Integrating training on maternal and newborn care through virtual simulations

Duration: 25:10 3383 transcript words EN summary

Video summary

The session introduces a significant advancement in healthcare training by integrating maternal and newborn care into virtual simulations, addressing the critical need for professionals skilled in caring for both patients simultaneously. Traditionally, these two areas have been taught as separate courses with little overlap, yet real-world scenarios often require health workers to manage situations where both mother and baby are ill at once. The presenters explain that while in-person training is essential, skills can decline rapidly without ongoing practice or supportive supervision. To combat this, the team has developed virtual simulations originally focused on newborn resuscitation and care for small infants, which they now expand to include preventing, diagnosing, and treating postpartum hemorrhage in mothers. This integrated approach aims to bridge the gap between isolated training modules and create a cohesive learning experience that reflects actual clinical realities where providers must balance the needs of two patients. A major portion of the presentation details the complex technical and design decisions required to merge these distinct curricula into a single, manageable game environment. The team mapped out an entire treatment flow starting from pre-birth through referral or discharge, allowing them full control over variables such as when complications arise for either patient within one continuous scene. To keep file sizes low and ensure accessibility on mobile devices without desktop computers, the developers simplified complex actions into basic mechanics like clickable options and drag-and-drop interactions. They also addressed how to realistically depict helpers in low-resource settings; rather than modeling full humanoids which would complicate animations and pacing, they made helpers invisible except for their hands performing specific tasks, requiring players to delegate instructions through dialogue or clicking on items to queue actions. Further challenges involved managing time management and communication within the simulation without compromising educational value. Since individual curricula often include "dead time" between checks that can be skipped in single-patient scenarios, combining both patients meant that idle moments for one became active duties for another, complicating strict 15-minute check-in intervals mandated by standard guidelines. The team resolved this by utilizing controlled time skips to maintain appropriate timing while ensuring players could focus on critical tasks without confusion from a rigid clock. Additionally, communication with the mother was kept generic and respectful rather than revealing specific next steps, preserving the learning curve for participants who must decide how to proceed based on symptoms alone. To avoid excessive complexity regarding scenarios like twins or simultaneous illness in both patients, the designers made pragmatic decisions to simplify these cases, ensuring that players could focus on mastering core competencies without being overwhelmed by rare edge cases. In conclusion, this virtual training platform serves as a powerful supplement to traditional in-person and online education, offering standardized content with objective feedback across multiple languages while remaining accessible globally. The integration of maternal care scenarios represents a sophisticated evolution from earlier prototypes built in OpenSim and later Unity, demonstrating how engineering teams can collaborate closely with subject matter experts to navigate difficult design choices regarding patient states, helper roles, and procedural flows. Although the simulation simplifies certain real-world complexities like simultaneous twin births or concurrent severe illness for both mother and child, these trade-offs were made intentionally to ensure that learners could effectively practice essential skills without unnecessary distraction. Ultimately, this project highlights how innovative use of virtual simulations can enhance patient safety by preparing healthcare teams in resource-limited settings where the ability to care for a dyad is often more common than specialized separate training might suggest.

