BMJ  2004;328:E301-E302 (12 June), doi:10.1136/bmj.328.7453.E301

BMJ USA: Editorial

Balancing mammography's benefits and harms

Are we overdiagnosing breast cancer?

Cancer screening is intuitively appealing. Common sense would dictate that early detection is good. If you had a silent, potential-life-threatening cancer, wouldn't you want to know as early as possible, when treatment might have the greatest chance of cure? Many would say, "Of course!" Some are even willing to pay a lot of money to have whole-body cancer screening with computed tomography—all for the sake of early detection.¹

Enthusiasm for cancer screening, however, may be based on misperceptions about the natural history of cancer and the benefits of early detection (see box). Indeed, some cancers may grow so slowly that they are unlikely to be diagnosed during one's lifetime, and others may regress. Detection of such cancers by screening, overdiagnosis, can lead to unnecessary, invasive treatment.

Overdiagnosis is thought to occur frequently in prostate cancer and neuroblastoma screening.^2,3 While concerns have been raised about overdiagnosis of ductal carcinoma in situ (DCIS) with screening mammography,⁴ little is known about the extent of overdiagnosis of invasive breast cancer because of screening.

In this issue of BMJ USA, Zahl et al suggest that overdiagnosis of invasive breast cancer occurs frequently with screening mammography (p 299). Studying breast cancer incidence rates in Norway and Sweden, they found that cancer incidence among women aged 50-69 years increased substantially after the implementation of screening programs (eg, 82% increase during the first year of screening in Norway). This would be expected to occur as a result of the programs. If screening detects more asymptomatic cancers at earlier ages (50-69 years), it is hypothesized that fewer women would be diagnosed with cancer at older ages (70-74 years), after screening has stopped. This is because some of the later cancers would have been diagnosed at an earlier, asymptomatic stage. However, the authors found no commensurate decline in breast cancer incidence among older women.

Common misconceptions about cancer screening All cancers progress Early detection is always a benefit Screening reduces the incidence of cancer Early detection implies reduced mortality Cancer screening is meant for patients with known symptoms Adapted from Gigerenzer10

How can we explain the persistently elevated cancer incidence rates among the older women in the study by Zahl et al? One explanation is overdiagnosis; screening detected cancers among younger women that would not have otherwise been diagnosed. Indeed, the authors assert that one in three breast cancers diagnosed by screening mammography is overdiagnosed.

Estimating the true extent of overdiagnosis, however, is challenging. The authors describe their design as a cohort study, but it is better characterized as an ecologic study, because they could not ascertain whether each woman actually received mammography. Many women probably continued to receive screening beyond age 70 years, either by invitation or self-referral, which would lead to spuriously increased breast cancer rates among the older, presumably "unscreened" group. In addition, screening was introduced gradually in Sweden; thus, it is difficult to know when to expect the hypothesized decline in incidence rate among women aged 70-74 years. Moreover, the authors could not adjust for confounding secular trends in breast cancer risk factors, such as estrogen therapy or delayed childbearing. Lastly, the observation period after introduction of the screening programs may have been too brief to observe the nadir of any subsequent decline in breast cancer incidence. Thus, a host of factors could have contributed to persistently elevated incidence among older women. While the study suggests that mammography will overdiagnose some breast cancer, the question remains: how frequently?

During busy office visits, it is difficult to thoroughly discuss with women the benefits and harms of mammography. Even if there were ample time, it would still be difficult. Nevertheless, we should strive to correct misperceptions whenever possible. Recommendations for screening mammography are based on evidence from randomized trials suggesting reduced breast cancer mortality among women aged 40 years and older. Yet many women overestimate the protective benefits of mammography⁵ and underestimate its possible risks, including the evaluation of false-positive mammograms and overdiagnosis leading to unnecessary mastectomy, radiation, or chemotherapy. Clinicians should describe potential benefits of mammography without candy-coating its plausible harms.⁶

Meanwhile, women can reduce the risk of false-positive mammography by receiving screening from the same facility over time, or by bringing previous films to a new facility for comparison. Menstruating women ideally should receive mammography during the follicular phase of the menstrual cycle (first two weeks).⁷ Astute clinicians will not recommend screening mammography for women with substantial medical comorbidity and short life expectancy, because the benefit is probably trivial while the risk of harm remains.⁸

The benefits and risks of medical interventions often exist in a tenuous balance. Despite their intuitive appeal, screening tests are no different than surgical or drug therapies in this regard. Current evidence suggests that the benefits of mammography outweigh the harms for most women over age 40 years.⁹ Mammography's benefits, however, probably come at the cost of some overdiagnosis of breast cancer.

Robert Wood Johnson Clinical Scholars Program, Department of Family Medicine, University of Washington, Seattle, WA jjfenton{at}u.washington.edu

University of Washington jelmore{at}u.washington.edu, Harborview Medical Center Seattle, WA, Robert Wood Johnson Clinical Scholars Program Seattle, WA

---

Competing interests: None declared.

JJF was a Robert Wood Johnson Clinical Scholar at the time of this work. The views expressed are those of the authors and not necessarily the Robert Wood Johnson Foundation.

Paper p 299

References

1. Schwartz LM, Woloshin S, Fowler FJ Jr, Welch HG. Enthusiasm for cancer screening in the United States. JAMA 2004;291: 71-78.[Abstract/Free Full Text]
2. Etzioni R, Penson DF, Legler JM, di Tommaso D, Boer R, Gann PH, et al. Overdiagnosis due to prostate-specific antigen screening: lessons from U.S. prostate cancer incidence trends. J Natl Cancer Inst 2002;94: 981-990.[Abstract/Free Full Text]
3. Schilling FH, Spix C, Berthold F, Erttmann R, Fehse N, Hero B, et al. Neuroblastoma screening at one year of age. N Engl J Med 2002;346: 1047-1053.[Abstract/Free Full Text]
4. Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson C. Incidence of and treatment for ductal carcinoma in situ of the breast. JAMA 1996;275: 913-918.[Abstract]
5. Black WC, Nease RF Jr, Tosteson AN. Perceptions of breast cancer risk and screening effectiveness in women younger than 50 years of age. J Natl Cancer Inst 1995;87: 720-731.[Abstract/Free Full Text]
6. Fletcher SW, Elmore JG. Clinical practice. Mammographic screening for breast cancer. N Engl J Med 2003;348: 1672-1680.[Free Full Text]
7. White E, Velentgas P, Mandelson MT, Lehman CD, Elmore JG, Porter P, et al. Variation in mammographic breast density by time in menstrual cycle among women aged 40-49 years. J Natl Cancer Inst 1998;90: 906-910.[Abstract/Free Full Text]
8. Walter LC, Covinsky KE. Cancer screening in elderly patients: a framework for individualized decision making. JAMA 2001;285: 2750-2756.[Abstract/Free Full Text]
9. 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(5 Part 1): 347-360.[Abstract/Free Full Text]
10. Gigerenzer G. Reckoning with risk: learning to live with uncertainty. London: Penguin Press, 2002.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?