Green tea and black tea come from the same plant, Camellia sinensis. Green tea is made from leaves and buds that skip the withering and oxidation steps used for oolong and black tea. A cup (245 g) has 2 calories, about 29 mg of caffeine and 172 mg of the catechin EGCG, according to USDA data.
The cup and the capsule are very different doses. EFSA estimated that European adults take in 90 to 300 mg of EGCG a day from green tea infusions on average, and up to 866 mg for the highest consumers, while green tea supplements supply 5 to 1,000 mg a day. Several trials below used capsules with 843 to 857 mg of EGCG a day, so their results do not show what a cup does.
The strongest-looking evidence for tea itself is observational. Large Japanese cohorts link 5 or more cups a day to a lower risk of dying, but drinkers may differ from non-drinkers in other ways. Caffeine is modest: lab tests of white, green and black teas found 14 to 61 mg per serving, and most brewed teas had less than brewed coffee.
Health benefits of green tea
None of the studies below reaches strong evidence, and many of the trials used extract or catechin capsules rather than brewed tea. The best-supported results are small drops in LDL cholesterol (2.19 mg/dL in 14 trials) and systolic blood pressure (1.98 mmHg in 13 trials), though a later pooled analysis of 4- to 24-week trials found no significant effect. Large Japanese cohorts link 5 or more cups a day to a lower risk of dying but cannot prove cause. Here is what 30 studies found, strongest evidence first, from cholesterol and blood sugar to weight, memory, cancer and liver safety, including the trials that found nothing.
Green tea drinks and extracts lowered LDL cholesterol by 2.19 mg/dL in a pooled analysis of trials
Meta-analysis of 14 randomized trials (1,136 people) of green tea beverages or green tea extract. Total cholesterol fell by 7.20 mg/dL and HDL did not change, and the results did not depend on dose, trial length or type of product. The LDL drop is small, and drinks and extracts were mixed together.
A green tea extract with 843 mg of EGCG a day for a year lowered LDL cholesterol by 4.1% in postmenopausal women
Placebo-controlled trial in 936 women who finished this substudy, using 4 capsules a day of green tea extract (EGCG capsules, not brewed tea). LDL fell 4.1% against a 0.9% rise on placebo and total cholesterol fell 2.1%, but triglycerides rose 3.6%. Total cholesterol dropped only in women who started high.
Green tea lowered LDL cholesterol by 0.64 mmol/L in a Cochrane review, but the result did not hold up when trials were dropped one by one
Cochrane review of 11 randomized trials (821 people) of green or black tea lasting at least 3 months; 7 used green tea, in different forms including extracts. Green tea also lowered systolic pressure by 3.18 mmHg and diastolic by 3.42 mmHg, but few small trials contributed. No trial reported heart attacks or strokes.
Green tea lowered systolic blood pressure by 1.98 mmHg and diastolic by 1.92 mmHg in a pooled analysis of trials
Meta-analysis of 13 randomized trials (1,367 people) of green tea, with drinks and green tea polyphenol products pooled. The drop showed up only in some subgroups: low-dose polyphenol, longer trials, or trials that ruled out the effect of caffeine.
Show 26 more studies
A pooled analysis of 4- to 24-week trials found no significant effect of tea on blood pressure or cholesterol
Meta-analysis of 14 randomized trials (798 people) of green or black tea in healthy and at-risk adults. The authors rate the evidence low to moderate, because most trials were too small to detect changes, and call for longer, larger trials.
Green tea lowered fasting blood sugar by 0.09 mmol/L and HbA1c by 0.30% in a pooled analysis of 17 trials
Meta-analysis of 17 randomized trials (1,133 people) of green tea or green tea extract. Fasting insulin fell by 1.16 µIU/mL, but only in the higher-quality trials. The drops are small, and the trials mixed drinks and extracts.
In a Cochrane review, green tea for 12 to 13 weeks changed weight by only 0.04 kg in trials outside Japan
Cochrane review of randomized trials of green tea preparations in overweight or obese adults, a term that covers extracts as well as drinks. Six trials outside Japan (532 people) found no significant change, while eight Japanese trials (1,030 people) were too different to pool and ranged from 0.2 to 3.5 kg of weight loss. The authors call the loss small and not significant.
