The causal inference framework and related methods have emerged as vital within epidemiology. This framework and associated analytic approaches facilitate the conduct of valid science using observational data. These approaches have helped catalyze knowledge development using existing data and also have addressed questions for which randomized controlled trials are neither feasible nor ethical. The study of normal childbearing processes and women who are medically low risk may benefit from more direct and deliberate engagement with the process of inferring causes and the use of methods appropriate for this undertaking. This article is the second in a series of 3 that review scientific challenges encountered in researching pregnancy, labor, and birth and approaches for addressing them. This article introduces 2 methods for causal inference (g-computation and instrumental variable analysis) to an audience of clinician-scientists, including references with further details. The causal inference framework and associated methods hold promise for generating strong, broadly representative, and actionable science to improve the outcomes of women who are medically low risk and their children.
Keywords: assumptions; causal inference framework; g-computation; instrumental variables; midwifery science; observational studies; physiologic childbearing science; secondary data analysis.
© 2018 by the American College of Nurse-Midwives.