Intended for healthcare professionals

CCBYNC Open access

Rapid response to:

Research

Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial

BMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g366 (Published 11 February 2014) Cite this as: BMJ 2014;348:g366

Rapid Response:

Re: Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial

When a study demonstrates a reduction of breast cancer mortality of approximately 10% (as in the CNBSS trial’s non prevalence screening rounds) or approximately 20% (as in the UK Age Trial and Malmo I) but does not reach statistical significance, the study had insufficient statistical power on which to base the statistical significance test as a 10% to 20% reduction effect is large enough to be statistically significant provided an adequate number of samples are collected. In such a situation, more samples were needed to further confirm the mortality reduction demonstrated from the few samples available and to allow the analysis of this mortality reduction effect to reach statistical significance provided that the additional samples follow the same trend of reduced breast cancer mortality. A lack of statistical significance is not the same as an insignificant effect. It is dangerous to ignore the fact that the UK Age Trial reported “Mortality reduction adjusted for non-compliance in women actually screened was estimated as 24%” [1]. It is dangerous to ignore the fact that the Malmo I trial reported that “women in the study group aged greater than or equal to 55 had a 20% reduction in mortality from breast cancer” and that “Cancers classed as stages II-IV comprised 33% (190/579) of cancers in the study group and 52% (231/443) in the control group” [2]. When you have clearly positive results demonstrating benefits from mammographic screening it is particularly dangerous to base one’s conclusions on the flawed statistical significance test.

Hopefully women will ignore Mr. Hefti’s opinions as I fear they may very well lead to preventable breast cancer deaths. I will not be investing any additional time and effort debating opinions supported by arguments based on no relevant research training such as: citing papers that don’t support the statements presented, producing quotations not included in the studies cited, ignoring or not comprehending the limitations of statistical significance testing while erroneously suggesting I would heavily focus on statistically significant findings supportive of my arguments (I think statistical significance should be ignored as much as possible) and so on. I would caution any woman from making a decision about breast cancer screening based on his opinions.

Mr. Hefti’s contribution to this debate has a negative impact on the legitimate concerns of those scientists keen to identify the actual shortcomings of mammographic breast cancer screening by contributing to the perception that those critical of mammographic screening are biased and must resort to weak and misleading tactics. I am keen to fully comprehend all of the shortcomings of mammographic breast cancer screening and much of my research interests are focused on overcoming those limitations. Although criticizing mammographic screening is helpful to me as motivation for my existing research interests and all my presentations on breast cancer screening include criticisms of mammography, it is also clearly important to accurately assess the mortality benefits of mammographic breast cancer screening, whatever they may be.

Jacob Levman, PhD
Centre of Excellence in Personalized Healthcare
Institute of Biomedical Engineering
University of Oxford

References:
[1] Moss SM, Cuckle H, Evans A, Johns L, Waller M, Bobrow L; Trial Management Group, "Effect of mammographic screening from age 40 years on breast cancer mortality at 10 years' follow-up: a randomised controlled trial", Lancet. 2006 Dec 9;368(9552):2053-60.
[2] Andersson I, Aspegren K, Janzon L, Landberg T, Lindholm K, Linell F, Ljungberg O, Ranstam J, Sigfússon B, "Mammographic screening and mortality from breast cancer: the Malmö mammographic screening trial", BMJ. 1988 Oct 15;297(6654):943-8.

Competing interests: No competing interests

22 April 2014
Jacob Levman
Researcher
Institute of Biomedical Engineering, Univeristy of Oxford
Parks Road, Oxford, OX1 3PJ