Green tea catechins or an EGCG-caffeine mixture lowered body weight by 1.31 kg in a pooled analysis
Meta-analysis of 11 studies of weight loss and weight maintenance, using catechin or EGCG-caffeine supplements. The effect was smaller with high habitual caffeine intake (over 300 mg a day), but that difference did not reach significance.
A review of 159 human studies found liver side effects only with concentrated green tea preparations in large doses, not with brewed tea
Systematic review of lab toxicology and 159 human intervention studies. Liver problems rose with dose and with catechin-rich preparations taken as large bolus doses, but not with brewed tea or with extracts in drinks or food. The authors derived a safe intake of 338 mg EGCG a day for solid doses. Three of the four authors work for Herbalife, a supplement company.
A green tea extract with 856.8 mg of EGCG a day for 12 weeks cut weight from 76.8 to 75.7 kg in women with central obesity
Placebo-controlled trial in 102 women using green tea extract (EGCG capsules, not tea). BMI, waist size and total cholesterol (down 5.33%) also fell, and no side effects were reported. The abstract reports the change inside the extract group, and the dose is high (see the liver warnings below).
A green tea extract with 302 mg of EGCG a day for 12 weeks did not lower weight more than placebo in obese women
Placebo-controlled trial in 78 obese women taking 400 mg capsules of green tea extract three times a day (capsules, not tea). Body weight fell 0.3% (0.15 kg), with no difference in BMI or waist size against placebo.
A green tea catechin drink plus exercise for 12 weeks cut abdominal fat by 7.7% against 0.3% with a control drink
Trial in 132 overweight and obese adults (107 finished) who exercised at least 180 minutes a week and drank a beverage with about 625 mg of catechins and 39 mg of caffeine, or a caffeine-only control. Body weight and total fat mass did not differ significantly. It was a catechin-fortified drink, not ordinary tea.
A year of green tea extract with 843 mg of EGCG did not change body fat or bone density in overweight postmenopausal women
Substudy of the Minnesota Green Tea Trial in 121 overweight or obese women (61 on extract, 60 on placebo), using decaffeinated extract capsules, not tea. BMI, fat mass and bone density did not differ, though some body fat measures fell more in women with a higher BMI.
2 g of matcha a day for 12 months did not change the main memory and daily-living scores in older adults with cognitive decline
Placebo-controlled trial in 99 people with subjective cognitive decline or mild cognitive impairment (49 on matcha, 50 on placebo). Social acuity, the reading of facial emotions, improved, and sleep quality showed only a trend (P = 0.088). Several authors work for a tea company (ITO EN) and a research firm (MCBI).
A review of 21 studies linked green tea to lower anxiety and better memory and attention, but not through one ingredient alone
Systematic review of 21 studies (4 randomized trials, 12 crossover studies, 4 cross-sectional studies and one cohort) of green tea, green tea extract, L-theanine and EGCG. The authors rate quality as good and say the cognitive effects appeared with caffeine and L-theanine together, with less effect from either alone.
A green tea extract added to a meal cut non-heme iron absorption from 12.1% to 8.9% in young women
Crossover test in 10 women aged 19 to 39 who ate meals with or without a phenolic-rich green tea extract (0.1 mmol) added to the meat, with iron labeled by isotopes. It was an extract added to food, not tea drunk with a meal.
In a year-long trial of 843 mg EGCG a day, the liver enzyme ALT rose 78.1% by month 6 in women with one UGT1A4 genotype, against 28.0% in the other
Secondary analysis of the Minnesota Green Tea Trial (1,075 postmenopausal women), in which women took high-dose green tea extract capsules or placebo for 12 months. The authors suggest genes may shape who is at risk of liver enzyme rises on high-dose extract.
Japanese adults drinking 5 or more cups a day had an 18% lower risk of dying (women) and 10% lower (men) in eight pooled cohorts
Pooled analysis of eight Japanese cohorts (313,381 people, 52,943 deaths over 17.3 years on average), comparing 5 or more cups a day of green tea as a drink with under 1 cup. Heart disease death was also lower (hazard ratio 0.82 in men and 0.75 in women). Tea drinkers may differ from others in ways the analysis cannot fully remove.
