Is Cologuard as good as a colonoscopy? What the trials and the USPSTF modeling show

"As good as" can mean two things, and they have different answers. One is how a single test compares: if you have a cancer or a large precancerous growth today, which test is more likely to find it? The other is how a screening program compares: if you repeat a stool test every few years from 45 to 75, or have a colonoscopy every 10 years, how much colorectal cancer death does each prevent over a lifetime?

The FDA's review of Cologuard Plus, approved October 3, 2024, separates them in its own warnings: "Programmatic performance of the Cologuard Plus test (i.e., benefits and risks with repeated testing over an established period of time) has not been studied. Non-inferiority or superiority of the Cologuard Plus test’s programmatic sensitivity as compared to other recommended screening methods for CRC and APL has not been established." The trials answer the first question. For the second, the USPSTF relied on modeling.

Both Cologuard trials gave adults who were scheduled for a screening colonoscopy a stool kit beforehand and compared the results, with colonoscopy and the lab report on anything removed as the reference. So the figures below are the share of colonoscopy's own findings that the stool test caught. Each trial also tested a fecal immunochemical test (FIT), a stool test for blood.

The FDA reviewed BLUE-C on a narrower population, excluding participants under 45 (the trial enrolled from 40) and anyone with a parent, sibling or child diagnosed with colorectal cancer. In that group of 18,911, the FDA reports sensitivity of 95.3% for cancer (81 of 85), 43.3% for advanced precancerous lesions and specificity of 90.7% in people with no cancer or advanced precancerous lesion. By stage, sensitivity was 88.0% for stage I cancers (22 of 25), 92.9% for stage II (13 of 14) and 100% for stage III (30 of 30). Both trials were funded by Exact Sciences, the maker, as each abstract states.

The pattern is the same in both trials and in the USPSTF's pooled review of four stool DNA studies (sensitivity 0.93 for cancer, 0.43 for advanced adenomas): the test is good at finding cancers that are already there and finds fewer than half of the larger precancerous growths, which are what colonoscopy removes to prevent cancer in the first place.

The warnings in the FDA's approval summary for Cologuard Plus translate the trial into counts. "Out of every 100 patients testing positive, approximately 3 patients will have CRC, 34 patients will have APL, 33 will have a non-advanced adenoma, and 30 will have no neoplastic findings." And: "Out of every 10,000 patients testing negative, approximately 2 will be falsely reassured that they do not have CRC. Out of every 100 patients testing negative, approximately 7 patients will be falsely reassured that they do not have APL."

False positives rise with age. The FDA review reports specificity of 97.8% at ages 45 to 49 and 85.5% at 75 and older, so roughly 2 in 100 people without an advanced finding tested positive in their late 40s, against about 15 in 100 from 75 (our arithmetic: 100 minus specificity). The same warnings say results "should be interpreted with caution in older patients as the rate of false positive results increases with age."

In these trials colonoscopy was the yardstick, so its own misses do not show. The USPSTF's evidence review does measure them: in four studies, colonoscopy's sensitivity for adenomas 10 mm or larger "ranged from 0.89 (95% CI, 0.78-0.96) to 0.95 (95% CI, 0.74-0.99)." Its harms are also counted. The USPSTF reports 14.6 major bleeding events and 3.1 perforations per 10,000 screening colonoscopies, and higher rates, 17.5 serious bleeding events and 5.4 perforations per 10,000, for colonoscopies done to follow up a positive stool test, "presumably because of fewer biopsies and adenoma removals" in screening colonoscopies.

A stool test that misses more per round can still be repeated more often. To compare whole programs, the USPSTF commissioned three CISNET models of a cohort of 40-year-olds screened from 45 to 75. The figures below are the mean across the models, per 1,000 people, from the Task Force's 2021 recommendation (Figures 1 and 2). The USPSTF states the assumption behind them: 100% adherence to every scheduled test and every follow-up colonoscopy.

