{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T23:36:35Z","timestamp":1791934595172,"version":"4.3.4"},"reference-count":66,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2022,1,31]],"date-time":"2022-01-31T00:00:00Z","timestamp":1643587200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,5,18]]},"abstract":"<jats:title>Summary<\/jats:title>\n                  <jats:p>This paper combines causal mediation analysis with double machine learning for a data-driven control of observed confounders in a high-dimensional setting. The average indirect effect of a binary treatment and the unmediated direct effect are estimated based on efficient score functions, which are robust with respect to misspecifications of the outcome, mediator, and treatment models. This property is key for selecting these models by double machine learning, which is combined with data splitting to prevent overfitting. We demonstrate that the effect estimators are asymptotically normal and $n^{-1\/2}$-consistent under specific regularity conditions and investigate the finite sample properties of the suggested methods in a simulation study when considering lasso as machine learner. We also provide an empirical application to the US National Longitudinal Survey of Youth, assessing the indirect effect of health insurance coverage on general health operating via routine checkups as mediator, as well as the direct effect.<\/jats:p>","DOI":"10.1093\/ectj\/utac003","type":"journal-article","created":{"date-parts":[[2022,1,31]],"date-time":"2022-01-31T05:21:29Z","timestamp":1643606489000},"page":"277-300","source":"Crossref","is-referenced-by-count":190,"title":["Causal mediation analysis with double machine learning"],"prefix":"10.1093","volume":"25","author":[{"given":"Helmut","family":"Farbmacher","sequence":"first","affiliation":[{"name":"TUM School of Management, Technical University of Munich (TUM), Arcisstr. 21, 80333 Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin","family":"Huber","sequence":"additional","affiliation":[{"name":"University of Fribourg, Department of Economics, Bd de P\u00e9rolles 90, 1700 Fribourg, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luk\u00e1\u0161","family":"Laff\u00e9rs","sequence":"additional","affiliation":[{"name":"Matej Bel University, Department of Mathematics, Tajovskeho 40, 97411 Bansk\u00e1 Bystrica, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Henrika","family":"Langen","sequence":"additional","affiliation":[{"name":"University of Fribourg, Department of Economics, Bd de P\u00e9rolles 90, 1700 Fribourg, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin","family":"Spindler","sequence":"additional","affiliation":[{"name":"University of Hamburg, Faculty of Business Administration, Moorweidenstr. 18, 20148 Hamburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,1,31]]},"reference":[{"key":"2022052010525449900_bib1","doi-asserted-by":"crossref","first-page":"1282","DOI":"10.1002\/sim.3016","article-title":"Mediation analysis via potential outcomes models","volume":"27","author":"Albert","year":"2008","journal-title":"Statistics in Medicine"},{"key":"2022052010525449900_bib2","doi-asserted-by":"crossref","first-page":"1028","DOI":"10.1111\/j.1541-0420.2010.01547.x","article-title":"Generalized causal mediation analysis","volume":"67","author":"Albert","year":"2011","journal-title":"Biometrics"},{"issue":"18","key":"2022052010525449900_bib3","doi-asserted-by":"crossref","first-page":"1713","DOI":"10.1056\/NEJMsa1212321","article-title":"The oregon experiment: Effects of medicaid on clinical outcomes","volume":"368","author":"Baicker","year":"2013","journal-title":"New England Journal of Medicine"},{"key":"2022052010525449900_bib4","doi-asserted-by":"crossref","first-page":"1173","DOI":"10.1037\/0022-3514.51.6.1173","article-title":"The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations","volume":"51","author":"Baron","year":"1986","journal-title":"Journal of Personality and Social Psychology"},{"issue":"1","key":"2022052010525449900_bib5","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1111\/rssa.12388","article-title":"The long-run effect of childhood poverty and the mediating role of education","volume":"182","author":"Bellani","year":"2018","journal-title":"Journal of the Royal Statistical Society: Series A (Statistics in Society)"},{"key":"2022052010525449900_bib6","doi-asserted-by":"crossref","first-page":"233","DOI":"10.3982\/ECTA12723","article-title":"Program evaluation and causal inference with high-dimensional