Online Program

Return to main conference page
Thursday, February 15
PS1 Poster Session 1 and Opening Mixer Thu, Feb 15, 5:30 PM - 7:00 PM
Salons F-I

Navigating Large-Scale Forest Plots Using R and Shiny (303680)

View Presentation View Presentation

David Ezekiel-Herrera, Oregon Health & Science University 
Nathalie Huguet, Oregon Health & Science University 
Miguel Marino, Oregon Health & Science University 
Jean P. O'Malley, Oregon Health & Science University 
Rachel Springer, Oregon Health & Science University 
*Steele Valenzuela, Oregon Health & Science University 

Keywords: R, Shiny, Data Visualization, Forest Plot

Forest plots are data visualizations traditionally implemented in meta-analyses, plotting point estimates from collected studies. However, with the advent of big data, forest plots are now being used to visualize multiple parameter estimates from a single model with many covariates. A mixture of several continuous variables and one or two categorical variables may appear digestible at first, but can easily overwhelm those who are not familiar with the model, especially as the number of predictors increases. We overcome this limitation by implementing R and Shiny, an interactive application that can filter down which predictors are pertinent and relevant to the user. Additionally, by placing an interactive application in the hands of fellow collaborators, they are able to digest the information at their own pace. We will describe easy-to-implement procedures for producing plots using R and Shiny and will provide a demonstration in the poster and laptop. We will apply this functionality to a study of the impact of gaining Medicaid insurance on health outcomes among a diabetic population derived from a multi-state electronic health record network.