Changing Musculoskeletal Extremity Imaging Utilization From 1994 Through 2013: A Medicare Beneficiary Perspective

Transcription

1 Musculoskeletal Imaging Original Research Gyftopoulos et al. Musculoskeletal Extremity Imaging Utilization Musculoskeletal Imaging Original Research Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved Soterios Gyftopoulos 1 Paul Harkey 2 Jennifer Hemingway 3 Danny R. Hughes 3 Andrew B. Rosenkrantz 1 Richard Duszak, Jr. 2 Gyftopoulos S, Harkey P, Hemingway J, Hughes DR, Rosenkrantz AB, Duszak R Jr Keywords: extremity imaging, imaging trends, imaging utilization, MRI, musculoskeletal imaging DOI:1.2214/AJR Received April 4, 217; accepted after revision April 21, 217. A. B. Rosenkrantz and R. Duszak, Jr., are supported by research grants from the Harvey L. Neiman Health Policy Institute. 1 Department of, NYU Langone Medical Center, 66 First Ave, New York, NY 116. Address correspondence to S. Gyftopoulos (Soterios.Gyftopoulos@nyumc.org). 2 Department of and Imaging Sciences, Emory University School of Medicine, Atlanta, GA. 3 Harvey L. Neiman Health Policy Institute, Reston, VA. AJR 217; 29: X/17/ American Roentgen Ray Society Changing Musculoskeletal Extremity Imaging Utilization From Through 213: A Medicare Beneficiary Perspective OBJECTIVE. The objective of our study was to assess temporal changes in the utilization of musculoskeletal extremity imaging in Medicare beneficiaries over a recent 2-year period ( 213). MATERIALS AND METHODS. Medicare Physician Supplier Procedure Summary Master Files from through 213 were used to study changing utilization and utilization rates of the four most common musculoskeletal imaging modalities: radiography, MRI, CT, and ultrasound. RESULTS. Utilization rates (per 1 beneficiaries) for all four musculoskeletal extremity imaging modalities increased over time: 43% (from to 633.6) for radiography, 615% ( ) for MRI, 758% ( ) for CT, and 5% ( ) for ultrasound. Radiologists were the most common billing specialty group for all modalities throughout the 2-year period, maintaining dominant market shares for MRI and CT (84% and 96% in 213). In recent years, the second most common billing group was orthopedic surgery for radiography, MRI, and CT and podiatry for ultrasound. The physician office was the most common site of service for radiography, MRI, and ultrasound, whereas the hospital outpatient and inpatient settings were the most common sites for CT. CONCLUSION. In the Medicare population, the most common musculoskeletal extremity imaging modalities increased substantially in utilization over the 2-decade period from through 213. Throughout that time, radiology remained the most common billing specialty, and the physician office and hospital outpatient settings remained the most common sites of service. These insights may have implications for radiology practice leaders in making decisions regarding capital infrastructure, workforce, and training investments to ensure the provision of optimal imaging services for extremity musculoskeletal care. M usculoskeletal extremity imaging plays a crucial role in the evaluation of a variety of bone, joint, and muscle conditions. Radiography, MRI, CT, and ultrasound contribute important and often complementary information used to make accurate diagnoses that lead to the selection of the most appropriate treatment of this patient population. A range of technologic advances occurred for all these modalities except radiography over the 2-decade period from through 213, and the impact of these new technologies on the diagnostic accuracy for clinically relevant pathologic findings has been well studied [1 4]. However, there remain persistent knowledge gaps in terms of how these advances affect other potentially important metrics such as the volume of imaging that is ordered, the subspecialties that could perform and bill for these services, and the locations where imaging is performed. The objective of this study was to assess temporal changes in the utilization of musculoskeletal extremity imaging for the Medicare fee-for-service population over a recent 2-year period. Materials and Methods This study did not constitute human subjects research and thus did not require institutional review board oversight. The data that we used had no protected health information and have been designated for public use by the Centers for Medicare & Medicaid Services (CMS). Data Billing data for the period from through 213 were obtained from CMS Medicare Physician Supplier Procedure Summary Master Files. These files contain aggregated Medicare Part B AJR:29, November

