Early involvement of embedded human factors practitioners is critical for an effective return on investment in problem solving and design endeavors. Through compelling case studies, this session explores how multidisciplinary collaboration and system science have facilitated robust interventions for prevalent healthcare challenges.
It's I'm honored to in to welcome Doctor Laurie Wolf back to ST Louis before becoming the Director of Human Factors Implementation at Carillion Clinic in Roanoke, Virginia. Doctor Wolf spent over 20 years at BJC healthcare first, as a senior consultant and manager of Ergonomics then as lead performance improvement engineer with operational excellence in her current role, she leads a comprehensive program using human factors principles to improve patient safety. She's published over 50 articles and worked with healthcare systems around the world, sharing her knowledge on how to make patients safer. Although she left BJC in 2017, her presence is still felt felt here when I'm faced with a complex safety event. I ask myself, how would Laurie Wolf look at this and I know many in this room do as well. Laurie changed the way I look not only at healthcare but how I see the world around me. And I know after you hear her talk today, you will too. Please join me in welcoming doctor Doctor Laurie. Thank you. Thank you so much. I am so excited and honored to be here today. Um I'm glad to be able to share some of the human factors principles and how it applies to healthcare. Um So I wanted to just start out by demonstrating, I got the remote clicker that has labels on it. So I'm hoping that I won't have the problems. It's just too bad that you have to have a Harvard degree and you still can't figure out how to use the clicker. So human factors issues are prevalent all around us every day. Uh My full disclosure is I have no, nothing to disclose everything you see is just gonna be some uh work that I do every day in trying to make things better for our patients. So I'm gonna talk to you a little bit about what human factors is and if not uh talk about the components that you can look for when you do human factors assessments. And mainly I'm going to uh work to tell you about some case studies. And so I'll give you examples of all these components and we'll look at some uh reactive ways of addressing issues that happen and as well as proactive predicting where things will go wrong before they go wrong. So just as a little introduction of myself, my career began in the automotive industry, I spent five years after that doing military contracts. So I dealt with anything that the human interacted with, whether it was cars or helicopter cockpits, tanks, uh missile systems. Um But I have never been as challenged as the human factors interface in health care. So I've been in health care now, almost 30 years and just love the complexity and fast pace and ever changing face of health care. And that's why you have to consider human factors in health care. And so uh I got my beginning, I will always have a special place in my heart for BJC because I learned how to do human factors here at BJC. Um We BJC was one of the first hospitals in the country to hire human factors people. And when I went to my very first conference, I assembled all five human factors people around the country. It was a representative from Duke Johns Hopkins, Mayo and BJC. And uh we collaborated to say we've got to stick together and understand human factors. How do we do this in health care? So we've been meeting over the years and now we have created a group called Human Factors Transforming health care. It's just a private little group that has grown to over 300 members now. So I'd encourage you join that group. It's free to join. And what we are is all embedded practical human factors, people working in health care. So you can toss anything out there on our uh Slack and blog and mirror. We have all kinds of ways of communicating now and you can get advice and help from other people that are guaranteeing having the same problems and issues that we're having. Wow, even with the label. I still have trouble. So this is my hospital now, it's called Carillion Clinic. We are located in the Blue Ridge Mountains in the Southwester part of Virginia. We have seven hospitals in our system, about 14,000 employees. And uh this is our flagship hospital. You can see lots of construction going on. Oops. Oh my gosh, I'm gonna switch to the one not label because this is bad. So we have lots of construction. Here's a brand new hole in the ground and this one's not even built yet here. My office is there with the overlooks these mountains here. So when I began by work of human factors, we were housed under our patient safety and quality group. A couple of years ago, we started uh developing human factors Center of Excellence. And that of course, we still work with patient safety and quality, but we're able to add components onto our center of excellence. Like the innovation team, we work in the Sim Center. So we have Sim Center people and human factors every other office. And in the back, we have our innovation engineers. We can draw something on a napkin, have an idea for a patient safety thing and they build it for us. They 3D print it, walk down the hall yesterday and they said here, look at this skin that they built with the 3D printer and it sure looks real. So it's amazing the things you just walk down the hall and see. So I'm really excited about the capacity of what we can do with a Human Factor Center of Excellence. This is our team. We have three phd S and two masters trained human factors people on our team. Uh four of us are embedded in the hospitals uh across those seven hospitals. And then doctor Sarah Parker is a professor of Health Systems Science at Virginia Tech. And so we do the research arm of human factors under the premise of Virginia Tech or our collaborations with them. So to step back and describe what we mean by human factors. If you Google, what is human factors, what are the human factors? This is what they give you. There's either five of them uh or there's 12 of them. Depends on what source you look at none of this is what I call human factors. What I care about is how much mental workload and task stacking can you do without making an injury or, or an error happened? So I'm looking at what are humans capable of or their limitations that, that cause them to make errors