Establishing the diagnostic accuracy of haemoglobin in faecal immunochemical tests (FITs) and establishing a receiver operating characteristic (ROC) curve for FITs in the English primary care symptomatic population: A cohort study in the South West of England

Who is involved?

Project Summary

Colorectal cancer is the fourth most common cancer in the UK, and the second leading cause of cancer-related deaths. Diagnosing colorectal cancer is difficult, as the symptoms are the same as many non-cancerous conditions.

The NICE guideline NG12 (2015) recommends that patients consulting their GP with ‘alarm’ symptoms of colorectal cancer are urgently referred for colonoscopy. However, not all patients with colorectal cancer have these alarm symptoms. Many have vague symptoms that do not warrant a colonoscopy under NG12. In 2017, a new NICE guidance DG30 suggested that faecal immunochemical tests (FITs) are used for patients with vague symptoms that could suggest colorectal cancer, but do not represent a great enough risk for an urgent referral. FITs measure the amount of haemoglobin (Hb) in a stool sample. A higher level of Hb in a stool sample may suggest bleeding in the bowel caused by cancer. We don’t know how high Hb in the stool should be before the patient is offered a colonoscopy, when the patient has these vague symptoms.

In this study, our primary aims are 1) to determine the optimum cut off point for Hb in FITs in a symptomatic primary care population, and 2) to estimate the diagnostic performance of FITs at detecting cancer in a symptomatic primary care population.

In the South West, FITs have been in use since June 2018. We will collect data on all FITs performed in the region during the 18-month study period. This will include the amount of Hb present in the patients’ samples, whether or not they were referred for colonoscopy, patient demographic data, the type of FIT used, and whether or not the patient was diagnosed with colorectal cancer within one year of their FIT. We will also collect data on the number and type of referrals and diagnoses in the region during the study period, and the number of FITs ordered from primary care during that time. We estimate that around 30,000 FITs will be performed during the data collection period.

This study will be complimented by a narrative review providing an overview of FIT use across the globe in primary care symptomatic patients and an East of England based mixed-methods cohort study on the diagnostic accuracy of FIT in symptomatic primary care patients.

Back to Projects