A Bayesian approach to the g-formula

Alexander P. Keil, Eric J. Daza, Stephanie M. Engel, Jessie P Buckley, Jessie K. Edwards

Research output: Contribution to journalArticle

Abstract

Epidemiologists often wish to estimate quantities that are easy to communicate and correspond to the results of realistic public health interventions. Methods from causal inference can answer these questions. We adopt the language of potential outcomes under Rubin’s original Bayesian framework and show that the parametric g-formula is easily amenable to a Bayesian approach. We show that the frequentist properties of the Bayesian g-formula suggest it improves the accuracy of estimates of causal effects in small samples or when data are sparse. We demonstrate an approach to estimate the effect of environmental tobacco smoke on body mass index among children aged 4–9 years who were enrolled in a longitudinal birth cohort in New York, USA. We provide an algorithm and supply SAS and Stan code that can be adopted to implement this computational approach more generally.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StateAccepted/In press - Jan 1 2017
Externally publishedYes

Fingerprint

Bayes Theorem
Bayesian Approach
Smoke
Tobacco
Body Mass Index
Language
Public Health
Parturition
Estimate
Potential Outcomes
Causal Inference
Causal Effect
Small Sample
Demonstrate
Epidemiologists

Keywords

  • Bayesian
  • causal inference
  • g-computation
  • semiparametric

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

A Bayesian approach to the g-formula. / Keil, Alexander P.; Daza, Eric J.; Engel, Stephanie M.; Buckley, Jessie P; Edwards, Jessie K.

In: Statistical Methods in Medical Research, 01.01.2017.

Research output: Contribution to journalArticle

Keil, Alexander P. ; Daza, Eric J. ; Engel, Stephanie M. ; Buckley, Jessie P ; Edwards, Jessie K. / A Bayesian approach to the g-formula. In: Statistical Methods in Medical Research. 2017.
@article{63ef2aa4bc3f456eb0b73ac3020f83a6,
title = "A Bayesian approach to the g-formula",
abstract = "Epidemiologists often wish to estimate quantities that are easy to communicate and correspond to the results of realistic public health interventions. Methods from causal inference can answer these questions. We adopt the language of potential outcomes under Rubin’s original Bayesian framework and show that the parametric g-formula is easily amenable to a Bayesian approach. We show that the frequentist properties of the Bayesian g-formula suggest it improves the accuracy of estimates of causal effects in small samples or when data are sparse. We demonstrate an approach to estimate the effect of environmental tobacco smoke on body mass index among children aged 4–9 years who were enrolled in a longitudinal birth cohort in New York, USA. We provide an algorithm and supply SAS and Stan code that can be adopted to implement this computational approach more generally.",
keywords = "Bayesian, causal inference, g-computation, semiparametric",
author = "Keil, {Alexander P.} and Daza, {Eric J.} and Engel, {Stephanie M.} and Buckley, {Jessie P} and Edwards, {Jessie K.}",
year = "2017",
month = "1",
day = "1",
doi = "10.1177/0962280217694665",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - A Bayesian approach to the g-formula

AU - Keil, Alexander P.

AU - Daza, Eric J.

AU - Engel, Stephanie M.

AU - Buckley, Jessie P

AU - Edwards, Jessie K.

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Epidemiologists often wish to estimate quantities that are easy to communicate and correspond to the results of realistic public health interventions. Methods from causal inference can answer these questions. We adopt the language of potential outcomes under Rubin’s original Bayesian framework and show that the parametric g-formula is easily amenable to a Bayesian approach. We show that the frequentist properties of the Bayesian g-formula suggest it improves the accuracy of estimates of causal effects in small samples or when data are sparse. We demonstrate an approach to estimate the effect of environmental tobacco smoke on body mass index among children aged 4–9 years who were enrolled in a longitudinal birth cohort in New York, USA. We provide an algorithm and supply SAS and Stan code that can be adopted to implement this computational approach more generally.

AB - Epidemiologists often wish to estimate quantities that are easy to communicate and correspond to the results of realistic public health interventions. Methods from causal inference can answer these questions. We adopt the language of potential outcomes under Rubin’s original Bayesian framework and show that the parametric g-formula is easily amenable to a Bayesian approach. We show that the frequentist properties of the Bayesian g-formula suggest it improves the accuracy of estimates of causal effects in small samples or when data are sparse. We demonstrate an approach to estimate the effect of environmental tobacco smoke on body mass index among children aged 4–9 years who were enrolled in a longitudinal birth cohort in New York, USA. We provide an algorithm and supply SAS and Stan code that can be adopted to implement this computational approach more generally.

KW - Bayesian

KW - causal inference

KW - g-computation

KW - semiparametric

UR - http://www.scopus.com/inward/record.url?scp=85043693367&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043693367&partnerID=8YFLogxK

U2 - 10.1177/0962280217694665

DO - 10.1177/0962280217694665

M3 - Article

C2 - 29298607

AN - SCOPUS:85043693367

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -