A Cognitive Autopsy - What's Wrong with Me?

SPEAKER_01 0:11

Hello and welcome back to a new episode of Home of Medicine Podcast in association with the Royal College of Physicians Edinburgh. My name is Ben Lovell. I am an acute medic and I work in London in the UK. And as usual, I'm joined by my co-host.

SPEAKER_02 0:26

I'm Amy Burbage, a consultant in acute medicine.

SPEAKER_01 0:30

And I'm really excited. This is our second episode with an amazing guest. Please go back and listen to the previous episode where we spoke to Professor Pat Croskery, who is an expert in clinical reason. And our first episode we did actually gave a really fantastic overview of a lot of issues about biases, thinking errors, and metacognition, which affect us on a day-to-day basis. And we're super excited to have Pat back with us again for one more episode where we're going to be talking through a case that really illustrates a lot of thinking errors and biases and facetes of clinical reasoning that I think we can all resonate with. So thank you so much, Pat, for being back with us today.

SPEAKER_00 1:12

Great. You'll be pleased to hear. So the case that I have is is uh actually from one of our books that um and it involved a uh I think she was about a 25-year-old woman. Uh, she came into the emergency department at about um quarter to 10 in the evening. And I mentioned that because this particular department used to close its doors at 10 and then refer all the patients on to the to the sort of mother hospital in town. So um that constraint uh I think had something to do with it, that we were short of time. So the patient arrived at about uh 9:30 or so. We were going to close at 10. The nurses were anxious to close the department at 10. So there's that little pressure to move quickly with the case. And the the woman's story was that uh, and she was with her husband who was very concerned, um, she had delivered a healthy baby about a week earlier, uh uh uh less than a week, uh, five or six days earlier. And her presenting complaint was that um she wasn't feeling well. And when you're trying to get her to be a bit more explicit about this, she said her body felt strange, that there was something going on, and she started crying, she became very tearful, uh, and uh and it was difficult to get uh uh any kind of quick history from her, uh uh, but just to let her kind of um um elaborate a little if if we could, and that's what I tried. And and sort of, can you describe what sort of feelings exactly are happening? Do you have pain anywhere? Is there any part of your body that hurts? Do you have an ongoing um uh wound that's healing or anything else that I should know? And she uh just got seemed to get increasingly upset and tearful and saying, What's wrong with me? I just don't know what's wrong with me. And every effort to try to get her to be more explicit wasn't working. And uh some of the nurses who were hovering around were getting uh were getting um um increasingly uh irritated that that that this wasn't working out quickly. And um uh I I actually had to leave the bedside for a short time just to pick up something. And one of the nurses grabbed me quickly when I when I went out and said, You know what this is, don't you? Uh and I said, No, I don't yet. And she said, She's got the baby blues, she's got postpartum depression. Or or she didn't say postpartum depression, but she has she said the baby blues. And uh I was familiar with the term. And and there were several other nurses in the background hovering there, and they said, Yes, the baby blues, and all of them were mothers, and they'd all gone through this process, and they were quite familiar with this sort of postpartum um mood change that can occur uh and so on. Uh um I uh my antennae went up a little bit because I could see what was emerging here. There was a bit of a rush to get her out. Um, and uh but the other thing was the I remember very clearly was the unanimity of the nurses, all mothers themselves, about four of them, all saying, like a chorus, this is uh postpartum depression or baby blues. So um I tried to go back to basics, uh, tried to review her medical history, when did it start? Uh did she have any significant history? And so on. When did this start? This had been coming on for about uh 24 hours, she thought. Her husband was entirely supportive. He was holding her hand, he was uh he was uh extremely concerned about her, and so she said, I just don't know what's wrong with me. I feel terrible and and crying. And uh uh so I began uh an examination, just going over her from head to toe to see if I could find anything as a starting point. And I went over her completely and couldn't really find any area of her body that was problematic, that was a source of pain or or anything. She just it did seem to be um, you know, leaning towards something that was clearly had clearly been adopted by the nurses. And I uh and the examination included um included uh uh uh a motor exam and tested her reflexes and so on, and head and neck exam, all of that. And I did notice on the on the um when I was testing her reflexes, so she had no uh patella reflex in either leg. And that is not a very specific finding. Um that doesn't point to much at all, but it was the only thing I could find really. Um so uh that was the situation. And then I can tell you that about a week before I saw this patient, I'd read an article published by a physician who'd had Guillain Baret syndrome. So uh that's called uh availability, um, in that recent knowledge or recent experience with something can influence the way you think. So availability is one of those biases. If you're in a pandemic and somebody comes in with a cough and you automatically assume it's COVID or automatically make any assumption, um, you do that because of what's called the availability