Is Time on Mammography’s Side?

Time passes quickly, as busy mammographers know, but will it pass quickly enough to bring the statistics and support necessary for tort reform? This solution seems to be the best that radiologists and experts can think of to solve the growing problem of negative revenue and increasingly scarce resources in the field.

Mammography is the gold standard in detecting breast cancer, but the modality is not perfect. The contrary public perception is one factor that has made breast cancer the leading condition for which patients file a medical malpractice claim. The resulting costs of malpractice insurance, in addition to poor reimbursement, mean that many facilities operate mammography centers at a loss. For some, the resulting business makes up for it; for others, the costs are overwhelming, and they must close their doors.

Facilities have implemented computer-aided detection (CAD), second readers, and improved patient communications, but these all cost money and time, both of which are in short supply.

Breast Cancer and Mammography

Approximately 200,000 new breast cancer cases are diagnosed in the United States each year, and about 40,000 women die from the disease.¹ Mammography is the best tool for breast cancer screening, but it is far from flawless. A report by the Institute of Medicine (IOM of Washington) and the National Research Council (Washington), “Saving Women’s Lives: Strategies for Improving Breast Cancer Detection and Diagnosis,” cited by WebMD (Elmwood Park, NJ), notes that in about one in 10 procedures, screening mammography misses about 17% of all breast cancers in women and might mistakenly label an abnormal area as a tumor where none exists.¹ Other studies indicate the error rate could be even higher, closer to 30%.

Richard Anderson, MD, chairman and CEO of the Doctors Co (Napa, Calif) likens finding breast cancers on a mammogram to “trying to find a snowstorm in a blizzard.” He says, “In the best of hands, there will be a 10%-15% false-negative rate. One third to one half of these are true negative; the mammogram is read correctly but doesn’t show cancer. One half to one third are due to radiology or communication error, where the radiologist read it correctly but it wasn’t communicated to the physician, or the physician didn’t act on it appropriately.”

Anderson notes that the US Preventive Services Task Force of the Agency for Healthcare Research and Quality (Rockville, Md) downgraded the evidence showing that mammography reduces breast cancer to fair, or a B, from good, an A. Even so, the agency, like many others, recommends screening mammography, with or without clinical breast examination, every 1-2 years for women aged 40 and older.

Efforts to encourage women to schedule these exams regularly have succeeded, perhaps too well. “Mammography has been oversold so that public expectation is unrealistically high,” says Elsie Levin, MD, director of the Faulkner-Sagoff Center for Breast Imaging and Diagnosis (Boston). “If a lump is found later, the patient blames mammography.”

Mammography does not prevent cancer; it can merely help detect it. Early detection leads to early treatment but not necessarily a cure. Though the idea is debated in some medical circles, many believe that early treatment could lead to decreased mortality. Yet, despite its value, some women might soon be denied the opportunity to have their breasts screened.

Fleeing the Field

The IOM report² found that the number of breast imaging specialists isn’t enough to keep up with the growing population of women over the age of 40, more than 1 million. It states, “The United States is rapidly losing its capacity to screen women for breast cancer using mammography, a trend that is causing many women to go without screening. Women across the country wait months to get screening mammograms because facilities are overstretched.” The report found that the number of US mammography facilities declined from more than 9,400 in 2000 to an estimated 8,600 in 2003, a drop of 8.5% per year.

More than 650 Mammography Quality Standards Act (MQSA)-certified screening centers have closed their doors in the past year because they couldn’t break even.³

Leonard Berlin, MD, chairman of the department of radiology at Rush North Shore Medical Center (Skokie, Ill), and professor of radiology at Rush Medical College (Chicago), offers additional statistics from the American College of Radiology (ACR of Reston, Va). “According to the ACR, between 700 and 800 mammogram facilities have closed in the past year in the United States. There is no question that some of it is financial, including malpractice, and part is due to fewer radiologists entering the specialty because of malpractice and reimbursement issues,” he says.

A study⁴ by the Georgia Board for Physician Workforce (GBPW of Atlanta) found that in October 2003, 13.7% of physicians had stopped providing high-risk procedures during the past year as a result of malpractice insurance premiums. This followed a 17.8% reduction in 2002. The primary services impacted were obstetrical services, mammography, surgical care, and trauma; 19% of these physicians were radiologists.

