This post accompanies the podcast “Simulation for Elite Team Performance,” which was recorded live at the Tactical Trauma 2024 conference in Sundsvall, Sweden. We are grateful to the organizing team for hosting us and allowing us to use the audio.
In this episode, Andrew Petrosoniak leads us through how simulation can be used beyond education for system design and improvement.
Listening Time – 20.12
Author – Andrew Petrosoniak
In emergency medicine and trauma care, we’re accustomed to acting fast and thinking on our feet. But I also spend a lot of time working on simulation, a powerful tool for training and investigating issues within clinical environments. Today, I want to share how simulation can transform our systems, prevent catastrophic failures, and help us deliver better care.
Let’s rewind to 2018. I was standing in the trauma bay at St. Mike’s, a Level 1 trauma centre in Toronto. We had just received a critically injured patient involved in an ATV accident—a four-wheeler rollover, to be precise. The patient was bleeding profusely, with a pelvic binder in place, and we knew we needed to act fast.
We prepared for intubation, established IV access, and ordered blood. But there was one major problem—no blood arrived. Minutes ticked by, 5 minutes, then 10 minutes, then 15. By the time the blood finally reached us, 17 minutes had passed. Watching someone actively bleed for 17 minutes feels like an eternity. While the patient ultimately did fine, the delay in blood delivery was unacceptable.
After that experience, I started asking colleagues if they had encountered similar delays, and many had. Like many others, our hospital had a well-established Massive Haemorrhage Protocol (MHP), but something wasn’t working. When I need to investigate a problem, I turn to simulation—not just for training purposes but to dig deeper into system failures.
Identifying System Failures Through Simulation
A few weeks after the incident, we ran simulations to understand why we were facing delays in getting blood to the trauma bay. What we observed during these sessions was eye-opening. Our nurses were task-overloaded—they were responsible for charting, establishing IV access, and, crucially, making phone calls to both the blood bank and the hospital’s locating service to request a porter. This two-step process was cumbersome, and sometimes the call to the blood bank was simply forgotten amidst the chaos of a trauma.
In response, we developed a new process: the call would be automatically forwarded to the blood bank when the nurse called the locating service. Simple, right? With just one call to make, the system would take care of the rest. However, before rolling out this new protocol, we wisely decided to test it using simulation.
During the test, something went wrong—the call kept dropping. It turns out that emergency phone lines can’t be forwarded, an issue no one had anticipated. Thankfully, this problem was discovered in a simulated environment where no patients were harmed. IT fixed the issue, but had we not tested the new process through simulation, this failure might have persisted for days or weeks, with staff blaming the locating service rather than the system itself.
This example highlights the concept of intelligent failure. By creating a venue where failure is safe, such as in simulation, we can identify and fix problems before they affect real patients.
Embracing Intelligent Failure
Most of us associate failure with negativity, especially in patient care. When something goes wrong, we internalize it and feel like we’ve personally failed. But failure can be a powerful tool if approached correctly. Amy Edmondson, the same expert who introduced the concept of psychological safety, popularised this idea of intelligent failure.
There are four key elements to intelligent failure:
- Novel Approach: Try something new. In our case, the novel approach automatically forwarded the call to the blood bank.
- Hypothesis: We hypothesized that this would reduce the likelihood of missed calls and speed up blood delivery.
- Benefit: The expected benefit was a more streamlined process, leading to faster delivery of blood products to the trauma bay.
- Minimal Risk: In simulation, the risks are contained, and no patients are harmed.
We can iterate, improve, and ultimately succeed by embracing intelligent failure. In fact, we saw a 21% reduction in time to blood delivery after implementing this simulation-informed process. Over the years, we’ve further refined the system, achieving a nearly 50% improvement. The simulation allowed us to safely test ideas, fail, and try again—without any real-world harm.
Data Integration in Simulation
We all know that sports teams use data to enhance performance. But in healthcare, we’re still catching up. Sure, I know how long a patient waits in the emergency department for an ankle sprain, but what about the quality of care during a cardiac arrest? How good was the CPR? How well did our team perform?
To answer these questions, we turned to data. We could objectively measure performance by integrating data from defibrillators and other devices into our simulations. For instance, during one simulation, we used data from a Zoll defibrillator to assess chest compression fraction (CCF)—the percentage of time during a cardiac arrest that chest compressions are being performed. The team achieved a CCF of 87%, well above the 80% target.
This objective feedback was invaluable. The team could see, in real-time, how well they had performed, and it boosted morale. No more subjective assessments—just clear, hard data.
