The Jay Allen Show on Safety FM
Why Learning From Accidents Does Little TO Improve Safety with Erik Hollnagel
January 24, 2023
Today's episode is a special one! We get to listen to a presentation from one of the pioneers of our industry. Sit back and relax and listen to what Erik Hollnagel has to say on The Jay Allen Show.
Here is what our AI thinks was said:

[00:00:00] :  this show is brought to you by safety FM Mhm Well hello and welcome to another episode of the J ALLen show, I hope everything is good and grand inside of your neck of the woods as we get to hang out on this lovely 24th of january of 2023 Yes, I'm still struggling with not saying 2022 don't worry about that, So how are things, how are things with you? How are things going? Is everything good as you're hanging out in this new year? I hope the answer to that question has been astounding Yesh, as we are taking a look around with everything that's going on now, a few weeks ago, I was actually hanging out in ST Petersburg florida at the safety to practical application conference and I was able to sit down and you know, take a listen to some of the things that was going on at the event, so with that being said, I have one today that I would love to share with you and this one is called Why Learning from accident does little to improve safety. Think about that, why Learning from accidents does little to improve safety Yes, that's the name of the episode. Now, listen, this is a presentation by Erik cole Nagle, he is the senior professor of patient safety. So as you sit back today, take a listen to this really, really close attention to what's going on with erIC as erIC is very well known throughout the industry and take a listen to his words of wisdom that he has in regards of what is going on within the world of safety and just kind of within the world in general, just take a listen and you'll notice Towards the end there are some questions that come in from audience members. I'm gonna go ahead and leave those in for this particular cut where you can take a listen to what erIC has to say. So there you go. Let's start off right now. Here's ERIC holder with his little presentation coming from the safety to practical application conference that took place in ST Petersburg florida before before starting. That's one thing I need to say that is if you ever unlucky enough to get an email from me, a department says the difference between what you can imagine what can happen is larger than you can imagine and that happened to me last year. I woke up one morning lying on the floor and had a stroke and never imagined that. So I'm still suffering from that since my speech is slightly incomprehensible, but at least I have a good excuse now, if you don't understand me at all. And when, as I said, prepare for this, I had the, the usual number of slides without cut them down dramatically. I realized today it doesn't really matter because in the old days you would show a slide and you talk about it, but now you show slides and everybody takes a picture and so it does matter if you don't understand me, it's worth it if you don't understand what you took a picture of. Don't come back and ask me about it please, because I'm not sure I have the answer. So this is what I'm going to talk about, why learning from accidents doesn't help to, to improve safety And uh get to the next one. So I think I'll start here because we talked a lot about safety and of course we all know that safety is the problem. But I think I wasn't here all day yesterday, but I think you're discussing all the water's safety actually. So the only thing you can say is that the lack of safety is very costly or just taking some, some statistics you can find, you know, the traffic accidents is performed for them. The most costly 508 a billion dollars. That's quite a lot of money. Even in europe, I won't even mention Dennis Chroma, which is uh, I can't even do the calculation in my head. So I think, I mean, I look down for definitions of safety and I find many different definitions and I'm not sure whether safety is a quality or quantity, he is here to safety is freedom from unacceptable risk. That's a quality and safety is the activity of ensuring that accidents are avoided, avoided. And you know, believe it or not, there's something called, which has another definition. This is if this concept that includes all measures and practices taken to preserve their life health at bodily executive individuals. So it's about protection and it's about things not not happening. So you need to ask what, what is safety actually because there are so many definitions around And as you know, we now we it's now common to talk about 22 interpretations which are called 61 and 62 and safety one is that the number of things that go along is as small as possible and that leads to the to the to the chocolate zero. That is no accidents. And I think the opposite that we cannot with safety is a condition whether or not the word I want Is as high as possible and I call that Kingdom Kingdom is a lesson for 100. So the target is 100% of everything. Everything you do should go well. And I think that's what we're interested because if something goes well, it can't it can't fail at the same time. you might as well turn around instead of privilege from field. That's why I have the coin here, you have a biased coins that comes up and hits all the time. That's