety review (2025) | Government of Canada | Canada | 1 | A | The regulator that licenses more than 3,000 turmeric products, warning about its own licensed category. All 12 Canadian cases had too little clinical information or other possible causes. |
| TGA final decision on curcumin label warnings (December 2023) | Australian Government | Australia | 1 | A | Published industry objections alongside its reasons. Focused on harms, not benefits. |
| Medsafe early warning on turmeric (2018) | New Zealand Government | New Zealand | 1 | A− | No product to sell. Based on a single spontaneous report plus known pharmacology. |
| DILIN turmeric case series (2023) | NIH (NIDDK) cooperative agreement with NCI support (full text) | US | 2 | A− | Expert-adjudicated causality and chemical testing of products; no conflicts declared. Only 10 cases, so it cannot estimate how common injury is. |
| Nelson et al. 2017, J Med Chem | NCCIH and NIH Office of Dietary Supplements grant U41AT008706 for two authors (full text) | US | 2 | B+ | No competing interest declared; a critique of a compound NIH itself has funded heavily. A chemistry review, not a clinical meta-analysis, and written before several depression trials. |
| Cheng et al. 2025 network meta-analysis | Taiwan Ministry of Science and Technology (full text) | Taiwan | 2 | B | Public funding; authors declare none. Network methods combine indirect comparisons of many small trials, which can inflate effects. |
| Forsyth et al. lead-adulteration studies (2019–2024) | Stanford University; project funded by the Stanford King Center on Global Development (King Center) | US / Bangladesh | 2 | A− | Laboratory testing of real market samples; no competing interests declared. Covers loose spice in South Asia, not capsules sold in Western pharmacies. |
| Fusar-Poli et al. 2020 meta-analysis | University of Catania; funding statement not accessible to us | Italy | 3 | B− | Academic team with no product. But pooled a handful of small trials, several using maker-supplied branded extracts, and reported an implausibly large anxiety effect (g −2.62). |
| WFSBP / CANMAT nutraceutical guideline (2022) | Professional societies; co-authors include the lead investigator of the BCM-95 depression trials, who declares grants or honoraria from curcumin makers | International / Australia | 3 | B− | Transparent grading and disclosures. The curcumin recommendation rests on one meta-analysis and was written partly by a researcher funded by curcumin companies. |
| Lopresti et al. 2014 BCM-95 trial | Murdoch University; BCM-95 from Arjuna Natural Extracts; described as having "industry support" (Andrade 2014) | Australia / India | 4 | C | Double-blind and placebo-controlled, with a mixed result reported honestly. Only 56 people and 8 weeks; the extract maker used it in marketing. |
| Cox et al. 2020 Longvida trial | Grant from Verdure Sciences, maker of Longvida; three authors declare supplement-industry funding | Australia / US | 4 | C+ | Pre-registered and states the funder had no role. Healthy older volunteers, not people with depression; curcumin group had raised blood glucose. |
Note on clustering: the safety evidence is spread across the EU, Canada, Australia, New Zealand and the US. The efficacy trials cluster in Australia, India, Iran, China and Thailand, and the most cited Australian trials come from one research group working with Indian and US extract makers.
What curcumin is, and its legal status
Turmeric is the dried underground stem (rhizome) of Curcuma longa, a plant in the ginger family native to Asia (Health Canada). Curcumin is the main curcuminoid in turmeric and makes up up to about 5% of it (Nelson 2017). The other two curcuminoids are demethoxycurcumin and bisdemethoxycurcumin.
Forms that matter
- Culinary turmeric: the spice in food. Health Canada and the TGA say turmeric in typical food amounts has not been identified as a safety concern (Health Canada).
- Standardised extracts: concentrated curcuminoids. Ordinary curcumin is poorly absorbed; in one human study it was not detected in most blood samples even after a 4 g dose (Nelson 2017, full text).
- "Enhanced bioavailability" products: extracts combined with black-pepper piperine, lecithin (Indena's Meriva), solid lipid particles (Verdure's Longvida), turmeric oils (Arjuna's BCM-95) or nano-formulations. NCCIH notes that liver damage "has been reported in some people who have consumed these bioavailable formulations" (NCCIH). A German study of 125 products in five countries found 34% did not state their curcuminoid content (Rahim-Mahdy 2025).