We didn’t stop there. We also used data to redesign clinical spaces. By running simulations and tracking where staff spent most of their time during trauma resuscitations, we could optimize the layout of our trauma bays. Equipment was moved to the areas where it was most needed, reducing wasted time and effort. Simulation wasn’t just improving individual performance—it was improving our entire system.
Scaling Impact with Simulation
Not all of us have the luxury of working with the same team every day, like a sports team does. In trauma care, our teams are often ad-hoc, with different people working different shifts. Training becomes a challenge when you can’t predict who will be there on any given day. But that doesn’t mean we can’t scale the impact of simulation across a larger, variable workforce.
At St. Mike’s, we’ve introduced a CPR coach role, and we used simulation to train the team in this new role. The best part? We zoomed the simulations out to those who couldn’t be physically present and video-recorded them for future learning. This allows us to scale our educational efforts and ensure that even shift workers can access the same training.
We also use simulation when designing or renovating clinical spaces. Rather than conducting lengthy orientation sessions to teach staff where equipment is located, we design the space intelligently from the start. Equipment is placed where it’s most likely to be needed, minimizing cognitive load and freeing staff to focus on critical decision-making.
The Future of Simulation in Healthcare
The future of healthcare lies in using simulation not just for training individuals and teams but for improving entire systems. By embracing intelligent failure, integrating data, and scaling the impact of simulation, we can create safer, more efficient environments for both patients and healthcare providers.
At St. Mike’s, we’ve seen first-hand how simulation can revolutionize trauma care, and we’re excited to continue pushing the boundaries of what’s possible. Simulation allows us to test new ideas, identify problems before they reach patients, and continuously improve our processes.
So, the next time you think of failure, try to view it through the lens of opportunity. Simulation provides a safe space for failure, and through failure, we learn, adapt, and ultimately succeed.
If you’re interested in learning more or discussing how simulation can be implemented in your practice, feel free to contact me. Together, we can harness the power of simulation to improve healthcare for everyone.
Podcast Transcription
When I’m, not doing emergency medicine or trauma care, I spend a lot of time doing simulation work. And so that’s what we’ll be talking about today.
I want to take you back to 2018. I was standing in the trauma bay at St. Mike’s. We had a sick patient that had just rolled in. They’re bleeding. It’s clear. They have a pelvic binder on. They’d just been involved in an ATV accident, a four wheeler rollover. So you know how sick these patients can get.
And I had just called for blood. And we were getting prepped, ready for intubation, we have the pelvic binder on, we’re getting IV access. All of the things that need to happen to take care of a sick patient are rolling. Except one thing, the blood products. So I’d called for it, five minutes goes by, no blood.
Ten minutes goes by, no blood. 15 minutes. No blood. 17 minutes. Finally, somebody rolls in with a cooler, but we don’t have blood in our trauma bay. It’s just above us in the blood bank upstairs. And you can imagine 17 minutes when you’re watching someone bleeding feels like probably 17 days.
And I’m not the most patient of people either. So I can’t say that my, my performance was at its best in that moment. There were some F bombs in a polite way, a questioning way, but the patient did fine.
I started asking some colleagues, Hey, have we had some problems with this? And we identified that we were having some issues in our space, at our institution. We’re a level one trauma hospital in Toronto. A couple of colleagues had also been, telling me that they had the same problem.
And I thought, okay, one of the things, one of the tools that I use to start to investigate and look at problems is simulation. And many of you are probably familiar with it in the, training environment. But we use it also in an investigative way to understand problems better.
And so we started to run, a simulation. a few weeks later to look and see what are our problems with our massive haemorrhage protocol? Why are we having delays getting blood? And one of the things that we heard, we watched as we were observing these simulations, which were effectively the same as a authentic trauma resuscitation.
One of the things that we heard about was our nurses were telling us that they were task overloaded. They were in charge of charting, they were getting IV access, but at the time, they were also in charge of calling the blood bank, and calling our locating through the hospital, which would then deploy a porter.
Somebody would then move, from blood bank and move the blood back and forth to the places that they need. That was the protocol. And so, somebody came up with this great idea,we do need to have a call, we need to make sure that, locating gets notified. But why don’t we just automatically forward the call on to blood bank? Because what we were hearing was the nurses were forgetting to make one of those calls sometimes. And so, great. Let’s do that. Nurse makes one call. Actually, you know what? Let’s not make it a nurse. So we actually brought in a clerical, so a nonclinical staff. So now that’s what we have at St. Mike’s and they make one call, they call, our locating service and then that gets forwarded on as a default, right?
