{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T20:03:25Z","timestamp":1792785805568,"version":"4.5.0"},"reference-count":32,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T00:00:00Z","timestamp":1785196800000},"content-version":"vor","delay-in-days":208,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"European Union\u2014Next Generation EU","id":[{"id":"https:\/\/ror.org\/019w4f821","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,20]]},"abstract":"<jats:title>Summary<\/jats:title>\n                  <jats:p>Quantifying the causal effect of a treatment on a disease is a crucial task in medical science for the administration of effective therapies. Typically, such causal effects are inferred from multivariate data that are collected on patients and recorded in the form of categorical variables, including risk factors involved in disease progression, treatment assignments, and disease status. This feature motivates an approach to causal inference based on categorical Directed Acyclic Graphs (DAGs), which provide an effective framework for causal reasoning in complex multivariate settings. In this context, traditional DAG-based methods assume population homogeneity and accordingly attribute a unique causal effect to all subjects. However, this assumption is often unrealistic in clinical contexts, since patients may exhibit heterogeneous characteristics, possibly linked to unmeasured features. To address this issue, we propose a Bayesian nonparametric methodology based on a Dirichlet Process mixture of categorical DAGs, which allows treatment effects to vary across individuals because of underlying clustering structures in the data. We develop a Markov chain Monte Carlo algorithm for Bayesian posterior inference and evaluate our methodology through simulation studies. We then analyze patients affected by HER2+ breast cancer undergoing therapies that may cause cardiotoxic side effects. Importantly, our findings show that approaches neglecting population heterogeneity may produce biased results, since they can over- or under-estimate the risk of cardiotoxicity across patients.<\/jats:p>","DOI":"10.1093\/biostatistics\/kxag026","type":"journal-article","created":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T04:52:39Z","timestamp":1785214359000},"source":"Crossref","is-referenced-by-count":0,"title":["Bayesian nonparametric mixtures of categorical directed graphs for personalized causal inference"],"prefix":"10.1093","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7911-2942","authenticated-orcid":false,"given":"Federico","family":"Castelletti","sequence":"first","affiliation":[{"name":"Department of Statistical Sciences, Universit\u00e0 Cattolica del Sacro Cuore , Largo Gemelli, 1 ,Milan 20123,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Laura","family":"Ferrini","sequence":"additional","affiliation":[{"name":"Department of Economics, Management and Statistics, Universit\u00e0 degli Studi di Milano-Bicocca , Piazza dell\u2019Ateneo Nuovo, 1 , Milan 20126,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,7,28]]},"reference":[{"key":"2026102110232538200_kxag026-B1","doi-asserted-by":"publisher","first-page":"1293","DOI":"10.1093\/jnci\/djs317","article-title":"Risk of heart failure in breast cancer patients after Anthracycline and Trastuzumab treatment: a retrospective cohort study","volume":"104","author":"Bowles","year":"2012","journal-title":"J Natl Cancer Inst"},{"key":"2026102110232538200_kxag026-B2","doi-asserted-by":"publisher","first-page":"658","DOI":"10.1007\/s11336-024-09969-2","article-title":"Learning Bayesian networks: a copula approach for mixed-type data","volume":"89","author":"Castelletti","year":"2024","journal-title":"Psychometrika"},{"key":"2026102110232538200_kxag026-B3","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1002\/sim.9599","article-title":"Bayesian graphical modeling for heterogeneous causal effects","volume":"42","author":"Castelletti","year":"2023","journal-title":"Stat Med"},{"key":"2026102110232538200_kxag026-B4","doi-asserted-by":"publisher","first-page":"ujae067","DOI":"10.1093\/biomtc\/ujae067","article-title":"Joint structure learning and causal effect estimation for categorical graphical