A Structural Approach to Selection Bias

TLDR

Selection bias in epidemiology encompasses various biases that arise from conditioning on a common effect of two variables, a structure also shared with other biases such as adjustment for variables affected by prior exposure. The authors argue that selection bias across different study designs originates from conditioning on a common effect of two variables, one being exposure or its cause and the other being outcome or its cause. They illustrate selection bias in case‑control and cohort studies and propose a structural classification that distinguishes biases from conditioning on common effects versus biases from common causes of exposure and outcome. This classification provides a unified approach to adjust for selection bias.

Abstract

The term "selection bias" encompasses various biases in epidemiology. We describe examples of selection bias in case-control studies (eg, inappropriate selection of controls) and cohort studies (eg, informative censoring). We argue that the causal structure underlying the bias in each example is essentially the same: conditioning on a common effect of 2 variables, one of which is either exposure or a cause of exposure and the other is either the outcome or a cause of the outcome. This structure is shared by other biases (eg, adjustment for variables affected by prior exposure). A structural classification of bias distinguishes between biases resulting from conditioning on common effects ("selection bias") and those resulting from the existence of common causes of exposure and outcome ("confounding"). This classification also leads to a unified approach to adjust for selection bias.