POS-2026-3015 : OptiFIT: Optimising faecal immunochemical testing in early diagnosis of colorectal cancer
Bowel cancer, or colorectal cancer, is the third most common cancer worldwide. Early diagnosis makes treatment easier. Many patients get diagnosed due to symptoms like bleeding from the back passage, changes in bowel habits, weight loss, or tummy swelling. These symptoms can be misleading, as they may come from various causes. NICE advises GPs to use Faecal Immunochemical Testing (FIT) for referring patients with suspected bowel cancer. This test detects tiny amounts of blood in stool, a potential sign of bowel cancer. FIT works well, but many low-risk individuals are being referred unnecessarily, raising concerns. This lowers testing capacity for high-risk patients and subjects low-risk patients to invasive procedures.
To support doctors, we developed a mathematical model called Colofit. It uses FIT results, age, sex, and blood tests to predict bowel cancer risk. Colofit finds high-risk individuals, speeds up urgent referrals, and reduces costs compared to using FIT alone. We have found more ways to enhance its effectiveness. Our project will ensure it operates well nationwide by doing two things:
1. Test how the Colofit model works across the UK
2. Create a new model, OptiFIT, using data from underserved communities
We will assess the Colofit model using data collected from across England. This will highlight areas for improvement. Next, we’ll develop OptiFIT. We’ll include factors like blood-thinning medications that can raise bleeding risks.
This research will help us prioritise people for urgent tests. It will speed up diagnoses. It will use fewer NHS resources. We will give GPs real-time risk information on colorectal cancer. We will integrate this into their systems to improve patient care. We will share our findings openly. We’ll work with NHS groups, patients, and the public to make sure everyone hears our results.
- Study lead: Joe West
- Organisation: University of Nottingham
- Project type: Research
- Start date: 30 September 2026
- View project progress, open code and outputs