There are many reasons researchers may want to compare the cost of care at VA to non-VA commercial hospitals. This webpage explains some possible methods.
Comparing VA and Non-VA Inpatient Costs
Researchers interested in comparing the cost of VA and non-VA inpatient stays may assume they can compare the Managerial Cost Accounting (MCA) Discharge file to non-VA inpatient data, such as Medicare or state all-payer claims data; however, without making adjustments to the data, you would be comparing apples and oranges. The reason is quite simple: VA and commercial hospitals define a hospital discharge differently.
Consider this example where Joe, an older Veteran, is admitted to a VA hospital in a medical bedsection/treating specialty with a broken hip. During the stay, Joe receives surgery, spends a few days in the ICU, recovers in a medical bedsection, and then spends time in the nursing home before he is discharged to home. In VA, all of this care is reported in a single inpatient discharge record. However, if Joe was admitted to a commercial hospital instead, he would not be transferred from medicine to a nursing home. Rather, Joe would be discharged from the medical/surgical hospital, and his care in a nursing home would be captured in a separate discharge record. Commercial hospitals would have two discharge records for Joe, whereas the VA reported this information in one record.
If you want to compare VA stays (or discharges) to non-VA stays, you need to consider how to define inpatient stays with VA data. HERC Average Cost data addresses this issue by using VA PTF/CDW inpatient data to define medical-surgical inpatient stays in a way comparable to non-VA data (e.g., Medicare Medpar dataset or HCUP data). HERC average cost data also then calculates costs in a way comparable to non-VA care by using Medicare relative value weights for Diagnosis Related Group (DRG) based on diagnosis and procedure codes. Because HERC Average Costs use non-VA relative value weights, researchers looking to compare costs should consider if these data are appropriate for their research question (learn more).
Comparing VA and Medicare Costs
Medicare is the largest health care payer in the U.S., and data on Medicare-funded health services are publicly available and nationally representative, so it is the most frequently used benchmark for cost comparisons. In addition, VA researchers have access to VA/CMS data through VIReC (VA intranet only: https://vaww.virec.research.va.gov/Index-VACMS.htm). Therefore, this page describes Medicare and VA cost data, the differences between these data, what can be done to make them more comparable, and the evidence from existing cost comparisons.
Medicare Cost Data
Reimbursements
One estimate of the cost of Medicare-sponsored services is the total Medicare reimbursement paid to the provider. This amount is also referred to as the Medicare-allowed payment. The total reimbursement is distinct from the amount that Medicare pays the provider because it includes both the payment by Medicare and the copayments and deductibles made by the patient. In this document when we refer to the Medicare reimbursement, we mean the total amount paid by both patient and sponsor. The total Medicare reimbursement represents the cost of care from the perspective of the health care sponsor.
The Medicare reimbursement rate often reflects policy decisions about reimbursement and may not necessarily represent the value of the goods, labor, and capital used to produce the service. Medicare provides greater reimbursement to physicians for an hour spent providing a procedure than for an hour spent in evaluation and management based on greater perceived intensity of effort for procedures as determined by skill, effort, judgment, and stress.
Medicare reimbursement rates differ by setting and region. For example, the reimbursement for a given hospital stay is higher at a teaching hospital than at a community hospital that does not train medical residents. Medicare rates are greater in geographic areas that have higher labor costs or greater malpractice insurance premiums. As a result, the analyst must consider which reimbursement rate is appropriate for the chosen study.
Cost-Adjusted Charges
An alternative to provider reimbursement is the cost-adjusted charge. The total charges incurred by the patient are obtained from a hospital bill. These charges can be adjusted by the ratio of cost-to-charges obtained from the hospital cost report. This cost-adjusted charge is regarded as a useful proxy for the actual cost of care from the perspective of the hospital.
The cost-adjusted charge does not necessarily equal the reimbursement to the provider. Medicare reimburses hospitals on a flat rate based on diagnosis (Diagnostic Related Group) and the hospital’s characteristics, so the cost-adjusted charge may be greater or less than this amount.
Choosing a Medicare Benchmark
These are several reasons to compare VA costs to Medicare costs, and the purpose of the study may determine which Medicare benchmark should be chosen.
- Make or buy decisions
The analyst may wish to learn if it is cheaper for VA to provide the service or purchase it from non-VA providers at Medicare payment rates. In this case, the appropriate benchmark is Medicare reimbursement to the provider. - Technical efficiency
Other studies may wish to learn if VA provides a given service with similar technical efficiency at similar cost as a non-VA provider. If charges for the same service stay can be found in Medicare data, the cost-adjusted charges may represent an appropriate proxy. Otherwise efficiency studies may need to examine total hospital production cost, based on Medicare cost reports, with a stochastic frontier model. - Generalizability
Cost-effectiveness research evaluates whether a health care innovation generates sufficient value to justify its cost. VA studies can be more informative if they can be generalized using non-VA costs. Although research might take place within VA, the analyst may wish to estimate what the intervention would have cost if it had been provided outside the VA in order to make inferences about what is cost effective in the health care system at large. Reimbursements or cost-adjusted charges might be used for this purpose.