cheating. That's what you want to do. Want to cheat. So you get hits all the time and you work of course. So I think this this brings me one way of phrasing. It is the question is not to be a lot to be, is to do a lot to do. Should we focus on what we should do or what on what we shouldn't do. I think Kathy is safety. We focus on what we shouldn't do and we said people don't do this, don't do that. Here's another procedure and tell you what we shouldn't do. But I think it's more important. It's much easier to tell people what they should do and they like it more, Nobody likes to be told they don't do that if you want to design a world environment is more important to tell them what to do and then tell them what not to do. So that's why I take the liberty of distorting Shakespeare little bit here. And so I think it should be to do or not to do. I won't do the whole story liquid. I should I should be happy about that Still 51 which was also called protective safety. It's about protection and and and and the the the metaphor is the glass but it's filled with the things that go wrong and you want to have it empty. There's a stereo vision. So you you look at what's in the class and try to understand it. You try to find find ways of removing, removing it. It's a classical way of doing safety. Of course you can have the opposite where you say you still use the analogy with the class but you want to have a full now of things that go well so that's not a calm, productive productive safety because if things go well you make money also and you produce it's a win win situation. If things go well, you don't have any accidents. You also produce and you create revenue and that that's that's that's what safety too is about. So try to think of this these jars and whether you want to empty your jar, you want to feel it and other we're looking at it is I'm grateful for a doctor in the Netherlands who is wonderful analogy. We try to improve safety by looking at situations where something went wrong and if something goes wrong, you say you're you're almost always say there was a lack of safety. So how can you start to safety in a situation when you say there was a lack of safety, you can't start anything if it isn't there. And I like her analogy. Can you understand what heavy marriages? By analyzing divorces alone? Of course you can't, you can say when you when you're married, you want to try to avoid a divorce, but that's not good enough. I think I speak other experience. I've had one divorce and that wasn't good enough. So, but I hope I never have to have another one. So so and and trump said was to come up with some practical advice for how to do it. So, so I thought I I'm an academic really. So I don't know how to do things and I know we're not not allowed to say philosophy here I am. Okay thank you. So so this is nothing to the philosophy this I'm sure you know this one stop think act and to me the I call this is a stop think act For the 51 version it's aimed at the shop and the people who work there who do the work and something happens to say stop think and then act I think what you need is We need another version which is for the planter into 62 version and that's what you should do when you have an incident uh an accident some reportable event you should say stop and say are you looking for courses efficiency just to find the cause and close the case or you're trying really trying to understand what what went on. That's what I called for owners and then think do you want to do you want to limit the constraint work as imagined? What do you want to facilitate and support work has done and act. I think you should engage with people at this event because they know what's going on and get them involved in implementing the changes that you that you together agree are the right things to do to make sure that work will go well and so that's my suggestion for how to how to do it practically think about these stop think act and interpretation hell a few more sites actually your honor sir. You want to see you go on now. Okay This is just repeating what I said about 61 safety too so safety oneness protective safety and you can see the promises that we need, we need to understand why accidents happen and we define safety by its absence. I think generations rolled out in the paper some years ago but nobody pays attention to that other thing about is swiss cheese and they never read any of his other stuff, which is quite good and that's okay. He also also safety is defined mobil's absence than by the presence and he's absolutely right. So safety is it's like this if something bad happens you want to get away from. And the problem with that is I'm a psychologist. You know that if you want to get away from something, which direction you take will work because it'll take you away from a bad thing. That's why that's why that's why you need to think when something happens it doesn't, it actually better is what you do is not just getting something and getting rid of it getting away from it. So when you look at the way we normally manage safety, I think you can say you manage safety by snapshots, snapshots are association where something goes wrong, go wrong and and they happen rarely and their snapshots of the system not working and how and what you want to do actually I should have a point and I use this this building pointer, it doesn't work here on the