Posted by: Margaret McCartney7 September 2012
Over-diagnosis is fact. A paper from Norway, published in the Annals of Internal Medicine this month1, has pointed this out clearly: if you perform screening mammography in 2,500 women over 10 years, you will prevent one death from breast cancer – you will also find six to 10 women with breast cancer who will be treated for it, but who would not have been harmed or maimed by their ‘cancer'.
As our scans and tests have become more sensitive, we have not accompanied this with an understanding of which are the ‘abnormalities' that are clinically significant. If that's a difficult thing for GPs to get our heads around, it's just as hard for patients. Doctors have long known the skill in watchful waiting – on knowing when to intervene with, say, minor LFT abnormalities. The more scans and X-rays we do, the more potential problems are discovered.
And here we come to the ‘popularity paradox'. If we find abnormalities that were never going to maim or harm, patients may be pleased to find themselves cured after the ‘treatment' has been administered. Yet all this patient stands to gain are the side-effects of medical interventions, not benefits.
The worse a screening test is, the more over-diagnosis is created, and the more people who believe that their life has been saved because of it. But when women get the facts about breast screening, many realise their treatment may not have been necessary.
Read the BMJ and you will find the testaments from rightfully angry women who have endured unpleasant surgery and radiotherapy without knowing, from the start, that the need for their treatment was uncertain.
A duty to explain
But it's hard for people to understand the complexities, partly because of our societal understanding of cancer. Early diagnosis is seen to equal better diagnosis.
We need to better distinguish the prompt attention to symptoms from the screening of a person without any complaints. Breast screening has been shown to find abnormalities, such as ductal carcinoma in situ, that are hardly ever seen outside of the screening programme.
Just because the risks and benefits of screening are hard to explain doesn't mean we shouldn't try. I'd argue that we give patients better information about the potential hazards when we proceed to even minor surgery than we ever do about breast screening, even though the harms from screening may be much bigger.
There is currently a review of breast screening under way in the UK, led by Professor Mike Richards and Harpal Kumar, chief executive of Cancer Research UK. However, they have previously said ‘we want targeted action to be taken to improve the numbers of people attending screening,' and that the Government should be ‘finding ways to encourage [people] to participate'.
Is this really the way forward? I don't think so. The problems of over-diagnosis occur in many types of screening, not just breast screening. Of course, the private screening sector is also culpable – for all the CT scans, MRI or ultrasound done on well people, and the abnormalities of uncertain significance that will be found as a result. It is up to the Government to lead in ensuring NHS screening is cost-effective. It simply isn't possible to go on as we are, and the sooner we give ethically sound information about screening and make clear it's a choice, not a stipulation, the better.
Dr Margaret McCartney is a GP in Glasgow
Kalager M, Adami H, Bretthauer M et al. Over-diagnosis of invasive breast cancer due to mammography screening: results from the Norwegian screening program. Ann Intern Med 2012;156:491-99