{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T18:31:52Z","timestamp":1770748312831,"version":"3.49.0"},"reference-count":37,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T00:00:00Z","timestamp":1770681600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Public Health"],"abstract":"\n Background<\/jats:title>\n In real-world medical data, selection bias can significantly impact the performance of machine learning models, potentially leading to distorted outcomes. However, research aimed at mitigating selection bias remains relatively limited.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n In this study, we evaluate the effectiveness of Propensity Score Matching (PSM) in reducing selection bias and assessing its impact on classification performance in imbalanced medical data. Specifically, we apply PSM alongside five undersampling, three oversampling, and three hybrid sampling techniques to three medical datasets: rapidly progressive dementia prediction (ADNI,\n n<\/jats:italic>\n = 628, events = 51), hypothyroidism prediction (UCI,\n n<\/jats:italic>\n = 3,772, events = 3,481), and cardiovascular disease prediction (Kaggle,\n n<\/jats:italic>\n = 253,680, events = 23,893), each exhibiting varying degrees of demographic selection bias. We train and compare six classification models to assess the impact of each resampling technique on model performance. The magnitude of selection bias is quantified using the standardized mean difference (SMD), while model performance is assessed using the Area Under the Receiver Operating Characteristic Curve (AUROC), the Area Under the Precision-Recall Curve (AUPRC), accuracy, precision, recall, F1-score, specificity, calibration curves, Brier score, and decision curve analysis.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n The results indicate that PSM reduces SMD within the dataset, maintains stable classification performance, and enhances the internal validity of the model under conditions of limited or moderate demographic imbalance.<\/jats:p>\n <\/jats:sec>\n \n Conclusion<\/jats:title>\n These advantages suggest its potential for improving model reliability and facilitating better generalization to external datasets in real-world medical applications. However, in datasets with extreme selection bias or when overly restrictive matching is applied, PSM can degrade model performance, underscoring the importance of choosing strategies that account for dataset characteristics.<\/jats:p>\n <\/jats:sec>","DOI":"10.3389\/fpubh.2026.1747762","type":"journal-article","created":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T07:20:41Z","timestamp":1770708041000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Evaluating the sampling effect of propensity score matching for reducing selection bias in medical data"],"prefix":"10.3389","volume":"14","author":[{"given":"Minji","family":"Roh","sequence":"first","affiliation":[{"name":"Interdisciplinary Graduate Program in Medical Bigdata Convergence, Kangwon National University","place":["Chuncheon, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sujin","family":"Yum","sequence":"additional","affiliation":[{"name":"Interdisciplinary Graduate Program in Medical Bigdata Convergence, Kangwon National University","place":["Chuncheon, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gihun","family":"Joo","sequence":"additional","affiliation":[{"name":"Kangwon National University","place":["Chuncheon, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jae-Won","family":"Jang","sequence":"additional","affiliation":[{"name":"Department of Neurology, Kangwon National University Hospital","place":["Chuncheon, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hyeonseung","family":"Im","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Kangwon National University","place":["Chuncheon, Republic of Korea"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,2,10]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"761","DOI":"10.1093\/ehjqcco\/qcab070","article-title":"Generalizability of randomized controlled trials in heart failure with reduced ejection fraction","volume":"8","author":"Lim","year":"2022","journal-title":"Eur Heart J Qual Care Clin Outcomes"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1393","DOI":"10.1016\/j.cmi.2023.08.019","article-title":"Are we enrolling the right patients? A scoping review of external validity and generalizability of clinical trials in bloodstream infections","volume":"29","author":"Ong","year":"2023","journal-title":"Clin Microbiol