By Mauro Proença —
Few things are more comforting than a hot cup of coffee or tea on a cold day. But what if the heat that makes the drink so enjoyable also increases your risk of cancer?
Nowadays, it seems like everything causes cancer. A quick internet search reveals that dairy products increase the risk; the same goes for Teflon cookware, food preservatives, and sun exposure, among many others. Some claims are well supported; others stem from associations found in observational studies and are influenced by chance, bias, or confounding factors.
However, observational studies are not inherently flawed. They can provide important evidence about risk factors, especially when findings remain consistent across different studies. The challenge is understanding what those associations actually tell us and how much weight they deserve.
The case of hot beverages illustrates this well. In 2016, the International Agency for Research on Cancer (IARC) classified very hot beverages, defined as those consumed above 65°C (149°F), as "probably carcinogenic to humans." Nearly a decade later, researchers published a large prospective study in the United Kingdom that found substantially higher risks of esophageal squamous cell carcinoma (ESCC) among people who preferred hotter beverages. More recently, a pooled analysis of nearly 1 million UK adults reported a roughly threefold higher risk among people who preferred very hot rather than warm drinks.
That sounds alarming, but does it mean that very hot drinks actually cause cancer?
What did IARC discover?
In 2016, experts from IARC concluded that consuming very hot beverages could likely cause esophageal cancer—largely on the idea that repeated exposure to very hot liquids causes thermal injury to the esophageal lining (mucosa), contributing over time to cancer development. According to IARC's Monograph, across three systematic reviews, the working group found that higher temperatures were associated with approximately twice the odds of developing ESCC.
While the evidence consistently linked hotter beverages to higher ESCC risk, researchers usually assessed temperature through subjective preference descriptions and adjusted for a variety of confounders. Evidence for a biological mechanism was also limited; IARC considered the genotoxicity evidence weak. Combined with limited animal evidence [1], this led IARC to classify very hot beverages as "probably carcinogenic to humans" (Group 2A).
For those unfamiliar with the IARC classification system, Group 2A generally indicates limited evidence of carcinogenicity in humans combined with either sufficient evidence from experimental animals or strong mechanistic evidence. In this case, however, IARC also considered the animal evidence limited. IARC’s classifications concern hazard, an exposure's potential to cause harm, rather than the magnitude of risk in a specific situation. Professor Sir David Spiegelhalter of the University of Cambridge’s Public Understanding of Risk group stated:
“Being six miles up in a plane is a hazard, but not very risky ... In the case of very hot drinks, the IARC concludes they are probably hazardous, but can’t say how big the risk might be. This may be interesting science, but it makes it difficult to construct a sensible response.”
The 2025 Study
Almost a decade after IARC’s assessment, the first large prospective study to test the question in a Western population was published in the British Journal of Cancer. It followed 454,796 adults in the UK, where tea and coffee drinking is close to universal, and examined how preferred beverage temperature and daily consumption volume related to esophageal cancer incidence.
Participants classified their preferred hot-beverage temperature as "very hot," "hot," "warm," or "do not consume hot beverages," and reported how many cups of tea or coffee they drank per day. Over a mean follow-up of 11.6 years, researchers identified 710 cases of esophageal adenocarcinoma and 242 cases of esophageal squamous cell carcinoma (ESCC).
Among people who preferred very hot drinks, the risk of esophageal squamous cell carcinoma (ESCC) increased steadily with the amount they drank. Compared with non-drinkers and those who preferred warm beverages, risk was about 2.5 times higher with up to four cups a day, nearly 5 times higher with six to eight cups, and more than 5.5 times higher with more than eight cups. The pattern was the same for tea, coffee, or both. No similar increase was seen for esophageal adenocarcinoma, the more common form of esophageal cancer in the West.
The authors concluded that consuming larger quantities of very hot beverages was associated with a higher, dose-dependent risk of ESCC, but not adenocarcinoma, a pattern consistent with earlier studies from South America and Iran, where most esophageal cancer cases are also ESCC. They also noted that beverage consumption and temperature preferences were self-reported and could have changed over the follow-up period; as an observational study, this analysis cannot establish causality.
Given those limits, how much further does this take us toward showing that very hot beverages actually cause esophageal cancer?
The New Study
The most recent study, published in the International Journal of Cancer, pooled data from nearly 980,000 participants across two large UK cohorts: the UK Biobank and the Million Women Study (MWS), a prospective cohort of 1.3 million women recruited between 1996 and 2001. Researchers collected sociodemographic, anthropometric, lifestyle, and health data at intervals of approximately three to five years.
Together, the analysis included 977,282 participants. In both cohorts, roughly 43% and 44% of participants, respectively, reported drinking six or more hot beverages a day, while about 15% preferred them very hot. Over a mean follow-up of 11 to 14 years, researchers identified about 1,045 cases of ESCC and 1,303 cases of esophageal AC. The researchers looked at risk in two ways:
-
Temperature: ESCC risk was 1.8× higher for those who preferred hot drinks and 3.5× higher for those who preferred very hot drinks. There was no similar increase in adenocarcinoma.
-
Amount consumed: after accounting for temperature, people who drank six or more hot beverages a day had about 1.7× the ESCC risk of those who drank fewer. Conversely, even after accounting for amount consumed, temperature still mattered: compared with warm drinks, risk was 1.7× higher for hot drinks and 3.2× higher for very hot drinks.
In short, both hotter drinks and greater consumption were independently associated with higher ESCC risk.
Assuming the relationship is causal, the authors estimated that 14% of ESCC cases could be prevented if people who drink very hot beverages switched to hot ones. The practical message is simple: both temperature and frequency appear to matter, and allowing very hot drinks to cool could reduce risk. There are important limitations. The researchers did not actually measure beverage temperature; participants reported whether they preferred their drinks “hot” or “very hot.” Those labels are subjective, so two people choosing “very hot” may have been drinking beverages at quite different temperatures.
Should we avoid very hot drinks?
It may be reasonable to drink tea and coffee at a lower temperature, but it's worth understanding the nuance in this research. The evidence linking higher temperatures to ESCC risk is consistent, yet we still lack confirmation of causality. A major complication is that beverage temperatures were self-reported, and different people likely have different perceptions and tolerances of what counts as "hot" or "very hot."
As Muy-Teck Teh, a professor of oral molecular oncology, noted, it remains unclear whether people who tolerate or prefer very hot drinks differ in thermosensation or mucosal sensitivity that could be associated with ESCC risk. These limitations don't invalidate the findings, but they do mean we should treat the results as evidence of a robust association rather than definitive proof that beverage temperature causes cancer. Alcohol and smoking are also linked to esophageal cancer; finding an association between beverage temperature and ESCC doesn't mean that temperature is the main contributor to the disease.
In an interview with the BBC, Cancer Research UK also noted the difficulty of determining exactly how long a beverage needs to cool before it reaches a safer temperature. For those concerned, the recommendation was to wait until the drink reaches a comfortable temperature rather than consuming it while it is still very hot.
Nevertheless, beverage temperature is an easily modifiable factor, which makes it a change many people could realistically adopt. This doesn't mean giving up hot drinks entirely on a cold day—few things are as enjoyable as hot chocolate on a cold, rainy night. Instead, given the consistent association and the simple measure available to reduce exposure, letting a drink cool slightly before drinking it may be a reasonable precaution.
[1] "Limited evidence" means that chance, bias, or confounding could not be ruled out with reasonable confidence.
Category
