Mammographic screening: the case for informed choice
Professor Michael Baum defends his controversial assertion that women are coerced into having breast screening
In certain circles I appear to have been branded a troublemaker and misogynist who wishes 'to kill one in every 1,000 women' (an accusation heard by the viewers of a Newsnight programme hosted by Jeremy Paxman in 2000), by daring to question the virtue of breast cancer screening. This article presents a further opportunity to defend myself, while defending womankind from coercive social engineering and unintended harm.
The evidence upon which national screening programmes were established emerged from a group of clinical trials completed over a decade ago and which are subject to constant reworking, reanalyses and wrangling between screening zealots and the screening sceptics. Most of the controversy seems to miss the point and so far I seem to have failed in opening eyes to a subtle and cruel deception inherent in screening.
The harms and benefits of this intervention have to be weighed up, but until now those decisions have been taken by paternalistic agents of the state rather than by the woman herself.
'Public health' is a summation of harms and benefits that are asymmetrically distributed among the population, so that the undoubted benefits to a small minority are at the expense of hidden harms to the majority. In the cases of seat belts or fluoridisation of the drinking water few suffer harm, but with medical screening of all kinds the potential for damage is huge.
The public health
The undoubtedly well-intentioned public health schemes such as the NHS breast screening programme aim to reduce breast cancer mortality by offering routine mammography to healthy individuals: a potential public good. But the public receives highly conflicting messages about its effects, hotly debated in the media on every new published estimate, couched in terms guaranteed to be misunderstood by the majority.
The latest reported '44 per cent reduction in breast cancer mortality in women aged 40-69' since screening was established1 was headlined in the media2. However, the news stories failed to mention there were also falls in breast cancer mortality attributed to improvements in treatment in those who were not screened.
Furthermore, this fall in mortality predated the start of screening in the UK and also applied to those under 50 who had not been invited for screening3. The public needs to be told about all the outcomes of screening in terms it can understand, achieved for example in Raffle and colleagues' report (on cervical screening) which failed to hit the headlines4.
All that aside, I am prepared to accept mammographic screening does in fact lead to about a 20 per cent relative risk reduction (RRR) in breast cancer mortality5, but the nub of my argument concerns the translation of RRR into absolute numbers so an intelligent woman can weigh up the chances of benefit versus harm.
The meaning of harm
Harm is often dismissed as a price worth paying for the perceived general good. Harms suffered by individuals may be physical, emotional, social, financial or psychologica · 6,7. These may be 'only' temporary around the time of having a test and while waiting for results. At the other extreme, there may be lifelong damage.
Slaytor et al reviewed 58 pamphlets used in breast-screening programmes and found they all used relative risk information about the benefits in preference to absolute risk reductions. Six pamphlets incorrectly advised that women who have regular screening mammograms every two years halve their chances of dying of breast cancer8.
Such 'framing' manipulations are highly persuasive in getting people to take tests9. There is also evidence from all screening programmes that when people are offered more detailed information about their own personal risks they are less likely to opt for tests10,11.
The meaning of benefit
The rise in the incidence of breast cancer is used in the NHS breast screening invitation to justify screening12, yet the lay public are not advised that this rise can be attributed largely to the introduction of screening which leads to an artificial increase because of the overdiagnosis of borderline pathology and ductal carcinoma in situ.
The numbers needed to screen to prevent one death, or absolute risks of dying of breast cancer, are not mentioned. The US Preventive Task Force research found it is necessary to screen 1,224 women aged 40-74, or 1,792 women younger than 50, for 14 years to prevent one death from breast cancer13. They stated that a 'more important concern' was 'whether the magnitude of benefit is sufficient to outweigh the harms'.
Ductal carcinoma in situ accounts for 20 per cent of screen-detected 'cancers'. It has an uncertain natural history, is a mystery to the laywoman14 and is an 'early' stage of disease that results in a 20-50 per cent mastectomy rate15. None of this is presented to women invited for screening.
The general public therefore has grossly inaccurate perceptions about the risk of dying of breast cancer and the benefits of screening. The uncertainties, harms, limitations and consequences of finding pathology of borderline significance are kept secret.
Of course the majority of screening episodes end with the woman feeling reassured, yet how many realise that their risk in any one year of developing breast cancer is two in 1,000 and that even that degree of reassurance has to be tempered by an interval cancer rate?
The shift towards informed decision-making is as much values based as evidence based, and driven by those with experience including patients with iatrogenic morbidity. The aim is not 'pursuit of good uptake' but to see a development of flexible decision aids to meet individuals' desires for balanced information and need for support.
In a recent angry exchange about screening for cervical cancer in the BMJ16 we were asked to look upon screening as 'insurance to avoid catastrophic consequences of an unlikely event'.
At least I know what my premiums for my life insurance are and I know the pay-off is guaranteed.
Women are not informed of
their premiums in the 'cost' of screening and the pay-off is a lottery with the odds about one in five for the unlikely event of deaths from breast cancer, which are about one in
26 if they are lucky enough to
avoid other causes of death up to the age of 8517.
1 Tabar L et al. Mammography service screening and mortality in breast cancer patients: 20-year follow-up before and after introduction of screening. Lancet 2003; 361:1405-10
2 In brief: Screening 'halves breast cancer death rate'. The Independent April 25, 2003; p. 10
3 Jatoi I, Miller AB. Why is breast cancer mortality falling? Lancet Oncology 2003;
4 Raffle AE et al. Outcomes of screening to prevent cancer: analysis of cumulative
incidence of cervical abnormality and
modelling of cases and deaths prevented.
BMJ 2003; 326:901-4
5 Blanks RG et al. Effect of NHS breast screening programme on mortality from breast cancer in England and Wales, 1996/8: comparison of observed with predicted mortality.
BMJ 2000; 321:665-9
6 Davey C et al. Insurance repercussions of mammographic screening: what do women think? Medical Science Monitor 2003;
7 Rakovitch E et al. A comparison of risk perception and psychological morbidity in
women with ductal carcinoma in situ and early breast cancer. Breast Cancer Research and Treatment 2003; 77:285-93
8 Slaytor E, Ward JE. How risks of breast cancer and benefits of screening are communicated to women: analysis of 58 pamphlets. BMJ 1998; 317:263-4
9 Edwards A et al. Presenting risk information: a review of the effects of 'framing' and other manipulations on patient outcomes. Journal of Health Communication 2001; 6:61-82
10 Wragg E et al. Information presentation and decision to enter a clinical trial: a hypothetical trial of hormone replacement therapy.
Soc Sci Med 2000; 51:453-62
11 Edwards A et al. Personalised risk communication in health screening programmes [review]. Cochrane Library 2003 (Issue 1).
Oxford: Update Software
12 Health Promotion England/NHS Cancer Screening Programmes. Breast screening.
The facts. London: HPE, 2001
13 Humphrey LL et al. Breast cancer screening: a summary of evidence for the US Preventive Services Task Force. Ann Intern Med 2002; 137:347-60
14 Anderson TJ et al. Comparative pathology of breast cancer in a randomised trial of screening.
Br J Cancer 1991; 64:108-13
15 NHS Breast Screening Programme/British Association of Surgical Oncology Breast Group. An audit of screen detected breast cancers for the year of screening April 1999 to March 2000.
NHS cancer screening programmes, May 2001
16 Sasieni PD. Think of screening as insurance.
17 Bunker JP et al. Putting the risk of breast cancer in perspective. BMJ 1998; 317:1307-9