Japanese adults drinking more than 5 cups a day had a 13 to 17% lower risk of dying over 18.7 years
Cohort of 90,914 adults aged 40 to 69 (12,874 deaths) who reported how much green tea they drank. Heart disease death was lower in men and women, stroke and respiratory deaths were lower in men, and cancer death did not differ.
Women drinking 5 or more cups a day had a 23% lower risk of dying in the Ohsaki study, and men 12%
Cohort of 40,530 Japanese adults aged 40 to 79, followed for up to 11 years (4,209 deaths), asked how often they drank green tea. Death from heart disease was also lower, most clearly for stroke, but death from cancer was not.
In two Shanghai cohorts, green tea drinkers had a 14% lower risk of dying from heart disease
Men and women in the Shanghai Men’s and Women’s Health Studies (6,517 deaths over 8.3 and 14.2 years). The link with death from any cause was weaker (hazard ratio 0.95, range 0.90 to 1.01) and stronger in never-smokers (0.89), and cancer death was not related to green tea.
Japanese adults drinking 4 or more cups a day had a 20% lower risk of stroke than those who seldom drank green tea
Cohort of 82,369 adults aged 45 to 74 followed for 13 years (3,425 strokes), with 2 to 3 cups a day linked to a 14% lower risk. Coffee was also linked to a lower stroke risk, and neither drink was linked to coronary heart disease.
Older Japanese adults drinking 5 or more cups a day had a 27% lower risk of dementia than those drinking under 1 cup
Cohort of 13,645 adults aged 65 or older followed for 5.7 years, with dementia taken from care insurance records; 8.7% developed it. The result held after excluding people who already reported memory complaints.
A review of 113 cohort studies found little evidence that tea drinking changes cancer risk
Meta-analysis of prospective studies covering 26 cancer sites and 153,598 cancer cases, with black, green and total tea analyzed separately. Apart from a link between green tea and lymphoid cancers, no association was consistent, and the other signals were weak.
Each extra daily cup of green tea made no difference to esophageal cancer risk (odds ratio 1.00)
Dose-response meta-analysis of 14 studies (5,057 cases among 493,332 people), including case-control studies. In women one more cup a day went with an odds ratio of 0.79, and in men 1.03. The authors warn the result may be affected by few studies and bias.
Japanese women drinking 5 or more cups a day had a 21% lower risk of stomach cancer, but men showed no lower risk
Pooled analysis of six Japanese cohorts (219,080 people, 3,577 stomach cancer cases). In women the lower risk was clearer for the distal part of the stomach (hazard ratio 0.70), and no link was seen for the proximal part in either sex.
Burning-hot tea plus 15 g or more of alcohol a day went with 5 times the risk of esophageal cancer
China Kadoorie Biobank cohort of 456,155 adults (1,731 cases over 9.2 years). It asked about tea in general, not green tea alone. Hot tea combined with smoking also raised risk (hazard ratio 2.03), while tea temperature was reported once, at the start.
A review of 16 studies found tea did not affect iron status in people with adequate iron stores
Review of 16 studies of tea in general, not only green tea. In groups with a high rate of iron deficiency, tea drinkers had lower ferritin or hemoglobin, but the link disappeared after adjusting for diet in all but one study.
A single high dose of EGCG raised the liver enzyme ALT 138-fold in mice
Male mice were given EGCG by stomach tube: one dose of 1,500 mg/kg, or two daily doses of 750 mg/kg, which raised ALT 184-fold and caused liver cell death. These are very large bolus doses of purified EGCG, so they show a mechanism, not a safe limit for people.
In lab experiments, EGCG turned toxic amyloid clumps into harmless oligomers
Test with purified alpha-synuclein and amyloid-beta, the proteins that clump in Parkinson’s and Alzheimer’s disease. Pure EGCG blocked the formation of toxic aggregates and produced nontoxic ones instead. No people were studied, so it does not show that drinking tea changes these diseases.
Who should be careful with green tea
Most people can drink green tea freely. Take a little care if one of these applies:
- Green tea extract and EGCG capsules. The concern is the capsule, not the cup. EFSA found in 2018 that 800 mg a day or more of EGCG from supplements raised liver enzymes in trials, and rare liver injury has also been reported after tea itself. The US Pharmacopeia advises taking extract with food and avoiding it with a liver problem. A cup has about 172 mg.