Legal status by country
| Place | Status | Source |
|---|---|---|
| European Union | Traditional herbal medicine for mild digestive problems only, such as fullness and flatulence. Monograph revised in 2019; a 2025 review found no reason to change it. No depression indication. | EMA; EMA addendum |
| Australia | Listed medicines with curcumin as an active ingredient must carry a liver-injury warning. New rules began 1 March 2024 with a transition until 1 March 2025. More than 600 medicines listed on the Australian Register of Therapeutic Goods contained curcuma species or curcumin in 2023. | TGA; RACGP newsGP |
| Canada | More than 3,000 licensed products, authorised for joint pain, inflammation and as antioxidants, not depression. Health Canada expects all licensed labels to list liver-injury symptoms. Most are not authorised for under-18s or people with biliary disorders. | Health Canada |
| Italy | After a hepatitis cluster, the Ministry of Health required label advice against use with biliary problems, in pregnancy and breastfeeding or for long periods, and banned liver, digestive and joint function claims (as summarised by the TGA). | TGA; Lombardi 2020 |
| France | The food agency ANSES set a maximum of 153 mg curcumin a day from supplements for a 60 kg adult (for standard, not enhanced, forms) and advised against use with bile duct disease. | TGA summary of ANSES 2022 |
| United States | Sold as a dietary supplement without pre-market proof of benefit. The US Pharmacopeia recommended in 2026 that its turmeric monographs carry a rare-liver-problems caution. | USP 2026 |
How it might work, and the chemistry critique
- Proposed (mainly animal and laboratory): trial authors cite effects on serotonin and other monoamines, inflammation, oxidative stress and the stress-hormone (HPA) axis (Lopresti 2014). None of these has been shown to explain a mood effect in people.
- In humans: one Chinese trial in 108 men on antidepressants reported lower inflammatory markers (IL-1β, TNF-α), higher BDNF and lower saliva cortisol with curcumin (Yu 2015). This is a single small study and the markers are indirect.
- The chemistry critique: Nelson and colleagues reviewed curcumin's medicinal chemistry and concluded it is unstable, reactive and poorly absorbed, and "a highly improbable lead". They classed it as a PAINS (pan-assay interference compound) and IMPS (invalid metabolic panacea): it gives false "hits" in many lab tests by sticking to proteins, clumping and interfering with readouts. They noted NIH had awarded more than $150 million to curcumin-related projects (Nelson 2017). This does not prove that curcumin cannot affect mood, but it means lab and animal studies cannot be taken as support.
- Bioavailability as a business: makers answer the absorption problem with patented formulations. Verdure says Longvida is absorbed up to 65 times better than "non-optimized" curcumin, a company claim (Nutritional Outlook). Regulators now link the same higher exposure to liver risk (see below).
What works and what does not
| Claimed benefit | Verdict | Evidence | Key caveat |
|---|---|---|---|
| Major depression, as an add-on to antidepressants | MIXED | SMD 1.03 better than antidepressant alone (Cheng 2025); positive 12-week Thai trial (Kanchanatawan 2018) | Few, small, short trials; high heterogeneity |
| Major depression, on its own | INSUFFICIENT | Not significant vs antidepressants (Cheng 2025); partial effect vs placebo (Lopresti 2014) | Do not replace prescribed treatment |
| Depressive symptoms in people with other illnesses | MIXED | Included in pooled analyses (Fusar-Poli 2020); no effect on depression in obese adults (Esmaily 2015) | Inconsistent results |
| Anxiety | INSUFFICIENT | Large pooled effects with extreme heterogeneity (Fathi 2024); moderate effect in 2 low-bias trials (Fornaro 2026) | No trials in diagnosed anxiety disorders that we found |
| Mood and fatigue in healthy older people | INSUFFICIENT | Longvida trials (Cox 2015; Cox 2020) | Funded by the extract maker; not depression |
Benefits by claim
Does curcumin work for depression?
The first meta-analysis, from the National University of Singapore, pooled six trials with 377 people. Curcumin lowered depression scores more than placebo, with an SMD of −0.34 (−0.56 to −0.13) (Ng 2017). In plain terms, that is a small effect. The authors noted that all trials lasted only 4 to 8 weeks and that there were too few studies to test for publication bias.
A 2020 Italian meta-analysis found a larger effect: Hedges' g −0.75 (−1.11 to −0.39) across 10 studies and 531 people, and an anxiety effect of −2.62 (Fusar-Poli 2020). An effect that large, from a few small studies, should be treated with great caution. The authors asked for cautious interpretation and more trials in Western countries.
