A1328
Title: Estimation of cluster-specific causal effects on spatially associated survival data using SoftBART
Authors: Indrabati Bhattacharya - Florida State University (United States) [presenting]
Durbadal Ghosh - Florida State University (United States)
George Rust - Florida State University (United States)
Debajyoti Sinha - Florida State University (United States)
Abstract: The aim is to propose a novel Bayesian approach to estimate causal effects in spatially clustered survival data. Using soft Bayesian additive regression trees (SBART), a nonparametric regression is introduced for a log-Normal survival model that accommodates spatial associations among unknown cluster effects through a directed acyclic graph autoregressive (DAGAR) model. A two-stage approach is employed, which entails estimating the propensity score in the first step and incorporating it as a confounder of the outcome model in the second step. In the simulation study, the method is compared with existing approaches under various simulation scenarios, including both correctly specified and misspecified outcome models, to demonstrate the superior performance of the method. The method is then applied to analyze the causal effect of treatment delay (TD) on post-treatment survival of breast cancer patients from the Florida Cancer registry (FCR). The analysis produces the county-specific as well as state-wide assessment of the causal effects while accommodating spatial association among counties.
