Coffee Science Debunked

Coffee science debunked: what counts as evidence?

Start by asking what the researchers actually measured. A change in a cell, a temporary metabolic response and a reduction in diagnosed disease are different findings. A claim becomes misleading when it quietly swaps one for another.

Observational studies follow people’s habits and health without assigning their coffee intake. They can identify associations, but other differences between drinkers and non-drinkers may influence the result. Controlled trials assign an intervention and a comparison, making them better suited to testing cause and effect, although their size, duration and chosen measurements still matter.

A review is only as useful as the evidence it assesses. Combining observational studies does not turn them into an experiment, and an authoritative assessment may reasonably conclude that the answer remains uncertain.

For the wider collection of everyday claims, use coffee myths and facts. Here, the focus is on what scientific evidence can actually establish.

Does coffee cause cancer, or prevent it?

Verdict: neither a blanket cancer warning nor a cancer-prevention promise is justified.

In June 2016, the International Agency for Research on Cancer, the World Health Organization’s cancer agency, changed coffee’s classification from Group 2B, “possibly carcinogenic”, to Group 3, “not classifiable as to its carcinogenicity to humans”. The full assessment appeared in IARC Monographs, Volume 116, published in 2018. Calling it a “2016 monograph” confuses the evaluation date with the publication date.

Group 3 does not mean “proved harmless”. It means the available evidence does not support classifying coffee as carcinogenic to humans. Equally, the earlier classification should not be presented as if it remains the current assessment.

Temperature is a separate question

IARC classified drinking very hot beverages, above 65°C, as Group 2A, “probably carcinogenic to humans”, principally concerning cancer of the oesophagus, the food pipe. This applies to very hot beverages generally, not uniquely to coffee.

The relevant temperature is the drink as swallowed. A brewing-water temperature tells you neither how hot the coffee will be when you drink it nor your personal cancer risk. Let a scalding drink cool rather than treating rapid drinking as part of the recipe.

What about protective associations?

IARC reported lower risks of liver and endometrial cancer among coffee drinkers in epidemiological research. Those findings should not become “coffee prevents cancer”: observational associations do not establish that prescribing coffee would produce the same benefit.

Evidence strength: a major expert assessment drawing on human, animal and mechanistic research. The temperature hazard and the protective associations answer different questions; neither supplies an individual risk calculation.

Does coffee boost metabolism enough to cause weight loss?

Verdict: a measurable metabolic effect is real, but the guaranteed slimming claim is not.

Metabolic rate is the rate at which the body uses energy. In a controlled study published in 1989, 100 mg of caffeine increased resting metabolic rate by approximately 3% to 4% over 150 minutes. That supports a short-term physiological effect, not a promise about what happens to body weight over months.

A more substantial coffee trial, published in 2020, assigned 126 adults with overweight and reduced insulin sensitivity to instant coffee or a coffee-like placebo for 24 weeks. It found no significant improvement in insulin sensitivity, although the coffee group had a modest reduction in fat mass relative to placebo. The trial was sponsored by Nestlé Research, a relevant detail when assessing its independence.

It would therefore be inaccurate to say coffee has no metabolic effects. It would be equally inaccurate to turn those findings into a reliable weight-loss prescription for everyone.

Notice the endpoint: energy expenditure, fat oxidation, fat mass and body weight are not interchangeable measurements. A study can support one without establishing the others.

Evidence strength: controlled human evidence supports short-term effects, with some longer trial evidence suggesting modest body-composition changes. That is narrower than the commercial promise that a particular coffee will make its drinkers slim.

Can coffee detox your body?

Verdict: “detox” is not a demonstrated benefit unless the claim specifies what is removed and shows that removal matters.

Ask three questions: which toxin, measured where, and compared with what? “Cleansing” without these details is not a testable explanation.

The US National Center for Complementary and Integrative Health reports no compelling research supporting detox diets for eliminating toxins or managing weight. It also describes weaknesses in the available studies. This is an assessment of detox programmes generally, not a direct trial of every coffee product.

Coffee’s association with better liver outcomes does not establish that it flushes unidentified toxins from the body. Nor would feeling more alert demonstrate toxin removal.

