Post-Polypectomy Surveillance: Are We Recommending Repeat Colonoscopy Too Early?
Why are surveillance intervals often too short after 1–2 nonadvanced adenomas? Review the 2026 study, guideline adherence and opportunities to improve care.
A five-year recommendation after two small adenomas: Appropriate caution or unnecessary surveillance?
A 58-year-old patient undergoes a high-quality screening colonoscopy. Two small tubular adenomas are completely removed, bowel preparation is adequate, and there are no additional colorectal cancer risk factors.
The endoscopist recommends another colonoscopy in five years.
The decision may seem conservative. Yet under the 2020 US Multi-Society Task Force (USMSTF) recommendations, the appropriate surveillance interval for this clinical scenario is generally seven to ten years.
Why, then, might a shorter interval still be recommended?
This question lies at the centre of a study published in Clinical Gastroenterology and Hepatology in September 2026.
Investigators examined surveillance recommendations for 41,606 patients who underwent removal of one to four nonadvanced adenomas within Kaiser Permanente Northern California.
Despite gradual improvement after the 2020 guideline update, approximately three in ten patients with one or two nonadvanced adenomas continued to receive recommendations for repeat colonoscopy earlier than the guideline interval.
The findings highlight a persistent challenge in colorectal cancer prevention: translating updated surveillance evidence into routine endoscopic practice.
However, the study assessed recommendations rather than completed procedures. It identifies potential over-surveillance, not a measured increase in unnecessary colonoscopies or evidence of harm from those recommendations.
The 2020 guidance changed the interval for low-risk adenomas
Post-polypectomy surveillance aims to identify patients who may benefit from subsequent colonoscopy based on their colorectal neoplasia risk.
The recommended interval depends on findings at the index examination, particularly polyp number, size and histology, together with examination quality and relevant patient characteristics.
In 2020, the USMSTF revised several surveillance recommendations.
For patients with one or two tubular adenomas smaller than 10 mm, completely removed during a high-quality colonoscopy, the recommended interval changed from five to ten years to seven to ten years.
For patients with three or four small tubular adenomas, the recommendation changed from three years to three to five years.
The 2020 US Multi-Society Task Force (USMSTF) guidelines extended recommended surveillance intervals for patients with low-risk adenoma findings compared with earlier guidance.
For patients with 1–2 nonadvanced tubular adenomas, the recommended interval increased from 5–10 years to 7–10 years.
Similarly, for patients with 3–4 nonadvanced tubular adenomas, the recommended surveillance interval was extended from 3 years to 3–5 years.
These changes reflect a shift toward less intensive colonoscopic surveillance for selected patients with low-risk adenoma findings, recognizing that shorter intervals may not always provide additional clinical benefit.
The recommendations assume a high-quality baseline colonoscopy, complete polyp removal, and no other factors requiring a different surveillance strategy.
Both revisions provided an opportunity to reduce surveillance frequency in appropriately selected patients.
The guidance assumes a high-quality examination with adequate bowel preparation, complete visualization and confident polyp removal.
It should not be applied automatically to patients with hereditary colorectal cancer syndromes, inflammatory bowel disease or other circumstances requiring different surveillance approaches.
These boundaries are essential when interpreting the study's results.
A large integrated health system offered a test of real-world adherence
Jeffrey M. Hendel and colleagues conducted a cross-sectional study of colonoscopy surveillance recommendations from 2019 through 2024.
The analysis included 41,606 patients aged 45–75 years who underwent colonoscopy with removal of one to four nonadvanced adenomas.
Among them, 33,787 had one or two nonadvanced adenomas, while 7,819 had three or four.
The procedures were performed by 281 endoscopists.
Nonadvanced adenomas were defined as tubular adenomas smaller than 10 mm without high-grade dysplasia or villous features.
All included patients were considered at average risk for colorectal cancer, with adequate bowel preparation and complete colonoscopy to the cecum.
Patients with previous colorectal polyps or colorectal cancer, inflammatory bowel disease, hereditary colorectal cancer syndromes or a family history of colorectal cancer were excluded.
The researchers also excluded those with advanced adenomas, serrated lesions, proximal hyperplastic polyps or at least five nonadvanced adenomas.
Surveillance recommendations were obtained through manual review of post-colonoscopy documentation.
The primary outcome was whether the recommended interval fell within the applicable USMSTF range.
The investigators also assessed the use of different years within each interval range and evaluated patient and endoscopist characteristics associated with nonadherence.
Importantly, no electronic decision-support intervention was introduced during the study period.
This was an assessment of clinical recommendations over time, not a randomized evaluation of a strategy to improve them.
One or two adenomas: Adherence improved, but the gap persisted
The most important finding involved patients with one or two nonadvanced adenomas.
Before the guideline revision, adherence to the then-recommended five- to ten-year interval was 88% in 2019.
After the guideline changed, adherence to the new seven- to ten-year interval was 62% in 2020.
The rate subsequently improved.
Adherence to the USMSTF-recommended 7–10-year surveillance interval showed a gradual improvement following the introduction of the 2020 guidance.
In 2020, adherence was 62%, increasing to 68.1% in 2021 and reaching 71.9% by 2024.
