Recently the issue of a woman’s breast density became front-page news. So now seems to be a good time to clarify what the controversy is and to critically evaluate the degree to which patients should be educated by their physicians concerning their breast density.
This past August, New York State became the fourth state to pass a Breast Density Inform Law. The legislation is simple enough. After a screening mammogram, radiologists must now—in addition to transmitting the result of her test to both the referring physician and the patient—also directly inform a patient of her breast density. Since density affects a mammogram’s accuracy, it is an independent risk factor for developing breast cancer and can be useful in guiding personal healthcare decisions. Therefore, it is important that women have this information.
Still, certain insurance groups and medical organizations opposed the bill’s passage and continue to oppose its intent. How can this be?
First, some facts: Having dense breasts, as opposed to predominantly non-dense breasts (also called predominantly fat-containing breasts), is determined by genetic and hormonal influences and is quite common. Sixty-six percent of premenopausal women and 25 percent of postmenopausal women have breasts dense enough to affect mammographic accuracy. Only a mammogram can determine density—not physical examination nor breast size nor texture. So, a woman would need a mammogram just to know her density.
It has been almost 10 years since I published two studies analyzing 27,825 consecutive patient examinations and reported that mammograms were highly accurate in women with non-dense breasts, detecting 98 percent of cancers. However, as density increases, mammographic sensitivity decreases, and in women with the densest breasts, only 40 percent of cancers were detectable by mammography. In fact, of all the factors evaluated, breast density was the most statistically significant reason a mammogram would fail to detect cancer.
Understanding this limitation, I performed a screening ultrasound on all women with dense breasts and diagnosed an additional 42 percent of women with cancer who would have been falsely labeled healthy by their mammogram. The false positive rate for ultrasound was only 5 percent, a percentage that I, and almost any woman you would ask, believe is an acceptable tradeoff for finding large numbers of otherwise undetected breast cancer. The AMA gave the study a Scientific Paper of the Year Award for 2002, and during the past 10 years, multiple investigators, including multi-institutional studies, have corroborated these findings.
The medical community is fully aware of these findings, and now accepts all of them as fact. However, this knowledge has not been adequately transmitted to the general public. While the FDA has mandated via the Medical Quality Standards Act that radiologists directly transmit mammogram results to the patient, there is no necessity to report the degree of breast density. Therefore, when describing a mammogram result to a patient as “normal,” the radiologist may be only 40 percent accurate in his or her diagnosis. Yet the patient is never informed of her mammogram’s degree of accuracy, nor have patients been educated that breast density in itself is a significant risk factor for developing breast cancer. If a patient were to be informed, she could then be an educated participant in her own medical care and have opportunity to discuss with her doctor the advisability of additional screening, such as with ultrasound or MRI. This lack of disclosure to the patient has a direct impact upon her own health care decisions.
Why would anyone oppose this bill?
Well, there are four major reasons for dissent. Examining each reason brings appreciation as to why the bill is necessary.
•The first reason for dissent is the fear of unnecessary additional testing and invasive procedures. There is concern about a lack of adequate validation (read: higher survival rates) of additional screening methods such as ultrasound and MRI. Some have suggested that these additional screens should not be performed until there has been a randomized, controlled trial proving a benefit. However, I would argue that while a randomized controlled trial is optimal, and ultimately the only way to prove that screening brings a survival benefit, it is somewhat disingenuous, and perhaps reckless, to suggest waiting. Because of their cost and complexity, randomized controlled trials will likely never be performed—and even if done, it would take 10 to 15 years for survival data to become available.
Those who argue against using screening ultrasound, which is acknowledged to diagnose an additional 42 percent of cancers missed by mammography, know full well that randomized, controlled trials will not be performed any time soon, if at all. The question is, What can we do, or should we do, in the absence of randomized, controlled data? What we do know is that ultrasound finds these additional cancers at a small size and low stage similar to those cancers found by mammography, which has been proven to save lives.
Utilizing ultrasound or MRI as a screening test requires the leap-of-reason that finding cancers early saves lives. True, not all lives will be saved with early detection, but since we do not know in advance which lives won’t be saved, should we abandon looking for those cancers that can save a woman’s life? Short of a randomized controlled trial, that is the best we can offer patients today. Shouldn’t patients know that?
Simply put, we as physicians can find large numbers of early-stage, treatable cancers that mammography misses in women with dense breasts.
Patients have a right to know this, and only then can they have a meaningful discussion with their physician as to whether additional screening is warranted.
•The second reason for dissent is that it has been argued that no consistent definition of breast tissue exists and there is no clear scientific evidence that breast tissue alone is a factor for high risk of breast cancer. While evaluation of the degree of density based on mammographic appearance is subjective, the presence of dense vs. non-dense is not. Numerous rigorous scientific publications have used breast density categories to evaluate for cancer risk, and just about all have found a significantly elevated risk of cancer in women with dense breasts. To argue otherwise and characterize density evaluation as subjective, non-reproducible, and of no clinical concern is to fixate on the less important issue of quantifying density while ignoring the increased risk density confers.
•The third reason for dissent: Physicians are correctly worried about false positive results with additional ultrasound or MRI. But who should be weighing the benefit of cancer detection vs. the anxiety and inconvenience of a false positive result? It should be the patient herself. It has been my experience that the vast majority of women would choose additional screening so as not to miss an early breast cancer while enduring the possibility of a false positive result. A “false positive” is certainly preferable to a “missed positive.”
•Finally, everyone worries about cost: The New York State bill is about patient education. The insurance reimbursement provision for screening ultrasound in this bill was removed prior to the State Senate and Assembly vote. Legislators did not want to add any costs to our health system, although it is possible—and, according to many, likely—that finding breast cancer earlier may actually balance out costs or even save money in the long run.
Compounding the controversy is the media’s misinterpreting of scientific studies and feeding misinformation to their readers.
In August, a paper was published reporting that high mammographic density was not associated with increased death compared with lower density.
Physicians and the media interpreted this as proof that there was no reason to pay any special attention to those with dense breasts; rather, they should be screened the same as those with non-dense breasts.
Even the American College of Radiology touted this as good news for patients. However, careful analysis of the study shows that the non-dense group of women were mammographically screened less often than the dense group. If both groups had been screened equally, it is reasonable to expect that the women with non-dense breasts would live longer—which, when comparing the two groups, is another way of saying that the women with dense breasts would die sooner. Since this hypothesis was not evaluated in the study, one cannot make the argument that women with dense breasts should be screened in the same fashion as those with non-dense breasts.
It is unfortunate that legislative measures have been enacted to compel doctors to transmit information to patients, since doctors (including me) never wish to have the legal system impose upon them how they should practice.
However, in this instance, legislative efforts to inform patients is about welcoming patients into the decisionmaking process when we as physicians have some—but not all—answers. At this point, only with information about their own bodies will women have a meaningful understanding upon which to base their breast health care decisions.