Unproven for older women, digital mammography saps Medicare dollars

Introduction

In 2005, when patients at the Plano, Texas mammography clinic Women’s Diagnostic of Texas began cancelling appointments, the reason why was hardly a mystery.

Earlier that year, a Dallas hospital that competes with the clinic had hung a large billboard announcing its new digital mammography machine along the busy North Central Expressway that many Plano commuters drove to work.

Digital mammography manufacturers and hospitals had long run advertisements in women’s magazines suggesting digital mammograms were more effective at finding breast cancer than traditional film tests. They played on emotions with mother and child beach scenes, or showed radiologists and patients peering into a computer monitor at a breast image. “No cold metal. No paper gowns. No old, slow film imaging,” one advertisement proclaimed.

Until the Dallas billboard arrived, managers at Women’s Diagnostic were satisfied with traditional film mammography machines, said Gerald Kolb, then chief knowledge officer at the Plano clinic. But soon patients began leaving for that Dallas hospital in droves.

“We didn’t want to spend the millions and millions of dollars to make the conversion to digital,” Kolb said. “We had to.”

By 2006, Women’s Diagnostic had replaced its film machines with five Fuji digital units and added a digital system to view and store images. The pricetag: around $1.5 million.

The Plano clinic is hardly alone. The fact is, hospitals and clinics nationwide have spent billions converting their mammography units since 2000, when General Electric released the first digital machine to the U.S. market.

But experts say the newer technology has not been proven to improve breast cancer detection, particularly among women 65 and older who make up the vast majority of female Medicare beneficiaries and about a tenth of the Plano clinic’s business.

In fact, for traditional Medicare-aged women, the results of a five-year National Cancer Institute sponsored trial released in 2005 suggest the machines find no more cancers in older women than film mammograms. The trial, which was first reported in The New England Journal of Medicine, was one of the largest to-date to compare digital to film mammography.

The lack of evidence that digital mammography benefits women over 65, however, has not kept radiologists from using it to screen Medicare patients.

An analysis of a six-year sample of Medicare billing data obtained by the Center for Public Integrity and The Wall Street Journal shows that despite its lackluster performance, digital mammography has become the new standard of care in breast imaging for women 65 and older.

Medicare claims for digital mammography skyrocketed during the six years of billing data the Center examined. In 2003, physicians, clinics and hospitals billed Medicare for only around 426,000 digital mammograms claims, according to extrapolation from the sample. In 2008, that number rose to almost 6 million.

The same six years experienced a marked drop in film mammography claims, as the new technology took off. In 2003, film mammograms made up all but 4 percent of breast cancer mammography claims. By 2008, digital mammography had trumped film, rising to nearly 53 percent of all mammography claims.

The transition has been expensive: the digital procedure has been reimbursed at the national rate of $129, compared to $78 for a standard film mammogram. A statistical sampling of Medicare data by the Center suggests digital mammography may have increased the cost of breast cancer screening by more than $350 million from 2003 to 2008, further depleting the coffers of the rapidly expanding health care program for the aged and disabled that desperately needs to shed costs to survive.

Despite the fact that Medicare claims for mammograms fell slightly between 2005 and 2008, the total cost to Medicare for providing mammography increased by 17 percent to $636 million.

Medical researchers say digital mammography is one of a number of tests that, when considered together, are helping drive Medicare costs while doing little or nothing to make Americans healthier.

Liquid-based cytology, a modern cervical cancer screening test that has replaced traditional pap smears, is one example. A large randomized trial in 2009 found that the new test is no better at finding cancers than the pap smear, yet it is more expensive, with Medicare paying laboratories as much as $37 for the liquid test compared to under $15 for a standard pap smear. In 2008, liquid-based cytology accounted for more than 93 percent of claims for cervical cancer screens. The Center’s statistical sampling of Medicare data suggests the increased use of this test since 2003 has cost the program an additional $90 million.

Another high-profile example: the prostate-specific antigen, or PSA, screening test for prostate cancer. In December 2002, the U.S. Preventive Services Task Force, an independent panel of medical experts, cast doubt on the PSA, saying there is insufficient evidence to recommend for or against the test. The task force suggested men talk to their doctors about whether or not they should be screened, since treatment for prostate cancer carries risks, including incontinence and impotence. In 2008, the task force recommended that men 75 and older stop receiving the PSA test, since it was less likely that men of that age would die from prostate cancer.

