A major nationwide study has revealed that the majority of bowel cancers diagnosed late, after a false-negative colonoscopy could have been prevented or detected earlier. The study is published in Endoscopy.
The research is the largest and most detailed review of post-colonoscopy colorectal cancers (PCCRCs) to date. It analysed 1,724 cases across 126 NHS hospitals in England and found that nearly 70% of these cancers were avoidable. Almost half (44%) of patients experienced harm as a result of delayed diagnosis, including premature death in 8% of cases.
Over 35,000 colorectal cancers are diagnosed in England each year, with the vast majority of these diagnosed through colonoscopies delivered within the NHS. When cancer is diagnosed at an early stage, treatment is more effective and long-term survival is substantially improved so it is a priority to increase rates of early diagnosis in the NHS.
Colonoscopy is a vital test used both to diagnose bowel cancer and to prevent it by removing pre-cancerous growths. However, some cancers still occur after a previous colonoscopy. These PCCRCs represent a missed opportunity to catch disease earlier or prevent it altogether.
Until now, reviews of such cases in England have been inconsistent and incomplete. This new audit, funded by Bowel Cancer UK, Cancer Research UK, Health Data Research UK and the National Institute for Health and Care Research (NIHR) Oxford Biomedical Research Centre, created the first national system for identifying PCCRCs and understanding why they arise.
By linking cancer registry and hospital records, the study was able to capture cases even when diagnosis and original colonoscopy took place at different hospitals. Local clinical teams then reviewed each case using a standardised template based on international guidelines.
Key findings:
- Most cases of PCCRCs involved missed lesions. Two-thirds of cases were categorised as possible missed cancers, even when the quality of the original colonoscopy had been judged as adequate;
- PCCRCs were disproportionately found in harder-to-examine parts of the bowel, such as the transverse colon and the hepatic and splenic flexures;
- About one in four cases also involved patient, clinical, or administrative issues. This included patients declining further tests, delays in booking repeat procedures, or inappropriate follow-up decisions;
- More than four in ten patients experienced moderate to severe harm, including premature death.
The study highlighted several areas where changes could reduce the risk of PCCRCs:
- Better bowel preparation and repeat testing where initial procedures were incomplete;
- Enhanced training and referral pathways for complex polyp removal.
- Improved photo documentation, with audits of image quality as a routine part of service delivery;
- Closer monitoring of high-risk patients, including those with inflammatory bowel disease, Lynch syndrome, or a history of bowel cancer or polyps.
Dr Nick Burr, Consultant Gastroenterologist at Mid Yorkshire Teaching NHS Trust and first author of the study, said ‘This is the largest study of its kind, and it shows clearly that most of these cancers could have been avoided. The audit gives us the evidence we need to improve services, prevent harm, and ultimately save lives.’
Around 1,200 PCCRCs are estimated to occur each year in England. Acting on these findings could enable earlier diagnosis, or prevent over 800 cases annually, saving lives and reducing the burden on NHS cancer services.
Rachel James, who has Lynch syndrome, a genetic condition that can increase the risk of bowel cancer, and was later diagnosed with cancer following a false-negative colonoscopy, said ‘On the one hand I feel incredibly lucky, knowing that I have Lynch syndrome and the fact that I was on the bowel cancer screening program did mean that following successful surgery, and amazing support from the endoscopy team at North Tyneside General Hospital, I was cancer free. But the impact this has had on us all as a family, the waiting on test results, the recovery, and my own quality of life post-surgery as well as the cost to the NHS has been significant. It is frustrating that this could have avoided and it worries me that I’m back on 2 yearly screening colonoscopies when clearly this wasn’t enough, if I get bowel cancer again the outcome will certainly be worse. I have heard of so many people in similar situations that haven’t had as positive an outcome as I have had and while we say in the Lynch community that knowledge is power, there is little point knowing you have Lynch if the screening programs aren’t fit for purpose.’
Eva Morris, Professor of Health Data Epidemiology at Oxford Population Health and senior author of the study, said ‘The NHS is uniquely positioned to find and tackle these missed cancers because it generates incredibly rich datasets. Their linkage and use enables this study’s audit system to simply and efficiently identify all PCCRCs and by learning from each and every case, set new standards, strengthen training, make colonoscopy safer and, most importantly, diagnose cancer earlier, or even prevent it, for hundreds of people each year. This simple, data driven approach, could not only save valuable resource but also many, many lives. Unfortunately, whilst we’ve shown this approach is successful in improving diagnostic services it is not yet routine clinical practice in the NHS. This is a huge missed opportunity to help, cost effectively, improve cancer outcomes. We can, and should, be doing more with NHS data to save lives.’
Genevieve Edwards, Chief Executive at Bowel Cancer UK, said ‘Most people can feel confident that their colonoscopy is a safe and effective test, but this important study shows there’s still room for improvement. For the small number of people affected, missing a cancer diagnosis can have devastating consequences. This new auditing tool has the potential to be a game changer for the NHS as it gives specialists a practical way to spot mistakes and make colonoscopies even safer. We urge the NHS to adopt it nationwide because when bowel cancer is diagnosed early, it is treatable and curable.’