NHS Breast Screening Programme response to Gøtzsche and Olsen Letter in the Lancet

Published 19th October, 2001

This year's work by Gøtzsche and Olsen is based on a further re-assessment of the results from the same seven studies of mammographic screening that the authors had already examined in their 2000 report. No new studies are included in their re-assessment, nor have their conclusions altered materially.

The main difference between the 2000 and 2001 reports is the way the authors classified the quality of the studies (as high, medium, poor or flawed). They classified no studies as high quality, two as medium quality, three as low quality and two as flawed. Gøtzsche and Olsen concluded that the two studies they classified as medium quality showed no benefit whereas the studies they classified as low quality or flawed showed substantial benefit from mammography; in terms of a reduction in mortality from breast cancer.

The way Gøtzsche and Olsen classified studies was based on criteria that would not be agreed by many experts in the field. Indeed many researchers would classify all seven studies as of similar quality, and when the results from all seven studies are combined, there is clear evidence of the benefit from mammography.

Gøtzsche and Olsen also claim that supplementary data they examined show that screening leads to more "aggressive" treatment of breast cancer. It is difficult to evaluate these claims, as they are based on small subgroups of individuals from two studies only classified as poor quality studies by Gøtzsche and Olsen. Furthermore, more "aggressive" treatment of breast cancer might be expected to be beneficial for women with breast cancer. Gøtzsche and Olsen did not investigate directly whether or not what they regard as "aggressive" treatment was beneficial.

In summary, the new analyses by Gøtzsche and Olsen provide no fresh evidence about the efficacy of mammographic screening for breast cancer. Overall the seven studies reviewed by the authors show that mammographic screening reduces mortality from breast cancer. However, Gøtzsche and Olsen used criteria that might not be generally accepted to classify the quality of each study and concluded that the two studies that they considered to be of the highest quality did not, by themselves, show a statistically significant benefit.

Julietta Patnick, National Coordinator of the NHS Breast Screening Programme adds;

"The NHS Breast Screening Programme has been in existence since 1988, and women are now starting to reap the benefits of a "steady state" programme. The programme is now outperforming the original Swedish Two Counties trial with an overall standard detection rate of 1.14 (Swedish Two Counties trial detection rate was 1.0), and as well as the improved SDR, more cancers are being detected at an earlier stage."

"We estimate that screening now is saving on average 1,250 lives a year. On October 31st 2001, we will publish the NHS Breast Screening Programme's annual review. The following is a preview of some of the most up-to-date statistics that show an increase in overall detection rate with the majority of cancers detected on a subsequent screen, and a significant decrease in benign biopsies (indicating a less invasive and more informed approach to screen-detected abnormalities):

  • An 8.6 per cent increase in the number of cancers detected - 9,525 in 99/00 compared to 8,771 in 98/99.
  • An 8.6 per cent increase in the number of small invasive cancers (less than 15 mm) - 4,041 in 99/00 compared to 3,722 in 98/99. Cancers smaller than 15mm are impossible to feel with the human hand.
  • The number of in situ cancers detected in 99/00 was 2,009 compared to 1,733 in 98/99.
  • 85 per cent of breast cancers are now being diagnosed without an operation.
  • The vast majority of breast cancers (over 60 per cent) are diagnosed in women attending for a subsequent screen."

"Gøtzsche and Olsen call for breast screening to be tested again in 'large, well-conducted randomised trials with the all-cause mortality as the primary outcome'. While, in principle, we would always welcome further scientific validation through such a trial, we feel it would be unethical to carry this out in the UK where screening is so well established and an integral part of this country's efforts to reduce the death toll from breast cancer, as, in a trial a number of women would have to form part of a control group and not be screened..."