Search

Purpose: In 2009, the results of two randomized controlled trials refuting the effectiveness of vertebroplasty compared with sham procedures were published in a leading journal. The purpose of the present study was to evaluate the impact of these randomized trials on subsequent volume and utilization rates of vertebral augmentation (VA) in the United States.Methods: Using nationwide Medicare Part B databases, Current Procedural Terminology, version 4, codes for thoracic and lumbar vertebroplasty and kyphoplasty were studied from 2006 to 2013 (codes 22520 to 22525). The total volumes of procedures were determined and utilization rates were calculated. Volumes and rates by provider specialty were also studied.Results: The total volume of VA procedures peaked in 2008 at 101,807 and thereafter fell steadily to 80,940 in 2013. The utilization rates per 100,000 beneficiaries also showed a similar trend. Radiologists performed the largest number of VA procedures in 2013 (33,618 procedures [42%]), followed by orthopedic surgeons (19,886 procedures [25%]). After 2009, vertebroplasty volumes decreased sharply. Kyphoplasty volumes increased in 2011, after an initial decrease in 2010. The divergent trend in the volumes of the two procedures persisted through 2013.Conclusions: After the publication of the two trials' results in 2009, vertebroplasty volumes and rates decreased sharply. However, there is an emerging trend toward performing more kyphoplasty procedures, mitigating the decrease in total volume of VA procedures. Radiologists have the strongest role in performing these procedures among all medical specialties. [ABSTRACT FROM AUTHOR]

Purpose: To compare recent trends in the use of percutaneous and surgical approaches to treating abdominal abscesses in a large population.Methods: The nationwide Medicare Physician/Supplier Procedure Summary Master Files for 2001 through 2013 were searched. Current Procedural Terminology-4 codes were selected for the four types of abdominal abscesses that had distinct codes for both open surgical and percutaneous drainage-appendiceal, peritoneal, subphrenic, and liver. Medicare specialty codes were used to determine if the procedures were performed by radiologists or other nonradiologist physicians. Trends in use of the two approaches were compared.Results: In 2001, a total of 14,068 abdominal abscesses were drained percutaneously. This volume increased progressively every year thereafter, reaching 28,486 in 2013 (+102%). Open surgical drainage volume was 8,146 in 2001, decreasing progressively to 6,397 in 2013 (-21%). In 2001, 63% of all abdominal abscesses had been drained percutaneously; by 2013, this figure had risen to 82%. In 2001, radiologists had performed 90% of all percutaneous abdominal abscess drainages; this percentage share increased to 97% in 2013. Of all abdominal abscesses treated in 2013 in Medicare patients, 79% were treated by radiologists.Conclusions: Use of percutaneous drainage of abdominal abscesses has steadily increased, whereas use of open surgical drainage has declined. The vast majority of these abscesses are now treated percutaneously. Radiologists are a strong majority of those performing the procedures. Although this database does not provide information on outcomes, percutaneous drainage is another good example of radiology-related value. [ABSTRACT FROM AUTHOR]

Purpose: To study recent outpatient imaging trends in private offices and hospital outpatient departments (HOPDs), to determine if shifting between the two has occurred. Concern is currently focused on whether reduced reimbursements and other factors might lead to a shift to higher-cost HOPDs.Methods: The nationwide Medicare Physician/Supplier Procedure Summary Master Files for 2001 to 2013 were studied. All Current Procedural Terminology codes for MRI, echocardiography, nuclear medicine, ultrasound, and CT were selected, and procedure utilization rates per 1,000 Medicare beneficiaries were determined for each year. Medicare location codes identified the settings where the scans were performed.Results: Total utilization rates, per 1,000 beneficiaries, of all these examination types in private offices, grew from 478 in 2001, to 874 in 2008 (+83%), and then declined to 503 in 2011 (-42%), primarily as a result of code bundling. No further bundling occurred in 2012 or 2013, but the decline continued in those years, to 462. In HOPDs, the total rate rose from 416 in 2001, to 523 in 2008 (+26%), followed by similar bundling-related declines, to 418 (-20%) in 2011. But in 2012 and 2013, in contrast to private office trends, the HOPD rate increased to 447. The ratio of private office to HOPD advanced imaging was 1.67 in 2008, declining to 1.03 in 2013. In addition, individual modality shifts away from offices and into HOPDs were quite apparent.Conclusions: In recent years, a shift has occurred in utilization of all advanced imaging modalities, from private offices to HOPDs. This change could portend a loss of access for patients and an increase in costs. [ABSTRACT FROM AUTHOR]

