Most cancers found in the years after a clearing colonoscopy were already there. Corley and Kaminski showed that adenoma detection rate is how that risk moves. ACG/ASGE 2024 made ADR, sessile-serrated-lesion detection rate, bowel-prep adequacy, and recommended-interval adherence the four priority indicators. Cecal intubation still matters. Withdrawal time does not replace detection.

Experienced teaching points

Clinical Pearls

ADR is now screening plus surveillance plus diagnostic, age 45 and up. Leave out positive FIT, mt-sDNA, IBD, genetic syndromes, and therapy of a known neoplasm.
Minimum ADR is 35% (about 40% in men, 30% in women). It is a floor, not a trophy. Corley's Kaiser data still show cancer risk falling as ADR rises through the 30s and 40s.
SSLDR minimum is 6%. If the whole unit is under that, talk to pathology before you blame the withdrawal.
Adequate prep in at least 90% of outpatient exams. Inadequate prep is a repeat within a year, not a 10-year stamp.
Average withdrawal in a normal colon is 8 to 9 minutes. That number cannot substitute for ADR or SSLDR. A short withdrawal plus a low ADR is the inspection problem.

Clinical Bottom Line

Question	2026 practical answer
What are the priority indicators?	ACG/ASGE 2024: adenoma detection rate, sessile serrated lesion detection rate, bowel-prep adequacy, and recommended screening/surveillance-interval adherence. Measure these four first.
Who is in the ADR denominator?	Patients aged 45 years or older having screening, surveillance, or diagnostic colonoscopy. Exclude positive noncolonoscopy screening tests, IBD, genetic cancer syndromes, and exams done to treat a known neoplasm.
ADR target	At least 35% overall. Performance characteristics still differ by sex (about 40% men, 30% women). This is a minimum. Higher is better for cancer prevention (Corley, Kaminski).
SSLDR target	At least 6%, same denominator idea as ADR. Details of mucus-cap anatomy live on the SSL page.
Prep	Adequate enough to follow the recommended interval in at least 90% of outpatient exams. Split-dose or same-day. Inadequate: repeat within 1 year, not a 10-year letter.
Cecal intubation	Still an indicator: at least 95%, with photographs of the appendiceal orifice and ileocecal valve. It is no longer a priority indicator in units already above that line. Technique: ileum page.
Withdrawal time	Average 8 to 9 minutes in normal colons without biopsy or therapy. Not a substitute for ADR. A low ADR with a short average withdrawal is an inspection problem.
Interval cancer	Most post-colonoscopy CRCs are missed or incompletely resected lesions, not biological rockets. Raise detection, wash the right colon twice, and finish the polyp. Intervals: surveillance table.

What is a post-colonoscopy colorectal cancer?

WEO calls it post-colonoscopy colorectal cancer (PCCRC): a CRC diagnosed after a colonoscopy in which no cancer was found, up to the next due date. About half to two-thirds of those tumors sit in the missed-lesion bucket. Another chunk is incomplete resection. A minority are true new, fast biology, often serrated and right-sided.

That is why quality is not a customer-service score. Kaminski's Polish screening program (2010) showed interval-cancer risk falling as ADR rose. Corley's Kaiser cohort (2014) put a number on it: each 1% absolute ADR gain was associated with about a 3% drop in interval cancer and a 5% drop in interval-cancer death, across ADR quintiles from the high 10s into the 40s. The 2024 document exists because those curves did not flatten at the old 25% floor.

This page is the dashboard. Optical diagnosis of serrated lesions is the SSL companion. CADe is a separate argument on the CADe page; AGA 2025 made no recommendation for or against routine CADe. Do not use a box on the tower as a substitute for ADR.

How do I calculate ADR in 2024?

The old screening-only ADR is retired as the primary measure. Too many lists are mixed, and surveillance/diagnostic exams carry adenomas too. ACG/ASGE 2024 wants one ADR:

Numerator: patients with at least one conventional adenoma removed and documented by pathology. Sessile serrated lesions do not count toward ADR. They have their own rate.
Denominator: age 45 years or older, indication screening or surveillance or diagnostic symptoms.
Exclude: positive FIT, mt-sDNA, or other positive noncolonoscopy screening tests (those lists are enriched and would inflate you); IBD surveillance; polyposis and other genetic syndromes; colonoscopy for therapy of a known neoplasm.

Minimum acceptable ADR is 35%. Sex-specific floors of about 40% (men) and 30% (women) remain reasonable because prevalence differs. Do not game the number by excluding hard indications or by calling every SSL an adenoma. The point of the new denominator is honesty, not a higher trophy.

FIT-positive follow-up colonoscopy is a different list. Its ADR should be higher. Do not dump those patients into the screening ADR to look better, and do not use the 35% floor as the FIT-positive target.

