The use of prognostic scores for causal inference with general treatment regimes

Research output: Contribution to journalJournal articlepeer-review

Standard

The use of prognostic scores for causal inference with general treatment regimes. / Nguyen, Tri Long; Debray, Thomas P.A.

In: Statistics in Medicine, Vol. 38, No. 11, 2019, p. 2013-2029.

Research output: Contribution to journalJournal articlepeer-review

Harvard

Nguyen, TL & Debray, TPA 2019, 'The use of prognostic scores for causal inference with general treatment regimes', Statistics in Medicine, vol. 38, no. 11, pp. 2013-2029. https://doi.org/10.1002/sim.8084

APA

Nguyen, T. L., & Debray, T. P. A. (2019). The use of prognostic scores for causal inference with general treatment regimes. Statistics in Medicine, 38(11), 2013-2029. https://doi.org/10.1002/sim.8084

Vancouver

Nguyen TL, Debray TPA. The use of prognostic scores for causal inference with general treatment regimes. Statistics in Medicine. 2019;38(11):2013-2029. https://doi.org/10.1002/sim.8084

Author

Nguyen, Tri Long ; Debray, Thomas P.A. / The use of prognostic scores for causal inference with general treatment regimes. In: Statistics in Medicine. 2019 ; Vol. 38, No. 11. pp. 2013-2029.

Bibtex

@article{d5ecc372942244e5bff7c1bf5dd54209,
title = "The use of prognostic scores for causal inference with general treatment regimes",
abstract = "In nonrandomised studies, inferring causal effects requires appropriate methods for addressing confounding bias. Although it is common to adopt propensity score analysis to this purpose, prognostic score analysis has recently been proposed as an alternative strategy. While both approaches were originally introduced to estimate causal effects for binary interventions, the theory of propensity score has since been extended to the case of general treatment regimes. Indeed, many treatments are not assigned in a binary fashion and require a certain extent of dosing. Hence, researchers may often be interested in estimating treatment effects across multiple exposures. To the best of our knowledge, the prognostic score analysis has not been yet generalised to this case. In this article, we describe the theory of prognostic scores for causal inference with general treatment regimes. Our methods can be applied to compare multiple treatments using nonrandomised data, a topic of great relevance in contemporary evaluations of clinical interventions. We propose estimators for the average treatment effects in different populations of interest, the validity of which is assessed through a series of simulations. Finally, we present an illustrative case in which we estimate the effect of the delay to Aspirin administration on a composite outcome of death or dependence at 6 months in stroke patients.",
keywords = "causal inference, multiple treatment exposures, observational study, prognostic score",
author = "Nguyen, {Tri Long} and Debray, {Thomas P.A.}",
year = "2019",
doi = "10.1002/sim.8084",
language = "English",
volume = "38",
pages = "2013--2029",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "JohnWiley & Sons Ltd",
number = "11",

}

RIS

TY - JOUR

T1 - The use of prognostic scores for causal inference with general treatment regimes

AU - Nguyen, Tri Long

AU - Debray, Thomas P.A.

PY - 2019

Y1 - 2019

N2 - In nonrandomised studies, inferring causal effects requires appropriate methods for addressing confounding bias. Although it is common to adopt propensity score analysis to this purpose, prognostic score analysis has recently been proposed as an alternative strategy. While both approaches were originally introduced to estimate causal effects for binary interventions, the theory of propensity score has since been extended to the case of general treatment regimes. Indeed, many treatments are not assigned in a binary fashion and require a certain extent of dosing. Hence, researchers may often be interested in estimating treatment effects across multiple exposures. To the best of our knowledge, the prognostic score analysis has not been yet generalised to this case. In this article, we describe the theory of prognostic scores for causal inference with general treatment regimes. Our methods can be applied to compare multiple treatments using nonrandomised data, a topic of great relevance in contemporary evaluations of clinical interventions. We propose estimators for the average treatment effects in different populations of interest, the validity of which is assessed through a series of simulations. Finally, we present an illustrative case in which we estimate the effect of the delay to Aspirin administration on a composite outcome of death or dependence at 6 months in stroke patients.

AB - In nonrandomised studies, inferring causal effects requires appropriate methods for addressing confounding bias. Although it is common to adopt propensity score analysis to this purpose, prognostic score analysis has recently been proposed as an alternative strategy. While both approaches were originally introduced to estimate causal effects for binary interventions, the theory of propensity score has since been extended to the case of general treatment regimes. Indeed, many treatments are not assigned in a binary fashion and require a certain extent of dosing. Hence, researchers may often be interested in estimating treatment effects across multiple exposures. To the best of our knowledge, the prognostic score analysis has not been yet generalised to this case. In this article, we describe the theory of prognostic scores for causal inference with general treatment regimes. Our methods can be applied to compare multiple treatments using nonrandomised data, a topic of great relevance in contemporary evaluations of clinical interventions. We propose estimators for the average treatment effects in different populations of interest, the validity of which is assessed through a series of simulations. Finally, we present an illustrative case in which we estimate the effect of the delay to Aspirin administration on a composite outcome of death or dependence at 6 months in stroke patients.

KW - causal inference

KW - multiple treatment exposures

KW - observational study

KW - prognostic score

U2 - 10.1002/sim.8084

DO - 10.1002/sim.8084

M3 - Journal article

C2 - 30652333

AN - SCOPUS:85060146705

VL - 38

SP - 2013

EP - 2029

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 11

ER -

ID: 218395518