Does Your 80-Year-Old Patient Really Need That Colonoscopy?

A large share of screening colonoscopies in adults over age 75 years were performed on individuals with a limited life expectancy, a group that stands to benefit little and carries a greater risk for complications, findings from a large health system showed.

In the cross-sectional study with a nested cohort of more than 7,000 patients undergoing colonoscopy for screening, 30% of those ages 76-80 years had a life expectancy below 10 years, a percentage that jumped to 71% for those ages 81-85 years, and to 100% for the smaller group of patients age 86 and up, reported Michael Rothberg, MD, MPH, of the Center for Value-Based Care Research at the Cleveland Clinic in Ohio, and colleagues.

Significant findings on colonoscopy were infrequent, the group detailed in JAMA Internal Medicine, as part of the journal's "Less Is More" series.

Among the full study population, 37.7% had a non-advanced neoplasia and 5.7% had an advanced neoplasia, including 0.2% with colorectal cancer (CRC). Detection of advanced neoplasia increased with each age group (from 5.4% to 6.2% to 9.5%, respectively; P=0.02).

But as the benefits of CRC screening can take at least 10 to 15 years (the interval for an adenoma to progress to CRC), screening those with a limited life expectancy is of little benefit and must be weighed against the potential harms, the study authors explained.

In the current study, "patients with life expectancies of fewer than 10 years were very unlikely to have their cancers treated," Rothberg and co-authors wrote in their discussion. "Furthermore, older patients had markedly increased overall endoscopic adverse events, including serious gastrointestinal adverse events, hospitalizations, and perforations."

Adverse events (AEs) leading to hospitalization within 10 days of the procedure were common, at 13.58 per 1,000 patients, and this rate increased with age. Overall, patients with a limited life expectancy had double the rate of complications.

The U.S. Preventive Services Task Force recommends against CRC screening for those over age 85, and recommends individualized decision-making for people ages 76-85, but the Task Force "does not stipulate how the practice might be implemented," the researchers noted.

While using estimated life expectancy to determine who stands to benefit is a "better measure" than chronological age alone, the current study showed physicians "often ignored" life expectancy and offered colonoscopy to people expected to live fewer than 10 more years, added Rothberg and co-authors.

As guidelines have shifted to screen for CRC earlier in life, they suggested that reducing over-screening in older adults with a limited life expectancy could reduce burden on the healthcare system.

"Of course, stopping screening in older adults may be challenging," the researchers acknowledged. "Physicians must first calculate life expectancy (at least for those younger than 85 years) but may not have the tools to do so. The tool by Cho et al offers one potential approach. Future studies should test the effects of making such tools available in electronic medical records."

After exclusions, the study from Rothberg and colleagues included 7,067 patients older than 75 years undergoing outpatient screening colonoscopy from January 2009 to January 2022 at multiple Cleveland Clinic sites in Ohio and Florida, with 82% of the cohort ages 76-80 years, 14% ages 81-85 years, and 4% age 86 or older. A random review of 150 of these colonoscopies (50 for each age group) confirmed that 96% were indeed for screening colonoscopy.

People were excluded if they had incomplete data, an indication for colonoscopy other than screening, had a prior colonoscopy in the past 5 years, or had a history of CRC or inflammatory bowel disease.

In the end, patients had two comorbidities on average, were a median age of 78 years, 56% were women, and over three-fourths were white.

The study's primary outcome was the proportion of colonoscopies performed on patients with a life expectancy under 10 years, with estimates determined via the tool by Cho et al based on their comorbidities. Other outcomes included findings on colonoscopy and adverse events from the procedure.

When analyzed by sex, 39% of men and 23% of women ages 76-80 years underwent a screening colonoscopy despite a life expectancy of under 10 years. For those ages 81-85 years, these percentages were 82% and 61%, respectively. Only 3% of colonoscopies were performed on people with a life expectancy below 5 years.

Of the 15 patients with CRC detected, 10 received no anticancer treatment (patient or clinician's preference). One of nine patients with a life expectancy under 10 years underwent treatment, as compared with four of six patients with a life expectancy of at least 10 years.

Severe AEs were more common as patients increased in age: 2.42 per 1,000 patients ages 76-80 years; 3.92 per 1,000 ages 81-85; and 11.07 per 1,000 age 86 and up.

Similarly, perforations in the three groups occurred at rates of 0.52, 1.95, and 3.69 per 1,000 people, respectively, with a rate seen in the oldest group that "exceeds the quality measures" from the American Society for Gastrointestinal Endoscopy/American College of Gastroenterology Task Force on Quality in Endoscopy, according to the authors.

Limitations cited by the researchers included the single-center design, that the study only included patients who actually underwent screening, and that the specialty of physicians ordering the colonoscopy was not identified. Assuming that some patients received post-procedure care outside the health system meant AEs were probably undercounted, they noted.

Comment