Using self-report measures in neurobehavioural toxicology: can they be trusted?

Questionnaires are one of the most common methodologies used in research on neurobehavioural effects in occupational and environmental health, most commonly for gathering information on demographic characteristics, psychological or neurological symptoms, mood state, or exposure to hazards. Questionnaires are self-report measures, so by definition are subjective, although their degree of subjectivity depends on the phenomenon they are measuring. For some phenomena questionnaires are used because they are convenient but the information can be obtained from other sources. For other phenomena questionnaire or self-report is the only way of obtaining the information, for example, feelings and experiences, mood or emotions. Questionnaires are essential tools in psychological and neurobehavioural research as they can tap into aspects of nervous system function that cannot be readily measured in other ways. Despite the obvious need for self-report measures, there are a number of serious issues that threaten their validity as effective indicators of neurobehavioural function. This paper considers the implications of some of the major problems with self-report measures, focusing particularly on current approaches to measurement of symptoms and mood. It includes issues relating to validity of measures such as demand characteristics, malingering and under or over reporting, individual differences and problems of language and question style. It also includes issues relating to the interpretation of self-report measures, the relationship between self-report and performance measures, whether they reflect primary or secondary effects and whether they can be used as diagnostic criteria for neurobehavioural functional effects of occupational or environmental exposure. The paper looks at some of the current approaches to overcoming these problems including using interviews and observational methods and improving psychometric qualities of these measures. Self-report measures are important tools in our arsenal of measures of the neurobehavioural effects of occupational and environmental exposure, but they need to be used with care.