Transcript: Anatomic Pathology and Surgical Pathology

Episode information

Date of session: 12/16/20

Facilitator
Triona Henderson, MD, MPH
Centers for Disease Control and Prevention

Didactic speaker
Vinita Parkash, MBBS, MPH
Yale New Haven Hospital
vinita.parkash@yale.edu

Transcript

TRIONA HENDERSON: So good afternoon. My name is Triona Henderson, and I'm the clinical pathologist and the facilitator of this ECHO pilot project. I extend a warm welcome to you from the Division of Laboratory Systems at the Centers for Disease Control and Prevention in Atlanta, Georgia. Thank you for joining us today.

The topic for this interactive web discussion is anatomic pathology and surgical pathology. Our subject matter expert is Dr. Vinita Parkash, who will present today's topic. Dr. Parkash is an Associate Professor of Pathology, Gynecology, and Reproductive Sciences at Yale School of Medicine, and an instructor in Health Policy and Management at the Yale School of Public Health where she teaches quality and safety in healthcare organizations. Thank you for joining us today, Dr. Parkash.

So here is some technical information about the ECHO session. So before introductions and the presentation, we will have — so most of you are muted. If you are able to use your video, which enhances interaction, we would really appreciate it. It's very helpful to put a name to the face while we are having dialogue.

All microphones are now currently muted during this discussion section. Please unmute yourself to speak. If you are connecting by phone only and would like to speak, please announce yourself by name and organization before you begin.

If you are experiencing any technical difficulties, you can reach out directly via private chat to Johanzynn Gatewood, labeled as DLS ECHO Tech, and she's waving to you right now. She will do her best to respond to your issue.

Finally, we've designed these relevant sessions based on your evaluation. We would like to encourage you to complete this final post-session evaluation. For those of you requiring P.A.C.E.^® credits for continuing education, you are offered that option at the end to either receive a certificate of participation or your P.A.C.E.^® certificate. If you have any additional comments or questions or any feedback for us on these ECHO sessions, please send us an email directly at DLS, that's DLSEcho@cdc.gov.

So again, and many of you have heard these. How are these sessions different than a regular expert lecture teleconference or webinar? It's because their case discussions, and just open discussion being the main feature. Subject matter experts hope to share some of their work that may be transferable to you in your individual laboratories and institutions. These ECHO sessions will focus exclusively on the clinical laboratories and clinical laboratory issues in the United States and US territories.

Once again, we value your feedback, and discussion amongst all of you that ensues and would encourage you to share your own experiences and challenges on this topic. We thank you for your interest and participation. So here is the line out of the process today. So there will be a presentation by the subject matter expert. Then there will be clarifying questions by me. Then we will open up the floor for ideas, shared experiences, and comments by our subject matter expert. Then we'll have closing comments and then adjourn.

Today's session is being recorded. If you do not wish to be recorded, please disconnect now. Closed captioning is now also available and you can find the link in the Zoom chat box. Here is the biosketch of Dr. Parkash. She's received her undergraduate degree in chemical engineering from the University of Nevada. She attended medical school at the University of Toledo. Following residency in Anatomic and Clinical Pathology at East Carolina University.

Really? I did not know. Is that correct? No, I didn't think so. I was like, wait. No. We will have to fix that, Vinita. Now we'll invite Dr. Parkash to begin her presentation. Dr. Parkash?

VINITA PARKASH: Easily introduced. I've been at Yale all my professional life.

HENDERSON: That's what I thought. I was like, wait.

PARKASH: Well, you know. What are you going to do? Sometimes things happen. All right. Should I get started?

HENDERSON: The floor is yours.

PARKASH: Thank you. Can everyone hear me? OK. So I'm going to be talking about diagnostic error and anatomic pathology. So the National Academies of Medicine define diagnostic error as a diagnosis that is either inaccurate, not communicated to the patient, or not timely. So as you can see, the mnemonic there is A-C-T. So accurate, communicated to the patient, and timely. And diagnostic error has been recognized recently as a leading concern in US healthcare by a number of organizations, including ECRI, which is a patient safety organization that is federally funded and works with the AHRQ.

