Is an access crisis on the horizon in mammography?
By Phyllis Maguire
A convergence of problems including poor reimbursement and proposed federal legislation has some physicians talking about looming access problems in mammography.
In several parts of the country, internists complain that patients face serious delays in scheduling mammograms. Mammography screening centers in these markets are closing, while remaining centers are having trouble finding radiologists to read mammograms.
In New York City, for example, the average wait time for first-time screening mammograms has jumped from 14 days in 1998 to more than 40 days this year. And in parts of Florida, scheduling delays now run three months.
For now, many parts of the country have yet to experience any mammography-supply problems. Nonetheless, experts say, a number of factors could cause those problems to spread.
Radiologists say that inadequate reimbursement and high liability have made mammography unprofitable. As a result, they say, some of their colleagues are simply closing shop and walking away from the procedure. And they claim that new legislation being considered by Congress could make the situation even worse.
While so far radiologists are the most alarmed, other physicians are beginning to take note. Delays in both scheduling screening and in getting results, they warn, could prevent timely diagnosis—and put everyone at risk.
Access problems in mammography may come as something of a surprise, particularly because the nation sees the procedure as the key weapon in the fight against a major cancer.
By all accounts, screening mammography is one of health care's most dramatic success stories. Breast cancer screening levels among eligible women in the mid-1990s reached 66%, a stellar achievement when less than 30% of eligible Americans get screened for colon cancer. Among the general public, only cervical cancer screening—with rates close to 80%—has more patient buy-in as a preventive measure.
While physicians in many parts of the country say that success continues with no delays for their patients, others tell a different story.
Scheduling delays in Winter Park, Fla., for example, run up to 90 days, said Cecil B. Wilson, MACP, Immediate Past Chair of the College's Board of Regents and a solo practitioner. When one patient needed additional views, he recalled, she spent three days trying to get through to a facility on the phone to schedule an appointment.
"If these tests keep getting postponed, you may miss a tumor at a point where it could be curable," he pointed out. "It's a major concern."
In Philadelphia, one hospital resolved its six-month backlog of mammograms by simply shutting down its mammography facility altogether, said Eileen K. Carpenter, ACP Member, a general internist with Health Associates of South Philadelphia. And across the state in Pittsburgh, some physicians say they're waiting several weeks to see mammography reports, instead of a few days just a year or two ago.
In some areas, delays seem to follow socioeconomic lines. While average waiting times in New York have reached five-year highs, according to D. David Dershaw, MD, director of the breast imaging section at New York's Memorial Sloan-Kettering Cancer Center and president of the national Society of Breast Imaging, women in New York who can afford to pay $350 out of pocket are finding no wait for boutique mammography services.
In other parts of the country, physicians report that private mammography sites are closing, with radiologists instead offering mammography only at local hospitals or shutting down altogether. Melvyn L. Sterling, FACP, Governor for the Southern California Region II Chapter, for instance, said the mammography group he relies on—which specializes in mammography, not general radiology—is considering closing. Group officials told him they have lost between $250,000 and $300,000 a year for three years in a row.
Radiologists blame emerging access problems on factors that are all-too-familiar to most internists: low reimbursement, tough regulations and a frightening degree of liability.
Fees for screening mammograms, radiologists claim, range from marginal to inadequate, with Medicare now paying only slightly more than $82. (Before 2002, Medicare paid only $69.)
The cost of performing screening mammograms, on the other hand, is about $87 in a freestanding clinic and $105 in a hospital setting, according to the American College of Radiology (ACR). In high-cost areas like New York City, costs can run as high as $140.
Reimbursement, however, is not the only issue driving problems with mammography supply. As one of the country's most highly regulated medical procedures, mammography is subject to the Mammography Quality Standards Act (MQSA). That legislation, which Congress first passed in 1992, requires radiologists who read mammograms to earn 15 category 1 mammography-specific CME credits every three years.
In addition, the law requires every mammography site to undergo state inspection and keep records documenting quality control programs. The mammography group that Dr. Sterling refers to in California, for instance, told him it spends as much as $60,000 a year on quality assurance.
In addition, radiologists who read mammograms—one of the specialty's toughest interpretive jobs—face steep liability. While missed breast cancer diagnoses are a major cause of lawsuits for physicians of all kinds, radiologists take by far the biggest hit.
According to the 2002 Breast Cancer Study, which was issued by the Physician Insurers Association of America (PIAA), internists were named in 7% of the surveyed failure-to-diagnose breast cancer suits brought during the 1990s. (Family physicians were named in 11% of cases and gynecologists were named in 29%.) Radiologists topped the list, however, being named in 40% of all failure-to-diagnose breast cancer claims.