The most recent review (19 trials) again found benefit, SMD −0.76 for diagnosed depression, but heterogeneity was 94.5% and the authors judged the result "fragile" on sensitivity analysis (Musazadeh 2026). A 2021 meta-review co-written by the BCM-95 trial lead found "supportive" evidence for curcumin but warned of possible "significant publication bias" (Sarris 2021).
An independent psychiatrist's critique of the first four trials concluded they provided "no convincing evidence" that patients do better on curcumin than placebo. He pointed to small samples, three different extracts with inconsistent results, and a lack of randomisation or blinding in the most favourable study (Andrade 2014).
The individual trials
- BCM-95, 8 weeks, 56 people: both groups improved in the first 4 weeks; curcumin pulled ahead only from weeks 4 to 8 (p=0.045 on the main self-rated scale). The authors described "partial support" (Lopresti 2014). Arjuna, the extract maker, later promoted the result in a press release targeting the European market (Arjuna 2015).
- Curcumin, two doses, or curcumin plus saffron, 12 weeks, 123 people: the three active groups combined beat placebo (p=0.031), but no dose or combination beat another (Lopresti 2017). The main comparison pooled three active arms against one placebo arm.
- Thai add-on trial, 12 weeks, 65 people: curcumin (500 rising to 1,500 mg a day) beat placebo on the MADRS from week 12, with effects stronger in men (Kanchanatawan 2018).
- Chinese add-on trial, 6 weeks, 108 men: lower Hamilton and MADRS scores with 1,000 mg a day (Yu 2015). We could not access the funding statement.
- Indian comparison with fluoxetine, 6 weeks, 60 people: no significant difference between curcumin, fluoxetine and both together, with no placebo group (Sanmukhani 2014). A trial this small cannot show that curcumin is "as effective" as an antidepressant, though trade press later described an earlier trial that way (SupplySide).
What do guidelines say?
No national guideline we found recommends curcumin for depression. The WFSBP/CANMAT taskforce rated curcumin 500–1,000 mg a day "Provisionally Recommended" for mild to moderate major depression, based on the 2020 meta-analysis (k=10, n=531) (WFSBP/CANMAT 2022). The same guideline lists the BCM-95 trial lead among its authors, with declared grants or honoraria from several curcumin suppliers (see the money section). NCCIH, by contrast, says the evidence is not enough to conclude (NCCIH).
Does turmeric help anxiety?
A 2024 meta-analysis of 8 trials (567 people) reported SMD −1.56 (−2.48 to −0.64), with heterogeneity of 95.6% (Fathi 2024). When a 2026 network meta-analysis kept only low-risk-of-bias trials, curcumin's anxiety effect shrank to −0.58 (−0.99 to −0.18) from just 2 trials and 89 people, and overall confidence in the evidence was "low to very low" (Fornaro 2026). Most anxiety data are secondary outcomes in depression trials or come from people with physical illnesses. One crossover trial in 30 obese adults found less anxiety but no change in depression (Esmaily 2015).
Mood, stress and fatigue in healthy people
Two Australian trials of Longvida in healthy older adults reported less fatigue and some improved mood scores (Cox 2015; Cox 2020). The 2020 trial was funded by Longvida's maker, Verdure Sciences, and the curcumin group had raised blood glucose. These were not people with depression.
Risks and all side effects
The Italian cluster
Between late 2018 and mid-2019, about 20 hepatitis cases in Tuscany and Lombardy were linked to turmeric supplements, and Italy's Ministry of Health ruled out contaminants (NutraIngredients 2026). In seven Tuscan cases reported up to September 2019, every case involved "high bioavailability" formulations and high curcuminoid doses, and most improved after stopping. A systematic review in the same paper found 23 more published cases (Lombardi 2020). The EU herbal committee noted these cases but found no liver toxicity at the low doses in its monograph, so it did not change the monograph (EMA addendum 2025).
Australia's warning
In August 2023 the TGA reported 18 liver-injury reports, including one death, and said risk "may be higher" with enhanced-absorption or high-dose products (RACGP newsGP). Its final decision cited "at least three deaths" and rejected a "wait and see" approach. Three clinical bodies supported a label warning; five industry representatives opposed it or gave mixed support (TGA).
How common is it?
