Lesser Known Facts about Nested Case-Control Designs

Nested case-control designs (or equivalently, incidence density sampling designs) are are a common approach for reducing the costs the costs of exposure assessment in a prospective epidemiologic study. Exposure data are obtained from all cases and a pre-specified number of controls randomly chosen at each failure time from all subjects who had entered the study but had not failed or left it yet [1]. The subjects selected across failure times are confusingly referred as ‘controls’. However, unlike traditional matched case-control designs, these ‘controls’ may include failures and some controls may be selected multiple times across different failure times. There can be a large redundancy between the controls.