models","volume":"80","author":"Castelletti","year":"2024","journal-title":"Biometrics"},{"key":"2026102110232538200_kxag026-B5","doi-asserted-by":"publisher","first-page":"1289","DOI":"10.1007\/s10260-021-00579-1","article-title":"Structural learning and estimation of joint causal effects among network-dependent variables","volume":"30","author":"Castelletti","year":"2021","journal-title":"Stat Methods Appl"},{"key":"2026102110232538200_kxag026-B6","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1016\/j.ejca.2023.02.019","article-title":"Anthracycline-induced cardiotoxicity \u2013 are we about to clear this hurdle?","volume":"185","author":"Dempke","year":"2023","journal-title":"Eur J Cancer"},{"key":"2026102110232538200_kxag026-B7","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1007\/s10549-021-06280-x","article-title":"Trastuzumab-induced cardiotoxicity: a review of clinical risk factors, pharmacologic prevention, and cardiotoxicity of other HER2-directed therapies","volume":"188","author":"Dempsey","year":"2021","journal-title":"Breast Cancer Res Treat"},{"key":"2026102110232538200_kxag026-B8","doi-asserted-by":"publisher","DOI":"10.1162\/99608f92.8102afed","article-title":"From controlled to undisciplined data: estimating causal effects in the era of data science using a potential outcome framework","volume":"3","author":"Dominici","year":"2021","journal-title":"Harv Data Sci Rev"},{"key":"2026102110232538200_kxag026-B9","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1080\/01621459.1995.10476550","article-title":"Bayesian density estimation and inference using mixtures","volume":"90","author":"Escobar","year":"1995","journal-title":"J Am Stat Assoc"},{"key":"2026102110232538200_kxag026-B10","doi-asserted-by":"publisher","first-page":"209","DOI":"10.1214\/aos\/1176342360","article-title":"A Bayesian analysis of some nonparametric problems","volume":"1","author":"Ferguson","year":"1973","journal-title":"Ann Statist"},{"key":"2026102110232538200_kxag026-B11","doi-asserted-by":"publisher","first-page":"1344","DOI":"10.1214\/aos\/1069362752","article-title":"A characterization of the Dirichlet distribution through global and local parameter independence","volume":"25","author":"Geiger","year":"1997","journal-title":"Ann Statist"},{"key":"2026102110232538200_kxag026-B12","doi-asserted-by":"publisher","first-page":"965","DOI":"10.1214\/19-BA1195","article-title":"Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion)","volume":"15","author":"Hahn","year":"2020","journal-title":"Bayesian Anal"},{"key":"2026102110232538200_kxag026-B13","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1023\/A:1022623210503","article-title":"Learning Bayesian networks: the combination of knowledge and statistical data","volume":"20","author":"Heckerman","year":"1995","journal-title":"Mach Learn"},{"key":"2026102110232538200_kxag026-B14","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1023\/A:1009769707641","article-title":"Extensions to the k-means algorithm for clustering large data sets with categorical values","volume":"2","author":"Huang","year":"1998","journal-title":"Data Min Knowl Discov"},{"key":"2026102110232538200_kxag026-B15","doi-asserted-by":"publisher","first-page":"600","DOI":"10.1200\/jco.2013.31.15_suppl.640","article-title":"Prognostic value of HER2 on breast cancer survival","volume":"31","author":"Katzorke","year":"2013","journal-title":"J Clin Oncol"},{"key":"2026102110232538200_kxag026-B16","author":"Kvisgaard","year":"2024"},{"key":"2026102110232538200_kxag026-B17","doi-asserted-by":"publisher","author":"Lee","year":"2021","DOI":"10.48550\/arXiv.2009.09036"},{"key":"2026102110232538200_kxag026-B18","doi-asserted-by":"publisher","first-page":"e1583","DOI":"10.1002\/wics.1583","article-title":"The how and why of Bayesian nonparametric causal inference","volume":"15","author":"Linero","year":"2022","journal-title":"Wiley Interdiscip Rev Comput Stat"},{"key":"2026102110232538200_kxag026-B19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v042.i10","article-title":"poLCA: an R package for polytomous variable latent class analysis","volume":"42","author":"Linzer","year":"2011","journal-title":"J Stat Soft"},{"key":"2026102110232538200_kxag026-B20","doi-asserted-by":"publisher","first-page":"1980","DOI":"10.1016\/j.amjcard.2013.08.026","article-title":"Review and meta-analysis of incidence and