Cochrane for Clinicians: Putting Evidence into Practice

Should We Offer Routine Breast Cancer Screening with Mammography?



FREE PREVIEW Log in or buy this issue to read the full article. AAFP members and paid subscribers get free access to all articles. Subscribe now.


FREE PREVIEW Subscribe or buy this issue. AAFP members and paid subscribers get free access to all articles.

Am Fam Physician. 2003 Jul 15;68(2):260-262.

This clinical content conforms to AAFP criteria for evidence-based continuing medical education (EB CME). EB CME is clinical content presented with practice recommendations supported by evidence that has been systematically reviewed by an AAFP-approved source. The practice recommendations in this activity are available at www.update-software.com/abstracts/ab001877.htm.

Clinical Scenario

A multiparous, 45-year-old woman attends your clinic for an annual physical examination. She does not have a family history of breast cancer and is not currently using estrogen therapy.

Clinical Question

Should we offer this patient routine breast cancer screening with mammography?

Evidence-Based Answer

While there is good evidence that mammography starting at age 50 reduces breast cancer mortality, this meta-analysis1does not provide evidence of a mortality benefit for mammography screening in women aged 40 to 49. Controversy exists over whether to include in the analysis trials with methodologic flaws that would strengthen the evidence of a benefit.2Nevertheless, there is a growing consensus that physicians should encourage the use of screening mammography in women aged 40 and older. The review did not address screening intervals or clinical breast examination.

Cochrane Abstract

Background. Mammographic screening for breast cancer is controversial, as reflected in greatly varying national policies.

Objectives. To assess the effect of screening for breast cancer with mammography on mortality and morbidity.

Search Strategy. The sources of data included MEDLINE (May 16, 2000), the Cochrane Breast Cancer Group's trial register (January 24, 2000) and reference lists, letters, abstracts, and unpublished trials. Authors were contacted.

Selection Criteria. Randomized trials comparing mammographic screening with no mammo-graphic screening.

Data Collection and Analysis. Data were extracted independently by both authors.

Primary Results. Seven completed, eligible trials involving 500,000 women were identified. The two best trials provided medium-quality data and, when combined, yield a relative risk for overall mortality of 1.00 (95 percent confidence interval [CI], 0.96 to 1.05) after 13 years. However, the trials are underpowered for all-cause mortality, and CIs include a possible worthwhile effect as well as a possible detrimental effect. If data from all eligible trials (excluding flawed studies) are considered, the relative risk (RR) for overall mortality after 13 years is 1.01 (95 percent CI, 0.99 to 1.03).

The best trials failed to show a significant reduction in breast cancer mortality (RR, 0.97; 95 percent CI, 0.82 to 1.14). If data from all eligible trials (excluding flawed studies) are considered, the RR for breast cancer mortality after 13 years is 0.80 (95 percent CI, 0.71 to 0.89). However, breast cancer mortality is considered an unreliable outcome and biased in favor of screening. Flaws are the result of differential exclusion of women with breast cancer from analysis and differential misclassification of cause of death.

Reviewers' Conclusions. The currently available reliable evidence does not show a survival benefit of mass screening for breast cancer (and the evidence is inconclusive for breast cancer mortality). Women, clinicians, and policy makers should consider these findings when they decide whether or not to attend or support screening programs.


These summaries have been derived from Cochrane reviews published in the Cochrane Database of Systematic Reviews in the Cochrane Library. Their content has, as far as possible, been checked with the authors of the original reviews, but the summaries should not be regarded as an official product of the Cochrane Collaboration; minor editing changes have been made to the text (http://www.cochrane.org)

Did the authors address a focused clinical question? Yes.

Were the criteria used to select articles for inclusion appropriate? Yes.

Is it likely that important relevant articles were missed? No.

Was the validity of the individual articles appraised? Yes.

Were the assessments of studies reproducible? Yes.

Were the results similar from study to study? No. There was statistical heterogeneity across studies. Evidence for reduction in breast cancer mortality was seen only when poor-quality studies were included. The analysis was under-powered to detect benefits in all-cause mortality, and results suggested a possible benefit or harm from mammography on all-cause mortality.

How precise were the results? Breast cancer mortality results were reported as “unreliable.”

Can the results be applied to patient care? Yes.

Do the conclusions make biologic and clinical sense? Yes. Breast cancer and death from breast cancer are less common in younger women, and mammography is less sensitive in women younger than age 50.3Furthermore, other studies have shown that benefits of earlier screening typically do not become apparent until after age 50.4,5

Are the benefits worth the harms and cost? Yes. While there are psychologic harms from false-positive results, the overall benefit in mortality and cost-effectiveness supports screening in women over age 50. There is less benefit in screening women under age 50, and the decision on whether to screen should be based on the patient's level of risk, concern about breast cancer, and concern about false-positive results.

