Podcast: Misdiagnosis during the COVID-19 pandemic
Transcript
Advertisement Are you looking for your next adventure or for your next place to call home? Come explore British Columbia with the help of Health Match BC. We're a free health professional recruitment service funded by the Government of British Columbia. We're currently recruiting for physicians of all specialties on behalf of BC's publicly funded health employers. If you're a physician or other health professional looking to make a change, we can help. Visit us at www.healthmatchbc.org for more information, and to speak with one of our recruitment consultants.
To shingles, age isn't just a number. Do you have patients 50 or older? They're at higher risk of getting shingles. Don't wait. Talk about shingrix with your patients over 50 today. Shingrix is indicated for the prevention of herpes zoster or shingles in adults 50 years of age or older, consult the product monograph at gsk.ca/shingrix/pm for contraindications, warnings and precautions, adverse reactions, interactions, dosing administration information. To request a product monograph or to report an adverse event please call 1-800-387-7374. Learn more at thinkshingrix.ca.
Dorian Deshauer: For the past few months, COVID-19 has been dominating the news cycle and our social media feeds. It's also been front of mind for physicians, especially those responsible for diagnosing it while also trying to keep up with protocols and guidelines. With so much talk about the disease, is COVID-19 distracting? Is it affecting the way physicians diagnose? Is it leading to diagnostic error? I'm Dr. Dorian Deshauer, deputy editor for CMAJ. In today's episode, we'll try to answer those questions. You'll hear from Dr. Justin Morgenstern, an emergency doctor who has spent a lot of time analyzing physician decision-making. He digs into the many factors that influenced the way doctors diagnose, including cognitive bias. But first up, you'll hear my conversation with two physicians, Dr. Alex Kobza, and Dr. Brandon Budhram, who describe how they misdiagnosed a patient back in April of this year. She presented with classic COVID-19 symptoms, but her diagnosis turned out to be something else entirely. And it took the physicians quite a lot of time to get to the correct diagnosis. Welcome to CMAJ Podcasts.
Brandon Budhram: Thanks for having us Dorian.
Alex Kobza: Thanks so much for excited to be here.
Dorian Deshauer: So can you tell us a little bit about yourselves and where you work? Starting with Alex.
Alex Kobza: My name is Alexandra Kobza, and both Brandon and I are second year internal medicine residents at McMaster University. We're really thrilled to be here today.
Brandon Budhram: My name is Brandon Budhram. And not only are we co-residents in the internal medicine program, we're also the co-leaders of the patient safety committee here at McMaster University. And it was really in that capacity that we became interested in this case.
Dorian Deshauer: Well, let's get right into it. You've written up the details of a case that's really timely in this COVID pandemic. So could we start off by telling our listeners about the patient who was admitted and what information you had about her?
Brandon Budhram: Yeah, so our patient was a 40 year old female who is otherwise quite healthy. She presented to the hospital in April of 2020, with a 10 day constellation of symptoms that included mildly productive cough, low grade fever, fatigue, and shortness of breath on exertion with oxygen saturations that were going to the high 80s. Now notably, she didn't have any COVID contacts or recent travel that we listed. She was, however, a personal support worker with exposures to long term care facilities. Now, as we know, many of these facilities were hit quite hard during the pandemic, but fortunately, her workplace did not have any outbreaks at that time. And that's all the, you know, the relevant information we acquired on the first pass.
Dorian Deshauer: And just with that basic information, were you suspecting COVID?
Alex Kobza: Absolutely. So at this point, COVID-19 was certainly the leading diagnosis. Now, I do want to bring your attention to Brandon's earlier point that the patient presented in April of 2020, which many of us will remember what's really the peak of COVID-19 cases so far in Canada. And if you recall, during this time, the sheer quantity of COVID-19 related information, both with respect to policies and medical literature, was just really overwhelming. I think that the pandemic inspired a significant amount of fear related to the diagnosis of COVID-19. And it's probably safe to say for both ourselves, and many of our colleagues that we had some degree of tunnel vision.
Dorian Deshauer: So here you have this person you're suspecting COVID, you're putting on your personal protective equipment and you decide to order some tests. So what do you do next?
Brandon Budhram: We did what you would expect in terms of the standard battery of tests when she was admitted. So that included bloodwork, chest X ray, and a nasal pharyngeal swab (NPS) looking for multiple respiratory pathogens, obviously, including SARS-CoV-2. The bloodwork had multiple findings that were compatible with the diagnosis COVID-19 including the significant leukopenia, low albumin and elevated inflammatory markers, including the CRP and the LDH. These markers that we commonly associated with COVID-19, the chest X ray, further pushed us in that direction showing bilateral interstitial markings. But it should be noted that while these investigations were compatible with the diagnosis of COVID-19, and we put a lot of emphasis into them at the time, they're incredibly nonspecific tests. And then lo and behold, our first NPS was in fact negative.
