Coffee linked with lower throat cancer risk — but higher laryngeal cancer risk

Design

Dose-response, site-specific systematic review and meta-analysis of observational studies

Population

30 studies of coffee intake and head and neck cancer

Primary outcome

Adjusted odds of head and neck cancer by intake and subsite

Effect

Overall odds ratio 0.78; oral 0.71, pharyngeal 0.80; laryngeal higher at 1.13

Smoking, alcohol and HPV are the established head and neck cancer risks; coffee's role has been unclear. This dose-response meta-analysis of 30 studies pooled adjusted odds ratios by intake level and tumour site.

Higher intake was associated with lower overall head and neck cancer risk (odds ratio 0.78), with the clearest reductions at three or more cups a day for oral and pharyngeal cancer. The striking exception was the larynx, where coffee was associated with a modestly higher risk (odds ratio 1.13). So the effect is not uniform across the upper aerodigestive tract.

This is observational and cannot establish cause. Coffee intake correlates with smoking and alcohol, and residual confounding or reverse causation could drive either direction. It is a signal worth knowing when a patient asks, not a basis to recommend coffee for cancer prevention — and certainly not reassurance that overrides the dominant, modifiable risks of tobacco and areca nut, which matter most in India.

• Higher coffee intake was associated with lower overall head and neck cancer risk (odds ratio 0.78, 95% CI 0.72–0.86).
• Reduced risk was clearest for oral (0.71) and pharyngeal (0.80) cancer at 3 or more cups a day.
• Laryngeal cancer risk was modestly higher with coffee (odds ratio 1.13, 1.04–1.24).
• Do not recommend coffee for prevention; keep the focus on tobacco, areca nut and alcohol cessation.

Why it matters

It answers a common patient question with nuance, and warns against reading a single protective number across all head and neck sites.

Don't overread it

This was observational — coffee drinkers differ in smoking and alcohol use, so the associations cannot establish that coffee causes or prevents these cancers.

The statistics, in plain English

An odds ratio of 0.78 suggests about a fifth lower odds, but opposite directions at different sites (0.71 oral versus 1.13 larynx) are a clue that confounding may differ by subsite. Observational associations like this cannot show coffee itself changes risk.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.