Is Cologuard Accurate? The Real Numbers on Cancer Detection and False Positives
In the trial that won Cologuard FDA approval, the test found 92.3% of colorectal cancers and 42.4% of advanced precancerous polyps in nearly 10,000 average-risk adults. That single split tells you almost everything about how to think about it.
The test is excellent at catching cancer that already exists. It is much less reliable at finding the polyps that might quietly turn into cancer over the next decade.
That distinction matters because most colorectal cancer screening is really about removing precursors before they become anything serious. So "is Cologuard accurate?" depends on which question you are actually asking the test.
How the Numbers Compare to Other Screening Options
Cologuard is a stool DNA test. It looks for genetic and methylation markers shed by tumor cells along with blood in stool, then runs the result through an algorithm. Its accuracy is judged against colonoscopy, which is treated as the reference standard because it both visualizes the colon and removes polyps in a single procedure.
| Test | Detects cancer | Detects advanced polyps | False positive rate |
|---|---|---|---|
| Cologuard (original) | ~92% | ~42% | ~10–13% |
| Next-generation Cologuard | ~94% | ~43% | ~7–9% |
| FIT (single annual test) | ~67–74% | ~23–24% | ~4–5% |
| Colonoscopy | reference standard | reference standard | not applicable |
Instalab's Cologuard kit ($599) is FDA-approved for average-risk adults 45 and older and is recommended every 3 years per ACS screening guidelines.
The picture is consistent across multiple large trials. Cologuard catches more cancer and far more advanced polyps than the simpler FIT stool blood test, but it produces more false positives. Colonoscopy still wins on both ends, at the cost of bowel prep and a procedure under sedation.
What "92% Sensitivity for Cancer" Actually Means
When the pivotal trial enrolled 9,989 people and ran both Cologuard and colonoscopy on every participant, 65 had colorectal cancer. Cologuard flagged 60 of them. FIT flagged 48.
That gap (92% vs 74%) is the strongest reason a stool DNA test exists. It catches early-stage cancers FIT misses, including the small T1 tumors where older meta-analyses peg FIT sensitivity at just 40%.
The advantage holds in real-world data too. In a study of Alaska Native adults, where colorectal cancer rates are among the highest in the world, Cologuard detected 100% of the cancers found at colonoscopy versus 80% with FIT.
The next-generation Cologuard, validated in a 2024 trial of 20,176 adults, edged the cancer sensitivity up to 93.9% (95% CI 87.1 to 97.7) and held its specificity for advanced neoplasia at 90.6%. The improvement is modest on cancer detection. The bigger gain is in fewer false positives.
Where the Test Misses
The same trial that found 92% of cancers found only 42% of advanced precancerous polyps. That number is the test's central limitation. Six in ten advanced adenomas, including some that would have progressed to cancer over the following 5 to 10 years, looked normal on Cologuard.
Sensitivity tracks closely with polyp size. In an automated-assay validation study, sensitivity climbed from 57% for advanced precancers measuring 1 cm or larger, to 73% for those over 2 cm, to 83% for those over 3 cm. Smaller advanced adenomas, especially flat or sessile serrated ones, are the hardest for any stool test to find.
That said, Cologuard is far better at catching one specific lesion type that traditional screens fail on. Sessile serrated polyps over 1 cm sit on the right side of the colon and bleed unpredictably; the methylated BMP3 marker that's part of Cologuard detected them 66% of the time in a head-to-head comparison, versus 10% with FIT. These polyps account for an estimated one-third of colorectal cancers despite being among the easiest for FIT to miss entirely.
What Happens After a Positive Result
About 1 in 7 Cologuard tests comes back positive. The follow-up is always a colonoscopy. The question is what that colonoscopy actually finds.
In a primary-care study of Medicare patients overdue on screening, 14.7% tested positive on Cologuard, and 96% completed the recommended follow-up colonoscopy. Of those positives, 8% had cancer, 43% had advanced adenomas, and 31% had non-advanced adenomas. About 8 in 10 positive Cologuards revealed some kind of neoplasia worth removing.
Other real-world cohorts have found similar yields, though the proportion classified as "false positive" depends on what you count. If your definition is "no cancer or advanced polyp," then roughly 7 in 10 in some primary-care series fall into that bucket. If your definition is "no clinically actionable finding at all," many of those same patients had non-advanced polyps that were still removed during the diagnostic colonoscopy.
There is one more subtlety. Knowing that a stool DNA test came back positive seems to make the follow-up colonoscopy itself more thorough.
In one center, withdrawal times were 19 minutes after a positive Cologuard versus 13 minutes for blinded screening colonoscopies, and the rate of right-sided flat lesions detected went from 9% to 40%. The test, in other words, primes the gastroenterologist to look harder.
What the False Positive Cost Actually Is
Cologuard's specificity in negative colonoscopies was 89.8% in the original trial and 92.7% in the next-generation version. That translates to roughly 7 to 13 unnecessary colonoscopies per 100 people screened, depending on the version and the population.
That is the real cost of the test's higher cancer sensitivity. FIT's specificity sits around 95%, which means fewer false alarms but also fewer cancers and far fewer advanced polyps caught.
The trade-off is genuine. Both tests are recommended in current guidelines, and the right pick depends partly on which kind of error you would rather make.
Stool DNA also catches roughly twice as many advanced precancerous polyps as FIT alone, with one Dutch screening cohort finding 46% versus 27%. So the false-positive penalty buys real precancer detection, not just more cancer detection.
Where Cologuard Fits in Real Decision-Making
Cologuard is best understood as a high-sensitivity cancer detector that doubles as a moderate-sensitivity polyp detector. It is FDA-approved for average-risk adults 45 and older and recommended every 3 years per current ACS guidelines.
Cologuard fits well when:
- You are average-risk, due for screening, and would rather start at home before considering colonoscopy. The 92% cancer sensitivity is meaningfully higher than what FIT delivers, and the test arrives at your door.
- You have been putting off screening. Data on previously-noncompliant Medicare patients suggests Cologuard substantially improves uptake among people who would otherwise stay overdue.
Colonoscopy makes more sense when:
- You have a family history of colorectal cancer, a known genetic predisposition, or a history of prior polyps. Stool DNA testing is validated only for average-risk screening; higher-risk patients need direct visualization.
- You have symptoms like rectal bleeding, a sustained change in bowel habits, or unexplained anemia. Screening tests are for asymptomatic people; symptoms require a diagnostic workup, which means colonoscopy.
What to Take Away About Cologuard's Accuracy
Cologuard is genuinely accurate for what it claims to do. It catches more than 9 in 10 cancers and substantially more advanced polyps than the older stool blood tests it largely replaced. It misses about 6 in 10 advanced precancerous polyps, generates a positive result in roughly 1 in 7 average-risk adults, and is meant to send most of those people on to colonoscopy.
If you understand those numbers, the test is a reasonable choice. If you expect it to function like a colonoscopy, you will be disappointed by both directions of error.
No referral needed. Results reviewed by a physician.