- Pregnancy and caffeine. A cup has about 29 mg of caffeine, but servings ranged from 14 to 61 mg in lab tests. EFSA and ACOG put the limit in pregnancy at 200 mg a day from all sources, and the FDA cites 400 mg a day for most adults.
- Low iron. Green tea can lower absorption of non-heme iron, the type in plant foods and supplements: from 12.1% to 8.9% in a small test with an extract. A review found tea did not change iron status where iron stores were adequate, so the concern is small for most people. If you have anemia or low iron, ask your clinician.
How to choose, store and eat green tea
- Brewed tea: this is the form the cohort studies measured, with the lowest risks at 5 or more cups a day. A cup (245 g) has about 29 mg of caffeine and 172 mg of EGCG on the USDA record.
- Matcha: you drink the whole ground leaf. In a 2003 lab test it gave far more EGCG per gram of dry leaf than a common green tea, and a 12-month trial gave older adults 2 g a day.
- Extract capsules and catechin drinks: trials used 843 to 857 mg of EGCG a day in capsules, close to the 800 mg a day that EFSA links to raised liver enzymes. A cup has about 172 mg.
Typical steeping uses about 2 g of leaf per 100 mL of water at 61 to 87 °C for 30 seconds to 3 minutes, and steeping too hot or too long gives a bitter, astringent cup. In lab tests, steep time changed how much caffeine ended up in the cup. Producers store unrefined green tea cold, at 0 to 5 °C in low humidity, to extend its shelf life. Let very hot tea cool: IARC classed drinks above 65 °C as probably carcinogenic to the esophagus, and in a China cohort the highest risk came when burning-hot tea was combined with alcohol.
Sources
- Beverages, tea, green, brewed, regular (FDC ID 171917). USDA FoodData Central.
- EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS). Scientific opinion on the safety of green tea catechins. EFSA J. 2018.
- Oketch-Rabah HA, Roe AL, Rider CV, et al. United States Pharmacopeia (USP) comprehensive review of the hepatotoxicity of green tea extracts. Toxicol Rep. 2020.
- Hu J, Webster D, Cao J, Shao A. The safety of green tea and green tea extract consumption in adults: results of a systematic review. Regul Toxicol Pharmacol. 2018.
- Caffeine. European Food Safety Authority.
- Spilling the beans: how much caffeine is too much? US Food and Drug Administration.
- Moderate caffeine consumption during pregnancy. American College of Obstetricians and Gynecologists. Committee Opinion 462.
- Chin JM, Merves ML, Goldberger BA, Sampson-Cone A, Cone EJ. Caffeine content of brewed teas. J Anal Toxicol. 2008.
- Samman S, Sandström B, Toft MB, et al. Green tea or rosemary extract added to foods reduces nonheme-iron absorption. Am J Clin Nutr. 2001.
- Temme EH, Van Hoydonck PG. Tea consumption and iron status. Eur J Clin Nutr. 2002.
- Weiss DJ, Anderton CR. Determination of catechins in matcha green tea by micellar electrokinetic chromatography. J Chromatogr A. 2003.
- IARC Monographs evaluate drinking coffee, maté, and very hot beverages. International Agency for Research on Cancer, press release no. 244. 2016.
- Green tea. Wikipedia.
History of green tea
Tea began in China. Legend credits the mythical Emperor Shennong, and Lu Yu’s The Classic of Tea, written in the Tang dynasty around the 8th century, is an important early text. DNA work on 392 cultivated tea plants suggests the plant was domesticated more than once: China tea, Chinese Assam tea and Indian Assam tea appear to come from three separate events.
In Tang China, tea was often pounded into powder and shaped into cakes. In the Song dynasty, people whisked powdered tea in bowls, a drink called mo cha, and Japanese Zen monks who visited Chinese temples met the practice there.
In 1391 the founding Ming emperor banned the production of compressed tea. Loose leaf steeped in hot water became the normal way to drink tea in China, and powdered tea faded out there. Chinese green tea was originally steamed, but after the early Ming dynasty it was usually pan-fired in a dry wok.
