Collider bias undermines our understanding of COVID-19 disease risk and severity

2020/11/12 by Gareth J Griffith, Tim Morris, Matthew Tudball +11 · 1 voice · 3 citations
Mathematics · Medicine · #COVID-19 Clinical Research Studies #COVID-19 epidemiological studies #SARS-CoV-2 and COVID-19 Research

paper · pdf · doi:10.1038/s41467-020-19478-2

openalex publication_date 2020/11/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

Numerous observational studies have attempted to identify risk factors for infection with SARS-CoV-2 and COVID-19 disease outcomes. Studies have used datasets sampled from patients admitted to hospital, people tested for active infection, or people who volunteered to participate. Here, we highlight the challenge of interpreting observational evidence from such non-representative samples. Collider bias can induce associations between two or more variables which affect the likelihood of an individual being sampled, distorting associations between these variables in the sample. Analysing UK Biobank data, compared to the wider cohort the participants tested for COVID-19 were highly selected for a range of genetic, behavioural, cardiovascular, demographic, and anthropometric traits. We discuss the mechanisms inducing these problems, and approaches that could help mitigate them. While collider bias should be explored in existing studies, the optimal way to mitigate the problem is to use appropriate sampling strategies at the study design stage.

Citations

• Negative Controls
• Endogenous Selection Bias: The Problem of Conditioning on a Collider Variable
• Does the Recruitment of Offline Households Increase the Sample Representativeness of Probability-Based Online Panels? Evidence From the German Internet Panel
• Collider scope: when selection bias can substantially influence observed associations
• Sensitivity analysis for inverse probability weighting estimators via the percentile bootstrap
• Thinking Clearly About Correlations and Causation: Graphical Causal Models for Observational Data
• To Explain or to Predict?
• Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China
• Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study
• Fundamental principles of epidemic spread highlight the immediate need for large-scale serological surveys to assess the stage of the SARS-CoV-2 epidemic
• Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal
• Causal Diagrams for Epidemiologic Research

Cited by

• Technology readiness levels for machine learning systems
• Comorbidities, multimorbidity and COVID-19
• Evaluating sources of bias in observational studies of angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker use during COVID-19: beyond confounding

Discussions

• For instance, if both our exposure and outcome predict participation (ie, collider bias), this can bias the relationship between these variables. (Image from doi.org/10.1038/s414...) [bsky, 0 points, 1 comments]