VANGUARD ROUNDTABLE #6: THE PROMISE AND PITFALLS OF INJURY PREDICTION

Can human performance practitioners, armed with the best-of-breed technology and heaps of data, reliably “predict injury”? In this episode, we explore why this has become such a controversial topic, how to vet technology solutions claiming to enable injury prediction, and what needs to happen in the future to evolve these methods. Our panel of performance experts weigh in on this debate and discuss the new approaches to (and often controversial claims of) injury prediction.    

CONNECT WITH THE PARTICIPANTS

 

GET THE PODCAST

The Vanguard Roundtable Podcast is now available for you to catch up on every episode anytime, anywhere. Follow us on SpotifyApple, and Google Podcasts.

FULL TRANSCRIPT: The Promises and Pitfalls of Injury Prediction

Emma Ostermann 00:04 Welcome to the Vanguard roundtable podcast, where we discuss the latest trends driving the human performance industry forward. I’m your host, and in this episode, we discuss the promise and pitfalls of entry prediction. We explore why this has become such a hot and controversial topic, how to bet technology solution claiming to enable injury prediction and what needs to happen in the future to evolve these methods. Today’s roundtable guests include Garrett Bullock, Assistant Professor Wake Forest School of Medicine. Garrett is a former professional baseball player, physical therapist and quantitative epidemiologist. His specific research entails risk and risk mitigation lifespan, impact of Exercise and Sport and the influence of clinical decisions on patient care. He’s worked with multiple national and international professional athlete organizations and lakes. Next up Darin Peterson. Darin has served as the human performance director at the US Marine Corps School of Infantry East since 2007. He drank the Marine Corps kool aid from day one and has been hooked ever since. He’s also a proud father to three amazing boys. And finally, John Cortes human performance consultant at Fusion Sport. John has a Bachelor’s Degree in Health Sciences athletic training from the University of Central Florida and a master’s degree in Exercise Science from Syracuse University. John has worked with professional organizations including the NFL and MLB and in the collegiate setting. The views expressed today are those of the individual guests and do not necessarily reflect the position of fusion sport, or the guests organizations. Enjoy the episode. The Vanguard roundtable podcast is brought to you by fusion sport maker smartabase. smartabase is the premier human performance optimization platform for elite sport teams and military organizations. smartabase is built on an infinitely configurable framework that allows you to create an adaptable solution to support your unique strategy, process and culture. With ultimate flexibility smartabase helps data driven sport and military organizations continuously leverage the latest science and technology to improve athlete performance and service member combat readiness to see how smartabase can help you visit fusion sport.com Garrett, Darin and John, welcome to the Vanguard roundtable podcast. Before we dive into the questions today, I would like to hear a little bit about yourself. I’m going to kick it over to Garrett, you want to give us a quick update on who you are. Garrett Bullock 02:23 Yeah, so I’m a system professor in the Wake Forest School of Medicine, my whole appointments in orthopedic surgery, rehabilitation, bioengineering and biostatistics. And then I also have appointments within University of Oxford in the center of sport UK. I’m working with Olympic athletes, and my focus. I’m also a physical therapist and my focus in research and applications and implementation of researches into risk risk mitigation and prediction in sports. Emma Ostermann 02:49 Excellent. Darin, want to give us a brief bio. Darin Peterson 02:52 Hey, how are you guys doing? So brief bio, on the human performance director at the School of infantry East here camp Geiger for the United States Marine Corps. Background originally athletic training into strength and conditioning, kind of both. We can wrestle with you can actually be both or not, but been there. I’ve been at the School of entry 17 years. And yeah, that’s, that’s a current bio. Thanks. Emma Ostermann 03:22 Excellent. And finally, John, would you like to do a brief introduction? John Cortez 03:26 Yeah. Thanks, Emma. My name is John Cortez. I’m one of the human performance consultants at Fusion sport. Prior to at fusion. I was an athletic trainer for 14 years, working in professional settings like the NFL and Major League Baseball, as well as in various collegiate settings and also in PG clinics. Emma Ostermann 03:46 Excellent. Garrett, Darin and John, we’re so excited to have you guys here today with us. We’re gonna go ahead and dive right into it. Darin, I’m going to turn to you for this first question. And have you kick us off for why has the term injury prediction become such a hot topic and polarizing topic and human performance? Um, Darin Peterson 04:05 yeah, I appreciate the question. I think the prediction is a hot topic right now, because a lot of folks want to get ahead of this game, right? They want to get ahead. They want to identify things that could potentially happen. I think it’s a little tough, to be honest, in the military setting, in the military setting, because a lot of the research that currently exists is in athletics. I don’t I don’t know if that data is directly transferable. So I think, to your second point, as far as it being a polarized topic, or or something that’s not really even understood is how do we predict or how do we necessarily associate prediction with current prefer progressive models, right? Uh huh. How do we get in a human performance type model? The programming that we provide? How do we predict what’s going to happen from that whether it’s an ACL tear, we can pretty much do that. In a setting where everything’s controlled in the military setting, for us, particularly, nothing’s controlled, and the enemy always gets a vote. That’s the same in sport. However, I think the the end results a little bit different. If you if you miss the mark. Garrett Bullock 05:29 Yes, I would like to jump in on on the controversy is that I would, I would vote that, I would say almost all