Applied Radiology

Dr. Feig is Director, Breast Imaging, Department of Radiology, University of California Irvine (UCI) Medical Center and Professor of Radiology, UCI School of Medicine, Irvine, CA.

Breast-imaging services frequently lose money for radiology practices. When the costs of performing and interpreting these studies exceed reimbursement rates, breast imaging must be subsidized by other areas of the practice. This problem took root in the late 1980s when cost was seen as a barrier to utilization of screening. ¹ Several studies had shown that screening mammography could be performed for <$50 when batch reading was used. ^2,3 Based on such published data, the 1992 Medicare reimbursement rate for screening mammography was set at $56.76. ⁴ This decision meant that facilities that could not feasibly implement batch interpretation would lose money on screening mammography. Medicare made the financial problem even worse by simultaneously setting the reimbursement rate for diagnostic mammography at only $63.24, only 11% higher than the rate for screening, even though interpretation of a diagnostic study takes 5 times as long as a screening study.

By 1999, the Medicare rate for screening mammography had increased only slightly to $68.00. The rate for diagnostic mammography had shown a proportionately greater increase to $80.00. Neither rate had kept up with the added costs of inflation, the requirements of the Mammography Quality Standards Act of 1992 (MQSA), or higher standards of care. ⁵

Diagnostic mammographic workups became increasingly complex, requiring additional views and correlation with ultrasound. All diagnostic patients, and even many screening patients, expected to discuss examination results with the radiologist. Referring physicians expected the radiologist to phone them with all positive results. Comparisons with previous studies became the standard of care. Retrieval of earlier studies performed outside the facility entailed additional costs. ⁶ Finally, MQSA required: 1) continuing medical education (CME) in breast imaging for radiologists, technologists, and medical physicists; 2) more frequent quality control tests of equipment; 3) medical audit for interpretative assessment outcome; 4) tracking of biopsy results; and 5) written reports in layman's terms to patients. ⁷

Other total Medicare fees for breast imaging services (professional and technical components combined) in 1999 were: consultation on outside films, $38; ultrasound, $64; cyst aspiration, $128; wire localization, $132; stereotactic biopsy, $559; and ultrasound-guided biopsy, $305. ⁴ In order to determine the profit or loss from the professional component of each of these breast-imaging procedures, Enzmann et al ⁸ measured the time for radiologist's interpretation/performance. This time included viewing images, dictating and signing reports, filling out paperwork for tracking and patient letters, consulting with clinicians, informing patients of results, and performing procedures. The reported results were: Screening mammography, 5 minutes; diagnostic mammography, 25 minutes; breast ultrasound, 25 minutes; outside consult, 25 minutes; interventional procedure, 25 minutes. Based on these times and professional component reimbursements, Enzmann et al ⁸ calculated a profit of $20 for interpretation of each screening mammogram and $70 for performance of each interventional procedure. These profits were overshadowed, however, by losses for interpretation of each of the following examinations: Diagnostic mammography, $47; breast ultrasound, $28; outside consult, $55.

Enzmann et al ⁸ also surveyed the financial performance of breast imaging at 7 academic hospitals. These investigators found an average professional component loss of $346,000 per institution and $99,750 per full-time equivalent breast imager. Losses were greater at facilities that had higher ratios of diagnostic to screening mammograms.

In 2001, the American College of Radiology (ACR) conducted a survey of the technical cost of screening mammography. The survey, under the direction of Harvey L. Neiman, MD, the Chair of the ACR Board of Chancellors, found the average technical cost to be $105.57, with an average of $87.00 in radiology offices and $125 in hospitals. All of these costs exceeded the $50 Medicare technical component reimbursement. Logan-Young et al ⁹ reported that the combined technical and professional costs for either screening or diagnostic mammography, breast ultrasound or consultation on outside imaging exceeded the Medicare or other insurance reimbursement. ⁹

Medicare reimbursement rates for breast imaging may be compared with 2001 rates for other imaging tests: Magnetic resonance (MR) of the brain, $500; contrast-enhanced computed tomography (CT) of the abdomen, $320; noncontrast CT of the head, $220; 3phase bone scan, $230; aortogram, $500; barium enema, $100; and chest X-ray, $33.

