Invited commentary: interpreting associations between exposure biomarkers and pregnancy outcome

Abstract

Levels of exposure biomarkers vary among individuals because of differences in both environmental exposure and metabolism. However, the ultimate interest is in providing information about the impact of modifying environmental exposures through regulation or behavior change. Using these levels in studies of pregnancy outcomes, as nicely illustrated by the study of Kadhel et al. in this issue of the Journal (Am J Epidemiol. 2014;179(5):536-544), has the usual strength of being integrative across multiple pathways but may reflect reverse causality, in which the underlying disease alters biomarker levels or shared physiological determinants of the biomarker level and the health outcome. Specifically, biomarkers may vary because of spatial differences in exposure, behavioral differences affecting exposure, and metabolic differences across members of the study population. Proper interpretation of such studies calls for a clearer understanding the sources of variation in exposure to more fully consider confounding and reverse causality due to metabolic differences within the study population.

Keywords: biomarkers; fetal growth; pregnancy; preterm birth; reverse causality.