or be lack of productivity. So that is what we learn in school and what we're um working towards now and how we do that. Our model is uh human factors is a half combination of psychology and engineering. We put those two sciences together in order to understand the human and what they're capable of and apply that to either technology equipment, whether you're designing a uh a helicopter or an IV pump, it's the same principles and concepts that you use or the process you wanna design, design something that humans are capable of doing in order to be productive and comfortable and safe and what's right for our staff will make it safe for our patients. So sometimes it's easier to describe what human factors is not, instead of what it is, human factors will never eliminate human error. Humans make mistakes no matter what. And we'll show you some examples of how that can happen. So what we try to do is make a process or an environment that is resilient to those errors. We don't focus on education. Uh Although that's certainly part of what we have to do, but it's not everything. If you only educate and say, hey, don't make that mistake, it's going to happen again. We don't focus on one individual. We focus on a systems approach and I'll show you some of the components and it ties well into what Jessica had talked about earlier. Um And show you how to do that. It is a uh human factors is a scientific degree. Uh not a uh a weekend class, although you can certainly teach principles and behaviors. But to dig down deep into some of the complex problems that we run into in health care, you need to understand those human capabilities and limitations in order to predict where the errors are going to happen. And as a result of that you get human factors professionals with specialty areas. That's one of the things I like about our team. We have a psychologist, we have an exoskeletal uh ergonomic specialist, we have a data visualization uh expert. And so it shows in the kinds of projects and the depth of projects that we are able to do. Yeah. Hey, what's going on here? Mhm This is the best joke ever. I don't even have to talk. So how did I know that? First off, did you get it? All right. Yeah. And then you go here and then you go here. Did you find the mistake? You will read? Wait, I can't even get to write you and you will just read last. Did you catch that? So we can tell this is where human factors, we know how people read or left to right and up and down and we know where you're going to make a mistake. And where do you think they put the most important things on electronic medical records? Where's the patient? MRN and the patient number? Where is it usually right up here? So be looking the next time you go open your medical records and try to understand the design of them when we're designing uh BPAs, I'll show you an example of that. So it's easy to see. We can predict you're gonna do this wrong. I bet 99% of the people in here fell for this trick. So the kind of work we do some of it's reactive. These are the four kind of major buckets that we work on. We of course, respond to every adverse event and trends and do analysis. We still do root cause analysis in different methods. Um One of the newer projects we have is a whole committee on, it's called the Gross Committee. Get Rid of stupid stuff. Um So that's a, that's a great, great, great project. We love that committee. It's a fun one. We meet once a week and we have a million uh projects going on and then we get together and talk about Gross stuff. Um We also, of course, are working with our uh it people and we have the regular electronic medical record issues. And one of the things that we're so strong on now is usability tests. So before we get purchase equipment, we understand, well, if you can compare product A to B, which one is more, more prone to uh errors, we don't always get the products exactly as we want, but we've understood those project products and then able to train or be aware that this is a potential risk of an error that can happen. So this is an example of the number one thing I want you to take away from our, from my pre my discussion here is go to the workplace, see how work is actually done. This is an example where the landscape architecture, person made a nice sidewalk right here. But where do you think people actually walk? Nobody uses that nice sidewalk the way it was designed and intended. It's because humans do things easy and we try to do things efficiently and, and that makes us do work arounds. Health care is expert at work arounds, they will get the job done no matter what. So we have to go to the work and see how people are really doing it every single time I've ever done this, it has ne there's always some gap between what we imagine work is based on our policies and procedures versus what you actually see. So first thing you have to do is identify what this gap is. But the next thing is ask why that gap is there and that's where human factors work begins trying to figure out why we do the work arounds that we do. There's another characteristic of human behavior is our tendency to drift towards risky behavior. We get used to things. If you're a skydiver, you're used to the risk. Um If your radiology, uh intervention area, you may be a little used to handling radiation and uh you might get a little lax in procedures and policies. So the easy example, here is the speed limit if I'm driving in Nebraska, five hours to get to the farm and it's a long straight road in the middle of the night. Do you think I go the speed limit? Probably not, but maybe we do in a hopefully in a school zone where it's 25 maybe you go 15 there. So it's situational specific and that is why health care is so exciting to me, it's very different. You have to customize it. We heard there later earlier today that it has to be individualized and yet still the same and it's very complex. So what I'm gonna give you examples of today is how human factors can help chip away and make things better. So if you call it, this is uh my example of death by 1000 paper cuts. If you could come up with the ideal electronic medical records, say that exists and it does way doesn't now I know. But even, and if you had it, there would still be a million other irritants that bother you. And so while the brilliant people are fixing out, figuring out how to fix electronic me medical records, what we are trying to do is let's fix some of