The estimates for diagnostic error vary. We don't really know what they are. Some people estimate as many as four million cases, resulting in 1.7 million deaths in the United States. I'm not sure I quite buy that. But importantly, diagnostic error is a very expensive error. It's very expensive both in human cost and financial cost.

So this chart here shows the analysis of over 10,000 cases from Covaris, which is a malpractice carrier. And it breaks up all of their cases, those 10,000 cases, claims, over a five-year period from 2013 to 2017. As you can see, diagnosis-related errors is the number one error that results in claims. And what did I do? Sorry, I went backwards.

And in addition to that, what is important here is to see that the indemnity paid for diagnostic errors is very high because these are generally high severity errors. So if you look at the National Association of Insurance Commissioner's scores, an abnormally large number of cases fall in the six to nine range, which is either severe disability or death. So there's very heavy human cost with respect to patients, and also heavy human costs for the provider. We accept systems errors a little bit easier than diagnostic errors, which are very personal to us and related to clinical judgment.

This is the breakup of by the area where in the patient journey the error happened. And as you can see, lab tests are a very large component of it. Whether it's ordering the lab test, performance, received or transmittal, or interpretation of test results by the care provider. And that makes sense, right? I mean, tests drive 70% of the diagnosis. So it makes sense that if there is a misdiagnosis, that area would be susceptible.

And the top three diseases that are associated with diagnostic error are cardiovascular, infection, and cancer. And cancer is primarily where anatomic pathology comes in. And we're the final word, so we carry an unduly high number of percentage of non-dismissal for cases. If we make a diagnosis, there's a 40% chance that it will go to payment. And we have a very high number of outlier awards, which is more than a million dollars in awards, and it's only second to GYN.

So as I said, we're a low volume field relative to lab medicine. Lab medicine is about 7 billion tests a year. So if you think about it, one test per population or person in the world. Anatomic pathology is just 350 million tests in the US annually, which is one per person in the United States, more or less. But we're a very high impact field. We're a final word for certain high impact diagnoses, primarily cancer. And that's why we have this God complex.

So this was a painting — I went into the oncologist's office and a patient had made this painting, so I asked her if I could take a picture of it. So because we're a high impact field, it stands to reason we're a high penalty field. The misdiagnosis is devastating. If we inaccurately make a false negative diagnosis, the cancer is missed, it progresses, it can sometimes result in death.

If we give a false positive diagnosis, that is also bad because it results in very toxic therapies for the patients. Cancer is an aggressive disease, so the patients either get additional surgery that they didn't need with attendant side effects or potentially chemo and radiation therapy, which might have long-term effects. Even an ambiguous diagnosis is problematic. So if I say atypical cells, or I can't make a diagnosis, that also imposes a penalty on the patient because the patient ends up having to repeat invasive procedures because most of our biopsies require an invasive procedure.

Not advancing. So it's important to note how and where errors occur in anatomic pathology in order to be able to decide how we are going to reduce them. So this is the test cycle that everybody is familiar with, with the pre-analytic, analytic, and post-analytic phase. And I've also put in the five part cycle that some people use. So there is a pre-pre and a post-post added to the cycle. And this is because it includes whether the diagnosis is communicated to the patient.

Most post-analytic analysis will stop at, did the physician get the report, and then it becomes the physician's problem. And the pre-pre phase is basically in the clinician's head. Did they select the right test? And while that might not directly be our problem, if you look at the health system as a whole, we are perhaps situated to address that overuse or inaccurate utilization.

So in lab medicine, pre-analytic errors are responsible for about 70% of all errors. In anatomic pathology, it's stated to be 40%. In our lab it's actually 15%. So why is that? Well, primarily because we're a small field. So most of our specimens generate from two or three types of sites. So the OR, some surgical physician offices, whereas lab tests are done in every office, every floor, everywhere. So it's just lower opportunity for error.