screen. But you see the you can see there's a sort of a little text about his presence of safety. So I think what you need to to understand what happens up there. That's what we want to do. That's what we built. That's what we want to support and and to do that. It's not enough to look at the absence of safety. You need to look at what what happens up there and understand what what on earth happened up there. If you want to do something about it, what happens when you look at accidents? We have, I've got glasses on. Many of us have got glasses and we have got different glasses on. I haven't got a label of mine, but you can see the people have glasses with labels on human air levels, safety, culture classes, maintenance, technical malfunction, latent conditions, root causes, whatever. And I think this was mentioned by thought also, I think we all notice that we have this from hundreds of course. He said that's why we look at accidents and learning from accidents. We need to knowledge of the course and to eliminate the course. But I think one thing we should keep in mind is that when shared that was in 1931 and what was world like in 191931 compared to the world conditions we have today. I don't think they compare at all. So using this philosophy from, from 1931 doesn't actually help us because every every method and every philosophy is to and every model is developed to meet the problems of its time. We should we should use methods and wonders admit the problems are all time another problems in 1931. even though we've gotten used to it and that's why they still teach. So I think it's worth keeping that in mind. So that's how we think this is I know you have root go back here and it does something similar. This is a danish cartoonist who did that and I like that because it it shows how you can think in terms of causes and effects. This is a lazy guy. This is in the old days before we had electricity and remote points for all of the light bulb. So he wanted to be able to switch off the light when he when he was going to see who invented this. And you can sort of reasons ruled from from, from where he put his finger to a new reason what's going to happen. And you can also reason backwards if if there I suddenly goes out, you can understand why what could go wrong on the way and that that's that's a nice easy way of thinking that the world is not like that anymore. Now he's got a remote control and and it can fail in many different ways And it may also work and work if you got, he had a wife who could, you know put out the light but not this guy's gotta talk instead. The dog is not smart enough to do that. I guess so. I think we have this, what I call linear thinking is so ingrained. We have absolutely everything around us in medicine. You talk about the virus is the germs make a diagnosis. We have the physics and chemistry, which first I think in the nail service and what happens, this happens something happened before that. And uh, you have in genetics, you have it in, in geology. We try to understand why you have earthquakes and volcanoes. And the interesting thing is you look at that is that every time you have a big earthquake that come up and say, we we we don't understand how this happens. We have to revise our theories. I think it's the same with volcanoes actually. They don't really know. You have in criminal investigations, you try to find out who did what and why. I mean that's why sheriff All this is a great in speaking. So, so when things go wrong, we we immediately say what was the reason for that was the cause. And and one of the popular ones is human failure. A technical failure. If you can't figure out what what the hell it is. You say it's an act of God on examination. I think depending on who you are, what you want to talk about. So I saw that some years ago there was a was a was a bridge, a motorway bridge that collapsed in Pakistan. And I I have a new spirit clip from it. And the guys who said that we, we never had something that that happened before. So it must be an act of God. I don't know the authority's board that I hope not. So the reasoning for 1st 51 is like this. When something happens, it's because we know that all components will have hardware or software will fail sooner or later. Humans make errors and they always will. There will always be unexpected and situations and combinations can hide sneak falls and other floors. And conversely when things go well, it's because systems are well designed and perfect, perfectly maintained and people people behave as expected or and as instructed as the Children to work is done is the same as workers imagined and procedures are complete and correct. I'm sure you agree with that. Designers can anticipate and prepare for every contingency. That's why we don't have accidents and that's that's the way to make sure we don't have accidents. So how do we explain failures? Well, well, well look, I'm sorry, Charge apologies for the screen here. You know, there was uh, it produced tv from years ago that that said had this picture and he said the title of that is where my earrings, not like it. So we look for failures and and we have sort of a standard repertoire of courses that will be applied and again, combination of the glasses, we have different glasses on and one important concept is what you look for is what you find because you have something in mind and you look for it