Last talk of the day is quality indicators in Barrett's esophageal cancer. Again, my disclosures, objectives, a few slides on why we should be measuring quality. The meat of the talk is going to be on quality indicators in Barrett's esophagus surveillance and endoscopic eradication therapy. And then next steps in terms of incorporating these quality measures or indicators in practice, what needs to be done. So why measure quality? The obvious answer to this is that high quality endoscopy, we presume, will deliver better health outcomes for everyone. And this is perhaps critically important as we shift from a volume-based reimbursement paradigm to a value-based practice and reimbursement paradigm. So we are placing more and more emphasis on value as opposed to volumes. And we know that physicians will likely be required to track and report their performance during endoscopy. This is happening in other fields. And at some point, reimbursement for endoscopy will be linked to reporting and our performance on these quality measures. So it behooves us to be at the forefront of the quality indicator development rather than having other perhaps ill-conceived measures foisted on us by providers or agencies who don't know much about endoscopy. So we should remain at the forefront of this process. We know that Barrett's esophagus is common, surveillance is recommended, and that healthcare costs to prevent ESE as well as from ESE mortality are immense and formidable. And we also know that our current guidelines, as we've just discussed, support screening, surveillance, endoscopic eradication therapy in Barrett's-related neoplasia and dysplasia. We have discussed surveillance is essential to detecting dysplasia and cancer. We are increasingly using more of the ablative technology because we are finding more dysplasia. And therefore, it is important that we identify and validate some of these quality indicators. This is important for benchmarking performance as well as improving outcomes. So the first attempt at developing quality indicators for Barrett's esophagus was made by a group of physicians, gastroenterologists, famous names, who met at DDW, I believe, in 2013, and the paper was published in 2015. There were 27 experts who used a modified Delphi method to discuss evidence and identify quality indicators. And they used this Delphi method of multiple iterations of these statements to achieve at least 80% agreement and also graded the evidence as high, low, and very low. And they came up with a few quality indicators. So I have divided these into screening and diagnosis. So for screening and diagnosis, the first statement was if Barrett's is suspected, the squamous-culmonary junction, the gastroesophageal junction, and the location of the diaphragmatic hiatus should be recorded. So again, we've talked about this identifying and recording the landmarks. Second, if Barrett's is suspected, the endoscopy should document the extent of suspected Barrett's using prior criteria. Again, we've discussed this today. Third, the normal appearing and normally located squamous-culmonary junction or Z-line should not be biopsied. We talked about that earlier on today and hopefully have convinced you as to why it can lead to more issues and why we think it is not a condition associated with morbidity or mortality. The fourth statement in surveillance is surveillance biopsy should be taken every two centimeters in four quadrants throughout the endoscopically involved segment. Statement five, if biopsies show no dysplasia, follow-up endoscopy should be recommended no sooner than three to five years. And in fact, this recommendation or this quality marker is also in the Institute of Medicine shortlist of quality indicators for surveillance in Barrett's esophagus. And I'll provide you with some data in this regard against a lot of agreement, but not much evidence. Statement six, we talked about this during surveillance biopsies from visible lesions should be obtained and processed separately from four-quadrant biopsies. We talked about is it good to biopsy, is it good to endoscopically resect? I think it should be as per the experience of the endoscopist. More important thing is that we identify them and we sample them and we put them in a separate bottle. So we know that, unfortunately, there is room for improvement in terms of compliance with surveillance recommendations. This was a large database study from a pathology database. And I know there has been a separate study from the GI quick database, which is showing slightly better numbers. But unfortunately, our compliance is 50% in this paper. In Dr. Vani's paper, I think it was in the 70s or high 70s. When you look at the number of biopsies taken every two centimeters, the compliance was around 50%. The longer the barrage segment, not surprisingly, the compliance actually fell. So you see it went from almost 80% to just 10% in longer segments. Most importantly, it is consequential. You detect less dysplasia if you biopsy less. So if you are not adherent, you are going to pick up less dysplasia, even though it exists. So you're going to miss dysplasia a lot. Let's move on to some data on how often are we appropriately recommending endoscopic follow-up in those with non-dysplastic and low-grade dysplasia, non-dysplastic barrage. This was from the GI quick registry. They looked at a five-year endoscopy record here and about 58,000 endoscopies mean barrage length of two centimeters. And this was the distribution of the recommended follow-up interval in those with non-dysplastic barrage. So you see only about a half were being recommended at three years, some more than three years, and a substantial number less than three years. So those are the ones that are being recommended less than three years. And even if you assume that endoscopies at three and six months may have been recommended for say erosive esophagitis, almost a quarter were still being non-compliant or were calling their patients back for endoscopy despite non-dysplastic barrage at less than three years. So that's surveillance in terms of taking biopsies or bringing patients back. The next concept I would like to introduce is something called