| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EDITORIAL COMMENT |
Washington University School of Medicine
St. Louis, MO
clon{at}tci.wustl.edu
The need for a paradigm shift is usually signaled when large-scale studies that appear to be well designed produce inconsistent and contradictory results. Contradiction forces investigators to rethink the adequacy of long-cherished assumptions and conventional study designs. Without the constructive force of contradiction, investigators blame inconsistency on small samples, rather than questioning the adequacy of their basic assumptions and methods of analysis. For example, psychosomatic medicine has long sought to identify the personality traits that are predictive of specific diseases, such as type A personality traits for heart disease. Other work has sought to identify the personality traits that are predictive of mortality from all causes. However, health is more than the absence of disease. Healthy configurations of personality predispose to longevity and effective utilization of medical treatment, whereas unhealthy personalities predispose to mortality as a result of susceptibility to diseases or to poor utilization of medical treatment. Mortality is better predicted by the absence of positive emotions than by the presence of negative emotions (1). In turn, positive emotions are the result of well-integrated configurations of personality that lead to good insight, judgment, and satisfaction with life (2). As a result of the multiple psychosomatic processes that influence mortality, studies of the average effects of differences among individuals in personality are likely to lead to inconsistent results. The inconsistencies are likely as a result of ignoring the differential personality dynamics that influence susceptibility to disease, utilization of health-promoting activities, and conscientious adherence to medical care.
For example, two large-scale studies, one in this issue of Psychosomatic Medicine and one in the last, report contradictory results about the personality traits that predict mortality in individuals over 65 years of age. Wilson and colleagues (3) identified over 6000 individuals over 65 years of age in a general community-based census regardless of their medical health status. During a 6-year follow-up, nearly 40% died. People who had high levels of neuroticism or low levels of extraversion were more likely to die during the follow-up. These effects of personality were largely explained by associated differences in the cognitive, social, and physical activity at baseline.
In the study published in the last issue, Weiss and Costa (4) followed over 1000 individuals over 65 years of age who were in frail health and participating in a Medicare demonstration project. The participants lived in the community but were required to have a recent history of significant use of health care services and to need help with their activities of daily living. Nearly 40% of their subjects also died during a 5-year follow-up period. In contrast to the findings of Wilson and colleagues (3), high neuroticism or low extraversion did not predict increased mortality. Rather, Weiss and Costa (4) observed that high neuroticism and agreeability were weakly associated with survival. Their strongest finding was that people with high levels of conscientiousness were more likely to survive than those with average or low levels of conscientiousness. There was no difference in survival between individuals who were average or low in conscientiousness. Weiss and Costa (4) did not report data about associated differences in cognitive, social, and physical activity at baseline or in adherence to medical care during the follow-up.
Weiss and Costa (4) suggest that findings about conscientiousness are simply more consistent than findings about neuroticism and other personality traits used to predict mortality. This may be true but provides no answers to help us understand the frequent inconsistent findings about how personality influences mortality. It seems likely to me that multiple nonlinear processes interact within individuals to influence mortality. Specifically, high conscientiousness is an important influence on adherence to recommendations for medical treatment, whereas neurotic introverts have less healthy lifestyles, which increase their susceptibility to diseases. To test these hypotheses, future research on personality and mortality should to be extended in three major ways. First, predictors of health, susceptibility to disease, and adherence to treatment recommendations should be differentiated because they are distinct psychosomatic processes (5). Second, both positive and negative emotionality should be measured to assess quality of life, which is an important index of integration of personality and the risk of mortality (1,2). Third, analyses should focus on the nonlinear interactions of variables within individuals, rather than on the average effects of variables that differ among individuals (6).
Unfortunately, before the development of modern methods for studying nonlinear dynamical systems, many researchers became accustomed to analyzing the average effects of individual variables that differ among individuals. Most clinicians, and many academicians, still do not recognize that the average effects of differences among individuals may be almost irrelevant to what is happening in any particular person. "Linear" thinking systematically neglects the interactions that are crucial for understanding the development of the individual as a person over time and different situations. When inconsistent results occur, investigators who are accustomed to thinking in a linear way often attribute the contradictions to small sample sizes or differences in experimental details. In other words, there is strong resistance to facing the fact that contradictory findings may require a paradigm shift in which there is fundamental change in our way of understanding the complex processes of psychosomatic development.
Ironically, linear thinking has often been used to describe the structure of differences among individuals in their personality by means of factor analysis, which assumes that measured variables have an integral scale with the same average effects throughout their range. This assumption proved inadequate in both studies of personality (3,4). In fact, the differences among individuals may tell us little about the processes that are active within a person as they adapt to cognitive, social, and physical challenges with increasing age or when they must interact with peers, family members, and medical authorities (7). Fortunately, there are ways of measuring personality and its within-person dynamics that are based on an understanding of psychosomatic principles, rather than factor analysis (2).
Recent research has shown that many genetic and environmental influences interact in complex adaptive systems that influence the development of personality and many common disorders that increase mortality. As a result of nonlinear interactions among genetic and environmental factors, the average effect of any one gene is inconsistent from study to study (2). Consistent findings are recognized only when complex patterns of genetic and environmental effects within individuals are studied but not when single variables that differ among individuals are studied. For example, the personality trait of novelty seeking is increased only when individuals have particular sets of genes that increase the excitability of prefrontal neurons and the person is treated as a child with cold and punitive discipline (8,9). When interactions among genes or among genetic and environmental variables are ignored, the pattern of inconsistency across studies is inevitable.
Likewise, longitudinal research has shown that focus on the average effects of individual variables that differ among individuals in groups often fails to produce consistent results when the behavior under study depends largely on complex configurations of variables within individuals (6). Studies of the average effects of differences among individuals often do not permit any valid conclusions about their effects in a specific person. As clinicians, we recognize the uniqueness of interacting processes within the person as a whole when we take a careful narrative history of a person’s lifetime development. We take individual life histories in order to try to understand the unique interactions of many processes that operate within a person as he or she adapts to a changing internal and external environment with age, illness, and treatment. The same attention to dynamic psychosomatic processes within individuals is needed to clarify the multiple ways that personality influences mortality.
DOI:10.1097/01.psy.0000189130.15870.93
 |
REFERENCES |
|---|
 TOP REFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
