biostudies-literatureLi KMUS National Institutes of HealthPut VA Data to Work for VeteransUS Department of Veterans Affairs under the research priority to “Put VA Data to Work for Veterans”NHLBI NIH HHSUS Department of Veterans Affairs under the research priority to "Put VA Data to Work for Veterans"NIH HHSUS Department of Veterans Affairs409-423https://www.ebi.ac.uk/biostudies/studies/S-EPMC12844582biostudies-literatureUnknown33(2)<h4>Objective</h4>To develop a transfer-learning Bayesian sparse logistic regression model that transfers information learned from one dataset to another by using an informed prior to facilitate model fitting in small-sample clinical patient-level prediction problems that suffer from a lack of available information.<h4>Methods</h4>We propose a Bayesian framework for prediction using logistic regression that aims to conduct transfer-learning on regression coefficient information from a larger dataset model (order 105-106 patients by 105 features) into a small-sample model (order 103 patients). Our approach imposes an informed, hierarchical prior on each regression coefficient defined as a discrete mixture of the Bayesian Bridge shrinkage prior and an informed normal distribution. PerformancJournal of the American Medical Informatics Association : JAMIATransfer-learning on federated observational healthcare data for prediction models using Bayesian sparse logistic regression with informed priors.PMC12844582R01 HL16995422-D4VVA ORD 22-D4VSchuemie MJSuchard MALi KMReps JMNishimura AfalseTransfer-learning on federated observational healthcare data for prediction models using Bayesian sparse logistic regression with informed priors.<h4>Objective</h4>To develop a transfer-learning Bayesian sparse logistic regression model that transfers information learned from one dataset to another by using an informed prior to facilitate model fitting in small-sample clinical patient-level prediction problems that suffer from a lack of available information.<h4>Methods</h4>We propose a Bayesian framework for prediction using logistic regression that aims to conduct transfer-learning on regression coefficient information from a larger dataset model (order 105-106 patients by 105 features) into a small-sample model (order 103 patients). Our approach imposes an informed, hierarchical prior on each regression coefficient defined as a discrete mixture of the Bayesian Bridge shrinkage prior and an informed normal distribution. Performanc2026-01-01T00:00:00Z2026 Feb2026-06-09T03:18:35.616Z2026-06-09T03:11:50.374ZS-EPMC128445824114429810.1093/jamia/ocaf146