Cost and radiation savings of partial substitution of ultrasound for CT in appendicitis evaluation: a national projection.

OBJECTIVE: The costs of an ultrasound-CT protocol and a CT-only protocol for an appendicitis evaluation are compared. For the ultrasound-CT protocol, patients with right lower quadrant abdominal pain undergo an ultrasound examination. If it is positive for appendicitis, they are sent directly to surgery, avoiding CT. MATERIALS AND METHODS: A comparative effectiveness research study was conducted. The costs of imaging tests, excess surgeries, and excess surgical deaths for the ultrasound-CT protocol and the costs of imaging tests and excess cancer deaths in the CT-only protocol were estimated. Data sources were Centers for Medicare & Medicaid Services (CMS) datasets, national hospital discharge surveys, radiology information system cases, and U.S. Census Bureau life tables. A meta-analysis and sensitivity analyses were also conducted. RESULTS: The meta-analysis showed a positive predictive value of 92.5% for CT and 91.0% for ultrasound. Analysis of CMS files showed that utilization of CT was almost exactly 2.0 examinations (one abdominal and one pelvic) per patient and for ultrasound was almost nil. The cost of this imaging protocol was $547 per patient, whereas the cost of a limited ultrasound study would be $88 per patient. For the total U.S. population, the cost savings in imaging minus the cost of extra surgeries and extra surgical deaths is $24.9 million per year. Following model VII proposed by the Committee on the Biological Effects of Ionizing Radiation (BEIR), which is known as "BEIR VII," the avoidance of a 12.4-mSv exposure for 262,500 persons would prevent 180 excess cancer deaths. The value of the years of life lost would be $339.5 million. The sensitivity analyses indicate that the cost savings are robust. CONCLUSION: An ultrasound-CT protocol for appendicitis evaluation offers potentially large savings over the standard CT-only protocol. There are moderate savings from using a less expensive imaging technique despite extra surgeries and large savings from radiation exposure avoided.

VPAC1 receptors for imaging breast cancer: a feasibility study.

UNLABELLED: VPAC1 encodes G-protein-coupled receptors expressed on all breast cancer (BC) cells at the onset of the disease, but not on benign lesions. Our extensive preclinical studies have shown that (64)Cu-TP3805 has a high affinity for VPAC1, is stable in vivo, and has the ability to distinguish spontaneously grown malignant BC masses from benign lesions. Our long-term goal is to develop (64)Cu-TP3805 as an agent to perform in vivo histology, to distinguish malignant lesions from benign masses noninvasively and thereby avoid patient morbidity and the excess economic costs of benign biopsies. METHODS: (18)F-FDG obtained commercially served as a control. (64)Cu-TP3805 was prepared using a sterile kit containing 20 µg of TP3805. Radiochemical purity and sterility were examined. Nineteen consenting women with histologically proven BC were given 370 MBq of (18)F-FDG. One hour later, 6 of these patients were imaged with PET/CT and 13 with positron emission mammography (PEM). Two to 7 d later, 6 PET/CT patients received 111 MBq (± 10%) (n = 2), 127 MBq (± 10%) (n = 2), or 148 MBq (± 10%) (n = 2) of (64)Cu-TP3805 and were imaged 2 and 4 h later. Thirteen PEM patients received 148 MBq (± 10%) of (64)Cu-TP3805 and were imaged 15 min, 1 h, 2 h, and 4 h later. Standardized uptake value (SUV) was calculated for PET/CT patients, and PUV/BGV (PEM uptake value/background value) was calculated for PEM patients. Tumor volume was also calculated. RESULTS: The radiochemical purity of (64)Cu-TP3805 was 97% ± 2%, and specific activity was 44.4 GBq (1.2 Ci)/µmol. In 19 patients, a total of 24 lesions were imaged (15 invasive ductal carcinoma, 1 high-grade mammary carcinoma, 3 lobular carcinoma, 1 invasive papilloma, and 4 sentinel lymph nodes). All lesions were unequivocally detected by (64)Cu-TP3805 and by (18)F-FDG. The average tumor volume as determined by PET/CT with (64)Cu-TP3805 was 90.6% ± 16.1% of that with (18)F-FDG PET/CT, and the average SUV was 92% ± 26.4% of that with (18)F-FDG. For PEM, the tumor volume with (64)Cu-TP3805 was 113% ± 37% of that with (18)F-FDG and the PUV/BGV ratio was 97.7% ± 24.5% of that with (18)F-FDG. CONCLUSION: (64)Cu-TP3805 is worthy of further investigation in patients requiring biopsy of suggestive imaging findings, to further evaluate its ability to distinguish malignant lesions from benign masses noninvasively.