Overall, this represents a 9.9-percentage-point increase between 2020 and 2024, suggesting progressive adoption of the extended surveillance recommendation in clinical practice.
Despite this improvement, approximately 28% of cases remained nonadherent in 2024, highlighting a continuing gap between guideline recommendations and real-world practice. The reported figures do not establish whether nonadherence reflected earlier or later surveillance recommendations.
Between 2021 and 2024, each additional calendar year was associated with approximately 6% lower odds of receiving a recommendation outside the guideline range.
Nevertheless, 28%–32% of patients with one or two nonadvanced adenomas were advised to return in less than seven years during this period.
None were advised to wait longer than ten years.
Thus, nonadherence in this subgroup reflected recommendations that were shorter than the guideline range rather than excessively delayed surveillance.
The gradual improvement is encouraging, but approximately one-quarter to one-third of recommendations remained shorter than the guideline interval several years after its introduction.
This suggests that guideline publication alone does not necessarily produce complete changes in clinical practice.
It does not establish why individual endoscopists selected shorter intervals or whether all deviations were clinically inappropriate.
Three or four adenomas: A different adherence pattern
The results were substantially different for patients with three or four nonadvanced adenomas.
Following the 2020 update, 90%–91% received recommendations within the three- to five-year interval.
This high adherence persisted through 2024.
Approximately 2.5% were advised to return earlier than three years, whereas about 7% received recommendations longer than five years.
The distinction is important because recommendations outside the guideline range were not all shortened in this group.
Compared with patients who had one or two adenomas, those with three or four showed more consistent adherence after the revised recommendations were introduced.
The study cannot determine whether this difference reflected confidence in the evidence, differences in perceived risk or other aspects of decision-making.
However, it identifies the one-to-two-adenoma group as a particularly relevant target for future implementation efforts.
Rather than treating surveillance nonadherence as a uniform problem, quality-improvement initiatives may benefit from identifying specific recommendations that are most frequently applied inconsistently.
Endoscopists favoured the boundaries of recommended ranges
Another distinctive observation concerned how clinicians used interval ranges.
For patients with one or two nonadvanced adenomas, approximately 46%–47% received a ten-year recommendation, while 20%–25% received a seven-year recommendation.
No more than 1.5% were assigned an eight-year interval, and none received a nine-year recommendation.
The same pattern appeared among patients with three or four nonadvanced adenomas.
Approximately 51%–52% received a three-year recommendation, while 38%–40% received a five-year recommendation.
No more than 1.4% were assigned four years.
Thus, endoscopists generally selected the shortest or longest interval rather than a value between the boundaries.
The investigators suggested that this pattern may be relevant to future guideline design.
One possibility is that discrete recommendations could be easier to implement than broad ranges.
However, the study did not compare range-based recommendations with specific-year recommendations in a randomized intervention.
It therefore cannot establish that replacing interval ranges would improve adherence or clinical outcomes.
The observation is best understood as a potentially useful insight into how recommendations are translated into practical decisions.
Which patients were more likely to receive earlier recommendations?
The investigators used multivariable logistic regression to evaluate factors associated with recommendations shorter than the guideline interval.
Among patients with one or two nonadvanced adenomas, several characteristics were associated with earlier recommendations:
Older age and male sex.
Body mass index of at least 30.
Previous smoking.
Greater comorbidity burden.
Certain differences in endoscopist experience.
Patients with a Charlson Comorbidity Index score of at least three had approximately 30% higher odds of receiving a shorter-than-recommended interval than patients with a score of zero.
Endoscopists with 15–24 years of experience had approximately 20% higher odds of assigning a shorter interval than those with fewer than 15 years.
The association was not statistically significant among endoscopists with at least 25 years of experience.
Shorter-than-recommended intervals were less frequent among patients undergoing diagnostic rather than screening colonoscopy, Hispanic patients and non-Hispanic Asian patients.
None of the examined factors was significantly associated with nonadherence among patients with three or four nonadvanced adenomas.
These findings identify associations, not causes.
For example, the analysis does not establish that smoking history or obesity independently warrants more frequent surveillance after complete removal of one or two small tubular adenomas.
Nor does it demonstrate that clinician experience directly causes shorter recommendations.
Unmeasured clinical considerations and differences in practice may have influenced the observed patterns.
Why might clinicians continue to recommend earlier surveillance?
The authors discussed several possible explanations in their accompanying GI & Hepatology News interview.
These included incomplete awareness of revised recommendations, uncertainty about the supporting evidence, established practice habits, medicolegal concerns, patient-related factors and financial incentives in some practice environments.
These possibilities should not be mistaken for measured causal explanations.
The study did not determine which motivations were responsible for individual recommendations.
An additional consideration is the clinical difference between assigning an interval based solely on polyp findings and making a decision using the complete patient context.
Guideline adherence is meaningful only when the correct guideline applies to the patient and the examination.
A shorter interval may be appropriate if bowel preparation was inadequate, resection was uncertain or a patient has risk factors placing them outside the average-risk population.
However, the study deliberately selected patients with adequate preparation and complete examinations, and excluded several major high-risk conditions.
