As a full-time breast radiologist and the administrator who oversees the standards and qualifications of every mammography center in the United States, I am disappointed in recent news coverage of the Canadian National Breast Screening Study, a flawed research project that resurfaces every few years, grabbing headlines each time.
AUTHORS NOTE: The Salt Lake Tribune published this an an Op-Ed on Sunday, March 23, 2014. You can read this on their website as well.
Not only has it undermined confidence in screening mammography — the only test proven in multiple randomized, controlled clinical trials to save lives — it has confused the medical community, as well. Many primary care providers lack the expertise to sift through contradictory news reports and screening recommendations in order to counsel their patients. A brief review of the facts is in order.
Since the widespread implementation of screening mammography in the early 1990s, the death rate from breast cancer has decreased by more than 30 percent. This has been confirmed many times over by well-regarded studies from around the world. In fact, here in Utah, researchers at Intermountain Healthcare have shown that women are more likely to survive breast cancer if their tumors are found during mammography, rather than during a clinical exam.
Unfortunately, this message isn’t getting through to Utah women. Many mistakenly believe that if they have no family history of the disease, they don’t need a mammogram. In fact, 85 percent of breast cancers occur in women with no significant family history. Yet, Utah women have the second-lowest screening rate in the country. Every year, approximately 1,200 Utah women are diagnosed with breast cancer and 250 die.
Screening mammography can help reduce those numbers, but only if women understand the truth: There are fundamental problems with the study they’re hearing about in the news.
First, before the randomization process even began, all participants were given a clinical breast exam, and some women with advanced disease were allowed into the screening group. This is a violation of the randomization process known as contamination bias. This explains why women in the screening arm of the Canadian study did not do as well as similar women in other trials around the world. In fact, there were many more women with advanced cancer in the screening group than one would expect with appropriate randomization.
Second, the interpretation and quality of the mammograms were immediately called into question. Not only did the interpreting radiologists receive limited training before the trial, the technologists performing the exams were not experienced with proper positioning. The quality of the images was poor, prompting the reference physicist to withdraw his support of the trial, saying: "The quality of the images ... was far below state of the art, even for that time (early 1980s.)"
During my 23-year career, I have interpreted more than 300,000 mammograms and diagnosed several thousand cases of breast cancer. I have been fortunate to work with top-notch technologists who understand the importance of positioning and technique. They know a bad mammogram is worse than no mammogram. Even the best radiologist cannot find a small cancer if the images are not taken properly.
Every time the Canadian study reaches a significant anniversary, it incorporates data that are updated, but still flawed. And every time, the study is greeted as though it was the result of cutting-edge scientific inquiry. It’s not.
My fellow radiologists and surgical colleagues continue to stand by the recommendations of the American College of Radiology, American College of Surgeons, American College of Obstetrics and Gynecology, American Cancer Society, and National Cancer Institute: Women should begin annual screening mammography at age 40.