Evidence strength: weak evidence for commercial detox programmes, with no basis in the sources reviewed for treating ordinary coffee as a general detox treatment. A credible product-specific claim would need controlled human evidence measuring the named substance and a meaningful health outcome.

Are mould and mycotoxins making coffee unsafe?

Verdict: mycotoxins are real food-safety hazards, but that does not validate blanket scare-marketing about coffee.

Mycotoxins are substances produced by certain moulds. Ochratoxin A, often shortened to OTA, is relevant to coffee and other foods. Finding the toxin is not the same as finding living mould, and neither result should be described without its measurement context.

EFSA’s 2020 risk assessment identified kidney toxicity and kidney tumours in animal studies, while finding that the available human epidemiological evidence could not establish a causal relationship with adverse effects. EFSA also identified concerns about dietary exposure. Dismissing OTA as imaginary would therefore be as misleading as declaring every coffee dangerous.

The missing step in many alarming claims is exposure: how much is present, how much of the food or drink is consumed, and over what period?

Coffee is already subject to contaminant controls

Maximum levels for OTA apply to coffee. The legal framework differs across the UK: Great Britain uses assimilated Regulation (EC) No 1881/2006, as amended, while the EU framework applicable in Northern Ireland is Regulation (EU) 2023/915. The latter should not be described as a UK-wide assimilated regulation.

Those controls are a reason to ask precise questions, not a guarantee that contamination can never occur.

For a “mycotoxin-free” claim, ask for the named toxins, batch identification, laboratory method, numerical results and reporting limits. “Not detected” is bounded by a test’s detection capability; it is not proof of absolute absence.

Evidence strength: substantial animal toxicology and regulatory risk assessment establish the hazard. They do not, by themselves, demonstrate that choosing a coffee carrying a special marketing label improves human health.

Does coffee’s acidity explain stomach trouble?

Verdict: coffee can aggravate symptoms, but acidity is not a complete explanation.

Reflux happens when stomach contents travel back towards the throat. The NHS lists coffee among the foods and drinks that can make it worse. That is different from saying every coffee drinker develops reflux or that coffee’s own acidity is necessarily the cause.

There is also a distinction between acid already present in a drink and gastric acid secretion, the stomach’s production of acid. A controlled physiological study published in 2014 compared two coffee blends in nine healthy volunteers. The dark-roasted blend stimulated less gastric acid secretion than the medium-roasted blend.

That is an interesting mechanism, but a small study measuring stomach acidity does not establish that every dark roast prevents heartburn. Nor does it prove that a “low-acid” product will relieve symptoms in people with diagnosed reflux.

For the chemistry behind acidity claims, see the pH myth. Roast-related caffeine claims belong in does dark roast have more caffeine?, rather than being used as a shortcut to predicting stomach comfort.

If coffee repeatedly brings on symptoms, consider reducing or avoiding it and discuss persistent heartburn with a pharmacist or GP. Seek medical advice if heartburn occurs most days, treatment is not helping, food sticks when swallowing, vomiting is frequent or weight falls without explanation.

Evidence strength: human physiological evidence supports effects on acid secretion, while NHS guidance recognises coffee as a possible symptom trigger. Neither supports a universal rule based only on acidity.

How to check the next scientific coffee claim

Use this sequence before accepting either a frightening headline or a reassuring product promise.

1. Write down the exact claim. “Contains an antioxidant” and “prevents disease” require different evidence.
2. Identify what was tested. Was it brewed coffee, isolated caffeine, an extract, a cell culture or a commercial blend?
3. Find the comparison. A change after drinking coffee is less informative without knowing what happened in the comparison group.
4. Check the endpoint and duration. A brief physiological response cannot establish a lasting health outcome.
5. Read the limitations and funding. Sponsorship is relevant, but does not automatically invalidate a result. Look at design, reporting and independent confirmation rather than using funding as the entire verdict.

The most useful distinction is between evidence of a mechanism and evidence of a benefit or harm in people. Coffee science debunked properly preserves that distinction, including when the honest answer is “not established”.