data","volume":"85","author":"Belloni","year":"2017","journal-title":"Econometrica"},{"key":"2022052010525449900_bib7","first-page":"1","article-title":"An IPW estimator for mediation effects in hazard models: With an application to schooling, cognitive ability, and mortality","volume":"57","author":"Bijwaard","year":"2018","journal-title":"Empirical Economics"},{"key":"2022052010525449900_bib8","article-title":"The causalweight package for causal inference in R","volume":"493","author":"Bodory","year":"2018"},{"key":"2022052010525449900_bib9","article-title":"National longitudinal survey of youth 1997 cohort, 1997\u20132017 (rounds 1\u201318)","author":"Bureau of Labor Statistics at the US Department of Labor","year":"2019"},{"issue":"4","key":"2022052010525449900_bib10","first-page":"389","article-title":"The effect of change of health insurance on access to care","volume":"35","author":"Burstin","year":"1998","journal-title":"Inquiry"},{"issue":"5","key":"2022052010525449900_bib11","doi-asserted-by":"crossref","first-page":"885","DOI":"10.1162\/REST_a_00431","article-title":"New evidence on the finite sample properties of propensity score matching and reweighting estimators","volume":"96","author":"Busso","year":"2014","journal-title":"Review of Economics and Statistics"},{"issue":"3","key":"2022052010525449900_bib12","doi-asserted-by":"crossref","first-page":"406","DOI":"10.1016\/j.econlet.2014.06.036","article-title":"The effect of health insurance coverage on the reported health of young adults","volume":"124","author":"Cardella","year":"2014","journal-title":"Economics Letters"},{"key":"2022052010525449900_bib13","doi-asserted-by":"crossref","first-page":"C1","DOI":"10.1111\/ectj.12097","article-title":"Double\/debiased machine learning for treatment and structural parameters","volume":"21","author":"Chernozhukov","year":"2018","journal-title":"Econometrics Journal"},{"key":"2022052010525449900_bib14","doi-asserted-by":"crossref","first-page":"261","DOI":"10.2307\/2527916","article-title":"Analysis of covariance: Its nature and uses","volume":"13","author":"Cochran","year":"1957","journal-title":"Biometrics"},{"key":"2022052010525449900_bib15","doi-asserted-by":"crossref","first-page":"F28","DOI":"10.1111\/ecoj.12420","article-title":"The effects of two influential early childhood interventions on health and healthy behaviour","volume":"126","author":"Conti","year":"2016","journal-title":"The Economic Journal"},{"key":"2022052010525449900_bib16","first-page":"353","article-title":"Machine learning in the estimation of causal effects: Targeted minimum loss-based estimation and double\/debiased machine learning","volume":"21","author":"D\u00edaz","year":"2020","journal-title":"Biostatistics"},{"issue":"3","key":"2022052010525449900_bib17","doi-asserted-by":"crossref","first-page":"661","DOI":"10.1111\/rssb.12362","article-title":"Causal mediation analysis for stochastic interventions","volume":"82","author":"D\u00edaz","year":"2020","journal-title":"Journal of the Royal Statistical Society: Series B (Statistical Methodology)"},{"issue":"6","key":"2022052010525449900_bib18","doi-asserted-by":"crossref","first-page":"453","DOI":"10.1016\/S0749-3797(18)30141-7","article-title":"The effect of health insurance coverage on the appropriate use of recommended clinical preventive services","volume":"13","author":"Faulkner","year":"1997","journal-title":"American Journal of Preventive Medicine"},{"key":"2022052010525449900_bib19","article-title":"Identification and estimation of causal mechanisms and net effects of a treatment under unconfoundedness","volume":"4237","author":"Flores","year":"2009","journal-title":"Institute of Labor Economics (IZA)"},{"issue":"4","key":"2022052010525449900_bib20","doi-asserted-by":"crossref","first-page":"502","DOI":"10.1007\/s11606-007-0127-2","article-title":"Risk of cardiovascular events and death: Does insurance matter?","volume":"22","author":"Fowler-Brown","year":"2007","journal-title":"Journal of General Internal Medicine"},{"issue":"2","key":"2022052010525449900_bib21","doi-asserted-by":"crossref","first-page":"315","DOI":"10.2307\/2998560","article-title":"On the role of the propensity score in efficient semiparametric estimation of average treatment effects","volume":"66","author":"Hahn","year":"1998","journal-title":"Econometrica"},{"key":"2022052010525449900_bib22","doi-asserted-by":"crossref","first-page":"2052","DOI":"10.1257\/aer.103.6.2052","article-title":"Understanding the mechanisms through which an influential early childhood program boosted adult outcomes","volume":"103","author":"Heckman","year":"2013","journal-title":"American Economic