GI Endoscopy · 7 min read
FIT Accuracy: Good for Cancer, Weak for Advanced Adenomas, Blind to Large SSL
Annual FIT is an accepted screening program. It is not a cleared colon. The miss that matters is the nonbleeding advanced polyp.
Lee 2014 pooled FIT sensitivity for colorectal cancer at 79% with specificity 94%. Imperiale 2014, using colonoscopy as the reference, put FIT sensitivity at 73.8% for CRC, 23.8% for advanced precancerous lesions, and 5.1% for sessile serrated polyps 1 cm or larger. A negative FIT is reassuring about a bleeding cancer. It is not a statement about a pale right-colon SSL. A positive FIT is a colonoscopy, not a second FIT.
Experienced teaching points
Clinical Pearls
- Lee 2014: pooled FIT sensitivity 79% and specificity 94% for CRC in average-risk adults. Lower cutoff, higher sensitivity, more colonoscopies.
- Imperiale 2014: FIT 23.8% sensitive for advanced precancerous lesions, 5.1% for SSL 1 cm or larger. That is the counseling sentence.
- Positive FIT: colonoscopy. Do not repeat the stool test to see if it goes away.
- FIT programs are annual. A one-time negative FIT is not 10 years of coverage.
Clinical Bottom Line
| Finding | What FIT actually does |
|---|---|
| Colorectal cancer | Pooled sensitivity 79%, specificity 94% (Lee 2014, PMID 24658694). Imperiale's one-sample FIT: 73.8%. |
| Advanced adenoma / advanced precancerous lesion | About 1 in 4. Imperiale: 23.8% for advanced adenomas plus large serrated polyps. This is the gap. |
| SSL 1 cm or larger | 5.1% in Imperiale 2014. FIT is almost blind to nonbleeding serrated lesions. |
| Positive test | Colonoscopy. Not a second FIT. Not a delay until the next birthday. |
| Negative test in a FIT program | Repeat on the program's schedule, usually yearly. It does not replace surveillance colonoscopy after polyps. |
How accurate is FIT for cancer?
Lee 2014 (PMID 24658694) meta-analyzed 19 studies of FIT in asymptomatic average-risk adults. Pooled sensitivity for CRC 0.79 (95% CI 0.69-0.86), specificity 0.94 (0.92-0.95). Sensitivity was better below 20 micrograms of hemoglobin per gram of stool (0.89) than at 20-50 (0.70), with the expected specificity tradeoff. One sample performed like multiple samples in that analysis. Brand and cutoff matter more than folklore about "three cards."
Imperiale 2014 (PMID 24645800) is the colonoscopy-verified head-to-head with multitarget stool DNA in 9989 average-risk adults. One-sample FIT (OC FIT-CHEK, cutoff 100 ng/mL buffer) found 48 of 65 cancers (73.8%). That is a real cancer test. It is also the number you should quote instead of "FIT never misses cancer."
Why do advanced polyps slip through?
FIT antibodies detect human hemoglobin. A bleeding cancer is in range. A 20 mm nonbleeding adenoma is often not. Imperiale: sensitivity 23.8% for advanced precancerous lesions, 46.2% for high-grade dysplasia, 5.1% for sessile serrated polyps 1 cm or larger. The last number is the counseling problem. A patient with a negative FIT has not been cleared of a mucus-capped right-colon SSL. SSL page.
That does not make FIT a bad program. It makes FIT a cancer-detection program that has to be repeated. A one-time negative stool test offered as if it were a colonoscopy is the error.
What do I do with the result?
- Positive: colonoscopy. Document the FIT in the indication. Do not wait for a "confirmatory" stool test.
- Negative, average-risk, FIT program: again next year. USPSTF 2021 keeps annual FIT as a Grade A option for adults 45 to 75 (Davidson, PMID 34058355).
- Prior advanced adenoma, SSL, or piecemeal resection: this person is in a colonoscopy surveillance program. FIT is not a substitute. Intervals: colonoscopy quality page.
- Symptoms, iron-deficiency anemia, or a mass: this is not a screening FIT. Scope.
Multitarget stool DNA (Cologuard in Imperiale) was more sensitive for cancer (92.3%) and for advanced precancerous lesions (42.4%) and less specific. It is a different test on a different interval. Do not mix the numbers.
Selected references
- Lee JK, Liles EG, Bent S, Levin TR, Corley DA. Accuracy of fecal immunochemical tests for colorectal cancer. Ann Intern Med. 2014;160:171.
- Imperiale TF, et al. Multitarget stool DNA testing for colorectal-cancer screening. N Engl J Med. 2014;370:1287-1297.
- Davidson KW, et al. Screening for colorectal cancer. US Preventive Services Task Force. JAMA. 2021;325:1965-1977.
Last reviewed September 20, 2026. Written for clinicians counseling average-risk screening, and for the ones who need to walk a positive FIT to the endoscopy unit instead of repeating the stool test.
For your teaching file
Save this article as a PDF
Drop your email and we'll open a print-ready version you can save as a PDF — and you'll start getting our weekly GI endoscopy newsletter.