Epidemiologic textbooks and methodological papers define multiple causal effects. These causal effects can differ substantially; yet, the causal effect of interest is rarely specified in published epidemiologic studies perhaps because their distinctions are underappreciated. Here, we provide an organizational schema that distinguishes causal effects based on six characteristics. We use simple numeric examples to demonstrate the variability across effects and show why specifying the causal effect is necessary for an accurate intervention interpretation even under the simplest scenarios. The objective of our schema was to illuminate the distinguishing characteristics of various causal effects and clarify their interpretation, thus guiding epidemiologists in choosing an appropriate causal effect to estimate.
From the aDepartment of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY; and bEpidemiology, Worldwide Safety & Regulatory, Pfizer Inc., New York, NY.
The authors report no conflict of interest.
Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). This content is not peer-reviewed or copy-edited; it is the sole responsibility of the author.
Editors' note: A commentary on this article appears on page 98.
Correspondence: Nicolle M. Gatto, WSR Epidemiology, Pfizer Inc., 235 East 42nd Street MS 219/9/1, New York, NY 10017. E-mail: firstname.lastname@example.org.