Review"},{"key":"2022052010525449900_bib23","first-page":"2401","article-title":"Ratio of mediator probability weighting for estimating natural direct and indirect effects","volume-title":"Proceedings of the American Statistical Association, Biometrics Section","author":"Hong","year":"2010"},{"key":"2022052010525449900_bib24","doi-asserted-by":"crossref","first-page":"920","DOI":"10.1002\/jae.2341","article-title":"Identifying causal mechanisms (primarily) based on inverse probability weighting","volume":"29","author":"Huber","year":"2014","journal-title":"Journal of Applied Econometrics"},{"key":"2022052010525449900_bib25","doi-asserted-by":"crossref","first-page":"179","DOI":"10.1080\/07350015.2014.937437","article-title":"Causal pitfalls in the decomposition of wage gaps","volume":"33","author":"Huber","year":"2015","journal-title":"Journal of Business and Economic Statistics"},{"key":"2022052010525449900_bib26","doi-asserted-by":"crossref","first-page":"180","DOI":"10.1162\/REST_a_00632","article-title":"Why do tougher caseworkers increase employment? The role of program assignment as a causal mechanism","volume":"99","author":"Huber","year":"2017","journal-title":"Review of Economics and Statistics"},{"key":"2022052010525449900_bib27","doi-asserted-by":"crossref","first-page":"441","DOI":"10.1111\/rssa.12279","article-title":"Direct and indirect effects of training vouchers for the unemployed","volume":"181","author":"Huber","year":"2018","journal-title":"Journal of the Royal Statistical Society: Series A (Statistics in Society)"},{"key":"2022052010525449900_bib29","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1093\/pan\/mps040","article-title":"Identification and sensitivity analysis for multiple causal mechanisms: Revisiting evidence from framing experiments","volume":"21","author":"Imai","year":"2013","journal-title":"Political Analysis"},{"key":"2022052010525449900_bib28","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1214\/10-STS321","article-title":"Identification, inference and sensitivity analysis for causal mediation effects","volume":"25","author":"Imai","year":"2010","journal-title":"Statistical Science"},{"key":"2022052010525449900_bib30","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1162\/003465304323023651","article-title":"Nonparametric estimation of average treatment effects under exogeneity: A review","volume":"86","author":"Imbens","year":"2004","journal-title":"Review of Economics and Statistics"},{"key":"2022052010525449900_bib31","doi-asserted-by":"crossref","first-page":"602","DOI":"10.1177\/0193841X8100500502","article-title":"Process analysis: Estimating mediation in treatment evaluations","volume":"5","author":"Judd","year":"1981","journal-title":"Evaluation Review"},{"key":"2022052010525449900_bib32","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1007\/s10994-011-5266-3","article-title":"Statistical analysis of kernel-based least-squares density-ratio estimation","volume":"86(3)","author":"Kanamori","year":"2012","journal-title":"Machine Learning"},{"key":"2022052010525449900_bib33","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1186\/1742-5573-1-4","article-title":"A further critique of the analytic strategy of adjusting for covariates to identify biologic mediation","volume":"1","author":"Kaufman","year":"2004","journal-title":"Epidemiologic Perspectives & Innovations"},{"key":"2022052010525449900_bib34","doi-asserted-by":"crossref","first-page":"937","DOI":"10.1002\/pam.21853","article-title":"Identifying mechanisms behind policy interventions via causal mediation analysis","volume":"34","author":"Keele","year":"2015","journal-title":"Journal of Policy Analysis and Management"},{"issue":"9673","key":"2022052010525449900_bib35","doi-asserted-by":"crossref","first-page":"1447","DOI":"10.1016\/S0140-6736(09)60239-7","article-title":"Public policy for the poor? A randomised assessment of the Mexican universal health insurance programme","volume":"373","author":"King","year":"2009","journal-title":"Lancet"},{"key":"2022052010525449900_bib36","article-title":"High-dimensional l2boosting: Rate of convergence","volume":"arXiv:1602.08927","author":"Luo","year":"2016","journal-title":"arXiv preprint"},{"issue":"9","key":"2022052010525449900_bib37","doi-asserted-by":"crossref","first-page":"1656","DOI":"10.1377\/hlthaff.2008.0701","article-title":"Greater use of preventive services in us health care could save lives at little or no cost","volume":"29","author":"Maciosek","year":"2010","journal-title":"Health Affairs"},{"key":"2022052010525449900_bib38","first-page":"159","article-title":"On