MethodsA Graphical Description of Partial ExchangeabilitySarvet, Aaron L.a; Wanis, Kerollos Nashata,b; Stensrud, Mats J.a,c; Hernán, Miguel A.a,d,eAuthor Information From the aDepartment of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA bDivision of General Surgery, Department of Surgery, Western University, London, Ontario, Canada cDepartment of Biostatistics, University of Oslo, Oslo, Norway dDepartment of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA eHarvard-MIT Division of Health Sciences and Technology, Cambridge, MA. Submitted May 16, 2019; accepted January 20, 2020. A.L.S. was supported by grant T32 MH017119 from the National Institute of Mental Health. M.A.H. was supported by grant R37 AI102634 from the National Institute of Health. M.J.S. was supported by The Research Council of Norway, grant NFR239956/F2 Analyzing clinical health registries. The authors report no conflicts of interest. Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). The authors Sarvet and Wanis contributed to this study equally. Correspondence: Kerollos Nashat Wanis, Department of Surgery, Western University. London Health Sciences Centre, Rm. C8-114, London, ON N6A 5A5, Canada. E-mail: email@example.com. Epidemiology: May 2020 - Volume 31 - Issue 3 - p 365-368 doi: 10.1097/EDE.0000000000001165 Buy SDC Metrics Abstract Partial exchangeability is sufficient for the identification of some causal effects of interest. Here we review the use of common graphical tools and the sufficient component cause model in the context of partial exchangeability. We illustrate the utility of single world intervention graphs (SWIGs) in depicting partial exchangeability and provide an illustrative example of when partial exchangeability might hold in the absence of complete exchangeability. Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.