Seminar Series: Jay Kaufman, PhD

How to Break Something by Fixing It

Jay Kaufman, PhDKaufman_headshot.jpg

Professor
Department of Epidemiology, Biostatistics and Occupational Health
McGill University

Abstract/Keywords:

One of the interesting developments in epidemiologic methods over the last decade is the increasing infusion of quasi-experimental techniques from econometrics, such as diffs-in-diffs, regression discontinuity and instrumental variables. Epidemiology and economics are fields that each have their own rich and distinct traditions of data analysis and causal discovery, and so the breaking down of disciplinary silos allows for a comparison of these traditions and tools. I will describe some of the differences in worldview between these fields, and then illustrate with one concrete example, which is the handling of clustering, as both an analytic challenge (hierarchical data) and as a design strategy matching). Causal tools from one discipline reveal the limitations and dangers of the practices that characterize the other. I review the good, the bad and the ugly from fixed effects designs like exposure-discordant siblings, and how insights from both epidemiology and economics balance the pros and cons of this method in health research.

Short Biography:

Jay S. Kaufman is Professor and Graduate Program Director at in the Department of Epidemiology, Biostatistics and Occupational Health at McGill University (Montreal, Quebec).  He also holds current appointments at the University of Chile, The University of Michigan and The University of North Carolina.  Dr. Kaufman earned a doctorate in epidemiologic science from the University of Michigan  (1995). After a post-doctoral position at Loyola Stritch School of Medicine (Chicago, IL) from 1995-1997, he was Medical Epidemiologist at Carolinas Medical Center (Charlotte, NC) from 1997 to 1999. From 1999 through 2008 he held a positions as Assistant and Associate Professor at the University of North Carolina School of Public Health at Chapel Hill and as Faculty Fellow of the Carolina Population Center before beginning his current position at McGill, where he also held a Canada Research Chair in Health Disparities (2010-2017).  Dr. Kaufman's work focuses on social epidemiology, analytic methodology, causal inference and on a variety of health outcomes including perinatal outcomes and cardiovascular, psychiatric and infectious diseases.  He is an editor at the journal “Epidemiology” and co-editor of the textbook “Methods in Social Epidemiology” (2nd Edition, 2017).  He served as President of the Society for Epidemiologic Research in 2020-2021.