The American Medical Association (AMA of Chicago) has found the problem to be nationwide, citing 20 states in “full-blown medical liability crisis.” These include Arkansas, Connecticut, Florida, Georgia, Illinois, Kentucky, Massachusetts, Mississippi, Missouri, New Jersey, Nevada, New York, North Carolina, Ohio, Oregon, Pennsylvania, Texas, Washington, West Virginia, and Wyoming. The analysis found that physicians are retiring early, moving out of state, or quitting certain medical procedures, such as delivering babies.

MRI: Not for the Masses Unlike CAD, MRI is not seen as a solution for the masses. The modality is prohibitively expensive, and currently, it suffers a high false-positive rate. However, it also offers high sensitivity and is quite useful for select patients, particularly women with high-risk and/or high-density breasts. At the Saddleback Memorial Medical Center, every woman with a new breast cancer diagnosis has an MRI recommended. “Most patients agree,” says the center’s Peggy Ann Pugh, MD, DABR. “The exam helps to gather as much information as possible about the cancer, such as whether it’s unifocal or multifocal, which then determines treatment.” Studies are currently under way examining exactly how useful MRI is, but even after the results are published, cost could be a limiting factor. Richard Anderson, MD, of the Doctors Co predicts that if MRI were to become the standard, it would multiply the cost of screening for breast cancer by several hundred percent. “However, the increase in detection might be a small amount,” he says, asking, “Would the value justify the trade-off?” -RD

Levin shares that patients scheduling a breast imaging exam at the Sagoff Center during the day are put on a 13-month waiting list. “Locally,” she says, “some facilities have closed, and our volume has increased as a result.”

Malpractice Costs

The IOM blames low payment rates from Medicare and high malpractice insurance costs as possible factors in centers closing and also claims that a poor financial outlook is keeping new physicians from going into radiology.¹

In 2002, Medicare reimbursement for screening mammography was $81.81, which included both the professional fee ($35.48) and the technical component ($46.33). If CAD was used with screening, the government paid an additional $17.74. Private insurance generally follows Medicare’s example.³ The mammography reimbursement, however, does not cover the expenses, and facilities are forced to run their mammography centers at a loss.

Malpractice insurance adds additional costs, and depending on the state, it can be overwhelming. “The cost of malpractice insurance is based on the claims experience,” explains Larry Smarr, president of the Physician Insurers Association of America (PIAA of Rockville, Md). “There are many such claims on failure to diagnose cancer via mammography.”

The most recent compilation of Data Sharing Information released by PIAA⁵ reveals that malignant neoplasms of the female breast continue to be the condition for which a patient most frequently files a medical malpractice claim. In 1995, radiologists represented 24% of defendants in breast cancer malpractice cases; in 2002, it rose to 33%.

Breast cancer is the second-most expensive condition in terms of indemnity dollars, next to claims resulting from neurologically impaired newborns. More than 41% of all claims involving breast cancer result in indemnity payment to the claimant. The total indemnity for a claim rose 45.3% from an average of $301,460 in 1995 to $438,047 in 2002. The average expense paid to defend these cases rose 75.8%.

More than 46% of cases in the breast cancer study resulted in a large loss-a payment of $250,000 or more in indemnity.⁴ Indemnity payments of $1 million or more accounted for 5.8% of paid cases.

Radiologists had an average indemnity of $346,247, with a total indemnity of $63,709,403. The prevalent associated issue for radiologists was “Mammogram Misread,” indicated in 75.2% of radiology claims. It was also the most expensive associated issue, with $57 million in indemnity. The associated issue, “Negative Mammogram Report,” was involved in 35.1% of 2002 claims, up from 25.8% in 1995.

The resulting increase in claims and awards has caused malpractice insurance premiums to increase as well, particularly in states where there has been no effective tort reform.

Respondents to the GBPW survey⁴ reported that malpractice insurance premiums had increases of 25%-50% on average in 2003, following a 20% increase in 2002. The specialties most affected were obstetrics/ gynecology, general surgery, radiology, neurology, and enterology medicine.

Anderson cites statistics⁶ indicating that since 2000, mean rates for medical malpractice have increased 10%-20% annually. These averages obscure increases of 100% or more in some venues with unlimited liability, in contrast to average increases of 5%-10% in states that have passed effective tort reform statutes.

