Interpretation of epidemiological surveys of asthma.

Two particular issues make the interpretation of epidemiological studies in asthma problematic. The first is the lack of any clear definition of asthma. This is a perennial area of controversy. Thirty-eight years ago a Ciba Foundation guest symposium addressed this issue and suggested a solution. However, as J. G. Scadding, one of the participants of that symposium, pointed out after further consideration of the problem, what they had proposed was a description, not a definition. Since then, further attempts have been made but with little progress. They remain descriptive rather than definitive and have become, if anything, vaguer. The second problem has been the widespread failure to be precise about hypotheses or to define more precisely the hypothetical influences on asthma. Examples of this are the notions of 'inflammation' and 'atopy'. Standardization of methods for epidemiological studies of asthma is likely to provide a more rigorous framework for the comparison of results and the testing of hypotheses. Nevertheless, the development of such protocols should itself be seen as a hermeneutic device rather than an assertion of established knowledge.