{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T02:04:36Z","timestamp":1780452276433,"version":"3.54.1"},"reference-count":55,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"DOI":"10.13039\/100000957","name":"Alzheimer\u2019s Association","doi-asserted-by":"crossref","award":["AARG-23-1077773"],"award-info":[{"award-number":["AARG-23-1077773"]}],"id":[{"id":"10.13039\/100000957","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000049","name":"National Institute on Aging","doi-asserted-by":"publisher","award":["K02AG076883"],"award-info":[{"award-number":["K02AG076883"]}],"id":[{"id":"10.13039\/100000049","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,2,13]]},"abstract":"<jats:title>Summary<\/jats:title>\n                  <jats:p>Comparing outcomes across treatments is essential in medicine and public policy. To do so, researchers typically estimate a set of parameters, possibly counterfactual, each targeting adifferent treatment. Treatment-specific means are commonly used, but their identification requires a positivity assumption: every subject has a nonzero probability of receiving each treatment. This assumption is often implausible, especially when treatment can take many values. Causal parameters based on dynamic stochastic interventions offer robustness to positivity violations. However, comparing these parameters may fail to reflect the effects of the underlying target treatments because the parameters can depend on outcomes under nontarget treatments. To clarify when two parameters targeting different treatments yield a useful comparison of treatment efficacy, we propose a comparability criterion: if the conditional treatment-specific mean for one treatment is greater than that for another, then the corresponding causal parameter should also be greater. Many standard parameters fail to satisfy this criterion, but we show that only a mild positivity assumption is needed to identify parameters that yield useful comparisons. We then provide two simple examples that satisfy this criterion and are identifiable under the milder positivity assumption: trimmed and smooth-trimmed treatment-specific means with multivalued treatments. For smooth-trimmed treatment-specific means, we develop doubly robust-style estimators that attain parametric convergence rates under nonparametric conditions. We illustrate our methods with an analysis of dialysis providers in New York State.<\/jats:p>","DOI":"10.1093\/biomet\/asag017","type":"journal-article","created":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T12:42:13Z","timestamp":1772800933000},"source":"Crossref","is-referenced-by-count":0,"title":["Comparing causal parameters with many treatments and positivity violations"],"prefix":"10.1093","volume":"113","author":[{"given":"A","family":"McClean","sequence":"first","affiliation":[{"name":"Artificial Intelligence Department, Ataraxis AI , 1239 Broadway, Suite 1502 , New York, New York 10001,","place":["U.S.A"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Y","family":"Li","sequence":"additional","affiliation":[{"name":"Artificial Intelligence Department, Ataraxis AI , 1239 Broadway, Suite 1502 , New York, New York 10001,","place":["U.S.A"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0098-8816","authenticated-orcid":false,"given":"S","family":"Bae","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"M","family":"McAdams DeMarco","sequence":"additional","affiliation":[{"name":"Department of Surgery, New York University , 1 Park Avenue , New York, New York 10016, \u00a0 yiting.li@nyulangone.org \u00a0 sunjae.bae@nyulangone.org \u00a0 mara.mcadamsdemarco@nyulangone.org","place":["U.S.A."]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9056-2047","authenticated-orcid":false,"given":"I","family":"D\u00edaz","sequence":"additional","affiliation":[{"name":"Division of Biostatistics, New York University , 180 Madison Avenue , New York, New York 10016, \u00a0 ivan.diaz@nyu.edu","place":["U.S.A"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7642-9773","authenticated-orcid":false,"given":"W","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health , 2024 E Monument Street, Baltimore , Maryland 21205, \u00a0 wenbowu@jhu.edu","place":["U.S.A"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,3,13]]},"reference":[{"key":"2026060221291144000_asag017-B1","volume-title":"Efficient and Adaptive Estimation for Semiparametric Models","author":"Bickel","year":"1993"},{"key":"2026060221291144000_asag017-B2","author":"Branson","year":"2024"},{"key":"2026060221291144000_asag017-B3","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1023\/A:1018046112532","article-title":"Stacked