Two caveats come from the Task Force itself. Stool DNA every 3 years "did not provide an efficient balance of the estimated lifetime number of colonoscopies vs the estimated life-years gained, compared with other options for stool-based screening," though it "is still estimated to provide a reasonable balance" compared with no screening. And yearly stool DNA buys its near-colonoscopy result with more follow-up: about 850 more follow-up and surveillance colonoscopies per 1,000 people than yearly FIT. The modeling also predates Cologuard Plus: the companion modeling study (Knudsen et al., JAMA 2021) lists the stool DNA test it simulated at 0.94 sensitivity for cancer and 0.91 specificity, taken from the 2021 evidence review.

Real life is the gap in this table. On our reading, a program that assumes every positive stool test is followed by a colonoscopy says nothing about how the two compare for someone who would skip that colonoscopy.

ACP's guidance is the outlier. It lists FIT or high-sensitivity guaiac testing every 2 years, colonoscopy every 10 years, or sigmoidoscopy every 10 years plus FIT every 2 years, and states that clinicians "should not use stool DNA, computed tomography colonography, capsule endoscopy, urine, or serum screening tests." Its announcement and abstract do not give a reason specific to stool DNA, and we did not read the full statement. Our colorectal cancer screening page sets out each body's starting and stopping ages.

The FDA approved the original Cologuard on August 11, 2014 for adults 50 and older, widened it to 45 and older on September 20, 2019, and approved Cologuard Plus on October 3, 2024 for adults 45 and older at average risk. The FDA's approval summary for Cologuard Plus states it is "NOT indicated" for people with a personal history of colorectal cancer or advanced precancerous lesions; a positive result from another screening test within the last 6 months, 12 months for a FIT or fecal occult blood test, or 36 months for a FIT-DNA test; a family history of colorectal cancer, "defined as having a first-degree relative (parent, sibling, or child) with a CRC diagnosis at any age"; inflammatory bowel disease; or a hereditary cancer or polyposis syndrome such as Lynch syndrome or familial adenomatous polyposis. It "is not a replacement for diagnostic colonoscopy or surveillance colonoscopy in high-risk individuals."

Screening is also for people without symptoms. Its warnings say not to provide a sample during diarrhea or with known blood in the urine or stool, and that "Unexpected bleeding should be discussed with your healthcare provider."

A positive stool test is the first half of the screen. The USPSTF: "Positive results on stool-based screening tests require follow-up with colonoscopy for the screening benefits to be achieved."

Medicare. Medicare.gov (read 2026-10-28) covers a multi-target stool DNA test every 3 years for people aged 45 to 85 at average risk with no symptoms, and states that after a positive result "Part B also covers a follow-up colonoscopy as a screening test." You pay nothing for it if your provider accepts assignment, but if a polyp is removed, "you pay 15% of the Medicare-approved amount for your provider's services," plus a 15% coinsurance to the facility in a hospital outpatient setting or ambulatory surgical center. The underlying regulation, 42 CFR 410.152(l)(5), sets Medicare's share for the screening colonoscopies described in 42 CFR 410.37(j), those that involve removal of tissue, at 85% for 2023 through 2026, 90% for 2027 through 2029 and 100% from January 1, 2030.

Private plans. In FAQ Part 51 (January 10, 2022), the Departments of Labor, Health and Human Services and the Treasury state that non-grandfathered group health plans and health insurance issuers, the ones section 2713 of the Public Health Service Act applies to, "must cover and may not impose cost sharing with respect to a colonoscopy conducted after a positive non-invasive stool-based screening test" for people described in the USPSTF recommendation, for plan or policy years beginning on or after May 31, 2022.

ACP says clinicians should select a test in consultation with the patient, "based on a discussion of benefits, harms, costs, availability, frequency, and patient values and preferences." Our colorectal cancer screening page carries each body's ages and test list, quoted and dated, and your personal screening plan lists the screenings that apply at your age and sex.

Questions worth asking: does anything in my personal or family history rule out a stool test, how often would I repeat it, and what would a positive result cost under my plan. Past 75, the Cologuard Plus warnings say the decision to screen "should be made on an individualized basis in consultation with your healthcare provider"; our guide to mammograms after 75 works through the same question for breast screening. For the numbers that sit alongside screening, see the blood tests worth getting every year.