2 Gyftopoulos et al. Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved billing claims data, including service counts stratified by procedure code using the Healthcare Common Procedure Coding System (HCPCS), place of service, and self-reported provider specialty. The Neiman Imaging Types of Service ( NITOS) radiology claims classification system was used to identify the HCPCS codes that correspond with nonvascular extremity imaging: NITOS body region of extremity excluding codes with a vascular focus [5]. Service counts were identified for MRI, ultrasound, CT, and radiography nonvascular extremity imaging codes for each year and normalized to annual Medicare beneficiary enrollment numbers to obtain utilization rates per 1 Medicare fee-for-service beneficiaries [6 11]. To avoid double counting, we excluded technical only claims, and focused on only services with professional and global claims. So, for example, in the situation of a nonradiologist who owns an MRI unit and bills for technical services, only the professional services billed by the interpreting radiologist would be counted. Annual utilization rates for each modality were stratified by the specialty group that performed the imaging and the place of service. The specialty groups were categorized as radiology, orthopedic surgery, podiatry, primary care, nonphysician providers, emergency medicine, physical medicine and rehabilitation, rheumatology, unclear specialty designation (e.g., multispecialty group practice; hereafter referred to as unclear ), and all other specialties combined (hereafter referred to as other ). The market share of all billed nonvascular extremity imaging was determined for the most common performing specialty group for each modality. Sites of service were categorized as follows: inpatient hospital, hospital outpatient, physician office (including both radiologist and nonradiologist office-based imaging centers), emergency department, and all other specialties combined (hereafter referred to as other ). Temporal changes for all of these parameters were assessed. Statistical Analysis Data analysis was performed using statistics software (SAS, version 9.1, SAS Institute) and Excel 217 for Macintosh (version 15.31, Microsoft). Results Imaging Modality Growth Between and 213, utilization rates per 1 Medicare fee-for-service beneficiaries increased for all four musculoskeletal extremity imaging modalities: radiography, 43% from to 633.6; MRI, 615% from 5.4 to 38.6; CT, 758% from 1.2 to 1.3; and ultrasound, 5% from 1.8 to 1.8 (Table 1). For MRI, the growth was initially more rapid from through, with a subsequent plateau (change of +638% from through, followed by a change of 3% from through 213). In comparison, radiography, CT, and ultrasound showed more consistent growth throughout the 2-year interval. Specialty Distribution by Modality and orthopedic surgery were, in that order, the two most common billing specialties for radiography throughout the time period (Fig. 1). Utilization per 1 beneficiaries for radiology increased 41% from to 314.5, whereas utilization for orthopedic surgery increased 49% from to The market share of all performed radiography examinations remained overall stable during this time period for both specialties (for radiology, ranging from 5.5% to 49.6%; for orthopedic surgery, ranging from 28.4% to 29.5%). was the most common billing specialty for MRI throughout the time period with a 546% increase in utilization per 1 beneficiaries from 5. in to 32.3 in 213 (Fig. 1). maintained a massive market share throughout, ending at 84% in 213. Beginning in, providers with an unclear specialty and orthopedic surgeons had the second and third greatest market shares, respectively, ending in 213 with a 1% share and a 4% share. was the most common billing specialty for CT throughout the time period with an 8% increase in utilization per 1 beneficiaries from 1.1 in to 9.9 in 213 (Fig. 1). maintained a market share of greater than 9% throughout this time period, ending at 96%, with no other specialty having a market share of greater than 1% in any year. TABLE 1: Extremity Imaging in Medicare Fee-for-Service Beneficiaries, by Modality and Billing Specialty Group Modality and Billing Specialty Group 213 Change Between and 213 (%) Radiography Utilization rate per 1 beneficiaries Total examinations a, no. (%) 14,269,75 21,353, ,26,974 (5.5) 1,598,591 (49.6) 47 Orthopedic surgery 4,46,854 (28.4) 6,297,317 (29.5) 56 Primary care 929,462 (6.5) 717,642 (3.4) 23 Nonphysicans 2662 (.) 