Longitudinal association of serum carotenoids and tocopherols with hostility: The CARDIA study

Tetsuya Ohira, Atsushi Hozawa, Carlos Iribarren, Martha L. Daviglus, Karen A. Matthews, Myron D. Gross, David R. Jacobs

Research output: Contribution to journalArticlepeer-review

10 Scopus citations


Hostility is a personality trait associated with increased risk of coronary heart disease. No study has reported the association between hostility and antioxidants, which may be mediators for atherosclerosis. CARDIA (Coronary Artery Risk Development in Young Adults) Study participants were 3,579 men and women 18-30 years of age in 1985-1986. Serum carotenoids and tocopherols were measured at years 0 and 7, and hostility was measured at years 0 and 5. Analysis of covariance was used to test for covariate-adjusted differences in serum carotenoids and tocopherols across quartiles of hostility. After adjustment for age, gender, race, serum lipids, and baseline of the dependent variable, the mean carotenoid values at year 7 of the lowest and highest quartiles of hostility score at year 0 were 3.9 and 3.3 μg/liter for alpha-carotene (p < 0.001), 9.1 and 8.0 μg/liter for beta-cryptoxanthin (p < 0.001), and 50.6 and 46.8 μg/liter for the sum of four carotenoids (p < 0.001). Hostility scores at year 0 were unrelated to year 7 lycopene and tocopherols. In contrast, neither year 0 carotenoids nor tocopherols predicted the hostility score at year 5. High hostility predicted future low levels of some serum carotenoids, which may help to explain the association of hostility and cardiovascular risk observed in other epidemiologic studies.

Original languageEnglish (US)
Pages (from-to)42-50
Number of pages9
JournalAmerican journal of epidemiology
Issue number1
StatePublished - Jan 2008


  • Antioxidants
  • Carotenoids
  • Health behavior
  • Hostility
  • Longitudinal studies


Dive into the research topics of 'Longitudinal association of serum carotenoids and tocopherols with hostility: The CARDIA study'. Together they form a unique fingerprint.

Cite this