 Hi, everybody. This is Donna Foster, but the Patient Safety Movement Foundation here to bring you another COVID-19 update. We're here today to talk about the impact that the COVID-19 pandemic has had on diagnostic errors in medicine. We are excited to be joined today by Dr. Hardeeb Singh, who is a professor of medicine at the Baylor College of Medicine and at the Houston VA Medical Center. Welcome, Dr. Singh. How are you today? Hi, good. Thanks for having me. Great. Can you tell us a little bit about your background? Sure. I'm a general internist and a patient safety researcher. I do research on improving diagnosis and how to use technology to improve healthcare. Excellent. Well, we know that you have recently written an amazing paper to talk about diagnostic errors and the impact that this COVID-19 pandemic has had on that. Tell us a little bit more, though, about the history of diagnostic errors in this country and across the world. What kind of medical errors are you seeing happening as a result of this? So, diagnostic errors are one of the most common types of errors and can happen in any setting in any specialty. Our work shows about 5% of U.S. adults will have a diagnostic error at least once every year. And then there was a National Academies of Medicine report in 2015 which said every one of us is going to have a diagnostic error at least once in our lifetimes. It's a very common problem and it affects several processes of care and it involves clinicians thinking within healthcare systems which are themselves very complex. So, it's been a little bit of a patient safety challenge that we've had to solve over the last decade, figuring out how to make things work better in a complex health system and how to reduce errors. And right now diagnosis is getting a lot of attention because of the pandemic because, of course, testing issues have come up quite significantly. And so, what else are you seeing right now in terms of this pandemic? What other impacts on diagnoses other than COVID are you seeing? Yeah, so, you know, the COVID is an interesting challenge because it has exposed healthcare system vulnerabilities. So, not only are clinicians sort of stressed but health systems are stressed, the processes of care are stressed. And we know from patient safety literature when things are stressed around us, defense mechanisms fall and we are more vulnerable to making errors. What the COVID-19 pandemic has done is it's sort of introduced many types of diagnostic errors, some of which are, you know, sort of unique to the pandemic and others are known cognitive fallacies. So, some things that we have known before that can happen. So, in this paper that we just recently published, we describe eight different types of diagnostic errors that can happen as a result of the COVID pandemic. Now, each of these has a sort of a unique flavor and one patient could have multiple types of diagnostic errors and I'll quickly give you an overview of the types of issues that we're seeing. One of the diagnostic errors, perhaps the most common one that we have been seeing is the classic diagnostic errors. We call these classic because we came up with a terminology that everybody could understand. So, classic diagnostic error is, you know, patient comes to you with symptoms that are very suggestive of COVID disease but you're not able to diagnose them correctly either because the testing is not available or the testing is false negative. That means I have respiratory symptoms and I'm hypoxic and I, you know, have trouble breathing but I cannot get a test that confirms that I'm positive. So, that's sort of the classic diagnostic error we're seeing. A false negative test happened. A false negative test happened even before but in this type of error, if you tell me I'm negative, I'm just going to maybe go around and infect a few other people so it's really important to know about classic diagnostic errors. The second type of diagnostic error that we've seen is that of patients who have sort of atypical symptoms. So, patients who present with, you know, some non respiratory symptoms for instance nausea, diarrhea, belly pain, with some trouble with smelling and tastes have come up, anastomia. So, there's some unusual patterns where people who have mild disease just have some trouble smelling and tasting and these could be also COVID. So, we call these atypical anomalous types of diagnostic errors. That means the patient has COVID but the presentation is mostly atypical and we miss the diagnosis because of the atypical presentation. The third type of diagnostic error that we are seeing is something what we call is anchoring. So, for instance, somebody comes to me and says they've got respiratory symptoms but I can get a test done and I think well you must be having COVID but they don't have COVID. So, we've seen instances where sinusitis has been missed, tuberculosis has been missed because it can present to the respiratory disease, pneumonia has been missed. So, you know, just sometimes when we just anchor on COVID but it's truly not COVID. Another fourth type of diagnostic error we've seen is sort of secondary complications. So, we know that COVID can cause other types of complications such as pulmonary embolus. So, somebody's in the ICU, they deteriorate. It may not be COVID that's causing their deterioration. It actually may be pulmonary embolus. We just recently found a lot of these multi-system inflammatory syndromes that have been described in children that we can easily miss and that is also as a result of COVID. The fifth and sixth type of diagnostic error has a lot of them have been in the news recently because we have seen patients where they may not go to the emergency room or to have their procedure if they are really needing one because they are fearful that they will get infected or they don't want to overwhelm the healthcare system. So, we've had cases in which patients have had strokes and heart attacks sitting at home and have not sought attention because of these reasons. Similarly, we've had patients where their test was canceled. They were getting a workout for cancer diagnosis. They were supposed to have a biopsy or a mammogram or a colonoscopy but that got canceled and they didn't actually have it done. So, that's another reason. So, we call these acute and chronic collateral types of damage errors. That's the terminology that we came up with. Another type of error that we are seeing is that of strain. So, when the health system is