This site is intended for health professionals only

GPs to refer more patients with low-risk cancer symptoms under new NICE guidance

GPs are being urged to refer and test thousands more patients for potential cancers under new guidelines rubber-stamped by NICE experts today.

The guidelines take a radical new approach, with the aim of giving GPs greater freedom to investigate patients with potential signs and symptoms of cancer and decrease the time before diagnosis.

The guidelines recommend GPs are able to request some ‘direct access’ tests urgently – within two weeks or even 48 hours for some rare cancers – which NICE advisors said would streamline the patient’s journey.

But GP critics said the guidelines risked referring too many patients at very low risk of cancer and missing those at the greatest risk because the new symptoms-based approach did not take into account patients’ baseline risk.

The guidelines are expected to lead to an increase in urgent referrals for potential lung cancer, upper gastrointestinal (GI) cancers, such as stomach cancer, and bowel cancer in particular – although NICE has not forecast exactly how many more patients would expect to go through these pathways.

Despite this, NICE said any increased costs would likely be evened out by savings resulting from diagnosing people earlier, so that their treatment is less complicated and more likely to be life-saving.

The main recommendations are largely unchanged from the draft guidelines published last November, and effectively lower the risk threshold at which patients are referred for cancer investigation – from a positive predictive value (PPV) of roughly 5% to 3% – as well as getting GPs to order tests for people with clusters of symptoms suggestive of potential cancer but with a lower PPV than this.

Professor Willie Hamilton, clinical lead for the guidelines development group and a GP in Exeter, told Pulse the guidelines should give GPs more flexibility to investigate when a patient’s presenting symptoms are only vaguely suggestive of cancer.

Professor Hamilton said: ‘GPs know the symptoms of cancer [but] what has been difficult up to now has been being able to act on that knowledge. There were patients with mild symptoms that just might have been cancer that it was more difficult to investigate.

‘So now that we have slightly increased both the types of tests and number of tests that are undertaken it is now easier for GPs to test.’

Professor Hamilton insisted the symptoms-based approach was not prescriptive and that he did not expect GPs to have to do much more work as a result of the recommendations to do more testing in primary care.

He said: ‘I don’t think it will be remotely unmanageable because some testing was already been going on, we’ve been doing blood tests, CA125 tests. We will have some responsibility for seeing the faecal occult blood test results, and making sure we act on the results. Similarly for gastroscopies.’

However, Professor Julia Hippisley-Cox, professor of epidemiology and general practice at the University of Nottingham, said it was ‘disappointing’ that the guidelines panel had not addressed concerns over the complexity of recommendations.

Professor Hippisley-Cox, who has been working on producing the validated QCancer risk tool that has already been adopted on GP computer systems, said: ‘They haven’t addressed the large number of concerns about whether such long and complicated guidance could ever be implemented by GPs in everyday clinical practice.’

The guidelines development group also faced criticism for not recommending GPs consider risk factors alongside symptoms in weighing up the patient’s risk of cancer.

Dr Nick Summerton, former NICE advisor and a GP in East Yorkshire, said it was ‘bonkers’ to ignore risk factors.

Dr Summerton said: ‘To ignore risk factors is – to be kind to them – bonkers. When any person arrives in the surgery they have a baseline risk of cancer – prior probability –  according to features such as age, gender, ethnicity, family history, lifestyle – for example, smoking – and co-morbidities.

‘We then collect other information on, perhaps, symptoms to enhance this probability. Thus to ignore baseline risk will mean increased referral of those at low risk and reduced referral of those at high risk – given the same symptoms. This is frankly stupid.’

However, Professor Hamilton said the addition of risk factors such as family history or smoking should not change the decision on whether a patient was investigated as the symptoms alone should be enough.

NICE – Suspected cancer: recognition and referral


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.