Now, let's start drilling down into the quality and safety issues I've outlined. We will begin with Dr. T.R. Levin, addressing quality indicators for colonoscopy. Dr. Levin is a research scientist at the Kaiser Permanente Northern California Division of Research and Assistant Clinical Professor of Medicine at the University of California, San Francisco. He's the clinical lead for colorectal cancer screening for the Permanente Medical Group. Dr. Levin currently serves as a member of the ASGE Quality Assurance and Endoscopy Committee. T.R., the floor is yours. Thank you so much. I really appreciate this opportunity to be here and present for you guys today. We covered a lot of good rationale for some of the information I'm going to be sharing this morning. And there's a lot of material here, but I think we can also boil it down to some really few key concepts that I really want to make sure everyone takes home. So I just want to disclose that I have some research support from both my medical group and also Freenom. Learning objectives today, we want to describe the quality indicators for colonoscopy, demonstrate the practical value of measurement of colonoscopy quality, summarize updated colonoscopy surveillance recommendations, and discuss challenges in colonoscopy quality measurement. We'll start with the quality indicators common to all procedures, then go into colonoscopy quality indicators, both pre-colonoscopy, intra-colonoscopy, and post-colonoscopy, and then focus on improving colonoscopy quality and acknowledge some of the challenges that occur with quality measurement. So in reference to the diversity, equity, and inclusion, we really want to emphasize that high-quality colonoscopy is exhortation for all of our patients. When colonoscopy is not done with high quality, patients do not experience good outcomes, and it's really important that we deliver high-quality colonoscopy for everyone. And also, just to note that the benchmark for adenome detection is different for men and for women. So why does quality of colonoscopy matter? Colonoscopy is the most common endoscopic procedure in the U.S. Colonoscopy is one of the most effective procedures we do because it's associated with reduced colorectal cancer incidence and mortality. We've already seen that there's variable performance of endoscopists, and this variation does impact the effectiveness. So defining the elements of high-quality endoscopy and developing evidence-based quality measures and indicators is a priority, not just within gastroenterology, but also for organizations that pay for our services as well. So the issue that we're concerned about is post-colonoscopy colorectal cancer, also known as PCCRC, or interval cancers. These are not rare, and you can see up to between 3 and 8% or almost as high as 9% of patients will experience a cancer that occurs within a short interval of a prior colonoscopy. So why do these occur? There's a lot of theoretical explanations for why this might be happening. There's certainly biological variation in how quickly tumors in the colon will grow. You might have incomplete removal of polyps. Sometimes the polyps are technically very limited, very difficult to detect. There's technical limitations in the detection. So the mucosa could be hiding behind folds and hard to expose. There could be flat adenomas that might be very hard to see. The bowel prep could be inadequate, and the examination technique suboptimal. So Doug Robertson actually looked at this in a large cohort of 9,000 participants in eight large colonoscopy studies, and he found that probably the majority of the interval cancers in this cohort were missed lesions occurring within a short interval of a colonoscopy. And another 24% might have been a new lesion, and 19% were due to incomplete removal at polypectomy. So here's a poll question. Which of the following is not a priority quality indicator for colonoscopy? Frequency of documentation of an appropriate indication. Frequency of appropriate surveillance recommendations. Frequency with which colonoscopy bowel preparation is deemed adequate. Frequency of sequel intubation with photodocumentation of landmarks. And frequency with which adenomas are detected in asymptomatic average risk individuals. So again, this is a little bit of a tricky question because all of these things are important, but some of them are really priority quality indicators. So we're looking at which one is not a priority quality indicator. I think you guys are a pretty smart group already. Let me see if you kind of know the answer. You'll hear the answer after I finish the next section of the talk. So there are, I'm going to kind of reference several of the ASGE documents that refer to quality indicators. And so there are certain quality indicators that are common to all gastrointestinal endoscopic procedures. 