The recommendations, however, did not stem the PSA tide. Medicare billing for the tests among men 65 and older rose about 18 percent from 2003 to 2006, though the numbers have fallen off some since then. In 2008, about one in seven Medicare-aged men received the test. From 2003 to 2008, PSA tests cost Medicare $372 million.

“We are living in a time when a lot of medical interventions have been oversold, and [digital mammography] is another one,” said Dr. Russell Harris, a professor and preventive medicine expert at the University of North Carolina School of Medicine. “What’s happened is that the people who make the machines, who benefit by selling newer machines, have triumphed.”

And on that path to triumph, the makers of digital mammography machines had plenty of help. The story behind the rise of digital mammography is a tale of intense industry marketing, direct-to-consumer advertising, political lobbying, and strategic campaign donations to politicians who shepherded beneficial Medicare reimbursement rates through Congress, creating the financial incentive for clinics and hospitals to replace film machines with digital.

A Promising New Technology

In the mid-1990s, when imaging manufacturers met with the Food and Drug Administration to discuss clearing digital mammography systems for marketing, the technology had been under development for more than a decade, with prototypes from General Electric, Fischer Imaging, Fuji Medical Systems, and the Danbury, Conn.-based Trex Medical involved in clinical trials at hospitals.

From the beginning, some doubted digital mammography would improve cancer detection, but many others believed that a clearer digital image would allow radiologists to identify cancers that remained invisible in traditional film mammograms, improving early detection rates.

The government had been an early partner with private industry in the “war” on breast cancer. General Electric, for example, partnered with the Defense Advanced Research Projects Agency (DARPA), the U.S. Army, and the National Naval Medical Center in Bethesda, Maryland, which received an early prototype of the company’s Senographe 2000D digital mammography machine for testing.

“The search for SCUDs in the deserts of Iraq … has more in common with the search for tumors than you might, at first, think,” then-CIA Director James Woolsey said in 1994, during an announcement that the Pentagon, the CIA and other government agencies would turn over military surveillance system technology to private companies developing cancer screening tools.

Even so, the FDA approval process for digital mammography dragged on, pitting manufacturers and breast cancer advocates against the agency, which serves as the gatekeeper for new medical devices. To gain approval, the FDA does not require manufacturers to prove their products perform better than technologies they replace. Instead, manufacturers must prove that new machines are as good and as safe as those already on the market. Cost is not a consideration.

Breast cancer advocates blamed the slow pace on what they called a culture of foot-dragging and secrecy at the FDA. Manufacturers said the FDA set the standards too high to prove digital mammography was equal to film screening, which would require large expensive trials.

To the consternation of breast cancer advocates, American women would not be the first to experience digital mammograms. In 1998, Trex Medical gained approval to market its digital mammography system in the European Union. In 1999, GE gained access to E.U. markets. That same year, the FDA denied the application of Trex Medical, after determining that its digital mammography system was not equivalent to film systems.

In 1999 testimony before the Senate Health, Education, Labor and Pensions Committee, Amy Langer, then executive director of the National Alliance of Breast Cancer Organizations, warned that FDA missteps could cause manufacturers to give up on digital mammography. “Consumers have read and heard about digital imaging technology for years now,” Langer said, “and women want to try it.”

That same year, Morgan Nields, then the chief executive of Fisher Imaging, told USA Today, “Let’s approve it, and then let the market determine the role of digital mammography.”

Not long after it denied the Trex Medical application, the FDA in 1999 suggested digital mammography manufacturers change their approach and file for market approval under criteria set aside for new devices instead of the criteria for devices that replaced other technology. Soon after, GE won market approval based on a trial of only 625 women. In 2000, the 2000D Senographe went on the U.S. market.

GE celebrated the release of the digital machines with an advertising campaign during the 2000 Summer Olympics in Sydney, Australia. The 30-second television spot featured three middle-aged female radiologists who referred to the machine as “a major new breakthrough in the fight against breast cancer.” The machine, they said, caused less discomfort and less anxiety, while providing radiologists with clear images and “more information.” “That makes a difference to me as a doctor, and as a woman,” said Dr. Debra Mitchell, an Oklahoma radiologist, who stood at a GE mammography machine dressed in a patient’s gown.

