Primary Submission Category: Applications in Health and Biology
Overcoming an extreme positivity violation to distinguish the causal effects of surgery and anesthesia via a separable effects model
Authors: Caleb Miles, Amy Pitts, Caleb Ing, Ling Guo,
Presenting Author: Caleb Miles*
The U.S. Food and Drug Administration has cautioned that prenatal exposure to anesthetic drugs during the third trimester may have neurotoxic effects; however, there is limited clinical evidence available to substantiate this recommendation. One major scientific question of interest is whether such neurotoxic effects might be due to surgery, anesthesia, or both. Isolating the effects of these two exposures is challenging because they are observationally equivalent, thereby inducing an extreme positivity violation. To overcome this, we adopt the separable effects framework of Robins and Richardson (2010) with a novel perspective, shifting the focus from starting with a known intermediate variable of interest to starting with known separable components of treatment that are of interest. In particular, under this framework, surgery and anesthesia are the known separable components of the exposure, and we identify the effect of anesthesia (alone) by blocking effects through intermediate variables that are assumed to completely mediate the causal pathway from surgery to the outcome. We apply this approach to data from the nationwide Medicaid Analytic eXtract (MAX) from 1999 through 2013, which linked 16,778,281 deliveries to mothers enrolled in Medicaid during pregnancy. Furthermore, we assess the sensitivity of our results to violations of our key identification assumptions.