23, in fact, 9 pre-procedure, 5 intra-procedure, and 9 post-procedure. But the priority quality indicator is common to all GI endoscopic procedures. So it's the frequency with which endoscopy is performed for an indication that is basically included in the published standard list of appropriate indications. The frequency with which prophylactic antibiotics are administered for appropriate indication. And the frequency with which management of antithrombotic therapy is formulated and documented before the procedure. So those are things we should all be doing for all of our endoscopic procedures. The quality indicators for colonoscopy and the priority quality indicators for colonoscopy here are adenoma detection rate, which we've been hearing about, and we'll talk some more, frequency with which colonoscopies follow recommended post-polypectomy and post-cancer resection surveillance intervals, and the frequency with which visualization of the cecum is noted by either the presence of landmarks and also with photo documentation of the landmarks. So we really want to strive for documenting that you've actually reached the cecum, it's very important. So there are other quality indicators. And most of them, as we've heard, are process measures. And we'll go through those in a minute. So the pre-procedure quality indicators. So indication, which we've talked about, informed consent is obtained, including discussions of risks, benefits, and alternatives. Number three, frequency with which colonoscopies follow recommended post-polypectomy and post-cancer resection surveillance intervals, and then number four, frequency with which ulcerative colitis and Crohn's colitis surveillance is recommended within the proper intervals. And this is the, again, the surveillance intervals is one of the priority areas. So here's another poll. We're going to talk about surveillance intervals at this point. When should you recommend surveillance in an average risk individual found to have one 8-millimeter adenoma and one 9-millimeter adenoma? These are, to clarify, these are tubular adenomas on a high-quality colonoscopy. Okay, so a lot, so it seems like the modal response was probably five years, although quite a few people are doing three years. Actually, the correct answer is seven to 10 years, and this is based on more recent data. And here's another question. When should you recommend surveillance in an average risk individual found to have one 10-millimeter tubular adenoma on a high-quality colonoscopy? Well, looks like this one was a little bit of an easier one for the group, but yeah, three years is the correct answer. So we've known this for quite a while, and this has been looked at in several different studies. This is one example from the prostate, lung, colon, and ovarian cancer screening study led by Rocky Shone. But there's both, in clinical practice, there's both overuse and underuse of surveillance. So, for example, in this cohort of patients, 26% of those with no adenomas were having another colonoscopy within five years. So this would represent overuse of colonoscopy in those patients. And then underuse, because only 58% of those with advanced adenomas have another colonoscopy within five years. And remember, the guidelines call for three-year surveillance for those patients. So it's very important to get your surveillance colonoscopy intervals correct. And it's tricky, and I'll have a suggestion for some workarounds that will help you remember what the intervals actually should be. So this is a key quality indicator. The appropriate recommendation for timing of the next colonoscopy should be documented more than 90% of the time. So if colonoscopy is done with inappropriate surveillance intervals, it's not cost-effective, it increases risk, it would be considered an inappropriate indication. It can lead to both overuse in low-risk patients and underuse in higher-risk individuals. And we've seen that, as I showed you in that data. And reassuringly, those of us who are gastroenterologists and have advanced training in colonoscopy are less likely to use the incorrect intervals. But non-gastroenterologists may recommend intervals much shorter than they probably should be. And so it's important to measure the intervals, measure how you're documenting them. So Samir Gupta led the Multi-Society Task Force on an update of these guidelines in 2020. And I'll kind of go through those for a minute. This is a flow sheet that I actually keep on my phone, because when my pathology comes back, sometimes I need to check it, even though I've been a co-author on some of these guidelines in the past. There's a lot of information here, and it can be hard at times to keep track of it all. So the important thing is, the most important first step is to have a high-quality colonoscopy. Make sure it's complete to the cecum. Make sure you have an adequate bowel prep, where you can detect polyps greater than 5 millimeters. You want to make sure your colonoscopist has an adequate adenoma detection rate, and we'll talk a little bit about that, that you have complete polyp resection. So all these intervals all apply to really a high-quality exam where you're confident that you've gotten the polyp completely removed. Then you risk stratify based on the size, number, and histology of the polyps that were present. So anyone with a normal colonoscopy or fewer than 20 hyperplastic polyps, less than 10 millimeters, can be deferred for 10 years for their next colonoscopy. The update to the guideline was a 7- to 10-year interval for people with 1- to 2-small adenomas less than 10 millimeters. 