prediction in sport, and then the the poor translation into military, which is definitely a different environment has been is it’s not been as the models and that worked very well, to not working at all at this point. I think some of that is due to the methods. And then I think a lot of it’s also due to what companies and proprietary systems are touting versus the actuality of what they’re actually used for, and what the actual tool can be implemented within clinic or performance settings. Yeah, John Cortez 06:05 I think there’s a lot of, I guess, confusion around the what, when people are saying injury prediction, you know, what are? Are they saying that it’s going to happen on this Thursday at two o’clock? Because that’s what some people’s mindset is, and, and others are saying that, you know, this is this is a, we’re just looking at probability, this is a risk this might happen. And this is when you’re more likely? Or if you have these mitigating factors, that then you’re more likely or less likely, but people have their there’s not a consensus on that definition. And then I think that’s what causes a lot of the controversy is that people have their, you know, their, in their mindset, what they think it is, somebody thinks it’s another thing, so then they’re, they’re working to achieve theoretically the same thing. But they’re going at it with different principles in some ways. Darin Peterson 06:54 Yeah, I would agree with that. And I think I would caveat with I think predictability, for us, is moving away from like, what you mentioned with pay might happen Thursday, I think what we’re trying to do is is lump folks into bends more of, hey, the posterior chain in our military tends to be pretty weak. So we know, over time, we need to always increase posterior chain strength, we need to increase power production, those types of things. It helps us with our programming more than it does actually pinpointing and even saying to the, to the member, you have a high risk of injury in this setting, or in this type, or this location to your body. Because sometimes that mindset just throws them off a little bit too. Garrett Bullock 07:46 Yeah. And John, I think you’re you’re different, you’re the thing, distinguishing between classification of like you’re going to be injured, you’re not gonna be injured on Thursday versus risk is really important. And that prediction in terms of risk, risk management is definitely something these tools can be used for. But not, I think we need to do a much better job of educating of, you’re going to get injured in the next week, or you are definitely going to have an ACL tear, or UCL tear is not how these models or these tools work. And I think we need I think if we can do a better job of educating and using more shared decision making a viewer higher risk. And when you do this, this and this, and we can decrease it by 20%, say within posterior chain strengthening, doing some type of glute ham Essentrics works kind of that kind of classic kind of methodology. And with the shared working between strength coach at Pt MD and other stakeholders, I think that could help actually use the models and the tools and the way they really should be used. Emma Ostermann 08:41 Absolutely. And I think that you guys brought up a lot of good points and Garrett and one of the points that you brought up is the buckets that that people tend to be thrown in with these buckets. Do you think they’re two completely separate things between sport and military? Meaning, if you’re trying to do these predictive modeling, whatever, however you’re doing it, and you’re trying to sell, say, either to military or to sport, do you think those buckets are even in line to even come close to giving some sort of good information? Darin Peterson 09:10 So I would say no, I would say that they’re, they’re not in line, I would argue, or want to discuss with anybody that thinks that they’re directly related anything from an athletic population, trying to relate that to the military population. There’s some correlations because people want to coin the term tactical athlete, however, the training is completely different. The end result or the end desired end state is a little bit different those outcomes and those behaviors that we want them to possess. Training for an event for two to three hours on a Saturday, maybe for a game is a little bit different than a six month deployment or you know a special ops type mission that might last a few days, two weeks. So I don’t think you can been anything when it comes to like the military setting, but you can maybe bend some things with offensive linemen are predisposed to this or shortstop might be predisposed to these types of things because they do the repetitive movements over time. And those things tend to happen in that fashion. Again, my take on it would be I don’t, I don’t think you can translate one to the other. There’s a lot of folks out there. And as mentioned, there’s there’s some systems out there that they claim that that can be done. I’m not, I’m not on board with that right now. Garrett Bullock 10:37 You’re not have to agree that I think some of the, the principles that we’ve learned through athletics can be used in the other setting, but they need but the context is very different. And that in case, in case of prediction, what we’re talking about, we need to develop completely different models, and not just take another model, say for ACL injury in soccer, and then plug it into military. John Cortez 11:00 Yeah, and I think, you know, the, when we look at sport, like sport has a lot more, in some ways, like, it could be a little bit more customized. In some fashions were in the military settings. You know, when we look at, let’s say, footwear, in the military setting, you’re not getting custom footwear, you know, for a while, you know, these kind of these kind of these boots kind of bother my feet, you know, can I get something custom made? You’re not getting anything custom made? They what’s issued is what’s what’s issued. But that’s going to create, I mean, we all know that everybody is structurally has their slight differences. And while there may be some small things here, and there that are that are done, there’s at the end of it there is that sort of external component that we can’t change it the the boot is the boot. So, you know, does that increase their their