Generalized spatiotemporal modeling and causal inference for assessing treatment effects for multiple groups for ordinal outcome.

Hierarchical hurdle model;health professional shortage;superiority score;ordinal outcome;antibiotic overuse;generalized spatiotemporal model

Soutik Ghosal

University:University of Louisville

Department:Bioinformatics and Biostatistics

关键词: Hierarchical hurdle model;    health professional shortage;    superiority score;    ordinal outcome;    antibiotic overuse;    generalized spatiotemporal model;   

Others  :  https://ir.library.louisville.edu/cgi/viewcontent.cgi?article=4194&context=etd

美国|英语

来源: The Universite of Louisville's Institutional Repository

【 摘 要 】

This dissertation consists of three projects and can be categorized in two broad research areas: generalized spatiotemporal modeling and causal inference based on observational data. In the first project, I introduce a Bayesian hierarchical mixed effect hurdle model with a nested random effect structure to model the count for primary care providers and understand their spatial and temporal variation. This study further enables us to identify the health professional shortage areas and the possible impacting factors. In the second project, I have unified popular parametric and nonparametric propensity score-based methods to assess the treatment effect of multiple groups for ordinal outcome. I have conducted different simulation scenarios and compared the performance of those methods. In the third project, I have introduced a generalized spatiotemporal model to identify the antibiotic medication overuse in Kentucky. In this project, I used the Medicaid data to understand the spatial and seasonal variation of the antibiotic overuse for children insured by Kentucky Medicaid.