{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,9]],"date-time":"2026-01-09T15:16:17Z","timestamp":1767971777599,"version":"3.49.0"},"reference-count":45,"publisher":"Oxford University Press (OUP)","issue":"3","license":[{"start":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T00:00:00Z","timestamp":1753315200000},"content-version":"vor","delay-in-days":21,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000024","name":"Canadian Institutes of Health Research","doi-asserted-by":"publisher","award":["FDN-16726"],"award-info":[{"award-number":["FDN-16726"]}],"id":[{"id":"10.13039\/501100000024","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,7,3]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n               <jats:p>Individualized treatment rules tailor treatments to patients based on clinical, demographic, and other characteristics. Estimation of individualized treatment rules requires the identification of individuals who benefit most from the particular treatments and thus the detection of variability in treatment effects. To develop an effective individualized treatment rule, data from multisite studies may be required due to the low power provided by smaller datasets for detecting the often small treatment-covariate interactions. However, sharing of individual-level data is sometimes constrained. Furthermore, sparsity may arise in 2 senses: different data sites may recruit from different populations, making it infeasible to estimate identical models or all parameters of interest at all sites, and the number of non-zero parameters in the model for the treatment rule may be small. To address these issues, we adopt a 2-stage Bayesian meta-analysis approach to estimate individualized treatment rules which optimize expected patient outcomes using multisite data without disclosing individual-level data beyond the sites. Simulation results demonstrate that our approach can provide consistent estimates of the parameters which fully characterize the optimal individualized treatment rule. We estimate the optimal Warfarin dose strategy using data from the International Warfarin Pharmacogenetics Consortium, where data sparsity and small treatment-covariate interaction effects pose additional statistical challenges.<\/jats:p>","DOI":"10.1093\/biomtc\/ujaf082","type":"journal-article","created":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T15:12:09Z","timestamp":1753369929000},"source":"Crossref","is-referenced-by-count":1,"title":["Sparse 2-stage Bayesian meta-analysis for individualized treatments"],"prefix":"10.1093","volume":"81","author":[{"given":"Junwei","family":"Shen","sequence":"first","affiliation":[{"name":"McGill University Department of Epidemiology, Biostatistics and Occupational Health, , Montr\u00e9al, Qu\u00e9bec H3A 1G1 ,","place":["Canada"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7225-3977","authenticated-orcid":false,"given":"Erica E M","family":"Moodie","sequence":"additional","affiliation":[{"name":"McGill University Department of Epidemiology, Biostatistics and Occupational Health, , Montr\u00e9al, Qu\u00e9bec H3A 1G1 ,","place":["Canada"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3382-9563","authenticated-orcid":false,"given":"Shirin","family":"Golchi","sequence":"additional","affiliation":[{"name":"McGill University Department of Epidemiology, Biostatistics and Occupational Health, , Montr\u00e9al, Qu\u00e9bec H3A 1G1 ,","place":["Canada"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,7,24]]},"reference":[{"key":"2025072411120336700_bib1","doi-asserted-by":"crossref","first-page":"Article 10","DOI":"10.2202\/1557-4679.1204","article-title":"Optimal dynamic regimes: presenting a case for predictive inference","volume":"6","author":"Arjas","year":"2010","journal-title":"The International Journal of Biostatistics"},{"key":"2025072411120336700_bib2","doi-asserted-by":"crossref","first-page":"371","DOI":"10.1002\/sim.1023","article-title":"Individual patient- versus group-level data meta-regressions for the investigation of treatment effect modifiers: ecological