Associations between dimensional personality measures and preclinical atherosclerosis: The cardiovascular risk in Young Finns study Tom Rosenström a, ⁎, Markus Jokela a , Claude Robert Cloninger b , Mirka Hintsanen c, a , Markus Juonala e , Olli Raitakari d , Jorma Viikari e , Liisa Keltikangas-Järvinen a a IBS, Unit of Personality, Work and Health Psychology, University of Helsinki, Helsinki, Finland b Center for Well-Being, Washington University School of Medicine, St. Louis, USA c Helsinki Collegium for Advanced Studies, University of Helsinki, Finland d Department of Clinical Physiology, Turku University Hospital and Research Centre of Applied and Preventive Cardiovascular Medicine, University of Turku, Turku, Finland e Department of Medicine, Turku University Hospital and University of Turku, Turku, Finland abstract article info Article history: Received 22 October 2011 Received in revised form 13 February 2012 Accepted 15 February 2012 Keywords: Personality Atherosclerosis Carotid intima-media thickness Behavioral medicine Multivariate analysis Temperament and character Objective: To assess how multidimensional personality–trait theories, such as the Psychobiological Model of Temperament and Character, and the Five-factor Model of Personality, are associated with subclinical athero- sclerosis as indicated by carotid intima-media thickness (IMT). The analysis was designed to tolerate non- linear development in which the same personality profiles can have multiple final outcomes and different an- tecedent profiles can have the same final outcome. Methods: 605 men and 844 women (average age 31.6 year, s.d. = 5.0, range = 24–39) provided data on IMT and traits of the psychobiological model, 725 men and 1011 women were assessed for IMT and the five- factor model (age 37.7 year, s.d. = 5.0, range = 30–45). Robust multidimensional Hotelling’sT 2 statistic was used to detect personality differences between participants with high IMT and others. Model-based cluster- ing method further explored the effect. Results: Those with a high level of subclinical atherosclerosis within the sample (highest IMT-decile) had a combined higher persistence (i.e., were perseverative or perfectionistic), more disorganized (schizotypal) character, and more antisocial temperamental configuration than others (P = 0.019). No effect was found for the five-factor model (P = 0.978). Traditional methods that did not account for multidimensionality and nonlinearity did not detect an association. Conclusion: Psychological well-being may have positive effects on health that reduce atherosclerosis in the population as a whole. Increased subclinical atherosclerosis was associated with a profile that combines known risk factors, such as cynical distrust and hostile tendencies. More frequent use of statistical procedures that can cope with non-linear interactions in complex psychobiological systems may facilitate scientific ad- vances in health promotion. © 2012 Elsevier Inc. All rights reserved. Introduction Surgeon and scientist John Hunter, “known for his anxious and argumentative nature (personality)” [1], noticed the association between stress and heart disease already in the 18th century by ob- serving his patients and his own state [1]. The role of stress, and per- sonality traits exposing individuals to stress, has been a topic of considerable interest in health psychology and behavioral medicine. A recent study found that Type D (distressed) personality [2] predicts clinical events after myocardial infarction, above and beyond disease severity and depression [3]. Type D is defined as a joint tendency to- ward negative affectivity and social inhibition. Several other studies have also found associations between different personality traits and cardiac risk factors [4–6]. In particular, extensive research has shown that personality components of hostility, including anger- proneness, cynical distrust, and paranoia, increased the risk of heart disease [7–15]. Likewise, components of personality related to anxiety-proneness, such as perfectionism, may be significant but are weaker and less consistent when their average effects are considered using linear regression analysis [16–19]. Following Gordon Allport's early definition of personality as “the dy- namic organization within the individual of those psychophysical sys- tems that determine his unique adjustments to his environment” [20], most theories of personality have embraced the idea of personality as Journal of Psychosomatic Research 72 (2012) 336–343 Abbreviations: IMT, carotid intima-media thickness; TCI, Temperament and Character Inventory; NEO-FFI, a version of the five factor personality inventory; NS, Novelty seeking; HA, Harm avoidance; RD, Reward dependence; P, Persistence; SD, Self-directedness; C, Co- operativeness; ST, Self-transcendence. ⁎ Corresponding author at: IBS, Unit of Personality, Work and Health Psychology, University of Helsinki (Siltavuorenpenger 1 A), P.O. Box 9, 00014, Helsinki, Finland. Tel.: +358 9 1912 9396; fax: +358 9 1912 9521. E-mail address: tom.rosenstrom@helsinki.fi (T. Rosenström). 0022-3999/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jpsychores.2012.02.003 Contents lists available at SciVerse ScienceDirect Journal of Psychosomatic Research