5- to 10-years for people with 1- to 2-cessile serrated polyps less than 10 millimeters. There's not really strong evidence that those small serrated polyps are actually at increased risk compared to the tubular adenomas, but they can be a lot harder to see, and it can be harder to be completely sure that you've removed them completely. So that was why the committee kind of hedged a little bit and gave a little bit of a wider interval of potential for earlier recall for people with cessile serrated polyps. 3- to 5-year interval for people in the 3- to 4-adenoma range, 3- to 4-small adenoma range, or 3- to 4-cessile serrated polyps less than 10 millimeters. If you have a hyperplastic polyp greater than or equal to 10 millimeters, there's a lot of concern that that may just be your pathologist not recognizing cessile serrated polyps, so we tend to treat those as a serrated polyp, cessile serrated polyp equivalent, so that's why those are in the 3- to 5-year range. And then we do 3 years for people with, you know, quote-unquote advanced adenomas, that would include 5- to 10-adenomas, 5- to 10-cessile serrated polyps, an adenoma for cessile serrated polyp greater than or equal to 10 millimeters, adenomas with villi cystology, tubular villi cystology, or high-grade dysplasia, and cessile serrated polyps with dysplasia, and then the very rare, but important to recognize, traditional serrated adenoma. And then you want to bring back people at 1 year if they have more than 10 adenomas, because there's probably a higher risk that you missed one because you were so busy removing that number of adenomas. One other point to remember is that if you have a piecemeal resection of a cessile serrated polyp or an adenoma more than 20 millimeters or 2 centimeters in size, those people you want to bring back sooner, at 6 months, because with piecemeal resection it's harder to be completely confident that you've removed the polyps completely. So let's go through some of the changes. I've kind of alluded to it. The biggest change was probably the change to 7- to 10-years instead of 5- to 10-years for people with a 1- to 2-small tubular adenomas, because in reality, based on available evidence, they have very similar risk for colorectal cancer compared to those who have a completely normal colonoscopy, and they are lower risk for colorectal cancer compared to the general population. So this is really a low-risk group, and they can safely extend their intervals at least to 7 and probably more often to 10 years. The MSTF reinforced the importance of surveillance colonoscopy in 3 years for those among patients with advanced adenomas because the evidence suggests increased colorectal cancer risk compared with patients with a normal colonoscopy, as well as compared to the general population. The Multi-Society Task Force recommends colonoscopy in 3- to 5- instead of 3 years for those with 3- to 4-adenomas because this is really likely a low-risk group if a high-quality colonoscopy has actually been performed. And the Multi-Society Task Force provides more specific recommendations for patients with serrated polyps. This is kind of an area in evolution, and there's been a lot learned in the more recent past about the importance of these lesions. There's a growing body of evidence that suggests that patients with large serrated polyps are at increased risk for metacrinous large serrated polyps, as well as colorectal cancer on follow-up. So what happens when you do a second surveillance, or when you, what happens with the findings at surveillance colonoscopy when planning the interval for the next surveillance? So let's kind of walk through that. If you have, if the, so it's very important to know what the baseline colonoscopy was and what was found in order to plan when the next surveillance should be based on what you find at the first surveillance colonoscopy. So it's 1- to 2-tubular adenomas, less than 10 millimeters, kind of the low-risk group. Remember, these people could be followed in 7- to 10-years. If you do your first surveillance and they remain low-risk, either they have a normal colonoscopy or they have the 1- to 2-tubular adenomas less than 10 millimeters, they're in the 10-year or 7- to 10-year category. As they start finding riskier lesions in that group, then you would approach the interval based on what that flow sheet would have shown you before. So 3- to 4-tubular adenomas, 3- to 5-years. An