The second is we don't have that selection of tests and ways to preserve tissue that you do. It's basically either fixed or unfixed. And so both of those make it a little bit easier for us to deal with. Our biggest problem was with specimen loss. At our institution, we have been able to reduce that immensely by basically putting in a computerized order and tracking system. And obviously, the investment was made because we had lost some critical specimens.

Now lab medicine analytic tests make up only about 15% of the errors, whereas in anatomic pathology, it's 30% and potentially even higher. And then post-analytic errors in lab medicine are only about 15%, whereas ours are about 30%. In our lab, they are around 45% and maybe higher. And again, that's because there's greater complexity to the results that we issue. And also, pathologists initiated additional testing, which then will inform the original diagnosis. So to my knowledge, we are the only test field that issues amended diagnoses because additional test results came back. We try to give an integrated diagnosis and so the loop is a little bit more complicated in that sense.

So, one other thing to think about is that a number of — there is disagreement amongst experts as to where analysis starts. So, a number of labs will classify grossing errors as pre-analytic. I disagree. I think that once the specimen has reached a professional's hand, like once the specimen has been accessioned, the test has begun, and so it starts with grossing. And in our lab, about 70% of the errors are lab grossing errors and about 30% are interpretive errors.

So digging a little bit deeply, lab processing errors happen for a couple of reasons. Our process is very manual and it has many hand-offs. So here is a uterus with a cancer. So, a pathologist's assistant or potentially a resident will gross this tissue and then the tissue will be loaded on a machine, and then it will make a block. So, the embedding will be done by a histotech, and a third histotech might cut it and make it into slides. So as we know in quality, more the hand-offs and more manual, the more likely the incidents of error.

The highest penalty errors in this setting are contamination errors. So what do I mean by contamination error? So this is a case that we had. And tissue from this case was transferred onto — so this is an endometrial curettage specimen — and tissue from this case transferred into another case. And this second specimen is an endocervical curettage specimen, and histologically, there is no way to know that the tissue doesn't belong to the patient.

Discovering these errors, I feel, has become harder over time because we've subspecialized. When we were generalists, the specimens were cut in different orders. You saw a GI specimen, and then you saw a GYN, and then maybe a lung. So it was a little bit easier to say, OK, this tissue didn't come from this case. But now that our breast pathologist only looks at breast cases, if there's a breast to breast contamination, you're just not going to know, at least not on histology.

These type of errors are more likely to occur with cases that what I refer to as litter fragment biopsies. They're just like litter, these tiny fragments of tissue and one piece can be transported there to another case at any point in a very manual process. The result of this is essentially an identity error. So the diagnosis of cancer might issue to the incorrect patient. Patient ends up having a surgery that they did not need. And in many cases, it might not be known because the slide has true tumor on it. There's no way to know whether it belongs to the patient or not.

And so some of these cases, I think, these are false positives. But we have no good way of knowing how many false positives we have. To my knowledge, there isn't literature looking at this. There is some literature with respect to what is called provenance testing. There are commercial companies that offer patients — so they take a swab, buccal swab. The patient takes their own swab and then it is sent off to the company and stored there, and then the biopsy is done. If the biopsy turns out to have cancer, then you do provenance or identity testing and the tumor tissue should match with the patient's normal tissue.

Most of us don't use these type of testing. In part because they're not reimbursed. This would be considered quality improvement or quality maintenance by CMS. Plus, it would be really tedious to do it for every single specimen that we do. So as of right now, I think it's primarily education and insisting that people follow protocols recognizing that education and telling people to do certain things when they're working at a rather rapid pace is not very effective. Education is low in the hierarchy of controls to reduce errors.

The second error that we deal with is an exhausted biopsy sample. So we're in a peculiar time, right? So what's happening is that the instruments to do biopsies are becoming smaller and smaller. So, we're getting tinier and tinier tissue fragments, but the amount of testing for that tiny tissue fragment has increased immensely.