bias. That diagnosis is very available to you because you've had recent experience with it. And this uh thought that I had of Guillaume Baret was available to me because I had recently read this paper by a physician who had it. So um from there, I said, the nurses said, um, well, uh um what are you gonna do? You know, we're about to close. I said, I'm gonna refer her to neurology at the major hospital. So I called neurology, and uh the resident who picked up the phone, I actually knew, who'd been with me on a previous rotation, and I said, uh, you're gonna think I'm a bit off the wall here, but I have a patient that I think may have Guillaume Beret. And the resident said, Yes, uh, Dr. Croskery, respectfully, you are off the wall. Do you know what the chances of this are? She's got postpartum depression. And I said, Yes, that's the consensus here that she has postpartum depression. This was now 10 o'clock at night. I said, Can you see her? Because I'm not comfortable sending her home. So she eventually agrees, uh, insisting that it is postpartum depression, uh, that I send the patient in to her. So I was able to get the patient out of the out of this uh clinic and uh into the main hospital. The nurses all shaking their heads and saying any fool can see that she has postpartum depression. That was at 10 o'clock at night. We got her out of the department. The next day I came into the same clinic uh for another shift, and uh I said, uh, did anybody follow up on the patient that we sent into the hospital? They said, uh and and of course, the other thing I want to add is that when I thought of Guillain Baret, I I considered something called the zebra retreat. I don't know if you're familiar with this, but uh there are about six reasons that go into the zebra retreat, which or zebra, uh, which stop you from making uh uh that diagnosis. And you can imagine what they are. They're outlined very well in in one of the cases, but you know, you think people won't believe this because it's so rare, or uh you're just using up valuable resources, you're missing the appropriate diagnosis, and so on. But anyway, uh I got in, nobody knew anything what had happened to her, so I called up the hospital, and the resident that I'd referred her to was no longer working, she'd gone somewhere else. Uh so I said, Well, um, I gave them the patient's name and said, Could you tell me what happened to the patient? And they said, shortly, they said, She's in the intensive care unit. And I said, Oh, uh uh, can you put me in, put me through to the unit? And they did. And I, can I speak to the nurse who's looking after the patient? Yes. And uh I talked to the nurse and I said, How is she? She said, she said, she's intubated right now. Um, and I said, Do you know what the problem is? She said, Yes, it's believed to be Guillaume Baret syndrome. So uh so within, it turned out, within two hours of where we were uh at the clinic, she had significantly deteriorated. If I'd sent her home, she almost certainly would have died, I think. So so what are the biases involved here? Well, one of them is availability on my part. I'd happened to be reading uh uh a very impressive article that a physician had written on what it was like to have Gimberry. So availability, but availability helping me, uh, because it it generated the idea. What were the other biases? Zebra retreat, for sure, which was if I make this diagnosis, as I tried to to the neurology resent, uh, I was more or less uh dismissed for having something too esoteric uh and unlikely. Um, the other biases that were interesting were the among the nurses, there was this uh kind of um groupthink going on. They were all on the same page. Uh they were all a bit intimidating of me because this was actually only about my first or second week in that clinic, and they were clearly, I think, going to get this new guy straight. The group biases are referred to as groupthink and bandwagon effect. And it's it's just an example of how other colleagues can influence you and and make you feel as though you're outside of what's acceptable uh in in um in thinking of a diagnosis. But um I the only reason I keep remembering that case is because it had such a good outcome. If I'd missed it, I doubt that I could have shown up the next day. And so uh you w when we go back over the case and and we identify the biases, um, the case is very good, very good for me because I vividly remember all the things that were going on. But um when you present the case, you can identify uh the biases as they come out. Availability, um zebra, uh the zebra retreat, um, bandwagon effect, groot think, um, anchoring you know, when she came in, as soon as she said that she'd been recently pregnant and she was tearful, you know, that was the that was the anchor. So that sort of cognitive autopsy of a striking case can be a good kind of teaching manoeuvre for for um trainees, I think.

SPEAKER_02 14:27

So actually, I guess that's what we try and do in the podcast, um, is we talk through a case and try and pick up on the biases, although you've introduced me to some today that I'd never heard of before. So uh thank you.

SPEAKER_01 14:39

Um we've never done Amy. We always talk about biases and we we tend to frame them very, very negatively. I really like Pat because it's quite a novelty think about biases being helpful, like the availability bias. A lay person's interpretation of the word bias is a negative one. You're biased against this. People assume you mean like racial or sexual or something, right biases you you have. But we're talking about a different cognitive biases in medical in diagnostic conundrums and things. I quite like that case for for displaying how biases can get you from A to B in a safe way, because I don't think we've ever done that on the podcast before that I can remember.