In some states, the cost of malpractice insurance isn’t the only issue-it’s also finding it. According to the GBPW, MAG Mutual (Atlanta), the principle provider of coverage in Georgia, reported that of the 20 insurers in the state in 2001 (which had $1 million or more in premiums), only three remain that accepted new physicians in 2003.⁴ In the states most severely affected, which include Pennsylvania, Nevada, West Virginia, Mississippi, Texas, and Florida, some physicians have been unable to find coverage at any price or have been forced into state-run plans.⁶

“Radiologists are having a harder time finding insurance. Some companies no longer offer it; others are less willing to take on the risk,” advises the PIAA’s Smarr. “Doctors need to be diligent and network for contacts. They can only keep trying.”

Turning to Tort Reform

Malpractice insurance is creating a cost that is not adding any value to patients, according to Etta Pisano, MD, FACR, professor of radiology and biomedical engineering and chief of breast imaging at the University of North Carolina School of Medicine (Chapel Hill). “Some would argue that it adds value by encouraging doctors to be more careful, but in reality, it creates a closed medical system that is loathe to discuss errors. Victims of mistakes deserve to be paid, but the current system does not do that; rather, it is more like a lottery with a lucky few winning large payments.”

The best solution touted-according to Pisano, other physicians, and the experts cited here-is tort reform, with many holding California’s Medical Injury Compensation Reform Act (MICRA) as the standard for which to aim. MICRA does not limit economic damages, which include lost earnings, medical care, and rehabilitation costs, but it does put a cap on noneconomic awards, such as pain and suffering, at $250,000. Damages are paid over the period they are intended to cover rather than in one lump sum, and limits are placed on attorney’s fees.

Considerable data shows that a $250,000 cap on noneconomic damages reduces malpractice premiums by 25%-30%, as the experience in California confirms.⁶

Peggy Ann Pugh, MD, DABR, a breast-imaging specialist at Saddleback Memorial Medical Center (Laguna Hills, Calif) can attest to this truth. “MICRA insurance reform has kept rates reasonable in California,” she says. “My malpractice insurance costs $15,000 a year, far more reasonable compared to others I’ve spoken with who pay $40,000, $60,000, even $80,000 annually.”

The Loss Leader

Unfortunately, tort reform has stalled in Congress; until it passes, doctors and facilities need to continue to provide patients with a full range of services as well as make a living. The Mammography Risk Management Manual recommends four areas for survival:

1. Retaining patients and workers;
2. Ensuring timely follow-ups;
3. Creating a plan for image-quality troubleshooting; and
4. Auditing mammography practice regularly.

Mammography is considered by most facilities to be a “loss leader.” Pugh notes that patients who come in for mammograms are more likely to bring in their family members or seek other services.

Indeed, other breast treatments lead to additional revenues. “In the long run, some of these women, roughly 10%-15% on a national level, will be recalled for diagnostic mammograms, which have higher reimbursement. Some will need biopsy, which also has additional remuneration. And some will have cancer,” says Berlin of Rush Medical College.

The system can, therefore, remain solvent even when the mammography center loses money, because other treatments will bring in revenue. “It’s good for the system, but not so much for the radiologist,” says Anderson of the Doctors Co.

Fortunately, radiologists are frequently employed within groups rather than solo private practice. “Groups need to offer mammography for their primary care physicians,” Pisano says, “and they absorb the costs because it generates additional revenue.”

Alternative Ideas

Despite these gloomy statistics, facilities need not assume they have to lose money or patients. To handle the recent increase in volume, the Sagoff Center now offers evening and weekend hours. “These time periods are more efficient because the results are not read immediately and more patients can be seen,” says Levin, adding that daytime patients have their results and any follow-up needs taken care of immediately.

Other ideas have come from overseas. The “Swedish model,” which is becoming popular in the European Union and much of Canada, offers women screening tests by skilled technologists at conveniently located sites or mobile units. Films are then sent to a central site where they are interpreted by radiologists dedicated to reading mammograms. If the patient needs additional evaluation, she goes to a central diagnostic center.³

Alternatively, the IOM report suggests consolidated clinics, also used in Sweden and the United Kingdom. Tests, interpretations, and some treatments are all done under one roof.

