Turmeric and ginger as anti-inflammatories: what the dose changes

A level teaspoon of ground turmeric, roughly 3 g, delivers about 70 mg of curcumin. The clinical trials that find an effect on joint pain use 500 to 1500 mg a day, as concentrated extract. That is a gap of at least seven to one, and it is the number most articles about anti-inflammatory spices never put on the table before everything else.

Keep the spice rack. But know what you are asking of it, and stop treating a capsule and a curry as two versions of the same thing.

A teaspoon of turmeric is not a curcumin capsule

Turmeric rhizome contains between 1 and 5 % curcuminoids, depending on variety, origin and how long the jar has been open. Curcumin itself accounts for roughly 75 to 80 % of those curcuminoids, the rest being demethoxycurcumin and bisdemethoxycurcumin. These figures are cited by the German Federal Institute for Risk Assessment in its 2021 opinion on curcumin in food supplements.

A level teaspoon of turmeric, 3 g, at 3 % curcuminoids gives 90 mg of curcuminoids, so roughly 70 mg of curcumin. In a rich 5 % powder you reach 116 mg. In a powder flattened by two years in a cupboard you can drop to 23 mg. To match a common 500 mg curcumin capsule you would need to swallow between 13 and 21 g of turmeric powder, four to seven teaspoons, in a single day, every day.

Nobody cooks like that. A generous curry for four rarely holds more than two teaspoons of turmeric in total, which puts each diner at around 35 mg of curcumin, only a fraction of which will reach the bloodstream. Our guide to the art of homemade curry gives the usual proportions for a paste that feeds four.

The consequence is blunt: trial results obtained with concentrated extracts do not transfer to the plate. When an article claims turmeric reduces inflammation and leans on those trials, it has quietly swapped one subject for another.

The acceptable daily intake sets the frame

Curcumin is also the food colour E100. In that capacity it was assessed by the Joint FAO/WHO Expert Committee on Food Additives in 2004, then by the European Food Safety Authority in 2010.

The toxicological starting point is a two-generation feeding study in rats. Body weight gain was reduced in the F2 generation at the highest dose, corresponding to 850 to 1100 mg per kilogram of body weight per day. The no observed effect level sat between 250 and 320 mg/kg. Applying the standard safety factor of 100, the acceptable daily intake was set at 3 mg of curcumin per kilogram of body weight per day.

For a 70 kg adult that works out at 210 mg of curcumin a day, from all sources combined: spices, food colours, supplements.

A teaspoon of 3 % turmeric therefore accounts for about a third of that daily allowance. Two well-filled teaspoons of a rich powder and you are brushing the ceiling. This is not an alarm, because the acceptable daily intake describes what can be consumed every day across a lifetime without detectable risk, not a toxicity threshold. But it puts the spice in its real position: culinary turmeric already occupies a noticeable slice of the envelope. There is no unlimited room to pile more on.

The European authority also estimates that the use of E100 as a colour already produces mean intakes of 0.2 to 0.6 mg/kg in adults, with maximum intakes of 0.4 to 1.5 mg/kg. In other words, part of the envelope is spent without anyone noticing, through industrial products coloured with turmeric. Knowing how to read a nutrition label helps you spot E100 in an ingredient list.

Black pepper: the effect is real, and that is exactly what makes it awkward

The advice to pair turmeric with black pepper is correct in principle. It deserves closer reading than it usually gets.

The reference study is Shoba and colleagues, published in Planta Medica in 1998. In human volunteers given 2 g of curcumin, adding 20 mg of piperine raised curcumin bioavailability by a factor of twenty. The 2000 % figure that circulates everywhere comes from there. In rats, in the same study, the factor was only 1.5.

The trouble is that this spectacular result does not reproduce reliably. In another human study cited by the German institute, with 2 g of curcumin and only 5 mg of piperine, the increase was down to a factor of two. In rats, across different piperine doses, factors of 1.2 to 1.7 have been reported, and one paper found no increase at all. The size of the effect depends on the piperine dose, on the ratio between the two substances, and probably on the individual.