adenoma greater than or equal to 10 millimeters in size or other advanced features, they would come back in 3 years. So if you had sort of this middle ground advanced situation with 3- to 4-tubular adenomas, less than 10 millimeters, then the recommended surveillance was 3- to 5-years. And again, they tend to follow the risk of the finding at that first surveillance. The important key is that if they had a high-risk lesion on the first colonoscopy, a high-risk adenoma, then your recommendation would have been three years. Then in all cases, the longest interval you're going to choose for that person is going to be five years. Then otherwise, if you find a more advanced lesion, then you might also follow them in three years. There was a really nice clinical practice update in gastroenterology last year that I would refer you to in the reference here, guiding you through some of the best practices in colonoscopy. It creates a nice little flow diagram of things you want to do to either avoid late surveillance for high-risk patients, or also avoid overuse of surveillance in low-risk patients. I think we've talked about the intervals pretty well, but it's all laid out for you here. But it's really key to hit surveillance right in that sweet spot. Then we're going to talk about the inter-procedure quality indicators. You want to make sure you document BALPREP and we're going to be spending a while talking about BALPREP operation a little later today. Sequel intubation, also key, with photo documentation, and the adenoma detection rate. This is the frequency with which adenomas are detected in average risk individuals undergoing screening. Withdrawal time is also a quality indicator. We recommend that it is measured. The average should be greater than or equal to six minutes in screening colonoscopies in which polyps are not removed. A lot of people will have long withdrawal times when they're removing multiple polyps or doing surveillance biopsies in inflammatory bowel disease. But the key for the six-minute interval is really for those with the negative results on the screening colonoscopy. Then you'd want to make sure if you're doing colonoscopy with an indication for chronic diarrhea that you're taking biopsies. We recommend tissue sampling done with the indication of surveillance in UCM Crohn's colitis. Make sure you're taking sufficient biopsies or doing close examination for those patients. Essentially, the other quality indicators that people with pedunculated or sessile polyps smaller than two centimeters should not be referred for surgery prior to attempted removal at colonoscopy. Those patients should not be having surgery. They should really be removed endoscopically. Cecal intubation, these are the landmarks that you want to identify. You've got the appendiceal orifice. Most people doing colonoscopy can find that. The ileocecal valve, those are some key landmarks to make sure that you can find them and also document them. Photo documentation, mandatory. Photograph both the appendiceal orifice and the ileocecal valve. If you're not certain, then enter the TI and take a picture in there. Performance targets, cecal intubation rate with photography greater than or equal to 90 percent. Cecal intubation rate of greater than 95 percent on screening in particular. Exceptions, we don't want people to overdo it, trying to get to the cecum if there's a poor prep, there's severe colitis, or the indication for the procedures that therapeutic intervention distal to the cecum. This is a study looking at administrative data, looking for quality measures associated with post-colonoscopy colorectal cancer. Really when there's a suboptimal colonoscopy completion rate, or essentially not reaching the cecum with greater than 90 percent, you can see a significant increase in the risk for interval colorectal cancer. Just emphasize the adenoma detection rate. Sometimes there's some confusion about how to measure this. This is calculated in screening colonoscopies, average risk individuals over the age of 50. We exclude diagnostic and surveillance colonoscopies from the denominator. This is the current best practice, although I'll talk a little bit about some potential changes that may occur in the future. This is probably the single most important quality measure in colonoscopy, and it's directly correlated with colorectal cancer incidents. The ADR targets overall, greater than or equal to 20 percent in women and greater than or equal to 30 percent in men. This is an update from prior lower ADR targets. The rationale for the targets, an ADR less than 20 percent is associated with a tenfold increase in interval cancer compared to if you have your ADR above 20 percent. It clearly makes a difference. This is the data from Doug Corley's work in our medical group. These quintiles actually are quite old, but at the time he was looking, this is where things broke out with the 33.5 percent as the cut point to get to quintile 5. As