Another idea, also cited in the IOM report, is to use trained, nonphysician screeners, such as physicians’ assistants, to help speed mammography interpretation and potentially improve accuracy; however, physicians are not yet ready to embrace such a plan. Pugh notes that this idea would not work in the United States where fears of malpractice are a large concern. “If I signed a report I hadn’t read myself, I would not be able to sleep at night,” she says.

Levin wonders if the readers would even be willing to take on the liability. “It takes many years, much experience, and a large volume to become an expert reader,” she notes. Studies would support this notion; one found that generalists had more abnormal interpretation rates for screening mammography, such as recall rates, than specialists-at 7.1% versus 4.9%. Similarly, specialists detected more cancers at all stages than generalists.⁷

Studies showing that radiologists’ error rates diminish as they read more have led countries, including Canada and the United Kingdom, to require that radiologists certified for mammography read a minimum of 2,500 a year. The United States’ MQSA requires only 960 mammograms read every 2 years. No official movement has been made to change this policy, but specialists know that experience increases accuracy.

Setting Standards

Many radiologists follow guidelines determined by themselves, their facilities, or other organizations. PIAA offers a number of recommendations, which might help to reduce liability. (See “PIAA Recommendations to Reduce Liability,” below.)

The MQSA lists specific guidelines to follow in communicating mammography results to patients.⁸ Improved communication has been found to reduce liability. Pisano recalls a seminar she attended years ago in which a lawyer explained that doctors who are sued tend to not have a good relationship with their patients. “Traditionally, radiologists don’t have a relationship with the patient, but breast imagers can and must make an effort to build one,” she advises. “Answer all questions, provide information, don’t rely on the primary care physician to relay everything. It takes time and money to establish a good relationship, but it increases care as well.”

Saddleback Memorial Medical Center follows its own best practices, which include guidelines for communication between the patient’s physicians and with the patient herself. No symptom will go unexplained. “If a patient has a lump, we do not stop testing until we know what it is,” she says.

The facility also employs double reading. “We used double readers before CAD, even though there was no reimbursement, because we saw value in it,” Pisano says. “Now CAD is reimbursed, so we use that as a double reader, which has the added value of freeing up the second radiologist to see patients directly and generate more revenue.”

The Sagoff Center also employs CAD. “We installed CAD 5 or 6 years ago when there was no reimbursement because we felt it was the best way to provide a second reader and therefore the best quality of care,” Levin says. “Now it is reimbursed but requires no additional time.”

New technologies, such as CAD, MRI, and digital mammography, could help to improve accuracy and might even help the bottom line. Though more expensive up front, these technologies could increase patient volume or detect enough additional cancers to pay for themselves over time. Because CAD is reimbursed, it brings in revenue immediately.

These advances also might eventually help to bring down insurance rates. “To the extent that they are able to improve diagnostic technique, there should flow a reduction in errors, claims, and, therefore, rates, but it would take time,” Smarr says. Time that some mammography centers might not have.

Renee DiIulio is a contributing writer for Medical Imaging.

References

1. Zwillich, T. Women facing “crisis” in mammographies. WebMD Health: June 10, 2004. Available at: http://my.webmd.com/content/Article/88/99962.htm. Accessed August 24, 2004.
2. Institute of Medicine. Saving women’s lives: strategies for improving breast cancer detection and diagnosis. June 10, 2004. Available at: http://www.iom.edu/report.asp?id=20721. Accessed August 27, 2004.
3. Bruce, R. Innovative programs cut through mammography’s crisis. Women’s Imaging Digital Community: Oct 17, 2002. Available at: http://www.auntminnie.com/default.asp?Sec=sup&Sub=wom&Pag=dis&ItemId=56385. Accessed August 24, 2004.
4. Georgia Board for Physician Workforce. The effect of rising medical liability premiums on physician supply and access to medical care in Georgia follow-up. Available at: http://gbpw.georgia.gov/vgn/images/portal/cit_1210/
   15895593Fact%20Sheet%20-%20Medical%20Liability
   %20Follow-Up%202003.pdf. Accessed August 24, 2004.
5. Physician Insurers of America. Breast Cancer Study. 3rd ed. Spring 2002.
6. Anderson, RE. Defending the practice of medicine. Arch Intern Med. 2004;104:1173-1178.
7. Sickles EA, Wolverton DE, Dee KE. Performance parameters for screening and diagnostic mammography: specialist and general radiologists. Radiology. 2002;224:861-869.
8. Mammography Quality Standards Act Regulations. US Food and Drug Administration, Center for Devices and Radiological Health. Available at: http://www.fda.gov/cdrh/mammography/frmamcom2.html#s90012. Accessed August 24, 2004.

