Observational study / 2014
In adults aged 80 and over, smoking was not linked to increased mortality risk, and long-lived smokers had biomarker levels similar to never smokers.
Data from 4,655 current and never smokers aged 50 and above from NHANES III were analyzed using a proportional hazards model. Smoking significantly increased mortality in most age groups, but it did not increase mortality risk for participants who were age 80 and over at baseline. Additionally, long-lived smokers (80+) had adjusted mean levels of inflammation, HDL, and lung function similar to those of never smokers.
Shows an association, not proof of cause and effect.
- Study type
- Cross-sectional and cohort analysis
- Who was studied
- 4,655 current and never smokers, ages 50 and above, from NHANES III
- What was measured
- Mortality risk and biomarkers (inflammation, HDL, lung function)
THE PAPER
Not all smokers die young: a model for hidden heterogeneity within the human population.
Levine M, Crimmins E. PLoS One, 2014.
This note is written from the study's published abstract. It summarizes what the authors reported; it is not medical advice and does not replace reading the full paper.