The “exposure-based cross-sectional” study design: a novel observational study design applicable to rare exposures

Abstract

Current epidemiological studies are either inefficient or very expensive and time-consuming when the exposure of interest is very rare. The ‘exposure-based cross-sectional’ study is a new design that can overcome this problem. The ‘exposure-based cross-sectional’ study starts with exposed and unexposed groups. Then, these two groups are compared to determine what proportion of each group have the disease and what proportion do not. It is as if we were conducting a reversed case–control study in which the positions of the disease and exposures are altered. Dissimilar to retrospective cohort studies, the ‘exposure-based cross-sectional’ study does not depend on the basic existing records. This study measures the disease ‘prevalence’ rather than the disease ‘incidence’. The ‘exposure-based cross-sectional’ study design was examined in several real-life epidemiological studies with binary and continuous outcomes. The ‘exposure-based cross-sectional’ study is an efficient, inexpensive, expeditious, and easy to conduct study design for rare exposures. It can be performed for both binary and continuous outcomes.