So as an example, when we were doing brain biopsies, we treated them as regular biopsies, and we would get three cuts on every biopsy. But now, since 2016, all brain tumors have to get a molecular diagnosis and we issue an integrated diagnosis. So we reviewed our process and we ended up deciding — we only cut a single section on our brain biopsies, and when we cut that single section, automatically ten unstained slides are produced because if the diagnosis is malignancy, then molecular studies can be activated. So this way we've been able to avoid what would happen in the past, which was a repeat invasive procedure.

Now interpretive errors are the ones that cause the greatest amount of tension and difficulty, and they fall into two categories. The first one is what we call a perceptual error or put another way, I missed it. My computer's stuck again. So here is a case, again, this is what I call the litter fragment biopsies. You've got thousands of fragments, and if one tiny fragment is cancer, humans miss things. And so you would miss it and of course, humans are subject to all sorts of cognitive pressures.

And so looking at this biopsy at 3 o'clock, you're more likely to miss it. It happens very infrequently so you're not looking for it. I mean, it's like those luggage checks that we go through when we're going through airports. Those people are really good at finding water bottles because that's what they find most of the times. They're actually not as good at finding guns as they should be, but that happens very infrequently. So sort of like that, we miss them, if it is a single fragment. I think that this is where AI will be really useful because computers don't tire.

A second type of perceptual error, which is perhaps even more difficult is when it occurs within the tissue itself. So the tumor itself is very sneaky. So here is a case of a gastric biopsy and the tissue looks inflamed. You can see there's chronic inflammation here. And actually this case did get signed out as chronic gastritis, but was reviewed and the other pathologist was a little more concerned about it. And so they did a keratin stain and all those brown dots are cancer cells that are infiltrating between normal cells.

And this was a very unusual case. The patient had breast cancer, lobular breast cancer, but she had not yet presented — this was her presentation of breast cancer. The vast majority of our inaccurate diagnoses are where an accurate diagnostic label was not applied to the case. So most of these are interobserver variability or variation amongst experts. But these still qualify as errors because they can result in a different treatment depending on who saw the case. But the patient either has the disease or doesn't have the disease. And so these do qualify as errors. They may not be errors that would meet a legal definition of error, as in a breach of standard of care, but they are errors from a patient's perspective.

So why does this happen? This happens because the natural history of disease is a continuum. So you've got the yellow, which is benign and change is occurring until you get to cancer, which is the red. But treatment is bend. Do something, don't do something, watch. And so we draw these lines based on risk. But if anybody can — something here versus something here would basically result in interobserver variability and potentially will result in intervention, depending on who saw it and not intervention in another person.

A very small subset of errors occur because there's a knowledge gap or failure, as in I don't know. What I don't know, I don't see. Most often this happens when somebody who's trained in one specialty is dealing with a lesion that generated in another specialty, and so they're not familiar with it. The vast majority of errors that we see are actually communication errors. For us, it's about 70%. And they fall into a few categories.

The first one is delivery errors. So report did not get delivered to the ordering physician. This is no longer a concern for most institutions. High tech required that to certify any EHR, they had to have a foolproof mechanism of making sure that the result was delivered to the physician. But that does not mean that the physician looked at it, so that's the problem. The problem is no longer electronic delivery, but did the physician actually look at it.

The other problem is with complexity of care, the report was not received by the treating physician. Joint Commission says that the report should get to the responsible and active care provider. Who do we send the report to? The interventional radiologist did the biopsy and our report goes back to the interventional radiologists. We're done. I mean, we delivered the report, but the patient's left the interventional radiologist who was just a proceduralist, and gone on to another physician. Did the report follow the patient? We don't know.

In the end, the important thing is that the cycle stops with the patient. And of the cancer malpractice cases, 16% of those cases are because the cancer diagnosis was not communicated to the patient. So this is our sort of reporting cycle. Patient comes to the physician, is sent to an interventionalist who does a biopsy, and then they throw it over the wall to me. I catch it. Then I send maybe a piece of it off to the molecular pathologist.