clinical predictors of Anthracycline cardiotoxicity","volume":"112","author":"Lotrionte","year":"2013","journal-title":"Am J Cardiol"},{"key":"2026102110232538200_kxag026-B21","doi-asserted-by":"publisher","first-page":"670691","DOI":"10.1155\/2015\/670691","article-title":"Statistical methods for establishing personalized treatment rules in oncology","volume":"2015","author":"Ma","year":"2015","journal-title":"Biomed Res Int"},{"key":"2026102110232538200_kxag026-B22","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1214\/cbms\/1362163748","volume-title":"Nonparametric Bayesian inference","author":"M\u00fcller","year":"2013"},{"key":"2026102110232538200_kxag026-B23","doi-asserted-by":"publisher","first-page":"647","DOI":"10.1214\/16-AOS1462","article-title":"Estimating the effect of joint interventions from observational data in sparse high-dimensional settings","volume":"45","author":"Nandy","year":"2017","journal-title":"Ann Statist"},{"key":"2026102110232538200_kxag026-B24","doi-asserted-by":"publisher","first-page":"249","DOI":"10.1080\/10618600.2000.10474879","article-title":"Markov chain sampling methods for Dirichlet process mixture models","volume":"9","author":"Neal","year":"2000","journal-title":"J Comput Graph Stat"},{"key":"2026102110232538200_kxag026-B25","volume-title":"Causality: Models, reasoning, and inference","author":"Pearl","year":"2000"},{"key":"2026102110232538200_kxag026-B26","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1007\/BF02595718","article-title":"Statistics and causal inference: a review","volume":"12","author":"Pearl","year":"2003","journal-title":"Test"},{"key":"2026102110232538200_kxag026-B27","doi-asserted-by":"publisher","first-page":"527","DOI":"10.1038\/s41597-023-02419-1","article-title":"A cardiotoxicity dataset for breast cancer patients","volume":"10","author":"Pi\u00f1eiro-Lamas","year":"2023","journal-title":"Sci Data"},{"key":"2026102110232538200_kxag026-B28","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1080\/01621459.2017.1390466","article-title":"Functional graphical models","volume":"114","author":"Qiao","year":"2019","journal-title":"J Am Stat Assoc"},{"key":"2026102110232538200_kxag026-B29","doi-asserted-by":"publisher","first-page":"322","DOI":"10.1198\/016214504000001880","article-title":"Causal inference using potential outcomes: design, modeling, decisions","volume":"100","author":"Rubin","year":"2005","journal-title":"J Am Stat Assoc"},{"key":"2026102110232538200_kxag026-B30","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1126\/science.3798106","article-title":"Human breast cancer: correlation of relapse and survival with amplification of the HER-2\/neu oncogene","volume":"235","author":"Slamon","year":"1987","journal-title":"Science"},{"key":"2026102110232538200_kxag026-B31","doi-asserted-by":"publisher","first-page":"559","DOI":"10.1214\/17-BA1073","article-title":"Bayesian cluster analysis: point estimation and credible balls (with discussion)","volume":"13","author":"Wade","year":"2018","journal-title":"Bayesian Anal"},{"key":"2026102110232538200_kxag026-B32","doi-asserted-by":"publisher","first-page":"ujae025","DOI":"10.1093\/biomtc\/ujae025","article-title":"Confounder-dependent Bayesian mixture model: characterizing heterogeneity of causal effects in air pollution epidemiology","volume":"80","author":"Zorzetto","year":"2024","journal-title":"Biometrics"}],"container-title":["Biostatistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/biostatistics\/article-pdf\/27\/1\/kxag026\/70230065\/kxag026.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biostatistics\/article-pdf\/27\/1\/kxag026\/70230065\/kxag026.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,21]],"date-time":"2026-10-21T14:23:33Z","timestamp":1792592613000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biostatistics\/article\/doi\/10.1093\/biostatistics\/kxag026\/8743782"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":32,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1,20]]}},"URL":"https:\/\/doi.org\/10.1093\/biostatistics\/kxag026","relation":{},"ISSN":["1465-4644","1468-4357"],"issn-type":[{"value":"1465-4644","type":"print"},{"value":"1468-4357","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,1]]},"published":{"date-parts":[[2026,1]]},"article-number":"kxag026"}}