Education and smoking: confounding or effect modification by phenotypic personality traits?

Ann Behav Med. 2009 Dec;38(3):237-48. doi: 10.1007/s12160-009-9142-3. Epub 2010 Jan 5.


Background: Little is known about whether educational gradients in smoking patterns can be explained by financial measures of socioeconomic status (SES) and/or personality traits.

Purpose: To assess whether the relationship of education to (1) never smoking and (2) having quit smoking would be confounded by financial measures of SES or by personality; whether lower Neuroticism and higher Conscientiousness would be associated with having abstained from or quit smoking; and whether education effects were modified by personality.

Method: Using data from the Midlife Development in the US National Survey, 2,429 individuals were classified as current (n = 695), former (n = 999), or never (n = 735) smokers. Multinomial logistic regressions examined study questions.

Results: Greater education was strongly associated with both never and former smoking, with no confounding by financial status and personality. Never smoking was associated with lower Openness and higher Conscientiousness, while have quit was associated with higher Neuroticism. Education interacted additively with Conscientiousness to increase and with Openness to decrease the probability of never smoking.

Conclusions: Education and personality should be considered unconfounded smoking risks in epidemiologic and clinical studies. Educational associations with smoking may vary by personality dispositions, and prevention and intervention programs should consider both sets of factors.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Adult
  • Aged
  • Confounding Factors, Epidemiologic
  • Educational Status*
  • Female
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Personality Inventory
  • Personality*
  • Risk Factors
  • Smoking / epidemiology*
  • Smoking / psychology*
  • Smoking Cessation / psychology*
  • Smoking Cessation / statistics & numerical data*
  • Socioeconomic Factors
  • Surveys and Questionnaires
  • United States