those other million things that are wrong. Um Our colleague at Cincinnati Children's Hospital, she calls it. Her analogy is mosquitoes. I can get one or two mosquito bites, slap them off and go on and do my job probably doesn't even interrupt me. But the more and more mosquito bites I get every time I get an irritant or something that I have to work around. It's irritating and eventually it does either makes your ear frustrated, might cause interference and can imp uh interfere with your productivity. So when we are going to try to find why things are irritating to people because that will help us fix it. These are the components we looked at so very similar to the components that Jessica talked about earlier today. But this is the system that a very, it's a patient safety systems approach. And I'm gonna show you examples of each one of these components here and then we'll get into some case studies to bring those to life. So the first one is organization, you know how I love policies and procedures, right? So this is an example of a policy. We were, we have a problem with our pressure injuries going up and I was in a meeting and they said all we need to do is train them to that policy. They're not doing what the policy said. So we said, all right. Well, let's look at it. We have over 2000 policies and to try to look up what the name is. It's um if you can find the right policy, this policy happens on page 16. This is what they want to train them on. So I find the right policy. I scroll down the page 16 and this is what I have. I purposely it's small, those colors mean nothing. It's complicated. And the nurse on the floor is gonna bring that out and read it. It's not gonna happen. So it may not be perfect. But at least what we did is made it easier. It's still on page 15. Still, still not perfect. But at least when you do get to an algorithm, it's a little bit easier to see. OK. Environment is another component you need to look at and this is my specialty area, noise and lighting and things that interfere about the environment. Maybe it's the layout of the floor, maybe it's the Pixes or army cell that's down the hall and we have a new design that I just heard about last week and they have a little pocket of PC US that didn't quite fit. So they put them on the other side of the fire door. So there's four rooms that have to go through the fire door just to get to the medications. You think that nurse isn't gonna put some medications in her pocket or stash supplies on the other side of the fire door. So, oh, I ramble on here. Um Let me stick to the script. So this is an example of lighting. Uh This was an error that happened where we grabbed the wrong syringe and the light in this medicine room was connected to the light, same light for the computer room right next door and the there were glare on the computer so they dim the lights. So it's not it doesn't cause glare on the computer. What it makes is a very dim medicine room. And so the light levels weren't where they should be. And here's where the mistake happened here. It's very dark there. And let me show you what it looks like. That shelf is up high. And this syringe, the wrong syringe was grabbed, they grabbed this one. It should have been that one. It was a dosage air. So I call that a layout thing. It's a lighting thing. It's about the environment. Wasn't that the nurse didn't know the difference of teaching them how the difference between A TB and an insulin sy syringe is not gonna help that problem. Uh tools and technology is another great example where usability can help understand how people use it frequency sequence that they use it. All of these things matter and how the the uh product should be designed. So I do have one computer interface example here. This was a BP A, you know, we love those. Um This was how our, it people built the uh program for us and they spent hours on this logo. Nice and pretty, right? So what happens to your eye? And we had to really fight to get that out of there because they had spent so long on it. But this, this version doesn't even tell you the test results except it gives you one little snapshot of the most recent one, but it doesn't give it to you over time, you have to get out of this program and come back in to understand it. And so we just redesigned it. We gave them test results over time at a glance, they could see it right away. And then it was very clear what the steps were that they had to do. And so we did a usability test on it and it showed us what, how to redesign it. OK? Then the final component I wanna share with you is about people and this our mental and physical capabilities, we need to understand all of what that is and how that contributes to injuries and errors that we have. So luckily, do you know chat GP T is just turned a year, what a year and a half and uh there's a new version coming out 2.0 hasn't even been out for six months. And now we have another version coming out, isn't it this week or next week? So as things are getting so smart, I have smart water, I have a smart water bottle that will talk to my phone and tell me when I'm thirsty. I have. But now goes on my watch, my colleague actually showed me an email. He wanted me to read on his watch. I went, are you kidding me? I can't even see uh like uh yeah, don't talk to me like that. I might have smart shoes. I have smart cars. I can get on the UK uh underground. And literally today this morning, I used smart dental floss and it's really good, by the way. So you should look it up. So chat GP T has its things and I'm gonna show you an example of how we used it, but machine learning and things at least right now that it doesn't reason it doesn't critically think it does a good job for the algorithms if you program it right? And you're aware of errors that it can cause. But uh I think there's a time and a place for it, but it's never going to be able to be critically thinking and reasoning through the complexity of health care and the ever changing pieces that we need it to be. So this is how we use Chat GP T in our department. Uh We had a, we, we are always collecting data and asking users to do scenarios and we record their, how they react and do they make errors. So this is a uh a picture of us collecting data. So my colleagues ask chat GP T, this is actually oops. Oh my goodness. Uh This is how they uh how they talk to chat GP T. So if you want to understand the conversation, it