Coffee myths and disputed claims
Coffee attracts a lot of confident-sounding claims. Each card below states what's actually supported, what's misleading, and — just as importantly — what's genuinely uncertain rather than forcing a simple true/false verdict where the real answer is more nuanced.
This is general information, not personal medical advice. Associations found in research are not guarantees for any individual, and evidence can change — see coffee health and caffeine for fuller topic-by-topic detail on the health-related claims below.
Outdated“Coffee dehydrates you.”
Official guidance counts coffee towards daily fluid intake. Caffeine is often described as mildly diuretic, but that is not treated as cancelling the drink out.
What's true
Caffeine is commonly described as mildly diuretic. Neither document read for this entry measures that; the study measured body water, which did not change.
What's misleading
That coffee is dehydrating overall. The NHS's guidance on drinks lists tea and coffee among the drinks that count towards daily fluid intake.
Important context
The study covered moderate intake by habitual drinkers. Larger amounts, people unused to caffeine, heat and exercise were not tested.
Current evidence
The NHS's guidance says water, lower-fat milk and lower-sugar drinks, including tea and coffee, all count as part of daily fluid intake, and advises caffeine in moderation. A 2014 controlled study of fifty men who habitually drank coffee found no significant change in total body water over three days of four cups a day compared with the same volume of water; it says nothing about women or about people unused to caffeine.
Misleading“Dark roast coffee has more caffeine than light roast.”
Not on the evidence read. In one study that brewed the same coffees identically at different roast degrees, the cups from light-to-medium roasts held the most caffeine and darker roasts held less.
What's misleading
Roast colour is popularly treated as a caffeine indicator; it mostly is not.
Current evidence
Species and dose move caffeine far more than roast does. Within that, one brewing study found caffeine in the cup highest from light-to-medium roasts and lower from darker ones, which its authors read as indirect evidence of some loss late in roasting. It is one study; measuring by scoop instead of by weight adds a further difference, because darker beans are less dense.
Misleading“Espresso always contains more caffeine than filter coffee.”
Per serving, a typical filter coffee (larger volume) often contains more total caffeine than a single espresso shot, even though espresso is more concentrated per millilitre.
Important context
Espresso has more caffeine per millilitre. What decides the total is how much ground coffee went into the drink: a single shot from 7 to 9 g carries less than a cup of filter from 15 g, while a double from 18 g can equal or exceed it.
Current evidence
Serving-size differences matter more than brew method for total caffeine per cup in most real-world comparisons.
Misleading“Strong-tasting coffee always contains more caffeine.”
How strong a cup tastes is mostly how concentrated it is and how it was roasted. Caffeine depends on the species and on how much coffee went in, so the two need not move together.
Current evidence
This is reasoning from definitions, not a finding: strength is concentration, and a dark roast tastes roasty at any strength. The caffeine guide reports what was measured for species, dose and roast.
Misleading“Decaf coffee has no caffeine at all.”
Decaf keeps a little caffeine. The US Food and Drug Administration's consumer guidance gives 2 to 15 milligrams in an 8-fluid-ounce cup as typical.
Current evidence
Decaffeination removes most of the caffeine, not all of it. The one legal limit CoffeeHQ has read is the European directive on coffee extracts, which allows instant coffee and other extracts to be labelled decaffeinated only if caffeine is no more than 0.3 per cent of their coffee-based dry matter; it does not cover roasted coffee, and no rule setting a percentage to be removed has been read. Studies of decaffeinated coffee bought from shops, read from their abstracts, found from none detected to just under 18 mg in a serving.
Disproven“Drinking coffee sobers you up after alcohol.”
No. The US Centers for Disease Control and Prevention says that when caffeine is used with alcohol it does not reduce the effects of alcohol on the body.
Safety note
Do not treat feeling more alert after coffee as a sign that it is safe to drive or do anything else that needs you sober.
What's misleading
Feeling more awake is mistaken for being less impaired.
Current evidence
The CDC's page on alcohol and caffeine says the effects of alcohol on your body do not change when it is used with caffeine, and that caffeine might make you feel you have more energy, or feel that the alcohol is affecting you less.
Outdated“Coffee causes stomach ulcers.”
The NHS gives the main causes of stomach ulcers as infection with H. pylori bacteria and regular use of anti-inflammatory painkillers. It lists coffee among food and drink that can trigger symptoms, not among causes.
Safety note
Stomach pain that persists or recurs needs a doctor, not a change of drink.