Purpose: Cerebral catheter angiography and endovascular neurointerventions (ENIs) were developed and refined by early pioneers in neuroradiology. Recently, with developments in the safety and efficacy of ENIs, other physician specialists have expressed strong interest in performing these procedures. Our purpose was to compare volume and utilization of ENIs, among the various specialties, from 2000 to 2013.Methods: Data from the Medicare Part B Physician/Supplier Procedure Summary Master Files for 2000 to 2013 were used to study ENI volume and utilization rates, by radiologists, neurosurgeons, neurologists, vascular surgeons, cardiologists, and other physicians.Results: From 2000 to 2013, the volume of intracranial ENIs increased: overall, from 2,439 to 7,424; for radiologists, from 1,956 to 3,993; and for neurosurgeons, from 237 to 2,900. Although cardiologists did not perform many intracranial ENIs in these years, they performed most of the carotid artery stenting procedures (4,097, which is 51% of the total 8,201 performed in 2013).Conclusions: Radiologists continue to maintain a strong presence in the field of neurointerventional radiology, particularly in percutaneous intracranial interventions, performing 51% of all intracranial procedures in 2013, down from 80% in 2000. However, neurosurgeons have made substantial inroads into ENI procedures, with their volume increasing from 10% to 33%, from 2000 to 2013. The overall volume of ENIs rose steadily from 2000 to 2013. [ABSTRACT FROM AUTHOR]

Purpose: To determine whether MR angiography (MRA) and CT angiography (CTA) have replaced diagnostic catheter angiography (DCA) in diagnosing peripheral arterial disease. Methods: Medicare Part B databases for 2002--2013 were reviewed. Current Procedural Terminology codes for extremity MRA, CTA, and DCA were selected. Physician specialty codes were used to classify providers as radiologists, cardiologists, or surgeons. Utilization rates per 100,000 Medicare beneficiaries were calculated. Results: Among all specialties, the combined utilization rate of all 3 types of angiography increased from 917 per 100,000 in 2002 to 1,261 in 2006 (+38%), after which it remained stable until 2010, and then declined to 1,010 in 2013. The overall rate of MRA and CTA together increased from 89 in 2002 to 440 in 2006 (+394%), after which it leveled off, and then gradually decreased to 331 in 2013. In 2013, 33% of the total procedures were MRA or CTA, up from 10% in 2002. Radiologists performed >85% of MRA and CTA examinations. Among radiologists, the DCA utilization rate decreased by 75% from 2002 to 2013, whereas among cardiologists and surgeons together, the overall DCA utilization rate increased by 64% from 2002 to 2010 before dropping somewhat in 2011. Conclusions: Among radiologists, MRA and CTA have replaced DCA in diagnosing peripheral arterial disease. Although overall utilization of DCA has remained steady, it has risen sharply among cardiologists and surgeons, while dropping sharply among radiologists. Given the increased utilization of DCA among cardiologists and surgeons despite noninvasive alternatives, self-referral continues to be of concern in the setting of increasing health care costs. [ABSTRACT FROM AUTHOR]