Learn more about Medicare Data
VA MCA Cost Data
VA uses an activity-based costing system called Managerial Cost Accounting (MCA) to allocate its actual expenses to specific hospital stays and outpatient encounters. The MCA National Data Extracts are databases that report the cost of each inpatient stay, outpatient visit, and prescription provided by VA.
There are a number of differences between MCA data and Medicare. Here, we focus on the comparison between MCA costs and Medicare reimbursements. Cost-adjusted charges are not available for all services although they are available for hospital-provided services, such as the hospital portion for an inpatient stay or the hospital component of outpatient services such as an ambulatory surgery or visits to a hospital clinic. It is not possible to obtain cost-adjusted charges for visits to an office-based physician or for physician services provided to patients in a hospital as physicians do not prepare cost reports. It is possible to use cost-adjusted charges in lieu of allowed payment for the hospital component of services, and many of the same issues about comparability must be addressed.
MCA vs Medicare: Inpatient Comparisons
There are several key differences between MCA cost and Medicare reimbursement for inpatient care:
- Definition of a hospital stay
A single record of care in the MCA Discharge file may include care provided in different facilities, for example, treatment in a nursing home and acute care hospital. The MCA Treating Specialty file contains separate records for each treating specialty, so there are separate records for the nursing home and acute care hospital. In the Medicare system, this would also be regarded as two separate stays. - Provider cost
MCA includes the facility and physician costs. Medicare reimburses the physician and facility separately. It is not appropriate to compare MCA cost to the Medicare-allowed payment to the hospital, as the latter does not include physician services, which may account for as much as 20% of the reimbursement for an inpatient stay. If the MCA estimate is used for comparison, both facility and physician-allowable payments from Medicare must be included in the benchmark. With some strong assumptions, it may be possible to estimate the MCA cost of an inpatient stay exclusive of the physician cost to compare it to the Medicare hospital reimbursement exclusive of physician reimbursements. - Capital cost
Capital depreciation is included in the MCA system, but the cost of financing VA capital acquisitions is not since the cost of borrowing to make capital purchases is not incurred by VA but by the Department of Treasury. In contrast, Medicare reimburses facilities the entire cost of capital including the interest on loans taken out to make capital purchases. Capital may account for up to 5% of Medicare reimbursement to hospitals. - Teaching cost
Both MCA costs and Medicare reimbursements reflect the effect of graduate medical education and other teaching costs. However, teaching intensity is higher in VA hospitals, on average, than in Medicare hospitals.
MCA vs Medicare: Outpatient Comparisons
Medicare reimbursement for outpatient care delivered in a facility (such as a hospital or ambulatory surgical center) includes payments to the physician and to the facility. The facility payment is nearly equal to the physician payment. As a result, total Medicare reimbursement is nearly twice as great for services provided at facilities than for the same services provided by an office-based physician.
Although some outpatient services can only be provided in a facility, a comparison between VA costs and Medicare reimbursement is significantly affected by the decision to use the facility or office-based reimbursement as the benchmark.
MCA vs Medicare: Outpatient Pharmacy Comparisons
It is difficult to compare pharmacy costs of VA to the Medicare system since VA negotiates a lower acquisition cost for most drugs. In addition, MCA pharmacy costs also include substantial hospital overhead, which complicates pharmacy cost comparisons.
Medicare outpatient pharmacy costs are available for patients enrolled in Medicare Part D. The reimbursement amount to pharmacies and payments by patients are available for prescriptions covered under Medicare Part D in the Prescription Drug Event file. Part D drug reimbursement is negotiated between manufacturers, pharmacies, and the large number of Medicare prescription drug plans, so it can vary substantially by plan. The total cost of a drug (including costs by the plan and the patient) may be considerably less than the amount paid to pharmacies. Drug manufacturers provide significant rebates to prescription drug plans.
VA and other Big 4 agencies pays significantly less for prescription drugs than other payers due to negotiated discounts. The Federal Supply Schedule (FSS) reports the drug prices for most federal agencies while the Big 4 prices are not always disclosed. Usual drug costs in the U.S. health care system such as Medicaid can be estimated at 121% of the FSS price. As an alternative, non-VA costs of drugs can be estimated by obtaining VA costs and multiplying by 152%, the amount by which Medicaid prices exceed VA prices. Medicare reimbursement amounts are also available separately for provider-administered drugs paid by Medicare under Part B (e.g. chemotherapy or drugs administered in a clinic).
| INPATIENT TREATMENT |
VA Managerial Cost Accounting (MCA) | Medicare Reimbursement Payments |
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| OUTPATIENT TREATMENT |
VA Managerial Cost Accounting (MCA) | Medicare Reimbursement Payments to Facility-Based Providers |
Medicare Reimbursement Payments to Office-Based Providers |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| OUTPATIENT PHARMACY |
VA Managerial Cost Accounting (MCA) | Medicare Reimbursement Payments |
|---|---|---|
|
|
|
|
|
|
Using HERC Average Costs to Compare VA and Non-VA Care
Given the hurdles to using MCA data to compare VA and Non-VA costs, for some projects, using HERC Average Cost data rather than MCA data to measure the cost of VA care may be more appropriate. While MCA data is created using a local, activity-based method, HERC Average Cost Data represents the estimated cost of each VA encounter. These estimates are created by distributing VA aggregate costs to patient encounter-level utilization using the hypothetical Medicare payment for each service.
