“Drinking very hot tea or coffee triples the risk of oesophageal (squamous cell) cancer, according to a study by Oxford Population Health's Cancer Epidemiology Unit”
Plain restatementA 2026 observational cohort study from Oxford Population Health's Cancer Epidemiology Unit reported that people who said they usually drank tea or coffee "very hot" had roughly three times the rate of oesophageal squamous cell carcinoma compared with people who said they drank the same beverages "warm."
This one largely checks out. A real study from Oxford Population Health's Cancer Epidemiology Unit, published in the International Journal of Cancer in 2026, followed 977,282 UK adults from the UK Biobank and Million Women Study for 11 to 14 years. People who said they drank their tea or coffee "very hot" had about three times the rate of oesophageal squamous cell carcinoma compared with people who said they drank them "warm," and the link did not appear for the other main type of oesophageal cancer. Two things are worth adding. First, this is an observational study based on people describing their own drink temperature, not measuring it, so it shows a strong association rather than proof that heat causes the cancer. Second, the baseline risk is low, roughly a 1% lifetime risk in the UK, so tripling it is a real but small absolute increase, and independent experts note that smoking, alcohol, obesity and reflux remain far bigger drivers of oesophageal cancer than drink temperature. The post's claim that some heavy very-hot drinkers face over five times the risk appears to come from a different 2025 study rather than this one, and the suggestion that drink temperature is "the biggest predictor" of this cancer is not supported. The researchers' own advice is simply to let a drink cool a little before the first sip.
[drifted from the evidence:] Drinking very hot tea or coffee triples the risk of oesophageal (squamous cell) cancer, according to a study [drifted from the evidence:] by Oxford Population Health's Cancer Epidemiology Unit
A [added by the neutral restatement:] 2026 observational cohort study [added by the neutral restatement:] from Oxford Population Health's Cancer Epidemiology Unit [added by the neutral restatement:] reported that people who said they usually drank tea or coffee "very hot" had roughly three times the rate of oesophageal squamous cell carcinoma compared with people who said they drank the same beverages "warm.
Red-tinted words in the claim drifted from the evidence. Green-tinted words are what a neutral restatement needs.
The trace / claim to source
- **The study exists and is correctly attributed.** Oxford Population Health's Cancer Epidemiology Unit, published in the International Journal of Cancer, DOI 10.1002/ijc.70654, lead author Keren Papier. This is not a fabricated source.
- **The "triple" figure is accurate for the comparison the study made.** RR 3.17 (95% CI 2.49 to 4.03) for "very hot" versus "warm," specific to squamous cell carcinoma.
- **The subtype qualifier is correct and important.** The claim correctly specifies squamous cell carcinoma rather than oesophageal cancer generally.
- **The sample size (977,282), the two cohorts, and the 11 to 14 year follow-up are all accurate.**
- **The "hot" versus "warm" figure of roughly double is accurate** (approximately 1.7 times).
- **The frequency finding is accurate.** Six or more hot drinks per day was independently associated with roughly 71% higher SCC risk after adjusting for temperature.
- **The adenocarcinoma null result is accurate.** RR 0.92 (0.76 to 1.12), no clear association.
- **The IARC 65°C Group 2A classification is accurately described.**
- **The caption's framing that the absolute risk remains low, and that the practical advice is to let drinks cool, matches what the researchers themselves said.** This is a meaningful point in the post's favour. ---
- **Unsupported causal inference (in the verb).** "Triples the risk" states a causal effect. This is an observational cohort study. Oxford's own press release uses "links to" and "associated with," and an independent expert explicitly described the findings as associations rather than causal proof. The distortion is modest because the claim does attribute it to a study, but the verb is stronger than the design supports.
- **Omitted comparator.** "Triples the risk" without "compared with people who drink the same beverages warm" invites readers to assume the baseline is the general population or people who avoid hot drinks. It is not.
- **Relative risk presented without absolute risk.** Tripling a roughly 1% lifetime risk is a different real-world proposition from tripling a common risk. The claim headline omits this; the post's caption does partially restore it.
- **Superlative overreach in the caption.** "The biggest predictor of one type of cancer turned out to be something as simple as how hot you like your tea" is not supported. Independent expert commentary identifies temperature as a relatively minor risk factor and smoking, alcohol, obesity and reflux as the dominant drivers of oesophageal cancer.
