A patient I treated years ago — a 54-year-old stonemason — was diagnosed with bladder cancer six months after his last appointment with me. He had smoked for 30 years and worked a decade around silica dust. He asked me a question I still think about: “Doc, did I do this to myself?” The honest answer, which I gave him gently, was: yes, partly. And the harder truth is that the tools to reduce that risk had existed for decades.

A comprehensive analysis published in Nature Medicine in April 2026 — produced by researchers at IARC/WHO and covering 185 countries — put precise numbers on what clinicians like me have long suspected. Of 18.7 million new cancer cases diagnosed worldwide in 2022, an estimated 7.1 million (37.8%) were directly attributable to 30 modifiable risk factors. That stonemason’s cancer was part of a global pattern, and the pattern is overwhelmingly preventable.

How the study was built — and why the numbers matter

The researchers used GLOBOCAN 2022 incidence data across 36 cancer sites and applied prevalence data from approximately 2012 to account for the well-documented 10-year latency between exposure and cancer onset. This methodological detail matters: when you see today’s cancer numbers, you’re looking at what people were exposed to a decade ago. The exposures people are accumulating today will show up in diagnosis rooms in the 2030s.

The 30 modifiable risk factors fall into four categories: behavioral (smoking, alcohol, high BMI, insufficient physical activity, smokeless tobacco, suboptimal breastfeeding), environmental (air pollution, ultraviolet radiation), infectious agents (9 pathogens including H. pylori, HPV, HBV, HCV, and Epstein-Barr virus), and occupational exposures (13 carcinogens including asbestos, benzene, and silica dust).

The population-attributable fraction (PAF) approach used here estimates what proportion of cases would theoretically be eliminated if exposure to a specific risk factor were reduced to zero. The combined PAFs — adjusted to avoid double-counting overlapping exposures — reveal both the scale of preventable cancers and where interventions can have the greatest impact.

The GBD 2023 Cancer Collaborators, publishing simultaneously in The Lancet (2025), provided complementary data covering 47 cancer types from 1990 to 2023 with projections to 2050. Their forecast: 30.5 million cancer cases per year by 2050 — a 60% increase from today — if current trends persist. The two analyses together make the case that the cancer burden is not fated; it is manufactured.

Tobacco: still the single largest cause

Tobacco smoking accounts for 15.1% of all new cancer cases globally — 3.3 million people in a single year. What surprises many people is the breadth: smoking is causally linked not just to lung cancer but to at least 15 other cancer types, including bladder, kidney, pancreas, esophagus, and several head and neck cancers.

In men, the burden is particularly concentrated in certain regions. In East Asia, Eastern Europe, South-East Asia, Northern Africa and Western Asia, and Southern Europe, smoking contributes to more than one in every five new male cancer cases. A separate population-based systematic analysis published in eClinicalMedicine (April 2026) used GLOBOCAN 2022 data to confirm that in never-smoking populations — particularly Asian women — lung cancer incidence is rising due to other environmental factors. Tobacco control alone, then, is not the complete picture, but it remains the most impactful single intervention available.

In men, 1,326,453 preventable lung cancer cases were linked to smoking in 2022 — 69.4% of all preventable lung cancers in that sex. The WHO MPOWER program has made measurable progress on reducing male smoking rates globally, but its full impact on cancer incidence will take another decade to manifest, given the latency involved.

Infections: the overlooked cause in lower-income regions

Infectious agents cause 10.2% of all cancer cases globally — approximately 2.3 million cases per year. This is a category that many high-income country residents barely consider, yet it represents the largest preventable cancer driver in 141 countries for women and in much of sub-Saharan Africa and South-East Asia for men.

The breakdown by pathogen is instructive. H. pylori — the bacterium responsible for most peptic ulcers — is the dominant factor in stomach cancer, accounting for 92.2% of preventable stomach cancer cases in women and 73.3% in men. Stomach cancer is the third most common preventable cancer globally. The good news: H. pylori prevalence has been declining in countries with improving sanitation and targeted test-and-treat programs. Japan and South Korea, which carry high baseline rates, have implemented mass eradication programs that are measurably reducing gastric cancer incidence.

HPV accounts for 91% of cervical cancer cases. A 2025 Cochrane network meta-analysis covering all four WHO-prequalified vaccines (Cervarix, Gardasil, Gardasil-9, and Cecolin) confirmed high efficacy against cervical intraepithelial neoplasia and invasive cancer. A 2026 study from The Lancet Primary Care found that vaccine effectiveness against CIN2+ in European population cohorts aligns with clinical trial data. The obstacle is not scientific — it is logistical and economic. In sub-Saharan Africa, where HPV-driven cervical cancer rates are highest, vaccine birth-dose coverage for HBV and catch-up HPV programs remain critically underfunded.

The three highest-burden pathogens:

• H. pylori — stomach cancer (92.2% of preventable cases in women)
• HPV types 16, 18, and 12 others — cervical, oropharyngeal, anal, vulvar, vaginal, penile cancers
• HBV and HCV — liver cancer (53.1% of preventable cases in men from infections)

Alcohol, BMI, and physical inactivity: the behavioral cluster

Alcohol causes 3.2% of all new cancer cases — roughly 700,000 per year. It is the second-largest preventable cause in Eastern Europe and the third-largest in six other regions. A 2025 meta-analysis published in BMC Cancer found that beyond raising cancer risk, alcohol consumption also reduces oncological treatment effectiveness and increases treatment toxicity. This is a finding that changes the clinical conversation: alcohol is not just a risk factor for developing cancer, it complicates treatment for those who already have it.