Single-center comparison of three chemoembolization regimens for hepatocellular carcinoma.

PURPOSE: Transarterial chemoembolization regimens for hepatocellular carcinoma (HCC) vary, without a gold-standard method. The present study was performed to evaluate outcomes in patients with HCC treated with doxorubicin/ethiodized oil (DE), cisplatin/doxorubicin/mitomycin-c/ethiodized oil (CDM), or doxorubicin drug-eluting beads (DEBs). MATERIALS AND METHODS: Patients received the same regimen at all visits, without crossover. Groups were compared based on Child-Pugh disease status, tumor/node/metastasis stage, and Barcelona Clinic Liver Cancer stage. Imaging outcomes were assessed based on modified Response Evaluation Criteria in Solid Tumors to calculate tumor response (ie, sum of complete and partial response), progressive disease (PD), and time to progression (TTP). RESULTS: A total of 228 infusions were performed in 122 patients: 59 with DE, 30 with CDM, and 33 with DEBs. The groups had similar Child-Pugh status (P = .45), tumor/node/metastasis stages (P = .5), and Barcelona Clinic Liver Cancer scores (P = .22). Follow-up duration was similar among groups (P = .24). Patients treated with DE underwent significantly more treatments (2.3 ± 1.4) than those treated with CDM (1.6 ± 0.7; P = .004) or DEBs (1.4 ± 0.6; P<.0001). Compared with DE (51%), tumor response was significantly more common with CDM (84%; P = .003) or DEBs (82%; P = .004). PD was significantly more likely with DE (37%) than with CDM (13%; P = .02) or DEBs (9%; P = .004). TTP was similar between groups (P = .07). CDM and DEBs were similar in regard to disease progression (P = .6) and response (P = .83). CONCLUSIONS: During a similar follow-up period, patients treated with CDM or DEB chemoembolization showed a significantly higher response rate and a lower incidence of tumor progression, with fewer required treatment sessions, than those treated with DE chemoembolization.

Noncardiac point-of-care ultrasound by nonradiologist physicians: how widespread is it?

PURPOSE: The use of point-of-care (POC) ultrasound by nonradiologist physicians has recently been advocated. The aim of this study was to see how widespread this practice is. METHODS: The Medicare Part B databases for 2004 to 2009 were used. Global and professional component claims for noncardiac ultrasound were tabulated, and utilization rates per 1,000 beneficiaries were calculated. Provider specialty was determined. Utilization rates of ultrasound by radiologists and other specialists were compared, and changes over the years were studied. RESULTS: In 2009, 425.3 Medicare noncardiac ultrasound examinations per 1,000 beneficiaries were performed (+21% since 2004). Of these, radiologists performed 233.7 (55%), and another 15.6 (4%) were done at independent diagnostic testing facilities, for which provider specialty could not be determined. The remaining 175.7 (41%) constituted POC ultrasound by nonradiologists. Between 2004 and 2009, radiologists' utilization rate increased by 17%, compared with 28% for nonradiologists. Radiologists' market share of noncardiac ultrasound was 56.6% in 2004 and 54.9% in 2009. Other major specialties involved in POC ultrasound and their 2009 rates per 1,000 and percentage increases since 2004 were cardiology (39.7 [+60%]), vascular surgery (34.9 [+36%]), primary care (27.2 [+11%]), general surgery (24.2 [+8%]), and urology (22.3 [+12%]). CONCLUSIONS: Between 2004 and 2009, there was a 21% increase in the overall utilization rate of noncardiac ultrasound. Point-of-care ultrasound by nonradiologists amounted to 41% of all studies done in 2009. Multiple nonradiologic specialties are involved, but radiologists' involvement is far higher than any other single specialty. Radiologists' ultrasound market share remained relatively stable between 2004 and 2009.

Medicare payments for noninvasive diagnostic imaging are now higher to nonradiologist physicians than to radiologists.