really strained, you pay more attention to maybe just the COVID patients and you sort of don't see as many or don't diagnose the non-COVID patients accurately. Imagine the work systems, you know, we've had to deploy a lot of physicians from other specialties to take care of patients in ICU or in hospitals. So, we've heard of places where patients are being taken care of predictions or surgeons who may not be really attuned to delivering care for a sick hospitalized patient with respiratory infection. So, those strain-related errors also happen. And then lastly, the unintended consequences. So, you know, telemedicine has introduced some distance between us and our patients and of course we may not be able to examine patients the same way. You probably also know that about 70% or so of communication is non-verbal communication and we lose that when we are behind masks and are not able to sort of, you know, talk directly either through, you know, we're on the phone or on the computer or maybe behind a mask and we're not able to sort of communicate properly. So, these indirect or in unintentional errors are happening and we call them unintentional errors due to use of, you know, telemedicine or PPE. Again, many of these are anecdotes and we are just starting to describe these errors. So, I think it's good for health systems, clinicians and patients to know about them so you can actually do something, you know, and be aware and sort of find these errors and do something about that. Well, that's a really great point because we do want to do something about it, don't we? What recommendations do you have? I mean, I'd like to, you know, if you could address both the clinicians and the administrators and what they need to do about this, but then also, what patients and families need to do? What can they do to keep themselves safe? So, let's start first with clinicians and administrators. What can they do to improve their processes and their systems? Yeah, I think one of the things is there's no single bullet. I like the way you sort of, you know, called out all three clinicians, health care administrators, patients, and I'm sure there's others as well that can, you know, help out because there's not a magic solution, but if all of us sort of work together and everybody sort of does something to prevent it, it's just the defense that comes up, I think we'll be able to prevent a lot of these errors. So, for clinicians and health care administration folks, I think it's really important to find some of these errors and learn from them, but we just don't have very good systems of measurement to try to identify what happened, why did it happen, and what can we learn from it. I want to know about all of these telemedicine and use this diagnosis. So, if I have, I just recently heard about a case where one patient got three different diagnosis from three different dermatologists for the same rash, and we know somebody else whose diagnosis was missed and they had a bad stroke that was not caught through, you know, when the initially were on telemedicine. I want to know about them. It needs to go beyond just anecdotes. So, I like the health systems to sort of build the capacity to identify and learn from these errors. The other things is, you know, we know that systems are vulnerable, so as clinicians I think we need to sort of ask for help more often. In our previous work, we have shown that clinicians, even in the midst of a lot of uncertainty and difficulty, they may not actually ask for help. And it's essential in times like this when we're, you know, relatively compromised because of either expertise or the health system capacity that we ask for help. So, you know, having sort of buddy systems for clinicians who are not used to giving care in a certain environment, it's a good idea. Asking for help is a good idea. You know, using technology in your normal healthcare resources that you use, information resources, is really a good time. Yes, we know that it's stressed and it may not be in somebody's workflow to keep looking up things of good knowledge resources, but that's where teams have been useful, right? So, healthcare teams, working in a healthcare team and having a team, good sort of a team around you to help you get the right data is also very useful. And I think there's other routines, sort of strategies, the sort of processes of care that are important. I think sharing data, creating sort of learning systems around this. So, if let's say hospital A has figured out how to sort of address some type of an issue, I think we need to have better learning platforms where we can translate, you know, best practices from one system to another, because we just don't do that very well in healthcare and patient safety in particular, and we need to sort of do that as well. Excellent. And, you know, last week we just interviewed a PhD from Switzerland who is working with patient safety and quality personnel to figure out what they can do during this pandemic to bring in real-time change. So, this is a great example of where they can be collecting data right now to identify how this is impacting hospitals. Yeah, absolutely. And I would say that, you know, I think you also talked about the patients, you know, what can patients do? I think it's really important for patients to realize, you know, the hospitals are open and the emergency rooms are open and if they have symptoms of concern, they need to seek help. This shouldn't be a time when, you know, if you're having a chest pain and, you know, going to your arm and you know what that might be, don't think, you know, that this is a place, this is a time to sort of stay home because the hospital is overwhelmed. You still need to seek help. And again, I think if, if sort of patients can also, you know, when things go wrong, tell us when they're wrong. Through telemedicine, follow-up or other ways, there are ways to collect data from patients also to learn what's, what's going on. Great. Well, this has been very informative. Thank you so much, Artie, for joining us today. And we're going to share your, your paper on our website and on our YouTube video and we will make sure that we have the link in this video as well so that people can access it easily. Thank you. Thanks for having me. Thank you. We appreciate it and I hope that we can have you back again in several months to find out where we're at. Absolutely. Looking forward to it. All right. Well, thanks so much for joining us and you have a great day.