How effective is screening for breast cancer?BMJ 2000; 321 doi: https://doi.org/10.1136/bmj.321.7262.647 (Published 16 September 2000) Cite this as: BMJ 2000;321:647
Reductions in mortality should not be the only marker of success
- Lennarth Nyström, senior lecturer in biostatistics (Lennarth.Nystrom@epiph.umu.se)
- Department of Epidemiology, Public Health and Clinical Medicine, Umeå University, SE-901 85 Umeå, Sweden
Papers p 665
Many countries initiated national screening programmes for breast cancer after the results of the Swedish two counties trial were published in 1985.1 This trial, which took place in Kopparberg and Östergötland, found that mortality from breast cancer fell by 31% after seven years' follow up of women who had been aged 40-74 at the beginning of the trial. These findings confirmed the results of an earlier trial by Shapiro et al.2 Finland and Sweden started their programmes in 1986, the United Kingdom in 1988, and the Netherlands in 1989. Those in charge of health policy in these countries now want evidence that these programmes are achieving their goals.
Can the results of randomised controlled trials be generalised, and can the impact be maintained when findings are put into practice? The crucial point of this process is the setting of the goal, which for screening with mammography is usually a reduction in mortality from breast cancer within a certain number of years. In practice, the goals have often been based on the results of the two counties trial or sometimes just the Kopparberg part of the trial (in the latter the reduction in mortality was higher). However, sometimes the goal is set without considering the basic characteristics of the trial—for example, in Kopparberg the attendance rate for …