PURPOSE: Radiologists have always been considered the physicians who "control" noninvasive diagnostic imaging (NDI) and are primarily responsible for its growth. Yet nonradiologists have become increasingly aggressive in their performance and interpretation of imaging. The purpose of this study was to track overall Medicare payments to radiologists and nonradiologist physicians in recent years. METHODS: The Medicare Part B files covering all fee-for-service physician payments for 1998 to 2008 were the data source. All codes for discretionary NDI were selected. Procedures mandated by the patient's clinical condition (eg, supervision and interpretation codes for interventional procedures, radiation therapy planning) were excluded, as were nonimaging radionuclide tests. Medicare physician specialty codes were used to identify radiologists and nonradiologists. Payments in all places of service were included. Overall Medicare NDI payments to radiologists and nonradiologist physicians from 1998 through 2008 were compared. A separate analysis of NDI payments to cardiologists was conducted, because next to radiologists, they are the highest users of imaging. RESULTS: In 1998, overall Part B payments to radiologists for discretionary NDI were $2.563 billion, compared with $2.020 billion to nonradiologists (ie, radiologists' payments were 27% higher). From 1998 to 2006, payments to nonradiologists increased by 166%, compared with 107% to radiologists. By 2006, payments to nonradiologists exceeded those to radiologists. By 2008, the second year after implementation of the Deficit Reduction Act, payments to radiologists had dropped by 13%, compared with 11% to nonradiologists. In 2008, nonradiologists received $4.807 billion for discretionary NDI, and radiologists received $4.638 billion. Payments to cardiologists for NDI increased by 195% from 1998 to 2006, then dropped by 8% by 2008. CONCLUSIONS: The growth in fee-for-service payments to nonradiologists for NDI was considerably more rapid than the growth for radiologists between 1998 and 2006. Then, by the end of 2008, 2 years after the implementation of the Deficit Reduction Act, steeper revenue losses had been experienced by radiologists. The result was that by 2008, overall Medicare fee-for-service payments for NDI were 4% higher to nonradiologists than they were to radiologists.

How widely is computer-aided detection used in screening and diagnostic mammography?

PURPOSE: The aim of this study was to determine how widely computer-aided detection (CAD) is used in screening and diagnostic mammography and to see if there are differences between hospital facilities and private offices. METHODS: The nationwide Medicare Part B fee-for-service databases for 2004 to 2008 were used. The Current Procedural Terminology(®) codes for screening and diagnostic mammography (both digital and screen film) and the CAD add-on codes were selected. Procedure volume was compared for screening vs diagnostic mammography and for hospital facilities vs private offices. RESULTS: From 2004 to 2008, Medicare screening mammography volume increased slightly from 5,728,419 to 5,827,326 (+2%), but the use of screening CAD increased from 2,257,434 to 4,305,595 (+91%). By 2008, CAD was used in 74% of all screening mammographic studies. During this same time period, the Medicare volume of diagnostic mammography declined slightly from 1,835,700 to 1,682,026 (-8%), but the use of diagnostic CAD increased from 360,483 to 845,461 (+135%). By 2008, CAD was used in 50% of all diagnostic mammographic studies. In hospital facilities in 2008, CAD was used in 70% of all screening mammographic studies, compared with 81% in private offices. For diagnostic mammography in 2008, CAD was used in 48% in hospitals, compared with 55% in private offices. CONCLUSION: Despite some operational drawbacks to using CAD, radiologists have embraced it in an effort to improve cancer detection. Its use has grown rapidly, and in 2008, it was used in three-quarters of all screening mammographic studies and half of all diagnostic mammographic studies. Women undergoing either screening or diagnostic mammography are more likely to receive CAD if they go to a private office than if they go to a hospital facility, although the differences are not great.

Endovascular repair vs open surgical repair of abdominal aortic aneurysms: comparative utilization trends from 2001 to 2006.