Those trends are sending ripples throughout the specialty. With business in diagnostic imaging booming, explained Memorial Sloan-Kettering's Dr. Dershaw, "radiologists are avoiding breast work. If you do 'neuro' or body CT, you get a lot more money, you're not being sued and you don't have all these hassles."
That observation has not been lost on future generations of radiologists, many of whom are avoiding mammography. According to the ACR, fellowship applications in mammography have fallen by 75% in some medical centers, while hundreds of breast imaging radiology positions nationwide are going unfilled.
In addition, 700 mammography sites around the country have closed over the last two years—a figure that represents about 7% of all sites. (The number of sites was previously estimated at close to 10,000.)
With fewer mammography centers open for business, hospital-based facilities are facing much stronger demand, even though mammograms in hospital-based facilities are already the most expensive to provide. This trend is worrisome, Dr. Dershaw explained, because hospital-based mammography facilities are "the most financially tenuous and endangered sites" of all.
As sites close, staff at mammography centers that remain run the risk of being overworked. Lori A. Bartholomew, PIAA's director of research, said that factor alone could fuel higher diagnosis failure rates. "You may now have radiologists reading more than they can do safely," she explained.
The debate over accuracy
Radiologists say that legislation currently before Congress threatens to make a tough situation even worse. Proposals designed to improve quality and protect patients, they say, may backfire, making it even harder for some patients to get a mammogram.
The proposals in question are part of legislation to reauthorize the MQSA. As Congress prepares to renew that legislation (it needs to be reauthorized every five years), legislators are considering mandatory self-assessments to boost radiologists' reading accuracy.
The requirement is a point of contention between radiologists, who not surprisingly oppose the measure, and consumer breast cancer advocacy groups. While consumer advocates acknowledge that access to mammography is becoming scarce in some communities, they are quick to argue that radiologists' rate of both false negatives and false positives is too high.
Analysts point out that false negative reading rates (not detecting cancers) are at least 10%—although some studies have pegged them as high as 40%. And a study published in the April 16, 1998, New England Journal of Medicine (NEJM) found that almost half of women who get a mammogram every year for 10 years get called back at least once for an unnecessary workup. That level of false positives, consumer groups say, vastly increases patient anxiety and mammography's high costs. (A study abstract is online.)
The solution, some consumer advocates say, is for radiologists to complete mandatory skills-based, hands-on self-assessments as part of their CME requirement. Radiologists, they claim, can best learn from feedback about how accurately they read mammograms.
Consumer groups want legislators to include mandatory hands-on self assessment for radiologists in the reauthorized MQSA, which was supposed be issued before Sept. 30, the end of the federal fiscal year. (While the legislation had not been reauthorized at press time, Congress had extended funds to continue its existing requirements.)
Self-assessment would entail radiologists looking at actual mammograms, either in seminars or via CD-ROMs, evaluating them and then comparing their readings to the actual results, getting real-time feedback on their reading accuracy. (For more on the benefits of feedback, see "When measuring quality in mammography, high-volume facilities not necessarily better." The ACR already sells a mammography interpretive skills assessment CD-ROM that features instant feedback and scoring—but completion of the ACR program is voluntary.
Radiologists successfully beat back a self-assessment mandate when the MQSA was first passed in 1992—and succeeded in defeating a similar provision when the bill was first reauthorized in 1998. An ACR spokesperson conceded, however, that there may now be strong enough Congressional support for the mandate to pass.
While radiologists claim they are most concerned that the self-assessment test results could be made public and discoverable in court, some say they also are concerned that passage of the proposed requirement will have a bigger impact on all of medicine. Edward A. Sickles, MD, professor in the department of radiology of the University of California, San Francisco, and chair of the ACR committee that designed its self-assessment program, said he finds the prospect of government-mandated physician self-assessment an ominous possibility—and not just for his specialty.
"To the extent that regulations get down to micromanaging radiology," he said, "it is that much more likely to result in micromanagement of other areas of medicine, including general internal medicine." Mandatory self-assessment for mammography, he said, would be analogous to Congress mandating annual documentation of clinical breast exam self-assessment for primary care physicians.
As radiologists and consumer advocates wait for Congress to decide on a self-assessment mandate, the bigger issue of access remains.
Some analysts are counting on the advent of digital mammography to help improve accuracy rates. Immediate past ACR president Valerie P. Jackson, MD, interim radiology chair at the Indiana University School of Medicine in Indianapolis, said that digital mammography can boost detection rates. The problem is that the technology also produces many false positives.
