• Report

Assessment of the potential impact of productivity changes on Medicare RVUs. Final report

Citation

Cromwell, J., McCall, N., Hoover, S., Urato, C., Cromwell, G., & Braun, P. (2007). Assessment of the potential impact of productivity changes on Medicare RVUs. Final report. (Prepared for U.S. Department of Health and Human Services, Washington, DC. ASPE Contract No. 100-03-0018). Research Triangle Park, NC: RTI International.

Abstract

In the early 1980s, Hsiao and colleagues at Harvard University developed a conceptual framework for defining resource costs of providing physician services (Hsiao, 1979; Hsiao, 1985). In doing so, the researchers needed to overcome several major methodological challenges: developing reliable and valid methods of measuring physician work; developing a method to align the scale of physician work across different specialties; and developing survey methods that could be used to obtain reliable estimates of work effort from national samples of physicians. The model for the Resource-Based Relative Value Scale (RBRVS) developed by Hsiao and colleagues contains three resource inputs for the provision of physician services: total work, an index of relative practice costs by specialty, and an index of amortized value for the opportunity costs of specialty training (Hsiao et al., 1992a). Based upon the successful development of this conceptual framework, the Centers for Medicare & Medicaid Services funded Hsiao and his colleagues at Harvard, in collaboration with the American Medical Association (AMA) to develop a resource-based relative value scale to be used as the basis of payment for physician services to Medicare beneficiaries starting in 1992. Relative values within families of codes (e.g., heart surgery, radiological scans) were then linked using a reference set of Multiple Points of Comparison (MPCs) (Morton, Kominski, Kahan, 1994). The new resource-based system was predicted to increase fees for physician visits by 15–45% while reducing the income of most surgical specialties by 10–20% (Hsiao et al., 1992b).