This site is intended for health professionals only

Is the early cancer diagnosis drive harming patients?


There’s no denying that we could do better when it comes to earlier cancer diagnosis. But the current drive being developed by our managerial and specialist colleagues will fail to improve patient outcomes and – moreover – lead to physical and psychological harm to a large number of individuals. What’s more, the extra pressure on practices will further damage patient care.

Colleagues report being inundated with requests for appointments from the worried well as a result of the Government’s Be Clear on Cancer campaign. For instance, more and more patients under 30 are presenting with rectal bleeding, reducing the availability of appointments and potentially putting off less pushy older patients in genuine need from coming to consult.

We must be wary of the harm overinvestigation can cause. No test is without risk – for example, conducting a colonoscopy to investigate rectal bleeding can result in a ruptured bowel wall. Moreover, a study in Australia found CT scans could be responsible for 40 fatal instances of cancer there every year.¹

The Ulysses effect should also be considered – one ill-advised investigation generally leads to another, which again increases risk to patients.

This, of course, couples with the psychological effects of overinvestigation. Numerous anecdotes from colleagues suggest, not surprisingly, that telling patients they might have cancer increases anxiety for their families, and can lead to their becoming depressed and taking time off work.

Diagnosis plans must incorporate the ideas and experience of GPs. If politicians, specialists and managers are so interested in earlier diagnosis by GPs, why do only two members of NHS England’s new cancer taskforce have any experience of general practice?

NICE should develop a GP advisory panel on cancer diagnosis, which could address the fact that current guidelines don’t allow us to weigh up the benefits and drawbacks of tests.

At the moment, some tests actually slow down diagnosis – the unreliability of chest X-rays has been widely discussed and ultrasounds don’t help diagnose pancreatic cancer unless the patient is extremely thin.

Don’t get me wrong: we should be pushing to improve our early diagnosis and cancer referral rates. But as long as those developing new national referral thresholds and targets ignore GP experience, we won’t get anywhere.

Dr Nick Summerton is a GP in East Yorkshire and a former NICE adviser



Dr Richard Roope is a GP in Fareham, Hampshire, and RCGP clinical lead on cancer

The burden of cancer, both to our nation’s health and the economy, is set to rise. With an ageing population and increasingly unhealthy lifestyles, it is thought that half of us will develop one or more cancers in our lifetime.

One-year survival rates (a proxy for early diagnosis) for a number of cancers have improved over the past eight years. However, with the exception of breast cancer, we are not closing the cancer survival gap on our European neighbours.

In 2011, a paper in The Lancet identified that cancer survival was persistently higher in Australia, Canada and Sweden than in England, Northern Ireland and Wales, and that the figures were consistent with later diagnosis in the UK.²

When comparing the UK system with that of Scandinavia, for example, one notable difference is the readiness to refer patients with suspected cancer. The average full-time GP could prevent 12 cancer deaths during their career if the UK improved its performance to the level of the best in Europe, and as a nation we could prevent 10,000 cancer deaths every year.

The current drive will make patients safer and reduce pressure on practices by eradicating problems in existing guidance. For example, current NICE guidelines have set the referral bar too high – they suggest referring patients whose symptoms and signs generate a positive predictive value of having cancer of 5%, which has generated a two-week referral conversion rate of 10%. The latest draft guidance suggests reducing this to a positive predictive value of 3% to generate a referral for further investigation.

NHS England’s early diagnosis programme will be targeted at high-risk, hard-to-reach groups, particularly populations with a high risk of lung cancer – probably the cancer type for which the UK has the worst outcomes, compared with the rest of the world. Statistics from Cancer Research UK show that diagnosing cancer earlier will save lives and money.

If GPs are able to investigate and make cancer referrals earlier, the outcomes for patients will be better, we will have fewer consultations, and the treatment will cost less. To me, that sounds like a winning situation all round.


1 Simpson G, Hartrick G. Use of thoracic computed tomography by general practitioners. Med J Aust 2007: 187; 43-6

2 Coleman M et al. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. The Lancet 2011; 377: 127-38