Purpose To determine whether recent reimbursement cuts have resulted in a shift of outpatient MRI from private offices to hospital outpatient departments (HOPDs); and to study office MRI utilization trends among radiologists and other specialists. Methods The Medicare Part B Physician/Supplier Procedure Summary Master Files were used. MRI codes were aggregated, and total MRI volumes from 2002 to 2012 were studied. Medicare place-of-service codes were used to identify studies performed in private offices and HOPDs and create trend lines. Specialty codes were used to categorize private office MRI users as radiologists, orthopedic surgeons, other physicians, and independent diagnostic testing facilities. Results Medicare office and HOPD utilization of MRI (all specialties) rose rapidly from 2002 to 2006, reaching 2,727,807 in offices and 2,355,641 in HOPDs. Thereafter, office volume steadily declined, whereas HOPD volume steadily increased. By 2012, more studies were done in HOPDs than in offices. Over the entire period from 2002 and 2012, office MRI volume among radiologists increased 27%, compared with 216% among orthopedic surgeons and 124% among other physicians. Conclusions Although the majority of Medicare outpatient MRI studies had previously been performed in private offices, recent years brought a shift, with more now being performed in HOPDs. This change will increase costs to payers, because reimbursements to HOPDs are generally higher than those to offices. Although radiologists perform the majority of MRI exams that are conducted in private offices, the rate of growth for such exams from 2002 to 2012 was considerably higher among orthopedic surgeons and other physicians than among radiologists. [ABSTRACT FROM AUTHOR]

Purpose The aim of this study was to examine recent trends in imaging utilization in emergency departments (EDs) in the Medicare population. Methods The 2002 to 2012 Medicare Part B databases were used. Imaging studies were categorized by modality. Medicare’s place-of-service codes identified those studies performed in ED patients. Specialty codes identified the specialties of the interpreting physicians. Utilization rates per 1,000 Medicare beneficiaries were calculated. Trends were assessed in plain radiography (XR), CT, noncardiac ultrasound, MRI, and nuclear medicine. Results XR and CT were the most widely used modalities in ED patients. From 2002 to 2012, the XR utilization rate per 1,000 increased from 248.7 to 320.0 (+29%), and CT increased from 57.2 to 147.9 (+159%). Utilization rates of the other modalities were much lower. Ultrasound increased from 9.5 to 21.0 (+121%), while MRI increased from 1.4 to 5.1 (+264%). Growth in these 4 modalities was continuous and did not show the flattening that has characterized the utilization trends in other places of service. Nuclear medicine use was very low and remained essentially flat. During the study period, CT accrued 91 new examinations per 1,000, followed by XR at 71 and ultrasound at 11.5. The vast majority of examinations were interpreted by radiologists. Conclusions Despite the cessation of overall utilization growth of the various modalities in recent years, ED utilization rates continued to increase. The greatest increases, in terms of accrued new examinations per 1,000, were seen in CT and XR. This suggests that radiologists and ED physicians need to work together to better manage imaging utilization. [ABSTRACT FROM AUTHOR]

Purpose: In 2009, the add-on codes for spectral Doppler and color flow Doppler echocardiography were bundled into the code for primary transthoracic echocardiography. The relative value units for the new single code were substantially lower than the previous sum for the 3 codes. The purpose of this study was to see how this affected the distribution of outpatient echocardiographic studies between cardiology offices and hospital outpatient departments (HOPDs). Methods: The 2005 to 2011 Medicare databases were used. All echocardiography Current Procedural Terminology codes were selected. Specialty codes identified those done by cardiologists (who do most echocardiographic studies). Place-of-service codes identified those done in offices and HOPDs. Procedure volumes and utilization rates per 1,000 were determined each year before and after bundling occurred in 2009. Results: Cardiologists' office echocardiography utilization rate rose from 219.5 per 1,000 in 2005 to 257.1 in 2008 (+17%), then dropped to 100.0 in 2009 (−61%) because of bundling. Their HOPD echocardiography rate rose from 72.2 in 2005 to 76.5 in 2008 (+6%), then dropped to 35.0 in 2009 (−54%). From 2009 to 2011, cardiologists' office echocardiography rate dropped again from 100.0 to 88.8 (−11%), while their HOPD rate increased from 35.0 to 46.1 (+32%). Conclusions: Echocardiography code bundling produced the expected sharp drop in outpatient claims from cardiologists in 2009. But after bundling, office echocardiography rates continued to drop, while HOPD rates increased. It seems that in this instance, code bundling led to the closure of many cardiology offices and a resultant shift of echocardiography from that lower cost setting to the higher cost HOPD setting. [Copyright &y& Elsevier]