regressions","volume":"24","author":"Breiman","year":"1996","journal-title":"Mach. Learn"},{"key":"2026060221291144000_asag017-B4","doi-asserted-by":"crossref","first-page":"F452","DOI":"10.1111\/j.1468-0297.2010.02387.x","article-title":"Active labour market policy evaluations: a meta-analysis","volume":"120","author":"Card","year":"2010","journal-title":"Econ. J"},{"key":"2026060221291144000_asag017-B5","doi-asserted-by":"crossref","first-page":"138","DOI":"10.1016\/j.jeconom.2009.09.023","article-title":"Efficient semiparametric estimation of multi-valued treatment effects under ignorability","volume":"155","author":"Cattaneo","year":"2010","journal-title":"J. Economet"},{"key":"2026060221291144000_asag017-B6","first-page":"3096","volume-title":"Proc. 36th Int. Conf. Neural Info. Proces. Syst","author":"Chen","year":"2022"},{"key":"2026060221291144000_asag017-B7","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":"Economet. J"},{"key":"2026060221291144000_asag017-B8","doi-asserted-by":"crossref","first-page":"20160016","DOI":"10.1515\/em-2016-0016","article-title":"Doubly robust estimator for indirectly standardized mortality ratios","volume":"6","author":"Daignault","year":"2017","journal-title":"Epidemiol. Meth"},{"key":"2026060221291144000_asag017-B9","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":"J. R. Statist. Soc. B"},{"key":"2026060221291144000_asag017-B10","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1111\/j.1541-0420.2011.01685.x","article-title":"Population intervention causal effects based on stochastic interventions","volume":"68","author":"D\u00edaz","year":"2012","journal-title":"Biometrics"},{"key":"2026060221291144000_asag017-B11","doi-asserted-by":"crossref","first-page":"449","DOI":"10.1111\/j.1467-985X.2004.apm12.x","article-title":"Statistical analysis of performance indicators in UK higher education","volume":"167","author":"Draper","year":"2004","journal-title":"J. R. Statist. Soc. A"},{"key":"2026060221291144000_asag017-B12","doi-asserted-by":"crossref","DOI":"10.1007\/b97848","volume-title":"A Distribution-Free Theory of Nonparametric Regression","author":"Gy\u00f6rfi","year":"2002"},{"key":"2026060221291144000_asag017-B13","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":"2026060221291144000_asag017-B14","doi-asserted-by":"crossref","first-page":"5260","DOI":"10.1002\/sim.5907","article-title":"Estimation of the effect of interventions that modify the received treatment","volume":"32","author":"Haneuse","year":"2013","journal-title":"Statist. Med"},{"key":"2026060221291144000_asag017-B15","author":"Hastie","year":"2024"},{"key":"2026060221291144000_asag017-B16","doi-asserted-by":"crossref","first-page":"645","DOI":"10.1080\/01621459.2017.1422737","article-title":"Nonparametric causal effects based on incremental propensity score interventions","volume":"114","author":"Kennedy","year":"2019","journal-title":"J. Am. Statist. Assoc"},{"key":"2026060221291144000_asag017-B17","doi-asserted-by":"crossref","first-page":"1229","DOI":"10.1111\/rssb.12212","article-title":"Non-parametric methods for doubly robust estimation of continuous treatment effects","volume":"79","author":"Kennedy","year":"2017","journal-title":"J. R. Statist. Soc. B"},{"key":"2026060221291144000_asag017-B18","doi-asserted-by":"crossref","first-page":"2021","DOI":"10.3982\/ECTA7372","article-title":"Irregular identification, support conditions, and inverse weight estimation","volume":"78","author":"Khan","year":"2010","journal-title":"Econometrica"},{"key":"2026060221291144000_asag017-B19","author":"Khan","year":"2024"},{"key":"2026060221291144000_asag017-B20","first-page":"1168","article-title":"Components of a difference between two rates","volume":"50","author":"Kitagawa","year":"1955","journal-title":"J. Am. Statist. Assoc"},{"key":"2026060221291144000_asag017-B21","first-page":"589","article-title":"A joint confidence region for an overall ranking of populations","volume":"69","author":"Klein","year":"2020","journal-title":"Appl. Statist."