took an iterative approach to get chat G BT to tell them how to program Python that they could do this experiment. And so Chat GP told them how to program it. They put it in Python and our experiment was uh on electronic adverse event reporting. So when something bad happens, you have to put it in electronically. And then we have a great team that looks at every single one of them, there's like 450 to 500 every week that they go through and they look at everyone, we have a meeting every day to look at what issues have happened the night before. We have a great process for that, but it all hands on depends on our staff putting them in. And right now our process with this older version here takes 10 minutes to put in one fall. So if my patient falls, I gotta stop my day, spend 10 minutes putting it in. So we have a new design that we're recommending and we worked with it on this one, but we wanted to prove whether it was any better or not because of course they have to reprogramming. It takes a long time. So they want proof that it works. So we use the program code in Python that chat GP D gave us. And what it did was put uh essentially what I think of is like a screen over the mat over the over the um Yeah, the same. What does it go? Oh my God. Yeah. So yeah, the form that you fill out. Thank you. And, and this puts a screen over and it tells me every time I'm at a mouse click and a scroll and all of that. And so I can compare the old one to the new one. It's kind of like a um spaghetti diagram, but it's electronic. It's pretty cool. And I can see that here. It's a little tighter. Look at, they have the old version has to come and go out here and there's very few of those except for the accept button over here. And what I can really do is it takes all that data automatically dumps it into time and key strokes and mouse clicks and then that's easy to dump into a uh analysis program. All of this was done in about an hour of programming because chat GP T helped. Normally, it probably would have taken us a week to get that done. This is an another example of this was a collaboration with process improvement in human factors. We were having trouble with our crash carts and we had an error. So this is in our SIM center. Simple. Oops, that was my fault. OK. Crash cart comes in. This is simulated fake patient, real nurse, fake doctor simulating a code, real nurse trying to find Narcan. This is her job is to find Narcan in the crash cart. She needs to do it quickly. We have cameras all over our SIM center so we can track her motions. Yes. So you see she's trying to find the right bottle. She's looking at all the bottles. Notice this piece of paper right up here that'll come in handy in a minute. See her answers are right there on that piece of paper, Narcan and where it's listed is right there. She didn't even look at it. People don't read signs or labels. Everybody clear. All right, everybody clear. Shocking shock, the liver begin the impressions. So the physician needs that Narcan right now and she's still looking for it. Ok? You get the gist. It takes a while to find it. It gets jumbled, it goes down the hall. So just by organizing it and making slots, we made it more obvious where Narcan is located. And then we were so we made the change. We have the same people come back in doing the same um scenario and we were able to decrease the time in half and the number of motions and manipulations around in half. So that's how we prove that usability can work or prove that a design is better. And also getting the word out. We talked about culture change. If people are willing to tell us there's a problem, they have to recognize that they're doing a work around and then tell us about it. So I gave a human factors uh like 20 minute um uh little class to our dermatology residents. Then the very next week I get these pictures back. It's great. They hate their overhead lamps, they bump their head on them and I said, well, why do you bump your head on them? It looks like they're perfect. Right? You got this little hand and you can move this all around and the arm did move all around. But if you look at where they actually need it, the patient sits in a chair and they need to look at the foot or something that light doesn't, uh manipulate. So they end up using their cell phone in one hand and using the light and then they only have one hand to work with. So by moving the exam area around, we were able to fix that a little bit. They also had this shelf that has nothing important on it and yet look, they can't even lay the patient back. So they're even doing gymnastics to evaluate the patient. OK. So that gives you a flavor of reactive things we react to because people are telling us something's wrong. But I love to bring the world of human factors into proactive we can. That's the whole purpose of understanding human capabilities and limitations. We know an error is going to happen before it ever does. So let's start working on that. So this is my front door and I have a grate right outside my front door. How many times do you think I drop my keys down that storm? Right. Right. So what I do now is I have a keypad. I don't even have to use keys anymore. It's been the best thing for me ever. So there's an error proof engineered out problem with my keys, made my life so much better. Uh retain guard wires. It's supposed to be a never event. Right. The way that these things are designed, you have a thumb and we actually have a 3d printed gizmo, but I couldn't pass it around to everybody. But you see how your thumb advances that guide wire. So it's going up into your heart. There's nothing to prevent it from slipping all the way in. Except you're supposed to remember to grab the end and not let it go through. But by redesigning it all we did, I went and drew it on a napkin. We go back to our innovation people in the back and they came up with a 3d printed gizmo that will not advance all the way to the end. But you need to do that. But what it does is forces you to grab it with one hand to pull off the, the um, the advancer and then it's already in your hand and they're good at that once it's in my hand, I'll remember that because I don't have to, I'm holding on to it. Otherwise I have to remember not to let it slip on through. So it's understandable how this can happen. And if we can error proof it, it's so much better. So we're trying to work with manufacturers to see if we can build