Measuring Returns to Healthcare - Methods, Results, and Code

Medicare claims data report payments made to healthcare providers, which provides information on healthcare utilization, including the types of care provided. Deidentified beneficiary information includes demographics and diagnoses. Medicare claims data is also well suited to measuring value because it is longitudinal and includes information about mortality. These data well documented at the Research Data Assistance Center, ResDAC, https://www.resdac.org/. This project uses these data to characterize providers (physicians, ambulance companies and hospitals) in the treatments they provide and their quality of care.

Source: Centers for Medicare & Medicaid Services (CMS)

Data & Supporting:

2012/06/11 Data

• Readme
• Medicare Claim Data
• NYS Data

Dataverse Files

• Readme
• Constructing hospital measures
• Primary analytic file from extracted Medicare claims
• Prepare Analytic file
• Fit OLS and 2SLS regressions
• Create tables and figures for main text and appendix
• Instructions for extracting Medicare claims from NBER
• Function to create leave-out instruments
• Supplementary SAS code edits to NBER Medicare SAS Code
• Set up directories and parameters
• Supplementary code to subset to ambulance cases
• cluster-se.r
• create-analytic-file-from-raw-medicare-files.do
• data-dictionary.do
• datadic.tab