and then you maybe you find it, it's not because it's actually what was there is because you looked for it. If you look for human error, you're bound to find it because they're always people in the system and they always do things that are different for they what you would like them to do because they are creative and adaptive and flexible and try to make things work. So what we do is we like what I call simple explanations and I call them monolithic explanations using the idea of monolith from Kubrick's 2001, you remember that? So we like monolithic explanations, you know, simple single explanations as one whole that you can't break down into anything. And we like Monolithic solutions also to go with the monolithic explanations because that's nice and easy as a 1-1 correspondence. So I just, just for fun, I tried to make up the timeline of these explanations. We have human factors started 19, complexity in 1984 with Charles Perrot's book and human error human error, I really say after team i in 1979 safety culture after Chernobyl and charity in 1986 that safety culture is surely monolithic explanation which says 60 40 that's it, situation awareness. In 1989 and more recently I saw somebody suggested organizational blindness. I'm not sure what that is, but that's as if you if you only overcome that, that's the main reason why things go wrong. But I also think we use them in in in is in stereotyped ways we use them either as a factual cause that that's what was there and that's why we're drawn or as a counterfactual forces say we think it was there and if it hadn't been there then it would have been fine. And we also use it as a hypothetical solution to say we should, we need more of that. We need more safety culture, we need more situation awareness or whatever it is then then then we need more resilience. I should say that because we've been working on resilience engineering for some years to uh and sometimes people say you you need more and that's what, that's what you feel when you start, people would interpret it in that way. So we just need more resilience, could you prove some resist citizens in the organization please. How can you, I mean I I saw some years ago they were trying to design cockpits that increased situation awareness, I'm not sure how that would work. And the problem is with all of these nice concepts, there's no theory behind them. They just sound extremely plausible and people buy them. But this is what I call it. And then eric from was um very famous US psychoanalyst think he said something is very, very appropriate. Is the question of certainty blocks a search for meaning. Another guy who wasn't a psychoanalyst, great philosopher said something much longer because he was, he didn't make it short so you can see here and I won't read it. And he says the first principle is any explanation is better than none. And I think that's quite true. The the cost creating drivers is just conditioned and excited by the feeling of fear. And conversely we can avoid that. If you look at safety too and say we're not interested in failures and white failures have, we're interested in in in things go well and therefore we should be interested why things go well. And then a nice thing of that is that if you unlike failures, if you, if you have failure and you want to get away from a bad situation, any direction is good. If you want to get a good situation, there's only one way to go, but you need to find it. So it is not avoidance. Its approach. We should be considering and when you manage safety, we shouldn't manage it by his absence, but by his presence we should look at what happens about, about the the the the access, the presence of safety. What what characterizes the presence of safety. And so here's the here's the logic of safety to you see when when when nothing happens, it's because when things go well is because humans find ways they all come. Design force interest is it's because performance, they adjust their performance to to immediate demands and the conditions. It's because pretty pretty procedures we used in the context of the situation. And it's because the inter event and things look like like they may go wrong. And if you see if you then look at why do accidents? We have accidents for the very same reasons. And we need to understand that we need to look at that better. Yeah. But the truth is if you look around it every day lows of what we do goes well from morning to night. If you go to the supermarket here and you will you look for something it's there on the shelf. And if it's if it's not to say why the hell wasn't there, what are they thinking of? But if it's there, you never say why is it there? Which you think you should because it's a small actually. And usually usually it's usually things will work and you should. You know the funny thing is we never stop. And and I would say worry about it, we never stop and and consider why things work. We always consider why they don't work because that's unexpected. It's we just expected to work. But in some sense, it's just as unexpected when things work as when it don't work. If you look at it in the right way. So so why do things then go well? And we have to have some sorry to see that. We have some theory about that. It comes from resilience engineering. You say things the world because we are able to respond in a flexible way to what happens. We're able to learn