PURPOSE: Within the past few years, endovascular aneurysm repair (EVAR) has come into use for the treatment of abdominal aortic aneurysms (AAAs). In many cases, EVAR has the potential to replace traditional open surgical repair (OSR), which is more invasive, risky, and expensive. The aim of this study was to determine to what extent EVAR is replacing OSR, whether the frequency of treatment is increasing with the advent of the less invasive approach, and which specialties are performing the procedures. MATERIALS AND METHODS: The Medicare Part B data sets for 2001 through 2006 were studied. Procedure volume and utilization rates per 100,000 Medicare beneficiaries were determined for the 7 Current Procedural Terminology, fourth edition, procedure codes that describe EVAR and the 4 codes that describe OSR for AAA. Medicare's physician specialty codes were used to ascertain the specialties of the physician providers. RESULTS: A total of 31,965 OSRs for AAA were performed in Medicare beneficiaries in 2001, dropping to 15,665 by 2006 (-51%). In contrast, EVAR was carried out in 11,028 instances in 2001, increasing to 28,937 by 2006 (+162%). The utilization rate per 100,000 for OSR dropped from 90 to 42 (a rate decrease of 48) during the study period, while the rate for EVAR increased from 31 to 77 (a rate increase of 46). The combined utilization rate per 100,000 of the two types of interventions for AAA (EVAR and OSR) decreased from 121 in 2001 to 119 in 2006. In performing EVAR, procedure volume and market share in 2006 by specialty were 1) 22,003 procedures by surgeons, a 76% share; 2) 3,287 procedures by radiologists, an 11% share; 3) 1,915 procedures by cardiologists, a 7% share; and 4) 1,732 procedures by all other physicians, a 6% share. CONCLUSIONS: Treatment for AAA seems to be an example of the responsible use of new technology by physicians. The newer, less invasive, and less risky procedure (EVAR) is replacing the older and more invasive procedure (OSR) to a considerable degree. However, the overall combined utilization rate of both types of AAA treatment has remained stable in the Medicare population. There is thus no evidence to suggest that the introduction of the newer approach has led to the overtreatment of patients. Although radiologists do have a role in EVAR, surgeons strongly predominate.

Ownership or leasing of CT scanners by nonradiologist physicians: a rapidly growing trend that raises concern about self-referral.

PURPOSE: The aim of this study was to examine recent nationwide trends in the ownership or leasing of computed tomographic (CT) scanners in private offices by nonradiologist physicians. METHODS AND MATERIALS: The Medicare Part B fee-for-service data sets for 2001 though 2006 were used to identify all CT scans performed in nonhospital, private-office settings. Ownership or leasing of CT scans was determined by tabulating all global and technical-component-only claims. Professional-component claims were excluded. The specialty of the owner or lessee was determined using Medicare's physician specialty codes. Procedure volume trends and growth rates among all nonradiologist physicians as a group were compared with those among radiologists. Individual specialty volume trends and growth rates were also studied. RESULTS: From 2001 to 2006, Medicare private-office CT scan volume in facilities owned by radiologists increased by 85%. CT scan volume in facilities owned or leased by nonradiologist physicians as a group increased by 263%. The nonradiologic specialties with the largest volumes in 2006 were primary care (192,255 scans), internal medicine subspecialties other than cardiology and medical oncology (184,991 scans), urology (125,850 scans), cardiology (104,739 scans), and medical oncology (61,976 scans). Excluding CT scans performed in independent diagnostic testing facilities (for which physician ownership cannot be determined), nonradiologists' private-office CT market share rose from 16% in 2001 to 28% in 2006. CONCLUSIONS: The majority of Medicare private-office CT scans are done in facilities owned by radiologists. However, nonradiologist physicians are acquiring or leasing CT scanners in increasing numbers, and the growth trend is much more rapid among them than it is among radiologists (85% among radiologists from 2001 to 2006, compared with 263% among nonradiologists). As a result, nonradiologists' market share has increased considerably. At a time when both cost containment and reduction in radiation exposure are urgent priorities, the self-referral opportunities resulting from this trend should be of concern to payers and policymakers.

Recent trends in mammography utilization in the Medicare population: is there a cause for concern?

CONTEXT: Recent published reports have shown a decline in the mammography screening rate in women over age 40, but it is not known whether this trend is a reason for concern in the Medicare population. OBJECTIVE: To study recent trends in mammography utilization in the Medicare population and determine how the newer digital mammography may be affecting mammography utilization. DESIGN AND SETTING: The Centers for Medicare & Medicaid Services Physician/Supplier Procedure Summary Master Files for 1996 through 2005 were examined to determine overall trends in mammography utilization, as well as trends in screening vs diagnostic and conventional screen-film vs newer digital examinations. Medicare Limited Datasets for 2002 to 2004 were used to determine 2-year mammography and multiple imaging rates in individual patients. MAIN OUTCOME MEASURE: Mammography utilization. RESULTS: Overall, the mammography utilization rate increased from 26,646 per 100,000 in 1996 to 39,363 per 100,000 in 2005, a 48% increase. The diagnostic mammography rate decreased by 39% (from 15,314 to 9,301), whereas the rate for screening mammography increased by 166% (from 11,332 to 30,062). Digital mammography increased from 2.2% of all mammography in 2002 to 10.4% in 2005. In both digital and film mammography, screening increased more rapidly than diagnostic mammography. CONCLUSIONS: The utilization rate of all mammography showed a substantial 48% increase between 1996 and 2005, and an 11% increase in screening mammography was seen between 2000 and 2005. Although the increase in mammography utilization is encouraging, the 2005 rate of 39,363 per 100,000 female Medicare beneficiaries seems to be well below American Cancer Society recommendations.

