Publication: Educational Note: Paradoxical collider effect in the analysis of non-communicable disease epidemiological data: a reproducible illustration and web application.
dc.contributor.author | Luque-Fernandez, Miguel Angel | |
dc.contributor.author | Schomaker, Michael | |
dc.contributor.author | Redondo-Sanchez, Daniel | |
dc.contributor.author | Jose Sanchez Perez, Maria | |
dc.contributor.author | Vaidya, Anand | |
dc.contributor.author | Schnitzer, Mireille E | |
dc.date.accessioned | 2023-01-25T10:26:28Z | |
dc.date.available | 2023-01-25T10:26:28Z | |
dc.date.issued | 2019 | |
dc.description.abstract | Classical epidemiology has focused on the control of confounding, but it is only recently that epidemiologists have started to focus on the bias produced by colliders. A collider for a certain pair of variables (e.g. an outcome Y and an exposure A) is a third variable (C) that is caused by both. In a directed acyclic graph (DAG), a collider is the variable in the middle of an inverted fork (i.e. the variable C in A → C ← Y). Controlling for, or conditioning an analysis on a collider (i.e. through stratification or regression) can introduce a spurious association between its causes. This potentially explains many paradoxical findings in the medical literature, where established risk factors for a particular outcome appear protective. We use an example from non-communicable disease epidemiology to contextualize and explain the effect of conditioning on a collider. We generate a dataset with 1000 observations, and run Monte-Carlo simulations to estimate the effect of 24-h dietary sodium intake on systolic blood pressure, controlling for age, which acts as a confounder, and 24-h urinary protein excretion, which acts as a collider. We illustrate how adding a collider to a regression model introduces bias. Thus, to prevent paradoxical associations, epidemiologists estimating causal effects should be wary of conditioning on colliders. We provide R code in easy-to-read boxes throughout the manuscript, and a GitHub repository [https://github.com/migariane/ColliderApp] for the reader to reproduce our example. We also provide an educational web application allowing real-time interaction to visualize the paradoxical effect of conditioning on a collider [http://watzilei.com/shiny/collider/]. | |
dc.identifier.doi | 10.1093/ije/dyy275 | |
dc.identifier.essn | 1464-3685 | |
dc.identifier.pmc | PMC6469301 | |
dc.identifier.pmid | 30561628 | |
dc.identifier.pubmedURL | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6469301/pdf | |
dc.identifier.unpaywallURL | https://academic.oup.com/ije/article-pdf/48/2/640/31517845/dyy275.pdf | |
dc.identifier.uri | http://hdl.handle.net/10668/13323 | |
dc.issue.number | 2 | |
dc.journal.title | International journal of epidemiology | |
dc.journal.titleabbreviation | Int J Epidemiol | |
dc.language.iso | en | |
dc.organization | Escuela Andaluza de Salud Pública-EASP | |
dc.page.number | 640-653 | |
dc.pubmedtype | Journal Article | |
dc.pubmedtype | Research Support, N.I.H., Extramural | |
dc.pubmedtype | Research Support, Non-U.S. Gov't | |
dc.rights.accessRights | open access | |
dc.subject | Epidemiological methods | |
dc.subject | causality | |
dc.subject | non-communicable disease epidemiology | |
dc.subject.mesh | Bias | |
dc.subject.mesh | Causality | |
dc.subject.mesh | Confounding Factors, Epidemiologic | |
dc.subject.mesh | Epidemiologic Methods | |
dc.subject.mesh | Humans | |
dc.subject.mesh | Logistic Models | |
dc.subject.mesh | Noncommunicable Diseases | |
dc.title | Educational Note: Paradoxical collider effect in the analysis of non-communicable disease epidemiological data: a reproducible illustration and web application. | |
dc.type | research article | |
dc.type.hasVersion | VoR | |
dc.volume.number | 48 | |
dspace.entity.type | Publication |