Endoscopies for all?
British plans for a national bowel cancer screening programme could do as much harm as good.
Health secretary John Reid recently announced plans for a national bowel cancer screening programme in the UK. ‘Screening is key to cutting deaths from cancer’, he told us cheerfully on 27 October (1). This seems like common sense: if you catch cancer early, it will be easier to treat, so fewer people will die. However, this is not quite the whole story.
In a couple of years, people in their sixties will be asked to send off a sample of their stools to be tested for traces of faecal occult blood (FOB), which is suspicious, but far from diagnostic, of bowel cancer. Patients who are ‘FOB positive’, probably about two per cent overall, will be asked to undergo a colonoscopy, a fibreoptic examination of the large bowel. This is a far from enjoyable day out at the hospital.
Patients are asked to stick to a special low-fibre diet the day before the procedure, then drink a potent laxative solution to clear the bowel. They are usually given strong sedation as the flexible telescope is passed around the bowel, which can be extremely uncomfortable. And colonoscopy is not risk-free: just under 0.5 per cent of patients will suffer from bowel perforation or haemorrhage, both potentially fatal complications. Probably about 10 per cent of those undergoing colonoscopy will turn out to have invasive bowel cancer.
The value of FOB testing is controversial, but is generally accepted to reduce mortality from colon cancer by about 15 per cent in populations tested every two years (2). So far, so good. But you can look at the figures another way. Taking a population of 10,000 screened individuals, offered biennial FOB testing for 10 years; if two thirds of the population undergo at least one FOB test in the 10-year period, 8.5 cancer deaths will be prevented. However, this would also result in 2800 participants undergoing colonoscopy, with 3.4 potentially fatal colonoscopy complications.
And there is another element of doubt. As H Gilbert Welch points out in his excellent new book Should I Be Tested For Cancer? (3), people who have had major surgery for bowel cancer have a higher chance of dying from a condition related to the surgery, such as post-operative pneumonia or bowel obstruction. These deaths will not be counted as cancer deaths, since they were due to treatment for the cancer rather than the cancer itself. So even our 8.5 lives saved are being whittled away.
In the UK, mortality rates from colorectal cancer have fallen by about 20 per cent in the past 10 years without screening, due to improved detection and treatment (4). But in the USA, where screening is practised widely (40 per cent of men and women above 50 have undergone testing for bowel cancer) overall mortality rates are not much better than in the UK. However, five-year survival rates for bowel cancer are considerably higher in the USA than in the UK.
This illustrates another problem with screening, known as lead-time bias. Cancers may be picked up earlier in their natural course, thus gaining improved five-year survival rates without necessarily an improvement in the cure rate. In other words, patients are living longer with the fear and anxiety that comes with knowing that they have cancer, but still dying at the same time they would have anyway. (For a good description of this phenomenon in relation to breast cancer screening see Professor Michael Baum’s article on spiked, What mammography misses.)
A large-scale, UK government-funded trial published preliminary results earlier this year, showing that screening would be feasible here (5). The trial studied two areas in the Midlands and two in Scotland. About 500,000 people were invited to take part, by sending in three stool specimens; over 50 per cent did so. This study is so far too short-term to conclude whether this will in fact cause a decrease in mortality rates from cancer. The authors claim that other studies have shown screening will decrease mortality rates, and their study proves screening is possible in the UK.
But there are major concerns about the introduction of a screening programme. Most obvious is the fact that endoscopy services in this country are already tremendously overstretched. Patients in the hospital where I work can wait months even for urgent colonoscopies, which in turn delays their treatment. John Reid has promised a massive expansion of colonoscopy services, but one suspects this will be swallowed up by the enormous demands of a nationwide screening programme. Patients who waited until they had symptoms of cancer may be pushed to the back of the queue.
It is odd that this government is so keen to promote strategies for finding more cancers, when we struggle to treat the patients we already have. The NHS is often caricatured as a ‘national sickness service’, concentrating on illness rather than encouraging people to be healthy. Unfortunately, despite the best efforts of Reid and his pedometer, everyone does get ill in the end. Perhaps we are better to not waste our days worrying about this inevitability – but asking our health service to deal with it when it happens.
Dr Liz Frayn is a surgical trainee in southwest England.
(1) DoH press release, Weds 27 Oct 2004
(2) ‘Screening for colorectal cancer using the faecal occult blood test’, Hemoccult (Cochrane review); Towler, BP et al; The Cochrane Library, Issue 4, 2004, UK
(3) Should I be tested for cancer? H Gilbert Welch, University of California Press 2004
(4) CancerStats- Large Bowel Cancer, Cancer Research UK, 2003
(5) ‘Results of the first round of a demonstration pilot of screening for colorectal cancer in the UK’, UK Colorectal Cancer Screening Pilot Group, BMJ 2004;329:133
To enquire about republishing spiked’s content, a right to reply or to request a correction, please contact the managing editor, Viv Regan.