and most importantly with greater patient comfort than traditionally trained fellows. This performance advantage was observed for roughly 30 patient‐based procedures, after which the skills for the traditionally trained group statistically caught up. On the basis of these positive results and others like it, some institutions, such as the Mayo Clinic, have adopted early training curricula around computer simulation, requiring all GI fellows to perform roughly 20–30 simulated colonoscopies prior to being allowed to begin patient‐based training. Could longer simulation training provide an even greater performance advantage seen in the simulator‐trained group? This is possible, as similar research to the study above has shown 10 hours of simulator training, imparting a measurable sbenefit in skills in up to 80 live cases [21]. However, other data examining the learning curves of performance metrics during simulation training have found that a trainee's performance on the simulator tends to plateau after roughly 20–25 cases, suggesting that a computer simulator has taught a trainee all it can during this length of training [28, 29]. This is likely due to the modest level of difficulty of the cases currently available on computer simulators [30, 31]. As the realism of looping, haptic feedback, and case complexity improves on these models, the benefits computer simulation training could conceivably extend well beyond the initial training of novices. Currently, however, it is recommended that computer simulation be used primarily for teaching early motor skills [22]. Simulators are also still prohibitively expensive ($75K–100K), and as a result are found primarily at larger academic teaching institutions. One solution that would allow for smaller training programs to reap the benefits of these teaching tools would be the development of regional training centers. This initial simulation training of 20–25 cases could easily be completed over a weekend course and sponsoring such courses could provide a return on an investment for institutions that have already purchased such devices. Regional training centers could also serve as testing centers. As computer simulators become more advanced, it is inevitable that they will become part of board certification in gastroenterology, where testing of competence in endoscopy skills will be eventually be required. Before this type of high‐stakes assessment could happen though, the complexity and measurable performance metrics of current endoscopy simulators would need to be greatly enhanced and separately validated for such testing purposes.
Figure 6.31 A trainee using a virtual reality colonoscopy simulator.
Figure 6.32 A static mechanical model, the colonoscopy Erlangen active simulator for interventional endoscopy (coloEASIE), is shown here.
Figure 6.33 Ex vivo models. In these images, harvested animal models are lain out on special platforms that configure the organs in the shape of a human colon. The Endo‐X trainer (a) is designed specifically to accommodate a harvested bovine colon while the Erlangen active simulator for interventional endoscopy model (EASIE‐R) (b) platform is shown here with a porcine digestive tract.
Static mechanical models such as the Erlangen active simulator for interventional endoscopy colonoscopy model (coloEASIE) consists of a tube of spiraled wired mounted to a platform that allows for practice in navigation and loop management (Figure 6.32) [32, 33].
Ex vivo colonoscopy models have also been developed and can be used for teaching early motor skills [33, 34] (Figures 6.33a,b). These models utilize harvested bovine or porcine colons that are laid out on a special platform in a human anatomical configuration. The use of these models are typically limited to more advanced training in skills such as therapeutic hemostasis devices or advanced endoscopy procedures such as EMR and ERCP. The ASGE, however, does offer annual training courses to first year fellows, teaching early endoscopy skills with the aid of ex vivo models, at their central endoscopy training center, the Interactive Training and Technology (ITT) Center in Downers Grove, IL. Another option is to utilize one of a number of commercial entities that offers the delivery and set up a temporary ex vivo training laboratory wherever it is needed. As the use of ex vivo simulation training for basic skills becomes more common, participation by first year fellows in standardized courses for colonoscopy could become commonplace as a precursor to starting patient‐based training.
In addition to the teaching above, the continuous formative assessment of early motor skills is an important part of training. This ideally should be done in a formative manner to help identify early bad habits such as overuse of the two‐handed scoping technique or to extinguish unsafe practices such as pushing while in red‐out. This should be done during the simulation phase of training (if used), otherwise very early in the patient‐based training experience to prevent bad habits from becoming ingrained.
In summary, computer, static models, or ex vivo animal models can be used effectively to teach basic endoscopy skills when used prior to beginning patient‐based training. Regardless of the type of simulation training used, it should be noted that it is not intended as a replacement to bedside teaching but simply a means to augment traditional training and possibly accelerate the acquisition of skills. There is no training that will ultimately better prepare one to perform colonoscopy on patients than actually performing patient‐based exams. If simulation models are not available, seeking out special courses such as those offered by the ASGE would be recommended. If these options are not possible, patient‐based training alone is still the standard and completely acceptable means to train these early skills.
Intermediate cognitive skills
The intermediate cognitive skills in colonoscopy hinge on a trainee's ability to recognize abnormalities and the decision‐making abilities of what to do about them. As discussed earlier in this chapter, the skill of recognizing patterns of pathology simply requires numerous encounters with various abnormal findings. As the trainee develops the ability to recognize patterns of pathology and their sometimes subtle differences, management decisions will become more refined as well. Instruction in this cognitive skill predominately rests on ensuring that fellows experience a wide variety of findings during patient‐based endoscopy. However, if patient exposure is the only means of education, a trainee's ability to recognize certain abnormalities could be limited due to patient selection biases or inadequate volume of certain abnormal findings (i.e., many polyps in a given practice but limited exposure to various presentations of inflammatory bowel disease). Instead, patient‐based training should be augmented with self‐directed study of photo atlases and multimedia resources that have been identified by instructors to ensure a wide variety (and more importantly, greater repetition) of pathology is experienced by the learner. One such media source is the GI Leap learning site of the American Society for Gastrointestinal Endoscopy. This online site (www.asge.org/home/advanced‐education‐training/online‐learning‐gi‐leap/gi‐leap‐lp) provides many examples of endoscopic images and videos along with case reports and explanations to accompany them. Contributions to the site are peer reviewed by its editorial board and review panel made up of nationally recognized names in gastroenterology.