Above all, and this is the point golden latte recipes skate past: raising the bioavailability of a substance also raises its toxic potential. The German opinion says so plainly. The reported liver injury cases associated with curcumin-containing products mostly involve preparations with improved bioavailability, frequently through added piperine.

In the kitchen the stakes stay modest. A turn of the mill over a curry contributes a few milligrams of piperine against a few tens of milligrams of curcumin. The pairing makes sense on taste alone, and causes no problem at these amounts. It is on a supplement label that the word piperine deserves careful attention, not in a frying pan.

What the authorities recognise, and what they do not

European Regulation 432/2012 establishes the list of health claims permitted on foods. It contains close to two hundred of them. Iron is there, so are zinc, oat beta-glucans, vitamin D and rye fibre.

Turmeric is not. Neither is ginger. No claim relating to inflammation appears at all, for any food. This does not mean the spices are inert. It means no dossier has cleared the level of proof required for a manufacturer to print “reduces inflammation” on a European package.

The Committee on Herbal Medicinal Products at the European Medicines Agency has published monographs on both plants. They repay reading closely, as long as you read what they actually say.

For turmeric, recognition rests on traditional use, for the relief of mild digestive complaints. The posology is 0.5 to 1 g of powder, two to three times daily, in adults only. Traditional use means there is insufficient clinical evidence, but the effect is plausible and the use has been documented for at least thirty years. No anti-inflammatory indication is granted.

For ginger the situation is different and more favourable. Prevention of nausea and vomiting in motion sickness falls under well-established use, meaning it is backed by published clinical studies, a category turmeric does not reach for any indication. Relief of mild spasmodic gastrointestinal complaints, bloating, temporary loss of appetite, minor joint pain and cold symptoms fall under traditional use.

Ginger therefore has one clinically established indication, and it is not inflammation. It is motion sickness.

Knee osteoarthritis, the one place where trials converge

If a human effect of concentrated curcumin exists, this is where it shows most clearly.

A meta-analysis of meta-analyses published in Phytotherapy Research in 2024 pooled eleven meta-analyses of randomised controlled trials. It concludes that curcumin significantly reduces the visual analogue pain score and the WOMAC pain, function and stiffness scores.

A Bayesian network meta-analysis in the Journal of Ethnopharmacology the same year covered 23 studies, seven countries and 2175 patients. Against placebo, curcumin lowered the visual analogue pain score by 1.63 points, with a confidence interval running from 2.91 to 0.45 points, and the total WOMAC score by 18.85 points. Patients on curcumin had fewer adverse events than those on non-steroidal anti-inflammatory drugs.

One detail in this literature is worth pausing on. A 2021 meta-analysis of eleven studies and 1258 participants compared high doses, from 1000 mg a day upwards, with lower ones. It found no significant difference in pain relief between them. If the effect does not track the dose above 1000 mg, that still gives no licence to extrapolate downwards to the few tens of milligrams in a spiced dish. No trial has tested that range.

On blood markers the picture is more mixed than the headlines suggest. A 2025 meta-analysis of 21 studies and 1705 knee osteoarthritis patients found significantly lower C-reactive protein and TNF-alpha under curcumin, but no significant difference in erythrocyte sedimentation rate, interleukin 1 beta, interleukin 6 or prostaglandin E2. Moving two markers out of six is not the same as reducing inflammation, and the distance between those two statements is exactly the distance between a finding and a slogan.

Ginger and joints: two meta-analyses, two conclusions

Ginger illustrates what thin evidence looks like in practice.

A meta-analysis in Osteoarthritis and Cartilage in 2015, covering five trials and 593 patients, found a statistically significant reduction in pain, with a standardised mean difference of 0.30, and in disability, at 0.22. Modest effects, then, but real ones. The same paper notes that patients given ginger were more than twice as likely to discontinue treatment as those on placebo, because of adverse events, mostly digestive. The authors rate the quality of the evidence as moderate.

A systematic review in Pain Physician in 2020 revisited the question with different inclusion criteria. From seven eligible trials and four retained for quantitative synthesis, it concludes there is insufficient evidence to support oral ginger over placebo for pain relief and function in knee osteoarthritis.