Two Is Better Than One SCANIS CEO Bob Chapman believes that, in theory, CAD could reduce exposure to lawsuits. Like mammography, computer-aided detection (CAD) is not perfect, but, when used together, the two get a little closer. Studies have shown that having mammograms interpreted by a second reader increases the cancer detection rate by about 15%, according to Bob Chapman, CEO of SCANIS Inc (Foster City, Calif). CAD acts as a second reader-a very experienced one. The results of one study suggest that adding CAD provides a clinically significant improvement of 19.5% in the cancer detection rate.1 Another found CAD improved detection by 20%.2 When used as intended, CAD is expected to increase the number of mammograms interpreted as positive to the extent that it points out abnormalities previously overlooked by the radiologist on unaided reading.1 Some are concerned that CAD will increase the false-positives and recall rate, but studies have shown any increase to be insignificant. Etta Pisano, MD, FACR, of the University of North Carolina School of Medicine, believes the data to indicate that using CAD can bring any radiologist’s level of reading up to an expert mammographer. Chapman explains that this is exactly the intent. The Mammex MammoCAD System from SCANIS, currently available in the United States for investigational use only, was developed with input from a team of expert mammographers and reads the image just as a physician would. “So the system offers the collective experience of more than just two doctors,” he says. Still, others are not completely convinced. “Early data is favorable, but more research needs to be done,” says Leonard Berlin, MD, of Rush North Shore Medical Center and Rush Medical College. That research might provide the data to begin affecting change in the field of mammography. “In theory, CAD could reduce exposure to lawsuits, but it takes a long time to develop statistics to change anything, whether it’s reimbursement, insurance rates, or even clinical practice,” Chapman says. Yet, he notes, CAD offers immediate value. “In light of malpractice statistics, if a CAD system detects two cancers a year that would otherwise have been missed, the system will have paid for itself,” he says. CAD mammography systems that are FDA cleared and available in the United States include: The ImageChecker platform from R2 Technology Inc (Los Altos, Calif); The Second Look line from CADx Systems Inc (Northborough, Mass), which was acquired by iCAD (Nashua, NH); and MammoReader by Intelligent Systems Software Inc (Boca Raton, Fla), which recently merged with Howtek Inc (Hudson, NH). -RD References Blue Cross Blue Shield Association’s Technology Evaluation Center. Computer-aided detection (CAD) in mammography. Assessment Program: December 2002. Available at: http://bcbs.com/tec/vol17/17_17.html. Accessed August 24, 2004. SCANIS Inc. MammoCAD by the numbers. Available at: http://www.scanis.com/physicians/administrator_justi.html. Accessed August 24, 2004.

PIAA Recommendations to Reduce Liability The Physician Insurers Association of America (PIAA) has developed recommendations based on study findings1 and prevalent problems that could help to reduce a physician’s liability. Not to be mistaken for standards of care, the recommendations include the following: For all physicians: Document all patient complaints relative to the breast. Record personal and family history of breast cancer. Inquire and request the results of any previous mammograms. Follow up with other physician consultants regarding test results. Do not delay appropriate diagnostic studies because of pregnancy. For radiologists If a mammogram results in a film of poor technical quality, repeat the study. If the mammogram results are equivocal, recommend a repeat study, additional views, or other imaging modalities as appropriate. Be sure an adequate physical examination is performed and documented. Compare the results of the present study to all previous studies performed. Promptly report findings to the referring physician. If the patient was self-referred, the results of the study should be conveyed directly to her. The mammography center at which the study was performed also is required to send the results to the patient. If there is any suspicion of an abnormality, the patient should be advised to consult promptly with her primary care physician. If a screening mammogram is being performed on a self-referred patient, be sure to do a thorough breast examination, and advise the patient of the importance of regular physical breast exams to complement the study. In cases of self-referral, ensure that the patient receives proper follow-up. -RD Reference Physician Insurers of America. Breast Cancer Study. 3rd ed. Spring 2002.