Do Unproven CAD Tools Get the Benefit of Reimbursement?

A recent study has found that as long as doctors can get reimbursed for new medical technology, the technology does not have to show clear-cut benefits for them to use it. The findings, published in the Archives of Internal Medicine, focused on a digital diagnostic tool—computer-aided detection (CAD)—that is used to detect breast cancer.
Installation of CAD technology can cost upward of $100,000, according to the researchers. Joshua Fenton, MD, a professor at University of California–Davis, led the study. He recently told Reuters Health that “it illustrates a bigger problem that exists in our society, where the rails are really greased to get new technologies into practice before we are sure that they are safe.”

Fenton used Medicare data to find that in 2001, CAD was used in less than 5% of screening mammograms. By 2003, the technology use quintupled. Reuters said that small studies have shown that CAD may better identify cancers than traditional mammograms but that the technology may also inadvertently flag plenty of women who don’t have cancer. The small studies also suggest that CAD would yield many more false positives than additional cases of actual cancer.
CAD critics say that the findings show “technology creep” in healthcare. Fenton casts some of the blame on manufacturers, such as R2 Technology Inc., which pioneered the use of CAD for mammography in 1998 with its ImageChecker system. R2 lobbied diligently to ensure Medicare coverage.

“This coverage was an important goal for the industry,” Fenton said. “Reimbursement was a major plan of their marketing.”

R2 Technology Inc. was acquired by Hologic in 2006. Although aware of Fenton’s article, Hologic declined to comment for this story. Another player in the CAD arena, iCAD, paints a different story than the one told by Fenton.
“It’s interesting that they continue to state that there’s not clinical evidence, because peer-reviewed published studies do show the clinical utility and the benefits of CAD in helping physicians to detect breast cancer earlier,” said Mary Robinson, director of marketing at iCAD.

She pointed to the fact that Fenton did not conduct a peer-reviewed study, and that “there are at least 25 peer-reviewed CAD studies that do speak to [its] clinical efficacy.” Robinson cited a landmark digital mammography study published in the New England Journal of Medicine in 2005, as well as another published in 2008 that looked at 118,000 instances of mammographies using CAD.

Robinson also questions the cost mentioned for CAD systems. “Mammography systems can [cost] anywhere from just under $100,000 to more than a couple hundred thousand. The mammography equipment—we don’t have any bearing on. Our solution is CAD, which is an algorithm. They cite $100,000. A CAD system’s average price is probably $50,000 or less.”

And what about reimbursement for CAD? “The Medicare national average is $12, so it’s not as though there’s tremendous financial incentive,” she says.

Ultimately, the question for end-users of mammography equipment is whether to buy additional software technology—i.e., CAD systems—to enhance the cancer screening process. “The person is going to have to have a mammogram, so you’re still going to have the mammogram equipment,” Robinson says. “Do you or don’t you elect to use this technology? At this point, the sentiment that we’re hearing from the clinical community is that [CAD] exists as a standard of care.”