Japan kept the powder. Tea seeds reached Japan in the early 9th century with the monks Saicho and Kukai. The Zen priest Eisai (1141 to 1215) brought tea seeds from China, and his 1211 book Kissa Yojoki, or Record of Drinking Tea for Nourishing Life, discussed tea and health. Shade-growing and stone-milling later produced matcha.
Most Japanese green tea is steamed rather than pan-fired, which gives it a sweeter, grassier flavor, and the vast majority of it comes from one cultivar, Yabukita, developed in the 1950s. All commercial tea made in Japan today is green tea, though black tea was briefly made there in the late 19th and early 20th centuries.
Sources
- Meegahakumbura MK, Wambulwa MC, Thapa KK, et al. Indications for three independent domestication events for the tea plant (Camellia sinensis (L.) O. Kuntze) and new insights into the origin of tea germplasm in China and India revealed by nuclear microsatellites. PLoS One. 2016.
- Green tea. Wikipedia.
- Matcha. Wikipedia.
Frequently asked questions about green tea
Is green tea good for you?
For most people, yes, as a drink. A cup (245 g) has 2 calories, 29.4 mg of caffeine and 172 mg of EGCG, and Japanese adults drinking 5 or more cups a day had an 18% (women) and 10% (men) lower risk of dying in a pooled analysis of 313,381 people. That is a link, not proof, and no green tea finding here reaches strong evidence. Trials of its effect on LDL cholesterol and blood pressure found only small changes.
Does green tea have caffeine?
Yes, about 29.4 mg in a cup (245 g) on the USDA record, or 12 mg per 100 g. In a lab test of 20 teas, white, green and black teas held 14 to 61 mg per serving and decaffeinated ones under 12 mg, which is less than most brewed coffee. The FDA cites 400 mg a day as an amount not generally linked to harm in most adults, and EFSA and ACOG put the limit in pregnancy at 200 mg a day from all sources.
How many cups of green tea a day is okay?
No trial here sets an ideal amount. The Japanese cohorts compared groups drinking 5 or more cups a day with under 1 cup, and a stroke cohort linked 4 or more cups a day to a 20% lower risk. Five cups at 29.4 mg of caffeine is about 147 mg a day, below both the 400 mg adult guide and the 200 mg pregnancy limit before other sources, though strong brews reached 61 mg a serving in lab tests. Extract capsules are a different matter.
Does green tea help you lose weight?
Barely, if at all. A Cochrane review found a change of 0.04 kg in trials outside Japan, a pooled analysis of catechin or EGCG-caffeine supplements found 1.31 kg, and a year of extract capsules did not change body fat in 121 women. Several of these used extract or catechin supplements rather than tea. A cup has 2 calories, so it can replace a sugary drink, but that is a swap effect, not a proven effect of green tea.
Does green tea lower blood pressure and cholesterol?
A little, in some trials. A pooled analysis of 13 trials found systolic blood pressure 1.98 mmHg lower and one of 14 trials found LDL cholesterol 2.19 mg/dL lower, but a later pooled analysis of 4- to 24-week trials found no significant effect. The drops are small and the trials mixed drinks with extracts. Tea is not a replacement for blood pressure or cholesterol medicine.
Can green tea hurt your liver?
Brewed tea rarely does, but extract capsules can. EFSA found that 800 mg a day or more of EGCG from supplements raised liver enzymes in trials, while a cup has about 172 mg, and a review of 159 human studies found liver side effects only with concentrated preparations in large doses. Rare cases have also been reported after tea itself. If you take extract, take it with food, and see a clinician about belly pain, dark urine or yellow skin or eyes.
Is matcha healthier than regular green tea?
No trial here compares them, but matcha is the whole ground leaf, so you drink the leaf itself. In a 2003 lab test it gave far more EGCG per gram of dry leaf than a common green tea. In a 12-month trial of 99 older adults, 2 g of matcha a day did not change the main memory and daily-living scores, and several authors worked for a tea company.
Does green tea affect iron absorption?
A little. In a test with 10 women, a green tea extract added to a meal cut non-heme iron absorption from 12.1% to 8.9%. A review of 16 studies found tea did not change iron status in people with adequate stores, only in groups with marginal iron. If you have low iron or take iron pills, ask your clinician.