Current evidence
The NHS's page on the causes of stomach ulcers names H. pylori infection and regular use of anti-inflammatory painkillers such as ibuprofen as the main causes. Its advice for people who have an ulcer is to avoid food or drink that triggers symptoms, which can include fatty or spicy foods, coffee and chocolate.
Depends on context“Coffee causes heart disease.”
The largest review of the published studies found coffee drinking more often associated with benefit than harm for cardiovascular outcomes, and called most of that evidence low or very low quality. It shows association, not cause, in either direction.
Safety note
This is general information, not a substitute for advice from your own doctor about your heart.
Important context
People with a diagnosed heart or rhythm condition may be advised differently by their own clinician.
Current evidence
A 2017 review in the BMJ concluded that coffee consumption seems generally safe within usual levels of intake. Its authors add that the evidence is mostly observational and of low or very low quality, that residual confounding could explain the associations, and that genetic studies found no evidence of a causal link with lower cardiovascular mortality. The FDA separately lists increased heart rate and palpitations among the effects of too much caffeine.
Depends on context“Coffee is bad for everyone with high blood pressure.”
The NHS's guidance on high blood pressure advises not drinking too much caffeine; it does not say to avoid coffee altogether. What applies to one person is a matter for their own clinician.
Safety note
Do not stop, start or change caffeine as a substitute for medical advice about blood pressure or its medication.
Current evidence
The NHS's page on high blood pressure lists, among things to avoid, drinking too much caffeine, and names coffee, tea and cola as drinks high in it; it gives no amount. The FDA lists high blood pressure among the symptoms of too much caffeine. Nothing read for this entry measures how large or lasting the effect of a cup is.
Depends on context“Coffee always causes anxiety.”
Not always. The FDA lists anxiety and jitters among the symptoms too much caffeine can cause, and says some conditions and medicines make people more sensitive to it.
Safety note
If coffee reliably brings on severe anxiety or panic, cutting back and speaking to a healthcare professional is a reasonable step.
Current evidence
The FDA's consumer guidance describes anxiety and jitters as possible symptoms of too much caffeine, alongside a raised heart rate, palpitations and insomnia, and says sensitivity varies with certain conditions and some medications. It gives no amount at which a given person is affected.
Misleading“Coffee is addictive in the same way as illicit drugs.”
Stopping regular caffeine can bring unpleasant withdrawal, which the FDA describes as not considered dangerous. The comparison with illicit drugs is not one any document read for this entry makes.
Current evidence
The FDA's consumer guidance says caffeine withdrawal is not considered dangerous but can be unpleasant, and advises people who use caffeine daily and want to cut back to do so gradually. No document comparing caffeine with other substances was read, so this entry makes no such comparison.
Depends on context“Coffee consumed early in the day never affects that night's sleep.”
Never is too strong. The European Food Safety Authority says a single 100 mg dose of caffeine may affect sleep in some adults, particularly close to bedtime; the NHS's advice for people with insomnia is no tea or coffee for at least six hours before bed.
Important context
People differ in how caffeine affects them; neither document gives a figure for how long it stays active.
Current evidence
EFSA's 2015 opinion, as the authority summarises it, says single doses of 100 mg may affect sleep duration and patterns in some adults, particularly when consumed close to bedtime. The NHS's insomnia page advises not drinking tea or coffee at least six hours before going to bed. Neither document says how early is early enough for everyone.
Misleading“Coffee prevents disease.”
Population studies associate coffee drinking with lower risk of several conditions. The authors of the largest review of them say the evidence is mostly low quality and may reflect confounding. An association is not prevention.
Safety note
Do not treat coffee as a substitute for medically recommended prevention or screening.
Current evidence
A 2017 review in the BMJ found coffee consumption more often associated with benefit than harm across many outcomes. Its authors write that most of the evidence comes from observational research of low or very low quality, and that genetic studies found no evidence that the associations with type 2 diabetes and mortality are causal.
Outdated“Coffee causes cancer.”
The World Health Organization's cancer agency found no conclusive evidence that drinking coffee causes cancer. In 2016 it moved coffee from 'possibly carcinogenic', its 1991 classification, to 'not classifiable'.
Safety note
The same evaluation classed drinking very hot beverages, above 65 °C, as probably carcinogenic to the oesophagus. That finding is about temperature, not about coffee.
Current evidence
The International Agency for Research on Cancer reviewed more than 1,000 studies and classified coffee drinking in Group 3, not classifiable as to its carcinogenicity to humans. It reported no carcinogenic effect for cancers of the pancreas, female breast and prostate, reduced risks for cancers of the liver and the lining of the womb, and inconclusive evidence for more than twenty others. The classification concerns hazard, not the size of any risk.
