To determine whether the Centers for Medicare & Medicaid Services (CMS) Physician Fee Schedule confers higher value for physician work in procedure and test codes than in Evaluation and Management (E/M) codes.
Medicare Payment Advisory Commission previously demonstrated that time for medical services is the dominant element in valuing physician work in the CMS Physician Fee Schedule. In contrast, a more recent analysis suggests that more relative value units (RVUs) per unit time are issued for work in procedure codes than in E/M codes. Both prior analyses had important limitations for evaluating a possible systematic differential valuation of medical services.
Data regarding RVUs, physician work times (minutes), and claims were obtained for all active level I Current Procedural Terminology (CPT) codes from 2011 CMS files. Linear regression was used to assess the associations of work time components and CPT category with work RVUs, including a model that weighted codes by the number of claims.
Included in the analysis were 6522 CPT codes (87 E/M codes, 6435 procedure/test codes). Compared with E/M codes, procedure/test codes did not have a significant difference in work RVUs adjusting for time (−0.631; 95% confidence interval, −1.427 to 0.166). The analysis also did not indicate a work RVU advantage specifically for Surgical CPT codes compared with E/M adjusting for time (−0.760; 95% confidence interval, −1.560 to 0.040). This pattern was not altered after weighting codes by the number of claims, indicating that an increase in RVUs per minute was not concentrated in a small number of highly utilized procedure codes.
We did not find evidence of a systematic higher valuation of physician work in procedure/test codes than in E/M codes in the CMS RVU system.
We sought to determine whether the Centers for Medicare & Medicaid Services Physician Fee Schedule confers higher value for physician work in procedure and test codes than in Evaluation and Management (E/M) codes. The findings indicate that work time explains 89% of the variance in relative value units and that procedure and test codes do not receive a systematic higher valuation than E/M codes.
*Department of Neurology Health Services Research Program, University of Michigan Health System, Ann Arbor, MI
†Private practice, Leesburg, VA
‡Department of Neurology, Henry Ford Hospital, Detroit, MI.
Reprints: Kevin A. Kerber, MD, MS, Department of Neurology, 1500 East Medical Center Dr, Ann Arbor, MI 48109. E-mail: firstname.lastname@example.org.
Disclosure: This work did not have a funder or sponsor. K.A.K. serves as an advisor for the American Academy of Neurology for tasks related to the RVS Update Committee and receives a stipend for this work (unrelated to work on this manuscript). M.R. services as a member of the RVS Update Committee (appointed by the American Academy of Neurology) and receives a stipend for this work (unrelated to work on the manuscript). G.L.B. serves as an alternate member of the RVS Update Committee (appointed by the American Academy of Neurology) and receives a stipend for this work (unrelated to work on the manuscript).