611,822 (2.9) 22,884 Physical medicine and rehabilitation 749 (.1) 43,688 (.2) 49 Podiatry 1,33,32 (7.2) 1,463,882 (6.9) 42 Rheumatology 29,273 (1.5) 287,718 (1.3) 37 Emergency medicine 66,93 (.5) 126,837 (.6) 9 Unclear 648,881 (4.5) 1,23,861 (4.8) ,94 (.8) 182,22 (.9) 55 MRI Utilization rate per 1 beneficiaries Total examinations a, no. (%) 173,368 1,31, ,268 (93.6) 1,9,158 (83.8) 572 Orthopedic surgery 1394 (.8) 55,126 (4.2) 3855 Primary care 198 (.6) 7855 (.6) 615 Nonphysicans (.) 916 (.1) NA Physical medicine and rehabilitation 6 (.) 1116 (.1) 18,5 Podiatry 36 (.) 2946 (.2) 883 Rheumatology 49 (.) 522 (.4) 1,149 Emergency medicine 12 (.) 1372 (.1) 11,333 Unclear 4692 (2.7) 129,831 (1.) (2.2) 6749 (.5) 77 (Table 1 continues on next page) 114 AJR:29, November 217

3 Musculoskeletal Extremity Imaging Utilization Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved TABLE 1: Extremity Imaging in Medicare Fee-for-Service Beneficiaries, by Modality and Billing Specialty Group (continued) Modality and Billing Specialty Group 213 was the most common billing specialty for ultrasound throughout the time period with a 28% increase in utilization per 1 beneficiaries from 1.3 in to 4. in 213 (Fig. 1). Beginning in, podiatry became the second most common specialty with a 9% increase in utilization from.2 in to 2. in 213. maintained the largest market share throughout the time period, although this market share decreased in half from 74% in to 37% in 213. Podiatry had the second largest market share beginning in, ending at 18% in 213. Change Between and 213 (%) CT Utilization rate per 1 beneficiaries Total examinations a, no. (%) 39, , ,569 (93.4) 333,13 (95.6) 811 Orthopedic surgery 38 (1.) 1692 (.5) 345 Primary care 376 (1.) 1631 (.5) 334 Nonphysicans 1 (.) 85 (.) 84 Physical medicine and rehabilitation (.) 5 (.) NA Podiatry 1 (.) 75 (.) 65 Rheumatology 5 (.) 3 (.) 5 Emergency medicine 7 (.) 396 (.1) 5557 Unclear 1592 (4.1) 9894 (2.8) (.5) 1689 (.5) 732 Ultrasound Utilization rate per 1 beneficiaries Total examinations a, no. (%) 55, , ,96 (73.9) 133,488 (36.8) 226 Orthopedic surgery 218 (.4) 33,895 (9.3) 15,448 Primary care 2349 (4.2) 33,431 (9.2) 1323 Nonphysicans 1 (.) 1,184 (2.8) 1,18,3 Physical medicine and rehabilitation 136 (.2) 16,1 (4.4) 11,738 Podiatry 4 (.) 66,55 (18.2) 1,651,275 Rheumatology 12 (.2) 36,21 (9.9) 35,215 Emergency medicine 23 (.) 258 (.7) 11,117 Unclear 3165 (5.7) 7854 (2.2) (15.2) 23,68 (6.4) 173 Note = increase, = decrease, NA = not applicable. a Paid claims. Site of Service by Modality The physician office was the dominant site of service for radiography with a 4% increase in utilization per 1 beneficiaries from in to in 213 (Fig. 2). The hospital outpatient department was the second most common site for radiography with a 34% increase in utilization from 89.8 in to 12.3 in 213. s were also the most common site of service for MRI, increasing 758% from 2.6 in to 22.3 in and subsequently declining slightly to 19.7 in 213 (Fig. 2). The hospital outpatient department was the second most common site for MRI, increasing 574% from 2.3 in to 15.5 in 213. The hospital outpatient, inpatient, and emergency department settings were the most common sites for CT utilization with all three locations seeing rapid growth during portions of the studied time period (Fig. 2). Hospital outpatient CT utilization per 1 beneficiaries increased 353% from.75 in to 3.4 in 213. Inpatient CT utilization increased 357% from.7 in to 3.2 in, followed by a slight decline ending at 3. in 213. Ultrasound utilization showed a 1644% increase in the physician office setting from.4 in to 7.5 in 213, being by far the most common site in 213 (Fig. 2). The next most common locations for