Consequences of inadequate reimbursement

The financial losses from performing and interpreting mammography have affected radiologists' ability to provide breast-imaging services to women. Faced with limited funding, hospital administrators and department chairpersons preferred to spend their budget on equipment and staffing for MR or CT rather than mammography. In 2000, a Society of Breast Imaging (SBI) survey found that the waiting time for mammography appointments had lengthened. ^10-13 Yet, the demand for breast cancer screening has increased as more and more women have complied with recent screening guidelines. Fewer residents now intend to specialize in breast imaging. Applications for fellowships in breast imaging have declined, which raises concerns for a future shortage of radiologists and technologists in the sub-specialty. ^14,15 Many breast-imaging facilities have closed and/or consolidated, which has lengthened waiting times for appointments and curtailed convenient access.

How have radiological societies addressed these problems?

To increase public awareness of these problems, several members of the SBI Economics and Practice Issues (EPIC) Committee held a press conference at the 2000 Radiological Society of North America (RSNA) Annual Meeting. This briefing was widely covered by the mass media in newspapers, magazines, and on national television. In the early spring of 2001, the leadership of the ACR, the American College of Surgeons (ACS), and SBI met with representatives of breast cancer advocacy groups in Washington, D.C. They also provided information to key members of Congress.

At that time, Medicare payment rates for screening mammography were determined by Congress. Medicare reimbursement rates for diagnostic mammography were set by the Health Care Finance Agency (HCFA), now known as the Centers for Medicare and Medicaid Services (CMS). ¹⁶ These provisions changed when the Benefits Improvements and Protection Act became effective in January 2002. The mammography equipment manufacturers had successfully lobbied Congress to pass this Act in December 2000. Through this law: 1) digital mammography reimbursement performed for either screening or diagnosis was set at 150% of the value of conventional mammography; 2) an additional $15 was allowed for the use of computed-aided detection (CAD); and 3) control of the reimbursement rate for screening mammography was switched from Congress to CMS. The manufacturers had lobbied for this bill without consulting the ACR. The passage of this bill made it more difficult to increase reimbursement for conventional screening mammography. ^4,16

Several members of Congress also realized that improved reimbursement was necessary to ensure the availability of screening mammography. In March 2001, Senator Tom Harkin introduced legislation to return control of the screening mammography reimbursement rate to Congress and to increase this rate to $90 for 1 year. ^17-19 During this time, the General Accounting Office (GAO) was to evaluate screening costs to determine whether the $90 rate should be continued. The Harkin-Snowe bill would also have provided government funding to increase the number of radiology resident positions and to train more radiologic technologists. It was anticipated that these provisions would enlarge the workpool of radiologists and radiologic technologists, some of whom could eventually specialize in breast imaging. Unfortunately, the Harkin-Snowe bill was never brought to a Congressional vote. Nevertheless, the national publicity resulting from the RSNA press conference, the meetings with breast cancer advocacy groups, and the introduction of the Harkin-Snowe bill stimulated Medicare to increase the reimbursement rate for screening mammography from $69.23 to $80.73 in 2002. With supporting data from the SBI Mammography Practice Survey, the ACR was able to convince the CMS to increase Medicare reimbursement for diagnostic mammography from $80 to $88 beginning in ^10-13,16 Rates increased further over the next several years. Reimbursement rates in the 2005 Medicare fee schedule are listed in Table 1.

Relative cost and cost-effectiveness

Even at a reimbursement rate of $90 per examination and at the current 59% compliance rate for annual screening, screening mammography would represent only 0.41% of all Medicare expenditures. ²⁰ If every woman aged ≥65 years were to have an annual mammogram, such screening would still represent only 0.68% of all Medicare costs. Even if every woman aged ≥40 years were to screen once a year, the cost would represent only 0.43% of all national healthcare expenditures. ²⁰

At present, breast cancer accounts for 3.9% of all causes of death among American women. Allocation of 0.43% of all national health-care costs for screening mammography would be amply justified. In terms of cost per year of life saved, the cost-effectiveness of screening mammography is somewhat higher than that of cervical cancer screening, comparable to treatment for hypertension and screening for osteoporosis, but much less than that for coronary bypass surgery, renal dialysis, or use of automobile seat belts and air bags. ^21,22

Strategies to improve efficiency and productivity

While the ACR and other national organizations work to increase mammography reimbursement, there are also strategies that individual radiology groups may implement to affect their own profits or losses for breast imaging. ^23-26 One solution might be to accept only self-paying patients. By law, all Medicare-participating providers must agree to accept the Medicare fee as payment in full. If radiologists prefer not to participate in the Medicare plan, they should know that recent regulations have placed limits on the amount of out-of-pocket fees that can be charged to a Medicare patient by a nonparticipating provider. Radiologists who do not accept Medicare should also know that proper coding and billing can increase reimbursement and that this requires a complete understanding of the Medicare reimbursement system. ²⁷ Radiologists who choose to accept payments from third parties other than Medicare should know that some, but not all, carriers permit balance billing.

