The HERC inpatient estimates represent the national average cost of a hospital stay given its Diagnosis Related Group, overall length of stay, and days in intensive care. The inpatient estimates are based on analysis of cost-adjusted charges in the 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.
|Rehab, MH, and LTC||X||X||X||X||X||X||X||X||X||X||X||X||X||X||X||X|
|Rehab, MH, and LTC||X||X||X|
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).
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.
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 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. 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 NDE. This dataset does not include patient diagnostic information, but it 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.
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 Steve Porter and Phil Franklin from MCA to report this issue. Per Phil, 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.
|Discharge with subtotals||X||X||X||X||X||X||X||X|
|Discharge with subtotals||X|
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.
In 2005 MCA files were released to VA researchers at the level of the intermediate product department for each encounter for inpatient (T-IPD) care. These files provide more detailed cost information for specific types of care. Previously, MCA National Data Extracts reported costs for six broad categories -- medicine/other, laboratory, pharmacy, radiology, nursing, and surgery. These cost sub-totals were dropped from the NDE files while new IPD files were created with a richer set of cost detail. The IPD extract files are available for all fiscal years beginning in 2003.
TRT-IPD (Inpatient intermediate Product Department). This dataset has one record for each separate MCA product department (IPD) utilized within an inpatient stay. Intermediate products represent work performed in each department and are bundled to make up the inpatient stay.
The IPD national extract files do not contain any formatting for the product department variable (IPD_NUM), so HERC has created a permanent SAS format library for the product department names for files FY 2003 through FY 2008. HERC has made this file available at the Austin Information Technology Center for use by researchers.
The Non-VA Medical Care files include information on covered care provided to VA patients by contract providers under the Non-VA Medical Care Program. The Non-VA Medical Care files overlap with other VA utilization data in two areas. Because Non-VA Medical Care data represent invoices rather than discrete stays, ongoing inpatient stays appear in Non-VA Medical Care data once the provider has begun to invoice VA. When an inpatient Non-VA Medical Care stay is authorized, a 'stub' record is created in the PTF Non-VA Hospitalization (NVH) file. Once VA has paid a claim for the stay, the stub record is updated to include additional information.
Data from the Non-VA Medical Care system input in SAS files at the VA Austin Information Technology Center (AITC). The files that contain Non-VA Medical Care are named in the form:
MDPPRD.MDP.SAS.FEN.FYyy where yy is the last two digits of the year in which the payment was processed.
Ancillary services provided to inpatients:
Acute hospital and community nursing home care is found in the inpatient file (INPT). This file has places to record 5 diagnosis codes, 5 procedure codes, and the "Pricer DRG", although there are many blank values.
There is an alternative method for accessing summaries of Non-VA Medical Care expenditure data, one that does not require an AITC time-share account. A "Non-VA Care Cube" has been created as part of the national Financial and Clinical Data Mart. The data cube shows combined payments from the four central Non-VA Medical Care files: Inpatient, Outpatient, Inpatient Ancillary, and Pharmacy. One can view and download a series of standard reports or create a unique "view". The data cube does not present individual encounter records. Rather, it shows payment totals summed within category (Purpose of Visit) and time period (month or fiscal year). The data cube is accessed through the VISN Support Services Center (VSSC) web site, also known as the KLFMenu. The site is accessible only through the VA intranet.
|Non-VA Medical Care Data||FY00||FY01||FY02||FY03||FY04||FY05||FY06||FY07||FY08||FY09||FY10||FY11||FY12|
|Inpatient Ancillary / Physician||X||X||X||X||X||X||X||X||X||X||X||X||X|
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.
This table presents the daily cost of stays in VA care. The data are not adjusted for inflation.
|Spinal cord injury||787||946||1,125||1,292||1,502||1,811||2,001||1,923||2,052||2,072||2,275||2,141|
|Inpatient substance abuse||532||587||695||606||602||625||663||676||708||716||705||728|
This table includes overhead and physician costs.
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.
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 firstname.lastname@example.org to obtain these.
|0.||Inpatient Medicine/ Surgery|
|2.||Inpatient Blind Rehabilitation|
|3.||Inpatient Spinal Cord|
|6.||Inpatient Substance Abuse|
|9.||Inpatient Long Term|
|10.||Psychosocial Residential Rehabilitation Programs (PRRTP)|
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.
|Category of Care||Bedsection / Treating Specialty codes|
|0) Acute medicine||1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 24 30 31 34 83 1E 1F 1H 1J|
|1) Rehabilitation||20 35 41 82 1D 1N|
|2) Blind Rehab||21 36|
|3) Spinal Cord Injury||22 23|
|4) Surgery||48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 65 78 97 1G|
|5) Psychiatry||25 26 28 29 33 38 39 70 71 75 76 77 79 89 91 92 93 94 1K 1L|
|6) Substance Abuse||27 72 73 74 84 90 1M|
|7) Intermediate Medicine||32 40|
|8) Domiciliary||37 85 86 87 88|
|9) Nursing Home||42 43 44 45 46 47 64 69 80 81 95 96 1A 1B 1C|
|10) PRRTP - Psychosocial Residential Rehabilitation Treatment Program*||25 26 27 28 29 38 39|
* 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