Misleading“Coffee is naturally calorie-free no matter how it's prepared.”
The calories in a coffee drink come almost entirely from what is added to it: milk, cream, sugar and syrups.
Current evidence
This entry is arithmetic, not a measurement: a drink is the sum of its ingredients, and milk, cream, sugar and syrup all carry calories that the coffee itself barely adds to. No figure is given because none was read for a specific drink; a product's own nutrition label is the place to find one.
Misleading“Instant coffee is not real coffee.”
Instant coffee is made from real brewed, roasted coffee that has had its water removed — it goes through additional industrial processing, but it is not a synthetic substitute.
Current evidence
See CoffeeHQ's instant coffee guide for the full production process.
Misleading“Robusta is always low-quality coffee.”
'Always' is the error. Robusta is a species with many varieties and clones, bred for different purposes; its reputation comes from how most of it has been used, not from a ceiling on the plant.
Important context
Much robusta goes into blends and instant coffee, which is where the reputation comes from. That describes a market, not every lot.
Current evidence
The published catalogue of robusta varieties records each clone's breeding and traits and treats the species as one with real differences between varieties. See CoffeeHQ's robusta cultivars page. No tasting comparison was read for this entry, so none is claimed.
Misleading“Oily-looking beans are fresher.”
Oil on the surface says more about how dark the roast is than about how fresh the coffee is. The roast date, where a bag prints one, is the direct evidence of age.
Current evidence
Roasters describe surface oil as a feature of darker roasts. CoffeeHQ has read no study relating surface oil to freshness, so this entry rests on that trade description.
Misleading“Crema thickness proves espresso quality.”
Crema is a foam, and how much of it there is depends on things other than how the shot tastes. A thick layer is not evidence of a good shot.
Current evidence
A review of espresso foam by two authors at a coffee company attributes crema chiefly to the carbon dioxide roasted coffee holds, cites measurements finding more of it in robusta than in arabica and more in dark roasts than in medium, and names the bicarbonate in brewing water as a further source of gas. So crema follows gas, and gas follows freshness, roast, species and water together. No study relating the amount of crema to how a shot tastes was read for this entry.
Unsupported“Coffee should be stored in the fridge.”
A fridge's humidity and odours from other food can degrade coffee faster than a cool, dark, airtight cupboard.
Current evidence
The usual reasoning is that coffee taken in and out of a fridge meets moisture and food odours each time. CoffeeHQ has read no study that tests fridge storage against a cupboard, so this is reasoning, not a finding.
Depends on context“Freezing coffee always ruins it.”
'Always' is not supported. Whether freezing helps depends on how the coffee is sealed and handled, and the published evidence is thin.
Important context
The practical worry is moisture: a cold bag opened in a warm room collects condensation.
Current evidence
See CoffeeHQ's guide to why coffee goes stale for what the one storage study read reported and what it did not test.
Outdated“Espresso should always take exactly 25-30 seconds.”
Shot time is one signal among several (dose, yield, taste) — a fixed universal time target doesn't account for different beans, ratios and roast levels.
Current evidence
The 25 to 30 seconds is a convention of espresso training, not a measured optimum: no document read for CoffeeHQ's espresso guides establishes a shot time that is right for every coffee.
Misleading“A pressure gauge reading proves good extraction.”
Pressure is only one variable among grind, dose, distribution and time — a correct gauge reading doesn't guarantee a well-extracted, good-tasting shot.
Current evidence
Espresso quality depends on the interaction of multiple variables; pressure alone doesn't capture channelling, uneven distribution or grind problems.
Misleading“More expensive equipment guarantees better coffee.”
Price buys convenience, repeatability and features. It does not make the coffee, the grind or the recipe right.
Current evidence
This is reasoning, not a finding: no comparison of cheap and expensive equipment was read. CoffeeHQ's equipment pages say for each item what it changes and when it is not needed.
Misleading“Single origin is always better than a blend.”
Blends are deliberately designed for balance and consistency and can be excellent; single origins highlight a specific character but aren't automatically higher quality.
Current evidence
Quality depends on the specific beans, processing and roasting, not on "single origin" versus "blend" as a category.
Marketing claim“Kopi luwak is automatically the best coffee in the world.”
'Automatically the best' is a claim about taste that no document read for this entry supports. A high price shows that something is scarce or sought after, not how it tastes.
Current evidence
CoffeeHQ has read no tasting study of kopi luwak and no report on how it is produced, so this entry says nothing about either. It is reasoning about what a price can and cannot show.