In the meantime, I throw back a report which goes back to the radiologist, who hopefully sends it back to the physician. But now the molecular — And well, the physician now refers to the patient to the oncologist. In the meantime, the molecular pathologist sends me something back. So I change my report, which I again send to the radiologist, who hopefully sends it to the doctor, who then hopefully forwards it to the oncologist. So this is a very convoluted report delivery process that we have. And so many hand-offs, so errors are likely.

The second type of error with respect to communication are comprehension and integration errors. So we have such long reports. Our breast cancer reports are 12 pages long. And this becomes really complicated when you are using the EHR because the HER — our reports are basically, we just repurposed our paper reports and put them into the computer. Looking at 12 pages on the computer is very different from looking at 12 pages of paper in your hand.

The other problem that the EHR creates is that people just cut and paste this. So one clinician cuts and pastes it into their record, and the subsequent clinician gets the result from that cut and paste rather than getting it from the actual pathology report. Also, the reports are truly incomprehensible to the clinicians. Every specialty has developed its own terminology. So this is a paper that my colleague did. This is when — I forget who was from Mars and who was from Venus — book had come out and surgeons misunderstood the pathologist's reports 30% of the times and this has been validated by other reports.

There is also the problem of inadequate clinician education. Mike Laposata has written about it. We've gone to organ based rather than subject based education. And so medical students don't get exposure to lab medicine or surgical pathology to an amount that is needed for patient management. So they learn it just by following their mentors and supervisors, and so there are gaps in the clinician's knowledge.

The other thing that is contributing to this is the explosive increase in information that we now have. So apparently medical knowledge is supposed to double every three months. So we would need an extra two months of the year to know what we need to know. So this is an example– This is Sue and Patrick Sheridan. This is a very well-known case. And I met Sue because I watched a movie. This movie, To Err is Human came out, it was tagged as a patient safety documentary. And this case was discussed, and I contacted Sue and she gave me permission to use this example.

So Patrick developed a neck mass. He went to see his doctor who referred him to a very reputable large institution where he saw a neuropathologist and this was the original pathology report that was delivered. Atypical spindle cell neoplasm. Please see note. Pending neuropathology consultation. The clinicians notes read that I have the pathology report which states that this is an atypical tumor. Very pleased with the outcome. The patient is sent back home.

But then here's the final report. The final report is that the patient has a malignant spindle cell neoplasm consistent with synovial sarcoma. Now if you translate the original report, atypical spindle cell tumor, it means very little to just someone who's using translation of English words. Not typical. OK, spindle cell, it's a tumor of some site abnormal growth of cells. To a general medical person who's not a soft tissue pathologist, it's not atypical. It's made up of spindle cells. Yes, it's a tumor. I knew it was a tumor. Most tumors are benign. Clinically, it was benign. So it's probably benign.

When a pathologist says atypical spindle cell tumor, this is what they look like. We are just freaking out. It's badness. We're praying that it's wrong. And additional studies will prove that it is or is not. And so this is a misunderstood test result. The pathologist did not think they'd issued a final result. They had communicated, in their mind, that they were worried about this tumor, but somehow the clinician understood it to not mean very much.

So the solutions to reduce error, artificial intelligence will certainly address perceptual errors. But their implementation and actually, legal issues as to if the AI machine says it's a gun, but you think it's a water bottle, which diagnosis takes precedence. And if the AI calls it a water bottle and it turns out to be a gun, who is going to take responsibility for that diagnosis.

Another possibility is having cases looked at by a second pathologist. Certainly that would help some, but it raises issues of efficiency. Groupthink. In groups, we tend to think alike. Also, there's some big dog effect where we defer to whoever is the senior most person, but that doesn't make the diagnosis right.

Coordination errors are perhaps addressed by creating diagnosis management teams. And this is something that I strongly support. The equivalent in anatomic pathology is multidisciplinary tumor boards, but there are issues of process standardization and efficiency. And the literature is mixed on whether it helps or not. I do think that the last two are really important. We need to change our pathology reports. It needs to be a lay report that anybody can understand. And we also need to explore direct communication of results to patients.

