Mammography: To See or Not To See

~ by Dennis Morgan, MD, Oncologist

There has been much controversy in recent years over just who should get screening mammography. The greatest contention is over what age to begin and how often to perform. Women will understandably bring a certain amount of emotion to the table. Ideally the medical community would bring curated facts to the table that inform a process of shared decision making. The major challenge for all parties involved is the curation process — not just knowing the facts, but making sense of them.

My overview of this subject comes at the behest of someone who recently underwent a harrowing encounter with mammography. As a medical oncologist my perspective is not necessarily neutral as I have an inherent wariness of the “slippery slope” of investigation and intervention that can lead to unintended, sometimes harmful, consequences. But I have no related service to promote or academic position to defend. Let me share my investigation of this controversy and invite comment.

Our first task is to separate fact from opinion. Opinion comes from personal values or professional goals. I would categorize the relevant literature into studies, reports and positions. Studies are original scientific investigations (facts), reports a critical analysis of such studies (interpretation of facts), and positions, i.e. opinions, about next steps. The landscape is dotted with any number of each. The area of hottest contention is the appropriate age bracket for screening. The value of any screening tool depends on the prevalence of the disease. In our case it is relevant the risk is proportional to age. Young women are  unlikely to have breast cancer, and the oldest women are more prone to getting it but also more often die of another condition. So the firestorm is over which of the ‘middle-aged’ women to screen.

Let’s take as our focal point the era before the publication in 2009 of the — infamous to some — U.S. Preventive Services Task Force (USPSTF) Recommendation Statement(1). This update of a 2002 paper is controversial for it’s radical departure from common practice. While supporting mammography for women age 50-74 it advised only a two-year, not annual, schedule. For women younger and older than this the task force was not persuaded of demonstrated benefit. Nor was any confidence expressed for the alternative imaging methods of digital mammography or MRI. Perhaps most shocking was their position that breast self-exam (BSE) is a waste of time and should not even be taught.

The blow back was swift and vehement, notably by The American Cancer Society (ACS) which maintains to this day that “Current evidence supporting mammograms is even stronger than in the past. In particular, recent evidence has confirmed that mammograms offer substantial benefit for women in their 40s.” Further, “Mammograms should be continued regardless of a woman’s age, as long as she does not have serious, chronic health problems … ”. (2).

And this from a letter in the New York Times by the chairwoman of the Breast Imaging Commission of the American College of Radiology and the president of the Society of Breast Imaging (3):

“Every medical organization experienced in breast cancer (including the American Cancer Society, American Congress of Obstetricians and Gynecologists, American College of Radiology, Society of Breast Imaging and National Accreditation Program for Breast Centers) recommends annual mammograms for women ages 40 and older.”

I think it is notable that every organization cited is either positioned as a patient advocate/protector or is a provider of the service. Perhaps neither would be inclined to retreat from a posture of vigilance.

Their letter was in response to an op-ed piece (4) by a co-author of Quantifying the Benefits and Harms of Screening Mammography — an MD, MPH faculty member of the Institute for Health Policy and Clinical Practice at Dartmouth(5). He cited data from the radiology community itself that the false positive rate for over ten years of annual screening is 50 percent.  He noted, “A screening program that falsely alarms about half the population is outrageous.” and “What about the benefit? Among those thousand women, 3.2 to 0.3 will avoid a breast-cancer death. If you don’t like decimals, call it 3 to 0.”. His paper is, I believe, the most comprehensive and impartial survey to date and is discussed in his interview in the ASCO Post(6).

There is a growing list of studies and reports that recommend a decrease in the use of mammography but the message seems as foreign to the American institutions cited above as the countries originating them.

From Scandinavia 2008 in the Cochrane Database of Systematic Reviews — the Nordic Cochrane report(7):

“If we assume that screening reduces breast cancer mortality by 15% after 13 years of follow-up and that over-diagnosis and over-treatment is at 30%, it means that for every 2,000 women invited for screening throughout 10 years, one will avoid dying of breast cancer and 10 healthy women, who would not have been diagnosed if there had not been screening, will be treated unnecessarily. Furthermore, more than 200 women will experience important psychological distress including anxiety and uncertainty for years because of false positive findings.”