- **"First large-scale evidence in a Western population" is overstated.** A separate UK Biobank analysis by US National Cancer Institute researchers was published in the British Journal of Cancer in February 2025, which assessed daily intake and preferred temperature of hot beverages in relation to ESCC among 454,796 UK Biobank adults and concluded that the findings provide new evidence that drinking hot or very hot beverages is a risk factor for ESCC in the UK . The Oxford study is the largest in a UK population, which is what the Oxford press release claims. "First" is the post's addition.
- **Mechanism stated with more certainty than the researchers used.** "Heat, specifically, appears to be the mechanism" compresses a stated hypothesis into a finding. The researchers said further work is needed. ---
- **The "over five times the baseline risk" figure for heavy very-hot drinkers could not be confirmed in the Oxford paper.** A very similar figure appears in the *different* 2025 British Journal of Cancer UK Biobank study, where consumer-magazine coverage reported that people who preferred "very hot" drinks and drank up to four cups a day had 2.5 times the risk, and those who drank eight cups daily had approximately 5.6 times the risk . That figure may have been imported from the earlier paper and attributed to the Oxford one. I could not retrieve the Oxford paper's joint temperature-by-frequency table to resolve this. Treat this specific number as unverified as stated.
- **Full text of the Oxford paper was not retrieved.** Effect estimates here come from the institutional press release and from an expert who read the paper and quoted exact confidence intervals. The adjustment set (which confounders were controlled for, and how) is not directly verified.
- **Whether "very hot" as self-reported corresponds to the IARC 65°C threshold is not established.** The post equates them. The study measured a subjective label, not a temperature, and misclassification is the researchers' own stated principal limitation.
- **The 70% versus 71% discrepancy** between the Oxford press release and trade coverage is trivial and reflects rounding of RR 1.71. ---
The study is real, recent, and correctly attributed. Oxford Population Health's press release states that drinking tea and coffee at a very hot temperature, rather than warm, is linked to a threefold increase in the risk of oesophageal squamous cell carcinoma, in a study by researchers at the Cancer Epidemiology Unit published in the International Journal of Cancer. The effect sizes reported in the paper: self-reported consumption of drinks at "very hot" temperatures was associated with a substantially higher risk of oesophageal squamous cell carcinoma (RR 3.17, 95% CI 2.49 to 4.03), but not adenocarcinoma (RR 0.92, 0.76 to 1.12) . After accounting for temperature, high-frequency consumption of six or more hot drinks per day also increased squamous cell carcinoma risk (RR 1.71, 1.51 to 1.94), with only minimal association for adenocarcinoma. The lead author's own phrasing: "Compared with people who reported drinking their beverages 'warm,' those who drank them 'hot' had around a 1.7 times higher risk of SCC, while those who drank them 'very hot' had around a three times higher risk," Papier told Fox News Digital. ---
Complete reasoning
The reply receipt is formatted for pasting into the thread where the claim is circulating.
Compact share page: verify.trueseeker.com/s/9f9d813bef91/ZT7Rv0SbcHBFfeQRnsEar-9
Ask this case
Answers come only from the case file above; nothing is added.
Does drinking very hot tea or coffee cause oesophageal cancer?
The study only shows an association, not proof of cause. It is an observational study, and its own authors describe the link as an association rather than a confirmed causal effect.
How much does the risk actually increase?
People who said they drank tea or coffee 'very hot' had about three times the rate of oesophageal squamous cell carcinoma compared with people who drank them 'warm.' This applies specifically to the squamous cell type of oesophageal cancer, not the other main type, adenocarcinoma, where no link was found.
Is a tripled risk something to worry about in real terms?
The baseline lifetime risk of this cancer in the UK is low, roughly 1 percent, so tripling it is a real but small absolute increase. Independent experts note that smoking, alcohol, obesity and reflux remain much bigger drivers of oesophageal cancer than drink temperature.
Is drink temperature the biggest risk factor for this cancer?
No, this is not supported by the case file. Independent commentary identifies temperature as a relatively minor factor compared with smoking, alcohol, obesity and reflux, which are the dominant known causes.
Where does the claim that heavy very-hot drinkers face over five times the risk come from?
This figure could not be confirmed in the Oxford study. It closely matches a figure from a separate 2025 British Journal of Cancer study using UK Biobank data, so it appears to have been imported from a different study rather than the one being fact-checked.