This makes the persistent prevalence of shorter recommendations noteworthy.
Even so, the researchers acknowledged that some endoscopists may have intentionally selected earlier follow-up based on perceived risks not fully captured in the analysis.
Consequently, a recommendation outside the nominal interval cannot automatically be labelled unnecessary.
The difference between potential over-surveillance and demonstrated harm
The study's outcome was the interval documented after colonoscopy.
It did not track whether patients actually underwent surveillance at the recommended time.
It also did not assess subsequent advanced adenoma detection, interval colorectal cancer, cancer mortality or complications caused by additional procedures.
This distinction places an important boundary around the conclusions.
Earlier recommendations may increase the likelihood of more frequent colonoscopy, but the study did not measure that downstream effect.
Potential consequences of unnecessary surveillance include additional bowel preparation, procedural burden, sedation exposure, procedure-related risks and demand on endoscopy capacity.
These are reasonable clinical and health-system considerations, not outcomes quantified by this analysis.
Conversely, the study does not establish that extending every shorter recommendation to seven or ten years would be safe in all patients.
Clinical circumstances outside the study's eligibility criteria may justify different follow-up.
The strongest supported conclusion is therefore that recommendations shorter than guideline ranges remain common in a carefully selected low-risk population.
Whether reducing those recommendations improves clinical outcomes or healthcare efficiency requires further evaluation.
Turning surveillance guidance into more consistent practice
The findings suggest several potential opportunities for quality improvement.
One approach discussed by the investigators is integrating guideline-based clinical decision support into the electronic health record.
A system could use documented polyp number, size, histology, bowel preparation quality and examination completeness to suggest an appropriate surveillance interval.
Automated results letters could communicate the recommendation consistently to patients and referring clinicians.
The authors also discussed requiring documentation of a specific reason when an endoscopist intentionally chooses an interval outside the recommended range.
Such an approach could preserve clinical flexibility while making deviations easier to evaluate.
Another possibility is incorporating guideline-concordant recommendations into institutional quality monitoring.
These strategies remain proposals rather than interventions tested by the study.
A future implementation trial could examine whether they improve recommendation accuracy without creating inappropriate restrictions on individualized care.
It would also be important to determine whether improved documentation translates into appropriately timed colonoscopies, patient understanding and better use of endoscopy resources.
What should clinicians conclude from a single-system study?
The investigation offers several strengths, including a large patient population, standardized inclusion criteria, detailed review of recommendations and data from 281 endoscopists across multiple calendar years.
The period before and after the 2020 guideline update provides useful information about how adoption changed.
However, the integrated healthcare setting may not reflect other clinical environments, particularly fee-for-service systems or practices with different referral pathways and documentation processes.
The exclusion of patients with family history, serrated lesions and other risk modifiers also limits generalizability.
Some recommendations or clinical characteristics may have been misclassified. In addition, changes in healthcare delivery during the COVID-19 pandemic may have influenced recommendations in 2020 and 2021.
Most importantly, the cross-sectional observational design cannot establish causation or determine which interventions would improve adherence.
Clinicians should therefore interpret the findings as evidence of an implementation gap, not proof that a particular surveillance policy has caused patient harm.
Clinical Takeaway: Appropriate surveillance is not simply earlier surveillance
The 2026 Kaiser Permanente study demonstrates that adoption of updated post-polypectomy guidance remains incomplete, particularly after removal of one or two nonadvanced adenomas.
Although adherence improved from 68.1% in 2021 to 71.9% in 2024, approximately three in ten patients continued to receive recommendations shorter than the guideline interval.
By contrast, adherence exceeded 90% among patients with three or four nonadvanced adenomas.
GastroAGI perspective: The principal clinical signal is a quality-of-care opportunity. Endoscopists should apply established surveillance recommendations to appropriate patients while documenting legitimate reasons for individualized departures.
The evidence does not support eliminating clinical judgment or assuming every shorter interval is inappropriate.
Future research should determine whether decision support, improved communication and systematic quality monitoring can reduce unnecessary surveillance recommendations while preserving appropriate colorectal cancer prevention.
Successful surveillance depends not only on detecting and removing polyps, but also on recommending the right follow-up at the right time.
Primary Reference
Hendel JM, Jensen CD, Merchant SA, et al. Endoscopist Adherence to Post-polypectomy Surveillance Interval Recommendations Following Removal of 1-4 Non-advanced Adenomas. Clinical Gastroenterology and Hepatology. Published online September 2026.
DOI: https://doi.org/10.1016/j.cgh.2026.08.036
Supporting Source
Brunk D. Endoscopists set shorter-than-guideline intervals for 3 in 10 patients with 1-2 adenomas. GI & Hepatology News. 12 October 2026.
Guideline Reference
Gupta S, et al. Follow-up After Colonoscopy and Polypectomy: A Consensus Update by the US Multi-Society Task Force on Colorectal Cancer. American Gastroenterological Association, 2020.
References
We are pioneers in clinical intelligence, dedicated to helping gastroenterologists harness the power of artificial intelligence to drive precision, efficiency, and patient growth.