Learning digital mammography proves difficult but worth the effort

Throughout its 30-year history, Diagnostic Imaging has tracked issues surrounding the adoption and development of new imaging technology. Rarely, however, have we looked closely at the question of the learning curves associated with this new technology. This month marks an exception. Our cover story on digital mammography recall rates (page 30) explores why some facilities may be seeing a jump, albeit temporary, in digital mammography recalls and suggests some solutions for keeping those rates down.

That recall rates sometimes rise as facilities shift to digital mammography should come as no surprise. Inevitably, there are learning curves

as radiologists adjust to new technological advances. This process has been happening ever since radiology entered its modern multimodality era, and possibly before. New technologies present more and better information, but they take time to master. We should not expect digital mammography to be any different.

So why is this an issue now? Two factors seem to be at work.

One is the rapid adoption of digital mammography in the U.S. In 2005, 450 units were sold, and that figure doubled to 900 in 2006. Sales were up again in 2007 and hit 1650, according to SCAN, Diagnostic Imaging's business newsletter. Vendors expect this kind of growth to taper off this year, but unit sales will still be substantial and are likely to hit the 2000 to 2100 range.

To the extent that a surge in recall rates occurs during film-to-digital transition, the problem will be magnified during periods in which the rate of digital adoption is high.

The other factor is the attention all matters involving imaging tend to get in politics and the lay media. One of the latest examples was an article in The New York Times in April headlined, "In shift to digital, more repeat mammograms." In a survey of 10 radiologists, eight reported that during the transition from film to digital, recall rates went up in women who were ultimately found to have nothing wrong.

Of course, the newspaper did a nice job of exploring the fear and anguish felt by women who are called back for more imaging, even though they are advised that most repeat tests are normal. "I know it's not a small thing, the anxiety," one radiologist was quoted as saying. "Patients are practically in tears because they're so worried." But, he added, "In the long run, it's going to be to everybody's benefit."

That last point is a key one for breast imagers to keep in mind as this transition to digital mammography continues, and we underscore it in our cover story. The article by Dr. Stamatia Destounis details the experience of the Elizabeth Wende Breast Center in Rochester, NY, in adopting digital mammography. Her institution has been moving to digital imaging since 2003. Initially, recall rates remained steady and on par with film, but they jumped up in 2006 to about 8% (still below the national 10% rate) with the adoption of additional digital mammography systems. The following year recall rates dropped to 7.7%, and the center is anticipating further reductions this year.

Destounis concludes that generally, centers should anticipate a six-month period to adjust for workflow changes, PACS training, prior (probably film) and current (digital) image discrepancies, and simply learning how to read mammograms digitally after learning how to read them on film.

But making the transition is worth the effort. Already, the Digital Mammographic Imaging Screening Trial (DMIST) has found that digital mammography is better than screen-film at spotting cancer in women under the age of 50 who have dense breasts. It is likely that further gains will emerge as digital mammography becomes more widely used and better understood.