Ownership or leasing of MRI facilities by nonradiologist physicians is a rapidly growing trend.

PURPOSE: To study growth trends in the ownership of magnetic resonance imaging (MRI) examinations by nonradiologist physicians who either own the equipment outright or are involved in scan leasing arrangements. METHODS AND MATERIALS: Medicare Part B data sets from 2000 through 2005 were reviewed, and procedure codes for MRI examinations were selected. The focus was on only those procedures performed at nonhospital, private-office facilities. Using Medicare's physician specialty codes, all such studies were categorized according to the specialties of the physicians who performed them. Ownership was determined by including only those claims for global or technical-component-only reimbursement. Physicians owning or leasing MRI facilities would use one or the other of these two types of claims. Professional-component-only claims were not included. Procedure volumes and growth trends were compared among radiologists and other specialists. RESULTS: From 2000 to 2005, private-office MRI examinations performed by radiologists increased by 83%. During the same period, private-office MRI examinations performed by nonradiologist physicians, either through owning or leasing the equipment, increased by 254%. Excluding studies performed by independent diagnostic testing facilities (for which physician ownership cannot be determined), nonradiologists' share of the private-office MRI market rose from 11% in 2000 to 20% in 2005. The nonradiologic specialties most actively involved in performing MRI were orthopedic surgery (161,296 Medicare studies in 2005), neurology (63,363 studies), primary care (58,092 studies), internal medicine subspecialties (34,317 studies), and neurosurgery (20,712 studies). CONCLUSIONS: In the private-office setting in 2005, radiologists performed most MRI examinations. However, the growth rate from 2000 to 2005 among nonradiologist physicians was far higher, 254% compared with 83% among radiologists. Because scans performed by nonradiologists through ownership or leasing are subject to self-referral, the much more rapid growth among those physicians should be of concern to policymakers and payers.

Rapid multiplanar abdominal survey using MRI with the steady-state free-precession technique.

PURPOSE: To retrospectively evaluate the sensitivity, specificity, and positive and negative predictive values of steady-state free-precession (SSFP) survey MRI of the abdomen. MATERIALS AND METHODS: A total of 375 consecutive outpatients underwent abdominal MRI at 1.5T. Excluding diffuse metastatic disease, 110 patients had at least one other clinically important finding. The SSFP survey included contiguous 5-mm-thick axial, sagittal, and coronal slices (total 90 slices) obtained during a total of 90 seconds of free breathing. Studies were reviewed by two experienced MRI readers independently, randomized, blinded, and at different sittings. The chi-squared test was used to compare SSFP to full MRI for showing clinically important findings. In a subset of 30 patients, confidence intervals (CI) were calculated to compare the accuracy of SSFP and full MRI as predictors of biopsy result. RESULTS: SSFP detected 87.3% of clinically important findings and 93.3% of malignancies reported on the full MRI, with a 1.5% false-positive rate. Significant association was shown between SSFP and full MRI for clinically important findings (P < 0.0001). Compared to biopsy, accuracy of SSFP was high (85% +/- 12.7%), though not as high as full MRI (93.3% +/- 8.8%). CONCLUSION: SSFP provides a rapid survey of the abdomen, with good sensitivity and few false positives.

Vertebroplasty in the United States: guidance method and provider distribution, 2001-2003.