neoplasia detection rate. This is similar to adenoma detection rate, which I think many of us are familiar with from the colonoscopy literature. And neoplasia detection rate, the definition is proposed as the proportion of patients who have high-grade dysplasia or cancer out of all the patients undergoing surveillance for barrage at their first surveillance endoscopy. So either at the first screening when surveillance is done or at the next first surveillance endoscopy. NDR has excluded low-grade dysplasia at this time, given the higher inter-observable variation. But there is a alternate metric of dysplasia detection rate, which would be low-grade plus high-grade dysplasia. And it is proposed to be a measure of the efficiency of surveillance endoscopy in barrage. And it may integrate a lot of these variables that we just talked about. How well are we inspecting, which goes to barrage inspection time? How well are we biopsying? Are we being compliant with the Seattle Protocol? Are we using white light endoscopy or chromoendoscopy? Because all of this is geared towards detecting dysplasia or neoplasia. And therefore, the dysplasia detection rate has been proposed. What should an ideal NDR be? So there are two studies I would like to highlight. The first one was systematic review and meta-analysis published in GUT, which looked at over 10,000 patients from 11 studies. And they came up with an NDR of 7%, of which 3% were cancers, and the remaining 4% were high-grade dysplasia. But they also separately looked at low-grade dysplasia, and that was about 10%. Now, these studies were largely from referral centers, expert centers. And we took a different approach, wherein we looked at a population-based cohort in southeastern Minnesota, all of whom underwent endoscopy done in a defined geographic cohort of over 1,000 patients. And we came up using the same definition with a slightly lower neoplasia detection rate of about 4.9%, likely because this was more community-driven, with an EAC prevalence of about 1.8%, but a very similar low-grade dysplasia rate of 10%. So this is where NDR likely should be, the proportion of patients undergoing their first surveillance who have either high-grade dysplasia or cancer detected. Remember, we are excluding patients who are symptomatic in terms of dysphagia, anemia, weight loss, etc. Now, NDR has been proposed as a metric, but a number of things need to be done before it is accepted as a validated metric. And the most important is correlating it with a clinically relevant outcome. So just like adenoma detection rate has been correlated, so the higher the adenoma detection rate, the lower the post-colonoscopy colorectal cancer, and the lower the colorectal cancer mortality. We have not done that yet with NDR, and ideally, we would want to correlate that with the post-endoscopy esophageal neoplasia or misdysplasia rate, as well as with EAC mortality. We also have to define the threshold at or above which the NDR should be, just like we have defined ADR thresholds for men and women as quality metrics. So that remains to be done. This was the first step that we took towards correlating. This was a group at Mayo Clinic and Indiana University, wherein we did a systematic review of meta-analysis, and we tried to correlate the NDR proportion of patients with hybrid dysplasia or cancer at initial surveillance endoscopy with the post-endoscopy Barrett's neoplasia rate. That is, the number of patients who had esophageal adenocarcinoma or hybrid dysplasia more than a year after their initial endoscopy, and what we found on meta-regression was that there was a significant negative correlation. So for every 1% increase in NDR, the post-endoscopy Barrett's neoplasia rate decreased by 3.5%. So again, this is something that needs to be done with DDR, which would be high-grade and low-grade, and esophageal cancer. What are some challenges with NDR? So unlike screening colonoscopy, screening and surveillance for Barrett's is relatively low volume. ESE is a rarer disease than colorectal cancer. So correlating the NDR or the DDR with the post-endoscopy Barrett's neoplasia or ESE rate is going to be very challenging at the individual endoscopist level. And we may have to do this at a practice or a hospital level. But again, these are steps that need to be taken. Let's switch gears and talk about quality indicators for endoscopic eradication therapy. And again, this was a conference, an effort led by Dr. Vani, Dr. Kumunduri, looking to develop quality indicators for endoscopic eradication therapies in Barrett's esophagus. And they used a fairly similar methodology with some modifications. So this was the RAND UCLA appropriateness methodology, also called RAND. This is modified Delphi. A bunch of experts got together, developed quality of care metrics. They had the opportunity to discuss their judgments between rounds. There was also a face-to-face group meeting, similar to an NIH consensus conference. Of note, level one evidence was not available to answer most questions. They developed procedures, they developed quality indicators pre-, intra-, and post-procedures. So what was, I'm not going to go through all of these, but perhaps some of the relevant ones. So for patients, this is pre-procedure, for patients in whom a diagnosis of dysplasia has been made, the rate at which the reading is made by a GI pathologist or confirmed by a second pathologist prior to endoscopic eradication therapy should be 90%. This is a process method. And we have gone through this with Dr. Thotta. She has showed us that the rate of inter-observer agreement between pathologists for low-grade dysplasia is pretty poor. It does improve somewhat when you change from a biopsy to an EMR, but it still remains on the poorer category. When you resect visible lesions, you do upstage your diagnosis or downstage your diagnosis in over half of these patients. So again, resection of any visible lesion is a very important quality metric. Now, importantly, do we really know what an expert GI pathologist is? And this was a very impressive effort that was published in Gut in 2020. And basically what it involved was a group of 51 pathologists from across the world. You