High BMI and insufficient physical activity together drive a significant portion of preventable cancers, particularly in women. For breast cancer, 28.7% of preventable cases link to high BMI and 32.9% to insufficient physical activity. A comprehensive 2025 systematic review by Watts et al. (NCI) — covering 226 articles and 1.5 million cancer data points — confirmed that elevated BMI is positively associated with 19 cancer types, including leukemia, non-Hodgkin lymphoma, bladder cancer, and glioma. These associations were not previously included in major consensus reports.

The umbrella review by Filis et al. published in Critical Reviews in Oncology/Hematology (2025) analyzed 740 meta-analytic associations between physical activity and cancer outcomes. The evidence is strong: post-diagnosis physical activity is associated with lower all-cause mortality, lower recurrence rates, reduced cancer-related fatigue, and improved mental health in breast, prostate, colorectal, and hematological cancers. Exercise is not a supplement to cancer care. For many patients, it is part of the treatment itself.

Air pollution and ultraviolet radiation: the environmental drivers

Air pollution — specifically fine particulate matter (PM2.5) — accounts for 2.0% of all new cancers globally, representing 434,282 cases in 2022. A 2025 review in the British Journal of Cancer confirmed that PM2.5 from traffic, industry, and biomass combustion carries sulfates, polycyclic aromatic hydrocarbons, and heavy metals that are established lung carcinogens. The review found significant associations between PM2.5 exposure and both lung cancer incidence and mortality.

The regional distribution is telling. In East Asia, air pollution causes approximately 15% of all female lung cancers — a proportion comparable to the contribution of smoking in some high-income regions. In North Africa and Western Asia, it accounts for 20% of attributable lung cancers in men. In low-resource settings, indoor air pollution from biomass fuel cooking — primarily affecting women — adds a second layer of exposure that outdoor monitoring systems consistently miss.

Ultraviolet radiation (UVR) accounts for 1.1% of all new cancers globally but is the leading preventable cause in Oceania, where skin melanoma rates are among the highest in the world. In Australia specifically, UVR-driven melanoma is the top preventable cancer. Comprehensive sun protection policies — shade infrastructure, early childhood sun safety education, and consistent use of high-SPF sunscreen — are cost-effective interventions with strong evidence behind them.

What this means for prevention policy — and for your daily life

The most important message from this body of evidence is that prevention is not abstract. It is specific, regional, and actionable. Different countries face different dominant risk factors, and effective policy has to match that profile.

For high-income countries — North America, Western Europe, Australia — behavioral risk factors dominate: tobacco, alcohol, high BMI, and physical inactivity. The policy toolkit includes tobacco taxation and plain packaging, alcohol minimum pricing, urban design for active transportation, and food environment regulation to reduce ultra-processed food access.

For low- and middle-income countries — sub-Saharan Africa, South-Central Asia, much of Southeast Asia — infectious agents are the primary target. HPV vaccination scale-up, HBV birth-dose vaccine programs, H. pylori test-and-treat initiatives, and hepatitis C treatment access represent the highest-return interventions available.

Five evidence-based steps any individual can take now:

• Stop smoking — or never start. No modification yields a greater cancer risk reduction.
• Get vaccinated against HPV (if under 45) and HBV. Two vaccines, two cancers, one decision.
• Keep alcohol to a minimum. Any amount carries risk; the lower the exposure, the lower the risk.
• Move daily. Even moderate physical activity — 150 minutes per week — measurably reduces cancer risk and improves outcomes after diagnosis.
• Know your environment. If you work with asbestos, benzene, silica, or diesel exhaust, understand your legal protections and request regular medical screening.

The GBD 2023 projections estimate that 57.9% of cancer deaths in 2023 occurred in low- to middle-income countries — yet these same countries carry the lowest cancer surveillance capacity and the least access to prevention infrastructure. This is where investment must go.

I’ve spent 31 years watching what happens when disease reaches the point where a surgeon is the last option. The data in this research is not discouraging — it is directional. We know what causes most cancers. We know what stops them. The gap between knowing and acting is where lives are lost.

References

1. Fink H, Langselius O, Vignat J, et al. Global and regional cancer burden attributable to modifiable risk factors to inform prevention. Nat Med. 2026;32:1306-1315.

2. GBD 2023 Cancer Collaborators. The global, regional, and national burden of cancer, 1990-2023, with forecasts to 2050: a systematic analysis for the Global Burden of Disease Study 2023. Lancet. 2025;406(10512):1565-1586.

3. [Authors]. The global landscape of cancer burden attributable to tobacco smoking in 2022: a population-based systematic analysis. EClinicalMedicine. 2026. DOI: 10.1016/j.eclinm.2026.103873.

4. Watts EL, Gonzalez-Feliciano A, Gunter MJ, et al. Adiposity and cancer: systematic review and meta-analysis. medRxiv. 2025.

5. Fountoukidis G, Schiza A, Valachis A, et al. Effect of alcohol consumption on oncological treatment effectiveness and toxicity: a systematic review and meta-analysis. BMC Cancer. 2025. DOI: 10.1186/s12885-025-13694-z.

6. Filis P, Markozannes G, Chan DS, et al. Grading the evidence for physical activity and any outcome in cancer survivors: an umbrella review of 740 meta-analytic associations. Crit Rev Oncol Hematol. 2025;207:104602.

7. Bergman H, et al. Human papillomavirus (HPV) vaccination for the prevention of cervical cancer and other HPV-related diseases: a network meta-analysis. Cochrane Database Syst Rev. 2024/2025.

8. Wang M, Kim RY, Kohonen-Corish MRJ, et al. Particulate matter air pollution as a cause of lung cancer: epidemiological and experimental evidence. Br J Cancer. 2025;132:986-96.