Practice Pointers

When considering whether to implement a screening test, several factors must be considered: (1) Does early diagnosis lead to improved survival or quality of life, or both? (2) Are early-diagnosed patients willing partners in the treatment strategy? (3) Are the time and energy it takes to confirm the diagnosis and provide lifelong care well spent? (4) Do the frequency and severity of the target disorder warrant this degree of effort and expenditure?8

Reading the Numbers

Misclassification refers to errors in the categorization of an intervention or a disease state. Misclassification errors are present in most studies, but the consequences of this type of bias depend on whether the misclassification is random.6When misclassification occurs equally in two groups (e.g., equal proportions of women with breast cancer are excluded from both groups in the analysis), misclassification is “random” or “nondifferential.” However, if more cases of breast cancer were excluded from the control group than from the mammography group, the misclassification is not independent of the disease status and is termed “nonrandom” or “differential.” According to the Cochrane review authors, differential misclassification occurred in the New York trial.7The authors wrote: “We calculated that 853 women were excluded from the screened group because of previous breast cancer, compared with only 336 in the control group. If only 10 percent of these excluded breast cancer cases are added as breast cancer deaths after 18 years of follow-up, the breast cancer mortality becomes higher in the screened group than in the control group, since the difference in breast cancer mortality at that time was 44 deaths.”1

Reading the Numbers

View Table

Reading the Numbers

Misclassification refers to errors in the categorization of an intervention or a disease state. Misclassification errors are present in most studies, but the consequences of this type of bias depend on whether the misclassification is random.6When misclassification occurs equally in two groups (e.g., equal proportions of women with breast cancer are excluded from both groups in the analysis), misclassification is “random” or “nondifferential.” However, if more cases of breast cancer were excluded from the control group than from the mammography group, the misclassification is not independent of the disease status and is termed “nonrandom” or “differential.” According to the Cochrane review authors, differential misclassification occurred in the New York trial.7The authors wrote: “We calculated that 853 women were excluded from the screened group because of previous breast cancer, compared with only 336 in the control group. If only 10 percent of these excluded breast cancer cases are added as breast cancer deaths after 18 years of follow-up, the breast cancer mortality becomes higher in the screened group than in the control group, since the difference in breast cancer mortality at that time was 44 deaths.”1

The analysis primarily addresses the first question. There appears to be a benefit in reducing breast cancer mortality in the 50- to 69-year range, but this meta-analysis did not detect a significant mortality benefit in women aged 40 to 49. However, in a less restrictive meta-analysis, the U.S. Preventive Services Task Force (USPSTF) determined a summary RR for breast cancer mortality of 0.85 (95 percent CI, 0.73 to 0.99). The key issue is whether the recognized flaws of several of the mammography trials are serious enough to disqualify them from inclusion in a meta-analysis. The USPSTF determined that observed mortality reductions in the “flawed” trials were not likely to be explained by the biases potentially introduced by the flaws. The USPSTF included these trials, whereas the Cochrane review authors thought that the trials were “fatally” flawed and should be excluded. The Cochrane review did not include summary data on flawed trials, but other meta-analyses have demonstrated a weaker but significant benefit with screening initiated in the 40- to 49-year age group and evidence of increasing benefit and cost-effectiveness with age. Most women diagnosed with early-stage breast cancer are willing to undergo treatment9and report a favorable quality of life.10 Mammography in selected women is cost-effective and comparable to screening for cervical cancer.

When considering one of the included studies,11screening with mammography every two years beginning at age 50 prevented one breast cancer death for every 4,000 women screened in a given year, per 1,460 mam-mographic examinations, per 13.5 biopsies, and per 7.4 breast cancers detected.12 In the United States in 2001, approximately 203,500 new cases of breast cancer were diagnosed, and 39,600 women died of the disease.13 Given the frequency and severity of breast cancer, which is the second leading cause of cancer death in women,13 the expense and potential harm of mammography seem to be warranted.

Epidemiologic data from the United States and the United Kingdom demonstrate a dramatic decline in breast cancer mortality beginning in the late 1980s.14,15 It is not known, however, to what extent this benefit is the result of widespread use of mammography or improved treatment regimens with adjuvant chemotherapy and radiation therapy. Women who are diagnosed early and undergo treatment have lower mortality rates than women who do not. Identifying and screening women at high risk,16 improving the proportion of eligible women screened,12 and eliminating disparities in access to mammography and treatment are evidence-based ways to reduce breast cancer mortality.17

Sean P. David, M.D., S.M., is assistant professor of family medicine at Brown Medical School, Pawtucket, R.I., and a doctoral student in pharmacology at the University of Oxford, U.K. Dr. David's work as a C. Everett Koop Cancer Fellow, Robert Wood Johnson Generalist Physician Faculty Scholar, and National Institute on Drug Abuse Mentored Clinical Scientist examines genetic influences on smoking cessation and cancer prevention.

