Epidemiologic Methodology for Biomarkers (including ROC curve)

Eunice Kennedy Shriver National Institute Of Child Health & Human Development

Investigators

Linked publications & trials

Abstract

Biomarkers are an integral part of epidemiological research, making substantial contributions to our understanding of disease pathways and processes. New and emerging biomarkers are essential to this continued understanding. As such, novel study designs that reduce cost and leverage statistical efficiency are also a major focus of Division researchers (Vernet et al, Epidemiology 2019; Van Domelen et al, Biostatistics 2019; Schildcrout et al, Am J Epidemiol 2019; Perkins et al, Epidemiology 2019; Valdes Salgado, Agric Environ Med 2019). These methods reduce cost and enhance efficiency by pooling samples, collapsing calibration information prior to analysis, utilizing efficient modeling strategies, appropriately handling data below the limit of detection (Kim et al, Statistics in Biosciences 2019), or by measuring biomarkers in a principally designed subset of the full cohort, guided by the outcome of interest. Division researchers continue to adapt these efficient designs and develop analytic methods to reduce various forms of bias (Vernet et al, Epidemiology 2019; Van Domelen et al, Biostatistics 2019, Kim et al, Statistics in Biosciences 2019), measurement error (Perkins et al, Epidemiology 2019) and adapt outcome dependent sampling to a broad spectrum of study designs (Schildcrout et al, Am J Epidemiol 2019).