Nobody knows. Clinical trials have not reported organ toxicity, but most are far too small to detect a rare reaction (USP 2026). The TGA pointed out that detecting a reaction rarer than 1 in 10,000 needs more than 30,000 people. Herbal and dietary supplements now account for about 20% of drug-induced liver injury cases in the US network, with turmeric among the main named agents (Halegoua-DeMarzio 2024).
Other side effects
| Effect | What the evidence shows | Source |
|---|---|---|
| Nausea, vomiting, acid reflux, stomach upset, diarrhoea, constipation | The most common side effects of oral turmeric | NCCIH |
| Prolonged bleeding | Antiplatelet and NSAID-like effects reported | Medsafe |
| Raised blood glucose | Seen in the Longvida group in one 12-week trial | Cox 2020 |
| Gallbladder and bile duct problems | Not recommended with biliary disorders | Health Canada |
All interactions
| Interacting substance | What can happen | Seriousness | Source |
|---|---|---|---|
| Warfarin and other anticoagulants | INR rose from stable to over 10 within weeks of starting turmeric in a reported case. Medsafe says combined use "should be avoided". | Avoid | Medsafe |
| SSRI antidepressants, antiplatelets, NSAIDs | Medsafe lists these as medicines with which curcumin may prolong bleeding and says combined use "should be avoided". Relevant because people with depression often take SSRIs. | Avoid; ask a pharmacist | Medsafe |
| Medicines that can harm the liver; any regular medicines | Most published liver cases involved other medicines or supplements. Health Canada labels advise asking a professional first if taking medications. | Ask a doctor | Lombardi 2020; Health Canada |
| Tamoxifen, endoxifen, talinolol, sulfasalazine | Interactions seen in clinical studies at curcumin doses above the EU monograph's. Curcumin affects the CYP3A4 enzyme and P-glycoprotein transporter. | Ask a specialist | EMA addendum |
| Cancer drugs and immunosuppressants | France's ANSES warned of interaction risk. | Ask a specialist | NutraIngredients |
| Black pepper (piperine) | Combined to raise absorption; found in several US liver-injury products. | Caution | NCCIH; DILIN |
| Tacrolimus, paclitaxel | No meaningful change in drug levels in small studies of turmeric. | Low | EMA addendum |
Who should avoid curcumin supplements
- People with current or past liver disease: the TGA advised doctors against offering turmeric supplements to them (RACGP newsGP).
- People with gallstones or bile duct disease: not recommended (Health Canada).
- Pregnancy and breastfeeding: NCCIH says turmeric supplements in pregnancy "may be unsafe" (NCCIH). Italy's labels advise against use in pregnancy and breastfeeding (TGA summary). Lead-contaminated spice is a particular risk in pregnancy (Jahir 2021).
- Children and teenagers: most Canadian products are not authorised for under-18s, and no depression trials in young people were found.
- People on warfarin, other blood thinners, SSRIs or NSAIDs, and anyone due for surgery, because of bleeding risk (Medsafe).
- Anyone with moderate to severe depression or suicidal thoughts: trials excluded people with suicidal ideation (Sanmukhani 2014), and curcumin has not been tested in severe illness.
Doses studied and product quality
These are doses used in research or set by regulators. They are not recommendations.
| Setting | Dose | Source |
|---|---|---|
| Murdoch University depression trials (BCM-95 in 2014) | 250 mg or 500 mg twice daily for 8–12 weeks | Lopresti 2014; Lopresti 2017 |
| Add-on trials | 1,000 mg a day for 6 weeks; 500 rising to 1,500 mg a day for 12 weeks | Yu 2015; Kanchanatawan 2018 |
| Longvida mood trials | 400 mg a day (80 mg curcumin) | Cox 2020 |
| EU herbal monograph (digestion) | Maximum 4 g turmeric a day, equal to at most about 209 mg curcuminoids | EMA addendum |
| France (ANSES, supplements) | No more than 153 mg curcumin a day for a 60 kg adult, standard forms only | TGA summary |
| Acceptable daily intake (food additive) | 3 mg/kg body weight, set by JECFA in 2004; the TGA notes it predates the liver-injury reports | TGA |
Note the gap: depression trials used 500–1,500 mg a day, often of enhanced-absorption extracts, well above the EU monograph and French limits, which were set for safety rather than mood.
Product quality problems
- 34% of 125 products from five countries did not disclose curcuminoid content (Rahim-Mahdy 2025).
- In 2016, routine New York State sampling led Gel Spice to recall ground turmeric for excess lead (Food Safety News).
