Incorporating faecal haemoglobin measurement using the faecal immunochemical test (FIT) in the referral, triage and prioritisation pathway for patients with colorectal symptoms
James Falvey, Chris Frampton, Richard Gearry, Ben Hudson, Lucinda Whiteley
Patients with bowel cancer (colorectal cancer (CRC)) often have few or no symptoms until the cancer is advanced. Meanwhile, although bowel symptoms are a common problem, most patients with bowel symptoms do not have a serious underlying cause such as CRC. Consequently, symptoms on their own are not a good decider of who should undergo bowel investigation. The current access criteria for bowel investigation in New Zealand are largely based on bowel symptoms, and this symptom basis likely contributes to high demand for investigation (usually by colonoscopy), even though few of these people have a serious cause detected. Detecting blood in the bowel motion using the faecal immunochemical test (FIT) is a powerful way of determining who is at risk of bowel cancer. FIT testing is used in the NZ bowel screening program to identify individuals who are at risk of bowel cancer (detectable blood in bowel motion using FIT) even when they don’t have bowel symptoms. These cases are invited to have a screening colonoscopy and approximately 7% of them are found to have bowel cancer. This paper summarises the data that has been published internationally about the accuracy of FIT for CRC in patients with bowel symptoms (as opposed to those without bowel symptoms as per bowel screening). The statistical process used to summarise the data is called meta-analysis. We then used this summary data, in conjunction with recent referral and colorectal cancer diagnosis data from Canterbury (previously published in NZMJ) to develop new access criteria for bowel investigation for patients with bowel symptoms. The new access criteria are presented in the paper. We estimate that when compared with the current criteria, the proposed criteria would increase the number of cancer cases detected, while simultaneously reducing the number of invasive procedures (usually colonoscopy) performed. Using the previously published dataset on which our modelling is based, we estimate that the proposed criteria, when compared with the current criteria, would increase the proportion of colorectal cancer cases detected (from among those referred) from 90.5% to 97.1%, while reducing the number of colonoscopies needed to detect these cancers by a factor of almost a half. This is important as colonoscopy is a finite resource in New Zealand, costly, invasive, and should be used as carefully as possible to achieve the greatest benefit for the population. Reducing demand for symptomatic colonoscopy would shorten waiting lists and waiting times, facilitate faster diagnosis for cases with serious disease, and free capacity for other necessary procedures (including screening or surveillance scopes, interventional cases etc.). We hope that our paper will promote discussion regarding the use of FIT in symptomatic cases, and hasten the introduction of FIT in the diagnostic process for New Zealanders with bowel symptoms.
