The Mammography Debate
Am Fam Physician. 2002 Dec 15;66(12):2211-2212.
As someone who has had a front-row seat at the mammography debate over the past few years and thinks it unlikely that more informative data will be available anytime soon, I am ready to weigh in. I will comment on the controversy about whether there is benefit to screening, on what I think the research tells us, and on what I think it means.
Mammography has been commonly recommended as a tool to reduce breast cancer-related deaths in women over age 50, although there always have been skeptics. Most of the controversy over screening has centered on younger women—especially those 40 to 49 years of age. A few years ago, a consensus conference of the National Institutes of Health (NIH) recommended against universal screening in the younger age group. An advisory board of the National Cancer Institute (NCI) overruled this recommendation after the spectacle of congressional hearings, which included a threat to terminate NCI funding.
Now we have a slightly larger research base with further long-term follow-up from several studies, especially the Swedish trials. Building on these studies, there have been two overviews from Europe published in the Lancet and by the Cochrane collaboration, a review of the Swedish trials published in the Lancet, a review performed by NIH's Physician's Data Query panel, and a new recommendation1 from the U.S. Preventive Services Task Force (USPSTF). At first glance, the reviews and recommendations appear to be farther apart than ever, ranging from assessments that mammography does not work at any age to recommending it in all women older than age 40.
First, what about claims that the flaws in most of the studies leave us with no scientific basis to conclude that screening is effective? All groups agree that the studies are imperfect, but they disagree on whether the flaws invalidate the conclusions. The Cochrane reviewers hold that the mere presence of imperfections invalidates the studies. The USPSTF recognizes the flaws but considers whether they could have biased the results. Authors associated with the USPSTF recently published an article describing how this approach leads to a conclusion of benefit from screening mammography. I have confidence in the methods, criteria, and reasoning of the USPSTF.
Second, if you use the USPSTF approach to evidence, what does the research tell us?
Beginning at age 40, mammography alone provides a benefit in breast cancer mortality, with evidence weaker at younger ages and stronger at older ages.
The relative benefit is a 20 to 30 percent reduction in breast cancer mortality.
The absolute benefit increases with age because the risk of breast cancer increases with age; to prevent one breast cancer-associated death, approximately 1,500 younger women or 1,000 older women need to be offered screening.
There is no evidence that screening leads to an overall (all-cause) reduction in mortality, although the power of available studies to detect such a reduction is small.
The potential adverse effects of screening include false-positive results that require biopsy, anxiety, over-diagnosis of ductal carcinoma in situ, and the theoretic risk of radiation-induced breast cancer.
Research trials used a one- to two-year screening interval, but there is no evidence that supports a preferred interval.
There is no direct evidence that screening by only self-examination or clinical examination is beneficial.
Third, what does the research mean? Yes, there is a benefit in reduced breast cancer mortality beginning at age 40; there are significant adverse effects; there is no evidence for an overall mortality benefit; and we do not know how often screening should be performed or when it can be discontinued. Screening reduces relative risk by approximately one third, and although this reduction is enormous when multiplied by millions of women, the likelihood of benefit for an individual woman is small. In a woman 60 years of age whose underlying risk of death from breast cancer in the next 10 years is 0.9 percent, a risk reduction of 33 percent yields an absolute risk reduction of 0.3 percent. This translates into 333 women who need to be screened to prevent one death; the number of women older than age 40 who need to be screened is more than 1,000. This is not much different than the number needed to screen for cervical or colorectal cancer, but it highlights the need to balance the likelihood of benefit with possible harms.
I would prefer a clinical recommendation that is a simple “yes” or “no,” but the decision to screen for breast cancer using mammography is not simple. I suggest reading the recommendation and rationale statement from the USPSTF carefully, including the clinical considerations. The evidence shows a benefit, but the age at which the benefits justify the harms is a subjective judgment. The likelihood of benefit to individual women is modest but could be large for a woman who is detected with early breast cancer. At the same time, the likelihood of harm from false-positive results and invasive procedures and treatment is substantial. Only the patient can place values on the benefits and harms; many women may feel the possibility of reducing breast cancer deaths justifies the potential harms, but some may feel comfortable deferring screening until they are older. Exactly how often to screen and when to discontinue screening are clinical judgments. I conclude that physicians should inform women about the potential benefits, the potential harms, the likelihoods of both, and the limitations of the test. Family physicians should be experts at this. Then let the patient decide, and document her decision.
1. Screening for breast cancer: recommendations and rationale from the U.S. Preventive Services Task Force. Accessed November, 2002, at: www.ahrq.gov/clinic/3rduspstf/breastcancer.
Copyright © 2002 by the American Academy of Family Physicians.
This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP. Contact firstname.lastname@example.org for copyright questions and/or permission requests.
Want to use this article elsewhere? Get Permissions