},{"key":"2026060221291144000_asag017-B22","author":"Levis","year":"2024"},{"key":"2026060221291144000_asag017-B23","doi-asserted-by":"crossref","first-page":"2389","DOI":"10.1214\/19-AOAS1282","article-title":"Propensity score weighting for causal inference with multiple treatments","volume":"13","author":"Li","year":"2019","journal-title":"Ann. Appl. Statist"},{"key":"2026060221291144000_asag017-B24","doi-asserted-by":"crossref","first-page":"1363","DOI":"10.1111\/rssa.12529","article-title":"Performance assessment as an application of causal inference","volume":"183","author":"Longford","year":"2020","journal-title":"J. R. Statist. Soc. A"},{"key":"2026060221291144000_asag017-B25","doi-asserted-by":"crossref","first-page":"782","DOI":"10.1136\/bmjqs-2013-002608","article-title":"Differences in case-mix can influence the comparison of standardised mortality ratios even with optimal risk adjustment: an analysis of data from paediatric intensive care","volume":"23","author":"Manktelow","year":"2014","journal-title":"BMJ Qual. Saf"},{"key":"2026060221291144000_asag017-B26","doi-asserted-by":"crossref","first-page":"701","DOI":"10.1136\/bmjqs-2014-003358","article-title":"Simpson\u2019s paradox: how performance measurement can fail even with perfect risk adjustment","volume":"23","author":"Marang-van de Mheen","year":"2014","journal-title":"BMJ Qual. Saf"},{"key":"2026060221291144000_asag017-B27","doi-asserted-by":"crossref","first-page":"476","DOI":"10.1093\/restud\/rdad006","article-title":"Inference for ranks with applications to mobility across neighbourhoods and academic achievement across countries","volume":"91","author":"Mogstad","year":"2024","journal-title":"Rev. Econ. Studies"},{"key":"2026060221291144000_asag017-B28","doi-asserted-by":"crossref","first-page":"803","DOI":"10.1080\/01621459.1997.10474036","article-title":"Statistical methods for profiling providers of medical care: issues and applications","volume":"92","author":"Normand","year":"1997","journal-title":"J. Am. Statist. Assoc"},{"key":"2026060221291144000_asag017-B29","author":"Polley","year":"2024"},{"key":"2026060221291144000_asag017-B30","volume-title":"R: A Language and Environment for Statistical Computing","author":"R Development Core Team","year":"2026"},{"key":"2026060221291144000_asag017-B31","doi-asserted-by":"crossref","first-page":"596","DOI":"10.1214\/14-STS499","article-title":"Nonparametric bounds and sensitivity analysis of treatment effects","volume":"29","author":"Richardson","year":"2014","journal-title":"Statist. Sci."},{"key":"2026060221291144000_asag017-B32","doi-asserted-by":"crossref","first-page":"1393","DOI":"10.1016\/0270-0255(86)90088-6","article-title":"A new approach to causal inference in mortality studies with a sustained exposure period\u2014application to control of the healthy worker survivor effect","volume":"7","author":"Robins","year":"1986","journal-title":"Math. Mod"},{"key":"2026060221291144000_asag017-B33","first-page":"pp. 335","volume-title":"Probability and Statistics: Essays in Honor of David A. Freedman (Inst. Math. Statist. (IMS) Collect. 2)","author":"Robins","year":"2008"},{"key":"2026060221291144000_asag017-B34","doi-asserted-by":"crossref","first-page":"e0257003","DOI":"10.1371\/journal.pone.0257003","article-title":"Can we trust the standardized mortality ratio? A formal analysis and evaluation based on axiomatic requirements","volume":"16","author":"Roessler","year":"2021","journal-title":"PLoS One"},{"key":"2026060221291144000_asag017-B35","doi-asserted-by":"crossref","first-page":"103","DOI":"10.3102\/10769986029001103","article-title":"A potential outcomes view of value-added assessment in education","volume":"29","author":"Rubin","year":"2004","journal-title":"J. Educ. Behav. Statist"},{"key":"2026060221291144000_asag017-B37","doi-asserted-by":"crossref","first-page":"1319","DOI":"10.1016\/j.athoracsur.2019.09.010","article-title":"The use and misuse of indirectly standardized, risk-adjusted outcomes and star ratings","volume":"109","author":"Shahian","year":"2020","journal-title":"Ann. Thorac. Surg"},{"key":"2026060221291144000_asag017-B36","doi-asserted-by":"crossref","first-page":"1955","DOI":"10.1161\/CIRCULATIONAHA.107.747873","article-title":"Comparison of \u2018risk-adjusted\u2019 hospital outcomes","volume":"117","author":"Shahian","year":"2008","journal-title":"Circulation"},{"key":"2026060221291144000_asag017-B39","doi-asserted-by":"crossref","first-page":"412","DOI":"10.1097\/SLA.0000000000003602","article-title":"Comparing outcomes and costs of surgical patients treated at major teaching and nonteaching hospitals: a national matched analysis","volume":"271","author":"Silber","year":"2020","journal-title":"Ann. Surg"},{"key":"2026060221291144000_asag017-B38","doi-asserted-by":"crossref","first-page":"1446","DOI":"10.1111\/1475-6773.12156","article-title":"Template matching for auditing hospital cost and quality","volume":"49","author":"Silber","year":"2014","journal-title":"Health Serv. Res"},{"key":"2026060221291144000_asag017-B40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1214\/09-STS313","article-title":"Matching