ultrasound in 213 were the hospital outpatient (1.9) and inpatient (.96) settings. Discussion Our comprehensive analysis of musculoskeletal extremity imaging ordering trends using aggregate Medicare fee-for-service beneficiaries over the 2-year period from through 213 elicited several interesting findings. All four primary musculoskeletal imaging modalities increased in utilization during this time period, with MRI experiencing an initial 15-year span of rapid growth followed by a plateau, whereas the other modalities grew more consistently across the full interval. was the most common billing specialty group for all modalities throughout the 2-year period, maintaining dominant market shares for MRI and CT. The physician office was the most common site of service for radiography, MRI, and ultrasound, whereas the hospital outpatient and inpatient settings were the most common sites for CT. The hospital outpatient department was the second most common location for the remaining imaging modalities. The slower growth of radiography compared with the marked growth for the other modalities is somewhat surprising given radiography s initial screening role for numerous extremity conditions, including trauma, chronic pain, and tumors [12 21]. The trajectories for CT and ultrasound suggest that these two modalities are poised to experience continuous growth in the future, whereas MRI has plateaued, remaining relatively stable since, similar to findings of a recent study by Levin et al. [22]. We believe these findings reflect a shift in musculoskeletal management and patient care after the advent of advanced technologies on which physicians increasingly rely. This shift has been seen with cervical spine imaging in the emergency medicine setting where there has been a decrease in the utilization of cervical spine radiography and a concomitant increase in CT from to [23]. A recent thoracic imaging utilization study similarly observed AJR:29, November

4 Gyftopoulos et al. Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved decreasing use of chest radiography and increasing use of CT and CT angiography imaging over a comparable time period [24]. Over all 2 years, radiology was the dominant billing specialty for all imaging modalities. had dominant market shares for MRI and CT but a much smaller share for ultrasound. Radiologists relative market share for ultrasound was cut in half over the study s time period, and podiatrists, in particular, showed marked growth in utilization. The reasons for this rapid growth could be related to the various advantages of ultrasound imaging, including improved accessibility, portability, and low cost. However, a number of studies in the podiatry literature have touted ultrasound as a mechanism to enhance practice revenue, suggesting that this could be a dominant driver of this utilization change [25 27]. The rapid growth in utilization of ultrasound by podiatrists led to a change in Current Procedural Terminology (CPT) coding and valuation for extremity ultrasound services in, which coincided with a sudden drop in billing for these services by podiatrists in [28]. The physician office was consistently the most common site for imaging for all modalities except CT, which was most commonly used in the hospital outpatient and inpatient settings. The hospital outpatient setting was the second most common location for the remaining imaging modalities. If current trends continue, the hospital outpatient department could surpass the physician office as the most common site for MRI. Our findings are concordant with those of prior studies showing the emergence of the hospital outpatient department and decreasing use of the physician office for advanced imaging such as MRI [22, 29]. There are several potential implications of our study s findings on the future of extremity musculoskeletal imaging. Our work suggests that hospital administrators and radiology group leaders seeking to improve patient access should consider investing in advanced imaging modalities given what appears to be an increased and ongoing reliance on these tools for the care of patients with musculoskeletal conditions. Additionally, radiology leaders and management should focus resources on physician offices and hospital outpatient departments