Alex Kobza: I can kind of tell you what our next steps were. So as with many of these undifferentiated patients, she was initially treated empirically for a community acquired pneumonia with levofloxacin. But COVID-19 was definitely still at the forefront of our minds. After a couple of days, she showed some minimal improvement. So she was no longer febrile, although she was receiving Tylenol, and her oxygen saturation had improved marginally maybe from 89 to 92%. So at this point, we performed another NP swab, as well as the COVID specific sputum PCR, both of which came back negative. So at this point, we had three negative confirmatory tests, yet COVID-19 still led our differential diagnosis.
Brandon Budhram: Yeah, you know, we didn't really appreciate it at the time. But it's important to note that our instituted McMaster PCR assay demonstrates pretty phenomenal sensitivity and specificity, both on the order of over 95% when there's adequate technique. So to your point, you know, we had three negative PCRs. But we still suspected COVID-19, and we still had to explain her symptoms or bloodwork or chest X ray. So we we ordered a CT scan of her chest. That CT scan showed extensive bilateral ground glass opacities, which again to us was compatible with COVID-19.
Dorian Deshauer: So I was just wondering, at that point, were you starting to get reports about false negatives? In the nasal pharyngeal swabs? I can't remember the timeline of all this unfolding and the flow of knowledge around testing.
Brandon Budhram: Yeah, so the thought process, at least on our end at that time, was that you can either get a sample from the upper respiratory tract or the lower respiratory tract. Obviously, the NP swabs are upper respiratory, the sputum PCR that she produced was pretty good quality, and that would be a lower respiratory tract sample. But our thought was that, you know, if we really want a great sample, we need to get the bronchoscopy done. Like she's otherwise healthy, she'll tolerate a bronc. And we asked her respirology colleagues, and they kind of agreed, so that was our next step.
Dorian Deshauer: So it sounds to me like you really wanted to nail down the diagnosis, you weren't willing to say probable COVID based on what you had. You really wanted that lab diagnosis.
Alex Kobza: Exactly. It was, it was pretty unsettling to have this previously healthy woman coming in with quite severe symptoms, and to not have an explanation for those symptoms before we sent her home.
Brandon Budhram: Yeah she had the bronchoscopy. She actually tolerated it very well. And then she requested to be discharged after the procedure. She wasn't 100% better by any means. But she was discharged in stable condition. And we did ask her to just self isolate pending the results. We were really hoping to get kind of this clear cut diagnosis of COVID-19 onbroad, because that was the goal.
Dorian Deshauer: So basically, you sent her home thinking yes, we're gonna get the diagnosis, she'll stay home and self-isolate. And hopefully this will resolve and suddenly you get this really unexpected result.
Alex Kobza: Exactly. So I wish we could say that we started suspecting an alternative diagnosis, maybe at the time of the third, negative PCR, or ideally even the second, but really, I we can't say that at all. So just like we were saying, in all honesty, COVID-19 led our differential diagnosis until that bronchoscopy came back, and it showed, to our surprise, Pneumocystis jirovecii. So at that time, the diagnosis of PJP completely was not expected. But once we had that confirmatory test, all the dominoes kind of fell into place, and the picture was really quite clear.
Brandon Budhram: The next steps after that were obviously, we call the patient back to look for an underlying immunosuppressive condition naturally thinking about HIV. And then I was surprised to see what I found. I was actually on-call that weekend that she returned to the hospital. And at that time, not really thinking about COVID-19 anymore, we were able to do a much fuller and more comprehensive history and physical exam. That new exam and just overall assessment revealed this 50 pound weight loss over the previous year, intermittent night sweats. It's kind of like purple, painless lesion on your chest that looked like classic kaposi sarcoma, and an active HSV, two general herpes infection. She also disclosed these high-risk sexual activities 5 to 10 years prior. And unfortunately, like none of this information had been identified at the time of her initial admission when COVID-19 was the prime suspect. You know, of course, had it been identified, it would have drastically changed the trajectory of her state. Now, not surprisingly, first of all, she did come back confirming HIV. Her CD4 count was actually 10 with viral load of just under 700,000.
Dorian Deshauer: So you had the the diagnostic side of it, but what about the like, on a personal level? How did you feel when you got that result?