was alluded to, every one percent increase in the ADR was associated with a three percent decrease in the risk of cancer and also every one percent increase in the ADR was associated with a five percent decrease in the risk of fatal cancer. Improving your ADR can have as much impact on your patient outcomes as whether the patient actually had a colonoscopy or not. The range between providers is, at least at the time we were measuring this, very large and can make a big impact on patient outcomes. Withdrawal time is also a recommended measurement. There's some evidence that having a slower withdrawal time is associated with improved adenoma detection. The six-minute withdrawal time in many series, it's really separated low and high-level detectors. There's a statistically significant correlation between interval colorectal cancer and withdrawal time shorter than six minutes. Its primary value as a quality metric may be in correcting low performers. When you start to measure ADR in your practice, if you have someone who's a low performer, then you also want to check their withdrawal time. The first thing to do is to tell them to slow down if they're withdrawing too fast. This is just graphical representation of what I just told you, which is that the withdrawal time predicts the number of adenomas found. This is in a series of 12 endoscopists. This is from Dr. Shockett's paper, looking at 51 gastroenterologists, showing that as the average withdrawal time extended the interval colorectal cancer rate, decreased. Clearly, there's a plateauing at around seven or eight minutes and above that, it didn't necessarily make that much difference in terms of interval cancer rate. Post-procedure quality indicators, so this is tissue sampling. Post-procedure, sorry, would be the incidence of perforation after colonoscopy, looking at all indications, and also look at screening and polyp surveillance. The frequency with which post-polypectomy bleeding is managed without surgery, and the frequency with which appropriate recommendations for the timing of repeat colonoscopy is documented. As you can see, there's several of these quality indicators all related to surveillance. It's important to get that right. How do we improve the quality of our colonoscopies? One thing that's widely recommended is having a report card. Essentially, you can improve what you don't measure. That's very key to measure it and provide feedback to the endoscopist in your practice. Let them know that these things are important. Particularly relevant is adenoma detection. You can see just by providing feedback in this series, there was an improvement in adenoma detection before the intervention and during the intervention. SQL intubation documentation went up, adenoma detection went up, and also proximal adenoma detection went up. The other thing is to train the leaders in detecting adenomas. There's a variety of training approaches. We can talk some more about it during the Q&A, but people can be trained and they can improve. Providing feedback in education, as you can see, clearly training made a big difference in this group of endoscopists and significantly higher than people who did not get the training. There's a variety of devices you can affix to the end of your scope to help flatten out the folds. You can also recommend it as retroflection, particularly in the right colon to look for the polyps that tend to hang out behind folds as well as caps and other things. These also may make the biggest difference in the low detectors. If you have low detectors, these are things that can help make them aware that there's a lot of stuff they might be missing. Just to go through some of the best practices all on one slide, definitely want to use split prep, and we'll be spending a while talking about that. High-definition scopes, second look in the right colon with a retroflex, cold snaring for polyps 3-9 millimeters. This helps ensure complete removal. Patients with benign complex polyps should be referred for endoscopic resection, not surgery. Definitely want to have clear and detailed post-procedure documentation and follow guidelines when assigning the screening and surveillance intervals, and those are summarized there. You really think of QI as a multi-component process, definitely with report cards and feedback, educational interventions, review technique, as well as using technology to help you get the best quality colonoscopy. As you think of moving along on your quality journey, you can think of three tiers of quality. Tier 1 would be where all providers are meeting the minimal benchmarks. You're using split preps, you're measuring your withdrawal time, you're measuring your ADR, you're meeting the ADR benchmarks, you've identified your load detectors, and you've developed interventions to improve their detection. Everyone's doing the minimum procedure documentation, and you're adhering to the guideline recommended surveillance intervals. As