can see where these pathologists came from. Most of the major continents are represented. And they all reviewed 55 barax esophagus biopsies along the diagnostic spectrum. And there was a reference panel of four pathologists and the study group of 51 pathologists. And the ultimate outcome was the concordance with this expert panel diagnosis of these 51 pathologists and what was the predictors of either concordance or discordance. And really the bottom line is that if you have a pathologist with at least five years of experience, consumerate with age, you would have less odds of making a mistake or not agreeing with the expert opinion. On the other hand, if you are working at a district general hospital, you are more likely to make a mistake compared to those working in an academic hospital or even a private hospital for that matter. The next issue that they tackled was center an endoscopist volume versus competency. This is very relevant with regard to utilization of endoscopic eradication therapies outside of tertiary care centers. And there were several unknowns in the field. How does one achieve competence? How many procedures do you have to do before you are declared quote unquote competent? What is the threshold number of cases required to achieve competence? What is the definition of a high volume center? And what is the relationship between endoscopist or center volume and outcomes? So I'm going to walk you through some recent evidence in terms of center and endoscopist volume and outcomes. So from the UK RFA registry, a nice paper was published which showed that if you define a high volume endoscopic eradication center as more than 40 EET cases per year and compare them to low volume centers, you achieve a higher rate of CRD and CRIM, reduced rates of rescue EMR, which means that as you are performing ablation, you are having to resect these lesions, which leads to the question that did you miss them, and is associated with lower recurrence rates. This is from the United Kingdom RFA data. There is also data to suggest that competency in RFA is achieved if you do 18 supervised cases, because studies have shown higher CRD and CRIM rates in more than 18 versus less than 12. Now remember, this is not a consistent story. There are some studies, even from the United States, which from the United Kingdom, which did not find an association between volume and CRD and CRIM rates. The US RFA registry also demonstrated increasing volume, leading to reduced sessions to achieve CRIM. But they did not see association with rates of CRD or CRIM once you exceed 30 patients. There is a strong correlation between endoscopist RFA volume and CRIM rate. And finally, a recent VA study from the United States showed that if you get treated in a high volume center, and they divided RFA volume across the nation into quartiles, and they compared the lowest quartile of the RFA experience to the highest quartile, you have lower recurrence in those which are in the highest quartile of RFA centers. So again, if you go to a center which has higher volume, you may have a higher likelihood of better outcomes such as higher rates of CRD, higher rates of CRIM, and maybe even lower recurrence rates. So indeed, the BSG and PSGE guidelines are suggesting that expert centers be defined as those which have at least 10 new Barrett's new dysplasia cases being treated per year, and that to gain competency or proficiency in EMR and RFA, you are doing at least 30 supervised cases of each. Of course, a high volume center is not defined just by the volume of cases it performs. It should really be a high quality practice wherein you have endoscopist with training in imaging, reception, and ablation, you have access to expert pathologists and you have access to expert surgeons. Now when we talk about intra-procedure quality metrics, you talk about the rate at which the Barrett segment is inspected using high definition white light endoscopy. Again, this is a process measure, target rate of 95%. The rate at which landmarks and length of Barrett's are documented using the Prague criteria, presence and absence of visible lesions is documented, process measure, performance target 90%. Intra-procedure, we talked about this, rate at which endoscopic resection is performed in patients with visible lesions, process measure, 90%. This is relatively new, and this is something that individuals or centers, for that matter, who perform Barrett's endoscopic eradication therapy should be striving for and perhaps monitoring, rate at which complete eradication of neoplasia is achieved by 18 months, 80%. Rate at which complete eradication of intestinal metaplasia is achieved by 18 months, 70%. So this comes from systematic reviews and meta-analyses of different ablation modalities, and they have come up with a fair amount of data. Rate at which complete eradication of intestinal metaplasia is achieved by 18 months, 70%. So this comes from systematic reviews and meta-analyses of different ablation modalities, and they have come up with a fairly conservative number. It was not 12 months, but they extended it to 18 months because patients may have trouble with follow-up, et cetera. So they gave us some legal. So in conclusion, in this era of value-based and quality-based healthcare, we should continue to refine and continuously evaluate our quality indicators, not only for Barrett's diagnosis and surveillance, but also for endoscopic eradication therapy. Implementation, the impact of implementation of these quality measures on hard, relevant patient outcomes, such as remission of intestinal metaplasia, remission of dysplasia, progression to cancer, adverse events, mortality, should be carefully studied. And finally, we need to address challenges which are germane to the process of measurement and evaluation. So how do we consistently measure these metrics? How do we monitor these metrics? How do we report these metrics? And most importantly, how do we validate these metrics? How do we correlate them to relevant patient care outcomes? Thank you. Thank you.

endoscopy

Barrett's esophageal cancer

quality measurement

value-based reimbursement

neoplasia detection rate

endoscopic eradication therapy