So we leverage what works in in, high stakes environments are, fall back onto a default and the likely that will, as long as the defaults designed properly, it will work very well. And so we went forward. We said, okay, let’s have them forward on the call to blood bank after they initially call, locating and nobody will ever forget anything.
Perfect. You know what? Why don’t we just run this as a simulation just before we go live? But this is great. And so we run the simulation and I look over the clerical and the clerical is looking at me like, And they’re pointing at the phone. I’m like, okay, what happened? So the call dropped. I’m like, all right, keep running the simulation.
So I’m like, that’s interesting. The call dropped fine. We’ll just probably a problem with the phone. And we do it again, and then the call drops again. I’m like, okay, let’s just pause the simulation. Let’s just see what’s actually happening here. We try it again. And sure enough, we find out from IT, we call IT, the call keeps dropping.
And they said, oh, that phone, because it’s an emergency line, can’t be forwarded, but we can do some back end work and we’ll sort it out. So, they sorted it out, but you can imagine if you guys work in anywhere like our institution where things go wrong, sometimes you send in an incident report and then that incident report goes through the paper shredder right on the other side of where that incident report gets filed.
And you can imagine that this would not have been flagged for days, weeks, months, because somebody would have just been like dropping F bombs about how stupid locating is and how they can’t do their job, right? But in fact, this was an actual problem with the system. And we identified this using simulation and It’s this idea we created a venue for failure to happen.
And we’re going to talk a little bit about that. I think most of us, when we talk about and we think about simulation, we think about simulation as a means to improve team performance, individual performance. But I think we can talk about it as a means of improving our systems,
so that it makes it easier for each of you to do your jobs. And we’re going to talk about three areas. We’re going to talk about intelligent failure. We’re going to talk about data integration and scaling impact. These are the three areas we’re going to focus on. I think all of us have failed.
but when we think about failure, we think about it, I think, in a negative way, right? When I say failure to you, does this sort of come to mind? When we think about the care that we deliver, that each of you are delivering on a regular basis, and something doesn’t go right, we internalise it, we feel negative about it.
That’s how we’ve used that word. But failure doesn’t need to be so negative. In fact, it can be harnessed in a way that’s quite positive. And we can think about trying something, failing, building a system around it and trying again. And eventually at some point there’s success. And we don’t have to look any further than the experts in failure and ultimately success.
And a nice nod to our Swedish teams here from Volvo who have designed, the safest automobiles in the world. But this is what they do on a regular basis. They crash test their cars. And effectively create a venue for failure on purpose. They celebrate it. Because no one wants dead people in an automobile.
We’ll tolerate something bad happening to a crash test dummy, which makes sense. And we can create a venue for failure. We can then make changes, so that when we put that car on the road, it’s gonna be successful in an automobile collision.
That’s fine. But we can expect that for as long as there are cars on the road, there will be crashes. Now, maybe Elon Musk and automated cars. And that may not be the case in a few years, but for now, when there’s humans behind a wheel, there’s going to be problems. And so this is this idea of intelligent failure.
And this might be a term that some of you are familiar with. It’s a term that has been popularised more recently by, Amy Edmondson, the same person that’s popularised the term psychological safety. And I think it’s a really great framework or great idea that’s really composed of four areas.
First, you need a novel approach. You need to. bring something new to the table. Just like when I introduced to you this idea in our trauma bay, why don’t we have a call that forwards off of our phone? And then you need to have a hypothesis. We thought that would make it easier for our teams.
That the chance of forgetting to make the call would be lower. And then, there needs to be a benefit, and we expected that the benefit would be that people would not forget that our process of getting time to blood would be much less. And finally, the risk needs to be as small as possible.
Now, the beautiful thing about using simulation is you don’t need to mitigate the risk because the risks are contained within the simulation, right? There’s no patient that was harmed when we were trialing this new idea. And it’s these four principles that we can use.
We can shift away from using failure as a negative term. There’s a real upside to our teams in being okay with failure. The idea that we move into using failure with, a bit more of a positive connotation is really quite helpful. One, we know that we’re going to fail personally at times, but if we can say, Hey, listen, all failures aren’t bad.
and you know what failure is part of the job that we do. And if we can do it intelligently and we can shift that into a venue where that happens and simulation affords us that opportunity, then great things can happen for our team. And what we did,we scaled this up and in a more comprehensive way.
We published this a few years ago, we looked and built out, and looked at how our MHP or massive haemorrhage protocol, the care of the bleeding patient worked, and we looked at in real patients. using simulation, okay, so we, we actually looked at what happens before and after we used a simulation based approach to solving this problem, getting blood to these bleeding patients.
