Regression Discontinuity Designs in Public Health Research

Regression discontinuity designs offer an internally valid approach for causal inference without need for randomization. Regression discontinuity designs can be implemented when an exposure is assigned at least in part based on a threshold rule: the party with > 50% of the votes wins in a two-party election; the HIV patient with a CD4 count below 500 cells is offered therapy; residents downstream of a point pollution source swim in contaminated water. Historically, regression discontinuity designs have been underutilized in public health and medical research. However, the last few years have seen burgeoning use of this method.

The presentation will review the theory behind regression discontinuity designs and their implementation, with a focus on examples in public health research.

Registration for this webinar is required. Visit the NCI website to learn more and to register. 

This session will be recorded and be available here.