methods for causal inference: a review and a look forward","volume":"25","author":"Stuart","year":"2010","journal-title":"Statist. Sci"},{"key":"2026060221291144000_asag017-B41","author":"Susmann","year":"2025"},{"key":"2026060221291144000_asag017-B42","first-page":"55","article-title":"On causal inference in the presence of interference","volume":"21","author":"Tchetgen Tchetgen","year":"2012","journal-title":"Statist. Meth.: Med. Res"},{"key":"2026060221291144000_asag017-B43","volume-title":"Semiparametric Theory and Missing Data","author":"Tsiatis","year":"2006"},{"key":"2026060221291144000_asag017-B44","volume-title":"2022 USRDS Annual Data Report: Epidemiology of Kidney Disease in the United States","author":"U.S. Renal Data System","year":"2022"},{"key":"2026060221291144000_asag017-B46","doi-asserted-by":"crossref","first-page":"25","DOI":"10.2202\/1544-6115.1309","article-title":"Super learner","volume":"6","author":"van der Laan","year":"2007","journal-title":"Statist. Appl. Genet. Mol. Biol"},{"key":"2026060221291144000_asag017-B45","doi-asserted-by":"publisher","DOI":"10.2202\/1557-4679.1043","article-title":"Targeted maximum likelihood learning","volume":"2","author":"van der Laan","year":"2006","journal-title":"Int. J. Biostatist"},{"key":"2026060221291144000_asag017-B47","volume-title":"Asymptotic Statistics","author":"van der Vaart","year":"2000"},{"key":"2026060221291144000_asag017-B48","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4757-2545-2","volume-title":"Weak Convergence and Empirical Processes","author":"van der Vaart","year":"1996"},{"key":"2026060221291144000_asag017-B49","first-page":"7","article-title":"On model selection and model misspecification in causal inference","volume":"21","author":"Vansteelandt","year":"2012","journal-title":"Statist. Meth.: Med. Res"},{"key":"2026060221291144000_asag017-B50","doi-asserted-by":"crossref","first-page":"1090","DOI":"10.1097\/MLR.0000000000001645","article-title":"Hospital-specific template matching for benchmarking performance in a diverse multihospital system","volume":"59","author":"Vincent","year":"2021","journal-title":"Med. Care"},{"key":"2026060221291144000_asag017-B51","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1214\/aoms\/1177730385","article-title":"On the asymptotic distribution of differentiable statistical functions","volume":"18","author":"von Mises","year":"1947","journal-title":"Ann. Math. Statist"},{"key":"2026060221291144000_asag017-B52","first-page":"509","article-title":"Intervention treatment distributions that depend on the observed treatment process and model double robustness in causal survival analysis","volume":"32","author":"Wen","year":"2023","journal-title":"Statist. Meth.: Med. Res"},{"key":"2026060221291144000_asag017-B53","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v077.i01","article-title":"ranger: a fast implementation of random forests for high dimensional data in C++ and R","volume":"77","author":"Wright","year":"2017","journal-title":"J. Statist. Software"},{"key":"2026060221291144000_asag017-B54","doi-asserted-by":"crossref","first-page":"487","DOI":"10.1093\/biomet\/asy008","article-title":"Asymptotic inference of causal effects with observational studies trimmed by the estimated propensity scores","volume":"105","author":"Yang","year":"2018","journal-title":"Biometrika"},{"key":"2026060221291144000_asag017-B55","doi-asserted-by":"crossref","first-page":"1055","DOI":"10.1111\/biom.12505","article-title":"Propensity score matching and subclassification in observational studies with multi-level treatments","volume":"72","author":"Yang","year":"2016","journal-title":"Biometrics"}],"container-title":["Biometrika"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/biomet\/advance-article-pdf\/doi\/10.1093\/biomet\/asag017\/67348705\/asag017.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/113\/2\/asag017\/67348705\/asag017.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/113\/2\/asag017\/67348705\/asag017.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T01:29:37Z","timestamp":1780450177000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biomet\/article\/doi\/10.1093\/biomet\/asag017\/8520887"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":55,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,2,13]]}},"URL":"https:\/\/doi.org\/10.1093\/biomet\/asag017","relation":{},"ISSN":["0006-3444","1464-3510"],"issn-type":[{"value":"0006-3444","type":"print"},{"value":"1464-3510","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026]]},"published":{"date-parts":[[2026]]},"article-number":"asag017"}}