Preprint
Sep 2026
RECaST-Surv: A Calibrated Borrowing Method for Survival Endpoints in Unequal Randomized Trials
RECaST-Surv is proposed, a Bayesian transfer-learning framework for time-to-event outcomes that extends the RECaST method to survival settings and maintains near-nominal Type I error across challenging settings while improving power by roughly 10%--12% over standard analyses.
De-Hua Bi, Arlina Shen, R. V. van Eijk et al.
· 0 citations