Overdiagnosis and the harms associated with unnecessary procedures is becoming a vibrant subject in today’s health-care dialogue, with serious implications for providers and patients alike. A new study from the Norwegian Screening Program concluded that 15% to 25% of breast cancers identified on mammography would never have been clinically significant in a woman’s lifetime.1 What are the implications of these findings, particularly within the larger discussion of overdiagnosis?
More Confusion Than Clarity?
Clearly, this is an area of ongoing controversy, not only within the oncology community, but also in the overall health-care dialogue. It is an important discussion; however, I’m not convinced that the recent study out of Norway sheds significant light on the issue of overdiagnosis. It was a laudable effort, but it may add more confusion than clarity to the dispute.
For one, so many clinicians and statisticians have argued publically in numerous letters to the editor about the study, which calls attention to potential problems with the Norwegian report. This was essentially a cohort-controlled study, which used historical comparison groups that were screened in disparate counties around Norway, leading to a lot of statistical assumptions, corrections for lead-time bias, and other issues that proved problematic. Also, many of the patients in this study were screened before the advent of digital mammography, which is a higher-contrast and more sensitive test than film mammography.
Different Perspective
That said, I’m not a statistician, so I would address the public-health issue of overdiagnosis from a different perspective. First and foremost, dying of breast cancer is a serious public health issue, and any pursuit to reduce overtreatment needs to be balanced within the context of that clinical reality. We all strive to get more accurate screening data and reduce the harms associated with potential overdiagnosis.
The focus in the United States is based on recognition that although mammography is an imperfect screening test, it does save lives. For me, the more compelling challenge is not to analyze mammography studies that looked at populations from 20 years ago, but to try to develop better technologies to improve our ability to screen patients with more sensitive and predictive tests in a cost-effective manner.
For instance, I recently saw a woman who had been getting mammograms every 6 months because of her family history. Shortly after her last normal mammogram, she presented with invasive lobular breast cancer that had spread to 20 of 37 lymph nodes. This illustrates that mammography can fail even those women who are hypervigilant about their breast health. Lobular breast cancer, which represents about 15% of all breast disease, is only one of several breast cancers that are frequently missed on screening mammography.
We can, of course, increase the sensitivity of mammography by complementing the test with ultrasonography, and in selected populations, adding breast MRI. However, this increased sensitivity also increases the number of potential false-positives, so again, I believe that we must continue to wisely use our current screening technologies and intensify our efforts to develop more accurate tests.
Proper Context
The Norwegian investigators, and others pursuing the same path, try to make the case that many of the cancers found on screening will not end up to be life-threatening. Consequently—as a public health-care policy— we can safely limit the amount of screening mammography, and by doing so, reduce waste and the associated harms such as unnecessary biopsies and lumpectomies. The debate over cost-effective screening will continue, but this study does not make a convincing case that the potential harms of mammography outweigh the benefits.
The issue of overdiagnosis is a complicated and nuanced challenge. We, the doctors who order and provide these tests, debate among ourselves about what is the best and most cost-effective way to screen and treat our patients. But when studies make headlines in the lay press, without proper context and analysis, they can do disservice to patients who are seeking information to help them make educated decisions about their health care. ■
Disclosure: Dr. Seidman reported no potential conflicts of interest.
Reference
1. Kalager M, Adami HO, Bretthauer M, et al: Overdiagnosis of invasive breast cancer due to mammography screening: Results from the Norwegian screening program. Ann Intern Med 156:491-499, 2012.