Two teams, one question, two answers. This is not a methodological scandal, it is what happens when the trial count is low and quality uneven. It simply means that ginger’s effect on joints, if there is one, is too small or too unstable to be seen reliably with the data available.

Where ginger is solid, by contrast, is nausea. A 2021 meta-analysis of fourteen studies and 1417 participants shows reduced postoperative nausea severity and less recourse to rescue antiemetics. Another, covering fourteen randomised trials and 1506 patients, confirms the reduction in nausea without a significant difference on vomiting. That is consistent with the European monograph.

The contraindications spice articles leave out

Turmeric stimulates bile secretion. That is precisely why the European monograph advises against it in cases of bile duct obstruction, cholangitis, liver disease, gallstones and any other biliary condition. The same mechanism routinely sold as a digestive benefit becomes a problem when the duct is blocked.

The same monograph states that use in children and adolescents under 18 has not been established for lack of adequate data, and that if symptoms persist beyond two weeks a doctor should be consulted.

For ginger, the European monograph excludes medicinal use during pregnancy and breastfeeding, along with use under 18 for most indications and under 6 for motion sickness. These are dosed medicinal preparations, not a knob of ginger grated into a stir-fry, but the gap between the two narrows once you are drinking three strong infusions a day.

On curcumin supplements, the 2021 German opinion details suspected liver injury cases. Estimated intakes ranged from 40 to 1425 mg of curcumin a day, with time to symptoms running from eight days to eight months. An Italian series describes seven cases at 250 to 1812.5 mg a day over two to eight weeks. In the lowest-dose case, 250 mg in a 60 kg patient, the intake corresponded to 4 mg/kg, barely above the acceptable daily intake. In two thirds of the reported cases, intake ran at 1.6 to 7 times that figure.

These numbers need careful reading. The opinion explicitly concludes that no causal relationship between curcumin intake and the observed effects can be deduced from the available data, and that many of the preparations contained other ingredients. But the direction is clear, and it contradicts the idea that a turmeric supplement is harmless by nature because turmeric is a spice. Our page on dietary supplements for vegetarians applies the same grid to deficiencies that are actually documented.

Cinnamon carries a problem almost nobody mentions

Articles recommending a daily spoon of cinnamon on your porridge skip a step.

Chinese cinnamon, known as cassia, contains coumarin, a substance that affects the liver at high and prolonged doses. Ceylon cinnamon, Cinnamomum verum, contains only traces. The gap is roughly a hundredfold. In analyses by the Austrian Agency for Health and Food Safety, cassia cinnamon measured 1880 to 3260 mg of coumarin per kilogram, against a maximum of 185 mg/kg for Ceylon cinnamon.

The tolerable daily intake for coumarin is 0.1 mg per kilogram of body weight, so 6 mg a day for a 60 kg person. With cassia at 2570 mg/kg, that threshold is reached with about 2.3 g of cinnamon, which is one teaspoon. With Ceylon cinnamon you would need more than thirty grams.

Ground cinnamon sold in European supermarkets is predominantly cassia, cheaper and blunter in flavour. A study in the Journal of Agricultural and Food Chemistry in 2013 had already shown that the cinnamon in American flavoured foods and supplements was probably Indonesian cassia. An unlabelled jar will not tell you which one you bought. The sticks are distinguishable by eye: Ceylon cinnamon rolls up in fine multiple layers, cigar-like, while cassia forms a thick tube of a single bark.

None of this condemns cinnamon. An occasional pinch poses no problem, and maximum levels in industrial foods are framed by the European flavourings regulation. But the daily habit of a cinnamon spoonful in porridge, recommended almost everywhere for its blood sugar effects, is worth pairing with a glance at which variety you bought.

On those glycaemic effects, incidentally, the data diverge. Several meta-analyses find reductions in fasting glucose and glycated haemoglobin in people with type 2 diabetes, around 15 mg/dl for the former. A 2023 dose-response meta-analysis, by contrast, found no significant effect on either, while observing a drop in triglycerides. The doses used in these trials are capsule doses anyway, often above a gram a day. The same gap between promise and data runs through our review of seven everyday spices.

