From the AGA Journals

Endoscopy experts review training, assessment evidence

View on the News

The foundation of high-quality endoscopy

Whitson, Williams, and Shah eloquently and thoroughly explore the principles central to successful training and review the latest understanding of best practices to apply them. Beyond fellows, this will have ongoing relevance to practicing endoscopists as they must learn new skills and in turn apply them in teaching others.

Jonathan Cohen, MD, FASGE, is a clinical professor of medicine at New York University Langone Health. He reported having no relevant conflicts of interest.

Dr. Jonathan Cohen

Feedback is an essential aspect of deliberative practice in my experience. The art of giving useful feedback requires one to be introspective, interactive, and iterative. Introspective: The instructor must assess the presession skill level and adjust the learning plan with observation in real time of student performance, taking primary consideration simultaneously for patient safety. From these inputs, the teacher decides a) what helpful information to convey, b) how best to deliver the message, and c) when best to do it. Interactive: “How best to deliver” will usually entail an approach to stop the action and ask the trainee to consider the present challenge and possible solutions rather than to dictate an action or demonstrate what to do. It is not always best to wait until after the procedure – contrary to what the authors favor in this article – but interruptions must be few and focused. Iterative: The lessons taught are building blocks that ideally relate to prior challenges and set the agenda for next learning goals.

The authors rightly emphasize the concern for cognitive overload. In practice, a maximum of one or two take-home lessons per mentored training session is a good rule of thumb; the converse should be equally emphasized, that every single proctored training examination ought to be mined for at least one relevant lesson, be it technical, cognitive, or another nontechnical pearl related to teamwork, professionalism, and so on.

Much simulator investigation to date, including my own, has focused on technical skills and performance outcomes – more work is needed especially in web, simulator, and even AI-based tools to teach cognitive skills for recognizing abnormalities, identifying them, and making real-time evidence-based decisions. The value of simulator-based teaching of endoscopic nontechnical skills, practice in troubleshooting common mishaps, and teaching by counter example of what not to do are other areas of great promise. There is not yet a prescribed, evidence-based guideline regarding which simulation devices should be used, at what stages of fellowship, and how often; however, the principles described in this paper provide the road map for how simulation ought to be integrated into endoscopic teaching.

Jonathan Cohen, MD is a clinical professor of medicine at New York University. He is the Editor of “Successful Training in Gastrointestinal Endoscopy,” 2nd ed. (Hoboken, N.J.: Wiley-Blackwell, 2022) and an investigator in ex vivo and computer endoscopy simulators. He is a consultant for Olympus America and Micro-Tech Endoscopy, receives royalties from Wouters-Kluwer and Wiley, and holds stock in GI Windows, Virtual Health Partners, ROMtech, and MD Medical Navigators.



Endoscopic training is increasingly complex as benchmarks for quality evolve and tools and procedures advance with innovation.

A team of experts, led by Matthew J. Whitson, MD, with Hofstra University/Northwell Health in Hempstead, N.Y., aimed to simplify challenges for educators and clinical endoscopists with a review of tools and techniques for education, as well as assessment methods.

Their review was published in the Techniques and Innovations in Gastrointestinal Endoscopy.

Giving feedback

Key steps to effective feedback include first talking about the goals for the endoscopy session, then careful observation, but minimal feedback during the endoscopy. Most of the feedback should come after the endoscopy, the authors wrote.

A paper by Walsh et al. demonstrated that with beginning endoscopists, giving feedback afterward led to long-term skill development as compared with short-term benefits of frequent feedback in the middle of procedures.

Feedback should be constructive and specific with emphasis on goals for the next procedure. It should be delivered in a respectful, nonthreatening way for greatest effectiveness.

Mastery learning

In this model, each trainee must achieve competence in specific skills to progress to the next level.

“For example, the trainee must master retroflexion in the stomach prior to attempting clip hemostasis in the stomach,” the authors wrote.

Repetitively practicing the skill is coupled with direct feedback.

Mastery learning is often paired with simulation so trainees can practice in a safe space before working with patients.

Cognitive load theory

Knowing the challenges of learners can help educators with instruction techniques. An important concept is cognitive load theory (CLT). CLT is focused on the working memory of a learner and the harm that an overload of information can have on learning. A learner’s working memory can process only a few pieces of information at any given time, the theory states.

One mitigation strategy by educators may be to assign a trainee a smaller piece of a specific task appropriate to the trainee’s skill level.

“For example, an early trainee endoscopist may be able to inject epinephrine for a bleeding vessel, but not be ready to perform effective BiCap cautery,” the authors suggested.

Different learning styles

Learning styles include visual, aural, reading/writing, and kinesthetic styles (when learners need to touch or manipulate to learn a skill).

“A study of surgical trainees demonstrated that kinesthetic learning was the most preferred unimodal learning style of those entering the field,” the authors wrote.

Dr. Whitson and coauthors gave examples of working with trainees with different learning styles.

A trainee who learns visually, they say, might need to learn about “loop reduction” by looking at images of alpha or beta loops or using ScopeGuide during endoscopy. A kinesthetic learner may need to feel a successful loop reduction with hands on the endoscope during simulation to understand the skill better.

“There is some suggestion that the millennial generation – the demographic of the current gastroenterology fellows – may have higher preferences for kinesthetic learning,” the authors wrote.

Role of simulation

The Accreditation Council for Graduate Medical Education, which oversees Gastroenterology Fellowship training, mandates simulation in gastroenterology education but does not specify endoscopic simulation. The American Board of Surgeons, however, does require their trainees to complete the flexible endoscopic curriculum, which is simulation-based.

Simulator use appears particularly helpful early in training. One study demonstrated that colonoscopy simulation has benefit in the first 30 colonoscopies in depth of insertion, independent completion, and ability to identify landmarks.

However, another study showed simulation after 50 colonoscopies has shown no benefit, the authors wrote. Finding uses for simulators beyond diagnostics will be important for justifying buying more of them for medical centers given the high cost.

Procedural volume

Dr. Whitson and colleagues wrote that using sheer volume of procedures as a measure of competency is falling out of favor and there is recognition in the field that competency will come at different times and at different volumes for individual trainees.

A study assessing competency in esophagogastroduodenoscopy (EGD), for example, demonstrated that, while most trainees will achieve independent intubation rates of the second part of the duodenum by 150 procedures, it will take between 200 and 250 for the average fellow to reach competency of all motor skills for a standard EGD, and 300 to become efficient (Gastrointest Endosc. 2019;90(4):613-20).

Assessment of skills has evolved from numbers of procedures to competency-based assessments to the development of direct observation tools.

Coaching for the practicing endoscopist

Most studies on coaching have focused on trainees, but coaching can be used with experienced endoscopists as well.

One study investigated use of direct verbal coaching to train experienced practitioners in water immersion colonoscopy “which resulted in shorter cecal intubation times, improved [adenoma detection rate], and less use of sedation during procedures,” the authors noted. Another study currently underway in the United Kingdom uses electronic feedback coupled with education and training to change behaviors to improve polyp detection performance in colonoscopy.

The authors noted that using one of these tools or strategies does not preclude using another.

“[I]n fact educators likely will recognize the utility of incorporating multiple of these techniques in the same endoscopy session with a trainee,” the authors wrote.

One author holds stock in Boston Scientific. The remaining authors disclose no conflicts.

Next Article: