Assessing the total effect of time-varying predictors in prevention research

Bethany Cara Bray, Daniel Almirall, Rick S. Zimmerman, Donald Lynam, Susan A. Murphy

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Observational data are often used to address prevention questions such as, "If alcohol initiation could be delayed, would that in turn cause a delay in marijuana initiation?" This question is concerned with the total causal effect of the timing of alcohol initiation on the timing of marijuana initiation. Unfortunately, when observational data are used to address a question such as the above, alternative explanations for the observed relationship between the predictor, here timing of alcohol initiation, and the response abound. These alternative explanations are due to the presence of confounders. Adjusting for confounders when using observational data is a particularly challenging problem when the predictor and confounders are time-varying. When time-varying confounders are present, the standard method of adjusting for confounders may fail to reduce bias and indeed can increase bias. In this paper, an intuitive and accessible graphical approach is used to illustrate how the standard method of controlling for confounders may result in biased total causal effect estimates. The graphical approach also provides an intuitive justification for an alternate method proposed by James Robins [Robins, J. M. (1998). 1997 Proceedings of the American Statistical Association, section on Bayesian statistical science (pp. 1-10). Retrieved from http://www.biostat.harvard.edu/robins/research.html; Robins, J. M., Hernán, M., & Brumback, B. (2000). Epidemiology, 11(5), 550-560]. The above two methods are illustrated by addressing the motivating question. Implications for prevention researchers who wish to estimate total causal effects using longitudinal observational data are discussed.

Original languageEnglish (US)
Pages (from-to)1-17
Number of pages17
JournalPrevention Science
Volume7
Issue number1
DOIs
StatePublished - Mar 1 2006

Fingerprint

Songbirds
Alcohols
Cannabis
Research
Epidemiology
Research Personnel

All Science Journal Classification (ASJC) codes

  • Public Health, Environmental and Occupational Health

Cite this

Bray, B. C., Almirall, D., Zimmerman, R. S., Lynam, D., & Murphy, S. A. (2006). Assessing the total effect of time-varying predictors in prevention research. Prevention Science, 7(1), 1-17. https://doi.org/10.1007/s11121-005-0023-0
Bray, Bethany Cara ; Almirall, Daniel ; Zimmerman, Rick S. ; Lynam, Donald ; Murphy, Susan A. / Assessing the total effect of time-varying predictors in prevention research. In: Prevention Science. 2006 ; Vol. 7, No. 1. pp. 1-17.
@article{269876c8a1f447708378d52d36f72c8f,
title = "Assessing the total effect of time-varying predictors in prevention research",
abstract = "Observational data are often used to address prevention questions such as, {"}If alcohol initiation could be delayed, would that in turn cause a delay in marijuana initiation?{"} This question is concerned with the total causal effect of the timing of alcohol initiation on the timing of marijuana initiation. Unfortunately, when observational data are used to address a question such as the above, alternative explanations for the observed relationship between the predictor, here timing of alcohol initiation, and the response abound. These alternative explanations are due to the presence of confounders. Adjusting for confounders when using observational data is a particularly challenging problem when the predictor and confounders are time-varying. When time-varying confounders are present, the standard method of adjusting for confounders may fail to reduce bias and indeed can increase bias. In this paper, an intuitive and accessible graphical approach is used to illustrate how the standard method of controlling for confounders may result in biased total causal effect estimates. The graphical approach also provides an intuitive justification for an alternate method proposed by James Robins [Robins, J. M. (1998). 1997 Proceedings of the American Statistical Association, section on Bayesian statistical science (pp. 1-10). Retrieved from http://www.biostat.harvard.edu/robins/research.html; Robins, J. M., Hern{\'a}n, M., & Brumback, B. (2000). Epidemiology, 11(5), 550-560]. The above two methods are illustrated by addressing the motivating question. Implications for prevention researchers who wish to estimate total causal effects using longitudinal observational data are discussed.",
author = "Bray, {Bethany Cara} and Daniel Almirall and Zimmerman, {Rick S.} and Donald Lynam and Murphy, {Susan A.}",
year = "2006",
month = "3",
day = "1",
doi = "10.1007/s11121-005-0023-0",
language = "English (US)",
volume = "7",
pages = "1--17",
journal = "Prevention Science",
issn = "1389-4986",
publisher = "Springer New York",
number = "1",

