Becker and Junkermann mention a 10% overdiagnosis rate as a possible harm resulting from mammography screening. In stating this, the authors have done a much better job than the current mammography information sheet of the German Joint Federal Committee, which is supposed to educate women about mammography but does not mention this very serious harm that can result from an early detection program.
The authors, however, fail to take overdiagnosis into account when they write that the mortality from breast cancer is 31 of 100 unscreened women compared to 20 of 100 screened women, corresponding to a 35% reduction in mortality.
(1) If the overdiagnosis rate is 10%, then, of 100 tumors detected by mammography, 10 would not have reached medical attention. The 20 deaths among these 100 women would therefore all be in the group of 90 women with potentially fatal tumors, which the authors compare to a group of 100 women whose tumors did reach medical attention. If we renormalize the 20 deaths among 90 women to a group of 100 women, we arrive at a figure of 22 deaths, and therefore to a 29% (not 35%) reduction in mortality.
(2) The authors have taken the 10% overdiagnosis rate from the article by Zackrisson et al. (1). These authors looked at the cumulative incidence 15 years after the end of the Malmö study. Immediately after the end of the study, however, the corresponding figure was 24%. Zahl et al. arrived at an even higher figure of 30% (2). The 10% overdiagnosis figure also seems too low because the overdiagnosis rate of in situ carcinoma alone is 13% (3). If the true overdiagnosis rate is 20%, then there would be 25 deaths resulting from 100 potentially fatal tumors detected by mammography, yielding a 20% reduction in mortality compared to no screening. If the true overdiagnosis rate is 30%, the corresponding numbers are 28 deaths and a 10% reduction in mortality. This uncertainty deserves mention just as much as the phenomenon of overdiagnosis itself.
References
- 1.Zackrisson S, Andersson I, Janzon L, et al. Rate of over-diagnosis of breast cancer 15 years after end of Malmö mammographic screening trial: follow-up study. BMJ. 2006;332:689–692. doi: 10.1136/bmj.38764.572569.7C. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zahl PH, Strand BH, Mæhlen J. Incidence of breast cancer in Norway and Sweden during introduction of nationwide screening: prospective cohort study. BMJ. 2004;328:921–924. doi: 10.1136/bmj.38044.666157.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Koch K, Weymayr Ch. Kritik der Krebsfrüherkennung. Der Onkologe. 2008;14:181–188. [Google Scholar]