A 53-year-old woman was referred for surveillance colonoscopy. She is a current smoker with a history of chronic kidney disease, chronic obstructive pulmonary disease, atrial fibrillation, and two diminutive hyperplastic polyps found on average-risk screening colonoscopy 3 years previously. Her prep at the time was excellent and she was advised to return in 10 years for follow-up. She has taken the day off work, arranged for a driver, is prepped, and is on your schedule for a colonoscopy for a “history of polyps.” Is this an appropriate referral and how should you handle it?
Most of us have had questionable referrals on our endoscopy schedules. While judgments can vary among providers about when a patient should undergo a procedure or what intervention is most needed, some direct-access referrals for endoscopy are considered inappropriate by most standards. In examining referrals for colonoscopy, studies have shown that as many as 23% of screening colonoscopies among Medicare beneficiaries and 14.2% of Veterans Affairs patients in a large colorectal cancer screening study are inappropriate.1,2 A prospective multicenter study found 29% of colonoscopies to be inappropriate, and surveillance studies were confirmed as the most frequent source of inappropriate procedures.3,4 Endoscopies are performed so frequently, effectively, and safely that they can be readily scheduled by gastroenterologists and nongastroenterologists alike. Open access has facilitated and expedited needed procedures, providing benefit to patient and provider and freeing clinic visit time for more complex consults. But while endoscopy is very safe, it is not without risk or cost. What should be the response when a patient in the endoscopy unit appears to be inappropriately referred?
The first step is to determine what is inappropriate. There are several situations when a procedure might be considered inappropriate, particularly when we try to apply ethical principles.
1. The performance of the procedure is contrary to society guidelines. The American Gastroenterological Association, American Society for Gastrointestinal Endoscopy, and American College of Gastroenterology publish clinical guidelines. These documents are drafted after rigorous research and literature review, and the strength of the recommendations is confirmed by incorporation of GRADE (Grading of Recommendations Assessment, Development, and Evaluation) methodology. Such guidelines allow gastroenterologists across the country to practice confidently in a manner consistent with the current available data and the standards of care for the GI community. A patient who is referred for a procedure for an indication that does not adhere to – or contradicts – guidelines, may be at risk for substandard care and possibly at risk for harm. It is the physician’s ethical responsibility to provide the most “good” and the least harm for patients, consistent with the ethical principle of beneficence.
Guidelines, however, are not mandates, and an argument may be made that in order to provide the best care, alternatives may be offered to a patient. Some circumstances require clinical judgments based on unique patient characteristics and the need for individualized care. As a rule, however, the goal of guidelines is to assist doctors in providing the best care.
2. The procedure is not the correct test for the clinical question. While endoscopy can address many clinical queries, endoscopy is not always the right procedure for a specific medical question. A patient referred for an esophagogastroduodenoscopy (EGD) to rule out gastroparesis is being subjected to the wrong test to answer the clinical question. Some information may be obtained from an EGD (e.g., retained food may suggest dysmotility or the patient could have gastric outlet obstruction) but this is not the recommended initial management step. Is it reasonable to proceed with a test that cannot answer the question asked? Continuing with the endoscopy would not enhance beneficence and might be a futile service for the patient. Is this doing the best for the patient?