Basically, that's based on three very important quality principles. The shortest distance between two points is a straight line. hand-offs increase errors. Redundancy reduces the risk of oversights. So instead of doing this, if we did that, that would be straight. So you've told the patient the report followed its routing and also got to the clinician, but hopefully we would avoid patient errors. Some of that is what drove the Cures Act, which was supposed to go into effect in November, now in April. But I still find it problematic because they're going to make reports available on the patient portals, but this doesn't improve integration or understanding of the report.

And I wrote this piece in US News and World Report basically saying that we need to come up with a mechanism where there is direct communication of reports to patients. Because the non-communication and even electronic communication of reports penalizes the most marginalized patients who tend to be more ER users who do not have a single care provider. They don't have access to internet to visit their patient portals.

In New Haven, we're dealing with a situation. We've gone all online for students. 20% of our students don't have reliable internet or computers to access learning. I imagine the parents are even worse off, perhaps. And obviously, it doesn't address comprehension errors. So to my mind, the pathologist and the laboratory professional are critical elements of the patient's diagnosis management team. And they're also the patient's doctors, so we should have direct communication of results.

And to that end, I'm involved in actually a community-based participatory research project where we are selecting the marginalized community to get their idea on how reports are best communicated to them in the hope that we can develop a structure where they do get their reports. And this is sort of being used for COVID testing so the positive results are not communicated by the patient's care provider. But we have a call center that is manned by nurses that is calling those results.

Thank you, and I think I went over time. I apologize. Just when I practice, it was a lot shorter. What can I say?

HENDERSON: No, thank you so much, Vinita. That was exceptional. And really highlighted a lot of points that I know our community and rural participants may have in their head. And I hope you guys are getting your questions ready. But I have a question, and I don't expect you to provide the answer. Then the world would be beautiful. But where do you see — especially now as you mentioned, COVID, you know we've seen a lot of things going on with the marginalized communities. But why do you think that the bridge between the pathologist and the patient has not been built in a lot of institutions and then only partially established in some?

PARKASH: So I find that very interesting. I grew up in India. My chair is from China. And we've actually discussed this. So in other countries, patients get their pathology report from the pathologist and basically then go to see the oncologist. I think in our healthcare system, first of all information is knowledge. So clinicians want to maintain control of that. From the pathologist's perspective, it's a non-reimbursed activity, so sure, I send it to the physician. It's easier to communicate with physicians than it potentially is with patients. So I think that our reimbursement system drives the process to some extent. And in doing so, it penalizes patients, especially marginalized patients.

HENDERSON: Thank you. And we're going to open the floor up now for feedback and questions. But before I do that, I want to welcome our VIP participants who participated in this ECHO session earlier in the year, Dr. Jim Crawford and Dr. Susan Weiss. If you want to see their sessions, you can log in to our website and see them. But please everyone, either you can raise your hand or you can use the chat box. Please remember when you are speaking, introduce yourself, and if possible, turn your camera on. Thank you.

JZ, I need you to help me man the — if not, I can ask another question.

PARKASH: Go for it.

HENDERSON: So in that final part of the presentation, as you spoke about getting to the patient. I know you've done the Fellowship with SIDM, and we've had discussions here of how to engage the patient. And we actually have a very active patient advocate on the CLIAC Federal Advisory Committee. And one of the questions that she always brings up is patients want to be able to read their reports and have access to it. And as you said, it needs to be in a form that they can read. Do you think this is a time where, probably now more than ever, that the patient safety advocates are necessary in helping to establish that bridge?

PARKASH: Yeah, so I think that pathologists and actually malpractice carriers and patient safety organizations need to collaborate on this. Because patients bear the ultimate cost of an inaccurate or lost test result. So they are invested in this. I actually think that malpractice carriers should be pushing for this because they bear the malpractice charges when a result is inaccurate or misplaced. I also think that pathologists should do it because of their commitment to patients.