Read the full video transcript

Hello everyone and welcome to the 4:30 to 5:00 pm session of the 2025 open simulator community conference. In this session, we are pleased to introduce the presentation integrating training on maternal and newborn care through virtual simulations. Our speakers are Dr. Rachel Yumorin and Dr. Mr. Matt Cook. Dr. is a neonitologist at the University of Washington and Seattle Children's Hospital where she is a professor of pra of pediatrics and the director of the neonatal education and simulationbased training program. Mr. Matt Cook is a research engineer with the neonatal education and simulationbased training, also called NEST lab at the University of Washington, where he leads development teams of engineers and health care professionals in identifying and implementing solutions to support health care teams and improve patient safety. Please check out the website found at conference.openimulator.org org for speaker bios, details of the sessions, and the full schedule of events. The session is being livereamed and recorded. So if you have questions or comments during the session, you may send tweets to opensimcc. Welcome everyone. Let's begin the session. Thank you. Thank you everyone. Um thank you so much Lear for that one warm introduction and it's great to see you all. Um it's amazing that it's been another year and I was just sharing with Lear that this uh opportunity comes around and every year I I think what what's the most important thing that we learned this year that I can share with this wonderful group of folks. So Matt and I are here again to give an update on our adventures in healthc care with virtual simulations and specifically focused on maternal care this year. So as you know there's a ongoing need for health care professionals who are skilled in maternal and newborn care. Um as a neonatlogist I've spent most of my career focused on the newborns. Not surprisingly, that's where my my heart is. But I also recognize that there are places in the world where and you know many places and many spaces in which you have a health professional caring for both the mother and the baby. And so thinking about how do we integrate that training in such a way that they can do the best for both of those uh patients has been a new focus of ours. Um, in-person training we know requires a lot of time and resources and without ongoing practice and what we call supportive supervision, which means that you could follow up with the learners to see how they're doing and remind them of things that they may have forgotten. The skills that we train for start to decline. And this decline can actually be quite rapid. Um and once you have that loss of skill then you can't really see the translation to clinical care. So as you all know we've been working in uh virtual essential newborn care simulations for the past probably now six or seven years. This work started out with prototyping in open sim and then eventually transitioned into a unity platform um for a few reasons that I'll share later. Um but we always remember our roots and we you know as we think about how we you know structure these simulations to help to support the care you know we're thinking about the gamification aspects. We're thinking about, you know, the importance of ensuring that they can be accessed in ways that our learners who are often in places where they may not have access to a desktop computer can, you know, get on to these. Um, we're thinking about how do we ensure that they are, you know, continuing to access the standardized content and the objective feedback. And so we have currently ongoing work and projects involving the left side of your screen which is the virtual essential newborn care simulations. Um with all the criteria that you see there um you know really trying to ensure that they you know can be accessed in different languages and that we can develop performance reports and that all of this is in a manageable file size. Um but now today we're going to talk about our move towards in integrating maternal care simulations within the same experience. Um and so briefly the simulations that we have currently are focused on how do you care for a newborn who is ill and needs resuscitation right after birth um or you know is small and needs extra care. And the new simulations are focused on how do you prevent, diagnose and treat a mother who may have bleeding after birth. And so we are leveraging curricula which are already in existence um and supported through multiple professional organizations including the World Health Organization uh but which historically have been taught as separate courses um you know they will attend a 3, four, five day course on maternal care with very little about the newborn and then attend another course on newborn care with very little mention about the care of the the the the mother or the parent. And so how do we take the attempt that's already there to reflect the areas of integration and actually fully bridge it by completely um within the experience ensuring that they have the opportunity to care for both members of the diet. So, I'm going to pass it on to Matt to talk a little bit more about our objectives and some of the lessons learned. Matt, take it away. >> Thanks, Rachel. So, our audience for this addition of maternal care, so beyond, you know, our previous audience of, you know, providers of newborns. So these are the basic emergency obstetric care providers and we have five um main focuses. Uh so first is postpartum hemorrhage prevention and make sure what those correct steps are. Uh how to identify uh postpartum hemorrhage and then of course how to treat it and what to do for normal care when it doesn't happen. And then finally is of course we uh put an emphasis on communication um with both with the mother and with uh colleagues um because apparently um doctors with the exception obvious uh Rachel here or our other colleagues at the University of Washington doctors seem to not be great at that. So, um, the first thing we did was map out the entire treatment flow of how to care for newborns