This study assessed the impact of a new Medicare Fee Schedule (MFS) on the revenues of teaching physicians. The new MFS, initiated in January 1992, is based on the Resource-Based Relative Value Scale; it altered the relative prices paid to physicians. Relative payment levels were greatly increased for visits (less intensive relative value units, [RVUs]), and reduced for most types of diagnostic tests and surgical procedures (more intensive RVUs). Relative payment levels were raised in rural areas and lowered in large urban areas. It was hypothesized that teaching physicians would be adversely affected by the MFS because they perform more high-technology procedures and provide comparatively fewer primary care services. In addition, they are located in large urban areas, and often experience customary charges that may be higher than the area-wide historical payments calculated for fee schedule transition. Further, some academic medical practices had not signed Medicare participation agreements and would be affected by MFS limits on balanced billing. The study analyzed all Medicare Part B claims associated with discharges during 1991-1993 from a sample of teaching and non-teaching hospitals. Part B revenues and selected price variables were assessed for different categories of service. No evidence was found to suggest that teaching physicians had been adversely affected by the MFS. A reduction in prices and the quantity of services per admission led to a similar percentage decline for inpatient revenues per admission in all types of hospitals. An increased shift of diagnostic tests to outpatient settings, a secular decline in some surgical procedures, and MFS-related coding changes led to a general decline in inpatient physician service intensity. Teaching physicians altered the mix of services provided, substituting less RVU-intensive visits for high-tech diagnostic tests.
The purpose of this study was to evaluate whether and how the MFS had a disproportionate impact on teaching physicians compared to non-teaching physicians. The study also addressed whether physicians in large teaching hospitals, especially academic medical centers, were more affected than physicians in small teaching hospitals.
Methods used by Medicare to reimburse physician services were fundamentally changed in January 1992. A new MFS was initiated, which was based on the Resource-Based Relative Value Scale and which dramatically altered the relative prices paid to physicians. Relative payment levels were greatly increased for visits (less intensive RVUs) and were reduced for most types of diagnostic tests and surgical procedures (more intensive RVUs). In addition, the MFS adjusted payments by a Geographic Practice Cost Index, which took into account the actual costs of physician practice in different parts of the country. This adjustment raised relative physician payment levels in rural areas and lowered them in large urban areas.
It was expected that the redistributive effects of the MFS would be particularly adverse for teaching physicians, who perform relatively more high-technology procedures and less primary care and who are located in large urban areas. Customary charges for teaching physicians also could be higher than the area-wide historical payments calculated for fee schedule transition. In addition, an unknown number of academic medical practices had not signed Medicare participation agreements, and the MFS limits on balance billing could further constrain revenues for some teaching physicians. Prior research had shown that physician responses to Medicare fee reductions will depend on the relative magnitude of income effects, such as an increase in service volume to offset a fall in revenue; and substitution effects, such as an increase in services to non-Medicare patients. Relatively little is known about physician response to Medicare fee increases. It was hoped that higher visit reimbursement levels under the MFS would lead physicians to encourage more contact with patients.
A stratified random sample of 701 hospitals was selected from Medicare's 1991 Provider of Service file. The strata were defined by teaching status of the hospital, ranging from academic medical centers to non-teaching hospitals. MedPAR records for all discharges from the sample hospitals during 1991-1993 were identified. Using the Health Insurance Claim number of these records, all Part B claims associated with these discharges were extracted.
Key revenue and price variables were constructed for each hospital-year: Part B revenues per admission, RVUs per admission, a price per service index, and a price per RVU. Each of these variables was created for overall inpatient services, and for five separate categories of service: evaluation and management services; high-tech surgical procedures (RVUs of 40 or more); general (all other) surgical procedures; high-tech tests (RVUs of 1.3 or more); and routine (all other) tests.
There was no evidence that teaching physicians were adversely affected by the MFS. Inpatient revenues per admission fell by the same percentage in all types of hospitals. Revenues fell because of a reduction in prices and the quantity of services per admission. The contributions of price per RVU and RVUs per admission to the declines in revenue were similar across all hospitals. A general decrease of inpatient physician service intensity resulted from a shift of diagnostic tests to outpatient settings, a secular decline in some surgical procedures, and MFS-related coding changes. Teaching physicians appear to have altered the mix of services provided (and/or billed for) by substituting less RVU-intensive visits for high-tech diagnostic tests. As a result, the actual decline in price per RVU was similar for teaching and non-teaching physicians.
Use of Results
The study documents the successful impact of Medicare's new RVU-based physician payment system on overall revenues and the substitution of less costly, more visit-oriented services for high-tech diagnostic services. The methods used will help to assess substitution and income effects associated with physician responses to future Medicare price changes and should result in more accurate estimates of Medicare program savings. The results suggest that price decreases will be associated with increasing volume but decreasing intensity per service and, thus, are important for monitoring future volume changes and determining the appropriate level of sustainable growth in physician services.
AGENCY SPONSOR: Health Care Financing Administration, Office of Strategic Planning
FEDERAL CONTACT: William Buczko
PHONE NUMBER: (410) 786-6593
PIC ID: 6666
PERFORMER ORGANIZATION: Center for Health Economics Research, Waltham, MA