Letters

Survival and reduction in mortality from breast cancer

BMJ 2000; 321 doi: https://doi.org/10.1136/bmj.321.7274.1470/a (Published 09 December 2000) Cite this as: BMJ 2000;321:1470

Impact of mammographic screening is not clear

  1. Michael Baum, professor (mbaum@ucl.ac.uk)
  1. Royal Free and University College Medical School, University College London, Academic Division of Surgical Specialties, Gower Street Campus, Department of Surgery, London W1P 7LD
  2. VA Outcomes Group, 111ECP, White River Junction, Vermont, VT 05009, USA
  3. Clinical Chemistry Department, Queen's Hospital, Burton-on-Trent, Staffordshire DE13 0RB
  4. Chemical Pathology Department, St Thomas's Hospital, London SE1 7EH
  5. Centre for Cancer Epidemiology, University of Manchester, Christie Hospital NHS Trust, Manchester M20 4QL
  6. Cancer and Public Health Unit, London School of Hygiene and Tropical Medicine, London WC1E 7HT
  7. Scottish Cancer Intelligence Unit, Information and Statistics Division, Trinity Park House, Edinburgh EH5 3SQ
  8. Demography and Health Division, Office for National Statistics, London SW1V 2QQ
  9. Department of Palliative Medicine, St Thomas's Hospital, London SE1 7EH

    EDITOR—We should all rejoice that there has been an improvement in survival and reduction in mortality for carcinoma of the breast, but Richards et al in their paper perpetuate the myth that this is related to the breast screening programme.1 The periods for comparison were 1981-5 and 1986-90.

    The Forrest report on mammographic screening was published in 1986,2 the first screening centres were established in 1988, and the country was not covered by the programme until 1990. Even the greatest zealots for mammographic screening would not expect an impact on mortality until 1997. The fall in mortality could therefore be attributed only to improvements in treatment, and it is relevant to note that the first overview of the trials of adjuvant systemic treatment were published in 1985.3 The only support for the assertion that the reduction in mortality can be attributed to the breast screening programme was a personal communication from S M Moss. Many people are of the opinion that mammographic screening is saving thousands of lives, but opinion alone does not provide sufficient data to support a publication in a prestigious journal such as the BMJ.

    References

    1. 1.
    2. 2.
    3. 3.

    Diagnostic practice in the United States is different

    1. H Gilbert Welch, editor, Effective Clinical Practice
    1. Royal Free and University College Medical School, University College London, Academic Division of Surgical Specialties, Gower Street Campus, Department of Surgery, London W1P 7LD
    2. VA Outcomes Group, 111ECP, White River Junction, Vermont, VT 05009, USA
    3. Clinical Chemistry Department, Queen's Hospital, Burton-on-Trent, Staffordshire DE13 0RB
    4. Chemical Pathology Department, St Thomas's Hospital, London SE1 7EH
    5. Centre for Cancer Epidemiology, University of Manchester, Christie Hospital NHS Trust, Manchester M20 4QL
    6. Cancer and Public Health Unit, London School of Hygiene and Tropical Medicine, London WC1E 7HT
    7. Scottish Cancer Intelligence Unit, Information and Statistics Division, Trinity Park House, Edinburgh EH5 3SQ
    8. Demography and Health Division, Office for National Statistics, London SW1V 2QQ
    9. Department of Palliative Medicine, St Thomas's Hospital, London SE1 7EH

      EDITOR—Richards et al seem to have made inferences about “deaths avoided” using data on five year survival.1 This measure is, however, powerfully affected by diagnostic practice and is not a reliable indicator of mortality.2

      In the United States the problem is best exemplified by prostate cancer. Five year survival has increased from about 40% in the 1950s to about 95% currently.3 Although it is tempting to conclude that we Americans have made major medical advances (and left the United Kingdom in the dust), the truth is that this largely reflects our diagnostic practice. As we aggressively seek and find early stage (and often innocuous) tumours, …

      View Full Text

      Sign in

      Log in through your institution

      Subscribe