As a result, digital mammography requires "a strong, knowledgeable radiologist who can decide what needs to be worked up and what doesn't." And given mammography's poor reimbursement, an investment in digital mammography would mean, she added, that facilities "would lose even more money."
In a few centers, consumer foundations have stepped in to bolster screening programs. At Seattle's Harborview Medical Center, for example, internist Joann G. Elmore, MD, head of general medicine and a co-author of the 10-year study published in NEJM, said that the hospital continues to offer mammography resources—including a patient coordinator and babysitting services—to indigent women, thanks to an $800,000 grant from the Avon Foundation.
But that kind of largesse can't fix underlying national problems. While screening delays affect all physicians, said New York's Memorial Sloan-Kettering's Dr. Dershaw, the challenges confronting mammography are really societal ones. The consumers who fought for guaranteed mammography screening coverage, he said, should now direct their considerable political muscle toward advocating for better reimbursement and liability protection.
For now, Dr. Dershaw offered this practical advice to physicians referring patients for a screening mammogram. "Order it today," he said, "because it may not be available tomorrow."
When it comes to referring your patients for a mammogram, where should you send them? Recent research shows that facilities that read high volumes of mammograms may not necessarily be the best choice.
The link between the number of procedures physicians perform and their accuracy is often taken for granted in health care. The Leapfrog Group, for instance, an influential coalition of health care purchasers, regularly touts the benefits of using hospitals that perform a greater number of surgeries and procedures such as coronary artery bypass grafts and angioplasty. The group says that the more of these procedures a facility performs, the better the quality and outcome.
When it comes to mammography, however, volume may not be a perfect quality indicator. In an intriguing study published in the Feb. 19, 2003, Journal of the National Cancer Institute (JNCI), researchers found that radiologists who read more mammograms did not necessarily do a better job of identifying cancers. (The study is online.)
In fact, the study failed to find a significant association between volume and reading accuracy, said Craig A. Beam, PhD, director of the biostatistics core at the H. Lee Moffitt Cancer Center and Research Institute at Tampa's University of South Florida. (Dr. Beam was one of the study's three co-authors.)
Researchers did identify, however, certain exploratory factors that seemed linked to more accurate readings. Radiologists who had recently finished training, for example, tended to read more accurately. So did radiologists working in facilities with diagnostic breast imaging exams and interventional procedures.
Another finding: Radiologists working in comprehensive breast screening or diagnostic centers and freestanding mammography centers scored higher in accuracy than radiologists working in multispecialty clinics or hospital radiology departments. And radiologists working in facilities where mammograms were read by two radiologists were also linked to higher accuracy.
Dr. Beam said that many of the study's findings point to feedback as a critical element for improving accuracy. In most cases, he pointed out, radiologists decide a mammogram is normal and never receive any feedback. They often don't get feedback about mammograms they read as abnormal, even though most found positives are false.
Radiologist Emily F. Conant, MD, chief of the breast imaging section at Philadelphia's University of Pennsylvania Medical Center and another co-author of the JNCI study, said she was not surprised that radiologists who work in facilities that offer both diagnostic and screening services may score higher in accuracy.
"We recommend a biopsy on Monday and find out the answer on Friday," she said. "That really helps hone our skills."
In choosing sites to refer patients for mammography, she noted, "I would shy away from ones that do only screening without any diagnostic work or continuum of care, or where breast images are leftovers read by whichever radiologist is rotating through."
An ACP Governor who asked not to be identified said his wife's diagnosis of breast cancer has made him a strong believer in comprehensive breast centers, facilities that typically offer different specialists, clinical breast exams and diagnostic techniques in addition to mammography. During a visit his wife made to one of these centers, mammography failed to pick up a lesion. A breast exam performed during the same visit, however, led to an immediate ultrasound and quick biopsy.
A lumpectomy and axillary dissection found a 7 mm tumor with good pathology and negative nodes. His wife—who was diagnosed at age 45—has been cancer-free for almost eight years.
"The mammogram showed dense breast tissue and would have been read as low risk," he said. "A year might have gone by and the risk of a bad outcome would have been worse."
Internist Archives Quick Links
Have questions about the new ABIM MOC Program?
One Click to Confidence - Free to members
ACP Smart Medicine is a new, online clinical decision support tool specifically for internal medicine. Get rapid point-of-care access to evidence-based clinical recommendations and guidelines. Plus, users can easily earn CME credit. Learn more