Features & Benefits
RAND Hospital Data provides subscribers with a curated subset of variables from CMS hospital cost reports, combining key metrics from hospitals in the United States and Puerto Rico – from 1996 onwards – in a single download.
Complex Data Simplified
Subscribers have great flexibility in their chosen product and can select the dataset that best suits their individual research or analytical needs. Customization capabilities include:
Multiple Years, One Download
Using raw CMS cost report data requires downloading and combining separate files for each year of interest. The RAND Hospital Data allows users to access all years of data from 1996 onwards in a single download, saving time and effort.
Key Metrics Calculated
Interested in analyzing profitability or other ratios but not sure what fields to use? The RAND Hospital Data include approximately 400 calculated metrics – operating margins, occupancy rates, cost-to-charge ratios, etc. - saving subscribers even more time.
Standardized Time Periods
Subscribers can download datasets standardized to different time periods, including calendar years (January-December), federal fiscal years (October-September), or the time periods chosen by individual hospitals.
Concept-Based Variable Names
In raw cost report data, variables are identified by their position on the form. For example, users interested in hospital operating expenses would need to find observations in the numeric values table where wksht_cd = ‘G300000’ and line_num = ‘00400’ and clmn_num = ‘00100’. RAND Hospital Data employs user-friendly variable names, so subscribers only need to look for a column called ‘operating_expenses’.
Data Outliers Corrected
Hospitals sometimes submit data that are clearly erroneous. To mitigate the challenges of working with these errors, we apply an automated outlier detection and correction algorithm. Subscribers can choose to download corrected or uncorrected datasets, leading to more accurate and reliable analysis results.
Geographic Summary Datasets
Subscribers interested in regional or national patterns can access datasets with select variables summarized at four geographic levels: county, core-based statistical area, state, and national.
Flexibility
Subscribers can download any datasets that best align with their research or analytical goals. This adaptability caters to diverse data analysis needs and methodologies.
Comprehensive Documentation
Each variable in the RAND Hospital Data is labeled and explained, providing subscribers with clear information about the source of each variable, how key metrics are calculated, and any relevant changes in variable definitions over time. This documentation enhances data transparency, facilitating better understanding and utilization of the data.