Danaei · PLoS medicine 2009 · comparative risk assessment modeling study · n=?

The preventable causes of death in the United States: comparative risk assessment of dietary, lifestyle, and metabolic risk factors.

Cited 2894 times in the scientific literature.

Level 3 - non-randomized controlled study

Comparative risk assessment modeling study integrating national health surveys, vital statistics, and meta-analytic relative risks (design analogy).

PubMed 19399161 · doi:10.1371/journal.pmed.1000058 · record verified 2026-08-29

What was done

The authors estimated the annual deaths attributable to 12 modifiable dietary, lifestyle, and metabolic risk factors in the United States for 2005 using comparative risk assessment methods. Exposure prevalence in the US population was drawn from nationally representative health surveys, while disease-specific mortality counts came from the National Center for Health Statistics. Etiological effect sizes by age and sex were obtained from systematic reviews and meta-analyses of epidemiological studies that adjusted for major confounders and regression dilution bias.

What was found

In 2005, tobacco smoking caused an estimated 467,000 deaths (95% CI: 436,000–500,000) and high blood pressure caused 395,000 deaths (95% CI: 372,000–414,000), accounting for roughly 1 in 5 and 1 in 6 adult US deaths, respectively. Overweight-obesity (216,000 deaths; 95% CI: 188,000–237,000) and physical inactivity (191,000 deaths; 95% CI: 164,000–222,000) each accounted for nearly 1 in 10 deaths. The leading dietary risks were high salt (102,000 deaths; 95% CI: 97,000–107,000), low omega-3 fatty acids (84,000 deaths; 95% CI: 72,000–96,000), and high trans fatty acids (82,000 deaths; 95% CI: 63,000–97,000). Alcohol use averted 26,000 cardiovascular and diabetes deaths (95% CI: 23,000–40,000) but caused 90,000 deaths (95% CI: 88,000–94,000) from other cardiovascular diseases, cancers, liver cirrhosis, pancreatitis, alcohol use disorders, injuries, and violence.

Why it matters

This study provides a unified, comparative accounting of preventable deaths in the US, establishing that smoking, hypertension, obesity, inactivity, and poor diet are the primary modifiable targets for public health intervention.

Limits

The findings depend on mathematical attribution modeling rather than direct intervention data. Effect sizes derive from observational epidemiological cohorts vulnerable to residual confounding. Exposure data rely on survey accuracy and self-report. Sample sizes for underlying surveys and meta-analyses were not reported in the abstract.

Cited by