Infect"},{"key":"B3","doi-asserted-by":"publisher","first-page":"e2128045","DOI":"10.1001\/jamanetworkopen.2021.28045","article-title":"Strategies to turn real-world data into real-world knowledge","volume":"4","author":"Hong","year":"2021","journal-title":"JAMA Netw Open"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1680","DOI":"10.1056\/NEJMp2200089","article-title":"Real-world evidence\u2014where are we now?","volume":"386","author":"Concato","year":"2022","journal-title":"N Engl J Med"},{"key":"B5","doi-asserted-by":"publisher","first-page":"306","DOI":"10.1136\/ard-2022-222626","article-title":"From real-world electronic health record data to real-world results using artificial intelligence","volume":"82","author":"Knevel","year":"2023","journal-title":"Ann Rheum Dis"},{"key":"B6","doi-asserted-by":"publisher","first-page":"751","DOI":"10.1002\/cpt.2988","article-title":"Clinical pharmacology applications of real-world data and real-world evidence in drug development and approval-an industry perspective","volume":"114","author":"Zhu","year":"2023","journal-title":"Clin Pharmacol Ther"},{"key":"B7","doi-asserted-by":"publisher","first-page":"287","DOI":"10.3322\/caac.21714","article-title":"An overview of real-world data sources for oncology and considerations for research","volume":"72","author":"Penberthy","year":"2022","journal-title":"CA Cancer J Clin"},{"key":"B8","unstructured":"Hern\u00e1n\n MA\n \n \n Robins\n JM\n \n \n Causal Inference: What If\n \n 2024"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3761822","article-title":"Sample selection bias in machine learning for healthcare","volume":"6","author":"Chauhan","year":"2025","journal-title":"ACM Trans Comput Healthc"},{"key":"B10","doi-asserted-by":"publisher","first-page":"04016134","DOI":"10.1061\/(ASCE)GM.1943-5622.0000808","article-title":"Analysis of the influences of sampling bias and class imbalance on performances of probabilistic liquefaction models","volume":"17","author":"Hu","year":"2017","journal-title":"Int J Geomech"},{"key":"B11","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1016\/j.ins.2019.08.062","article-title":"Neighbourhood-based undersampling approach for handling imbalanced and overlapped data","volume":"509","author":"Vuttipittayamongkol","year":"2020","journal-title":"Inf Sci"},{"key":"B12","doi-asserted-by":"publisher","first-page":"2897","DOI":"10.3390\/cancers14122897","article-title":"Bias and class imbalance in oncologic data\u2014towards inclusive and transferrable AI in large scale oncology data sets","volume":"14","author":"Tasci","year":"2022","journal-title":"Cancers"},{"key":"B13","unstructured":"Learning pattern classification tasks with imbalanced data sets\n \n 193\n 208\n \n \n Nguyen\n GH\n \n \n Bouzerdoum\n A\n \n \n Phung\n SL\n \n \n \n \n Yin\n P\n \n \n Vukovar\n In-Teh\n Pattern Recognition\n \n 2009"},{"key":"B14","doi-asserted-by":"publisher","first-page":"109960","DOI":"10.1109\/ACCESS.2021.3102399","article-title":"A comparative performance analysis of data resampling methods on imbalance medical data","volume":"9","author":"Khushi","year":"2021","journal-title":"IEEE Access"},{"key":"B15","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1093\/biomet\/70.1.41","article-title":"The central role of the propensity score in observational studies for causal effects","volume":"70","author":"Rosenbaum","year":"1983","journal-title":"Biometrika"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1112","DOI":"10.1093\/ejcts\/ezy167","article-title":"Statistical primer: propensity score matching and its alternatives","volume":"53","author":"Benedetto","year":"2018","journal-title":"Eur J Cardiothorac Surg"},{"key":"B17","first-page":"131","article-title":"Using propensity score matching to resolve the class imbalance problem in medical data","volume-title":"Proceedings of the Korean Institute of Information Scientists and Engineers (KIISE)","author":"Roh","year":"2023"},{"key":"B18","doi-asserted-by":"publisher","first-page":"478","DOI":"10.3988\/jcn.2023.0289","article-title":"Predicting the progression of mild cognitive impairment to Alzheimer's dementia using recurrent neural networks with a series of neuropsychological tests","volume":"20","author":"Park","year":"2024","journal-title":"J Clin Neurol"},{"key":"B19","doi-asserted-by":"publisher","first-page":"3637","DOI":"10.1109\/JBHI.2024.3368500","article-title":"MAD-former: a traceable interpretability model for Alzheimer's disease recognition based on multi-patch attention","volume":"28","author":"Ye","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"B20","doi-asserted-by":"publisher","first-page":"6451","DOI":"10.1109\/JBHI.2024.3473541","article-title":"Enhancing