Alcohol and Cancer: Why WHO Says There Is No Safe Level

What does WHO say about alcohol and cancer?

In January 2023, WHO together with IARC (the International Agency for Research on Cancer) published their position in The Lancet Public Health: "When it comes to alcohol consumption, there is no safe level that does not affect health." IARC places alcohol in Group 1 — the highest carcinogenicity category — alongside tobacco smoke, asbestos, and ionising radiation.

This is not a new classification: IARC included alcohol in Group 1 as far back as 1988. What was new was the emphasis: no amount of alcohol is safe with respect to cancer risk, and half of all alcohol-attributable cancer cases in the WHO European Region arise from light and moderate consumption — less than 1.5 litres of wine, 3.5 litres of beer, or 450 ml of spirits per week.

How many cancer cases worldwide are linked to alcohol?

Rumgay and colleagues (Lancet Oncology, 2021) conducted a population-based study of the global cancer burden attributable to alcohol in 2020. Key findings:

• 741,300 new cancer cases in 2020 — 4.1% of all cancers globally.
• By consumption level: heavy drinking (>60 g/day) — 346,400 cases; moderate (10–60 g/day) — 103,100 cases; light (<10 g/day) — 41,300 cases.
• By cancer site: oesophagus — 189,700; liver — 154,700; female breast — 98,300 cases.
• Men accounted for 77% of all attributable cases.

By 2021, an updated analysis (Danpanichkul et al., Alimentary Pharmacology & Therapeutics, 2025; GBD 2021 dataset) found that mortality from alcohol-attributable cancer reached 343,370 deaths — 51% more than in 2000, even though age-standardised mortality rates were declining. The rise in absolute numbers reflects global population growth and increased alcohol consumption in low- and middle-income countries.

Which cancer types does alcohol cause?

IARC has established a causal link between alcohol consumption and seven cancer sites:

• Oral cavity and pharynx cancer
• Laryngeal cancer
• Oesophageal cancer (squamous cell)
• Liver cancer (hepatocellular carcinoma)
• Colorectal cancer
• Female breast cancer

Rumgay and colleagues (Nutrients, 2021) report the increase in relative risk per 10 g/day of alcohol (approximately one standard drink): oesophageal cancer — RR 1.25 (95% CI: 1.12–1.41); colorectal cancer — RR 1.07 (1.05–1.08); breast cancer — RR 1.07 (1.05–1.09); oral cavity cancer — RR 1.15 (1.09–1.22).

In the United States in 2019, alcohol was linked to approximately 95,000 new cancer cases (5.4% of all cancers) and around 24,000 deaths (Islami et al., CA: A Cancer Journal for Clinicians, 2024). The largest absolute number was breast cancer: approximately 44,000 cases, representing 16% of all breast cancer cases in the country.

Is there a safe dose? What meta-analyses show

A meta-analysis by Jun and colleagues (Epidemiology and Health, 2023) pooled data from 106 studies with follow-up periods of 10–47 years. At light drinking levels (0.01–12.4 g/day, less than one standard drink), overall cancer risk did not reach statistical significance (RR 1.02, 95% CI: 0.99–1.04). However, risk was elevated for specific cancer sites even at this dose:

• Oesophageal cancer: RR 1.39 (95% CI: 1.10–1.75)
• Breast cancer: RR 1.05 (95% CI: 1.04–1.07)
• Colorectal cancer: RR 1.04 (95% CI: 1.02–1.07)

At moderate drinking (12.5–24.9 g/day — approximately 1–2 standard drinks per day), the combined risk of all cancers was RR 1.08 (1.04–1.12). At heavy drinking (50+ g/day) — RR 1.39 (1.29–1.49) for all cancers combined; for oesophageal cancer — RR 3.94 (3.04–5.10).

A separate systematic review of breast cancer (Sohi et al., Alcohol: Clinical and Experimental Research, 2024), covering literature through November 2023, confirmed that risk is elevated even below one drink per day, with no lower threshold at which risk disappears. The maximum relative risk increase at high consumption levels was 20–25%.

What happens when alcohol is stopped?

The IARC Working Group (NEJM, 2023; Gapstur, Lauby-Secretan et al., 15 scientists from 8 countries) analysed evidence on cancer risk reduction following reduced or stopped alcohol consumption. Conclusions:

• Sufficient evidence: reducing or stopping alcohol decreases the risk of oral cavity and oesophageal cancer.
• Limited but positive evidence: for laryngeal, colorectal, and breast cancer.

The Canadian Centre on Substance Use and Addiction (CCSA, 2023), based on analysis of approximately 6,000 studies, issued updated guidance: 3–6 drinks per week already constitutes moderate risk, including cancer risk; more than 7 constitutes high risk. One bottle of wine per week carries a cancer risk equivalent to approximately 5 cigarettes per week for men, or 10 for women.