Strengthening an instrumental variable: applications to health services research and multicenter clinical trials

In the first part of the talk,  I will discuss strengthening a continuous instrumental variable (IV) in the design of a matched observational study. We study how strengthening an IV may shorten the partial identification bounds for the sample average treatment effect (SATE) in an IV-based matched cohort study. Unlike the usual effect ratio estimand, SATE does not depend on who is matched to whom in the design, although a strengthened-IV design has the potential to narrow its partial identification bounds. We applied the method to studying a triage system that directs mothers to hospitals with varying capabilities using excess travel time as an IV.  In the second part of the talk, I will introduce how to leverage a naturally strengthened, binary IV to assess the generalizability of IV-based estimates. Under a monotonicity assumption, a valid binary IV nonparametrically identifies complier average treatment effect, whose generalizability is often under debate. In many studies, there may exist multiple versions of a binary IV, for instance, different nudges to take the same treatment in different study sites in a clinical trial. I will introduce a novel nested IV assumption and study the identification of the average treatment effect among two latent subgroups: always-compliers and switchers.  We apply the proposed method to the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial and study the causal effect of colorectal cancer screening and its generalizability. 

Link to the paper: Full article: Manipulating an Instrumental Variable in an Observational Study of Premature Babies: Design, Bounds, and Inference and [2405.07102] Nested Instrumental Variables Design: Switcher Average Treatment Effect, Identification, Efficient Estimation and Generalizability