Regression discontinuity design studies: a guide for health researchers
Sebastián Calónico, Neal Jawadekar, Katrina Kezios, Adina Zeki Al Hazzouri
Abstract
As randomized controlled trials are not always feasible, quasi-experimental methods, such as regression discontinuity design, can expand the scope of clinical investigations aimed at causal inference in observational settings. However, clinical researchers are likely to be less familiar with and have less training in quasi-experimental designs. This article focuses on implementation and provides a detailed checklist, glossary, and guided example for how to conduct an analysis of regression discontinuity design, with the aim to help clinical researchers read, conduct, and interpret regression discontinuity design and, overall, to encourage its wider adoption in clinical practice.