The Controversy About Screening Mammography
By Marcus M. Reidenberg, MD, FACP
Weill Cornell CERT
The conflict between the U.S. Preventive Services Task Force recommendation for screening mammography for asymptomatic women every 2 years starting at age 50 (1) and that of the American Cancer Society and other organizations to screen annually starting at age 40 (2) is related to the values attached to the data by the different groups. What these groups ignored is the individual woman’s risk.
The actual facts are that most studies show that following the introduction of screening programs, fatal breast cancers decrease by about 15% (2). The average woman’s risk of getting breast cancer is about 1 in 8, the accumulation of risk over an entire lifetime. Only one 40-year-old woman out of 69 will get breast cancer by the time she is 50. With present day treatment, many of these cancers won’t progress to kill the woman even if they are detected later than they would be detected by screening. Unfortunately, some will progress irrespective of early diagnosis and treatment. Thus, screening is only of value in detecting those cancers that would progress if not caught early and are responsive to treatment. Because breast cancer is uncommon in young women, one needs to ask 1904 women between ages of 39-49 to be screened to prevent 1 death (2).
These statistics are for average women. There are several risk factors that increase the risk of breast cancer. Breast cancer in a first degree relative raises the risk 4 fold. If one never was pregnant or had her first pregnancy over the age of 30, the risk doubles or triples. There are many other risk factors each contributing a small amount of additional risk. One can learn a woman’s risk and compared to an average woman’s risk by the National Cancer Institute’s Breast Cancer Risk Assessment Tool at: http://www.cancer.gov/bcrisktool/. For those women at higher than average risk, the benefit of screening may be better than the average statistics while for those at lower risk, the benefit to harm ratio will be less.
The harms of screening are false positive mammograms with the subsequent medical work-ups and the psychological after-effects. About 9% of mammograms give false positive results. Thus, for 39-49 year-old women being screened, 186 false positives occur for each cancer found. Since 20-30% of those women need breast biopsies to make the diagnosis, 38 to 56 women receive breast biopsies for each case of cancer detected. It has been estimated that half the women having annual screening mammograms for 10 years have had at least one false positive mammogram (4). For older women, the number of false positives is fewer per cancer found but now the problem is overdiagnosis (3). Cancers too small to be detected by more traditional ways may grow so slowly that another disease such as heart disease or lung cancer is fatal before the breast cancer causes trouble.
Women worry about breast cancer, even after the positive mammograms are proven to be false. In one study, three months after the breast evaluation proved the lesion seen was not cancer, 21% of women worried so much they said it affected their mood. Seventeen percent of those who had highly suspicious mammograms were so worried that it compromised their ability to engage in their daily activities 3 months after the work-up showed they did not have cancer (5).
The difference between screening annually and screening every 2 years is that screening every two years reduces the false positives by 50% and the early cancer detection by 19% (2). Different people evaluate the importance of these numbers differently, hence the conflict in recommendations.
An undercurrent of skepticism about the value of screening has been present for a while (6,7). Is it the earlier diagnosis, the continual improvement in treatment or both that has lowered the mortality? A new publication compared breast cancer deaths in European countries that started screening a decade before other countries. The countries were: Sweden (early) compared to Norway (late), Netherlands (early) compared to Belgium (late), and Northern Ireland (early) compared to Republic of Ireland (late). In each country, breast cancer deaths decreased over time. But the death rate fell at the same time and at the same rate in each country in the pair irrespective of when screening was started. These data call into question if screening has really improved outcomes or if improved treatment rather than earlier diagnosis is the actual cause of the improved outcomes in the time after screening has been implemented (8). More good data are needed to be sure of the answer.
All of the organizational recommendations about age to start screening mammograms and frequency of follow-up are based on the same data. Thus, the differences in the recommendations must be based only on personal value systems and the subjective weighting of the same facts. It is unfortunate that, in this era of “personalized medicine,” none of the recommendations take into account the individual woman’s easily determined risk of breast cancer using the N.C.I.’s tool on its web site. Since some women with risk factors have greater than average risk, others will have less than average risk. More women at lower than average risk will have to be screened to detect one cancer than those of average risk for which the above statistics have been given. Women at low risk will have more false positive mammograms for each case of cancer detected than women at high risk. The benefit to harm ratio of screening mammograms varies with the risk of the woman to get the cancer. Considering the overall benefits and harms together, one can conclude that screening may be of value for women at higher than average risk and is of less, if any, value for women at lower than average risk.
People who weigh the same facts differently can produce different recommendations from the same set of facts. Organizations making recommendations for mammography screening should, at least, acknowledge the subjectivity with which they weigh each fact and that a woman’s specific risk of breast cancer affects the benefit to harm ratio of screening.
Update March 3, 2014
A new study from Canada followed women for 25 years after being in a screening mammography study (9). The death rate from breast cancer in the mammography screened group was the same as the rate in the control group that had annual physical exams but no X-ray screening. Thus screening mammography X-rays did not lower the mortality from breast cancer in these women compared to those who had a breast physical exam alone. One reason why older studies showed benefit from screening mammograms while this one did not is that good adjuvant therapy including tamoxifen was available in Canada for all the patients during the time of this follow up. Good adjuvant therapy has made an enormous improvement in survival after breast cancer.
References:
- US Preventive Services Task Force. Screening for breast cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2009 Nov 17;151(10):716-26, W-236.
- The American College of Obstetricians and gynecologists. Brea st Cancer Screening. Obstet Gynecol 2011;118(2):372-82.
- Welch HG, Black WC. Overdiagnosis in cancer. J Natl Cancer Inst. 2010 May 5;102(9):605-13.
- Fletcher SW. False-positive screening mammograms: good news, but more to do. Ann Intern Med. 1999 Jul 6;131(1):60-2.
- Lerman C, Trock B, Rimer BK, Boyce A, Jepson C, Engstrom PF. Psychological and behavioral implications of abnormal mammograms. Ann Intern Med. 1991 Apr 15;114(8):657-61.
- Olsen O, Gotzche PC. Cochrane review on screening for breast cancer with mammography. Lancet 2001; 358: 1240-3.
- Horton R. Screening mammography- an overview revisited. 2001; 358: 1284-5.
- Autier P, Boniol M, Gavin A, Vatten LJ. Breast cancer mortality in neighbouring European countries with different levels of screening but similar access to treatment: trend analysis of WHO mortality database. BMJ. 2011 Jul 28;343:d4411. doi: 10.1136/bmj.d4411.
- Miller AB,et al. Twenty-five year follow up for breast cancer incidence and mortality of the Canadian national Breast Screening Study. BMJ 2014; 348: g366 doi: 10.1136/bmj.g366.
Posted 9/8/2011
Updated 3/3/2014
This note can be found online at http://www.weill.cornell.edu/cert/patients/screening_mamography.html
Health information for everyone from the Weill Cornell/HSS CERT http://www.weill.cornell.edu/cert/patients/