[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120653-en":3,"doc-seo-120653-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120653,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","Emulate Randomized Clinical Trials using Heterogeneous Treatment Effect Estimation for Personalized Treatments - Methodology Review and Benchmark","Big data and (deep) machine learning enable digital medicine, yet they largely prioritize associations rather than causal intervention effects. Treatment effects can differ across patients, making heterogeneous treatment effect (HTE) estimation central for personalized treatment development. This work reviews and compares eleven recent HTE estimation methodologies, then conducts a comprehensive benchmark using nationwide healthcare claims data. The benchmark applies methods to Alzheimer’s disease drug repurposing, addressing key challenges and opportunities for translating advanced HTE models into real-world healthcare deployment.","Emulate Randomized Clinical Trials using Heterogeneous Treatment Effect Estimation for Personalized Treatments: Methodology Review and Benchmark  \nYaobin Ling M.S., Pulakesh Upadhyaya Ph.D., Luyao Chen M.S., Xiaoqian Jiang Ph.D., Yejin Kim Ph.D.  \nSchool of Biomedical Informatics, University of Texas Health Science Center at Houston, Fannin 7000, Houston, Texas  \nTelephone number: 713 500 3998  \nFax number: 713 500 0360  \n[E-mail address: ](E-mail address: yejin.kim@uth.tmc.edu)[yejin.kim@uth.tmc.edu](E-mail address: yejin.kim@uth.tmc.edu)  \nABSTRACT  \nBig data and (deep) machine learning have been ambitious tools in digital medicine, but these tools focus mainly on association. Intervention in medicine is about the causal effects. The average treatment effect has long been studied as a measure of causal effect, assuming that all populations have the same effect size. However, no “one-size-fits-all” treatment seems to work in some complex diseases. Treatment effects may vary by patient. Estimating heterogeneous treatment effects (HTE) may have a high impact on developing personalized treatment. Lots of advanced machine learning models for estimating HTE have emerged in recent years, but there has been limited translational research into the real-world healthcare domain. To fill the gap, wereviewed and compared eleven recent HTE estimation methodologies, including meta-learner, representation learning models, and tree-based models. We performed a comprehensive benchmark experiment based on nationwide healthcare claim data with application to Alzheimer’s disease drug repurposing. We provided some challenges and opportunities in HTE estimation analysis in the healthcare domain to close the gap between innovative HTE models and deployment to real-world healthcare problems.  \nKeywords  \nCausal inference, Target trial, Conditional average treatment effect, Drug development, Deep learning, Machine learning  \n1. Introduction  \nCausal inference discovers a cause of an effect. Although randomized experiments (e.g., randomized clinical trials, A/B test) are a de facto gold standard to identify causation, they are sometimes economically infeasible or unethical if intervention harms subjects [1] . The treatment effect estimation using observational data (e.g., real-world data) is an alternative strategy to emulate the randomized experiments and infer the causation. However, observation inevitably contains bias. A confounding variable is a variable that influences exposure to the treatment and outcomes(Fig. 1A) . It is one of the major sources of bias that can mislead us to draw a wrong conclusion that the treatment has effects on the outcome when it does not. [2] A statistical approach to reducing such bias in observational data for treatment effect estimation has long been studied in multiple disciplines. For example, a target trial framework in epidemiology and biostatistics has been focused on hypothesis testing to infer the average treatment effect by adjusting the confounders via matching or weighting [3–12](Fig. 1Bc, 1Bd)  \nFigure 1. Illustrations of treatment effects analysis in the medical science field. A. An example of a causal relationship. B. Estimating causal treatment effects from real-world data under the Neyman-Rubin framework. (a) Subjects in RCTs are randomly assigned to a treatment group anda control group, thus the subjects in both groups have similar characteristics. (b) Subjects in realworld data are not randomly assigned to a treatment group and control group due to disease indication. (c) Matching subjects in each group can reduce bias [13] . (d) Weighting subjects by  \ntheir propensity for treatment can create a comparable pseudo population [14,15](Details described in S.1.1) . (e) Neyman-Rubin causal effect calculation. C. Heterogeneous treatment effects vs. Average treatment effect. Patients are diverse and treatment effects vary. Estimating the average treatment effect (ATE) may oversimplify the heterogeneity of","cbCaimPtVUuMuHfR","https://ap.wps.com/l/cbCaimPtVUuMuHfR","pdf",919403,1,37,"English","en",105,"# Introduction\n## Causal inference and bias in observational data\n## Target trial framework and ATE vs. HTE\n# Preliminaries\n## Potential outcome framework","[{\"question\":\"Why is heterogeneous treatment effect (HTE) important for personalized treatments?\",\"answer\":\"Patients differ in characteristics and their treatment effects vary, so average treatment effect (ATE) can oversimplify real outcomes. Estimating HTE supports identifying effects for individuals or similar subgroups.\"},{\"question\":\"How does the work emulate randomized clinical trials using observational data?\",\"answer\":\"It uses a target trial protocol under the Neyman-Rubin potential outcome framework, adjusting for confounding to infer causal treatment effects from real-world data.\"},{\"question\":\"What methodologies and benchmark setting does the review focus on?\",\"answer\":\"The review compares eleven HTE estimation methodologies, including meta-learner, representation learning models, and tree-based models. It benchmarks them using nationwide healthcare claims data and applies the approach to Alzheimer’s disease drug repurposing.\"}]","Emulate Randomized Clinical Trials using Heterogeneous Treatment Effect Estimation for Personalized Treatments - Methodology Review and Benchmark | PDF",1785731179,93,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"emulate-randomized-clinical-trials-using-heterogeneous-treatment-effect-estimation-for-personalized-treatments-methodology-review-and-benchmark","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/emulate-randomized-clinical-trials-using-heterogeneous-treatment-effect-estimation-for-personalized-treatments-methodology-review-and-benchmark/120653/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is heterogeneous treatment effect (HTE) important for personalized treatments?","Question",{"text":75,"@type":76},"Patients differ in characteristics and their treatment effects vary, so average treatment effect (ATE) can oversimplify real outcomes. Estimating HTE supports identifying effects for individuals or similar subgroups.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the work emulate randomized clinical trials using observational data?",{"text":80,"@type":76},"It uses a target trial protocol under the Neyman-Rubin potential outcome framework, adjusting for confounding to infer causal treatment effects from real-world data.",{"name":82,"@type":73,"acceptedAnswer":83},"What methodologies and benchmark setting does the review focus on?",{"text":84,"@type":76},"The review compares eleven HTE estimation methodologies, including meta-learner, representation learning models, and tree-based models. It benchmarks them using nationwide healthcare claims data and applies the approach to Alzheimer’s disease drug repurposing.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]