Study Designs for Biobank-Based Epidemiologic Research on Chronic Diseases

Abstract

A review is given on design options to be considered in epidemiologic studies on cancers or other chronic diseases in relation to risk factors, the measurement of which is based on stored specimens in large biobanks. The two major choices for valid and cost-efficient sampling of risk factor data from large biobank cohorts are provided by the nested case–control design, and the case–cohort design. The main features of both designs are outlined and their relative merits are compared. Special issues such as matching, stratification, and statistical analysis are also briefly discussed. It is concluded that the nested case–control design is ­better suited for studies involving biomarkers that can be influenced by analytic batch, long-term storage, and freeze-thaw cycles. The case–cohort design is useful, especially when several outcomes are of interest, given that the measurements on stored materials remain sufficiently stable during the study.