Why is SSLDR now a priority indicator?

Right-sided interval cancers are disproportionately serrated. ADR does not capture them. New Hampshire and GIQuIC data showed SSLDR predicting PCCRC independent of ADR. About a third of endoscopists with an adequate ADR still sat under a 6% SSLDR in one registry cut. That is the hole.

SSLDR: percentage of the same-shaped denominator with at least one sessile serrated lesion on pathology. Minimum 6%, expected to rise. Hyperplastic polyps in the rectosigmoid do not count. If an entire group is under 6%, the first call is the pathologist, not a new attachment. Local SSL thresholds vary. Then wash mucus caps, look pale, and resect completely. Morphology and crypts: SSL page.

What else is priority besides detection?

Bowel-prep adequacy is now priority. Target: at least 90% of outpatient colonoscopies adequate to follow the recommended interval (BBPS typically 2/2/2 or better). Split-dose or same-day. An inadequate exam is a repeat within a year (US Multi-Society Task Force), not a "pretty good, see you in 10." Dirty prep is how the right-colon SSL disappears.

Recommended-interval adherence is the fourth priority indicator. Finding the polyp and then bringing the patient back in 3 years for a 7 to 10-year lesion is how you manufacture harm. The interval table lives at the surveillance page. Document why you shortened.

Cecal intubation remains an indicator (at least 95% of screening exams, photographs of the appendiceal orifice and ileocecal valve). It dropped off the priority list because most gastroenterologists already clear it. If your unit does not, it is still the first mechanical problem. Failed intubation is a incomplete exam, not a quality-neutral inconvenience.

Does withdrawal time still matter?

Yes, as a diagnostic when detection is low. Average withdrawal in normal colons without biopsy or therapy should be 8 to 9 minutes. Barclay 2006 made 6 minutes famous. Later work (Zhao, Desai, Butterly, Shaukat) pushed the useful average higher, and Shaukat tied longer withdrawal to fewer interval cancers. The 2024 document moved the recommended average to 8 to 9 minutes.

That is an average, not a stopwatch law for every patient. A 5-minute withdrawal in a perfectly prepared, left-colon-only look at a 28-year-old is not the same as a 5-minute withdrawal through a dirty right colon in a 68-year-old. Withdrawal time cannot replace ADR or SSLDR. If both detection rates are low and the average withdrawal is under 8 minutes, the inspection is the problem: pools not washed, flexures not seen, folds not pulled back, right colon not looked at twice.

Technique beats gadgets. Distal caps, water exchange, and position change help when the mucosa is actually being exposed. CADe finds polyps you already pointed the lens at. It does not look behind the fold you skipped.

How do interval cancers actually happen?

Three honest buckets:

Missed lesions. The majority. Flat, pale, right-sided, behind a fold, under residual stool. SSL and nonpolypoid adenomas. Second look in the right colon (reinsertion or retroflexion) is the cheap intervention.
Incomplete resection. Pohl's CARE study put a number on leftover neoplastic tissue after snare of 5 to 20 mm lesions, worse as size grew and for serrated histology. Cold snare the small ones completely. For 10 to 20 mm sessile adenomas, UEMR is often the better capture. Do not leave a margin because the list is running long.
New biology. Real, less common. Serrated pathway can move faster than the old 10-year adenoma story. That is an argument for SSLDR, not for nihilism about detection.

NordICC is why skeptics still argue. Invitation to a single colonoscopy cut 10-year CRC incidence 18% (intention-to-screen). Only 42% of invited people had the exam, and CRC-death reduction was not significant on that analysis. Adjusted per-protocol estimates were about 31% less incidence and 50% less CRC death if everyone invited had undergone colonoscopy, with lower-than-hoped detection and intubation. The quality document reads that trial as an underestimate of what a high-ADR exam does, not as permission to skip the dashboard.

What do I do if my ADR is low?

Audit and feedback first. Then, in order:

Prep. Split-dose. Same-day if the slot is afternoon. Measure adequacy.
Inspection. Wash, expose proximal fold faces, distend, double the right colon. Time will follow technique, not the other way around.
Recognition. SSLs. Subtle rims. Dedicate a few lists to unlearning "if it is not red and round it is not real."
Resection. A found polyp that is not gone is an interval-cancer seed.
Attachments and water exchange as adjuncts. CADe last, and only as an adjunct. See the CADe page.

Do not lengthen every withdrawal to 12 minutes as theater. Do not shorten surveillance intervals to hide a low ADR. Do not exclude FIT-positive patients from a screening ADR you already computed, and do not include them to inflate a new one.

Selected references

Last reviewed September 20, 2026. Written for clinicians who have to report colonoscopy quality numbers, and who want those numbers to mean fewer post-colonoscopy cancers rather than a dashboard ritual.

Topics

Colonoscopy

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.

← All articles