In a recent interview, Mitchell, an early adopter of GE digital mammography, said the company flew her to New York for the taping. She said the advertisement was not well-received in the scientific medical community because of the claims GE made about the machine. “It didn’t meet with great approval,” Mitchell said. “People thought by and large there wasn’t data to back it up.” Mitchell, though, remains a believer in digital mammography. “From my eyes’ perspective, it’s always seemed the way to go.”

The Digital Mammography Premium

Despite the FDA approval and the advertisements, GE faced a stiff challenge breaking into the imaging market. Digital systems in 2000 cost more than $350,000, as compared to about $75,000 for film machines. If insurance payments remained equal for digital and film mammography, it didn’t make sense for hospitals and clinics to make the jump to the new technology. Manufacturers needed a payment boost.

In most cases, the Centers for Medicare & Medicaid Services (CMS) make Medicare reimbursement decisions, which in turn drive the payment rates of private insurers. In the case of digital mammography, however, manufacturers did an end-run around the agency (then called the Health Care Financing Administration) with help from powerful friends in Congress.

In a recent interview, E. Clay Shaw, a former House Republican from South Florida, said he pushed for a digital mammography reimbursement premium after GE held a demonstration of the machine at the Reserve Officers Association near Capitol Hill. Shaw, then a conservative senior member of the powerful House Ways and Means Committee, said GE showed him two mammograms – one a grainy film image, the other a much brighter and clearer digital image.

“It was like it was just ten times ahead of the old style,” Shaw said of the digital image. “You could see the tumors much clearer. I was impressed with what I saw.”

In July of 2000, not long after the GE demonstration, Shaw, whose wife lost her mother to breast cancer, introduced the Breast Imaging Goes High Tech in the Future for Women Act, which would have required Medicare to pay $130 for digital mammograms. The bill died in committee, but Shaw was not discouraged. In September, Shaw introduced the Medicare Access to Digital Mammography Act of 2000, which set the fee at $155. Spencer Abraham, the former energy secretary and Republican senator from Michigan, sponsored a mirror bill to Shaw’s in the Senate.

The American College of Radiology opposed the bill over concerns that a digital premium would suck funding away from traditional mammography, and for awhile the bill languished. But Shaw’s digital mammography provisions were later attached to a Medicare bill that subsequently was itself absorbed by a Health and Human Services budget appropriations measure – a sequence of events that was first reported by The Miami Herald.

When that appropriations bill became law, it established a 50 percent reimbursement premium for digital screening mammograms over film mammograms. According to the Herald, Shaw also won a $1 million earmark for Fort Lauderdale’s Holy Cross Hospital to purchase a GE Senographe 2000D and to conduct a study on the machine’s effectiveness.

Shaw said GE’s demonstration sold him on digital mammography, but there had also been strategic campaign donations from GE to the former congressman. During the 2000 election cycle, Shaw accepted $5,000 from General Electric’s political action committee, including a $1,000 donation seven days after he introduced one of the bills to establish a digital mammography premium. Shaw had not been a large recipient of GE money prior to 2000, bringing in only $1,500 during the 1998 cycle.

During the recent interview, Shaw said GE campaign donations did not buy his support for digital mammography. “That never interfered with my judgment on anything. I always separated the two,” he said. “I never allowed any discussion of that in my office. I would not ever discuss campaign donations in my office.”

Abraham, who sponsored the similar bill to Shaw’s in the Senate, received $9,000 in campaign donations from the General Electric PAC during the 2000 election cycle, up from $1,000 in the previous cycle. Messages left at the international consulting firm Abraham now runs were not returned.

Allison Cohen, a spokeswoman for General Electric, said political contributions were in no way connected to Shaw and Abraham’s push for increased Medicare reimbursement. The role of General Electric is to manufacture the best mammography systems, not to influence reimbursement, Cohen said.

At the time, some medical imaging industry watchers said they suspected the hand of General Electric in the digital mammography bump. “I’m certain there was a lot of pretty intense lobbying by the blue meatball,” said Gerald Kolb, a long-time industry insider formerly of the Plano clinic, using industry slang for General Electric and its blue logo.