From Norway 2010 in the New England Journal — The Norwegian Breast Cancer Screening Program(8):

“The difference in the reduction in mortality between the current and historical groups that could be attributed to screening alone was 2.4 deaths per 100,000 person-years, or a third of the total reduction of 7.2 deaths …”

From Canada 2014 in the British Medical Journal — 25 year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial(9). Annual mammography screening did not reduce long-term breast cancer mortality in women 40 to 59 years of age. An independent commentary on this study observed that “If the … results are correct, the number of “cured” drops to 0.” (10).

From Switzerland 2013 in the New England Journal of Medicine — Abolishing Mammography Screening Programs? A View from the Swiss Medical Board(11): Tasked with a recommendation for all of Switzerland the panel made several observations.

1. Conventional recommendations are based on outdated trials that do not reflect the effect of modern treatment.
2. It was not at all obvious that benefits outweigh risks when one compares a generally accepted 20% reduction in mortality with a 21.9% rate of over-diagnosis.
3. They note women substantially overestimate the benefits. “It is easy to promote mammography screening if the majority of women believe  it prevents or reduces the risk of getting breast cancer and saves many lives through early detection of aggressive tumors. We would be in favor of mammography screening if these beliefs were valid. Unfortunately, they are not, and we believe that women need to be told so.” “The board, therefore, recommended that no new systematic mammography screening programs be introduced and that a time limit be placed on existing programs.”

Why such resistance to change given the benefits are less, and the risks more than previously appreciated? Evidence contrary to closely held beliefs is always hard to accept and confirmation bias, favoring reports that agree with an a priori position, is hard to avoid. In the case of mammography several forces marshal to stave off change. I would suggest that some combination of emotion, defensive medicine, lack of scientific understanding, and a profit motive are responsible for continued over-use of mammography in the US.

Let’s see if we can tie some of this together with a visit to the doctor’s office. The office staff advises the patient that she is due for screening mammography. Often accompanied by fear and hope the staff reinforces her decision by anecdotes about others whose ‘lives were saved’ by mammography. Her hope is that the result will be negative because she could then conclude that she does not have breast cancer. But this is not necessarily so. On the other hand if the mammogram is positive she would then think that while she does have breast cancer her life will be saved by this early detection. Again, neither of these assumptions are necessarily true.

There are four possible outcomes from mammography — a positive or negative reading, either of which may be true or false. These provide the data for Bayesian analysis that is the mathematical rationale for screening tests.

The false results are instructive as to why screening mammography may continue to be overused. With false positive results, further studies — additional imaging and biopsy — will, hopefully, declare the patient cancer-free after all. The patient is reassured, thankful for the vigilance of her physicians. Unnecessary treatment avoided.

With false negative mammograms the cancer may eventually surface by some other means and, when it does, everyone will have a second look at that mammogram. Assuming it wasn’t read in error (not the same as ‘false negative’), the patient will be told that mammograms miss 20% of breast cancers.  They’ll tell her it was a “good thing she was doing BSE” or good thing that some serendipitous event lead to discovery. The fact that the mammogram in her case was of no value will probably be over-looked.

What’s worse? A breast cancer diagnosis within a year or two of a false negative mammogram or no mammogram at all? The patient is likely to have considerable negative feelings and second guess her physician’s value more than the mammograms. Negative feelings about physicians drives lawsuits. Doctors know this and often practice “defensive medicine” — better to get a test of questionable value than face negligence accusations, no matter how unwarranted.

Doctors often do not understand the limitations of screening tests. Bayesian analysis gives answers that are not intuitive for patients or physicians. Physicians routinely over-estimate the chance of cancer based on a positive mammogram. We are all prone to attaching more significance to relative changes than to absolute values as with the Norwegian study cited earlier. Are we to heed the one-third reduction or the absolute difference between 2.4 and 7.2 deaths per 100,000 person-years?