both from what works and what doesn't work. We're able to monitor what goes on. There was this question and debate yesterday account don't remember who acted about monitoring measurement. And so officer was I thought of another example say when I think it was a lady who was the question and say, I always want to say to you cook immediate. Do you monitor or do you measure? I don't think you mentioned very much you monitor you simply, you know, you see you're still in the part and you sort of have the feeling this is right. There's no measurement of force and talk that tells you no, no. I need to put a little more of that into it. So we we monitor all the time. We we do it very well. And the last thing is we were able to anticipate anticipate to look ahead to see what's coming. And I think that's what I think will work because we're able to do these these four things what you call the systemic potentials and that's what you that's what you'd knowledge and and build and support in an organization because then you will also be safe and it will be very productive at the same time. I think that's good. So respond money learn and anticipate. So I told you what we shouldn't learn from. That's what Henrik wrote in 1931. So what should we learn from? We should learn from what happens and how it how it happens. We should look at what people actually do and and try and learn from them with the other day as we were walking down the street before a guy going up on the ladder to to take down one of the christmas decorations a little bit early. I think a very tall letter and there was another guy standing at the foot of the letter holding a letter and there was another guy in a tree nearby with a rope and the guy who climbed up the letter had a rope in his hand. And I thought that's not going to help very much reforms. This has got very strong hands. Uh, but when I see that, I always say someone say, still he's done it before and nothing. He hasn't hasn't killed themselves before. That's why he thinks it works. And that's why people do things they've done it before. And it worked. That's why they keep doing it and we need to, we need to look at the situation they're in and try to understand why they're doing it and and how they're doing it and also why why they come to to this, this habit and also maybe even a social norm, this is what you do. You don't be a sissy, don't tie it around your waist, holding your hand. And uh, so I think that's what, that's what we should learn from. We should learn from. I used the chance before work is done, not, not from what is imagined, That's very important. See what actually happens and try and learn from that. And sometimes sometimes you'd be surprised at what happens. I think we can also look at ourselves and say, how often do we think do we do things exactly the way they should be done? Because we know we're smart and you know, in this situation this is a good way of doing it. So we know we know this works by experience, so that's what we do and we know it, we know it's not dangerous because we've never been hurt. Fine. I, I remember once in South Africa had a bunch of of work inspectors and talk to them and I said to them, be honest, how many of you get into your car and start driving, put then put on the safety built in about half of them raised hands because you should, you know, of course you shouldn't do that, you should put on your build before it starts to drive. Even the first meter because we know where the most accidents happen. They happen in the 1st, 1st, 1st bit of the movement and many people sort of forget that because they've been doing it all their lives and they haven't had a crash yet, so it's perfectly safe to do it like that and and they save, you know, maybe half a second every time and I don't know what, what, what that amounts to do a lifetime. But you know what, what could happen if if if something goes wrong and it's more than half a second that you lose. So another thing that's important is is and when we deal with with with, with safety and every anything else that you say? I want to say the problems and solutions must match. So simple problems could possibly have simple solutions, but complex problems never have simple solutions and discuss discussing a complex problem as a simple problem by offering a simple solution. Actually, it doesn't make the problem any simpler. It only makes it almost certain that the solution is not going to work. So that's that's what you need to keep in mind. I know I'm making your life difficult, but that's why I'm here and I think that was the end of it. So, so we have time I think to take some questions. Diane is that correct? Yes, we do. Right. So I'm holding the mic who would like to ask a question. Hands up. Oh is it that scary? Oh thank you. Great. We have a volunteer come here dr paul Nagle big fan like we all are. Um could you comment on your thoughts about high reliability, organizing and safety to? Is this part of the same? Are there similarities, Are they completely different? I'd like to hear your thoughts sir. Your, he asked about the similarities between stations two and high the library different sensations. Well as far as I know the the idea of a sorrows came about after Charles Perrot's book when people said that that some organizations did have didn't have as many accidents as they thought they should have according to Charles parole. So I think a charro is still focused on what goes wrong and and and and and and noticing that. And and I have a friend you know you used to be a navy pilot And he told me one time that that he flew on fluent carriers and he told me one time when he when he read about a sorrow and they had one example was to carry and he said he couldn't understand because his comrades were crashing left and right. Why why did they call it the higher higher village organization? But the H. R. O. Was just clearing the deck before the planes took off and after they had landed and had nothing to do with his colleagues dying when they were flying. So he was very confused about that now. He's a professor at the business school in Bergen in Norway. I don't know if there's a moral in that probably not eric. Um Thank you again. I'll say what the last questioner said and thank you for being here. I know I think it's your first big trip and uh and thanks, we're all really honored that you're here with us because it's you've influenced so much of how we think. Um So thank you one thing that you said to me once that I still haven't reconciled that I wondered if you can comment on um and I hope I'm saying this right but as I think you're a purist and I practice safety one and safety two together. Um and I wonder if you can comment on whether you think that works or if you think that to be really safe in the future. We need to come to a place where we're just focusing on safety too. Thanks terry. I should shouldn't be, I'm not, I'm not to happy happy about the term safety and safety to but but they seem to have people said so. I would rather say well we should be concerned about this to understand how how work is done and make sure that we do everything we can so that it goes well and I wouldn't call that safety. Choosing my concern is to make sure that things work and go work. I think we all we all want that to happen. Did you answer your question hi um yesterday or Tuesday you said something that piqued my interest. You said um I had been thinking along these lines and testing out. You said I think we should not use the word safety but we should switch to safely an adverb. Um Could you expound more about that and where what that that might take us? And also will you write something about that? That's not deep in a book? I I am sorry I have to admit I'm actually working on a book on mine mining safety with two other guys who knows about mining. I don't know know anything about mining. I was professor at the School of Mining in France but they don't have any mining there. That's probably why they appointed me. So I didn't didn't need to know anything about mining and I didn't learn anything about mining there. So so we are writing this book and we started out by calling it managing mining safety. And after why we agreed we should call it managing mining safely. Because that's what about about managing mining that's what mining companies do. They want to manage their mining. So they so they get some revenue and some profit and they wanted to safely so that people people aren't killed on the ground because mining can be very dangerous. I had an experience and I went down a coal mine in Australia some years ago and I had a funny experience because you know, they drive up, drive up down in little cars and, and, and they just take any cars around and and take that and use it. And we got into your car trying to go up again and I tried tried to close the door, but it wouldn't close. And he said, well, this one doesn't work just outside the doors to drive and I felt extremely safe. But we did get up to the surface as you can see, we have time for one more question any anybody, I'm sure there's a soul here who would like to hear has a curiosity for, for Professor Hall Nagel. Oh, Oh john wilkes. You want to talk about the God of the Devil? Come on. No, I'm actually gonna ask a softball, what is your favorite book that you've written? The 1? The one I I think is my favorite is actor. The principle. Mm. Great. Why? Because I thought it was a good idea. Thank you. I, I don't think it's often we have the opportunity to luminary like yourself. So thank you. So there you go. That's gonna pretty much sum it up for us. I hope you've learned some things on this one because I think you had a lot of great information to share and some things to make you question things that you might be doing within your organization, taken listen to that move forward and see how you can make this world a little bit better than what it was the day before. It's always gonna be the important part as we get to hang out anyways. I've been your safety manager and host jay Allen and until next time be safe. The views and opinions expressed on this podcast are those of the host and its guests and do not necessarily reflect the official policy or position of the company. Examples of analysis discussed within this podcast are only examples. They should not be utilized in the real world as the only solution available as they are based only on very limited and dated open source information, assumptions made within this analysis are not reflective of the physician of the company. No part of this podcast may be reproduced stored in a retrieval system or transmitted in any form or by any means mechanical, electronic recording or otherwise. Without prior written permission of the creator of the podcast, J Allen