Validity and bias in epidemiological research

Abstract

Some of the major concepts of validity and bias in epidemiological research are outlined in this chapter. The contents are organized in four main sections: Validity in statistical interpretation, validity in prediction problems, validity in causal inference, and special validity problems in case–control and retrospective cohort studies. Familiarity with the basics of epidemiological study design and a number of terms of epidemiological theory, among them risk, competing risks, average risk, population at risk, and rate, is assumed. Despite similarities, there is considerable diversity and conflict among the classification schemes and terminologies employed in various textbooks. This diversity reflects that there is no unique way of classifying validity conditions, biases, and errors. It follows that the classification schemes employed here and elsewhere should not be regarded as anything more than convenient frameworks for organizing discussions of validity and bias in epidemiological inference. Several important study designs, including randomized trials, prevalence (cross-sectional) studies, and ecological studies, are not discussed in this chapter. Such studies require consideration of the validity conditions mentioned earlier and also require special considerations of their own. A number of central problems of epidemiological inference are also not covered, including choice of effect measures, problems of induction, and causal modelling.