and mothers. So, with the curriculums currently separated, we of course started by mapping out each individually. Um, and then we had to figure out how to combine them, which is much more difficult than it sounds. So, with this, uh, we we've got this pretty well done. And so, what we've done is we've mapped the entire treatment flow. Um, it starts from pre-birth and then we take this from uh to the either referral of the mother or baby or maybe both. Actually, not both. I'll get into that. um or discharge whether they're both healthy and taken care of. And uh we skip over the birth itself. Um but yes, from from then on um and then so what this allows us to do is include any given scenario we want. um where we have complete control over the variables uh when when the mother or the baby gets sick, how they get sick, when they get sick. Um and we are able to do this all in a single game scene um to help keep our file size low. And we are able to control this by defining the mothers and babies, you know, states at various points throughout our game flow. go to the next slide. So, with all these different actions that need to be take that need to take place, it gets complicated. There's so many different things the player needs to be able to do um and we need to do it on hands-on basis. And so we've been able to boil down those complex actions into some mobile friendly simple mechanics. Uh and these have been tested t tried and true through our previous VNC work uh through and um with um mechanics testing. So we won't get into that. But that sums up to we have basic clickables and for some objects we can drag and drop them to uh show how they interact with other objects. Um and these just trigger different animations and these actions can vary from the object from for object to object or from for the even for the same object throughout time. And then of course for communication we have we simplify that to um some basic uh binary dialogue options uh with some good good se options with a good option to choose and a less good option to choose. But um yeah as as we designed everything it since we had this architecture already working it all had to fit into this. All right, moving on. So, we had a lot of complicated design decisions to make. And the first we had we discussed very thoroughly was how to utilize helpers cuz these providers don't do everything in re in real life and they have to be able to rely on helpers. And so we had to design this, you know, designed how the helpers are going to fit into things. We actually explored the possibility of the users being having to do everything like what if they don't have helpers because we're designing for low resource settings. You know, it's possible they might not. Well, we decided to def we decided to make sure to lean into the use of helpers um because you know if they're alone that's unfortunate. They might have to muddle through. Um but as far as the learning objectives um we decided they forcing that complication um wasn't worth didn't add anything to learning what they were supposed to do um in response to these different issues. Um so we decided that when one patient is sick the the the helper has to care for the other. So this created some technical and logistical complications for our helper like okay so what do they look like? Do we have to model a full humanoid person in the room? Do we have just floating hands and a head and a torso? Um, what do they do after they complete a task? Do they we can't have them just stand there stationary because that's not very realistic. But we have but we they can't exactly go on to the next one because well, what is that next task supposed to be? we haven't assigned them a task, we what does that mean for our flow? Who's creating these animations? How is this all supposed to work? Um, and then of course if they move into the next thing, you know, how does the players pace move in? Cuz some players are going to be faster or slower than others. And there we're we're not exactly going with real time realism with these with these actions. Um, so ultimately what we decided is the helpers are invisible. uh except for the hands um for certain tasks. Um so in the picture we have the helper um listening to the patient's heart rate and tapping it out for the provider or the player to see. Um and so if they're invisible, well, how do we initiate communication with them? How do how do you do that intuitively? And so we decided that the when you call for help, a dialogue option is going to pop up and you have to give the helper simplified instructions. Um so something along the lines of go help the mother or go help the go help the baby. We are assuming that the helper is an expert. they know what to do. What we're focusing on with this with these instructions is that the is that we're practicing when to communicate and basically what to communicate about and just showing that the importance of doing so. And then for some actions, we have some basic delegation by clicking on an item. So if there's multiple it multiple actions the helper could be doing and you were supposed to be delegating what which of them to do we go click on an action or click on a thing instead of us being the one to inter interact with it. It's basically us queuing like hey that's next helper you know what to do. You're the expert go and do it. All right move on. So in addition to communicating with the helper, we also have to communicate with the mother. Um and in order to provide proper dialogue, in order to like be really truly realistic and proper, um that dialogue wouldn't really fit. Um we don't have the space for it. And we also don't want to give away what the next steps are. So, it's, you know, we're go, mother, we're going to sweep your uterus now. Like, okay, well, great. Now, the game's told us what to do next. That's not helpful for learning. So, when we can, uh, we try to include some more generic dialogue. So, we say, you know, we're going to do a procedure now, and this could this could be painful. We're sorry about that. Um, so something that doesn't give away, but something