When is a case-control study a case-control study?
Nancy E. Mayo, Mark S. Goldberg
Rehabilitation professionals rarely ask questions about the etiology of health events or outcomes and may not have formal training or relevant experience in the design of studies whose intent is to identify causal factors. The case-control study, which is one design used to answer questions about etiology, is particularly difficult to understand and research has shown that this study design label is often used incorrectly. This paper outlines the main features of case-control studies, with a particular focus on sampling strategies. The goal is to educate clinical rehabilitation colleagues about the fundamental principles of this powerful epidemiologic design. Examples illustrate how the parameters of cumulative incidence, incidence-density, and prevalence are estimated and the effect of sampling strategy on these parameters. Also shown is how sampling strategy affects conclusions drawn about the effects of an exposure on outcome. Even when used appropriately, case-control studies are methodologically complex to design and analyze to ensure an unbiased answer to the research question. The hypothetical and real-life examples given here could be used as course material to educate rehabilitation researchers.