Publicly available data from Medicare frequently requires extensive initial effort to extract desired variables and merge them; this package formalizes the techniques I've found work best. More information on the Medicare program, as well as guidance for the publicly available data this package targets, can be found on CMS's website covering publicly available data. See < https://www.cms.gov/data-research>.
The medicare package is a collection of functions and methods I've used to manipulate Medicare data and get it ready for analysis. This includes things like efficiently subsetting messy Cost Report data to pull desired variables, renaming variables in data that doesn’t come with headers, and finding more useful names for Provider of Service files from the early 2000’s that name variables sequentially from “PROV0001”.
Publicly available Medicare data often requires extensive preparation
and cleaning before any analysis can take place. Files are often raw
dumps of database tables, which the researcher is expected to subset and
merge to make a workable dataset. This package contains methods to
extract data from such datasets (e.g. Cost
Reports),
provide useful names for variables (Cost Reports and Provider of
Services
File),
and even parse data dictionary / layout files to extract variable names
for older datasets, where names in the raw data are essentially
Var1, Var2, Var3... (Provider of Services File).
medicare is under active development and available on CRAN. You can
install the latest release version of the package by using
install.packages("medicare")
You can install the development version of the medicare package using
devtools:
library(devtools)
devtools::install_github("robertgambrel/medicare")
Please let me know about any problems by opening an issue.
For detailed examples on how to use some of the functionality, check out the Vignettes, which show examples similar to what I’ve done in my own work.
browseVignettes(package="medicare")