As to a profit motive, we need not necessarily find villains here. I will be quick to recognize the honest efforts of those who make a living fighting cancer. Physicians need not be greedy to cling to a profitable activity but rather just trying to keep the doors open in this era of diminishing reimbursement for physician services. However, we would be naive to dismiss the notion of a “medical-industrial complex” i.e. a socio-economic force that organically organizes to preserve profit as the primary, if not only, motive.

So we are left with a debate that has powerful advocates on both sides. The debate is not whether mammography has any value. It is rather whether we are willing to limit its use as a screening method when the harm exceeds the benefit. And the harm in this sense is both personal and societal. Each life saved comes at some cost of over-treatment death from treatment including fatal heart disease from radiation, secondary cancers and a chronic state of anxiety amongst middle-class women. By analogy, consider automobile speed limits and death rates in pedestrian-involved accidents. We could lower speed limits until the chance of a pedestrian fatality is practically zero. But at some point livelihoods and lives are lost due to the lack of efficient transportation for work and emergencies.

Hopefully, we will develop screening methods for breast cancer that are more sensitive and more specific. Until then, women and their doctors should share the decision about mammography in individual cases based on an open discussion of both sides of the ongoing controversy. We should avoid bad choices based on fear and hope alone but rather employ new information to gain the most benefit for the risk from mammography.

We are left to wonder what indeed are the best practices? Many of the issues are covered in the three-way debate in the New England Journal — screen at age 40, age 50 or not at all(12).  Dr. Welch notes in his New York Times op-ed article(4): “It has been more than 50 years since the last randomized trial of screening mammography in the United States. Now that treatment is so much better, how much benefit does screening provide? What we need is a clinical trial in the current treatment era.”

We should at least have the courage to test in this country the hypotheses posed by breast cancer screening.

Bio

Dennis Morgan, MD is Assistant Clinical Professor University of Connecticut Health Center, Emeritus Staff Johnson Memorial Hospital and Medical Center Stafford CT and Past President Connecticut Oncology Association as well as Past Medical Director Phoenix Community Cancer Center, Enfield CT

References

1. Screening for Breast Cancer: U.S. Preventive Services Task Force Recommendation Statement. U.S. Preventive Services Task Force. Ann Intern Med. 2009;151(10):716-726.
2. American Cancer Society recommendations for early breast cancer detection in women without breast symptoms. 2014.
3.  Weighing The Value of Mammograms. Monsees B, Rebner M. The Opinion Pages. Letters. New York Times. Jan 2 2014.
4. Breast Cancer Screening: What We Still Don’t Know. Welch HG. The Opinion Pages. New York Times. Dec 29 2013.
5. Quantifying the Benefits and Harms of Screening Mammograph. Welch HG, Passow HJ. JAMA Intern Med. 2014;174(3):448-454
6. Confronting Uncertainty About the Harms and Benefits of Screening Mammography. Bath C. ASCO Post. Feb 15 2014, Volume 5, Issue 3.
7. Screening for breast cancer with mammography (Review). Gøtzsche PC, Jørgensen KJ. The Cochrane Library 2013, Issue 6.
8. Effect of Screening Mammography on Breast-Cancer Mortality in Norway. Klager M et al. N Engl J Med 2010; 363:1203-1210. Sep 23 2010.
9. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. Miller AB. BMJ 2014;348:g366.
10. Annual mammography screening did not reduce long-term breast cancer mortality in women 40 to 59 years of ag. Fletcher SW. ACP Journal Club | Volume 160 • Number 10. May 20 2014
11. Abolishing Mammography Screening Programs? A View from the Swiss Medical Board. Biller-Andorno N, M.D., Ph.D., Jüni P, M.D. N Engl J Med 2014; 370:1965-1967. May 22 2014.
12. Mammography Screening for Breast Cancer. Clinical Decisions. N Engl J Med 2012; 367:e31. Nov 22 2012.