bias rears its ugly head","volume":"21","author":"Berlin","year":"2002","journal-title":"Statistics in Medicine"},{"key":"2025072411120336700_bib3","doi-asserted-by":"crossref","first-page":"298","DOI":"10.1093\/jrsssc\/qlad096","article-title":"Variable selection for individualised treatment rules with discrete outcomes","volume":"73","author":"Bian","year":"2023","journal-title":"Journal of the Royal Statistical Society Series C: Applied Statistics"},{"key":"2025072411120336700_bib4","doi-asserted-by":"crossref","first-page":"1610","DOI":"10.1080\/01621459.2012.716344","article-title":"Consistent high-dimensional Bayesian variable selection via penalized credible regions","volume":"107","author":"Bondell","year":"2012","journal-title":"Journal of the American Statistical Association"},{"key":"2025072411120336700_bib5","author":"Canadian Network for Observational Drug Effect Studies","year":"2025"},{"key":"2025072411120336700_bib6","doi-asserted-by":"crossref","first-page":"465","DOI":"10.1093\/biomet\/asq017","article-title":"The horseshoe estimator for sparse signals","volume":"97","author":"Carvalho","year":"2010","journal-title":"Biometrika"},{"key":"2025072411120336700_bib7","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4614-7428-9","volume-title":"Statistical Methods for Dynamic Treatment Regimes: Reinforcement Learning, Causal Inference, and Personalized Medicine","author":"Chakraborty","year":"2013"},{"key":"2025072411120336700_bib8","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1146\/annurev-statistics-022513-115553","article-title":"Dynamic treatment regimes","volume":"1","author":"Chakraborty","year":"2014","journal-title":"Annual Review of Statistics and Its Application"},{"key":"2025072411120336700_bib9","doi-asserted-by":"crossref","first-page":"1509","DOI":"10.1080\/01621459.2016.1148611","article-title":"Personalized dose finding using outcome weighted learning","volume":"111","author":"Chen","year":"2016","journal-title":"Journal of the American Statistical Association"},{"key":"2025072411120336700_bib10","doi-asserted-by":"crossref","first-page":"656","DOI":"10.1093\/aje\/kwn164","article-title":"Constructing inverse probability weights for marginal structural models","volume":"168","author":"Cole","year":"2008","journal-title":"American Journal of Epidemiology"},{"key":"2025072411120336700_bib11","doi-asserted-by":"crossref","first-page":"1627","DOI":"10.1002\/sim.9318","article-title":"Preserving data privacy when using multi-site data to estimate individualized treatment rules","volume":"41","author":"Danieli","year":"2022","journal-title":"Statistics in Medicine"},{"key":"2025072411120336700_bib12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/978-3-540-79228-4_1","article-title":"Differential privacy: a survey of results","volume-title":"Theory and Applications of Models of Computation","author":"Dwork","year":"2008"},{"key":"2025072411120336700_bib13","doi-asserted-by":"crossref","first-page":"600","DOI":"10.5694\/j.1326-5377.2000.tb124127.x","article-title":"Consensus guidelines for warfarin therapy: recommendations from the Australasian Society of Thrombosis and Haemostasis","volume":"172","author":"Gallus","year":"2000","journal-title":"Medical Journal of Australia"},{"key":"2025072411120336700_bib14","doi-asserted-by":"crossref","first-page":"515","DOI":"10.1214\/06-BA117A","article-title":"Prior distributions for variance parameters in hierarchical models","volume":"1","author":"Gelman","year":"2006","journal-title":"Bayesian Analysis"},{"key":"2025072411120336700_bib15","doi-asserted-by":"crossref","DOI":"10.1201\/b16018","volume-title":"Bayesian Data Analysis","author":"Gelman","year":"2013"},{"key":"2025072411120336700_bib16","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1214\/12-AOS968","article-title":"Q-learning with censored data","volume":"40","author":"Goldberg","year":"2012","journal-title":"Annals of Statistics"},{"key":"2025072411120336700_bib17","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1002\/sim.4780020219","article-title":"Tests for