Kolb said it often doesn’t take much arm-twisting to push members of Congress toward supporting increased payments for breast cancer screening and treatments. “It is such a highly-charged issue with women,” Kolb said. “There are congressmen with fairly serious cancer in their families. Some of their wives have had it. So you don’t vote against breasts, that’s pretty clear.”

The legacy of the 2000 premium for digital mammography remains part of Medicare reimbursement policy, and mammographers have avoided cuts that have hit other imaging procedures. In 2005, Congress cut imaging reimbursement by $2.8 billion over five years as part of the Deficit Reduction Act. The cuts hit diagnostic imaging including CT scans, PET scans, ultrasounds and MRIs. Mammography, however, was excluded from the cuts. Medicare currently reimburses a bilateral digital screening mammogram at a national rate of $129, which is $51 more than a bilateral film screening mammogram, although there is some variance in payment rates vary across the country.

Ellen Griffith, a spokeswoman at CMS, said the agency does not have the authority to set prices based on the comparative effectiveness of medical technologies. Instead, CMS pays for tests and procedures based on the resources required to perform them, which in the case of digital mammography are significantly higher than film mammography because of the cost of purchasing and maintaining the machines.

“If it requires more resources, then we pay more,” Griffith said. “We don’t increase or decrease payments based on some kind of clinical assessment of efficacy.”

It does not appear that the 2010 health care law will change that. The law makes a push for federally-funded comparative-effectiveness research, but it also restricts Medicare from using that research to set pricing or limit coverage of a service.

Benefit for Young Women Fuels Jump Among Older Women

Despite the 2000 reimbursement bump, the increase in digital mammography was initially modest; in 2005, digital mammograms, although growing dramatically, still accounted for only 10 percent of Medicare mammography claims. By 2008, they would make up more than half.

If there was a tipping point for digital mammography, experts say it came in October 2005, when researchers writing in the New England Journal of Medicine released the results of a massive five-year trial funded by the National Cancer Institute, the government’s principal agency for cancer research. The Digital Mammographic Imaging Screening Trial (DMIST) gave digital and film mammograms to almost 50,000 women in the United States and Canada, and compared the ability of the two methods to find breast cancer.

The trial found that digital mammography detected more cancers in women 50 and younger, women with dense breasts, women who had not yet experienced menopause, and women in the midst of menopause.

But for older women, a 2008 paper on the DMIST data showed that among women 65 and older with fatty breasts, there was a “non-significant tendency toward film being better than digital mammography,” despite the additional cost of the newer technology.

Harris, at the University of North Carolina, finds the 2008 DMIST results disconcerting. “The trouble is that most breast cancer occurs in older women, not younger women,” he said “If we are going to go after younger women with dense breasts (by swapping film mammography machines for digital) we may actually do a disservice to older women with more cancers. That is the dilemma.”

Soon after the trial results were released, Dr. Etta Pisano, the principal investigator of the trial, told reporters that for most women there was no reason to seek out digital mammography because film was just as good. “We don’t have enough digital mammography for everyone to get it,” Pisano said, “and some women won’t benefit from it anyway.”

Since then, however, Pisano has changed her mind. Asked about the rapid increase of Medicare claims for digital mammography since 2005, Pisano said that after talking to radiologists who rely on the technology, she now believes digital mammography finds at least as many cancers in all women. Imaging processing has improved tremendously since the DMIST trial began in 2001, Pisano said. She also said the DMIST finding that film mammography was non-statistically better at finding cancers in women 65 and older was likely a statistical fluke, a “sort of an accident.”

“Now you would not get those results because you would use more modern image processing,” Pisano said. “That is my opinion and that is what I really do believe.”

Other prominent cancer prevention experts say improved process does not necessarily mean that digital mammography prevents more women from dying of breast cancer. In 2009, the U.S. Preventive Services Task Force, which was derided by many radiologists and breast cancer advocates for opposing routine mammograms for women aged 40 to 49, found there is insufficient evidence to assess the benefits or harms of digital mammography, regardless of a woman’s age.

Dr. Ned Calonge, the task force chair, said the group reviewed studies including DMIST and “found no evidence that digital was performing better than plain film” when judged in terms of health outcomes.