this and make it real. Another example is uh insulin pen. This is back in 2017, we knew that there's a potential for this insulin pen error. Has anybody heard of this one? It's not new, but maybe it's old enough that it needs to be new again because this just happened to us. Um So what happens here? This is the insulin pen. It's the problem here is in the kind of needle. So the pins look the same, but you go home with a different needle than what we train on in the hospital. So this is what we do in the hospital we take, you take off that front cover and then everything else is automated. The needle goes in, it retracts in, it's nice and safe. But when they go home, that's an expensive needle. Often times they use a different needle type and this one has that same outside cover, but there's a secret inside cover that goes over that and this one happens to be green. I have an example in my purse. That's a white one. So you can't even really tell that it's there. So the patient will inject themselves. And from what I understand is sometimes you can't even really feel whether you inject it or not. So they don't realize that they didn't get an injection. That medicine hangs out right there in the pin cap because they never took it off. And so what happens then that they never really got a medicine administrated, right? And so when they go back for their doctor's visit, they're like, I need to up my dose because it's not working. And so they up the up the dose and the doctor doesn't even realize that this kind of needle is being used because we're training them on this kind. We assume that it is done. So what we have to do is either go watch them and this is how we discovered this problem just about six months ago. Um And they said, show us how you did that and they brought their needles with them and they showed them thank goodness before they just up the dose or administer that higher dose. And because that that would still be wrong for the patient. So this is something we need to get the word out or fix it. So I think you guys know about this solution. We have weaker solutions here that are people based education policies, procedures. We wanna be over here where the stronger ones are forcing functions, engineer it out, make it impossible to do the wrong thing. All right. So I'm gonna show you now, I'm into my case studies. I'm gonna show you four case studies. This one happened when I very first came to Carillion. We had a portable ventilator that ended up in the MRI machine. And you know, how did that happen? Do we need to train people? So we did a big thorough road cause analysis. We use fish bone in all the ways and we came up with some brainstorm solutions. And when you put them along are our uh uh arrow about strengths of weaknesses. We have the weaker parts. Yes, we did some training. Yes. We looked at them, lay out and told them that you don't go in these certain zones. But the strong solution we came up with is to tether that thing to the wall. And that was our short term fixes a cable, but you can get around that right, you can cut that cable. So now it's, it's fastened to the wall. So that's a pretty strong intervention. Sure enough though, I thought I used that as an example of really strong six months after the you'd done this, some someone had to get in there and change the batteries and they used a crowbar and they were able to get that out. So when there's a will, there's a way and there's a work around. So gotta always continuous improve, right? So this is an example I'm excited about is our construction and you guys are used to so much construction going on. If we can get human factors early in the design, we can make a change with much lower cost than after the design is already built. The building is built. Now, we need to make some changes. It also not only does help us get the right design right? With the architects, it helps us know what kind of process changes need to happen before we ever move in. And we can start understanding that and getting the team to know what processes are, are needed. And then it goes a long way for culture change. Um, and I'll show you an example where they were fearful, the new environment wouldn't help them. And so they were able to see that. Yes, they were able to do the job they wanted to do in the new area. Uh We first start by bringing blueprints to life, full size cardboard, we can lock up almost anything in about less than a half a day. We have a cardboard extraordinaire. We call him, this is Nate. He built us a cardboard C arm even that we could do the maneuver around the operating room. Um But so we bring these, these uh blueprints to life and then we film scenarios in there and we watch how people do things. We usually have a scenario that's sort of standard um uh common scenario. And then we have an emergency scenario where the patient codes. This one is just an exam room, it is an ent children's room and the this is an important piece of equipment that the doctor needs. So they're going over here to get it look where the patient family is. So a little awkward, right? So uh the one of the most recent projects we just did was with our trauma surgeons and uh it is going into a new tower, they're building four trauma bays, which is exciting. Um But they're trying to use one clean, one supply room to access all four of these and accommodate them, they have these doors to pass through. So remember that that's gonna come into play later. So, um our job was to figure out what kind of equipment needs to go in the clean room and are they able to do the job that they need in the trauma rooms? And so we had uh we had two different surgical teams come in on two different days, 13 people on each 13 different people each day and we were able to have them go through scenario. I'm just gonna show you an example. This is our, we put a GoPro up top, we just had turn on the gopro, we have them do our scenarios. So you can see they're working and now here comes the patient in up here in this corner is the initial layout down here is where the team redesigns it. So they go through the scenario once and they interact with each other. You can see they're scurrying around. Um This one note here, the red hat is a pharmacist. Every role has a different colored hat because we analyze it with video later. Um So pharmacists is in this area and on this version over here, the pharmacist is in this corner. So you get a feel