A second option is to perform more interventional procedures. Reimbursement rates for percutaneous biopsy and aspiration are higher than those for mammography. In their efforts to perform more interventional procedures, radiologists must discourage biopsy of Breast Imaging Reporting and Data System (BI-RADS) 2 and 3 lesions. Radiologist time for performing an interventional procedure should be kept to a minimum by having the procedure room set up and the patient ready beforehand. A technologist or medical assistant should be with the patient at that time.

A third strategy to improve revenues is for the radiology department to negotiate a contractual change with its hospital. The justification for this change is that breast imaging is a loss-leader that leads to downstream profits for surgery, radiation therapy, and oncology. Screening mammography should also be perceived as a public health service, which is good public relations for the hospital.

The fourth means to improve the economic picture at a breast-imaging center is to improve productivity. Unlike CT and MR, the major cost in mammography is professional and technical labor, rather than equipment. Thus, the radiologist should not have to perform any non-interpretive task that could be performed by a medical assistant. Such activities include placing phone calls for the radiologist to speak with a referring physician, calling patients to return for additional imaging, handling paperwork, and placing screening films on the rotator and later removing them.

To ensure productivity, screening mammography should be performed at a different site and time from diagnostic mammography. For screening cases, batch reading is much more efficient than online interpretation. Excessive recall rates are an unnecessary inconvenience for patients and are unprofitable for the facility because diagnostic mammography will lose money. Screening callback rates can be kept at 5% to 10% without missing cancers when interpretive skills and technical quality are good. ²⁸

Retrieval of outside films for comparison adds roughly $15 to the cost of an examination, usually without any corresponding payment for additional review, dictation, filing, mailing, or monitoring. ⁶ To minimize these costs, some facilities require that screening patients must: 1) have had a previous screening conducted at the same facility; 2) bring their outside films with them; or 3) be receiving their first screening examination.

The number of "no shows" on the screening or diagnostic schedule must be kept to a minimum. Patients should be sent a computer-addressed reminder letter several weeks before each appointment. A medical assistant should phone each patient several days prior to her appointment. If "no shows" persist, overbooking the schedule should be initiated. When late arrivals are a problem, it may be helpful to overbook early slots and underbook slots for later in the day. The first patients for the morning should be asked to arrive at least 15 minutes before the technologists start work so that registration and gowning will not delay the mammograms.

When some types of supplementary views are routine for specific diagnostic problems (eg, spot compression magnification of the lumpectomy site), these views should be performed by the technologist without prior approval by a radiologist. In most instances, diagnostic cases are more efficiently interpreted on a film rotator rather than on a stationary viewbox. Breast-imaging studies and reports should be separated from the patients' other studies and stored in the breast imaging area. All reports should be stored electronically for display on a monitor. To conserve the radiologists' time, some facilities allow a technologist to perform breast ultrasound followed by radiologist review of hard copy images. Many radiologists have extremely justifiable concerns about potential diagnostic errors and medicolegal risks associated with this practice. For the sake of accuracy and efficiency, the same radiologist who interprets a patient's mammogram should also perform and interpret her breast ultrasound.

Despite the overall improvement in mammography payments between 2000 and 2004, Medicare reimbursements for the technical component of mammography performed at hospital outpatient departments were substantially reduced during that period, as a result of a program known as the Hospital Outpatient Prospective Payment System (HOPPS). ²⁹ This program made no sense because the costs of performing mammography at a hospital are actually much higher than those at a private office. This unreasonable disparity was eliminated in January 2005 when Medicare, as a result of a provision in the Prescription Drug and Medicare Reform Act passed by Congress, was required to increase payments for hospital outpatient mammography to equal those paid to private office facilities. This reform represented another major legislative victory for the ACR. ³⁰

Conclusion

While the ACR and other national medical organizations try to increase breast imaging reimbursement rates, individual practice groups can try to make breast imaging more cost-efficient while maintaining a patient-friendly breast imaging facility and retaining diagnostic accuracy.