semiparametric estimation of a path-specific effect in the presence of mediator-outcome confounding","volume":"107","author":"Miles","year":"2020","journal-title":"Biometrika"},{"key":"2022052010525449900_bib39","doi-asserted-by":"crossref","first-page":"691","DOI":"10.1093\/biomet\/78.3.691","article-title":"A note on a general definition of the coefficient of determination","volume":"78","author":"Nagelkerke","year":"1991","journal-title":"Biometrika"},{"issue":"6","key":"2022052010525449900_bib40","doi-asserted-by":"crossref","first-page":"999","DOI":"10.1016\/0277-9536(96)00016-0","article-title":"Do preventive health services reduce eventual demand for medical care?","volume":"43","author":"Nakanishi","year":"1996","journal-title":"Social Science & Medicine"},{"key":"2022052010525449900_bib41","first-page":"213","article-title":"Optimal Asymptotic Tests of Composite Statistical Hypotheses","volume-title":"Probability and Statistics","author":"Neyman","year":"1959"},{"issue":"12","key":"2022052010525449900_bib42","doi-asserted-by":"crossref","first-page":"1359","DOI":"10.1002\/hec.1226","article-title":"Health insurance coverage and the use of preventive services by Mexican adults","volume":"16","author":"Pag\u00e1n","year":"2007","journal-title":"Health Economics"},{"key":"2022052010525449900_bib43","first-page":"411","article-title":"Direct and indirect effects","volume-title":"Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence","author":"Pearl","year":"2001"},{"key":"2022052010525449900_bib44","doi-asserted-by":"crossref","first-page":"276","DOI":"10.1097\/01.ede.0000208475.99429.2d","article-title":"Estimation of direct causal effects","volume":"17","author":"Petersen","year":"2006","journal-title":"Epidemiology"},{"key":"2022052010525449900_bib45","article-title":"Insurance coverage and preventive care among adults","author":"Press","year":"2014"},{"issue":"4","key":"2022052010525449900_bib47","doi-asserted-by":"crossref","first-page":"365","DOI":"10.1080\/14034940701219642","article-title":"Preventive health screenings and health consultations in primary care increase life expectancy without increasing costs","volume":"35","author":"Rasmussen","year":"2007","journal-title":"Scandinavian Journal of Public Health"},{"key":"2022052010525449900_bib46","volume-title":"R: A Language and Environment for Statistical Computing","author":"R Core Team","year":"2020"},{"key":"2022052010525449900_bib48","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1093\/oso\/9780198510550.003.0007","article-title":"Semantics of causal DAG models and the identification of direct and indirect effects","volume-title":"Highly Structured Stochastic Systems","author":"Robins","year":"2003"},{"key":"2022052010525449900_bib49","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1097\/00001648-199203000-00013","article-title":"Identifiability and exchangeability for direct and indirect effects","volume":"3","author":"Robins","year":"1992","journal-title":"Epidemiology"},{"key":"2022052010525449900_bib50","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1080\/01621459.1995.10476494","article-title":"Semiparametric efficiency in multivariate regression models with missing data","volume":"90","author":"Robins","year":"1995","journal-title":"Journal of the American Statistical Association"},{"key":"2022052010525449900_bib51","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1080\/01621459.1994.10476818","article-title":"Estimation of regression coefficients when some regressors are not always observed","volume":"90","author":"Robins","year":"1994","journal-title":"Journal of the American Statistical Association"},{"key":"2022052010525449900_bib52","doi-asserted-by":"crossref","first-page":"688","DOI":"10.1037\/h0037350","article-title":"Estimating causal effects of treatments in randomized and nonrandomized studies","volume":"66","author":"Rubin","year":"1974","journal-title":"Journal of Educational Psychology"},{"issue":"2","key":"2022052010525449900_bib53","doi-asserted-by":"crossref","first-page":"390","DOI":"10.1002\/pam.21972","article-title":"The impact of health insurance on preventive care and health behaviors: Evidence from the first two years of the aca medicaid expansions","volume":"36","author":"Simon","year":"2017","journal-title":"Journal of Policy Analysis and Management"},{"issue":"6","key":"2022052010525449900_bib54","doi-asserted-by":"crossref","first-page":"1119","DOI":"10.1377\/hlthaff.2017.0293","article-title":"Three-year impacts of the affordable care act: Improved medical care and health among low-income