for. We're watching them actually go through and perform their uh scenarios and this is what we do with that. So we have the different colored hats. So this was where the pharmacist was in that corner. And now we move the pharmacist over here in this corner. So you can see the red is much hangs out here more often, the less motion for the uh pharmacist and they hang out by the uh surgeon who's right here. So they're, they have better access and can hear the surgeon. So by uh like a link analysis or Spaghetti diagram, it's automatically done for you with the video. So we can compare one layout to another. We also do a bump analysis. So anything uh equipment bumps into or the two people bump into each other that shouldn't. Uh we count those. And over here in the initial layout, uh there's, that was our bump pattern and the, the red is the uh pharmacy hats. This, the second layout is um same number of bumps. Ideally you'd want to see the number of bumps go down. It didn't really in this case, but at least it spread the bumps out a little bit and they weren't all gathered up right there. So we were also with the new design, we were able to reduce passovers, meaning the equipment going from one side of the room over the patient's darrow field to the other. So that was a good thing. Uh This is that doorway I showed you, it ended up being too small and we were able to uh before construction ever started to say we need a bigger door to go through there. We were expecting the crash cart to get through there. It had one inch clearance for that crash cart to go through and you know, they're having to do that in a hurry. So bigger doors was a success. Ok. The third case study I wanna show you is about centralized telemetry. This is a task, very visual. We get usually they're high school graduates that get three days of training, three half days of training. And then they sit in front of these monitors for 12 hours a day. They might get a break if I have 30 patients and I need a break to at my lunch, you're going to take over my patients for me. Now you have 60 patients while I go for my five minutes to train. The closest analogy I can think of with job description is air traffic controller. There's no air air traffic controller that works 12 hour days. They have much more rigid break schedules. So this is a lot of, a lot of very visually taxing job. What we did is looked at nine months of data, tried to figure out what's an average amount of alarms that go off and we wanted to recreate it in a safe environment because it was really hard to look at errors and timing in the real world here. So as best we could, we recreated this in a simulation, one hour simulation. Nice and safe in our SIM lab. But we had our real monitor techs come in and perform a one hour scenario for us. So we found in a typical one hour, 100 and 45 alarms go off in one hour's time, they typically watch 36 patient wave forms. And what's a kicker out of that 145 only 93 of those alarms are actionable. So all the rest of those 30% of them, the right thing to do is nothing. So why bother me with those 30%? Just give me what I need to to make my corrections. All right. So just so you get a feel for what these guys listen to. There are three categories of alarms. This is one medium. It's very important that you put the right alarm thing in the right category. And I want to show you this one, this is a crisis, alarm doesn't happen all the time. We did include it in our one hour because it's the most important crisis alarm. On average, these monitor techs will see one of those in a week. So yes, we did happen to put one at minute 32 in our one hour scenario here. And do you notice those two alarms are going off at the same time. And then this is the third category, a warning alarm. So you get a feel, yes, they sound different, but there's a lot of things that are under those categories. So we had 12 different monitor techs come in on their day off and go through our one hour test and we looked at how long did, what's their response time? How long does it take them to understand what to react to that alarm that they heard? And so the good news here is the crisis alarm that um the V tech where it's rarely critical, they responded correctly 100% of the time and within nine seconds. So that's good. But you got all the way over here, an advisory alarm is one of those that goes off without an auditory signal and it's just visual. So when you have a visually, very high taxing visual overload, just giving them a visual alarm is not gonna work. So 13 minutes to respond to a visual alarm and they were only accurate less than half the time. And so why is that? So that's what we always want to know. Why are they making those mistakes? And if you look at the human capabilities of vision, it's 30 degrees con of vision is what I can see that should sound familiar to some people that's that doesn't change. Um So what does 30 degrees con of vision look like for a monitor tech? It's not a whole lot is it how many patients are actually right there in my cone of vision? So it requires me to scan like looking in the rearview mirror with a as a driving test that's very similar to that. So if you look at what those monitors, those three monitors and the distribution of where the patient mistakes are happening or where my my errors are happening, that's monitor one just divided into quadrants. So the patients are here and here, I'm most accurate right there in front of me. So you can see with that cone of vision, that is why our accuracy is better in the middle. And then there's another um a way of measuring that we can do, that's called the NASA TLX workload index so that we can get their opinion on how much workload they think they're having during this one hour thing. And so what we did is compared it, the pre was a little practice session. It was only five minutes. So that one was much less of a workload than the full hour. And where we ask them the same, fill out that same form again at the end. So this score of almost 65 almost 70 is very high on the workload. And you're de you're at that high workload for 12 hours. That's a lot. And this is what tells us how to fix it. The time pressure was the most pressing most contributing to workload. And the other is, it was just very mentally taxing to them anyway. So if we can come up with an intervention that reduces the time pressure and how many things they have to remember at one