Medical Matters: How did cancer screening become a cause for concern?
For every 1,000 people screened, 49 people will be wrongly identified and subjected to unnecessary tests or treatment. Photograph: Mark Thomas/Science Photo Library
Recent concerns about bowel cancer screening at Wexford General Hospital are a reminder of our sometimes unrealistic expectations about what screening can deliver.
Screening rarely benefits the entire population and it can have negative consequences for individuals. At best, screening can identify some of the people who have a disease, but it cannot prevent disease. It can never provide a “yes” or “no” answer and being given the “all clear” at the end of the process does not mean you will not go on to develop cancer.
BreastCheck, the national breast cancer screening programme, is likely to have met considerably more scientific resistance were the programme to be proposed in 2016.
From being in many respects a “no-brainer” beneficial public health proposal, the evidence questioning the downside of breast cancer screening has been mounting.
The most serious concern is the potential for overdiagnosis or over-detection – that is, picking up cancers that never would have done harm, and that would never have come to light if the health system hadn’t gone looking for them.
This message runs counter to long-standing advice about the importance of “catching cancer early”. Indeed, a recent survey of nearly 2,000 men and women aged 50-80 across the UK revealed that almost 90 per cent thought cancer screening for healthy people was “almost always a good idea”.The reality is somewhat different: a review of breast screening in Britain estimated that for every woman who has her life saved by screening, three women will be overdiagnosed and treated for a cancer that would not have harmed them.
The Canadian National Breast Screening Study is one of the longest in existence. It had been following almost 90,000 women for 25 years when it reported in the British Medical Journal in 2014 that “annual mammography in women aged 40-59 does not reduce mortality from breast cancer beyond that of physical examination or usual care”. It also found that some 22 per cent of screening detected invasive breast cancers were over-diagnosed.
It found that women aged 50-69 years who attend mammography screening reduce their risk of dying from breast cancer by 40 per cent compared with women who are not screened.
Sense About Science is a UK charity that aims to make sense of scientific evidence in public debate. In a publication on the limitations of screening, it takes the example of a screening programme with 95 per cent accuracy looking for a condition that is present in 2 per cent of the population.
For every 1,000 people screened, 49 people will be wrongly identified and subjected to unnecessary tests or treatment (false positive results); 19 people will be properly identified which allows them avail of early treatment; and one person with the disease will be missed by screening (a false negative) and will not be diagnosed until symptoms appear.
A key issue influencing the success or otherwise of disease screening programmes is the quality of the test chosen. For breast cancer it seems earlier iterations of mammography were not sensitive enough for women with dense breast tissue, while more recent technologies seem to be overdiagnosing a particular form of disease called ductal carcinoma in situ.
And the principal reason for the failure to approve population screening in men for prostate cancer internationally is the lack of sensitivity and specificity of the blood test that measures prostate specific antigen.
Possibly the biggest conundrum associated with screening for disease is that at a population level there will likely be a benefit from quality screening programmes. However, at an individual level – as some patients in Wexford are painfully discovering – the disadvantages of screening may be all too obvious.