And with the three groups working together, hopefully there will be a change. With respect to terminology, I think pathologists need to work on that. And I think radiologists have started doing some of the work. So for mammography reports, they have to issue a lay report and a technical report so that the patient can understand it. They're also, I believe for radiology, there are some programs that people are developing that can take that technical report and translate it so that the patient can have access to a lay report.

But that's something that we definitely need to work on. I actually worry that our clinicians have difficulty with our report, like you know– So my gynecologists understand my report, my gynecological oncologist. But I've had instances where a general gynecologist has called me up and said, I don't know what this means. What should I do with this report? And that's on me.

I don't believe — you know, we need to separate record keeping from reporting. The purpose of a report is to communicate with the clinician, with the patient, so that the right next step can be taken. Billing and all of that is record keeping. And I think we need to look to technology to help us separate the two and refine the process.

HENDERSON: That's really good. Thank you for that. Aurora Rock. I saw you put your camera on. Did you have a comment or question? Maybe, maybe not. Dr. Crawford, can I put you on the spot for comments? Is everyone here, but not here today?

PARKASH: Well, certainly in the Northeast, we're expecting a really large storm.

HENDERSON: That's true.

PARKASH: People are trying to get whatever they need to get done so that they can get out. I know that some of my colleagues are trying to get some of tomorrow's work done today because we don't know what tomorrow will be like.

HENDERSON: I'm spoiled. I've forgotten the Northeast already.

PARKASH: Warm weather is the best.

HENDERSON: It is, and I am enjoying. Any final comments or questions from the line? I know we have some of our rural partners from Louisiana on the line. North Carolina. Any comments before we wrap up this pilot project? Vinita has been exceptional. I always love your presentations and your passion for quality improvement, which is also mine. So I definitely admire the work that you do. And this has been an awesome finale for this pilot project.

PARKASH: Thank you.

HENDERSON: You're welcome. Going once, going twice. Let me just check the comments. No.

PARKASH: So even though I went over, we're ending early. That's perfect. Christmas present.

HENDERSON: Exactly. It always happens like that, doesn't it? Everything works out. But to you Vinita, and then all the past presenters on the line and even to the participants, we really want to thank you for participating today. We really hope that you found this important to your work in your individual laboratories.

We first started this project at the beginning of 2021, and we weren't anticipating having to confront a global pandemic. But we appreciate each of you and your continued participation every month. We considered pivoting to COVID. But based on the feedback, a lot of you really wanted the reprieve of still maintaining some type of normalcy in the continuing education that you receive. And so we pushed ahead and continued this pilot project every month since we've gone into COVID everything.

Next year, we will begin analyzing the qualitative and the quantitative data that we've collected. And thank you for participating in the post-session surveys to determine our initiative's future, whether we will continue or will we modify the model. If you'd like to listen to the previous sessions, as I mentioned, we've posted the audio and transcripts on the ECHO website. So Extension for Community Health Care Outcomes Project website.

And if you have any feedback for us, please feel free to reach out to us via email, individually or to dlsecho@cdc.gov. So now we will adjourn. Thank you again. Thank you, Vinita. And happy holidays everyone. Oop, I think I see a question. Oh my goodness. Oh, so somebody would like your email address for direct questions to you.

PARKASH: Sure. It's vinita.parkash@yale.edu. So my first name dot last name at Yale dot edu. And [INTERPOSING VOICES] I know that this is sort of a left-field topic. So I am interested in hearing, sort of the other side, you know how people on the other side think so that we can come up with a solution for our patients. So please feel free to send as many emails.

HENDERSON: Perfect. And that was one of our main purposes for this, bringing you, the experts, to the community when they usually would not have had a chance to meet you at national meetings or other organization meetings that they may not have access to. So we really appreciate your time and your effort and then sharing your contact information and facilitating that open discussion. So thank you, everyone. And hopefully, we'll see you in 2021.

PARKASH: Thank you.

HENDERSON: Take care. Bye.