Development of a CMS Quality Measure for Outpatient Colonoscopy

New quality measures should be carefully considered for how they will be gamed and other unintended consequences. The goal of colonoscopy is to prevent colorectal cancer by endoscopic resection of precancerous lesions. The risk for postprocedural complications, especially bleeding, is related to polyp size, proximal location, and resumption of anticoagulation in multiple studies. Failure to account or adjust for those factors in the model could lead clinicians to refer patients with high-risk polyps for surgical resection. Unnecessary surgical resection of benign colorectal polyps is already an enormous problem in the U.S. Endoscopists are already disincentivized to remove large polyps by poor reimbursement compared with small polyps and fear of complications. Once endoscopists start getting cited by their facilities for complications after large-polyp resection, the solution will be simple: Send patients to surgery, despite higher cost and greater risk. Endoscopists know which polyps are high-risk, though apparently CMS doesn't. Further, patients who are elderly and have comorbidities may be referred for less-effective diagnostic strategies (e.g., barium enema) or therapeutic strategies with greater cost and risk (e.g., surgery). Just because a model can be validated in a retrospective cohort does not mean it produces a quality measure that improves outcomes when applied prospectively. In colonoscopy, we have learned that lesson with quality measures such as withdrawal time. To disincentivize colonoscopy and polypectomy in patients with the most challenging lesions, and who at times have serious comorbidities but would benefit the most from this approach compared with surgical therapy, will be a grave mistake.