early Alzheimer's disease detection through big data and ensemble few-shot learning","volume":"29","author":"Atitallah","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"B21","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1038\/s41582-022-00659-0","article-title":"Rapidly progressive dementias\u2014aetiologies, diagnosis and management","volume":"18","author":"Hermann","year":"2022","journal-title":"Nat Rev Neurol"},{"key":"B22","unstructured":"Data from: Thyroid disease\n \n \n \n Quinlan\n R\n \n \n UCI Machine Learning Repository\n \n 1986"},{"key":"B23","first-page":"605","article-title":"Hypothyroidism: diagnosis and treatment","volume":"103","author":"Wilson","year":"2021","journal-title":"Am Fam Physician"},{"key":"B24","unstructured":"Data from: Heart disease health indicators\n \n \n \n Teboul\n A\n \n \n Kaggle\n \n 2022"},{"key":"B25","first-page":"424","article-title":"OUPS: a combined approach using SMOTE and propensity score matching","volume-title":"2014 13th IEEE International Conference on Machine Learning and Applications","author":"Rivera","year":"2014"},{"key":"B26","doi-asserted-by":"publisher","first-page":"193","DOI":"10.3233\/ICA-2009-0314","article-title":"Hybrid sampling for imbalanced data","volume":"16","author":"Seiffert","year":"2009","journal-title":"Integr Comput Aided Eng"},{"key":"B27","doi-asserted-by":"publisher","first-page":"11349","DOI":"10.4088\/JCP.20f13681","article-title":"Mean difference, standardized mean difference (SMD), and their use in meta-analysis: as simple as it gets","volume":"81","author":"Andrade","year":"2020","journal-title":"J Clin Psychiatry"},{"key":"B28","doi-asserted-by":"publisher","first-page":"399","DOI":"10.1080\/00273171.2011.568786","article-title":"An introduction to propensity score methods for reducing the effects of confounding in observational studies","volume":"46","author":"Austin","year":"2011","journal-title":"Multivar Behav Res"},{"key":"B29","unstructured":"Chapter 10\n \n 531\n 42\n \n \n Cohen\n J\n \n \n Some Issues in Power Analysis\n \n 1988"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1282","DOI":"10.1001\/jama.2022.1820","article-title":"Collider bias","volume":"327","author":"Holmberg","year":"2022","journal-title":"JAMA"},{"key":"B31","doi-asserted-by":"publisher","first-page":"q902","DOI":"10.1136\/bmj.q902","article-title":"TRIPOD+ AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1124","DOI":"10.1001\/archneurol.2011.189","article-title":"Rapidly progressive Alzheimer disease","volume":"68","author":"Schmidt","year":"2011","journal-title":"Swiss Arch Neurol Psychiatry Psychother"},{"key":"B33","doi-asserted-by":"publisher","first-page":"293","DOI":"10.30773\/pi.2022.0188","article-title":"Risk factors for various cognitive function decline trajectories in adults over 40 years of age: a retrospective cohort study","volume":"20","author":"Han","year":"2023","journal-title":"Psychiatry Investig"},{"key":"B34","doi-asserted-by":"publisher","first-page":"bvac172","DOI":"10.1210\/jendso\/bvac172","article-title":"Hypothyroidism prevalence in the United States: a retrospective study combining national health and nutrition examination survey and claims data, 2009\u20132019","volume":"7","author":"Wyne","year":"2023","journal-title":"J Endocr Soc"},{"key":"B35","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1038\/s41569-021-00607-3","article-title":"Epidemiology of cardiovascular disease in Europe","volume":"19","author":"Townsend","year":"2022","journal-title":"Nat Rev Cardiol"},{"key":"B36","doi-asserted-by":"publisher","first-page":"459","DOI":"10.1001\/jamacardio.2021.5613","article-title":"Prevalence and trends of coronary heart disease in the United States, 2011 to 2018","volume":"7","author":"Lee","year":"2022","journal-title":"JAMA Cardiol"},{"key":"B37","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.jacasi.2021.04.007","article-title":"Epidemiological features of cardiovascular disease in Asia","volume":"1","author":"Zhao","year":"2021","journal-title":"JACC Asia"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1747762\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T07:20:47Z","timestamp":1770708047000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1747762\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,10]]},"references-count":37,"alternative-id":["10.3389\/fpubh.2026.1747762"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1747762","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,10]]},"article-number":"1747762"}}