Brandon Budhram: Yeah, I mean, it wasn't a great feeling Dorian. It was certainly a humbling experience. There are multiple cognitive errors at play during this case, but the most important one I want to emphasize was premature diagnostic closure. This represents the failure to consider other diagnoses once an initial diagnosis is proposed, which in our case was obviously COVID-19. Now, this has been well documented as one of the most common causes of diagnostic error in internal medicine, and then our premature diagnostic closure then open the gates for other diagnostic errors, including the incomplete history and physical exam, the overestimation of the nonspecific laboratory and imaging planning, as it related to COVID-19.
Justin Morgenstern: Now the clinicians here had COVID as their number one diagnosis, and whether that's a great diagnosis depends a lot on the community prevalence at the time. And you know, I don't know a lot about the specific neighborhood. But in general in April 2020, a PSW with exposure to long term care facilities, a fever, a cough, bilateral ground glass opacities on a CT, well, COVID had to be on the differential. And you know, you might debate but I think they were probably right to put it right at the top of the differential. So I'm Justin Morgenstern. I'm an emergency doctor from the Toronto area. Actually, I probably spend a lot more of my time outside of the hospital working on things like medical education, podcasts, and blogs, all centered around my website. As I watched through medical school and residency, I sort of got the sense that what might differentiate an average doc from a great doc is sort of what they do with those mistakes. And I'll tell you, I've started my career as distinctly average. And I realized, if I didn't study my mistakes, and maybe study the mistakes of the people around me, if I didn't try to learn from them, I was destined to be making the same mistakes over and over and over again. And when you think about it, that's pretty depressing. So it was that thought that really got me reading a lot about how our brains work, about the mistakes we make and sort of hopefully how we can avoid some of those mistakes. Kudos to these authors for being willing to talk about their mistakes in public. That's an essential part if we're going to get better at this if we're going to learn from each other. But what I think is a really interesting component of this case, is even though the authors think that they may have made some mistakes, and they really got focused in on COVID, and that might be a mistake, they still weren't satisfied with that diagnosis, they still went ahead with bronchoscopy at a time when bronchoscopy is really difficult to do. It's an aerosol generating procedure. And they did it in a patient who was actually well enough to go home and wait for the results. You know, if it was me, I could picture myself just waiting, right? Assuming this was COVID. And assuming that there was nothing else that I can do, they didn't close off their minds, they kept searching for a diagnosis. And to me, that's key. Even if cognitive errors are made, as long as you keep your mind open, you're always going to be able to recover from your errors. Early in COVID, these doctors had to deal with so much uncertainty, you know, we didn't know how accurate the nasal swabs were, we didn't know how accurate CT was, we didn't know all the symptoms that COVID might present with, we didn't know how many people in the community were actually sick. I mean, we still don't know a lot of those things. But how can we blame a clinician in April 2020, for misinterpreting the meaning of ground glass opacities on a CT, in the context of all that uncertainty. So then I definitely think there's some stuff that you might call a no fault error. And then you get into the system of medicine a little bit more. I mean, I don't know about you, but I've never spent a day at work, where I didn't feel busy, where I didn't feel rushed. And during COVID, you know, we that's our baseline as doctors, but think about the, you know, hundreds of extra hours of meetings and the hours of writing and learning these new protocols that came out, I think we were all stretched pretty thin. And although it's really important to learn about COVID as it came out, I think we underestimate the secondary effects of, you know, releasing a new protocol every three or four days from, you know, distracting us a little bit from the patient in front of us potentially. I think there's a large number of ways we could think about system errors. You know, we were all very, very worried about PPE. I don't know what was happening in their hospital at the time, but if they didn't have adequate PPE, well, then maybe that's a system error where you're not gonna be able to do a full exam. I bet you a lot of us were actually in the opposite. Maybe we were actually safe. We just felt anxious or rushed, in a way that people had not made us feel safe to see these new patients think the important thing is so often when we see error, we point fingers and blame individual doctors and yeah, all of us really, really need to try to learn and improve constantly, but that won't be good enough. If we don't also at the same time, try to improve the systems within which we work. Actually, honestly, for the most part, the human brain works great, I think about how we work, we work in a chaotic world full of uncertainty with bizarre presentations and incomplete information. And physicians still make the right diagnosis the vast majority of the time, but these errors are out there. And I think we do need to try to improve. So in my mind, there's sort of four key pillars that I tried to do at least. So the first would be, you know, there's a very long list of known biases, known cognitive errors that have been coming out of the psychology literature for decades. And I think we should, at least at bare minimum, learn what they are, you know, you're never going to be able to avoid confirmation bias, if you don't at least know what it is. Second, I do think it's important to spend a little bit of time thinking about our thinking, but there's a fancy term for that metacognition, but the basic idea is, you know, you have a patient with chest pain in front of you, you decide that patients having an MI (myocardial infarction). You got to take a step back and ask yourself, you know, what about the case made you think it was an MI? Was your logic sound? Are there any parts of the case? That don't really make sense? And that would lead me to my point number three, which is, you always have to ask, what else could this be? And maybe more important than that question is ask yourself, what doesn't fit? Why am I wrong, because once you decide on a diagnosis, it's way too easy to get locked in on that diagnosis and stop thinking. And a key component of scientific thinking is always looking for disconfirming evidence, don't try to prove yourself, right, try to prove yourself wrong. And all of a sudden, your logic, your thinking is going to be a lot stronger. And then finally, I think we need to make it very, very comfortable in medicine to talk about our mistakes. You know, doctors want to be perfect. Errors almost taboo in medicine, but we can't learn from our mistakes. We can't improve unless we're willing to admit that we've actually made some mistakes. You know, it's always better to learn from somebody else's mistakes, so you don't have to make it yourself. So once again, I'd also say kudos to these authors for being brave enough to talk about their mistakes publicly in CMAJ.