interaction in epidemiologic studies: a review and a study of power","volume":"2","author":"Greenland","year":"1983","journal-title":"Statistics in Medicine"},{"key":"2025072411120336700_bib18","doi-asserted-by":"crossref","first-page":"435","DOI":"10.1080\/01621459.2014.993077","article-title":"Decoupling shrinkage and selection in Bayesian linear models: a posterior summary perspective","volume":"110","author":"Hahn","year":"2015","journal-title":"Journal of the American Statistical Association"},{"key":"2025072411120336700_bib19","doi-asserted-by":"crossref","DOI":"10.1002\/9781118445112.stat02818","article-title":"Asymptotic Normality","volume-title":"Wiley StatsRef: Statistics Reference Online","author":"Hoeffding","year":"2014"},{"key":"2025072411120336700_bib20","doi-asserted-by":"crossref","first-page":"753","DOI":"10.1056\/NEJMoa0809329","article-title":"Estimation of the Warfarin dose with clinical and pharmacogenetic data","volume":"360","author":"International Warfarin Pharmacogenetics Consortium","year":"2009","journal-title":"New England Journal of Medicine"},{"key":"2025072411120336700_bib21","first-page":"1225","article-title":"Dynamic treatment regimes: technical challenges and applications","volume":"8","author":"Laber","year":"2014","journal-title":"Electronic Journal of Statistics"},{"key":"2025072411120336700_bib22","doi-asserted-by":"crossref","DOI":"10.1002\/9781118445112.stat05836","article-title":"Consistent estimator","volume-title":"Wiley StatsRef: Statistics Reference Online","author":"Liang","year":"2014"},{"key":"2025072411120336700_bib23","doi-asserted-by":"crossref","DOI":"10.1002\/9781118445112.stat05980","article-title":"Unbiasedness","volume-title":"Wiley StatsRef: Statistics Reference Online","author":"Link","year":"2014"},{"key":"2025072411120336700_bib24","doi-asserted-by":"crossref","first-page":"1079","DOI":"10.1177\/0962280217746191","article-title":"Decision making and uncertainty quantification for individualized treatments using Bayesian Additive Regression Trees","volume":"28","author":"Logan","year":"2019","journal-title":"Statistical Methods in Medical Research"},{"key":"2025072411120336700_bib25","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s10985-022-09554-8","article-title":"Privacy-preserving estimation of an optimal individualized treatment rule: a case study in maximizing time to severe depression-related outcomes","volume":"28","author":"Moodie","year":"2022","journal-title":"Lifetime Data Analysis"},{"key":"2025072411120336700_bib26","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1007\/s12561-013-9103-z","article-title":"Q-Learning: flexible learning about useful utilities","volume":"6","author":"Moodie","year":"2014","journal-title":"Statistics in Biosciences"},{"key":"2025072411120336700_bib27","doi-asserted-by":"crossref","first-page":"1255","DOI":"10.1080\/01621459.2017.1340887","article-title":"A Bayesian machine learning approach for optimizing dynamic treatment regimes","volume":"113","author":"Murray","year":"2018","journal-title":"Journal of the American Statistical Association"},{"key":"2025072411120336700_bib28","article-title":"Evaluating strategies for data sharing and analyses in distributed data settings","author":"Rassen","year":"2013"},{"key":"2025072411120336700_bib29","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1124\/mi.6.4.8","article-title":"The pharmocogenomics of Warfarin: closing in on personalized medicine","volume":"6","author":"Rettie","year":"2006","journal-title":"Molecular Interventions"},{"key":"2025072411120336700_bib30","doi-asserted-by":"crossref","DOI":"10.1002\/9781119333784","volume-title":"Individual Participant Data Meta-Analysis: A Handbook for Healthcare Research","author":"Riley","year":"2021"},{"key":"2025072411120336700_bib31","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1007\/978-1-4612-1842-5_4","article-title":"Causal inference from complex longitudinal data","volume-title":"Latent Variable Modeling and Applications to