Public data and python source code for curating and combining files from the [Hospitals data archive] (https://data.cms.gov/provider-data/archived-data/hospitals) maintained and publicly provided by the Centers for Medicare and Medicaid Services (CMS).
The CMS Hospitals Data Archive (aka Hospital Compare, Care Compare) contains historical data on thousands of hospitals. These datasets include data on hospital mortality and patient safety indicators (PSIs), healthcare associated infections (HAIs), payment and value of Care, timely and effective care, unplanned visits and readmissions, imaging efficiency, and results of the following programs: Hospital-Acquired Conditions Reductions Program (HACRP), Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS), Hospital Readmissions Reduction Program (HRRP), and Hospital Value-Based Purchasing program (HVBP). However, these data are spread across many differently formatted files, the formats and features of which have changed quarter-to-quarter and year-to-year. Even the spelling of features has not remained consistent.
This repository contains the original publicly available data provided by CMS, the source code for curating and aggregating the data, and the resulting data files (which includes hospital-specific datasets).
Note, this repository contains data for as far back as 2005. Due to varied updates to CMS' online platform over the past few years, hospital archive data predating 2015 are now difficult or impossible to find on CMS sites.
Coming soon ...
Kenneth J. Locey, PhD. Senior Clinical Data Scientist. Center for Quality, Safety & Value Analytics. Rush University Medical Center.