PURPOSE: To use the nationwide Medicare database to retrospectively evaluate the provider distribution for vertebroplasty, as well as the guidance method used in the United States. MATERIALS AND METHODS: Use of the Medicare database was exempt from Institutional Review Board review and informed consent; regarding HIPAA compliance, the Medicare database lacks unique patient identifiers. Using the 2001, 2002, and 2003 United States Medicare part B claims database, the authors studied CPT-4 (Current Procedural Terminology, fourth edition) procedure codes used for vertebroplasty procedures, including thoracic and lumbar vertebroplasty (22520 and 22521, respectively), treatment of additional levels (22522), and method of radiologic guidance (fluoroscopy, 76012; CT, 76013). For each of these codes, volume and physician specialty participation were tabulated. RESULTS: In 2001, 14 152 vertebroplasty procedures were reimbursed through Medicare. In 2003, the volume increased to 24 558 (+73.5%). In 2001 radiologists performed the majority (9864, 69.7%) of these procedures. The minority were performed by other specialists, mainly orthopedic surgeons (1792, 12.7%), neurosurgeons (1037, 7.3%) and anesthesiologists (736, 5.2%). In 2003, radiologists performed 86.8% more vertebroplasties than in 2001, and participation increased to 75.0% of the total. Radiologists had the greatest increase of all specialties from 2001 to 2003 (+86.8% vs: orthopedic surgery, +58.3%; neurosurgery, +55.7%; other, +46.6%; physiatry, +32.1%; neurology, +15.5%; anesthesiology, -11.8%). From 2001 to 2003, the number of additional levels treated by each of the three highest-volume specialties decreased (radiology, 23.0% to 20.8%; orthopedic surgery, 36.3% to 27.5%; neurosurgery, 28.2% to 27.0%). As a guidance method, fluoroscopy is used almost exclusively (98.7% in 2003). CONCLUSION: Radiologists performed the majority of Medicare-reimbursed vertebroplasty procedures in the United States in 2001-2003. Fluoroscopy is nearly universal as a guidance method.

The controversy over advanced cardiovascular imaging: relative roles of radiologists, cardiologists, and other physicians in CT and MRI of the cardiovascular system.

PURPOSE: There is considerable conflict between radiologists and cardiologists over who has the experience and knowledge to perform computed tomography (CT) and magnetic resonance imaging (MRI) of the cardiovascular (CV) system. The purpose of this study was to determine the relative roles of the different specialties in these procedures. METHOD AND MATERIALS: The nationwide 2003 Medicare Physician Supplier Procedure Summary Master Files were used to study all 8 codes for CT angiography and all 12 codes for magnetic resonance angiography and MRI of the heart. The database provides information on the number of studies performed in each code and the specialties of the physician providers. All professional component and global claims were tabulated, but technical-component-only claims were excluded because that would have led to double counting studies. RESULTS: In 2003, in the Medicare fee-for-service population, a total of 266,142 CV CT studies were performed. Radiologists performed 257,581 (96.8%), and cardiologists performed 1,246 (0.5%). Primary care physicians (PCPs) and radiation oncologists each performed more of these studies than cardiologists: 1,879 and 1,389, respectively. Independent diagnostic testing facilities (IDTFs) and multispecialty groups together performed 2,398 (0.9%); although Medicare lists each of them as a medical "specialty," the actual specialty of a provider cannot be determined. A total of 110,743 CV MRI studies were performed. Radiologists did 100,788 (91.0%), and cardiologists did 3,257 (2.9%). In addition, IDTFs and multispecialty groups together performed 4,644, PCPs 753, and radiation oncologists 537. CONCLUSION: Cardiologists had only minimal roles in CV CT and MRI in 2003. Their procedure volume was so small that it raises a question as to whether they could be gaining meaningful experience or providing meaningful training to fellows. Radiologists strongly dominated in the use of these modalities for imaging CV structures.

Current practice patterns and recent trends in breast biopsy among radiologists and surgeons.

PURPOSE: To ascertain the relative roles of radiologists and surgeons in breast biopsy. METHODS AND MATERIALS: The Nationwide Medicare Physician Supplier Procedure Summary Master Files for 1999 to 2004 were used to study the 4 Current Procedural Terminology, Fourth Edition, codes for breast biopsy. Using the Medicare specialty codes, physician providers were categorized as radiologists, surgeons, or others. Procedure volumes for each of the 3 groups were tabulated for each year of the study, and trends were assessed. RESULTS: Between 1999 and 2004, the breast biopsy utilization rate per 100,000 Medicare beneficiaries increased by 43%. In 2004, a total of 144,697 breast biopsies were performed, of which 124,423 (86%) were done with imaging guidance. From 1999 to 2004, procedure volume increased by 101% among radiologists, compared with 32% among surgeons. By 2004, radiologists performed 63% of all breast biopsies, compared with 32% by surgeons and 5% by others. Within the 2 codes that are specific for the use of imaging guidance, radiologists performed 72%. CONCLUSIONS: Radiologists predominate in breast biopsies, and their role has gotten progressively stronger in recent years.