The research, published in the British Journal of Cancer, shows that the bowel cancer calculator was better at predicting whether patients had bowel cancer – compared with relying on individual bowel cancer symptoms.
The calculator, available at www.qcancer.org, could help GPs identify people who are at increased risk of having undetected colorectal cancer so that they could be sent for referral or further tests.
The study is the first to independently evaluate a new model developed by researchers at the University of Nottingham and being considered for a pilot by the Department of Health as part of its plans to improve early diagnosis.
The QCancer calculator uses seven risk factors for bowel cancer in women, and nine in men, to predict their risk of having the disease, including age, presence of symptoms and lifestyle behaviours.
This study looked at over 2.1 million patients between 2000 and 2008 from which more than 3,700 cases of bowel cancer were identified.
Researchers found that using the QCancer calculator nearly three quarters of bowel cancers were identified in both men and in women.
But using the symptom of rectal bleeding – only 34 per cent of women and 40 per cent of men were diagnosed with bowel cancer. Similarly, using the single symptom of abdominal pain around one third of bowel cancers were picked up1.
Dr Gary Collins, study author and senior medical statistician at the University of Oxford, said: “Our research shows that the QCancer calculator clearly outperforms individual signs and symptoms as a way of predicting whether a patient has bowel cancer.
“It’s great to see that the government is considering this model along with others to help spot people with cancers that might otherwise go undetected. Ultimately tests like this will be likely to improve early diagnosis of the disease so that patients have the best possible chance of survival.”
In April, the national cancer director for England, Professor Mike Richards said he was in discussion with the researchers behind QCancer to trial the use of the risk calculator in GP practices.
The web-based calculator could be integrated into doctors’ computer systems to create a list of high-risk patients who could be called back for further tests.
Sara Hiom, director of cancer information at Cancer Research UK, said: “When bowel cancer is found at the earliest stage, there is an excellent chance of survival, with more than 90 per cent of people surviving the disease for at least five years.
“This study highlights a new approach to helping GPs make better decisions about which patients to refer for further tests. Ultimately we hope risk calculators like these could contribute to swifter diagnosis of bowel cancers. Encouraging more people to take up their invitation to bowel screening is also important.”
For media enquiries please contact the Cancer Research UK press office on 020 3469 8300 or, out-of-hours, the duty press officer on 07050 264 059.
- Collins, G S et al., Identifying patients with undetected colorectal cancer: an independent validation of QCancer (colorectal) British Journal of Cancer, (2012).
Notes to editors
1. The study used eligible patients on the THIN database between 1st June 2000 and 30th June 2008 who were followed up for a median of two years. The percentages given of new bowel cancers diagnosed were for the 10 per cent of women and men with the highest risk of the disease.
QCancer identified 71 per cent of bowel cancers in women and 74 per cent in men. Abdominal pain identified only 36 per cent of bowel cancers in women and 31 per cent in men.
QCancer calculator was developed by Professor Julia Hippisley-Cox, a clinical epidemiologist and GP and Dr Carol Coupland from the University of Nottingham in collaboration with medical software company ClinRisk Ltd. The team have also created the cardiovascular risk calculator QRISK.