HSAF Hospital Service Area Files - onetomapanalytics/Meta_Data GitHub Wiki

HSAF - Hospital Service Area Files

General description

  1. Database primary purpose - The Hospital Service Area (HSAF) data is a summary of calendar year Medicare inpatient hospital fee-for-service claims data. It contains the number of discharges, total days of care, and total charges summarized by hospital provider number and the ZIP code of the Medicare beneficiary.
  2. Overall data type - Health outcomes
  3. Dataset type - Longitudinal
  4. Data source - Claims
  5. Data level - Hospital level
  6. Geographic location of the data collection sites - United States
  7. Sponsor, manager, or home institution - Centers for Medicare & Medicaid Services (CMS)
  8. Date range - 2014 - 2019
  9. Geolocation data - Zip codes (based on the mailing address used for cash benefits to the beneficiary)
  10. Hospital identifiers - Medicare Provider Number
  11. Longitudinal tracking - Track providers through Medicare Provider Number
  12. Financial variables - Total charges
  13. Variables that are uniquely present in this dataset - The data is generated from MEDPAR (Medicare Provider Analysis and Review) input files using a standard IBM utility sort (Syncsort) to select and sum the specified data elements from the designated Medpar source files to produce the HSAF output file.
  14. Database caveats and limitations - The full dataset contains more records than most spreadsheet programs can handle, which will result in an incomplete load of data. So, the use of a database or statistical software is required.

Applicable methods

  1. Association methods, such as negative-binomial regression (1), linear regression (2), logistic regression models (3)

High-impact designs

  1. Enrichment of the HSAF dataset through linkage to other datasets, such as American Hospital Association Annual Survey database (AHA) (1)

Data dictionary

To access the HSAF data dictionary, click here

Variable categories

  1. Provider ID (i.e., Medicare Provider Number of the institution that rendered services to a beneficiary)
  2. Geolocation (i.e., zip Code)
  3. Days of care (i.e., derived by subtracting the date of admission from the date of discharge)
  4. Charges (i.e., total charges)
  5. Total Cases

Linkage to other datasets

Linkages can be established for any dataset that might have geolocation data (i.e., ZIP code) or the provider ID (Medicare number).