SPEAKER_02 15:17

No, it's absolutely and it the zebra retreat is a fascinating one. Um it prompts me to think of a case. I was uh working in the intensive care unit and there was a patient with sepsis, and I I don't know where I'd read it or listened to it. It was something on hemophagocytic lymphangiohysteocytosis, which is rare. And I said to the consultant I was still in training, have you thought about hemophagocytic lymphangiohistiocytosis? It was like, no, um, you must be crazy, you know, nobody ever has this. And the patient had it, which um was astonishing to me, to be honest. It's the only case I've ever seen. Um, but there must have been something in the story that led me to believe that it could be that, as well as the fact that I'd read about it or learnt about it. Um and also in ICU, you have time to think. Um so I guess I was doing my tight tea thinking a lot more, whereas when I go to the ED department or the acute medical unit and we've got 30, 40 people to see, it's really difficult to think.

SPEAKER_01 16:27

Right. Right.

SPEAKER_02 16:29

Really challenging.

SPEAKER_01 16:31

So the other thing that really comes up for me in that case, um, sorry, that it that I taunt to quickly mentioned was about the scenario, the the um the environment you're working in. Uh-huh. Because you've talked before about in cognitive autopsy terms, saying, well, what what's really going on when we're making these decisions? And one thing you've talked about is when you work in error-producing conditions, which seem to favor you to make a thinking error. And I work, I think I work 24-7 in an error pr error-producing condition. It seems it's sometimes that one, it's almost as if my clinical workspace has been designed to distract, disorientate, and otherwise steal my cognitive bandwidth, noise, distractions, you name it. Um, and think I think when you said, oh, this was a case of someone who came in 15 minutes before close time, that would stress me out as a clinician, especially had the weight of my colleagues going, you know, we're closing in 15 minutes, let's speed it up here. I would really and it would certainly make me more likely to take what I would feel the path of least resistance, which is, well, let's just send them home. I'll say Fionetta, if you feel worse, come back. All right, then let's go for it. And as you say, if she went into crashing ventilatory failure two hours later, they would have had a horrendous outcome. And it takes quite a lot of effort to go against um the tide of what everyone wants you to do. And I think working in a very fast-paced environment is a very error-producing uh environment, particularly corridor care, where I can't hear myself think, let alone hear what the patient's saying to me. How do we keep our heads? I don't know. I'm still searching for the answer. How do we keep our heads?

SPEAKER_00 18:09

I think uh um when you're in an error-producing particular uh uh condition, then it's useful to remind yourself of it and that you are more vulnerable to error, you're more vulnerable to heuristic use of heuristics and biases, and it's more likely that you will make a mistake. So any kind of breaks you can put on the process are probably worthwhile. In the acute phase, in the in in the chronic phase, you we really have to raise this with the people who control resources and just say, look, you know, if if you keep doing this, mistakes are gonna get made. These lead to serious uh morbidity and mortality. And and we might sound like broken records, but but but if you keep waving the flag in front of them, maybe eventually they'll see it. I don't know. We're not responsible for resources um as individuals. We we're responsible for the intelligent use of them, but we're also responsible for bringing them to other people's attention when they're insufficient.

SPEAKER_01 19:24

Yeah, I've certainly had moments in my career, and it's happening more commonly recently, where I'm thinking, I can't work under these conditions. Maybe it's not a bit bit precious, I can't work under these conditions. I can work under these conditions, is the honest truth, but I might not work very well. And I suppose what we're saying is we just have to keep saying that out loud. I can work under any conditions you give me, really, but my outputs are going to be getting much more varied, um, depending on how you know, in proportionate with how much stress and how little resources you're giving me to do this.

SPEAKER_00 19:54

It's it's it's exactly right. And and if you have to show them an inverted U function as say, this is me. You keep asking, you keep making demands on me, I'll keep producing up to a point, but then I will reach a plateau and I'll start to come down. And and that's exactly what happens.

SPEAKER_01 20:15

I think the three of us here on this podcast, we've all chosen careers which do favor those sorts of conditions. We must get something out of it. Uh you know, it's very it's very different per other person to person. You know, you put someone who's may maybe work in a clinic quite a lot, you put them in a corridor in A and E, they'd say, This is insanity, I'm leaving. And something, something in us feeds on it a little bit. We quite like that. There's a there's a there's the right amount of pressure where we're nice and busy. But when you get pushed out into I'm saturated, busy, and I'm starting to fail, I guess everyone has a boundary.

SPEAKER_00 20:49

That's right.

SPEAKER_01 20:51

Amy, um, before we conclude, can you share? You're quite a chill person. Can you please share what you do um when when you are more demands than you have ability and time to just reorientate yourself and make sure you're safe?

