Inpatient Data

HERC Inpatient Average Cost Data

The HERC inpatient estimates represent the national average cost of a hospital stay given its Diagnosis Related Group (DRG), overall length of stay, and days in intensive care. The inpatient estimates are based on analysis of cost-adjusted charges in Medicare funded stays of veterans in non-VA hospitals. Estimates are adjusted so that the estimates tally to actual national VA expenditures for that type of care.

Data location: on VINCI (\\vhacdwsasrds01.vha.med.va.gov\HERC) or on the SAS 9.4 Grid (on the server VHACDWSASGSUB2.VHA.MED.VA.GOV at /data/prod/HERC).

Learn More

• HERC's Average Cost Methods
• Register to Use HERC's Average Cost Data
• Guidebook: HERC's Average Cost Datasets for VA Inpatient Care

return to top

HERC Inpatient Person-Level Cost Data

The person-level cost datasets contain total annual costs for VA care received by each individual who used the VA health care system. Currently spanning FY98-present, these data allow researchers to determine the annual costs of individual patients without needing to extract and summarize stay and visit level information from numerous VA databases. These data were discontinued after FY16.

Learn More

• HERC's Average Cost Methods
• Register to Use HERC's Average Cost Data
• Guidebook to the HERC Person-Level Cost Datasets FY1998-Present
• Tabulation: Annual Totals for the FY 1998-2014 Person-Level Cost Datasets

return to top

MCA Inpatient Cost National Data Extracts

The US Department of Veterans Affairs (VA) uses the Managerial Cost Accounting System (MCA), formerly Decision Support System (DSS), for fiscal management and to determine the cost of patient care. MCA uses an activity based methodology to estimate the costs of care. MCA National Data Extracts (NDEs) have been created to facilitate access to workload and cost information. These extracts report costs of inpatient encounters provided by VA.

DISCH (Inpatient Discharge). This is the inpatient MCA cost dataset. Each record includes all of the costs associated with an inpatient discharge. Data that span multiple fiscal years are associated with the year in which the patient was discharged. HERC has created a Guidebook for the MCA cost NDEs. This dataset does not include patient diagnostic information, but can be merged to the Medical SAS inpatient datasets (i.e., Patient Treatment Files).

TRT (Treating Specialty). This dataset includes costs for inpatient care with one record for each patient for each treating specialty visited per fiscal period (month). HERC has created a discharge view of these data, so that there is a subtotal of different types of inpatient care.

OBS (Observation care). MCA records observation care in the Opat NDE, whereas observation care is recorded as inpatient care in the PTF. This dataset reformats all observation records into the inpatient (TRT) format.

Note

Utilizing the FY15 MCA data, we discovered there were 141 records where the raw length of stay (LOS) did not equal the max of (DISDAY-ADMITDAY) or 1.

We followed up with MCA to report this issue. Per MCA, the discrepant records were due to the MCA production database, the source for the NDE, which needed correction. Updates to programming have been implemented for the FY17 datasets. Since these changes affect the production databases the local site teams use for monthly processing, they were unable to make updates to prior year datasets. Only data resulting from current year processing were corrected and prior year processing was closed. We have left the discrepant records as-is in our resulting dataset.

Learn More

• MCA's Cost Methods
• Obtain Permission to Use MCA's Cost Data
• Guidebook: Research Guide to the Managerial Cost Accounting National Cost Extracts
• Guidebook: HERC's MCA Discharge Dataset with Subtotals for Inpatient Categories of Care

return to top

OMOP Inpatient Cost Data

Note: This is a new data transform. If you find issues when work with these data, please let us know: herc@va.gov

The Observational Medical Outcomes Partnership (OMOP) is a common data model that allows for comparison across disparate observational data sources. The idea is to transform data into a common model with similar data dictionaries to aid analysis. VA data in OMOP do not provide additional data, but rather pull from various sources across CDW to centralize VA data with a standard vocabulary. OMOP access can be requested through DART for research projects or ePAS for operations use. See the VHA Data Portal for more information (VA intranet only: http://vaww.vhadataportal.med.va.gov/DataSources/OMOPCDWData.aspx).