- In Bangladesh, media coverage, consumer education and Food Safety Authority enforcement cut the share of market turmeric with detectable lead from 47% in 2019 to 0% in 2021 (Forsyth 2023).
Follow the money: who makes it and who funded the evidence
Who makes and sells it
- Arjuna Natural (India): maker of BCM-95, founded in 1989 in Aluva, Kerala (supplier profile). It issued a 2015 press release using the 2014 depression trial to launch BCM-95 in Europe (Arjuna). US trade press described BCM-95 as manufactured by Dolcas-Biotech LLC (SupplySide).
- Indena (Italy): Milan-area botanical extract firm, operating since 1921 (Indena), maker of the lecithin formulation Meriva (SupplySide). We found no Meriva trial in diagnosed depression.
- Verdure Sciences (US): Noblesville, Indiana, maker of Longvida (Nutritional Outlook); funder of the 2020 Longvida mood trial (Cox 2020).
- Market size: the Canadian turmeric and curcumin natural-health-product market was estimated at about CAN$165 million in 2021, according to trade press (NutraIngredients). We found no independent global figure.
Who funded the trials and reviews
- BCM-95 trials (Australia): an independent reviewer described the 2014 trial as having industry support (Andrade 2014). The trial lead declares presentation honoraria or clinical-trial grants from Arjuna Natural, Dolcas-Biotech, Pharmactive Biotech Products, Verdure Sciences and Sabinsa, among others (WFSBP/CANMAT 2022; Sarris 2021).
- Guideline authors: the WFSBP/CANMAT lead author declares payments from many supplement and pharmaceutical firms, and the curcumin trial lead is a co-author (WFSBP/CANMAT 2022).
- Longvida trials: funded by Verdure Sciences; three authors declare supplement-industry funding (Cox 2020).
- Public funders: NCCIH and the NIH Office of Dietary Supplements funded two authors of the PAINS critique (Nelson 2017); NIDDK funds the liver-injury network (DILIN); Taiwan's science ministry funded the 2025 network meta-analysis (Cheng 2025). NCCIH itself is funded by US Congressional appropriations (NCCIH).
Documented integrity and regulatory events
- Australia's industry body, Complementary Medicines Australia, responded to the TGA warning by arguing that the cases showed "correlation" not causation, and that the person who died was also taking naproxen (NutraIngredients). The TGA went ahead with the warning (TGA).
- In 2025 Health Canada told licence holders to add liver-injury warnings to the labels of all its more than 3,000 licensed products (Health Canada). The US Pharmacopeia recommended a similar caution in 2026 (USP).
- Bangladesh's Food Safety Authority used rapid lead testing to enforce penalties against adulteration after the Stanford work (Forsyth 2023).
Research funding at a glance
12 sources graded. The counts below are the independence tiers assigned in the scorecard above. They count sources, not funding amounts or evidence quality.
Public funding alone does not prove independence, and commercial funding alone does not prove a claim false. Named funders, countries and relationships are in the scorecard and source notes.
The pattern is clear: the sources that most strongly support curcumin for mood are the ones closest to extract makers, while the most independent sources (regulators and NIH-funded researchers) focus on safety and on doubts about the chemistry. That pattern does not prove the positive trials wrong. It does mean they need independent replication before anyone relies on them.
Related research
For wider, evidence-graded context, see these Pure City Research guides:
- Major depressive disorder: assessment, treatment evidence and safety
- Supplements for stress, anxiety and depression: the evidence
- Saffron for depression
- St John's wort
- St John's wort vs saffron vs SSRIs
- SAMe (S-adenosylmethionine)
- Omega-3
- Generalised anxiety disorder
- Drug and supplement induced liver injury
- Anticoagulants and bleeding safety
- Gallstones
- The COA transparency audit of supplement brands
Frequently asked questions
Does curcumin work for depression?
Possibly a little, over a few weeks, mainly as an add-on to antidepressants. Meta-analyses of small trials are positive, but effect sizes vary from small to large and one recent review called its result "fragile" (Musazadeh 2026). NCCIH says the evidence is not enough to conclude (NCCIH).
Can turmeric replace an antidepressant?
No trial shows that. The only head-to-head trial had 60 people and no placebo group (Sanmukhani 2014), and a network meta-analysis found curcumin alone was not significantly different from antidepressants, with a wide margin of uncertainty (Cheng 2025). Talk to your prescriber before changing any treatment.