adults","volume":"36","author":"Sommers","year":"2017","journal-title":"Health Affairs"},{"key":"2022052010525449900_bib55","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1016\/j.neunet.2009.07.007","article-title":"Dimensionality reduction for density ratio estimation in high-dimensional spaces","volume":"23(1)","author":"Sugiyama","year":"2010","journal-title":"Neural Networks"},{"key":"2022052010525449900_bib56","doi-asserted-by":"crossref","first-page":"4567","DOI":"10.1002\/sim.5864","article-title":"Inverse odds ratio-weighted estimation for causal mediation analysis","volume":"32","author":"Tchetgen\u00a0Tchetgen","year":"2013","journal-title":"Statistics in Medicine"},{"key":"2022052010525449900_bib57","first-page":"1816","article-title":"Semiparametric theory for causal mediation analysis: Efficiency bounds, multiple robustness, and sensitivity analysis","volume":"40","author":"Tchetgen\u00a0Tchetgen","year":"2012","journal-title":"Annals of Statistics"},{"key":"2022052010525449900_bib58","doi-asserted-by":"crossref","first-page":"926","DOI":"10.1111\/j.1541-0420.2007.00766.x","article-title":"Causal mediation analyses with rank preserving models","volume":"63","author":"Ten\u00a0Have","year":"2007","journal-title":"Biometrics"},{"key":"2022052010525449900_bib59","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.2517-6161.1996.tb02080.x","article-title":"Regresson shrinkage and selection via the lasso","volume":"58","author":"Tibshirani","year":"1996","journal-title":"Journal of the Royal Statistical Society"},{"key":"2022052010525449900_bib60","first-page":"1","article-title":"Targeted maximum likelihood learning","volume":"2","author":"van\u00a0der\u00a0Laan","year":"2006","journal-title":"International Journal of Biostatistics"},{"key":"2022052010525449900_bib61","first-page":"Article 25","article-title":"Super learner","volume":"6","author":"van\u00a0der\u00a0Laan","year":"2007","journal-title":"Statistical Applications in Genetics and Molecular Biology"},{"key":"2022052010525449900_bib62","doi-asserted-by":"crossref","first-page":"224","DOI":"10.1097\/EDE.0b013e318281a64e","article-title":"A three-way decomposition of a total effect into direct, indirect, and interactive effects","volume":"24","author":"VanderWeele","year":"2013","journal-title":"Epidemiology"},{"key":"2022052010525449900_bib63","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1097\/EDE.0b013e31818f69ce","article-title":"Marginal structural models for the estimation of direct and indirect effects","volume":"20","author":"VanderWeele","year":"2009","journal-title":"Epidemiology"},{"key":"2022052010525449900_bib64","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1515\/2161-962X.1014","article-title":"Imputation strategies for the estimation of natural direct and indirect effects","volume":"1","author":"Vansteelandt","year":"2012","journal-title":"Epidemiologic Methods"},{"issue":"3","key":"2022052010525449900_bib65","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1007\/s10754-016-9189-1","article-title":"Health insurance coverage and self-reported health: New estimates from the NLSY97","volume":"16","author":"Y\u00f6r\u00fck","year":"2016","journal-title":"International Journal of Health Economics and Management"},{"key":"2022052010525449900_bib66","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2202\/1557-4679.1361","article-title":"Targeted maximum likelihood estimation of natural direct effects","volume":"8","author":"Zheng","year":"2012","journal-title":"International Journal of Biostatistics"}],"container-title":["The Econometrics Journal"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/ectj\/advance-article-pdf\/doi\/10.1093\/ectj\/utac003\/42756191\/utac003.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/ectj\/article-pdf\/25\/2\/277\/43772863\/utac003.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/ectj\/article-pdf\/25\/2\/277\/43772863\/utac003.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,17]],"date-time":"2024-09-17T12:59:26Z","timestamp":1726577966000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/ectj\/article\/25\/2\/277\/6517682"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,31]]},"references-count":66,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,1,31]]},"published-print":{"date-parts":[[2022,5,18]]}},"URL":"https:\/\/doi.org\/10.1093\/ectj\/utac003","relation":{},"ISSN":["1368-4221","1368-423X"],"issn-type":[{"value":"1368-4221","type":"print"},{"value":"1368-423X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,5]]},"published":{"date-parts":[[2022,1,31]]}}}