Causal rule ensemble method for estimating heterogeneous treatment effect with consideration of prognostic effects

Abstract

We propose a novel framework based on the RuleFit method to estimate heterogeneous treatment effect in randomized clinical trials. The proposed method estimates a rule ensemble comprising a set of prognostic rules, a set of prescriptive rules, as well as the linear effects of the original predictor variables. The prescriptive rules provide an interpretable description of the heterogeneous treatment effect. By including a prognostic term in the proposed model, the selected rule is represented as an heterogeneous treatment effect that excludes other effects. We confirmed that the performance of the proposed method was equivalent to that of other ensemble learning methods through numerical simulations and demonstrated the interpretation of the proposed method using a real data application.

Keywords

Heterogeneous treatment effect RuleFit randomized clinical trial machine learning metalearner

Get full access to this article

View all access options for this article.

References

Holland

. Statistics and causal inference. J Am Stat Assoc 1986; 81: 945–960.

Gail

Simon

. Testing for qualitative interactions between treatment effects and patient subsets. Biometrics 1985; 41: 361–372.

Wager

Athey

. Estimation and inference of heterogeneous treatment effects using random forests. J Am Stat Assoc 2018; 113: 1228–1242.

Breiman

Friedman

Olshen

, et al. Classification and regression trees. New York: Wardsworth, 1984.

Tsai

Wang

, et al. Subgroup analysis via recursive partitioning. J Mach Learn Res 2009; 10: 141–158.

Athey

Imbens

. Recursive partitioning for heterogeneous causal effects. Proce Nat Acad Sci 2016; 113: 7353–7360.

Breiman

. Random forests. Mach Learn 2001; 45: 5–32.

Athey

Tibshirani

Wager

. Generalized random forests. Ann Stat 2019; 47: 1148–1178.

Powers

Qian

Jung

, et al. Some methods for heterogeneous treatment effect estimation in high dimensions. Stat Med 2018; 37: 1767–1787.

10.

Chipman

George

McCulloch

. Bart: Bayesian additive regression trees. Ann Appl Stat 2010; 4: 266–298.

11.

Hill

. Bayesian nonparametric modeling for causal inference. J Comput Graph Stat 2011; 20: 217–240.

12.

Hahn

Murray

Carvalho

. Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion). Bayesian Anal 2020; 15: 965–1056.

13.

Friedman

Popescu

. Predictive learning via rule ensembles. Ann Appl Stat 2008; 2: 916–954.

14.

Bargagli-Stoffi

Cadei

Lee

, et al. Causal rule ensemble: Interpretable discovery and inference of heterogeneous treatment effects. arXiv:2009.09036v4, 2023.

15.

Yuan

Lin

. Model selection and estimation in regression with grouped variables. J R Stat Soc: Ser B (Statistical Methodology) 2006; 68: 49–67.

16.

Tibshirani

. Regression shrinkage and selection via the lasso. J R Stat Soc: Ser B (Methodological) 1996; 58: 267–288.

17.

Wan

Tanioka

Shimokawa

. Rule ensemble method with adaptive group lasso for heterogeneous treatment effect estimation. Stat Med 2023; 42: 3413–3442.

18.

Wan

. Causal-Rule-Ensemble-grfit-. URL https://github.com/Kwan12321/Causal-Rule-Ensemble-grfit-/blob/main/. GitHub repository. 2023.

19.

Fokkema

. Fitting prediction rule ensembles with R package pre. J Stat Softw 2020; 92: 1–30.

20.

Friedman

. Greedy function approximation: a gradient boosting machine. Ann Stat 2001; 29: 1189–1232.

21.

Therneau

Atkinson

. rpart: Recursive Partitioning and Regression Trees. URL https://CRAN.R-project.org/package=rpart. R package version 4.1.16., 2022.

22.

Breheny

. grpreg: Regularization Paths for Regression Models with Grouped Covariates. R package version 3.4.0., 2021.

23.

Tian

Alizadeh

Gentles

, et al. A simple method for estimating interactions between a treatment and a large number of covariates. J Am Stat Assoc 2014; 109: 1517–1532.

24.

Tibshirani

Athey

Sverdrup

, et al. grf: Generalized random forests. URL https://CRAN.R-project.org/package=grf. R package version 2.2.1., 2022.

25.

Powers

Qian

Hastie

, et al. causalLearning: Methods for heterogeneous treatment effect estimation. R package version 1.0.0., 2022.

26.

Hammer

Katzenstein

Hughes

, et al. A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with cd4 cell counts from 200 to 500 per cubic millimeter. New Engl J Med 1996; 335: 1081–1090.

27.

Juraska

Gilbert

, et al. speff2trial: Semiparametric efficient estimation for a two-sample treatment effect. URL https://CRAN.R-project.org/package=speff2trial. R package version 1.0.5., 2022.

28.

Saravolatz

Winslow

Collins

, et al. Zidovudine alone or in combination with didanosine or zalcitabine in HIV-infected patients with the acquired immunodeficiency syndrome or fewer than 200 cd4 cells per cubic millimeter. Investigators for the terry beirn community programs for clinical research on aids. N Engl J Med 1996; 335: 1099–1106.

29.

Hothorn

Hornik

Zeileis

. Unbiased recursive partitioning: a conditional inference framework. J Comput Graph Stat 2006; 15: 651–674.

30.

Shih

Tsai

. Variable selection bias in regression trees with constant fits. Comput Stat Data Anal 2004; 45: 595–607.

31.

Hothorn

Zeileis

. partykit: A modular toolkit for recursive partytioning in R . URL http://partykit.r-forge.r-project.org/partykit/. Journal of Machine Learning Research, 2015, 16: 3905–3909.

32.

Wang

Leng

. A note on adaptive group lasso. Comput Stat Data Anal 2008; 52: 5277–5286.

33.

Künzel

Sekhon

Bickel

, et al. Metalearners for estimating heterogeneous treatment effects using machine learning. Proce Nat Acad Sci 2019; 116: 4156–4165.

34.

Nie

Wager

. Quasi-oracle estimation of heterogeneous treatment effects. Biometrika 2020; 108: 299–319.

35.

Yadlowsky

Fleming

Shah

, et al. Evaluating treatment prioritization rules via rank-weighted average treatment effects. Arxiv. arXiv:2111.07966, 2021.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.47 MB