because these are the most common sites for musculoskeletal imaging. Finally, nonvascular extremity ultrasound is now being performed and billed more commonly by a wider range of nonradiology specialties, potentially without the same standards and qualifications as radiologists who perform and interpret this type of imaging after lengthy training and board certification in diagnostic imaging. Thus, payers and accreditation bodies should reexamine technical and performance standards for nonvascular extremity ultrasound imaging given the potential patient hazards of increased utilization of lower-value and lowerquality imaging. There were two main limitations of our study, both of which are related to the generalizability of our results. First, we studied only the Medicare fee-for-service population, which consists primarily of patients 65 years old or older. Our findings, which reflect imaging in this older patient population, may not be generalizable to a younger subset of patients. Second, our findings may not apply to patients with private insurance or patients without insurance who, besides typically being younger, may not always have the same access to imaging (e.g., precertification requirements for advanced imaging). Future work with other private payer datasets, such as the MarketScan Database (Truven Health Analytics), would be necessary to see if the trends reflected in our study would be seen in private payer and uninsured populations who are younger than 65 years old. Additional limitations of our study are associated with the data used. Provider claims rely on self-reported CPT and specialty codes. Also, the data cannot be used to assess patient outcomes or the quality of care. Conclusion Our comprehensive analysis of musculoskeletal extremity imaging ordering trends over the 2-year period from through 213 showed increases in both volume and per-beneficiary utilization for the main imaging modalities: radiography, MRI, CT, and ultrasound. was the dominant billing specialty for all imaging modalities with dominant market shares for MRI and CT. The physician office and hospital outpatient department were the most common locations for musculoskeletal imaging. These insights may have implications for radiology practice leaders in making decisions regarding capital infrastructure, workforce, and training investments to ensure the provision of optimal imaging services for extremity musculoskeletal care. References 1. Klauser AS, Peetrons P. Developments in musculoskeletal ultrasound and clinical applications. Skeletal Radiol ; 39: Mallinson PI, Coupal TM, McLaughlin PD, Nicolaou S, Munk PL, Ouellette HA. Dual-energy CT for the musculoskeletal system. 216; 281: Ramnath RR. 3T MR imaging of the musculoskeletal system. Part II. Clinical applications. Magn Reson Imaging Clin N Am ; 14: Ramnath RR. 3T MR imaging of the musculoskeletal system. Part I. Considerations, coils, and challenges. Magn Reson Imaging Clin N Am ; 14: Boileau P, Villalba M, Hery JY, Balg F, Ahrens P, Neyton L. Risk factors for recurrence of shoulder instability after arthroscopic Bankart repair. J Bone Joint Surg Am ; 88: Duszak R Jr, Kim DH, Pickhardt PJ. Expanding utilization and regional coverage of diagnostic CT colonography: early Medicare claims experience. J Am Coll Radiol ; 8: Parker L, Levin DC, Frangos A, Rao VM. Geographic variation in the utilization of noninvasive diagnostic imaging: national Medicare data,. AJR ; 194: Levin DC, Rao VM, Parker L, Frangos AJ, Intenzo CM. Recent payment and utilization trends in radionuclide myocardial perfusion imaging: comparison between self-referral and referral to radiologists. J Am Coll Radiol ; 6: Levin DC, Rao VM, Parker L, Frangos AJ, Sunshine JH. Recent trends in utilization rates of abdominal imaging: the relative roles of radiologists and nonradiologist physicians. J Am Coll Radiol ; 5: Moreno CC, Hemingway J, Johnson AC, Hughes DR, Mittal PK, Duszak R Jr. Changing abdominal imaging utilization patterns: perspectives from Medicare beneficiaries over two decades. J Am Coll Radiol 216; 13: Harvey L. Neiman Health Policy Institute website. Duszak R, Hughes DR. How many Medicare enrollees? CMS methodological changes and implications for research. commentary/how-many-medicare-enrollees-cmsmethodological-changes-and-implications-forresearch/. Published March 7, 216. Accessed March 3, Berquist TH, Dalinka MK, Alazraki N, et al. Bone tumors: American College of ACR Appropriateness Criteria. 2; 215( suppl): el-khoury GY, Dalinka MK, Alazraki N, et al. Metastatic bone disease: American College of ACR 116 AJR:29, November 217