point in time because they had to remember those alarms that didn't need it to be addressed. Every 30 seconds had to be revisited. All that just piles on your mental workload. So if we can reduce those, we would make that job easier. And we're working on that one. That's a work in progress. This human factors transforming health care. I told you about we have a collaborative working on this project trying to come up with guidelines to make these better. And then my fourth case study is little less low tech, but it still works. So we were having errors that were made in our interventional radiology department and four different services were re were requesting treatments done from interventional radiology. And these four different services had four different forms and the information that the technicians needed were on four different areas, the red parts right there. So all we did made the four different services sit down in a room and design a form that would satisfy the needs of all four people, one form the same for all four services. And then we tracked before we implemented the change, we tracked the number of errors that happened before versus after our change. So before the two weeks before we made the change, 60% of the patients that came through had errors. And afterwards, it was down to less than 2%. And when we investigated, of course, why did that error happen? Somebody somehow got an old, old form crept in the drawer. So as soon as, uh, we got all the old forms out of everybody's drawers, we've not had one error since. So what I'd like you to take away with you today is go to the workplace, observe and guarantee it's not what you expect to see, but don't stop there. Understand why, why are they doing it the way they're doing it? Um Look for the way we modify work. And I said, health care workers are great at workarounds. That's why we need to make sure any workaround. Anything, any uh too many labels means it's a bad, bad design or it could be improved. And most of all listen to your front line staff, they're the ones who use it every day. They know what's irritating to them. They may not always know how, what will fix it, but you can creatively come up with a fix. A lot of times they have ideas, you bring it to life, come back to them and show them, hey, is this what you meant? They said, well, this fix it and then ultimately change the environment. Don't expect the human to change or change the process, but leave the human alone, they're gonna act the way they act and they're going to make mistakes. So the moral of my story is, yeah. Now, how many times on the news do you hear about human error? There's always more to the story. So that is where our job begins is whenever someone says human error, I love those. Ok, let's figure out why. Ok, that's all I have today. Thank you so much. Uh We have a few minutes for questions. Get rid of that thing. It's on the way up there. What do you did? They does BJC have what we do here, what, what human factors we have like a T for that office here. Um I know we do have the capability to get some consul ergonomics consultations, uh NP uh process improvement uh but certainly not to the level. Um That Carillion has hi actually, our Human Factors center center of Excellence. One of our goals is to get a human factors engineer in every hospital across the country. That's our goal. But we do, we uh we've been just chipping away one at a low level at a time, one mosquito bite at a time. We've gone to four different hospitals uh last year and we, I bring my business manager with us and we try to talk about return on investment and out of the four hospitals we visited last year, three of them now have human factors positions open. One of them, Dana Farber Cancer Institute hired our intern as part of their very first human factors engineer. So it's growing human factors is new for health care and we have to figure out how to use us sufficiently and in the best way that we can. So, but it's coming a long way, Laurie, do you do consulting? We do consulting. We're just growing that area first. We've developed our team. It's taken us this long to develop this team and we're gelling. But we do, uh we've done some consulting for CDC um for dialysis that was a three year contract. So we do, we do some consulting, but mainly we like to collaborate with hospitals and you guys have a lot of great talent here. BJC is a marvelous place so we can collaborate with you guys and on projects. Hi, Laurie. Is there a pipeline for human factors engineers? I is, are they plentiful? Are there degree programs? I think that would be somewhat of a barrier. That's a great question. It's, are there enough human factors engineers to go around? I think is what question is and it, and when you get your masters or phd, you can go into many different uh industries. I loved automotive industry. I love helicopters and tanks is pretty cool. Um But to lure people into health care is, is uh a challenge and we're working on that. We have this intern program. We just got 11 amazing resumes for our intern. We had one opening but now we have three openings. So we, we had so much interest in it and such a success last year with getting our intern hired they, now this summer we're gonna get three of them. So it, it's, it's growing and if we can nurture those human factors programs and lure them into health care, that's what our goal is because it's exciting place to be. It's challenging though. So the short answer is, yeah, we could use more of them. We're working at it uh large to make change. That's a great question. So what can you do without a human factors engineer? There's so much you can do. So those components that, that seeps framework is something very common that a lot of human factors people in health care use, it was designed for health care. It came almost all of our tools come from the military or, or uh industry and uh but it's been adapted to health care. So look at those components that I have under those subcom. It is just a little reminder if you want kind of a checklist to take with you, that's some a good source and I'd be glad to provide you with those kind of sources. Um Nielsen's heuristics is also a really good uh guideline to use. There's 10 heuristics that you can use as human factors principles when you're designing um a computer interface or you know, it has things like um decrease clutter uh that the uh game I showed you, it shows you how to use what you need when the frequency of use and the sequence of use, all of those things are heuristics. So there's a lot you can do on your own and then you just call someone when you get stuck. That's, that's the idea. So, yeah. Great question. 