The distinction between judging imaging technologies based on patient outcomes rather than cancer detection is not a small one. Mammography proponents have long stressed the role that early detection plays in fighting cancer. But Calonge said it may be untrue that simply finding more cancers earlier equates to improved outcomes for women. It’s not the number of cancers that imaging detects, Calonge said, it’s making sure that they find the correct cancers, adding that all screening tests have benefits and harms.

One of the harms of all mammography tests, Calonge said, is they pick up a certain number of cancers that will never harm a woman, leading to needless biopsies and mastectomies in a certain number of women. He said it remains unclear which test – digital or film – better mitigates those risks. “Rather than embrace the new technology, we need to figure out if it works first,” Calonge said.

In a written response to questions about the DMIST trial and the rapid increase of digital mammography billing in Medicare, GE spokeswoman Allison Cohen said the technology has helped “increase the survival rates of millions of women around the world due to timely diagnosis.” Cohen also wrote that DMIST showed the better image contrast of digital mammography “increases the detectability of breast cancer for women with dense breasts who are at the higher risk for developing breast cancer.” She did not, however, reference women 65 and over.

“I’ll Never Have it Done Another Way”

Dr. Otis Brawley, the chief medical officer of the American Cancer Society, said excellent breast care is a question of doctors, not a question of digital versus film.

“Which machine you are using, I don’t really care,” Brawley said.

Even so, it’s clear that some Medicare-aged women prefer the digital test and believe it is better at finding cancers, suggesting that advertising and radiologists who use digital machines have been more effective than the DMIST results or the U.S. Preventative Services Task Force at influencing women’s health care decisions.

Veronica Casano, a 67-year-old retired nurse and social worker in Albany, New York, said she had her first digital mammogram two years ago, after her gynecologist found a “significant change” in a film mammogram. Casano, after waiting through a Christmas break, repeated the test at a digital facility. That test came up clear. Casano said the mammographer at the digital facility told her that the initial film mammogram was poorly exposed.

“Now I will never have it done any other way,” Casano said, adding that the benefit of digital is being able to immediately see if the image has been taken properly. “I had to wait through Christmas and New Year’s even, knowing that I had significant changes. That to me was not fun.”

Casano knows that digital has not been proven more effective at finding cancers for women her age, but she’s willing to live without the proof. “That’s life. I’m willing to accept it,” Casano said.

Sharon Huizenga of Milwaukee, 65, has only recently signed up for Medicare, and like Casano, she prefers digital mammograms. Huizenga reads the monthly Harvard Women’s Health Watch and tries to keep up with the latest data on preventive care. But it’s the images themselves that sold her on digital mammograms. She has seen both her film and digital mammograms and believes that “digital gave a lot more information to work with.”

After Huizenga was informed of the results of the DMIST trial and the U.S. Preventive Services Task Force recommendations, she said she places more stock in the recommendations of clinical radiologists than researchers and epidemiologists. “They know what is easier to look at,” Huizenga said, referring to radiologists.

But health care economists say the problem with “trusting your doctor” comes when radiologists have a financial incentive to prefer digital mammography over standard mammography, regardless of which provides better outcomes for women. When a hospital, clinic or radiologist owns an expensive piece of imaging equipment, economists say they have a financial incentive to maximize billing to help pay for it.

In their advertising, hospitals and breast clinics continue to push digital mammography for all women, not the specific sub-groups that benefitted in the DMIST trial. In one current campaign, LRGHealthcare, a hospital network in New Hampshire, recruited two elderly residents and the executive director of Peabody Home, an assisted living facility, to star in a print advertisement announcing its digital mammography machines. “Digital Mammography — Make the call that could SAVE YOUR LIFE!” the ad proclaims.

Meg Miller, the executive director of Peabody Home, who starred in the ad, defended pitching digital mammography for elderly women. “It’s a little bit more accurate, seemingly,” Miller said. “The digital seems to not have as many shadows.” And Miller said digital mammography cuts down on waiting time at the clinic, which she said will translate into more elderly women receiving an annual mammogram.

Medicare’s Digital Revolution

The greatest consequence of the DMIST trial may have been to sway radiologists who had stuck with film mammography to take the leap to digital. Medicare billing data shows that the largest jump in claims for digital mammography happened in the two years after 2005 when the trial results were released.