Dorian Deshauer: Maybe a question for both of you? How do you see authoritative medical information that you can trust like getting into the medical practice? And how does that interact with other kind of rumors and less reliable information that might be flowing either on social media or in a popular press?
Alex Kobza: You know, the information that's flowing from social media or the news is it's much more easily accessible. And I find that sometimes when I hear that, I wonder, Is it because I've missed something in the literature? Or is that truth? And it's kind of hard to find, you know, a good balance of what the accurate information actually is.
Brandon Budhram: Yeah, that's a, you know, that's a great question. Because I've even noticed in my personal life, you know, we have information that flows from multiple directions, like you commented on, like, we can get some in the workplace in the sense of like, from our staff, physicians, and from the hospital themselves. But then when you go home, depending on you know, who you are, and where you live, you get this kind of huge amount of information from social media, from your friends and your colleagues. It's extremely hard to streamline this information. I think in an ideal world, you'd have one source of information that is reliable, and I think that should probably come from the hospital setting. But it's really hard because it's convoluted with other information you're getting. I think at that time, in April of 2020, information was coming kind of from all directions. And personally, I wasn't sure at least, I didn't feel sure about what the, what the guidelines are, what we were supposed to be doing for these query COVID-19 patients.
Dorian Deshauer: What do you think this is teaching you? Or what are you learning from this in terms of the role of experts in society, and I mean, not just in medicine, but the role of the expert?
Brandon Budhram: I mean, for me, personally, it taught me that I need to just acquire one source where I can or you know. A few sources that I can streamline information. And then you know, more importantly, it taught me that this kind of thing will happen, like diagnostic errors do happen. And throughout our medical careers, both Alex and I appreciate that we're gonna have to use this opportunity to learn for the, for you know, for the rest of our medical careers, and a sense that we won't make this mistake again. And we'll kind of be more aware when future situations like this arise.
Alex Kobza: I also think that when communicating with the public, I found that having fewer people commenting on the state of the pandemic and what people should be doing was helpful. And particularly having people in health care kind of make those recommendations I think, gained, buy in from the public and people's trust and having kind of one unifying message was the way to go.
Brandon Budhram: And Dorian, I think we can all probably appreciate that, say in April of 2020, compared to now, the month being September now, like we feel much more comfortable with these patients that come into our hospital when it's suspected COVID-19 just because we have more experience with it. We kind of know more about the literature and the policies, things are not as rapidly changing. So you know, that improves our comfort level in a situation. It's not like in April of 2020. Like I don't know that this same mistake may have happened again.
Dorian Deshauer: Thanks so much, Brandon and Alex for taking the time to talk to me today and share this case with our listeners and our readers.
Alex Kobza: Thanks so much. It was great.
Brandon Budhram: Thanks for having us, Dorian. And I just want to, you know, echo the sentiment that we're extremely thrilled to be part of this and we want to thank you for having us.
Dorian Deshauer: Well, you're very welcome. I've been speaking with Dr. Alex Kobza and Dr. Brandon Budhram, second-year internal medicine residents at McMaster University in Hamilton, Ontario. They wrote a practice article with Dr. Naufal Mohammed. The article is published in the CMAJ. And you can read it on our website. You also heard from Dr. Justin Morgenstern, emergency physician in Toronto. And don't forget to subscribe to CMAJ Podcasts on Soundcloud or podcast app. I'm Dr. Dorian Deshauer, deputy editor for CMAJ. Thank you for listening.