Lead in turmeric: a sourcing problem

This risk is not theoretical, and it concerns buying rather than cooking.

To achieve a brighter yellow and move poor-quality rhizomes, some processors add lead chromate to turmeric. A study in Science of the Total Environment in 2024 analysed 356 samples collected in 23 cities across India, Pakistan, Sri Lanka and Nepal. Fourteen per cent exceeded 2 µg of lead per gram. Samples above 18 µg/g showed a lead to chromium molar ratio close to 1:1, the signature of lead chromate adulteration. In Patna, Karachi and Peshawar some levels topped 1000 µg/g, enough to project child blood lead levels ten times the American threshold of concern.

A 2026 paper in NPJ Science of Food, covering 503 samples from 34 cities in eastern India, found 30 % above the Indian limit of 10 µg/g, with a geometric mean of 48 µg/g in Bihar and a maximum of 6416 µg/g.

The encouraging part is that the problem yields when it is tackled. In Bangladesh, an intervention combining public information, outreach to traders and enforcement with rapid lead detection cut the proportion of market samples containing detectable lead from 47 % in 2019 to 0 % in 2021. Blood lead levels in mill workers fell by a median of 30 %.

In practice, for a purchase in Switzerland or the European Union, turmeric falls under regulatory contaminant limits and official controls. The risk concentrates on powders brought back from travel or bought outside monitored channels. Turmeric of an unusually intense yellow-orange, very cheap, with no traceability, deserves suspicion.

Chronic inflammation, and what actually moves it

Acute inflammation is a normal response to injury: redness, heat, an influx of immune cells, repair. Nobody is trying to suppress that.

What spice articles are aiming at is something else: low-grade chronic inflammation, measured by blood markers such as C-reactive protein, and associated in epidemiology with cardiovascular disease, type 2 diabetes and several cancers. That association is robust. The causal link running the other way, from a dietary reduction in markers to a reduction in risk, is far less established.

Work on the dietary inflammatory index gives the measure of what is at stake. The index scores a whole diet according to the inflammatory potential of its components. Meta-analyses of observational studies find consistent associations between a high index and coronary artery disease risk, with roughly a 38 % increase between the highest and lowest quintile, as well as with overweight and fatty liver disease. But what weighs in that index are the large masses of the plate: fibre, vegetables, fruit, unsaturated fats, sugars, ultra-processed foods. Not the pinch of spice.

That is deflating, and also freeing. The levers that count are known, measured and recommended by the World Health Organization: at least 400 g of fruit and vegetables a day, at least 25 g of naturally occurring dietary fibre, less than 10 % of energy as free sugars, less than 10 % as saturated fat, less than 5 g of salt. The Swiss dietary recommendations point the same way.

A well-built vegetarian diet already ticks several of these boxes. Our page on essential pantry legumes and the complete guide to plant proteins cover that groundwork. Spices come afterwards, and their first merit lies elsewhere.

What spices genuinely contribute, and it is not small

Turmeric and ginger make dull food edible. On that point there is no uncertainty at all.

A bland plate of lentils is a plate of lentils you do not cook again. Cauliflower roasted with turmeric and cumin, a broth sharpened with fresh ginger, a carrot soup with some character: these dishes repeat, and repetition is what builds a diet. Our guide to essential spices and herbs and the one on measuring spices in a curry cover that dimension.

Ginger and pepper also let you cut salt without the dish turning sad, which our page on seasoning without salt works through. Reducing sodium intake has a measured effect on blood pressure. The health benefit is real, it simply travels a different route from the one advertised.

A few technical points are worth knowing at the stove. Ground turmeric disperses in hot fat at the start of cooking, which develops its colour and avoids the earthy taste of raw powder tipped in at the end. Grated fresh ginger loses its bite under long cooking, so holding some back for the final minutes keeps the heat. Fresh rhizome and powder are not interchangeable on taste, the powder being warmer and less lemony. To keep a cut rhizome, the methods described for storing fresh herbs work here too, since grated ginger freezes very well.