Purpose: Recent proliferation of mobile diagnostic ultrasound (US) units and improved resolution have allowed for widespread use of US by more providers, both for diagnosis and US-guided procedures (USGP). This study aims to document recent trends in utilization for USGP in the Medicare population. Methods: Source data were obtained from the CMS Physician Supplier Procedure Summary Master Files from 2004 to 2010. Allowed billing claims submitted for USGP were extracted and volume was analyzed by provider type and setting. Compound annual growth rates were calculated. Results: The total utilization rate for all USGP was 2,425 per 100,000 in 2004 and 4,870 in 2010, an increase of 100.8% (+2,445 per 100,000) with a compound annual growth rate of 12.3%. The year 2010 represents the first year that nonradiologists as a group performed more USGP than radiologists, at 922,672 versus 794,497 examinations, respectively. Nonradiologists accounted for 72.2% (599,751 of 830,925) of the USGP volume growth from 2004 to 2010. Most 2010 claims were submitted by radiologists (n = 794,497; 46.3%) and surgeons (n = 332,294; 19.4%). The largest overall volume increases from 2004 to 2010 were observed among radiologists, surgeons, anesthesiologists, rheumatologists, midlevel providers, primary care physicians, nonrheumatologist internal medicine subspecialists, and the aggregate of all other provider types. Conclusion: The year 2010 represents the first year that nonradiologists performed more USGP than radiologists. From 2004 to 2010, radiologists and surgeons experienced only modest growth in USGP volume, whereas several other provider types experienced more rapid growth. It is likely that many procedures that were previously performed without US guidance are now being performed with US guidance. [Copyright &y& Elsevier]

Purpose: The rate of thoracic biopsies overall and by type of biopsy from 1998 to 2010 is not known. The aim of this study was to examine the utilization rate of various types of thoracic biopsies within the Medicare population. Methods: Source data was obtained from the CMS Physician/Supplier Procedure Summary Master Files from 1998 to 2010. Allowed primary claims submitted for percutaneous thoracic biopsy, bronchoscopic thoracic biopsy, and surgical thoracic biopsy were extracted. Annual volume and utilization rates were calculated and analyzed by biopsy type and provider type. Results: Total thoracic biopsy volume in 1998 was 176,125 and in 2010 was 167,911 (−4.7%). The utilization rate for all thoracic biopsies decreased from 5.47 per 1,000 in 1998 to 4.76 per 1,000 (−13.0%) in 2010. The percutaneous biopsy utilization rate increased 3.6% from 1998 to 2010, while the rate for surgical biopsy decreased by 20.9% and the rate of bronchoscopic biopsy decreased by 19.6% during the same time period. In 2010, radiologists performed 96.4% (58,679) of all percutaneous biopsies. Radiologists' thoracic biopsy market share increased from 26.2% (46,084 of 176,125) in 1998 to 35.0% (58,700 of 167,911) in 2010 (+33.6%). Conclusions: The overall rate of thoracic biopsy decreased from 1998 to 2010, with fewer biopsies being performed surgically and bronchoscopically and more biopsies being performed percutaneously. Radiologists are responsible for an increasing role in thoracic diagnosis, as they have increased market share of thoracic biopsies. These findings may be the result of changing trends toward less invasive procedures, changing patterns of reimbursement, and increased availability of percutaneous biopsy. [Copyright &y& Elsevier]

