Methods of assessing personality for epidemiological study.

"Personality has different meanings for theologians, philosophers and sociologists, and in psychology it has been used in many ways" wroteGW Allport some 50 years ago.' Writing today, he might well have included epidemiologists amongst those challenged to develop their own approach to the problem of personality. Popularly, personality is understood largely in terms of traits or types using categories like Eysenck's "introvert" or "neurotic".2 In a more technical sense, the term "personality" is a useful rubric for the more complex aspects of psychological function which account for consistent patterns of behaviour characterising individuals. Implied in this definition are concepts like intelligence, emotional disposition, and patterns of social interactions. This distinguishes personality tests such as IQ tests from physiological tests such as galvanic skin response and measures of the social and physical environment such as a life events scale. Personality assessment began with concern over mental health.3 The first personality tests were measures of general intellectual performance and were developed to identify children requiring remedial education.4 Tests of emotional adjustment were then developed for military personnel selection in World War I' These applications illustrate the two main ways personality assessment has been used. In the first, personality is conceived as the dependent variable of a pathological process. In the second, personality is an independent variable predicting behaviour. Early in the history of epidemiology, personality was considered almost exclusively as a dependent variable ofa disease process. This confined the types of legitimate hypothesis largely to those involving mental illness6 or poor intellectual performance.7 More recently, however, personality is being increasingly considered as an independent variable in a variety of health related processes, including disease processes,8 therapeutic processes,9 10 and public health programmes. 1" This simple development of thought has facilitated a much greater variety of hypothesis involving personality. For this reason it is becoming increasingly important for the epidemiologist to be familiar with the opportunities afforded by, and the limitations of, personality assessment.