5 Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved 2; 215(suppl): Musculoskeletal knee. J Am Coll Extremity Radiol 215; Imaging 12: Utilization 26. Poggio A. Maximizing office revenue: what you 14. DeSmet AA, Dalinka MK, Alazraki N, et al. Chronic ankle pain: American College of ACR 2; 215(suppl): Dalinka MK, Alazraki N, Berquist TH, et al. Chronic wrist pain: American College of ACR 2; 215(suppl): el-khoury GY, Dalinka MK, Alazraki N, et al. Chronic foot pain: American College of ACR 2; 215(suppl): Berquist TH, Dalinka MK, Alazraki N, et al. Chronic hip pain: American College of ACR 2; 215(suppl): Wise JN, Daffner RH, Weissman BN, et al. ACR Appropriateness Criteria on acute shoulder pain. J Am Coll Radiol ; 8: Tuite MJ, Kransdorf MJ, Beaman FD, et al. ACR Appropriateness Criteria acute trauma to the 2. Mosher TJ, Kransdorf MJ, Adler R, et al. ACR Appropriateness Criteria acute trauma to the ankle. J Am Coll Radiol 215; 12: Bancroft LW, Kransdorf MJ, Adler R, et al. ACR Appropriateness Criteria acute trauma to the foot. J Am Coll Radiol 215; 12: Levin DC, Rao VM, Parker L. Trends in outpatient MRI seem to reflect recent reimbursement cuts. J Am Coll Radiol 215; 12: Gan G, Harkey P, Hemingway J, Hughes DR, Duszak R Jr. Changing utilization patterns of cervical spine imaging in the emergency department: perspectives from two decades of national Medicare claims. J Am Coll Radiol 216; 13: Kamel SI, Levin DC, Parker L, Rao VM. Utilization trends in noncardiac thoracic imaging, J Am Coll Radiol 217; 14: Leone A. Nine ways to enhance office revenue. Podiatry Today website. article/5916. Published August. Accessed March 19, 217 should know. Podiatry Today website. www. podiatrytoday.com/maximizing-office-revenuewhat-you-should-know. Published December. Accessed March 19, Weil L Jr. How to improve patient care while increasing revenue. Podiatry Today website. www. podiatrytoday.com/how-improve-patient-care-whileincreasing-revenue. Published December 214. Accessed March 19, American College of website. ACR Coding Source September October : projected impact on the Medicare physician fee schedule for. Advocacy/Economics-Health-Policy/Billing-Coding/ Coding-Source-List//Sept-Oct-/Projected- Impact-on-the-Medicare-Physician-Fee- Schedulefor-. Accessed March 19, Patel BP, Levin DC, Parker L, Rao VM. The shift in outpatient advanced imaging from private offices to hospital facilities. J Am Coll Radiol 215; 12: Utilization Rate per 1 Beneficiaries Utilization Rate per 1 Beneficiaries Orthopedic surgery Orthopedic surgery AJR:29, November A B Fig. 1 Graphs show annual extremity imaging by modality in Medicare fee-for-service beneficiaries. Numbers of examinations reflect paid claims. Utilization rates are per 1 beneficiaries. A, Radiography. B, MRI. (Fig. 1 continues on next page)

6 Gyftopoulos et al. Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved Utilization Rate per 1 Beneficiaries Utilization Rate per 1 Beneficiaries Utilization Rate per 1 Beneficiaries Orthopedic surgery Podiatry AJR:29, November 217 C D A Fig. 1 (continued) Graphs show annual extremity imaging by modality in Medicare fee-for-service beneficiaries. Numbers of examinations reflect paid claims. Utilization rates are per 1 beneficiaries. C, CT. D, Ultrasound. Fig. 2 Graphs show annual extremity imaging sites of service by modality in Medicare fee-forservice beneficiaries. Utilization rates are per 1 beneficiaries. A, Radiography. (Fig. 2 continues on next page)

7 Musculoskeletal Extremity Imaging Utilization Downloaded from by on 3/3/18 from IP address Copyright ARRS. For personal use only; all rights reserved Utilization Rate per 1 Beneficiaries Utilization Rate per 1 Beneficiaries Utilization Rate per 1 Beneficiaries AJR:29, November B C D Fig. 2 (continued) Graphs show annual extremity imaging sites of service by modality in Medicare fee-for-service beneficiaries. Utilization rates are per 1 beneficiaries. B, MRI. C, CT. D, Ultrasound.