that's respectful and simple. Um, because it's important that we keep that word count down in our language simple because of course, you know, space and user or users don't read. Um, so yeah, keep that small. Okay. time management was a was a uh big challenge. It you you wouldn't think it would be, but when you start to really get into the weeds, it led to some really detailed discussions cuz when you have the neonatal care or the mother care individually, there's time lapses between things. So we're we're doing controlled cord traction um or we have to have actions that take place we we check in on the mother or the baby every 15 minutes. So individually we have a lot of dead time and we have we get to rely on time skips to maintain that appropriate timing and skip over dead time. Well, now suddenly when we're balancing both patients, what was dead time for one patient is now something now time you have to go spend on the other. Um, and the curriculum, we had to stick to the curriculum and the curriculum is very clear. You check on the patient every 15 minutes. And so this one little detail on the clock on the in the game complicated things and that's the clock on the wall. We have to have the clock on the wall because it's important especially for newborn resuscitation that we're able to keep track of time. And so that's it's keeping it's keeping real real time in the seconds. And so now suddenly we have we uh at players different paces we may go from one patient to the other and then we have to come back to that original patient on that 15minute interval. Well, suddenly the clock isn't reading 15 minutes either. Maybe it's most likely much too soon or much too late. But it creates problems if if someone's paying attention to the clock, which we probably aren't, but would be good if they did really um might throw some confusion in those learning objectives. Um, so ultimately what we decided was, thank goodness for the helper, time skips were okay. See all that hemming and hawing and we just went back to time skips anyway. Um, because when you're using that dead time or taking an action with the player, you're able to rely on the helper to go work on the other patient. And we're looking at the flow. We just found it still managed to work. It definitely did me require significant changes into where the time skips took place cuz what we had as a time skip in one curriculum or the other definitely didn't you didn't fit right in that same spot when trying to combine them. Hopefully that all made sense. All right. So, what do you do when the mother and the baby are both sick? What do we do uh if there are twins? The procedure says to twe check for twins. Well, what if we have a scenario where there's a twin? How does this what do you do first? How does this impact the flow in our prescribed actions of what is the correct next step? Well, we decided to just simplify this by saying we're never going to have twins. We'll we'll check for it as part of our procedure, but it's never going to happen. And our mother and baby are just not going to be sick at the same time. Um, so our helpers manage one patient and the provider will do the other. And in realistically, they'd never do both anyway. So that still works. and we decided just the learning objectives just didn't warrant that increased complexity. So, we just kept it simple. >> Thank you, Matt. You know, we we're going to have to wrap shortly, but I want to thank you for pointing those important uh go right ahead and wrap, please. >> All right. Talking faster uh variable tracking. So in pre in one scenario or the other there's basically prescribed you know what's next but in when you combine they go then the looping nature of the flow makes requires a lot more tracking of okay has this happened already. So we have to track a lot more variables based on what's already happened in order to make the flow work. So that just is technically trickier and the order in which the merging happens. So at this point, do you take care of the mother first or the baby first or what? Different things happen at supposedly the same time. So it's really important that you work with subject matter experts um to make those final calls over how things should fit in. So as a technical person without that expertise would never be able to make those decisions. Uh so to sum up um this virtual training is a really good ad add-on to in-person uh training and online training. Um and the addition of these maternal scenarios is complex and there's just a lot of design decisions we had to make in order to make it work. So I want to thank all our partners. um at the University of Logos and e- health for everyone for helping with our research and our development. Uh couldn't have done it without them. And thank you for everyone at everyone at the University of Washington um that helped with the development in this project. And uh thank you to all of you for being here. >> Oh, we are so thankful for both of you and those design decisions you made, they make perfect sense. While the mother and baby in life would get sick at the same time, you have to make those decisions so they're focused on task. We want to thank you and Rachel for a fascinating session and I'm so sorry to cut in on your your wonderful talk. Let's see. And of course, let's see. Here we go. >> No worries. I ramble. >> No, but that was exciting and I love that the failure-based learning. That's wonderful. So, as a reminder to our audience, you will want to check out the conference.opensecar.org to see what is coming up next on the schedule. You won't want to miss our next session, which will begin at 5:00 p.m. in this keynote region, and it's entitled Enhancing Secondary Students Self-Efficacy through Virtual Worldbased Learning. Also, we encourage you to visit the OCC 25 poster expo in the OCC Expo 3 region to find accompanying information on presentations and to explore the hyperrid tour resources in OCC Expo 2 region along with our sponsor and crowdfunder booths located throughout all of the OSCC Expo regions. Thank you again to our speakers and the audience.