Observational studies are a widely used and challenging class of studies. A key challenge is selecting a study cohort from the available data, or “pruning” the data, in a way that produces both sufficient balance in pre-treatment covariates and an easily described cohort from which results can be generalized. Visual Pruner is a free, easy-to-use R shiny web
application that facilitates both of these goals by letting analysts use updatable linked visual displays to refine a study’s inclusion criteria. By helping analysts see how pretreatment covariate distributions relate to the estimated probabilities of treatment assignment (propensity scores), the app lets analysts make pruning decisions based on patterns that are otherwise hard to discover. The app yields a set of transparent covariate-based inclusion criteria that can be easily applied in any statistical software package. Thus, while the app is interactive and facilitates iterative decision-making, it can also easily be incorporated into a reproducible-research workflow.
March 31, 2018