SPEAKER_02 21:05

Um so I guess I'm more chilled now. Um the last post-COVID, I think COVID, I was at my peak as a doctor. I felt like I was in my zone. You know, I was seeing patients in ED, there was the teamwork, it was almost like camaraderie. I I was in I was in that flow state. I loved my job. And then post-pandemic, about a year after I had a bit of that a come down, and then I hit a bit of a slump, to be honest, and work got busier and busier and busier and noisier, and the background noise of not just actual noise, but the noise of what could be going on, AI, which we haven't touched on actually, um news, media, you know, patients coming into hospital saying, This is what I've got wrong with me. This I've searched on AI. And this is what's wrong with me. And it was it just got so much. And I felt incredibly under pressure when the interruptions, people would say, There's 10, Amy, and Amy. Come on, Amy, Amy, there's 10 and Amy. And I'd be like, I don't even know where I am anymore. I haven't been to the toilet for like 12 hours. I think I'm in retention and need a coffee. And I really coped very badly. Um, and I think about 12 months ago, I hit a bit of a wall where I was a I thought I was a really bad doctor. I wasn't able to make decisions. And I think it was maybe decision fatigue, and not necessarily decision fatigue having done a long shift, but decision fatigue having done 20 years in ED. Um yeah, and I think, and I ended up taking um six months off work this year, which is the best thing I've ever done, and I've completely reset, and I've gone back to work in the same environment that I was in before. Um me and Ben working different. So I'm in Birmingham, Ben's in London, um, but still quite similar um departments where we work. But I only system two think at work at the moment because I cannot system one thing safely at work anymore. And I'm sure that creates a huge amount of frustration, particularly for the resident doctors I work with, because I will take a history and it will take me 30 minutes. I know that's not sustainable, but it's the only way at the moment that I can work. Um so instead of seeing 30 patients a day, I might see 10 or 12, much to the dismay of a lot of people. But that for me is how I practice safely at the moment. And it's it is things like working together as a team. So making, I speak out loud now, so I talk all of my diagnoses out loud with the doctors I'm working with or the nurses I'm working with. I ask who I'm working with, what do you think could be going on? So make it more of a team sport. Um and I now follow up every person I see and I reflect on them. So today, when I was at work, I went through all the cases I saw last week to see what I did well and what I didn't do well. Again, not sustainable, but I think there are some things from that we can maybe do. Certainly reflecting on cases that we've seen and learning from them, I think is a good tool. And certainly thinking uh thinking processes out loud. Um, I was on a wardrobe last week and the resident doctor challenged me. And I was like, Oh, he said, Why are you giving antibiotics? And I was like, Well, I've given antibiotics because the CRP is elevated and there's some something on the chest x-ray. And he went, Yeah, but you've scanned the lungs and there's nothing in the lungs. And I'm like, I know, but they've had a bit of a temperature. And at the end of the day, it came down to the fact that that's what the patient wanted. And it was really nice to be challenged, actually. So I think we need to encourage that more from our resident doctors, actually, to be challenged. Um, I haven't answered your question, Ben. I'm so sorry. I've just waffled on. But um, I think what I have learnt is we need to look after ourselves before we can look after others. So if I go to work and I'm tired, I need a coffee, I need to go to the toilet, I make sure that I do that. And I eat, I take breaks, and I make sure that the people around me do the same as well.

SPEAKER_01 25:33

Um sometimes it's put your own oxygen mask, please.

SPEAKER_02 25:38

Yeah, yeah. And I also realize that the hospital can strangely run without me. So, you know, I was actually really I was always worried if I hadn't seen everybody properly because I was like, oh, the hospital's gonna fall apart. Well, you know what? I had six months off and nothing happened. So, and I think that's that's a bit of self, you know, recognition as well and kindness, self-kindness and kindness to others.

SPEAKER_01 26:04

That's a nice uh that's a nice um sort of topic to end that to end the session on kindness to self and kindness to others. Um Pat, I cannot tell you how grateful we are that you've come on twice now to really enhance this podcast and uh give us some real authority on the subject. So we talk about quite a lot of the time, but um I think you've shown us things that we haven't quite understood and you that we've learned a lot. Um, and I particularly love, I've always been a fan of case reports. They stick in my head, clinical narratives stay in my head more than reading out a textbook. So when when something I learn comes attached to a patient, even a fictitious patient, in some state, it goes somewhere in my brain and tends to stay there. So thank you so much for that case you presented and for for bringing that bringing the subject matter alive in that way. It really means a lot, and I hope it means a lot to our listeners as well. Um, and on that note, thank you very much everyone for listening to uh an episode of Home of Medicine Podcast with the Royal College of Physicians Edinburgh. Please do um rate uh the podcast, please do subscribe and please do pass along to your friends and colleagues if you found it useful. And it's a goodbye from me.

SPEAKER_02 27:11

Bye.

SPEAKER_01 27:13

Bye bye.