Unsafe claim“Coffee enemas detoxify the body.”
No document read for this entry supports the claim, and the US National Cancer Institute records harms: changes in blood chemistry from too many enemas of any kind, and published reports of three deaths that may be related to coffee enemas.
Safety note
Coffee enemas are not a medically supported practice. This is safety information, not personal medical advice; speak to a healthcare professional about any concern they are meant to address.
Current evidence
The National Cancer Institute's patient summary of the Gerson regimen describes coffee enemas as taken to remove toxins from the body. It says taking too many enemas of any kind can change normal blood chemistry, and that reports of three deaths that may be related to coffee enemas have been published.
Unsupported“Green coffee extract supplements cause meaningful weight loss.”
A systematic review of the trials found a moderate effect of uncertain clinical relevance in studies it called poor in quality, and a US regulator settled charges against a marketer over the weight-loss claims.
Safety note
This is not a recommendation for or against any supplement. Speak to a healthcare professional before using a weight-loss supplement.
Current evidence
The review, published in 2011, pooled the randomised trials then available. Its authors called the results promising but the trials poor, and said the clinical relevance of the effect was not certain. The US Federal Trade Commission announced a settlement with a company that had marketed green coffee extract for weight loss on the strength of a study the commission described as flawed.
Marketing claim“Mushroom coffee delivers all its advertised health benefits.”
Advertised benefits run ahead of what has been shown. CoffeeHQ's guide to functional and trend coffee claims reports what one review of the trials on lion's mane found and how limited it was.
Current evidence
See the functional and trend coffee claims guide. This entry states no benefit and no harm of its own.
Misleading“"Italian roast" means the coffee came from Italy, and "French roast" means it came from France.”
Both are names for a dark roast. They describe how the beans were roasted, not where the green coffee was grown.
Current evidence
"Italian roast" and "French roast" are conventional dark-roast-level names used by roasters in various countries; they describe roast degree, not origin.
Misleading“"Espresso roast" is a distinct type or species of bean.”
"Espresso roast" describes a roast style (typically medium-to-dark, developed for espresso brewing), not a separate bean species or variety — almost any coffee can be roasted for espresso.
Current evidence
See CoffeeHQ's roasting-basics guide, which already distinguishes espresso roast/filter roast/omni-roast as brewing-intent labels, not fixed colours or bean types.
Depends on context“Arabica always has less caffeine than every Robusta coffee.”
Robusta contains more caffeine than arabica on average: roughly one and a half times as much in the one study CoffeeHQ has read that measured many green coffees of both. But "always" overstates it. In that study the lowest robusta held less than the highest arabica, and in the earlier one it quotes the ranges for the two species overlap, so an individual arabica can hold as much as an individual robusta.
Important context
The species-level generalisation (robusta usually higher) is well supported; the absolute "always, every sample" framing is not, and neither is a flat doubling.
Current evidence
One study of 74 green coffees measured an average of 1.8 per cent caffeine by dry weight in arabica and 2.9 per cent in robusta, about 1.6 times. It quotes two earlier sets of figures: 1.3 and 2.3 per cent (about 1.8 times), and means of 1.3 and 1.8 per cent (about 1.4 times) with ranges of 0.8 to 1.8 and 1.2 to 2.5 per cent. The authors conclude that caffeine alone cannot reliably tell the species apart.
Disproven“Coffee loses all its caffeine during roasting.”
Roasted coffee keeps most of its caffeine. One brewing study did find less caffeine in cups from darker roasts than from light-to-medium ones, so roasting is not without effect, but the claim that it removes all of it is false.
Current evidence
In the one study read that brewed coffees identically across roast degrees, every roast produced a caffeinated cup, with the most caffeine at light-to-medium roasts.
Outdated“Espresso tamping must use an exact, fixed pressure (e.g. precisely 30 lb/13.6 kg).”
No document read for CoffeeHQ's espresso guides sets a tamping force that must be hit. Once the bed stops compressing, pressing harder has nothing left to do; a level, repeatable tamp is the part that can be checked.
Current evidence
The fixed figure is a training convention. Two published statements were found. One is a sentence in a 2020 espresso study: its authors explored a range of tamp pressures, saw no appreciable variation in shot time or extraction yield, and standardised on a machine pressing at 98 newtons, about ten kilograms. They noted flavour differences without quantifying them. The other is the summary of a 2017 study, read in that summary only, which reports that tamping pressure had no detectable effect on how caffeine and a second compound were extracted. Nobody has published a tasting comparison.
Misleading“More bars of pump pressure always means better espresso.”