01 way in the back life. I'm sorry. Do you have to be? Oh yes. Uh Yeah, we have our what we call our stop gaps, right? If something bad happens and you try to make it never happen again. From that minute on, you have a stop gap, maybe you don't get to do it as well as you want to, but maybe it's a little short term thing you could do to often use workarounds for that. But then sometimes design take a long time. We actually had a design here. I don't know if anybody remembers. We had a triple lumen catheter clamp that was breaking and we worked with the manufacturer to redesign a clamp that had a little joint in it and that clamp hasn't broken since that redesign, but it took seems like seven years. Was it to make that redesign happen? So some of these long term things, a work around that is a solid way of making the error not happen is still better than leaving that risk out. I can't believe I just said that because I don't really like like workarounds, but sometimes health care. This is a hard place. Sometimes it's hard to come up with an error proof solution. And anybody that does those action items, the goal is every single event should have an at least one strong and they say one intermediate strength action item to keep it from happening again. But that is really tough. Great questions. I have a question. So in this year of tighter margins that we talked about earlier, how do you build the case to embed your Human Factors program? Um as part of an uh an organization safety profile. So that's a great question. So as best we can, every one of these projects, we try to come up with a return on investment. Sometimes it's cost avoidance in the case of construction, that was an easy one to have cost avoidance. Like in that pediatric area, we recommended rounded edges on everything in there because two and three year olds bump their heads well during COVID times, somehow that rounded edge requirement got um didn't go through and the new design came out with square corners six months into it after they, they had moved in, they had to replace all those corners um rounding them off because the kids were getting bumped heads. So sometimes after the fact it happens um but it's, it's and it's cost avoidance, which isn't as good as actual cost savings. Uh Sometimes it's productivity or workload like that mental workload change. There's some research out there that shows if your mental workload exceeds a certain amount that you're likely to make an error. So some of that we can do on literature, some of it is the softer side and you're not always going to get evidence, but if we can be involved with our colleagues and make a long term change like falls. Um I've done a lot of work in patient falls. If you can get a program that takes, it's a lot of work to get it going and up. But you can prove that you can make a difference, but it's maybe not solely human factors falls is awful hard to engineer out unless you spend a bunch of money on a soft floor and then it causes other problems. But um so we just keep chipping away at it and put together as many of these success stories as we can. This collaborative across the country has just assembled. Uh They put a call out for everybody to ask for case studies. And so there's some examples out there with cost savings and all kinds of savings from human factors being involved in in cases. I think that's out on the website now. So if not, I can, I'm glad to share any of this stuff with you guys. Hey, Laurie, it's Jody over here. Hey, so I, I think an interesting metric would be um the health care worker satisfaction, the frustration of, are you kidding me? I brought this up. It doesn't seem to get fixed and then you just add that times 1000 to the bedside caregiver and they're having to deal with you. Like, why is this fixed? Why is this shelf here? It seems like that would be a, a really wonderful metric that could also, you know, fall downstream and improve patient and well, employee satisfaction. Yeah, our VH AJ is trying the two, just went to a presentation last week and they're trying to get that physician or provider satisfaction score part of the scorecard that that would be creative because it's some a measure we're already doing and that needs to be a balance with some of these other scorecard measures. So that's hopeful. And then there are individuals, things we can do like the usability scores or the NASA TLX, we can as we go through and do our experiments, we can get that measure, but that's not as broad of coverage as say the physician satisfaction score would be. That's great. Another question. OK. So one more, do you have any more? Do we have time for one more question? If anyone has one, have you, have you done any work in uh normalized and um positive deviance and its impact and high reliability? Uh I think we're out of time for questions. She handed me the mic. No, no, I have no idea. That's why. No. Uh Let me see if I can make up an answer so normalized. What obedience? It's, it's, it's more so looking at what argument about artificial nails. Why is it acceptable to do things that are, we know are not good behavior but its impact. Yeah, that is such a great question. And some of these infection prevention dilemmas are just so difficult and, and I work with the Center for Health or center, the CDC that is trying to reduce infection for dialysis patients and home dialysis. It's so confusing on what do you call clean and dirty. Something can you know, a piece of equipment is uh magically clean, just bec or dirt, clean or dirty either way because the patient sat down or the patient got up, that makes it clean. So I don't know that I'm stumped on that one, but it's certainly something we need to look at and it's, it's again because things are complex. If it was so simple, why couldn't we just swab it and know that's, you know, it's, it's, it's categorized as dirty and that's the behavior that makes it go away. It's not that easy. I think we try to simplify things into checklists and what we're losing is that decision making, critical thinking in a checklist. So it's not that simple. Ok. We're definitely at time. Thanks again so much. This is fantastic. Ok.