We were able to reduce our time to blood delivery by about 21%, and we’ve now moved into probably about a 50 percent improvement over the last 5 or 6 years. Because we tolerated failure, but we created a venue for that to happen. We used simulation. To support that, to trial new ideas, everything should be an experiment.
And building on this data, we can use this data and we can integrate data into our simulations to really augment what we have access to. And those of you that are sports fans know that sports teams are doing this on a regular basis, right? They have billions of dollars behind them. So they must be doing something right.
They use data to support their practice. They just support their, their authentic play. And we use this now increasingly in healthcare. We still fax things in Canada. So we’re not fully there with the data piece quite yet. I don’t know if you guys do. But we have, faxes on a regular basis. It’s pretty funny.
I was at a meeting a few, weeks ago. And we’re talking about AI, which I’m going to briefly mention. We’re talking about AI and in the same sentence, somebody then talked about faxes and I’m like, I see what’s happening here. We haven’t yet figured it out, but I digress. So a few months ago, my colleague and I, Derek Mock.
decided, you know what, we know how long people wait for an ankle sprain in the emergency department, but we don’t know the quality of the CPR. The literally life saving metrics for delivering care in these, to these patients. I know how long you wait if you have an ankle sprain. I don’t know if I gave you the best CPR
and how well our teams performed. But also I wonder about just giving out data on actual authentic cases. Like how does that work? And so we ran simulations to evaluate that, to look at it, to feel it. Like, how does a team respond? Because this isn’t something that we’re used to. We’re used to faxing things.
We’re not used to getting actual good quality data. And so, we, this data is from a Zoll defibrillator, what I’m showing you here. we had a few other, different,printouts, and it really just didn’t resonate with the team. But this is what we landed on. After we ran some simulations, we could see that teams actually valued and really appreciated getting data for their performance.
And I thought, okay,we had a cardiac arrest, an authentic cardiac arrest, and this is some of the data from it a few months ago. I just after we had rolled this out and I thought why don’t you know what we’re gonna we do our debriefing after every cardiac arrest or nearly every cardiac arrest I brought the team together and it went well from a team perspective.
Unfortunately, the patient didn’t survive, but I thought we brought our team together and everybody, some people were a little bit mixed emotions, obviously around the tragic event that had occurred. And then I printed off this because it prints out, immediately after. I brought the team together and I said, guys, we could actually have a look and see how well we performed.
And I walked them through it and you can see. So the chest compression ratio or fraction CCF is 87%. The goal is above 80. So 80, 87 percent of the resuscitation of the cardiac arrest, we were doing chest compressions. The data doesn’t matter too much, but we, for the purpose of the conversation, but it went very well.
Cool. there’s some improvement, but we brought it together. The team was looking at it and you could just see people stood a little higher. They had a bit of a smile on their face and it really brought the team together because we were able to shift the focus right on to how well they had just performed.
We use this not only in our debriefing, but we use, data to inform, our, design of clinical spaces. This is some work we did a few years ago where this is a nurse moving around a simulated trauma scenario. And you can see the heat maps where they spend most of their time. So why not use this data to redesign your space better?
It’s hard to do this in real life. Like you can’t video, you can video a trauma, but it’s not as you can. Carefully design this when it’s in simulation so you can get exactly what you want. And so then you can imagine, you can see where the nurse spends most of their time. So why not make sure that their equipment’s right there, available to them.
And so by using data in your simulations, you can shift the focus, you can make it objective. You can improve the reception. People are okay with when you start to talk and say, Hey, listen, you know what, I think that the CPR could have been better because we have the data that tells us that it could be versus
I don’t think you were pressing that hard. You can see how that lands a little bit differently. And it’s really important that we be objective when we talk about ways of improving. And it exposes a bunch of unknowns that otherwise we wouldn’t necessarily understand. When we were looking at where the nurse moves around the trauma bay, we didn’t know that was what we were going to find.
And so I think the future is integration of data into our simulations, and that will toggle back and forth between authentic clinical scenarios, when we are actually providing care, and then the simulated environment. And hopefully at some point it will be seamless between those two. And finally, I want to talk about how we scale impact.
And I think some of you, not all of you, might work with the same team every day. The way that a sports team shows up to practice every day with the same people. And so when we borrow the concepts of sports, or high performing teams that are always the same group, it’s different than the team
on your right, which is a trauma team in Toronto. They’re composed of about a hundred and fifty people that come together as an ad hoc team.
Training a team that’s gonna be, you can’t predict who’s gonna be there on that day. We would all love to just be able to train like a athletic team that can come together and you know exactly who everybody is, you know everybody’s name all the time, but it’s different, right? And sometimes I hear, why would we run simulations for a team that we can’t even get the same people there all the time?
