An Introduction to Negative Control and Proximal Causal Learning

Negative controls are auxiliary variables not causally associated with the treatment or outcome of interest. In this talk, we first introduced a formal negative control study design and summarize existing negative control methods for detection, reduction, and correction of unmeasured confounding bias. We then introduced the proximal causal learning framework, a generalization of negative controls, which offers an opportunity to learn about causal effects when exchangeability based on measured covariates fails by formally accounting for the covariate measurements as imperfect proxies of underlying confounding mechanisms.