Islami · CA: a cancer journal for clinicians 2018 · Comparative risk assessment and population attributable fraction modeling study · n=1,570,975 cancer cases and 587,521 cancer deaths (US population aged >=30 years in 2014)

Proportion and number of cancer cases and deaths attributable to potentially modifiable risk factors in the United States.

Cited 1639 times in the scientific literature.

Level 3 - non-randomized controlled study

Population attributable fraction modeling combining national surveillance data and meta-analytic relative risks (design analogy)

PubMed 29160902 · doi:10.3322/caac.21440 · record verified 2026-08-29

What was done

The authors estimated the proportion and number of invasive cancer cases (excluding nonmelanoma skin cancer) and cancer deaths attributable to potentially modifiable risk factors across 26 cancer types among US adults aged 30 years and older in 2014. Evaluated exposures included active and secondhand cigarette smoking, excess body weight, alcohol consumption, dietary factors (red/processed meat intake, low intake of fruits, vegetables, dietary fiber, and calcium), physical inactivity, ultraviolet radiation, and six cancer-associated infections. Cancer incidence was sourced from CDC and NCI registries, mortality from the CDC, exposure prevalence from nationally representative surveys, and relative risks from published pooled analyses or meta-analyses.

What was found

In 2014, an estimated 42.0% of all incident cancers (659,640 of 1,570,975) and 45.1% of cancer deaths (265,150 of 587,521) were attributable to the evaluated modifiable risk factors. Cigarette smoking contributed the highest burden (19.0% of cases [298,970] and 28.8% of deaths [169,180]), followed by excess body weight (7.8% of cases, 6.5% of deaths) and alcohol intake (5.6% of cases, 4.0% of deaths). By cancer site, lung cancer had the highest attributable burden (184,970 cases, 132,960 deaths), followed by colorectal cancer (76,910 cases, 28,290 deaths).

Why it matters

This study provides comprehensive, contemporary benchmarks identifying that over 40% of US cancer cases and deaths are linked to modifiable lifestyle and infectious factors, reinforcing targets for preventive health policy.

Limits

The analysis relies on population attributable fraction mathematical modeling using observational relative risk estimates rather than direct interventional data. The overall preventable proportion may be underestimated because several established or suspected modifiable exposures could not be quantified due to data limitations.

Cited by