HERC has worked with the OMOP team to put cost data into the OMOP data model. The primary challenge with this transform is that the data in CDW that OMOP draws from may not be the same as the data used for VA accounting of costs. VA considers inpatient stays to be any visits in which a patient stays overnight in a facility, including post-acute care, nursing facility care, and domiciliary services. This includes episodes where a patient moves between these types of care. Outside VA, inpatient stays are generally considered separate from long-term care. To work around this, Managerial Cost Accounting TRT tables are used. These tables are broken into months, allowing for matching on different types of inpatient care to match OMOP definitions. For stays with more than one type of inpatient care (e.g. acute care hospital and inpatient rehab) costs are distributed evenly over the full length of each type of stay. As an example, a patient that spent time in acute care and then went to post-acute care will have this stay to broken up into two episodes to satisfy OMOP logic. The total cost is then broken into a per-day cost and assigned accordingly. Full logic on the transform is available on the VHA Data Portal.

return to top

VA Utilization Data

VA Patient Treatment File (PTF) Medical SAS Inpatient Data

The VA hospital discharge file is known as the Patient Treatment File (PTF). The PTF is similar to discharge datasets created for Medicare and the health officials in many states. It includes length of stay, type of hospital unit, diagnoses, procedures, and patient demographics. VA data are unique because they include long-term, psychiatric, and other types of care that may not be reported in other discharge datasets.

Learn More

• Tabulation: Inpatient Days And Inpatient Stays In Each Bedsection Fiscal Years 1993-2012
• Inpatient PTF Bedsection and CDA

return to top

VA Corporate Data Warehouse Inpatient Data

The VA corporate Data Warehouse (CDW) is a repository of national VHA data that contains comprehensive information about patient encounters with VHA, including clinical, financial, demographic, and benefit information. It is a relational database organized into a series of domains and tables within domains. These tables and domains can be connected using linking keys to form analytic datasets. Data are available from 1999-present.

The CDW Inpatient domain consists of three parts: Part 1 - Inpatient, Part 2 - Inpatient Census (bed occupancy), and Part 3 - Inpatient Fee Lodger; information on hospitalizations and nursing home care can be found in Part 1. See the VIReC Factbook Corporate Data Warehouse (CDW) Inpatient 3.0 Domain (VA intranet only: https://vaww.virec.research.va.gov/CDW/Documentation.htm) for more information. Part 1 (Inpatient) of the Inpatient domain contains detailed information on patient encounters, including date and location, bed sections, patient transfers, and procedure and diagnosis codes. The CDW Inpatient Domain is available on VINCI and can be accessed through DART for research projects or ePAS for operations users. More information about data access is available on the VHA Data Portal (VA intranet only: http://vaww.vhadataportal.med.va.gov/DataAccess/DataAccessOverview.aspx).

Learn More

• VIReC Factbook: Corporate Data Warehouse (CDW) Inpatient 3.0 Domain (VA intranet only: (https://vaww.virec.research.va.gov/CDW/Documentation.htm)
• VINCI Data Sources: CDW (VA intranet only: https://vaww.vinci.med.va.gov/VinciCentral/DataSources/Index)
• CDW page on the VHA Data Portal (VA intranet only: http://vaww.vhadataportal.med.va.gov/DataSources/CDW.aspx)
• VIReC’s CDW page (VA intranet only: https://vaww.virec.research.va.gov/CDW/Overview.htm)

return to top

VA Observational Medical Outcomes Partnership (OMOP) Data

The Observational Medical Outcomes Partnership (OMOP) is a common data model that allows for comparison across disparate observational data sources. The idea is to transform data into a common model with similar data dictionaries to aid analysis. VA data in OMOP do not provide additional data, but rather pull from various sources across CDW to centralize VA data with a standard vocabulary. OMOP access can be requested through DART for research projects or ePAS for operations use. See the VHA Data Portal for more information (VA intranet only: http://vaww.vhadataportal.med.va.gov/DataSources/OMOPCDWData.aspx).

CDW inpatient data is available on OMOP. A fill list of available OMOP domains is available on VINCI Central (VA intranet only: https://vaww.vinci.med.va.gov/VinciCentral/DataSources/Index).

return to top

VA Inpatient Community Care Data

The VA Community Care files include information on covered care provided to VA patients by contract providers under the VA Community Care Program, including care provided under the Veterans Access, Choice, and Accountability Act (VACAA; also called “Choice”, FY2015-2017) and the VA Maintaining Internal Systems and Strengthening Integrated Outside Networks (MISSION) Act (FY2019-present). For more information visit our VA Community Care Data page.

Learn More

• VA Community Care
• Non-VA Medical Care (Fee Basis) Data: A Guide for Researchers
• Technical Report #26: Use and Cost of Fee Basis Services in FY 2007 (VA intranet only: https://vaww.herc.research.va.gov/files/RPRT_645.pdf)
• VHA Office of Community Care

return to top

FAQs

What are the average daily costs of VA inpatient care?

This table presents the daily cost of stays in VA care. The data are not adjusted for inflation.

