Workshops‎ > ‎

Workshop 2

Applying Propensity Score Methods in a Large Healthcare Data Analysis Context
Summary:

This workshop will focus on learning observational data analysis approaches in a healthcare data analysis context, particularly aiming in demonstrating the implementation of propensity score analysis in a real-world data analysis context through a hands-on data analysis exercise; explaining how these analyses are different than conventional regression methods, and understanding assumptions/diagnostics of these models. The workshop will (i) describe the basic concepts, (ii) give specific software instructions (in R), with a live demonstration of an analysis with a real dataset, (iii) include a general discussion of the best practices and guidelines for applying these methods, and (iv) cover some of the advanced topics, such as the ‘high-dimensional propensity score’ algorithm, explaining the rationale, use/applications, and potential enhancements of this method with machine learning methods/interesting data dimensions in the large healthcare analysis context. 

Requirement:

The prerequisite is knowledge of multiple regression analysis and working knowledge in R (e.g., basic data manipulation and regression fitting; provided software codes will be annotated). Background in causal inference is not required. Participants are encouraged to bring a fully-charged laptop to follow the instructor’s software instructions during the workshop. Having a google account will be helpful in accessing the cloud-based computing platform.

Slides:
  • Live links:
    • Link for the slides will be live during the workshop event.