Causal mediation analysis with observational data: Considerations and illustration examining mechanisms linking neighborhood poverty to adolescent substance use

Kara Rudolph, Dana E. Goin, Diana Paksarian, Rebecca Crowder, Kathleen R. Merikangas, Elizabeth A. Stuart

Research output: Contribution to journalArticle

Abstract

Understanding the mediation mechanisms by which an exposure or intervention affects an outcome can provide a look into what has been called a "black box" of many epidemiologic associations, thereby providing further evidence of a relationship and possible points of intervention. Rapid methodologic developments in mediation analyses mean that there are a growing number of approaches for researchers to consider, each with its own set of assumptions, advantages, and disadvantages. This has understandably resulted in some confusion among applied researchers. Here, we provide a brief overview of the mediation methods available and discuss points for consideration when choosing a method. We provide an in-depth explication of 2 of the many potential estimators for illustrative purposes: the Baron and Kenny mediation approach, because it is the most commonly used, and a recently developed approach for estimating stochastic direct and indirect effects, because it relies on far fewer assumptions. We illustrate the decision process and analytical procedure by estimating potential school- and peer-based mechanisms linking neighborhood poverty to adolescent substance use in the National Comorbidity Survey Adolescent Supplement.

Original languageEnglish (US)
Pages (from-to)598-608
Number of pages11
JournalAmerican journal of epidemiology
Volume188
Issue number3
DOIs
StatePublished - Jan 1 2019

Fingerprint

Poverty
Research Personnel
Comorbidity

Keywords

  • adolescent
  • mediation
  • neighborhood
  • stochastic intervention
  • substance use

ASJC Scopus subject areas

  • Epidemiology

Cite this

Causal mediation analysis with observational data : Considerations and illustration examining mechanisms linking neighborhood poverty to adolescent substance use. / Rudolph, Kara; Goin, Dana E.; Paksarian, Diana; Crowder, Rebecca; Merikangas, Kathleen R.; Stuart, Elizabeth A.

In: American journal of epidemiology, Vol. 188, No. 3, 01.01.2019, p. 598-608.

Research output: Contribution to journalArticle

Rudolph, Kara ; Goin, Dana E. ; Paksarian, Diana ; Crowder, Rebecca ; Merikangas, Kathleen R. ; Stuart, Elizabeth A. / Causal mediation analysis with observational data : Considerations and illustration examining mechanisms linking neighborhood poverty to adolescent substance use. In: American journal of epidemiology. 2019 ; Vol. 188, No. 3. pp. 598-608.
@article{1ed79e13f8e049c6944bc5a41d35e41f,
title = "Causal mediation analysis with observational data: Considerations and illustration examining mechanisms linking neighborhood poverty to adolescent substance use",
abstract = "Understanding the mediation mechanisms by which an exposure or intervention affects an outcome can provide a look into what has been called a {"}black box{"} of many epidemiologic associations, thereby providing further evidence of a relationship and possible points of intervention. Rapid methodologic developments in mediation analyses mean that there are a growing number of approaches for researchers to consider, each with its own set of assumptions, advantages, and disadvantages. This has understandably resulted in some confusion among applied researchers. Here, we provide a brief overview of the mediation methods available and discuss points for consideration when choosing a method. We provide an in-depth explication of 2 of the many potential estimators for illustrative purposes: the Baron and Kenny mediation approach, because it is the most commonly used, and a recently developed approach for estimating stochastic direct and indirect effects, because it relies on far fewer assumptions. We illustrate the decision process and analytical procedure by estimating potential school- and peer-based mechanisms linking neighborhood poverty to adolescent substance use in the National Comorbidity Survey Adolescent Supplement.",
keywords = "adolescent, mediation, neighborhood, stochastic intervention, substance use",
author = "Kara Rudolph and Goin, {Dana E.} and Diana Paksarian and Rebecca Crowder and Merikangas, {Kathleen R.} and Stuart, {Elizabeth A.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1093/aje/kwy248",
language = "English (US)",
volume = "188",
pages = "598--608",
journal = "American Journal of Epidemiology",
issn = "0002-9262",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - Causal mediation analysis with observational data

T2 - Considerations and illustration examining mechanisms linking neighborhood poverty to adolescent substance use

AU - Rudolph, Kara

AU - Goin, Dana E.

AU - Paksarian, Diana

AU - Crowder, Rebecca

AU - Merikangas, Kathleen R.

AU - Stuart, Elizabeth A.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Understanding the mediation mechanisms by which an exposure or intervention affects an outcome can provide a look into what has been called a "black box" of many epidemiologic associations, thereby providing further evidence of a relationship and possible points of intervention. Rapid methodologic developments in mediation analyses mean that there are a growing number of approaches for researchers to consider, each with its own set of assumptions, advantages, and disadvantages. This has understandably resulted in some confusion among applied researchers. Here, we provide a brief overview of the mediation methods available and discuss points for consideration when choosing a method. We provide an in-depth explication of 2 of the many potential estimators for illustrative purposes: the Baron and Kenny mediation approach, because it is the most commonly used, and a recently developed approach for estimating stochastic direct and indirect effects, because it relies on far fewer assumptions. We illustrate the decision process and analytical procedure by estimating potential school- and peer-based mechanisms linking neighborhood poverty to adolescent substance use in the National Comorbidity Survey Adolescent Supplement.

AB - Understanding the mediation mechanisms by which an exposure or intervention affects an outcome can provide a look into what has been called a "black box" of many epidemiologic associations, thereby providing further evidence of a relationship and possible points of intervention. Rapid methodologic developments in mediation analyses mean that there are a growing number of approaches for researchers to consider, each with its own set of assumptions, advantages, and disadvantages. This has understandably resulted in some confusion among applied researchers. Here, we provide a brief overview of the mediation methods available and discuss points for consideration when choosing a method. We provide an in-depth explication of 2 of the many potential estimators for illustrative purposes: the Baron and Kenny mediation approach, because it is the most commonly used, and a recently developed approach for estimating stochastic direct and indirect effects, because it relies on far fewer assumptions. We illustrate the decision process and analytical procedure by estimating potential school- and peer-based mechanisms linking neighborhood poverty to adolescent substance use in the National Comorbidity Survey Adolescent Supplement.

KW - adolescent

KW - mediation

KW - neighborhood

KW - stochastic intervention

KW - substance use

UR - http://www.scopus.com/inward/record.url?scp=85066823290&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066823290&partnerID=8YFLogxK

U2 - 10.1093/aje/kwy248

DO - 10.1093/aje/kwy248

M3 - Article

C2 - 30561500

AN - SCOPUS:85066823290

VL - 188

SP - 598

EP - 608

JO - American Journal of Epidemiology

JF - American Journal of Epidemiology

SN - 0002-9262

IS - 3

ER -