Among nearly 10,000 Medicare beneficiaries, the likelihood of finding advanced polyps or colorectal cancer (CRC) on surveillance colonoscopy was low. Yet, among patients for whom any follow-up recommendation – either for or against colonoscopy – was available, the vast majority (87%) were advised to return for the procedure in the future, even if their life expectancy was limited or there were no significant findings on their surveillance colonoscopy.
“These findings suggest that recommending against future surveillance colonoscopy in older adults with low-risk colonoscopy findings and/or limited life expectancy should be considered more frequently than is currently practiced,” say Audrey Calderwood, MD, with Dartmouth-Hitchcock Medical Center in Lebanon, N.H., and colleagues.
Because of the lack of clear guidance about when to stop recommending colonoscopies to older patients, it is not surprising that physicians recommend surveillance even for patients with low life expectancy, Ziad Gellad, MD, with Duke University Medical Center, Durham, N.C., said in an interview.
“As someone who performs these procedures, I can tell you that it is not easy to tell patients that they are too old to get preventive care, especially patients in whom your only interaction is the procedure itself,” said Dr. Gellad, who wasn’t involved in the study.
The study was published online in JAMA Internal Medicine.
Key findings
For older adults, surveillance after prior findings of colon polyps is the most frequent indication for colonoscopy. Data suggest that an estimated 5.6 million adults older than 75 will undergo follow-up colonoscopy annually by 2024.
For older adults with polyps, current guidelines recommend individualized decision-making about surveillance colonoscopy. That includes weighing the potential benefits (identifying and removing meaningful lesions to prevent CRC) against the burdens and potential harms (such as bleeding or perforation).
While most colon polyps are not harmful, a subset of polyps, if allowed to grow, can develop into cancer over 10-15 years. This long biological time line highlights the importance of considering life expectancy in deciding for whom surveillance colonoscopy should be recommended, Dr. Calderwood and colleagues note.
Using data from the New Hampshire Colonoscopy Registry, which is linked with the Medicare claims database, they evaluated surveillance colonoscopy findings and follow-up advice according to severity of findings and patients’ estimated life expectancy for 9,831 adults (mean age, 73; 54% men).
Life expectancy was 10+ years for 57.5% of patients, 5 to less than 10 years for 35%, and less than 5 years for 7.5%.
Overall, 791 patients (8%) were found to have advanced polyps (7.8%) or CRC (0.2%) on surveillance colonoscopy.
Recommendations to stop or continue future colonoscopy were available for 5,281 patients (53.7%). Among them, 4,588 (86.9%) were recommended to return for future colonoscopy, even when there were no significant colonoscopy findings or the patient’s life expectancy was limited.
Compared with life expectancy of less than 5 years, longer life expectancy was associated with advice to return for future colonoscopy regardless of clinical findings, with adjusted odds ratios of 21.5 and 2.7, respectively, for life expectancy of 10 or more years and of 5 to less than 10 years.
Among patients with no significant findings, 95% of those with life expectancy of 10 or more years were recommended to undergo repeat colonoscopy down the road, compared with 58% of those with estimated life expectancy of less than 5 years.
Among patients expected to live 5 to less than 10 years, future repeat colonoscopy was recommended for 75% who had no significant findings, 82% with one or two small polyps, and 88% with multiple polyps, advanced polyps, or CRC.
The recommended time to repeat colonoscopy was greater than life expectancy for 6.6% of patients with less than 5 years of life expectancy and 6% with 5 to less than 10 years of life expectancy.