Going Against Medical Advice: PCPs’ Role in Reducing Colonoscopy Overuse

Recent efforts have been successful in increasing the rates of screening colonoscopy for colorectal cancer (CRC); currently, 65 % of U.S. adults over the age of 50 receive appropriate screening.¹ At the same time, there has been growing recognition of the problem of over-screening. As Kruse et al. highlight in this issue, many patients who receive a second colonoscopy do so too early.² The median length of time from a normal screening colonoscopy to a follow-up screening was only 6.9 years, as compared to the 10 years recommended by current guidelines.

Overuse of colonoscopies can result in complications, increased financial costs, and misallocation of limited resources. Van Hees et al. estimated that reducing the interval between screening colonoscopies from ten to five years resulted in a marginal benefit, but with an additional 1.3 complications and $573,000 in spending per 1,000 Medicare beneficiaries.³

In addition to documenting the problem of overuse, Kruse and colleagues provide evidence as to what might be driving overuse.² Through detailed chart reviews, they found that the endoscopist recommended follow-up intervals that were shorter than guideline recommendations in over half of the cases, including 62 % of normal colonoscopies and 75 % of examinations with hyperplastic polyps. This study confirms and expands on an older survey in which endoscopists were frequently found to recommended surveillance colonoscopy earlier than guidelines suggested.⁴ Given the findings by Kruse and colleagues,² what is the role of primary care physicians (PCPs) in ensuring properly timed CRC screening?

PCPs are critical in the CRC screening process, as they make the initial referrals and often the referrals for subsequent examinations. Older survey data, however, reveals many gaps in PCPs’ understanding of screening guidelines, leading to both overuse and underuse. Yabroff et al. showed that fewer than 20 % of PCPs properly recommended guideline-consistent screening across all modalities (colonoscopy, sigmoidoscopy, or fecal occult blood testing).⁵ Others have shown that PCPs are not familiar with guideline-recommended intervals for the next surveillance exam after patients have polyps removed.⁶ It is likely that PCPs’ recommendations for repeat colonoscopies are being driven to a large degree by endoscopist recommendations, and going against subspecialist advice can be challenging for a number of reasons.

First, PCPs may not independently know or be up to date on the latest guidelines for repeat screening and/or surveillance colonoscopies. Furthermore, as CRC guidelines become more complex and nuanced, PCPs may be uncertain of the distinctions they make regarding different colonoscopy findings. Presumed to have expert knowledge in the field, the specialist may seem more empowered to make recommendations, and as such, PCPs may defer to their advice.

Second, even in situations when PCPs know the guidelines and recognize that the endoscopist’s recommendations depart from recommended practice, PCPs may suspect that the endoscopist has a particular clinically based reason for his/her approach. A shorter interval may signal an initial suboptimal examination not relayed in the colonoscopy report; it may reflect concerns an endoscopist may have had about the particular patient based on family history or risk factors; or it may indicate that the endoscopist does not believe that guideline-recommended intervals are appropriate. Though the American Gastroenterological Association recommendation received a rating of moderate quality, some specialists may view the recommended intervals as too long.⁷ In this setting, going against the specialist’s recommendation could place the patient at increased risk.

Third, it can be challenging to go against patient expectations when they have been explicitly told by the specialist to follow up in a designated amount of time. It may take substantial amounts of time for PCPs to explain to their patients why they are altering these recommendations and may demand a level of certainty that PCPs do not feel. Underlying these considerations are PCPs’ worries about malpractice and professional liability for opposing a specialist’s recommendation. Survey data found that physicians were concerned that they could face legal risk by not offering screening colonoscopies to their patients, and this feeling, then, was directly associated with over-referral for colonoscopies.⁸ Even adherence to guidelines may not protect PCPs from the potential guilt and anxiety if a patient experiences a bad outcome.

In the face of these uncertainties, what can be done to empower PCPs to provide appropriate clinical care? The first step is working to standardize medical advice so that PCPs do not need to go against it. Both specialists and PCPs need to conform to evidence-based guidelines and effectively deliver the same message to the patient in order to prevent confusion and the resultant unnecessary testing. In the ABIM’s Choosing Wisely initiative, two of the five recommendations of the American Gastroenterological Association addressed over-screening. The first was a reminder to providers and patients not to repeat CRC screening for ten years after a high-quality negative colonoscopy in an average-risk individual. The second was not to repeat a colonoscopy for at least five years for patients who have one or two small (<1 cm) adenomatous polyps, without high-grade dysplasia, completely removed via a high-quality colonoscopy.⁹ It is too early to determine the impact of Choosing Wisely on overuse. The fact that these two recommendations made it onto the list, however, is a promising start.

When gastroenterologists deviate from the guidelines, PCPs may take an active role in trying to uncover the reason. In some cases, early recommendations by the endoscopist may be due to a legitimate clinical concern that may not be well documented in the colonoscopy and/or pathology report. Krist et al. carefully reviewed endoscopists’ colonoscopy reports, and documented poor communication between the specialist and the PCP regarding the procedure or its findings.¹⁰ Reports did not provide contextual data to account for the discordance between the endoscopist-recommended colonoscopy follow-up interval and that suggested by guidelines. Communication, therefore, is essential to ensuring appropriate timing of colonoscopy follow-up; however this may be challenging in the setting of busy clinical practices and when patients frequently change providers. Improving the documentation of clinical reasoning leading to recommendations for early repeat examinations and instituting interventions that facilitate communication between PCPs and specialists may improve the quality of care that PCPs are able to provide their patients.

Finally, PCPs have an important role in directing referrals appropriately. New quality metrics may be developed and reported that focus on how often endoscopists perform repeat colonoscopies before the indicated intervals. This information—together with metrics around adenoma detection rates and other quality metrics—may be used to help PCPs decide where to send patients for their procedures.

At the same time that PCPs work to avoid over-screening, it is critical that they remain vigilant regarding completion of appropriate follow-up examinations. Many patients with abnormal colonoscopy results fail to comply with timely surveillance.

As the problem of over-screening gains greater attention, PCPs can play an important role in preventing overuse of CRC screening and surveillance colonoscopies. This requires a concerted effort by both generalists and specialists to be knowledgeable of and adherent to evidence-based clinical guidelines and to be coordinated in our approach. In doing so, PCPs will ultimately be equipped to provide better high-quality, cost-effective care to our patients.