Is it safe to take curcumin with sertraline or another SSRI?
Trials gave curcumin alongside antidepressants without reported serious harm (Kanchanatawan 2018). But New Zealand's regulator lists SSRIs among medicines with which curcumin may prolong bleeding, and says combined use "should be avoided" (Medsafe). Ask a pharmacist, especially if you also take aspirin, an NSAID or a blood thinner.
Does turmeric help anxiety?
The evidence is weak. Pooled effects are very large but inconsistent, and in the two low-bias trials the effect was moderate and based on 89 people (Fornaro 2026).
Can turmeric supplements damage your liver?
Rarely, yes. Regulators in Australia, Canada and Italy have added warnings, and US cases have included a death (DILIN; Health Canada). The risk appears higher with enhanced-absorption products and in people carrying HLA-B*35:01. Turmeric used in cooking has not been identified as a concern.
Is turmeric with black pepper more effective for mood?
Piperine raises absorption (NCCIH), but we found no good trial showing it improves mood more. Products with piperine appeared in several liver-injury cases (DILIN), and the EU herbal committee noted pepper extracts in reported cases (EMA addendum).
Does turmeric thin the blood?
Curcumin has antiplatelet effects, and a warfarin user's INR rose above 10 after starting turmeric (Medsafe). Tell your doctor or anticoagulation clinic before taking it.
Is there lead in turmeric?
Some loose turmeric from South Asia has been adulterated with lead chromate to brighten its colour (Forsyth 2024), and US recalls have occurred (Food Safety News). The pigment is added to make poor-quality roots look brighter yellow (Forsyth 2019).
Sources and funding notes
What we could not verify: the funding statements of the Ng 2017 and Fusar-Poli 2020 meta-analyses, the Musazadeh 2026 review, and the Yu 2015, Kanchanatawan 2018, Sanmukhani 2014 and Lopresti 2017 trials (publisher pages were paywalled or blocked). We could not confirm from the paper itself who funded the 2014 BCM-95 trial, so we rely on an independent reviewer's description and the maker's press release; we could not open the 2017 trial's registry record. The NIH LiverTox page on turmeric was blocked by a bot check, so we used the DILIN paper instead. The TGA's own alert page and the UK Committee on Toxicity's 2024 statement did not load; we used the TGA final-decisions document and a professional news report. We did not verify Italian Ministry of Health documents directly (we used the TGA's summary and the Italian phytovigilance paper), the ownership structure of Arjuna, Indena or Verdure, or any global market size.
- NCCIH: Turmeric (April 2025) — US government fact sheet; funded by Congress (appropriations history).
- EMA: Curcumae longae rhizoma herbal monograph page — EU regulator.
- EMA/HMPC addendum to assessment report on Curcuma longa (19 November 2025) — EU regulator.
- Health Canada: Summary Safety Review, turmeric and curcuminoids for oral use — Canadian regulator.
- TGA: Final decisions, low-negligible risk changes to the Permissible Ingredients Determination 2023–2024 (December 2023) — Australian regulator.
- RACGP newsGP: Turmeric supplement warning after fatality (17 August 2023) — professional college news report of the TGA alert.
- Medsafe: turmeric/curcumin and warfarin early warning (2018) — New Zealand regulator.
- Halegoua-DeMarzio D et al. Liver injury associated with turmeric: ten cases from DILIN. Am J Med 2023 — NIDDK/NCI funded; no conflicts declared.
- Halegoua-DeMarzio D, Navarro V. Challenges in herbal-induced liver injury. Liver Int 2025 — academic review; no competing interest.
- Lombardi N et al. Acute liver injury following turmeric use in Tuscany. Br J Clin Pharmacol 2021 — University of Florence and Italian National Institute of Health authors; funding statement not accessed.
- Akhtar N et al. Hepatotoxicity with turmeric supplements: a comprehensive review by USP. Pharm Biol 2026 — US Pharmacopeia expert committee (non-profit standards body); no conflicts reported.
- Nelson KM et al. The essential medicinal chemistry of curcumin. J Med Chem 2017 (full text) — NCCIH/ODS grant for two authors; no competing interest.
- Ng QX et al. Clinical use of curcumin in depression: a meta-analysis. J Am Med Dir Assoc 2017 — Singapore academic authors; funding statement not accessed.
- Fusar-Poli L et al. Curcumin for depression: a meta-analysis. Crit Rev Food Sci Nutr 2020 — University of Catania; funding statement not accessed.