}

Bray, BC, Almirall, D, Zimmerman, RS, Lynam, D & Murphy, SA 2006, 'Assessing the total effect of time-varying predictors in prevention research', Prevention Science, vol. 7, no. 1, pp. 1-17. https://doi.org/10.1007/s11121-005-0023-0

Assessing the total effect of time-varying predictors in prevention research. / Bray, Bethany Cara; Almirall, Daniel; Zimmerman, Rick S.; Lynam, Donald; Murphy, Susan A.

In: Prevention Science, Vol. 7, No. 1, 01.03.2006, p. 1-17.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Assessing the total effect of time-varying predictors in prevention research

AU - Bray, Bethany Cara

AU - Almirall, Daniel

AU - Zimmerman, Rick S.

AU - Lynam, Donald

AU - Murphy, Susan A.

PY - 2006/3/1

Y1 - 2006/3/1

N2 - Observational data are often used to address prevention questions such as, "If alcohol initiation could be delayed, would that in turn cause a delay in marijuana initiation?" This question is concerned with the total causal effect of the timing of alcohol initiation on the timing of marijuana initiation. Unfortunately, when observational data are used to address a question such as the above, alternative explanations for the observed relationship between the predictor, here timing of alcohol initiation, and the response abound. These alternative explanations are due to the presence of confounders. Adjusting for confounders when using observational data is a particularly challenging problem when the predictor and confounders are time-varying. When time-varying confounders are present, the standard method of adjusting for confounders may fail to reduce bias and indeed can increase bias. In this paper, an intuitive and accessible graphical approach is used to illustrate how the standard method of controlling for confounders may result in biased total causal effect estimates. The graphical approach also provides an intuitive justification for an alternate method proposed by James Robins [Robins, J. M. (1998). 1997 Proceedings of the American Statistical Association, section on Bayesian statistical science (pp. 1-10). Retrieved from http://www.biostat.harvard.edu/robins/research.html; Robins, J. M., Hernán, M., & Brumback, B. (2000). Epidemiology, 11(5), 550-560]. The above two methods are illustrated by addressing the motivating question. Implications for prevention researchers who wish to estimate total causal effects using longitudinal observational data are discussed.

AB - Observational data are often used to address prevention questions such as, "If alcohol initiation could be delayed, would that in turn cause a delay in marijuana initiation?" This question is concerned with the total causal effect of the timing of alcohol initiation on the timing of marijuana initiation. Unfortunately, when observational data are used to address a question such as the above, alternative explanations for the observed relationship between the predictor, here timing of alcohol initiation, and the response abound. These alternative explanations are due to the presence of confounders. Adjusting for confounders when using observational data is a particularly challenging problem when the predictor and confounders are time-varying. When time-varying confounders are present, the standard method of adjusting for confounders may fail to reduce bias and indeed can increase bias. In this paper, an intuitive and accessible graphical approach is used to illustrate how the standard method of controlling for confounders may result in biased total causal effect estimates. The graphical approach also provides an intuitive justification for an alternate method proposed by James Robins [Robins, J. M. (1998). 1997 Proceedings of the American Statistical Association, section on Bayesian statistical science (pp. 1-10). Retrieved from http://www.biostat.harvard.edu/robins/research.html; Robins, J. M., Hernán, M., & Brumback, B. (2000). Epidemiology, 11(5), 550-560]. The above two methods are illustrated by addressing the motivating question. Implications for prevention researchers who wish to estimate total causal effects using longitudinal observational data are discussed.

UR - http://www.scopus.com/inward/record.url?scp=33646480615&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646480615&partnerID=8YFLogxK

U2 - 10.1007/s11121-005-0023-0

DO - 10.1007/s11121-005-0023-0

M3 - Article

VL - 7

SP - 1

EP - 17

JO - Prevention Science

JF - Prevention Science

SN - 1389-4986

IS - 1

ER -