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)David Ezekiel-Herrera, Oregon Health & Science UniversityNathalie 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.
|