Causal interaction trees: Finding subgroups with heterogeneous treatment effects in observational data

被引：14

作者：

Yang, Jiabei ^{[1
,2
]}

Dahabreh, Issa J. ^{[3
,4
]}

Steingrimsson, Jon A. ^{[1
]}

机构：

[1] Brown Univ, Dept Biostat, Sch Publ Hlth, Providence, RI 02912 USA

[2] Brown Univ, Sch Publ Hlth, Ctr Evidence Synth Hlth, Providence, RI 02912 USA

[3] Harvard TH Chan Sch Publ Hlth, Dept Epidemiol, Boston, MA USA

[4] Harvard TH Chan Sch Publ Hlth, Dept Biostat, Boston, MA USA

来源：

BIOMETRICS | 2022年 / 78卷 / 02期

关键词：

causal inference; doubly robust estimators; heterogeneity of treatment effects; machine learning; recursive partitioning; PROPENSITY SCORE; INFERENCE;

D O I：

10.1111/biom.13432

中图分类号：

Q [生物科学];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

We introduce causal interaction tree (CIT) algorithms for finding subgroups of individuals with heterogeneous treatment effects in observational data. The CIT algorithms are extensions of the classification and regression tree algorithm that use splitting criteria based on subgroup-specific treatment effect estimators appropriate for observational data. We describe inverse probability weighting, g-formula, and doubly robust estimators of subgroup-specific treatment effects, derive their asymptotic properties, and use them to construct splitting criteria for the CIT algorithms. We study the performance of the algorithms in simulations and implement them to analyze data from an observational study that evaluated the effectiveness of right heart catheterization for critically ill patients.

引用

页码：624 / 635

页数：12

共 28 条

[1] Recursive partitioning for heterogeneous causal effects [J].

Athey, Susan ;

Imbens, Guido .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2016, 113 (27) :7353-7360

[2] Doubly robust estimation in missing data and causal inference models [J].

Bang, H .

BIOMETRICS, 2005, 61 (04) :962-972

[3] The Highly Adaptive Lasso Estimator [J].

Benkeser, David ;

van der Laan, Mark .

PROCEEDINGS OF 3RD IEEE/ACM INTERNATIONAL CONFERENCE ON DATA SCIENCE AND ADVANCED ANALYTICS, (DSAA 2016), 2016, :689-696

[4] Double/debiased machine learning for treatment and structural parameters [J].

Chernozhukov, Victor ;

Chetverikov, Denis ;

Demirer, Mert ;

Duflo, Esther ;

Hansen, Christian ;

Newey, Whitney ;

Robins, James .

ECONOMETRICS JOURNAL, 2018, 21 (01) :C1-C68

[5] Hydrological prediction in a non-stationary world [J].

Clarke, Robin T. .

HYDROLOGY AND EARTH SYSTEM SCIENCES, 2007, 11 (01) :408-414

[6] The effectiveness of right heart catheterization in the initial care of critically ill patients [J].

Connors, AF ;

Speroff, T ;

Dawson, NV ;

Thomas, C ;

Harrell, FE ;

Wagner, D ;

Desbiens, N ;

Goldman, L ;

Wu, AW ;

Califf, RM ;

Fulkerson, WJ ;

Vidaillet, H ;

Broste, S ;

Bellamy, P ;

Lynn, J ;

Knaus, WA .

JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, 1996, 276 (11) :889-897

[7] Using group data to treat individuals: understanding heterogeneous treatment effects in the age of precision medicine and patient-centred evidence [J].

Dahabreh, Issa J. ;

Hayward, Rodney ;

Kent, David M. .

INTERNATIONAL JOURNAL OF EPIDEMIOLOGY, 2016, 45 (06) :2184-2193

[8]

Fithian W., 2014, ARXIV14102597

[9]

Hirano K., 2001, HLTH SERVICE OUTCOME, V2, P259, DOI [10.1023/A:1020371312283, DOI 10.1023/A:1020371312283]

[10] Causal Inference of Interaction Effects with Inverse Propensity Weighting, G-Computation and Tree-Based Standardization [J].

Kang, Joseph ;

Su, Xiaogang ;

Liu, Lei ;

Daviglus, Martha L. .

STATISTICAL ANALYSIS AND DATA MINING, 2014, 7 (05) :323-336

← 1 2 3 →