Tuesday, March 8, 2022

Debating colorectal cancer screening recommendations: too young, too often?

 - Kenny Lin, MD, MPH

Last year, the U.S. Preventive Services Task Force (USPSTF) updated its colorectal cancer screening recommendations, lowering the starting age for average-risk adults from 50 to 45 years; this change was reflected in the Putting Prevention Into Practice case study in American Family Physician's September 2021 issue. However, after reviewing the USPSTF statement and supporting documents, the American Academy of Family Physicians (AAFP) concluded that the evidence was insufficient to recommend a starting age younger than 50. Two editorials in the February issue of AFP outlined the arguments for and against starting routine screening at 45 years of age.

In the first editorial, Dr. Richard Wender argued that "lowering the starting age is a settled issue," noting that several organizations, including the American Cancer Society, the National Comprehensive Cancer Network, and the American College of Gastroenterology have all independently reviewed the data and come to the same conclusion as the USPSTF. He pointed out that "the incidence of colorectal cancer in 45 year-olds today is ... almost identical to the risk in 50-year-olds in 1979 when colorectal cancer screening was first recommended," and that nearly a quarter of deaths from colorectal cancer in the U.S. occur in individuals diagnosed between 45 and 54 years of age. Four microsimulation models have also concluded that starting screening at 45 years of age is the most efficient strategy to maximize life-years gained per colonoscopy regardless of the initial screening test used (including the multitarget stool DNA test discussed in the same issue of AFP).

The second editorial, by Drs. Corey Lyon, Alexis Vosooney, and Melanie Bird, elaborated on the AAFP's position. The authors noted that "many of the trials used in the modeling studies did not include individuals younger than 50 years or did not provide separate data for this younger age group, decreasing confidence in the data inputs." They also expressed concern about costs to patients and the health care system from implementing the USPSTF recommendation as opposed to optimizing screening in patients age 50 years and older: "Expanding screening to up to 80% of eligible patients 50 to 75 years of age would prevent three times as many colon cancer deaths at one-third of the cost [of routinely screening Americans 45 to 49 years of age]." (A previous AFP Community Blog post by Dr. Jennifer Middleton described two modestly successful outreach strategies to patients who were not up to date on colorectal cancer screening.) Finally, they argued that persistent disparities in colorectal cancer incidence and mortality in Black patients would be more appropriately addressed by improving insurance coverage and access to care in this population rather than lowering the age to start screening. 

While colorectal cancer screening tests remain underused by many patients, studies have also documented that screening colonoscopies are performed more often than necessary - for example, being repeated 9 or fewer years after an initial high-quality colonoscopy showed no significant pathology, in contrast to the American Gastroenterological Association's Choosing Wisely recommendation. A recent systematic review of 6 studies that estimated the rate of overuse of screening colonoscopy in U.S. populations found that it ranged from 17% to 25.7%. Overuse occurs when endoscopists recommend that patients have subsequent colonoscopies at intervals shorter than those supported by guidelines, and primary care physicians (PCPs) defer to subspecialists' recommendations. In an editorial, Drs. Archana Radhakrishnan and Craig Pollack explained the obstacles that PCPs face in going against subspecialist advice but argued that they can still "play an important role in preventing overuse of colorectal cancer screening and surveillance colonoscopies" by directing referrals appropriately and communicating with endoscopists about deviations from evidence-based practices.