Notes

This table includes overhead and physician costs.

return to top

How do you define an acute, short stay hospitalization in the VA?

For inpatient care, the VA does not distinguish between medical/surgical care and non medical/surgical stays. The VA keeps track of bedsections.  Bedsection is a VA-specific term analogous to a hospital ward; MCA uses the term treating speciality, but the values are the same as the bedsection. Because a patient can get transferred among bedsections multiple times within a single medical/surgical hospital stay, keeping track of bedsections provides us with a great amount of detail.

See the section, "Categorizing the CDR into 11 Inpatient Cost Accounts" for more on HERC's definition of medical surgical bedsection.

return to top

How do I obtain the DRG value weights from Medicare?

The Centers for Medicare and Medicaid Services (CMS) maintains a list of relative value units (RVUs) for inpatient hospital care. These RVUs are also known as the Diagnosis-Related Group (DRG) weight. DRGs are used to determine how much Medicare reimburses a hospital for providing care, as patients within each DRG category are expected to use a similar amount of hospital resources. A DRG is assigned by a hospital based on the diagnoses and procedures noted by physicians in a patient’s medical record.

Under the conventional DRG system, hospitals were reimbursed the same amount for treating a patient within that DRG, whether that patient was extremely sick or relatively uncomplicated. Recognizing that it was more expensive for hospitals to treat sicker patients, in 2008, CMS changed from DRGs to MS-DRGs, which stand for Medicare-Severity Diagnosis Related Groups. The MS-DRG reflects varying resource intensity to treat a condition through its classification of inpatient admissions into three mutually-exclusive categories: a base MS-DRG; a MS-DRG with care complications or comorbidities (CC); or a MS-DRG with major care complications or major comorbidities (MCC). Note that while these MS-DRGs all reference the same underlying condition, each of the three MS-DRGs will have its own MS-DRG number. A three-digit code is used to represent DRGs and MS-DRGs, but there is no way to match DRGs in the old system to the MS-DRGs, as a given DRG may have been split into two or more MS-DRGs. There were approximately 475 DRGs under the old system, and there are currently approximately 750 MS-DRGs.

The latest DRG relative value units, along with information on the average length of stay for that MS-DRG, can be found on CMS' web site, in Table 5 entitled, “List of MS-DRGs, Relative Weighting Factors and Geometric and Arthritic Mean Length of Stay.” MS-DRG's weights can be merged to a utilization file using the MS-DRG number. Therefore, in order to be able to use the MS-DRG weights, your dataset must include MS-DRG code numbers. If you are using these data to determine length of stay for your population of interest, you must choose between the average length of stay and geometric mean length of stay. The average length of stay (ALOS) is the simple arithmetic mean, and is calculated by adding up the length of stay for all patients with that MS-DRG and dividing by the number of patients. The geometric length of stay (GLOS) is calculated by multiplying all of the lengths of stay for all patients with that MS-DRG and taking the nth root of that number, where n is the number of patients. The GLOS has the effect of reducing the influence of outlier values, or patients with very high or very low length of stay.

An example of the type of data available from CMS can be seen in the DRG weight files for 1983-2007 and the MS-DRG weight files for 2008-2013. However, please refer to the files on the CMS website for the most up-to-date information. It is important to make sure that the DRG or MS-DRG weight file is the same year as the utilization data you are using. HERC also has historical DRG files since 1983; you can email herc@va.gov to obtain these.

return to top

How do you categorize the PTF bedsections into 10 categories?

Notes

Number 4 is surgery, but this becomes combined with category 0. The skip in number from 3 to 5 maintains consistency with average daily rates calculated before FY98.

return to top

How do you categorize the CDR into 11 inpatient cost categories?

Notes

* Stations with an approved PRRTP program include: 459 463 501 504 515 516 518 523 528 541 546 549 554 555 556 561 568 573 586 589 590 595 598 620 622 631 632 635 637 640 645 653 656 658 662 663 666 676 678 687 689

return to top

HERC Resources

• Guidebook: HERC's Average Cost Datasets for VA Inpatient Care
• Guidebook to the HERC Person-Level Cost Datasets FY1998-Present
• Guidebook: Research Guide to the Managerial Cost Accounting National Cost Extracts
• Guidebook: HERC's MCA Discharge Dataset with Subtotals for Inpatient Categories of Care
• Guidebook for the MCA Intermediate Product Department Files
• Non-VA Medical Care (Fee Basis) Data: A Guide for Researchers
• Technical Report #26: Use and Cost of Fee Basis Services in FY 2007 (VA intranet only: https://vaww.herc.research.va.gov/files/RPRT_645.pdf)

Last updated: April 13, 2021

return to top