Address correspondence to Sean P. David, M.D., S.M., Brown University Center for Primary Care and Prevention, 111 Brewster St., Pawtucket, RI 02860 (e-mail: sean_david@brown.edu). Reprints are not available from the author.

REFERENCES

1. Olsen O, Gotzsche PC. Screening for breast cancer with mammography. Cochrane Database Syst Rev. 2001:CD001877.

2. Humphrey LL, Helfand M, Chan BK, Woolf SH. Breast cancer screening: a summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med. 2002;137:347–60.

3. Tabar L, Fagerberg G, Duffy SW, Day NE. The Swedish two county trial of mam-mographic screening for breast cancer: recent results and calculation of benefit. J Epidemiol Community Health. 1989;43:107–14.

4. Nystrom L, Andersson I, Bjurstam N, Frisell J, Nordenskjold B, Rutqvist LE. Long-term effects of mammography screening: updated overview of the Swedish randomised trials. Lancet. 2002;359:909–19.

5. Humphrey L, Chan BK, Detlefsen S, Helfand M. Screening for breast cancer. Systematic Evidence Review no. 15. Rockville, Md.: Agency for Healthcare Research and Quality, 2002.

6. Hennekens CH, Buring JE, Mayrent SL. Case-control studies. In: Epidemiology in medicine. Boston: Little, Brown, 1987:147–8.

7. Shapiro S. The status of breast cancer screening: a quarter of a century of research. World J Surg. 1989;13:9–18.

8. Jaeschke R, Guyatt GH, Sackett DL. Users' guides to the medical literature. III. How to use an article about a diagnostic test. B. What are the results and will they help me in caring for my patients? The Evidence-Based Medicine Working Group. JAMA. 1994;271:703–7.

9. Wilson RE, Donegan WL, Mettlin C, Natarajan N, Smart CR, Murphy GP. The 1982 national survey of carcinoma of the breast in the United States by the American College of Surgeons. Surg Gynecol Obstet. 1984;159:309–18.

10. Lindley C, Vasa S, Sawyer WT, Winer EP. Quality of life and preferences for treatment following systemic adjuvant therapy for early-stage breast cancer. J Clin Oncol. 1998;16:1380–7.

11. Tabar L, Chen HH, Duffy SW, Krusemo UB. Primary and adjuvant therapy, prognostic factors and survival in 1053 breast cancers diagnosed in a trial of mammography screening. Jpn J Clin Oncol. 1999;29:608–16.

12. Colditz GA, Hoaglin DC, Berkey CS. Cancer incidence and mortality: the priority of screening frequency and population coverage. Milbank Q. 1997;75:147–73.

13. Centers for Disease Control and Prevention. Cancer facts & figures 2002. Accessed May 2003 at: www.cancer.org/downloads/STT/CancerFacts&Figures2002TM.pdf.

14. Peto R, Boreham J, Clarke M, Davies C, Beral V. UK and USA breast cancer deaths down 25% in year 2000 at ages 20–69 years. Lancet. 2000;355:1822.

15. Peto R. Mortality from breast cancer in UK has decreased suddenly. BMJ. 1998;317:476–7.

16. Moller P, Borg A, Evans DG, Haites N, Reis MM, Vasen H, et al. Survival in prospectively ascertained familial breast cancer: analysis of a series stratified by tumour characteristics, BRCA mutations and oophorectomy. Int J Cancer. 2002;101:555–9.

17. Swan J, Breen N, Coates RJ, Rimer BK, Lee NC. Progress in cancer screening practices in the United States: results from the 2000 National Health Interview Survey. Cancer. 2003;97:1528–40.

The Cochrane Abstract is a summary of a review from the Cochrane Library. It is accompanied by an interpretation that will help clinicians put evidence into practice. Sean P. David, M.D., S.M., presents a clinical scenario and question based on the Cochrane Abstract, along with the evidence-based answer and a full critique of the abstract. This series is part of AFP's CME. See “Clinical Quiz” on page 209.


Copyright © 2003 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 afpserv@aafp.org for copyright questions and/or permission requests.

Want to use this article elsewhere? Get Permissions


Article Tools

  • Print page
  • Share this page
  • AFP CME Quiz

Information From Industry

More in Pubmed

Navigate this Article