Nine bar is a convention, not a requirement, and a higher number on the box is not a better shot. Pressure and flow depend on each other and on the puck.
Current evidence
See CoffeeHQ's guide to espresso pressure and flow, which reports the one study of pressure and flow read and a second that pulled its shots at six bar.
Misleading“Bean-to-cup machines cannot produce good coffee.”
Modern bean-to-cup machines can produce good, consistent coffee when fresh beans, correct grind and clean equipment are used — convenience and full manual control are a genuine trade-off, but that's different from a fixed quality ceiling.
Current evidence
See CoffeeHQ's bean-to-cup coffee-maker profile and Change the Taste guidance, which treat it as a legitimate format with its own tuning levers, not a lesser one.
Depends on context“Pod and capsule coffee is always weak.”
Strength depends on the specific capsule and drink-size setting chosen — many systems offer stronger/ristretto-style capsules and shorter extraction settings that produce a notably stronger result than a long, diluted setting.
Current evidence
See CoffeeHQ's pod/capsule Change the Taste guidance on why long extractions from one capsule taste thin, and how capsule choice and volume setting change strength.
Depends on context“Water straight off the boil always "burns" coffee instantly.”
There is no evidence that near-boiling water ruins coffee as a rule. Many recipes use water just off the boil, and in the one controlled tasting CoffeeHQ has read, a trained panel found no appreciable difference between brews at 87, 90 and 93 °C once strength and extraction were matched.
Current evidence
See the CoffeeHQ water-temperature guide, which reports the controlled test and its limits: one coffee, one brewer, three temperatures.
Unsupported“Coffee cleanses or detoxifies the liver.”
No document read for this entry describes coffee cleansing or detoxifying the liver. Population studies associating coffee with lower risk of liver cancer are a different and much narrower finding.
Safety note
Do not use coffee as a substitute for medical care for any liver condition.
Current evidence
The International Agency for Research on Cancer's 2016 evaluation reported reduced risks of liver cancer in studies of coffee drinkers. That is an association in population studies, not a mechanism of cleansing, and it is not a treatment.
Misleading“Decaf is always made from inferior-quality beans.”
Decaffeination is a process applied to green coffee; nothing in it requires poor coffee. 'Always' is the error.
Current evidence
This is reasoning from what the process is. No survey of the coffees chosen for decaffeination was read, so nothing is said here about what roasters usually choose.
Depends on context“Cold brew is always less acidic than hot coffee, and always contains more caffeine.”
Cold brew measured lower in total acids than hot brews of the same coffee in the studies CoffeeHQ has read, and drinkers often describe it as smoother; concentrate is often stronger than typical hot coffee — but both depend heavily on ratio and dilution, so neither claim holds universally once serving strength is accounted for.
Important context
A cold-brew concentrate diluted heavily can easily have less caffeine per serving than a strong hot coffee; perceived acidity also varies by bean and steep time, not just brew temperature.
Current evidence
See CoffeeHQ's cold-brew dilution guidance and caffeine estimator, which treat both strength and perceived acidity as recipe-dependent rather than fixed properties of the method.
Disproven“Coffee is completely free from acrylamide.”
Coffee contains acrylamide. The European Food Safety Authority lists coffee among the foods it is found in and, with fried potato products, among the most important dietary sources for adults.
Safety note
General food-safety information, not personal medical advice.
Current evidence
EFSA describes acrylamide as a chemical that forms naturally in starchy food during high-temperature cooking, from sugars and amino acids present in the food, and says it is found in products such as potato crisps, French fries, bread, biscuits and coffee. Its 2015 assessment concluded that acrylamide in food potentially increases the risk of developing cancer for consumers in all age groups. This entry gives no amount per cup and no advice; neither was read.
Depends on context“Washed process coffee is always cleaner-tasting, and natural process is always sweeter.”
These are the trade's expectations of the two processes, not established findings: CoffeeHQ has read no controlled tasting of washed against natural coffee from the same fruit. Variety, climate, drying and roasting all change the cup as well, so neither word on a bag guarantees a taste.
Current evidence
See CoffeeHQ's processing pages, which report what the process guidance and the few trials read actually say and promise no taste.
Misleading“Coffee grown at higher altitude is always better quality.”
Studies do tend to find better tasting scores at higher altitude: in one systematic review, 12 of the 13 studies that reported taste found so. But the review notes results varied, altitude largely stands in for temperature, and quality also depends on variety, processing, farming and roasting.
Current evidence
See CoffeeHQ's guide to where coffee grows and why, which reports the review's counts and its caveat.