The article in this week's Lancet, "Is screening for breast cancer with mammography justifiable?", in which the Danish authors conclude that it is not, is based on a complex statistical analysis of previous mammography trials, and not on original research involving a new set of patients.
There is a trend within medicine at present to perform "meta-analysis" in an effort to produce evidence-based guidelines for doctors to follow in the treatment of patients.
A meta-analysis is a pooling of the results of past research, in this case all previous studies of the effect of screening with mammography on the reduction of breast cancer mortality, with a view to increasing the statistical relevance. As with any mixing of ingredients, however, the outcome is not always predictable.
There are a number of criticisms that can be levelled at the article. The authors are primarily statisticians and not clinicians. The main plank of their analysis is the lack of a perfect age distribution in the previous trials. This issue might be of concern if the trials were looking at treatment outcome, but is less relevant in the scientific assessment of screening procedures.
Other sources of bias in trials of breast-cancer screening are more relevant, and the Canadian, English and Scandinavian authors of the original studies have minimised these biases. Factors such as screening interval and duration of follow-up are key factors in this regard.
Breast Check, the recently introduced Irish national breast screening programme, has been meticulous in its planning: issues such as attendance rates, quality of the mammograms, accuracy in reading of the films and decisions for referral have all been factored into the equation.
The evaluation of the outcome of cancer screening at a national level is a long-term proposition. The Irish programme is clear about its aims and about the importance of quality control in the assessment of the screening programme. The screening will be limited to women aged from 50 to 64, the group in which the strongest benefits of screening have been demonstrated internationally. And the National Breast Screening Programme Committee has been careful to recruit pathologists, radiologists and surgeons with international experience in the area of screening.
It is important, however, to point out that the impact on mortality will be detectable only after following up the entire screening population after a minimum of 10 years. Screening is very much a long-term investment in the nation's health. Ultimately we should see a 20 per cent reduction in mortality from breast cancer in women aged from 50 to 64.
It can be difficult to quantify the individual benefit of screening for a particular patient. There are anxieties associated with the process, although Breast Check has assured women that those who have a positive mammogram will be quickly investigated.
Inevitably, there will be a number of false positive results (a suggestion of a tumor when in fact there is none present) which will impinge negatively on the individuals concerned.
For those women whose cancers will be picked up early and treated with minimal invasion, the process will be well worth it. Certainly, the research in the Lancet should not put women off screening, an invitation to which they will receive in the post over the coming months.