- Musazadeh V et al. Efficacy of curcumin in depression: GRADE-assessed meta-analysis. Ann Gen Psychiatry 2026 — Iranian universities; no competing interests declared.
- Fathi S et al. Curcumin and anxiety: meta-analysis. Clin Nutr ESPEN 2024 — academic; no competing interest declared.
- Cheng YC et al. Nutraceuticals for depressive disorder: network meta-analysis. Psychol Med 2025 (full text) — Taiwan Ministry of Science and Technology; authors declare none.
- Fornaro M et al. Nutraceuticals and phytoceuticals for major depressive disorder: network meta-analysis. Mol Psychiatry 2026 — academic; several authors declare pharmaceutical payments, one declares supplement-company research support.
- Sarris J et al. Phytoceuticals in psychiatric disorders: a meta-review. Can J Psychiatry 2021 — co-authored by the BCM-95 trial lead, who declares curcumin-company grants.
- Sarris J et al. WFSBP/CANMAT clinician guidelines on nutraceuticals and phytoceuticals. World J Biol Psychiatry 2022 (journal page) — professional societies; extensive author industry disclosures.
- Andrade C. A critical examination of studies on curcumin for depression. J Clin Psychiatry 2014 — independent commentary; no disclosure seen on the page.
- Lopresti AL et al. Curcumin for major depression. J Affect Disord 2014 — BCM-95 from Arjuna; industry support per Andrade.
- Lopresti AL, Drummond PD. Curcumin and saffron/curcumin for major depression. J Affect Disord 2017 — funding statement not accessed.
- Kanchanatawan B et al. Add-on curcumin in Thai patients with major depression. Neurotox Res 2018 — academic; funding statement not accessed.
- Yu JJ et al. Curcumin as add-on in major depressive disorder. J Clin Psychopharmacol 2015 — funding statement not accessed.
- Sanmukhani J et al. Curcumin vs fluoxetine in major depressive disorder. Phytother Res 2014 — Indian academic hospital; funding statement not accessed.
- Esmaily H et al. Curcumin for anxiety and depression in obese individuals. Chin J Integr Med 2015 — academic; funding statement not accessed.
- Cox KH et al. Solid lipid curcumin, cognition and mood in older people. J Psychopharmacol 2015 — Longvida trial; funding statement not accessed.
- Cox KHM et al. Lipidated curcumin, mood and working memory. Nutrients 2020 (full text) — funded by Verdure Sciences.
- Forsyth JE et al. Lead chromate pigments added to turmeric in Bangladesh. Environ Res 2019 — Stanford-led; project funded by the Stanford King Center on Global Development.
- Forsyth JE et al. Food safety enforcement reduces lead chromate adulteration of turmeric in Bangladesh. Environ Res 2023 — academic; no competing interests declared.
- Forsyth JE et al. Turmeric adulteration with lead chromate across South Asia. Sci Total Environ 2024 — academic; no competing interests declared.
- Jahir T et al. Reducing lead exposure among children and caregivers in rural Bangladesh. Environ Res 2021 — academic; no competing interests declared.
- Rahim-Mahdy H, Seifert R. Market and risk assessment of 125 turmeric supplements. Naunyn Schmiedebergs Arch Pharmacol 2025 — Hannover Medical School; senior author is the journal's editor.
- Arjuna Natural Extracts press release (12 November 2015) — manufacturer marketing.
- SupplySide Supplement Journal: BCM-95 curcumin improves depression symptoms — trade press.
- Nutrition Insight: Arjuna Natural supplier profile — trade directory.
- Indena company website and SupplySide: Meriva named best new product (2010) — company and trade press.
- Nutritional Outlook: Verdure Sciences launches Longvida (2010) — trade press reporting company claims.
- NutraIngredients: Health Canada calls for warnings on turmeric/curcumin labels (15 December 2025) — trade press.
- NutraIngredients-Asia: Complementary Medicines Australia response (29 August 2023) — trade press quoting an industry body.
- Food Safety News: ground turmeric recalled for excessive lead (2016) — non-profit news site.
Educational research reviewed 6 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.
Have a question — or want us to cover something?
Ask about anything on this page, or request the next deep dive: an ingredient, a supplement, or a health concern. We use published research, evidence syntheses, and regulatory guidance, with clear source links.
One daily research roundup
Get the topics, key findings and links from our new articles in one email. At most one digest a day, only when there is something new.