You will see “sensitivity” and “specificity” used a lot. They sound technical. They are not.
Picture 100 people with symptoms who do have bowel cancer. At the threshold of 10, the test picks up about 89 of them. It misses around 11.
Now picture 100 people who do not have bowel cancer. At the same threshold, about 80 get a normal result. Around 20 are flagged and turn out to be fine.
These figures come from a meta-analysis of fifteen studies covering 28,832 patients with symptoms. At 10 µg Hb/g, sensitivity was 88.7% and specificity 80.5%.
Move the line and both change, always in opposite directions. Drop it to 2 and sensitivity rises to 96.8%, so you miss far fewer cancers, but specificity falls to 65.6%, so many more people are investigated for nothing. Raise it to 100 or 150 and sensitivity falls to 68.1% and 66.3%, while specificity climbs above 93%.
That trade-off is the whole argument about thresholds. There is no setting that is good at everything.
One caveat worth being clear about: these figures come from people who had symptoms. In people who feel completely well, bowel cancer is much less common, so a raised result is less likely to turn out to be cancer. That is why we include a consultation with every result rather than leaving you to interpret a number alone.
Read the full article on what a normal result rules out ›NHS bowel screening is one of the most effective things the health service does, and if you are invited you should take it. Here is honestly how the two differ.
| NHS screening | This test | |
|---|---|---|
| Threshold reported against | 120 µg Hb/g in England, 80 in Scotland and Wales | 10 µg Hb/g |
| Who can have it | Ages 50 to 74 by invitation | Any adult, no referral |
| How often | Every two years | Whenever you choose |
| What you are told | Normal, unclear or abnormal | Your exact figure |
| Who explains it | A letter | A GP, by phone, included |
| Cost | Free | £75 one-off |
One point worth knowing: NICE guidance says FIT should be offered to someone with symptoms even if they have already had a normal result through NHS screening. A screening letter answers a different question.
Read the full article on what a normal screening result means ›No test is perfect and we would rather you heard this from us.
It detects blood. Polyps, inflammatory bowel disease and other conditions raise it too. Most raised results turn out not to be cancer.
On haemorrhoids and raised results ›Around 1 in 10 people with bowel cancer have a result below the threshold. If you have symptoms, a normal result is a reason to keep looking, not to stop.
On testing more than once ›Blood from the stomach or gullet is broken down before it reaches a sample. This test says nothing about those.
On anaemia and FIT testing ›If you are in the second list, that does not mean the test is wrong for you. It means it should be part of a proper assessment rather than something you do alone. You can book a consultation and we will arrange both together.
Each article is written by a GP, sourced, and lists its references so you can check them yourself.
Why a screening letter answers a different question from the one you may be asking.
Cancer detection rates range from 0.1% to 20.7% depending on where your figure falls.
Around 1 in 10 people with bowel cancer have a result below the threshold.
The evidence says haemorrhoids are an infrequent cause of a raised result.
A normal result does not close the question if symptoms carry on.
Polyps, inflammation and other causes, and where the test is weaker.
The number is the same for everyone, and that is deliberate.
One protein, and the two consequences that shape everything about the test.
The screening committee says the optimal figure is 20. Here is why it is not that yet.
Collecting twice and repeating later are different things with different evidence.
Plenty of people order a test and never send it back. How to get past that.
Some need a test. A few need examining instead. How to tell which is which.
The old advice was no. The evidence turned out to say the opposite.
It performs well here, but it cannot complete the picture on its own.
The test measures haemoglobin in a stool sample and reports a number: micrograms of haemoglobin per gram of faeces. That number is a fact about your bowel. Everything else is a decision about where to draw a line through it.
The NHS bowel cancer screening programme draws that line high. In England it is being lowered from 120 to 80 µg Hb/g, a rollout that began in February 2026 and completes by March 2028. Scotland and Wales are already at 80. Northern Ireland remains at 120.
When somebody has symptoms, NICE draws the line at 10 in England, Wales and Northern Ireland. Scotland moved to 20 in its 2025 referral guidance. Same kit, same analysers, same laboratory. The line moves by roughly an order of magnitude.
The gap is not a disagreement about the test. The UK National Screening Committee has concluded that the optimal screening threshold would be 20. Screening sits well above that because a national programme has to fit within the number of colonoscopies the NHS can actually perform. That is a defensible decision about a system. It is a poor description of what your individual sample can tell you.
A meta-analysis of 28,832 symptomatic patients found that at a threshold of 10 µg Hb/g, the test detected 88.7% of colorectal cancers. Read the same samples at 150 and it detected 66.3%. Nothing about the samples changed.
So we do two things differently. We report against 10, and we give you the actual number rather than a pass or a fail. Detection rates vary enormously across the range of figures that all get labelled positive, so knowing whether you are at 12 or at 400 is genuinely useful rather than a technical detail.
Lowering a threshold means flagging more people who turn out to have nothing serious. That is the honest trade, and it is the reason a GP telephone consultation comes with every result whatever the number is.
A raised result arriving by email at nine at night, with nobody to ask, is a bad way to find something out. If your figure is raised we contact you and offer the earliest slot we have rather than waiting for you to book one.
Your exact number, a written GP report, and a consultation included with every result.