To create the HERC Inpatient Average Cost data, we identify the medical/surgical components of stays in VA PTF/CDW Inpatient data consistent with non-VA hospital definitions and conduct a cost regression using length of stay, the highest diagnosis-related group (DRG) weight from multiple bedsections, and number of days in intensive care. Details on the methods are available in Chapter 4 of HERC's Average Cost Datasets for VA Inpatient Care. The SAS code for creating medical/surgical discharges is available upon request.
The HERC Outpatient Average Costs are calculated using the relative value units (RVUs) assigned to procedure codes and Medicare physician and facility reimbursement rates, adjusted to reflect actual expenditures in each outpatient care category. Details on methods are available in HERC's Outpatient Average Cost Dataset for VA Care (VA intranet only).
HERC Average Costs may not be appropriate for all studies, particularly as HERC data are less precise than MCA. Unsure which VA cost data source is best for your project? More information is available in the seminar VA Costs: HERC versus MCA.
Published Comparisons of VA and Medicare Costs
While existing literature comparing costs in VA and Medicare are limited, the evidence generally indicates that costs are lower in VA. Recent evidence comes from research by David Chan and colleagues (2023). Chan and team collected Medicare payments and VA costs using MCA data and HERC Average Cost data resulting from ambulance to VA and non-VA hospitals for dually eligible Veterans above age 65. Using an approach in which veterans are as good as randomly assigned by ambulance to VA or non-VA hospitals, they found that VA care reduces 28-day spending by 21%. They also found evidence of upcoding among non-VA hospitals, a focus on more expensive procedures, and lower rates of potentially cost-effective practices like making phone calls after emergency visits. Details on their work is summarized here.
Chan et al.'s findings are consistent with earlier analyses, which compared VA costs at 6 VA medical centers to the hypothetical fee-for-service Medicare payments for the same services at that time (Nugent, 2004). The authors found that VA provided care at a lower cost. A 2003 supplement in Medical Care includes papers with detailed comparisons for difference services, including acute hospital stays, outpatient care, nursing home, and other types of care (Nugent, 2003).
Other authors have used HERC Average Cost data to compare costs in VA and Medicare for specific services. Radomski et al. (2024) estimate the cost of low value care among Veterans dually enrolled in VA and Medicare. In FY18, the total cost of these low value services was $226.3 million: $62.6 million in VA and $163.7 million in Medicare. Pickering et al. (2022) quantified the cost of Low-value PSA testing and associated downstream services and compared these costs for dually-enrolled Veterans who received their PSA test through VA vs through Medicare. They found that Veterans who received a PSA test through Medicare incurred an additional $36 per person while Veterans who received a PSA test through VA incurred an additional $25 per person.
Additional examples of cost comparisons are available in Technical Report 42: Including Medicare Cost Data in VA Research.
Evaluation of VA Pharmacy Costs
It was estimated that VA’s Pharmacy Benefits Management (PBM) program saved VA $1.5 billion between 1995 and 1999 (Sales, 2005). VA prices for 20 medications commonly used by seniors are lower even than those negotiated by private PBM companies for the Medicare Prescription Drug Plan.
References
Chan, DC, Card D, Taylor L. Is There a VA Advantage? Evidence from Dually Eligible Veterans. American Economic Review. 2023;113 (11): 3003–43.
Nugent GN, Hendricks A, Nugent L, Render ML. Value for taxpayers' dollars: what VA care would cost at medicare prices. Med Care Res Rev. 2004 Dec;61(4):495-508.
Nugent G, Hendricks A. Estimating private sector values for VA health care: an overview. Med Care. 2003 Jun;41(6 Suppl):II2-10.
Pickering AN, Zhao X, Sileanu FE, et al. Assessment of Care Cascades Following Low-Value Prostate-Specific Antigen Testing Among Veterans Dually Enrolled in the US Veterans Health Administration and Medicare Systems. JAMA Netw Open. 2022 Dec 1;5(12):e2247180.
Radomski TR, Lovelace EZ, Sileanu FE, et al. Use and Cost of Low-Value Services Among Veterans Dually Enrolled in VA and Medicare. J Gen Intern Med. 2024 Jul 8.
Sales MM, Cunningham FE, Glassman PA, Valentino MA, Good CB. Pharmacy benefits management in the Veterans Health Administration: 1995 to 2003. Am J Manag Care. 2005 Feb;11(2):104-12.
Last updated: July 24, 2024
