Further TC Cuts Ahead?
Due to inequities in payment methodology, are providers of imaging services overpaid? That question is raised, along with others, in a provocative article by Winter and Ray,¹ two analysts for the Medicare Payment Advisory Commission (MedPAC), that was published in a recent issue of Health Affairs. In examining the growth in imaging and Medicare’s method of paying for imaging, the authors identify several weaknesses in the payment methodology that may lead to inaccurate rates, and they suggest some options to improve the system. While much of the information in the report had been previously published, the MedPAC analysts take this opportunity to reply in the affirmative to the above question. The authors first review the now-familiar data charting the outlier growth of imaging (61%), compared with all physician services (31%), between 2000 and 2005, as paid under the Medicare Physician Fee Schedule (MPFS). In fact, Medicare spending on imaging paid under the MPFS practically doubled between 2000 and 2005, from $6.4 billion to $12.3 billion. After describing the mechanics of physician reimbursement under the MPFS, the authors identify what they characterize as inaccuracies in the distribution of payments among all physicians, even after accounting for the DRA cuts, the discount for contiguous body parts, and impending declines in payment that will accrue through 2010 due to a recent revision to the methodology CMS uses to calculate practice-expense RVUs. Until 2007, payment for practice expenses not involving physician work—including imaging services—was based on physician charges from 1998. CMS began phasing in new methodology in 2007; when it is fully implemented in 2010, this will result in an estimated 9% decline in practice-expense RVUs for all imaging services. This decline notwithstanding, practice-expense RVUs for imaging services may still be overvalued due to the following shortcomings, the authors contend. Three System Weakness One weakness in the system, the authors believe, is the overstatement of equipment costs. Due to the expensive nature of high-end imaging technology, equipment costs account for a large percentage of practice expenses. The authors offer the example of an MRI of the brain, for which equipment costs account for 90% of the total direct cost of the procedure. For a chest radiograph, equipment costs account for just slightly more than half of the direct cost of such a procedure. The per-unit cost of a procedure is based on the number of minutes that the equipment is used multiplied by its cost per minute. The authors posit that recent advances in imaging technology that have resulted in shorter scanning times, such as 64-slice CT and 3T MRI, mean that the times per scan generated in 2002 and 2003 by the AMA’s RVU Update Committee (RUC) may result in an overvaluing of the practice expense for some procedures. They also point out that CMS uses a formula that spreads the cost of the technology over the number of minutes that the scanner is expected to be operating during its useful lifetime to derive the equipment cost per minute. Another fundamental assumption used by CMS is that the equipment is in operation 50% of the time that the site is open for business. The authors report that a MedPAC-sponsored survey in six markets found that MRI scanners are used during 91% of operating hours and CT scanners are used during 73% of operating hours (MedPAC recently voted to recommend that the secretary of HHS increase the equipment-use standard to 90% for MRI, CT, and PET scanners). The authors believe that CMS would be encouraging the efficient use of high-tech equipment with a 90% use rate. A second inequity of the system that favors higher costs for imaging services is the fact that CMS began using new practice-expense data in 2007 for the eight specialties that responded to CMS’ invitation to submit new data. The authors believe that this has resulted in all of those eight specialties, including radiology and cardiology, receiving a higher value for practice expenses than those specialties that did not submit new data. Cardiology’s hourly practice costs, for instance, more than doubled, from $82 per hour to $184 per hour. The authors believe that, had radiology not submitted new practice-expense data, imaging practice-expense RVUs would have declined by 20% instead of 9% by 2010. The authors believe that CMS needs current practice expense data from all specialties in order to achieve an equitable solution. The third point made by the authors relates to the three separate geographic practice-cost indices (GPCIs) used to account for differences in price inputs for physician costs. The authors point out that the GPCI does not recognize that individual services have different shares of inputs for which prices vary geographically (such as nonphysician staff and office space) and for which prices are uniform (such as equipment and supplies). Therefore, for services with above-average costs for equipment, such as imaging, the index adjusts too large a portion of the practice-expense payment; for services with low equipment costs, the index adjusts too little of the cost. The average share of practice expenses related to equipment for all physician services is 14.6%, but equipment and supplies account for 60% of the practice expenses associated with MRI of the brain. In assuming that 14.6% (rather than 60%) of the practice expense of an MRI is related to equipment, the GPCI allows too much of this payment to vary geographically. The authors believe that this may result in overpayment for imaging in high-cost areas and underpayment for imaging in low-cost areas. On the other hand, equipment accounts for just 7.5% of an office visit, so the index allows too small a share of the practice-expense payment for an office visit to vary geographically. The authors suggest that this may result in fewer office visits and excessive imaging in areas with high input costs. MedPAC has discussed an alternative practice-expense GPCI that would exclude equipment and supplies, at least for those services with high equipment costs, such as high-tech imaging. Options for Achieving Equity The authors conclude that inaccuracies in how Medicare pays for the practice-expense portion of physician services may be driving the overutilization of imaging. They ask CMS to request a review of CT and MRI codes by the RUC, to ensure accuracy; to acquire more current practice-cost data for all specialties; and to consider using an alternative GPCI that would recognize that different services have larger portions of costs that do not vary geographically.