Home > Research > Publications & Outputs > To Adjust or Not to Adjust?

Links

Text available via DOI:

View graph of relations

To Adjust or Not to Adjust?: When a "Confounder" Is Only Measured After Exposure

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

To Adjust or Not to Adjust? When a "Confounder" Is Only Measured After Exposure. / Groenwold, R.H.H.; Palmer, T.M.; Tilling, K.

In: Epidemiology, Vol. 32, No. 2, 31.03.2021, p. 194-201.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Author

Groenwold, R.H.H. ; Palmer, T.M. ; Tilling, K. / To Adjust or Not to Adjust? When a "Confounder" Is Only Measured After Exposure. In: Epidemiology. 2021 ; Vol. 32, No. 2. pp. 194-201.

Bibtex

@article{9afd13c35daf4718a96d39a6337afbad,
title = "To Adjust or Not to Adjust?: When a {"}Confounder{"} Is Only Measured After Exposure",
abstract = "Advice regarding the analysis of observational studies of exposure effects usually is against adjustment for factors that occur after the exposure, as they may be caused by the exposure (or mediate the effect of exposure on outcome), so potentially leading to collider stratification bias. However, such factors could also be caused by unmeasured confounding factors, in which case adjusting for them will also remove some of the bias due to confounding. We derive expressions for collider stratification bias when conditioning and confounding bias when not conditioning on the mediator, in the presence of unmeasured confounding (assuming that all associations are linear and there are no interactions). Using simulations, we show that generally neither the conditioned nor the unconditioned estimate is unbiased, and the trade-off between them depends on the magnitude of the effect of the exposure that is mediated relative to the effect of the unmeasured confounders and their relations with the mediator. We illustrate the use of the bias expressions via three examples: neuroticism and mortality (adjusting for the mediator appears the least biased option), glycated hemoglobin levels and systolic blood pressure (adjusting gives smaller bias), and literacy in primary school pupils (not adjusting gives smaller bias). Our formulae and simulations can inform quantitative bias analysis as well as analysis strategies for observational studies in which there is a potential for unmeasured confounding. ",
keywords = "Bias, Confounding, Mediation, Observational study",
author = "R.H.H. Groenwold and T.M. Palmer and K. Tilling",
year = "2021",
month = mar,
day = "31",
doi = "10.1097/EDE.0000000000001312",
language = "English",
volume = "32",
pages = "194--201",
journal = "Epidemiology",
issn = "1531-5487",
publisher = "NLM (Medline)",
number = "2",

}

RIS

TY - JOUR

T1 - To Adjust or Not to Adjust?

T2 - When a "Confounder" Is Only Measured After Exposure

AU - Groenwold, R.H.H.

AU - Palmer, T.M.

AU - Tilling, K.

PY - 2021/3/31

Y1 - 2021/3/31

N2 - Advice regarding the analysis of observational studies of exposure effects usually is against adjustment for factors that occur after the exposure, as they may be caused by the exposure (or mediate the effect of exposure on outcome), so potentially leading to collider stratification bias. However, such factors could also be caused by unmeasured confounding factors, in which case adjusting for them will also remove some of the bias due to confounding. We derive expressions for collider stratification bias when conditioning and confounding bias when not conditioning on the mediator, in the presence of unmeasured confounding (assuming that all associations are linear and there are no interactions). Using simulations, we show that generally neither the conditioned nor the unconditioned estimate is unbiased, and the trade-off between them depends on the magnitude of the effect of the exposure that is mediated relative to the effect of the unmeasured confounders and their relations with the mediator. We illustrate the use of the bias expressions via three examples: neuroticism and mortality (adjusting for the mediator appears the least biased option), glycated hemoglobin levels and systolic blood pressure (adjusting gives smaller bias), and literacy in primary school pupils (not adjusting gives smaller bias). Our formulae and simulations can inform quantitative bias analysis as well as analysis strategies for observational studies in which there is a potential for unmeasured confounding.

AB - Advice regarding the analysis of observational studies of exposure effects usually is against adjustment for factors that occur after the exposure, as they may be caused by the exposure (or mediate the effect of exposure on outcome), so potentially leading to collider stratification bias. However, such factors could also be caused by unmeasured confounding factors, in which case adjusting for them will also remove some of the bias due to confounding. We derive expressions for collider stratification bias when conditioning and confounding bias when not conditioning on the mediator, in the presence of unmeasured confounding (assuming that all associations are linear and there are no interactions). Using simulations, we show that generally neither the conditioned nor the unconditioned estimate is unbiased, and the trade-off between them depends on the magnitude of the effect of the exposure that is mediated relative to the effect of the unmeasured confounders and their relations with the mediator. We illustrate the use of the bias expressions via three examples: neuroticism and mortality (adjusting for the mediator appears the least biased option), glycated hemoglobin levels and systolic blood pressure (adjusting gives smaller bias), and literacy in primary school pupils (not adjusting gives smaller bias). Our formulae and simulations can inform quantitative bias analysis as well as analysis strategies for observational studies in which there is a potential for unmeasured confounding.

KW - Bias

KW - Confounding

KW - Mediation

KW - Observational study

U2 - 10.1097/EDE.0000000000001312

DO - 10.1097/EDE.0000000000001312

M3 - Journal article

VL - 32

SP - 194

EP - 201

JO - Epidemiology

JF - Epidemiology

SN - 1531-5487

IS - 2

ER -