Purpose: The aim of this study was to determine what proportion of noninvasive diagnostic imaging (NDI) work done by radiologists occurred in each of the 4 primary places of service where imaging is conducted. Methods: Medicare's Physician/Supplier Procedure Summary Master Files for 2000 to 2011 were the data source. Specialty codes were used to identify radiologists, and place-of-service codes identified studies done in hospital outpatient facilities, hospital inpatient facilities, private offices, and emergency departments (EDs). The applicable total professional component relative value units (RVUs) were assigned to each NDI Current Procedural Terminology code, and RVU rates per 1,000 Medicare beneficiaries were calculated. RVU rates reflect workload and costs and are therefore a better metric than utilization rates based on volume. Results: From 2000 to 2006, radiologists' RVU rates per 1,000 Medicare beneficiaries increased in each of the 4 primary venues. However, from 2006 to 2011, rates remained essentially flat in hospital outpatient and inpatient facilities and offices but continued to increase in EDs. Absolute RVU rate increases from 2000 through 2011 were 289 in hospital outpatient facilities, 218 in EDs, 194 in private offices, and 99 in inpatient facilities. In 2011, 19% of radiologists' workload occurred in offices; the remainder was conducted in the 3 hospital settings. Twice as much elective outpatient NDI work by radiologists was done in hospital outpatient facilities as in radiologists' private offices. Conclusions: Radiologists' workload in hospital outpatient and inpatient facilities and offices grew from 2000 through 2006, but no further growth occurred thereafter. EDs were the only venue where growth continued. The vast proportion of radiology NDI RVUs (81% in 2011) are produced in hospital settings. [Copyright &y& Elsevier]

Purpose: Diagnostic imaging services have been repeatedly targeted as a source of excess health care expenditure. In particular, MRI is considered a high-tech and high-cost imaging service that saw rapid increases in utilization in the early 2000s. However, the most recent trends in the utilization of MR are not known. The aim of this study was to quantify trends in MR utilization overall and by body system from 1998 to 2010 in the Medicare population. Methods: Medicare Part B data sets were obtained for 1998 to 2010 for all MR examinations performed in the Medicare population. Using Current Procedural Terminology codes, the total volume and utilization rates of all MR examinations were tabulated for each year of the study period. MR volume was then categorized by body system. Results: The utilization rate of MR examinations in the Medicare population was 73 per 1,000 beneficiaries in 1998, increased to a peak of 189 in 2008, and decreased to 183 in 2010. The compound annual growth rate from 1998 to 2008 was 10%. The utilization rate in 2010 represents a decrease of 3.1% from the 2009 utilization rate. The most frequently imaged body section in every year was the head, which accounted for 2,404,250 examinations in 2010, 37.3% of all MR examinations in that year. Conclusions: The overall MRI utilization rate sharply increased from 1998 until 2008 but then decreased in each of the next 2 years. A similar trend was noted for MR examinations performed in most body sections. These trends are likely to be the result of a number of possible causative factors. [Copyright &y& Elsevier]

Purpose: The aims of this study were to examine recent trends in the utilization of radionuclide myocardial perfusion imaging (MPI) and to reflect on their causes and their implications for radiologists. Methods: Nationwide Medicare Part B databases for 2000 through 2010 were used. Codes for primary MPI studies (including PET) were selected. Medicare specialty codes were used to identify MPI examinations done by radiologists, cardiologists, and other physicians. Place-of-service codes were used to identify examinations performed in offices versus hospital settings. Utilization rates per 1,000 fee-for-service beneficiaries were calculated. Trends were assessed by place of service and specialty. Results: The overall MPI utilization rate rose from 2000 through 2004, followed by a period of stabilization from 2005 to 2008. A peak of 88.0 per 1,000 was reached in 2006. In 2009 and 2010, a decline occurred, with the rate dropping by 13% to 76.9. In private offices, cardiologists'' utilization grew rapidly from 2000 through 2006, but growth stopped thereafter. Their rate peaked in 2008 at 50.6 but dropped to 44.4 by 2010 (−12%). Radiologists'' role in office MPI was minimal. In hospital settings, radiologists predominated in 2000. Their rate remained stable through 2004 but thereafter began to decline steadily, dropping by 35% by 2010. Cardiologists'' hospital-based utilization rate rose gradually, then flattened, but began to rise in 2009 and 2010. By 2010, cardiologists performed more hospital MPI examinations than radiologists. Conclusions: Radiologists'' initially predominant role in hospital-based MPI has eroded recently, while that of cardiologists has strengthened. This seems related to a shift among cardiologists away from office practice and into hospital affiliations. [Copyright &y& Elsevier]