Common mistakes

Treating a curcumin supplement as a spice. This is the central confusion. One belongs to everyday eating, the other is a dosed product that calls for the same caution as an over-the-counter medicine.

Assuming a daily golden latte delivers a therapeutic dose. A cup holds half a teaspoon of turmeric at best, a few tens of milligrams of curcumin.

Adding pepper and expecting a guaranteed twentyfold effect. The amplification ranges from twofold to twentyfold across studies, and a turn of the mill supplies very little piperine.

Buying any cinnamon for daily use. Cassia and Ceylon do not pose the same coumarin problem.

Assuming “natural” exempts a plant from contraindications. Biliary obstruction for turmeric, pregnancy for dosed ginger preparations: these are agency warnings, not precautionary boilerplate.

Counting on spices to offset the rest of the plate. This is where the ratio of effort to result is least favourable.

By profile

In good health and cooking with spices daily, nothing needs to change. You stay comfortably below the acceptable daily intake and get the main benefit, which is eating more vegetables and legumes.

With knee osteoarthritis, the case for a curcumin extract rests on the most favourable data in the file. Trials use 500 to 1500 mg a day for at least twelve weeks. Those doses exceed the acceptable daily intake, which makes it a decision to discuss with a doctor, particularly alongside existing treatment.

On anticoagulants or long-term medication, any curcumin or ginger supplement goes to the doctor or pharmacist first. Ordinary cooking does not raise this issue.

During pregnancy, cooking ginger stays cooking ginger. What the European monograph excludes is dosed medicinal preparations, capsules and concentrated extracts included.

A history of liver or biliary disease is a specific reason to avoid curcumin supplements and to limit high-dose turmeric courses.

Children belong to the food conversation, not the supplement one. The European monographs rule out medicinal use of both plants below 18 for lack of data.

What this evidence does not say

The acceptable daily intake rests on a rat feeding study, with the factor of 100 that accompanies that kind of extrapolation. It has not been recalculated from long-term human data, which do not exist.

The curcuminoid content of retail turmeric is not labelled. The 1 to 5 % range is wide, and the calculations in this article give orders of magnitude, not values for your jar.

The osteoarthritis trials use standardised extracts whose composition varies between brands, including improved-bioavailability formulations that are not equivalent to one another.

The liver injury cases are pharmacovigilance reports, without control groups and often involving multi-ingredient products. They justify vigilance, not a causal conclusion.

The ginger meta-analyses rest on a small number of trials of uneven methodological quality, as their authors point out themselves.

Frequently asked questions

Should I stop using turmeric? No. At culinary doses it stays below the acceptable daily intake and improves the eating quality of legume dishes. It is the capsule that calls for thought, not the spice.

Is fresh turmeric better than powder? On taste it is livelier and less earthy. On curcumin content it holds considerably less per gram, since it is still full of water. It is not a way to raise the dose.

How much ginger per day? No regulatory limit is published for culinary ginger. The European monograph frames dosed medicinal preparations, not cooking. One to three centimetres of fresh rhizome a day in food stays within ordinary dietary territory.

Does ginger tea do anything? For nausea, that is ginger’s best-documented indication, and a strong infusion comes close to it. For inflammation, no data support the claim. Our page on herbal teas and spice blends for digestion covers the usual preparations.

Does turmeric thin the blood? Effects on platelet aggregation are described in vitro and in animal models. In humans, at culinary doses, no data show a clinically relevant effect. With a supplement and an anticoagulant the question is real and belongs to a doctor.

Should pepper go in after cooking? Nothing about piperine requires it. It is a matter of taste: the volatile aromas of freshly ground pepper hold up better when added late.

Is Ceylon cinnamon healthier? On coumarin, unambiguously yes, at roughly a hundredth of the level. On glycaemic effects, nothing indicates superiority, and most trials were run with cassia.

Do these spices replace an anti-inflammatory drug? No, and no agency monograph suggests otherwise. Existing treatment is not adjusted without medical advice.

Sources