Bias attenuation results for dichotomization of a continuous confounder

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

Bias attenuation results for dichotomization of a continuous confounder. / Gabriel, Erin E.; Peña, Jose M.; Sjölander, Arvid.

I: Journal of Causal Inference, Bind 10, Nr. 1, 2022, s. 515-526.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Gabriel, EE, Peña, JM & Sjölander, A 2022, 'Bias attenuation results for dichotomization of a continuous confounder', Journal of Causal Inference, bind 10, nr. 1, s. 515-526. https://doi.org/10.1515/jci-2022-0047

APA

Gabriel, E. E., Peña, J. M., & Sjölander, A. (2022). Bias attenuation results for dichotomization of a continuous confounder. Journal of Causal Inference, 10(1), 515-526. https://doi.org/10.1515/jci-2022-0047

Vancouver

Gabriel EE, Peña JM, Sjölander A. Bias attenuation results for dichotomization of a continuous confounder. Journal of Causal Inference. 2022;10(1):515-526. https://doi.org/10.1515/jci-2022-0047

Author

Gabriel, Erin E. ; Peña, Jose M. ; Sjölander, Arvid. / Bias attenuation results for dichotomization of a continuous confounder. I: Journal of Causal Inference. 2022 ; Bind 10, Nr. 1. s. 515-526.

Bibtex

@article{e0d2294ddef94ea6ab0517e00c8f5a9d,

title = "Bias attenuation results for dichotomization of a continuous confounder",

abstract = "It is well-known that dichotomization can cause bias and loss of efficiency in estimation. One can easily construct examples where adjusting for a dichotomized confounder causes bias in causal estimation. There are additional examples in the literature where adjusting for a dichotomized confounder can be more biased than not adjusting at all. The message is clear, do not dichotomize. What is unclear is if there are scenarios where adjusting for the dichotomized confounder always leads to lower bias than not adjusting. We propose several sets of conditions that characterize scenarios where one should always adjust for the dichotomized confounder to reduce bias. We then highlight scenarios where the decision to adjust should be made more cautiously. To our knowledge, this is the first formal presentation of conditions that give information about when one should and potentially should not adjust for a dichotomized confounder.",

keywords = "bias, causal inference, dichotomized confounder",

author = "Gabriel, {Erin E.} and Pe{\~n}a, {Jose M.} and Arvid Sj{\"o}lander",

note = "Publisher Copyright: {\textcopyright} 2022 the author(s), published by De Gruyter.",

year = "2022",

doi = "10.1515/jci-2022-0047",

language = "English",

volume = "10",

pages = "515--526",

journal = "Journal of Causal Inference",

issn = "2193-3677",

publisher = "Walterde Gruyter GmbH",

number = "1",

}

RIS

TY - JOUR

T1 - Bias attenuation results for dichotomization of a continuous confounder

AU - Gabriel, Erin E.

AU - Peña, Jose M.

AU - Sjölander, Arvid

PY - 2022

Y1 - 2022

N2 - It is well-known that dichotomization can cause bias and loss of efficiency in estimation. One can easily construct examples where adjusting for a dichotomized confounder causes bias in causal estimation. There are additional examples in the literature where adjusting for a dichotomized confounder can be more biased than not adjusting at all. The message is clear, do not dichotomize. What is unclear is if there are scenarios where adjusting for the dichotomized confounder always leads to lower bias than not adjusting. We propose several sets of conditions that characterize scenarios where one should always adjust for the dichotomized confounder to reduce bias. We then highlight scenarios where the decision to adjust should be made more cautiously. To our knowledge, this is the first formal presentation of conditions that give information about when one should and potentially should not adjust for a dichotomized confounder.

AB - It is well-known that dichotomization can cause bias and loss of efficiency in estimation. One can easily construct examples where adjusting for a dichotomized confounder causes bias in causal estimation. There are additional examples in the literature where adjusting for a dichotomized confounder can be more biased than not adjusting at all. The message is clear, do not dichotomize. What is unclear is if there are scenarios where adjusting for the dichotomized confounder always leads to lower bias than not adjusting. We propose several sets of conditions that characterize scenarios where one should always adjust for the dichotomized confounder to reduce bias. We then highlight scenarios where the decision to adjust should be made more cautiously. To our knowledge, this is the first formal presentation of conditions that give information about when one should and potentially should not adjust for a dichotomized confounder.

KW - bias

KW - causal inference

KW - dichotomized confounder

U2 - 10.1515/jci-2022-0047

DO - 10.1515/jci-2022-0047

M3 - Journal article

AN - SCOPUS:85147151109

VL - 10

SP - 515

EP - 526

JO - Journal of Causal Inference

JF - Journal of Causal Inference

SN - 2193-3677

IS - 1

ER -

ID: 336824343