you get to Tier 2 and you start to improve beyond that, you develop interventions to improve bowel preparation for all your patients, develop interventions to improve polyp detection for all the providers. Clearly, even people who are meeting benchmarks, as you look at the variation between your group, you can see that there may still be opportunities to improve, and you definitely want to intervene to improve post polypectomy practices. Then Tier 3 is really optimizing quality. All providers are continuously monitoring and optimizing their polyp detection. We have optimized care for patients with large polyps, detailed procedure documentation, and monitoring adverse events and interval cancer rates. In the future, we can expect AI to probably join us in the endoscopy room and help make sure that we can find a lot of those small polyps that we may be missing. This is one device that's now available for purchase, and it does make a difference, particularly for the smaller polyps. Maybe not as much of a difference for the larger polyps, but if we're doing colonoscopy once every 10 years, the small polyps are important because they may eventually turn into big ones if we leave them in place. When we looked in the past, definitely the quality metrics are not necessarily universally measured. Kudos to those of you attending this course, because I assume you're going to go back to your practices and really start measuring quality on a more aggressive way than you may have been in the past. There's a lot of challenges with measuring colonoscopy quality, particularly ADR. You need to identify your screening colonoscopies. This may lead to the fact that there's a small number of procedures that you could use because of small sample size, you can get variation in results. There's been a proposal, it still needs to be vetted, but clearly looking at all indications for colonoscopy and looking at ADR for those, they tend to track very closely, both the screening ADR and then other colonoscopies for other indications. The problem is we don't really have good benchmarks for the other colonoscopies ADR. This is a stay tuned, probably coming in the next few years. Keep an eye on this space. There's other challenges. It's very labor-intensive. A number of groups have worked on automating the ADR measurement. There's some references there. Those are options. There's certain software, depending on which endoscopy software you're using, that could help you measure your ADR. Intentionally, it does not include serrated polyps. The recommendation is for a separate measure for sessile serrated lesions. ADR does definitely emphasize detection of small lesions. The other risk is that people may say, oh, I found my one polyp, now I'm done. There's some experts really looking at adenomas per colonoscopy as another important outcome. For surveillance recommendations, this is also labor-intensive because you need to know what was the indication for the procedure, how high quality was the bowel preparation, was the cecum intubated, and what were the findings and what was the recommendation you need to linkage the pathology results. You can't just look at the procedure note, you have to actually link to the pathology results as well. The recommendation here is to do an audit of a selected subgroup of exams rather than trying to report this out on all exams. Cecal intubation rate, same thing, does require a manual review. There are some automated systems that have been used. This is another example where you could audit a subsample of the exams just to get an answer. I would just say that despite these challenges, measurement of colonoscopy quality is the standard of care today. There was a report from a few years ago, there was a group of 10 colonoscopists in Illinois who were required by one of the large employers in their town to publicly report their ADR. Will other employers or payers require this? It may happen and it may come, so I think we need to be prepared for that. Practices should definitely develop workflows to facilitate quality measurement. We talked about the GI endoscopy software with quality reporting tools, auditing the cecal intubation and the surveillance recommendations. Just to remind you all, colonoscopy quality is associated with cancer incidence and mortality. Reporting colonoscopy quality can lead to improve quality. Focus on the priority quality indicators, appropriate indication, adenoma detection, cecal intubation rate, and appropriate surveillance recommendations. Low-level adenoma detection actually endangers patients, and there's many proven interventions to improve ADR and AI is coming to help us. I'll just leave you with this phrase, if you don't measure quality, you don't know if you've achieved it. Thank you very much.

Dr. T.R. Levin

colonoscopies

quality and safety

post-colonoscopy colorectal cancer

adenoma detection rate

training and feedback interventions

artificial intelligence in colonoscopy