Causality","author":"Robins","year":"1997"},{"key":"2025072411120336700_bib32","doi-asserted-by":"crossref","first-page":"189","DOI":"10.1007\/978-1-4419-9076-1_11","article-title":"Optimal structural nested models for optimal sequential decisions","volume-title":"Proceedings of the Second Seattle Symposium in Biostatistics","author":"Robins","year":"2004"},{"key":"2025072411120336700_bib33","first-page":"591","article-title":"Discussion of \u201cRandomization analysis of experimental data in the Fisher randomization test\u201d by D. Basu","volume":"75","author":"Rubin","year":"1980","journal-title":"Journal of the American Statistical Association"},{"key":"2025072411120336700_bib34","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1111\/rssb.12273","article-title":"Maximin projection learning for optimal treatment decision with heterogeneous individualized treatment effects","volume":"80","author":"Shi","year":"2018","journal-title":"Journal of the Royal Statistical Society: Series B (Statistical Methodology)"},{"key":"2025072411120336700_bib35","doi-asserted-by":"crossref","first-page":"2982","DOI":"10.1002\/sim.2768","article-title":"Covariate heterogeneity in meta-analysis: criteria for deciding between meta-regression and individual patient data","volume":"26","author":"Simmonds","year":"2007","journal-title":"Statistics in Medicine"},{"key":"2025072411120336700_bib36","first-page":"1531","article-title":"Estimating optimal dynamic treatment regimes with survival outcomes","volume-title":"Journal of the American Statistical Association","author":"Simoneau","year":"2020"},{"key":"2025072411120336700_bib37","article-title":"RStan: the R interface to Stan","author":"Stan Development Team","year":"2020"},{"key":"2025072411120336700_bib38","article-title":"Stan Reference Manual","author":"Stan Development Team","year":"2021"},{"key":"2025072411120336700_bib39","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton","year":"2018"},{"key":"2025072411120336700_bib40","doi-asserted-by":"crossref","first-page":"171","DOI":"10.1093\/biomet\/asp062","article-title":"On doubly robust estimation in a semiparametric odds ratio model","volume":"97","author":"Tchetgen\u00a0Tchetgen","year":"2010","journal-title":"Biometrika"},{"key":"2025072411120336700_bib41","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1016\/j.jmp.2018.12.004","article-title":"Shrinkage priors for Bayesian penalized regression","volume":"89","author":"Van\u00a0Erp","year":"2019","journal-title":"Journal of Mathematical Psychology"},{"key":"2025072411120336700_bib42","author":"Vivli","year":"2025"},{"key":"2025072411120336700_bib43","doi-asserted-by":"crossref","first-page":"636","DOI":"10.1111\/biom.12306","article-title":"Doubly-robust dynamic treatment regimen estimation via weighted least squares","volume":"71","author":"Wallace","year":"2015","journal-title":"Biometrics"},{"key":"2025072411120336700_bib44","doi-asserted-by":"crossref","first-page":"1205","DOI":"10.1111\/biom.13104","article-title":"Model selection for G-estimation of dynamic treatment regimes","volume":"75","author":"Wallace","year":"2019","journal-title":"Biometrics"},{"key":"2025072411120336700_bib45","volume-title":"Learning from delayed rewards","author":"Watkins","year":"1989"}],"container-title":["Biometrics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/biometrics\/article-pdf\/81\/3\/ujaf082\/63841041\/ujaf082.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biometrics\/article-pdf\/81\/3\/ujaf082\/63841041\/ujaf082.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T15:12:15Z","timestamp":1753369935000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biometrics\/article\/doi\/10.1093\/biomtc\/ujaf082\/8211795"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,3]]},"references-count":45,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,7,3]]}},"URL":"https:\/\/doi.org\/10.1093\/biomtc\/ujaf082","relation":{},"ISSN":["0006-341X","1541-0420"],"issn-type":[{"value":"0006-341X","type":"print"},{"value":"1541-0420","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,9]]},"published":{"date-parts":[[2025,7,3]]},"article-number":"ujaf082"}}