Association of stress, hostility and plasma testosterone levels.

OBJECTIVES Many studies assessing the role of sex hormones, like testosterone, on stress and hostility factors have been primarily conducted in selected atypical populations such as violent criminals as well as androgen users and abusers. Therefore, the main aim of the current study was to investigate the association between testosterone levels and two psychosocial variables: stress and hostility in a cohort of healthy individuals who were members of a health maintenance organization (HMO). METHODS At five quarterly visits, psychosocial scales and blood draws were collected. Psychological stress was measured by using several scales that assessed different types of stress, including daily hassles, major life events and perceived stress. Similarly, different aspects of hostility were measured, among them cynicism, hostile affect and aggressive responding. Plasma collected from each visit was used for testosterone level determinations. RESULTS Testosterone levels were significantly associated with stress in both males and females. However, whereas this association exhibited a "threshold effect" in males, it demonstrated a direct and continuous linear relationship between these variables in females. Hostility was not correlated with testosterone levels in neither males nor females. CONCLUSIONS These results suggest that testosterone levels in normal males and females may be more reflective of an intricate balance between physiological responding and emotional coping to stressors than the hostility profile of the individual.