optimism on predisease processes. This is a topic warranting further research, since the immune system can exhibit considerable fluctuation (plasticity) while remaining within normal operating range. A pessimistic explanatory style, namely the tendency to explain negative events in terms of internal, stable, global qualities in oneself, may represent a general risk factor for disease and early mortality. In one study, pessimistic explanatory style was measured at age 25 and health at ages 45 to 60 (Peterson et al., 1988). Those high in pessimistic explanatory style had significantly poorer health two to three decades later. A study of elderly people showed that those with a pessimistic explanatory style had compromised cell-mediated immunity (Kamen-Siegel et al., 1991), which may represent elevated risk for immune-related disorders in this vulnerable population.

Negative affective states adversely affect vulnerability and the courses of a range of health disorders. For example, chronic negative affect was associated with more severe respiratory illness (measured as mucous production) following experimentally induced exposure to a cold virus (Cohen et al., 1995). A meta-analysis revealed an association between prior depression and subsequent development of coronary heart disease (Booth-Kewley and Friedman, 1987). Among already diagnosed patients, those who became depressed in response to their diagnosed coronary artery disease were more likely to have a more debilitating course of illness and a repeat cardiac event, after controlling for other risk factors (Frasure-Smith et al., 1995).

The broad base of empirical evidence demonstrating the significance of these and related psychological states in pathways to many acute and chronic disease outcomes underscores the importance of continuing to explore their role in initiating, exacerbating, and moderating these predisease processes.

Behavioral Factors

A set of behavioral risk factors, including poor diet, little exercise, promiscuous and/or unprotected sexual activity, smoking, alcohol abuse, and drug abuse, is associated with a broad array of diseases. Continued attention to the development and modification of these behaviors is essential.

Research suggests intergenerational transfer of risk behaviors that point to the necessity for early intervention. For example, mothers who smoked during pregnancy were more likely to have adolescent daughters who smoked, even after controlling for the mothers' postnatal smoking histories (Kandel et al., 1994); this suggests that nicotine or other substances in tobacco released by maternal smoking may have affected the fetus, perhaps through nicotinic input to the dopaminergic system. These changes appear to predispose the brain during a critical period of its development to the



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement