Cross-sectional studies and causal inference - it's complicated.

In this issue of the Journal, Savitz and Wellenius (Am J Epidemiol. XXXX;XXX(XX):XXXX-XXXX) discuss the contribution of cross-sectional studies to causal inference when the data are used to address etiologic research questions. We elaborate on their thoughts with a discussion of the conditions needed for addressing etiology with the cross-sectional design, using a modern causal inference lens.