Dr. Larry Killebrew, an Oklahoma City radiologist said he wasn’t sold until he saw the clarity of the images available in machines made by Hologic, the Massachusetts company that is now the U.S. market leader in digital mammography. “I wanted to make sure this was the real deal before we jumped out there,” Killebrew said.

By 2007, Killebrew’s Oklahoma Breast Care Center network was entirely digital. Despite his late adoption of the technology, in 2008 Killebrew billed more digital mammography claims than any other doctor in the five percent Medicare billing sample examined by the Center for Public Integrity. Debbie Clark, the center’s executive director, said Killebrew’s 2008 Medicare billing brought in around $535,000, which accounted for roughly 20 percent of the breast center’s business.

Oklahoma Breast Care Center now includes two clinics in Oklahoma City and two recreational vehicles that each put on 100,000 miles a year bringing digital mammograms to businesses and rural hospitals throughout the state. Three small hospitals in the state also send their images to Killebrew to read, which would be impossible with film machines.

Killebrew understands the concerns of people who question the cost effectiveness of digital mammography, but he said if they could see the difference in quality between the images, they would see the difference. “If we can save women’s lives,” Killebrew said, “why would we even dare say that it’s not cost effective?”

Hologic uses stories like Killebrew’s to market its machines. In the fall 2009 edition of its annual “Images for Life” marketing material, Hologic tells the story of Hillcrest Baptist Medical Center in Waco, Texas, which switched to digital and increased its monthly mammogram screening by as much as 23 percent without adding staff or hours. At Oregon Imaging Centers, the Hologic marketing material said, the switch to digital led to a 22 percent increase in patient volume and “increases in additional imaging and diagnostic procedures, such as MRI, ultrasound and biopsy.”

But Clark, the executive director, said Oklahoma Breast Care Center is only a little more profitable after the conversion to digital because the overall cost of a digital operation, with overhead and maintenance, is much higher than film. “The profit margin you think would be there isn’t,” she said.

The profit margin for imaging companies, however, is another matter. In its 2005 annual financial report to investors, Hologic estimated that the U.S. mammography imaging equipment market “may grow to $400 million by 2008 from its introduction only four years ago.” The market, Hologic reported, is being fueled by digital mammography. It appears Hologic was correct. Seventy-three percent of all the mammography equipment in U.S. hospitals and clinics are now digital, reports the FDA. A Hologic spokesman declined to comment for this story.

Experts Say Medicare Digital Premium Makes No Sense

Regardless of which machine is better at finding breast cancers in older women, radiologists say digital technology has other benefits. Images can be manipulated, stored on computers, and beamed from clinic to clinic for second opinions, said Dr. Carl D’Orsi, director of breast imaging research at Emory University School of Medicine and co-chair of the American College of Radiology’s Committee on Breast Cancer. In addition, digital mammograms require less radiation than film mammograms.

During the switch to digital, some hospitals and clinics provided both digital and film mammograms, and some directed older women to the older technology. But as breast centers replace old machines, D’Orsi said film machines are getting harder to find.

Dr. Michael Klouda, medical director of the East Texas Medical Center Breast Care Center in Tyler, said it’s impractical to run a mammography unit with both digital and film systems. “If digital exists and you make the transition to it, you need to do it 100 percent,” Klouda said.

Since upgrading from film in 2007, Klouda said productivity at his clinic has risen dramatically. Digital has allowed the East Texas Medical Center Regional Healthcare System, of which Klouda’s clinic is a part, to run mammography clinics in six rural hospitals that do not have staff radiologists, which helps bring mammography to women who might otherwise go without. After technologists perform the screening, Klouda said images are beamed “back to the mother ship” and in some cases on to a physician in Los Angeles who interprets mammograms in his home office.

When speaking to radiologists about digital mammography they often compare the advance of digital mammography to digital photography – like with photography, the ship has sailed for digital mammography and there is no going back, they say. “It’s a digital world,” D’Orsi said.

The question that remains, however, is if Medicare should continue to incur the cost. In 2008, a paper published in Annals of Internal Medicine by the DMIST researchers found that regardless of its possible benefits to women, digital mammography is not cost effective when compared to film mammography.

“Over time, film mammography is going to cease to exist,” said Anna Tosteson, the lead author. “But here is one thing that’s certain – there is no evidence that one should pay a premium for digital mammography.”