Cross-direct effects in settings with two mediators

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Cross-direct effects in settings with two mediators. / Gabriel, Erin E.; Sjolander, Arvid; Follmann, Dean; Sachs, Michael C.

In: Biostatistics, Vol. 24, No. 4, 2023, p. 1017–1030.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Gabriel, EE, Sjolander, A, Follmann, D & Sachs, MC 2023, 'Cross-direct effects in settings with two mediators', Biostatistics, vol. 24, no. 4, pp. 1017–1030. https://doi.org/10.1093/biostatistics/kxac037

APA

Gabriel, E. E., Sjolander, A., Follmann, D., & Sachs, M. C. (2023). Cross-direct effects in settings with two mediators. Biostatistics, 24(4), 1017–1030. https://doi.org/10.1093/biostatistics/kxac037

Vancouver

Gabriel EE, Sjolander A, Follmann D, Sachs MC. Cross-direct effects in settings with two mediators. Biostatistics. 2023;24(4):1017–1030. https://doi.org/10.1093/biostatistics/kxac037

Author

Gabriel, Erin E. ; Sjolander, Arvid ; Follmann, Dean ; Sachs, Michael C. / Cross-direct effects in settings with two mediators. In: Biostatistics. 2023 ; Vol. 24, No. 4. pp. 1017–1030.

Bibtex

@article{74295035839f4f9faed631e01c764a13,
title = "Cross-direct effects in settings with two mediators",
abstract = "When multiple mediators are present, there are additional effects that may be of interest beyond the well-known natural (NDE) and controlled direct effects (CDE). These effects cross the type of control on the mediators, setting one to a constant level and one to its natural level, which differs across subjects. We introduce five such estimands for the cross-CDE and -NDE when two mediators are measured. We consider both the scenario where one mediator is influenced by the other, referred to as sequential mediators, and the scenario where the mediators do not influence each other. Such estimands may be of interest in immunology, as we discuss in relation to measured immunological responses to SARS-CoV-2 vaccination. We provide identifying expressions for the estimands in observational settings where there is no residual confounding, and where intervention, outcome, and mediators are of arbitrary type. We further provide tight symbolic bounds for the estimands in randomized settings where there may be residual confounding of the outcome and mediator relationship and all measured variables are binary.",
keywords = "Causal pathways, Multiple mediation, Symbolic bounds, BOUNDS",
author = "Gabriel, {Erin E.} and Arvid Sjolander and Dean Follmann and Sachs, {Michael C.}",
year = "2023",
doi = "10.1093/biostatistics/kxac037",
language = "English",
volume = "24",
pages = "1017–1030",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "4",

}

RIS

TY - JOUR

T1 - Cross-direct effects in settings with two mediators

AU - Gabriel, Erin E.

AU - Sjolander, Arvid

AU - Follmann, Dean

AU - Sachs, Michael C.

PY - 2023

Y1 - 2023

N2 - When multiple mediators are present, there are additional effects that may be of interest beyond the well-known natural (NDE) and controlled direct effects (CDE). These effects cross the type of control on the mediators, setting one to a constant level and one to its natural level, which differs across subjects. We introduce five such estimands for the cross-CDE and -NDE when two mediators are measured. We consider both the scenario where one mediator is influenced by the other, referred to as sequential mediators, and the scenario where the mediators do not influence each other. Such estimands may be of interest in immunology, as we discuss in relation to measured immunological responses to SARS-CoV-2 vaccination. We provide identifying expressions for the estimands in observational settings where there is no residual confounding, and where intervention, outcome, and mediators are of arbitrary type. We further provide tight symbolic bounds for the estimands in randomized settings where there may be residual confounding of the outcome and mediator relationship and all measured variables are binary.

AB - When multiple mediators are present, there are additional effects that may be of interest beyond the well-known natural (NDE) and controlled direct effects (CDE). These effects cross the type of control on the mediators, setting one to a constant level and one to its natural level, which differs across subjects. We introduce five such estimands for the cross-CDE and -NDE when two mediators are measured. We consider both the scenario where one mediator is influenced by the other, referred to as sequential mediators, and the scenario where the mediators do not influence each other. Such estimands may be of interest in immunology, as we discuss in relation to measured immunological responses to SARS-CoV-2 vaccination. We provide identifying expressions for the estimands in observational settings where there is no residual confounding, and where intervention, outcome, and mediators are of arbitrary type. We further provide tight symbolic bounds for the estimands in randomized settings where there may be residual confounding of the outcome and mediator relationship and all measured variables are binary.

KW - Causal pathways

KW - Multiple mediation

KW - Symbolic bounds

KW - BOUNDS

U2 - 10.1093/biostatistics/kxac037

DO - 10.1093/biostatistics/kxac037

M3 - Journal article

C2 - 36050911

VL - 24

SP - 1017

EP - 1030

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 4

ER -

ID: 318867484