gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula
Summary: Researchers are often interested in estimating the causal effects of sustained treatment strategies, i.e., of (hypothetical) interventions involving time-varying treatments. When using observational data, estimating those effects requires adjustment for confounding. However, conventional re...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2020-06-01
|
Series: | Patterns |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2666389920300088 |
id |
doaj-43386261bdea498989a09022ab9d55e2 |
---|---|
record_format |
Article |
spelling |
doaj-43386261bdea498989a09022ab9d55e22020-11-25T04:08:54ZengElsevierPatterns2666-38992020-06-0113100008gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formulaSean McGrath0Victoria Lin1Zilu Zhang2Lucia C. Petito3Roger W. Logan4Miguel A. Hernán5Jessica G. Young6Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USA; Corresponding authorSchool of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USADepartment of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA 02215, USA; Department of Medical Oncology, Dana Farber Cancer Institute, Boston, MA 02215, USAFeinberg School of Medicine, Northwestern University, Chicago, IL 60611, USADepartment of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USADepartment of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USA; Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USA; Harvard-MIT Division of Health Sciences and Technology, Cambridge, MA 02139, USADepartment of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA 02215, USASummary: Researchers are often interested in estimating the causal effects of sustained treatment strategies, i.e., of (hypothetical) interventions involving time-varying treatments. When using observational data, estimating those effects requires adjustment for confounding. However, conventional regression methods cannot appropriately adjust for confounding in the presence of treatment-confounder feedback. In contrast, estimators derived from Robins's g-formula may correctly adjust for confounding even if treatment-confounder feedback exists. The package gfoRmula implements in R one such estimator: the parametric g-formula. This estimator can be used to estimate the effects of binary or continuous time-varying treatments as well as contrasts defined by static or dynamic, deterministic, or random interventions, as well as interventions that depend on the natural value of treatment. The package accommodates survival outcomes as well as binary or continuous outcomes measured at the end of follow-up. This paper describes the gfoRmula package, along with motivating background, features, and examples. The Bigger Picture: Causal inference is a core task of data science. When data from randomized experiments are not available, data analysts often rely on nonexperimental (observational) data to estimate causal effects. The parametric g-formula is a statistical method to estimate the causal effects of sustained treatment strategies from observational data with time-varying treatments, confounders, and outcomes. Although this methodology was introduced in the 1980s, it has not been widely used due to the lack of open-source software. This article presents the gfoRmula package, an implementation of the parametric g-formula in R. The aim of this software is to facilitate the application of the parametric g-formula to complex, observational data to answer causal questions. Furthermore, this package helps provide a way to compare the performance of the parametric g-formula to other methods in the causal inference literature.http://www.sciencedirect.com/science/article/pii/S2666389920300088g-formulalongitudinal datacausal inferenceR |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Sean McGrath Victoria Lin Zilu Zhang Lucia C. Petito Roger W. Logan Miguel A. Hernán Jessica G. Young |
spellingShingle |
Sean McGrath Victoria Lin Zilu Zhang Lucia C. Petito Roger W. Logan Miguel A. Hernán Jessica G. Young gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula Patterns g-formula longitudinal data causal inference R |
author_facet |
Sean McGrath Victoria Lin Zilu Zhang Lucia C. Petito Roger W. Logan Miguel A. Hernán Jessica G. Young |
author_sort |
Sean McGrath |
title |
gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula |
title_short |
gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula |
title_full |
gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula |
title_fullStr |
gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula |
title_full_unstemmed |
gfoRmula: An R Package for Estimating the Effects of Sustained Treatment Strategies via the Parametric g-formula |
title_sort |
gformula: an r package for estimating the effects of sustained treatment strategies via the parametric g-formula |
publisher |
Elsevier |
series |
Patterns |
issn |
2666-3899 |
publishDate |
2020-06-01 |
description |
Summary: Researchers are often interested in estimating the causal effects of sustained treatment strategies, i.e., of (hypothetical) interventions involving time-varying treatments. When using observational data, estimating those effects requires adjustment for confounding. However, conventional regression methods cannot appropriately adjust for confounding in the presence of treatment-confounder feedback. In contrast, estimators derived from Robins's g-formula may correctly adjust for confounding even if treatment-confounder feedback exists. The package gfoRmula implements in R one such estimator: the parametric g-formula. This estimator can be used to estimate the effects of binary or continuous time-varying treatments as well as contrasts defined by static or dynamic, deterministic, or random interventions, as well as interventions that depend on the natural value of treatment. The package accommodates survival outcomes as well as binary or continuous outcomes measured at the end of follow-up. This paper describes the gfoRmula package, along with motivating background, features, and examples. The Bigger Picture: Causal inference is a core task of data science. When data from randomized experiments are not available, data analysts often rely on nonexperimental (observational) data to estimate causal effects. The parametric g-formula is a statistical method to estimate the causal effects of sustained treatment strategies from observational data with time-varying treatments, confounders, and outcomes. Although this methodology was introduced in the 1980s, it has not been widely used due to the lack of open-source software. This article presents the gfoRmula package, an implementation of the parametric g-formula in R. The aim of this software is to facilitate the application of the parametric g-formula to complex, observational data to answer causal questions. Furthermore, this package helps provide a way to compare the performance of the parametric g-formula to other methods in the causal inference literature. |
topic |
g-formula longitudinal data causal inference R |
url |
http://www.sciencedirect.com/science/article/pii/S2666389920300088 |
work_keys_str_mv |
AT seanmcgrath gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula AT victorialin gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula AT ziluzhang gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula AT luciacpetito gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula AT rogerwlogan gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula AT miguelahernan gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula AT jessicagyoung gformulaanrpackageforestimatingtheeffectsofsustainedtreatmentstrategiesviatheparametricgformula |
_version_ |
1724424130669838336 |