The data – being presented at the National Cancer Research Institute (NCRI) Cancer Conference in Liverpool this week – was collected for breast, bowel, lung, pancreatic, oesophageal and stomach cancers for 2001-2 and for 2007-8.
Information was collected from the General Practice Research Database on more than 14,400 patients aged 40 or over who had been diagnosed with any of the six different cancers and who had previously shown potential cancer symptoms. These symptoms were predetermined by an expert group. An example of symptoms for bowel cancer included things such as constipation, diarrhoea or rectal bleeding.*
In 2001-2 bowel (colon and rectal) cancer was on average diagnosed 96 days after patients first reported a symptom to a GP to when they were diagnosed. This dropped significantly to 75 days in 2007-8.
For oesophageal cancer in 2001-2 the average time to diagnosis after first GP visit was 59 days. This fell to 48 days in 2007-8. For pancreatic cancer in 2001-2 the average time to diagnosis after first GP visit was 63 days. This fell to 52 days in 2007-8.
Breast, stomach and lung cancers did show a drop, though not a significant one.
Breast cancer times to diagnosis fell from 27 days in 2001-2 to 25 days in 2007-8. Stomach cancer diagnosis times reduced from 88 days in 2001-2 to 77 days in 2007-8, and for lung cancer from 106 days in 2001-2 to 102 days in 2007-8.
Midway between these two time periods was the release of the 2005 NICE referral guidance for suspected cancer cases. These guidelines provide GPs with details of symptoms that should prompt them to send a patient for further tests.
Dr Richard Neal, lead researcher based at the North Wales Centre for Primary Care Research, Bangor University, said: “We found that diagnostic intervals can and do change over time. The reduction between 2001-2 and 2007-8 may in part be due to the roll out of the 2005 NICE referral guidance for suspected cancer. This gives clear guidance on which symptoms should prompt a doctor to refer a patient for further investigation. But there is considerable variation between cancers, with diagnostic intervals highest in those cancers which are more difficult to diagnose.
“Diagnostic intervals were longer for patients with harder to diagnose cancers and for those presenting with symptoms that did not qualify for an urgent referral. But diagnostic intervals remain long in most cancers, with considerable potential for further reduction. In particular, the diagnostic intervals for the 10 per cent of patients who are diagnosed most slowly remain very long for most cancers. And we do not fully know the effect of the reduction of diagnostic intervals on improvements in stage at diagnosis and long term survival.”
Sara Hiom, Cancer Research UK’s director of information, said: “It’s very encouraging to see that patients are, on average, being diagnosed more quickly for some cancers, offering a better chance of successful outcome. It’s clearly vital for GPs to have access to good quality information to make the best decisions for their patients. There is still room for considerable improvement though, and reducing the time to diagnose and treat is a critical part of improving outcomes for people with cancer.”
For more information contact the press office on 020 3469 8300 or, out of hours, on 07050 264 059.
Notes to editors
As part of the National Awareness and Early Diagnosis Initiative (NAEDI) and funded by the Department of Health (England) the researchers compared diagnostic intervals for different cancers before and after the introduction of the 2005 referral guidelines for suspected cancer produced by the National Institute for Clinical Excellence (NICE). These guidelines provide GPs with details of symptoms that should prompt them to send a patient for further tests.
The role of NAEDI is to coordinate and provide support to activities and research that promote the earlier diagnosis of cancer. These results will provide partners in the initiative a benchmark of current practice.