I don’t want perfection to be the enemy of good. I think that we can still get something out of it and we can still scale our impact. And I also think that we can’t just be making our decisions agnostic from the clinical environment. Boardrooms aren’t designed for life and death decisions, clinical spaces are. And I think that we can use simulated environments to make that happen. And we use immersive design to create spaces. We use simulations, like I showed you with the heat maps. We use, targeted simulations as we build out new or existing clinical spaces and we make changes to them.
Because I firmly believe that a lot of the time we spend training our clinicians, we spend time teaching them where things are in the space. What a waste of time. Most of you I don’t think, or any of you, have ever been to our trauma bay at St. Mike’s. I’m gonna take you on a brief tour.
Hands up if you can find the chest tube cart. We’ve designed this from the simulations, because we don’t need to waste our time doing an orientation session, right? You come in and it’s your first day in the job. We want to spend, if we’re going to run a simulation, we want to have you focus on how you’re going to set that space up, how you’re going to make the decisions about whether to put in a chest tube
I don’t want you in the middle of a trauma resuscitation, but can you, can someone get me the chest tube cart? I don’t want that to happen. What a waste of time. of useful effort, of cognitive bandwidth to have to remember where things are. We want to make it easy for people. And so we use simulation to help design that.
I love this quote, maybe a bit controversial for those that spend time thinking about teams, but from Jeff Bezos, I don’t know if you’ve heard of him. There’s a new app, it’s called Amazon. kicking around out there. He’s done okay. We should be trying to figure out a way for teams to communicate less with each other, not more. I want that to happen in our trauma bay. You know when a successful resuscitation happens, when people are just not talking very much. In fact, they know it’s implicit coordination or communication that’s happening. People, have a sense of what needs to happen. And it’s this idea that we can design spaces well for that.
We can scale impact with simulation. This is, we’re introducing a CPR coach into our environment. This is a way of harnessing or leveraging this so that it can actually be scaled to our entire team. We can’t get everybody there all the time. But we can simulate and show how it should work. And then broadcast this out. Make sure that people have access to the preferred way that we use this new role. We also run our simulations. We zoom them all to people who can’t, if you’re a shift worker, you can’t be in the ED all the time.
So we zoom them all out. And this has become part of our culture. And finally we video record and we capture this and then we display it for others. obviously we ensure that people are okay with this, but we, this is a video of an MHP that we ran. It has 2. 7 million views. it’s not even the best video that I’ve, it needs some work, but people, are interested in this work.
And so this is the idea that we can use simulation to scale in the educational domain. Here are the three things we talked about intelligent failure. We talked about data integration. We talked about how simulation can be used to scale. And I think this represents the future, not only using simulation to educate people as individuals, as teams, but as systems.
And I’ll end there. Please hit me up with an email. Happy to talk more.
The Guest – Andrew Petrosoniak
Dr. Andrew Petrosoniak is an emergency physician and trauma team leader at St. Michael’s Hospital and an Assistant Professor in the Department of Medicine at the University of Toronto. He has completed a Master of Science in medical education where he focused on the use of in situ simulation (practice in the actual workplace) in procedural skill acquisition.
Andrew’s field of research includes in situ simulation and simulation-based technical skill acquisition. His work focuses on usability testing and the identification of personnel- and systems-based safety threats within acute care medicine. He is the principal investigator of the TRUST study (Trauma Resuscitation Using in Situ simulation for Team Training) that includes a partnership with human factors experts to evaluate systems and processes during high-stakes trauma simulations.
He is an invited speaker both nationally and internationally on the topics of trauma, simulation and procedural skill acquisition.
Where to Listen
You can listen to our podcast in numerous ways, ensuring you never miss an episode no matter where you are or what device you’re using. For the traditionalists, Apple Podcasts and Google Podcasts offer easy access with seamless integration across all your Apple or Android devices. Spotify and Amazon Music are perfect for those who like to mix their tunes with their talks, providing a rich listening experience. If you prefer a more curated approach, platforms like Podchaser and TuneIn specialize in personalising content to your tastes. For those on the go, Overcast and Pocket Casts offer mobile-friendly features that enhance audio quality and manage playlists effortlessly. Lastly, don’t overlook YouTube for those who appreciate a visual element with their audio content. Choose any of these platforms and enjoy our podcast in a way that suits you best!
Tactical Trauma
Huge thanks to Fredrik Granholm and all at Tactical Trauma 24 for their very warm welcome and for letting us record this series of podcasts. This is a fantastic conference, and we would highly recommend you check it out when they advertise their next event.