risk? I mean, theoretically, yes. But that’s not being really assessed in a these models are in these in these prediction mindset. So of being able to take in some of these things that we can’t control, even though in athletics, we kind of can agree. Garrett Bullock 12:12 John, I think it’s funny you talking about boots, because we’re really literally doing the redoing the boots and military right now. So with my lab, so I can agree that we’re trying to take the the average size male and average size female, and then we’re trying to skip the standard deviation to create the boot for everyone. So that is a great example Emma Ostermann 12:31 of it. Awesome. Gary, I want to drive this next question, have you kicked us off for it, but there are a lot of terms used, whether it’s injury prediction, injury forecasting, injury, risk profiling, injury, mitigation, etc. In the end, they’re all working in probably probabilities based on data. Why are some methods more controversial or problematic than others? Garrett Bullock 12:51 Well, I if you want to take this, I think a great example is if you would take sports medicine or strengthen conditioning is that there’s these different tools out there and for prediction, you know, you have different machine learning tools, or you can go to deep learning or yo statistical. And that the for the, for the right person, the right time, those tools or the correct the right intervention, like in this case, we say logistic regression or gradient boosting machine or deep learning algorithm. But trying to generalize these into this one is the best for everything, or we need to use this one for this stakeholder group is I think, where some of that issue is coming from. So like, for example, smaller data that’s being more specialized, is going to be a much better with a statistical type of model. But if you have a high signal to noise ratio, with an overwhelming amount of data, maybe maybe a gradient boosting machine, or deep learning network, depending on what’s going on might be better. But then on top of that, you have to think about the end user of what they can interpret and how they how much autonomy they want to have in interpreting the model. So that’s where all kind of all that kind of meshes together, of trying to and I think that’s where like a lot of the ambiguity and the misunderstanding of this model versus this model, or this prediction versus that prediction, John Cortez 14:06 is something I like, like it in this scenario, that because what the injury prediction or forecasting or whatever you want to call it, the everybody’s trying to be 100% You’re not going to be 100%. And like that, that is in some ways where we’re we’re already like, essentially putting this with something that could be a very useful tool and a very negative light because we’re like, Well, an injury happened well yeah, like we’re not nobody is saying that it’s we’re gonna we’re gonna create these models so that they don’t occur if you don’t want it to occur stay in bed and don’t do anything. You won’t get injured out on the on the field, but that’s not going to happen, you know, what bet whether it’s a battlefield or you know, on the on the playing field, so we have to accept the fact that there it’s okay for these models to be essentially wrong. I mean, we look at look at weather patterns. I mean, we’ve been studying weather for years, we can’t get that, right. So all of a sudden, we’re gonna turn around and you know, in 2022, and figure out injury, mitigation forecasting risk coming, it’s not going to happen, we’ve got to, again, not not to say that let’s not work towards it. But let’s also look at it and give ourselves a break and understand where you know, where we can actually the confines in which we’ve been working. Darin Peterson 15:25 Right. I, I couldn’t agree more with what you’re saying, John. And I think what you said earlier, Garrett is is we also need to look at throughput. And what I mean by that is taking a small population and studying them over time is ideal. However, at least, in the military setting, we have a very large population with limited amounts of time. And trying to predict those types of things within a training environment, at least where I’m at, we go through 20,000 Marines a year going through training. So how do you test how do you collect data? And how do you do those things? For that many folks, and oh, by the way, you have time commitments levied by the command lobby by leadership, those, we don’t have two to three days to study, folks. So we got to get six to 700 people through in an eight hour period. I’m not, I think that just adds in compounds the problem, right, how you collect massive data sets, which are good data sets to have when it comes to predictability measures. But how do you collect them in a short finite period of time and have accurate valid data? John Cortez 16:40 Well, and I think they’re you bring up a good point is that, you know, there, we can measure a lot of things. But there comes a certain point that how much capacity do we have to review and analyze and synthesize this and, you know, that’s where everyone gets excited about machine learning, let’s just chuck it into a machine. And it’ll it’ll spit out what we need to know. But it doesn’t work that way, like the otherwise we could easily get replaced with with a machine and you wouldn’t have the, you know, the practitioner, the human element of it, that there’s a big factor that goes into that as well, that’s not going to be able to be spit out through these these machine models. So the there is, there is a lot that we need to we can look at, but we also have to look at it, it this kind of goes to the point I was making before in terms of being okay, with not not getting everything is we have to be okay with like, Alright, we’re gonna collect these five metrics, because I can I have the bandwidth to look at these five metrics, and we’re looking at these five metrics consistently. Because if we look at these five, this year, these five next year, then we don’t have enough longitudinal data there in order to actually come up with anything. So then that that also lessens our, the, I guess, how strong our any sort of forecasting or our predictions become. Darin Peterson 17:55 Yeah, I agree. John, I think the other thing we need to look at and is actionable data, if we’re collecting data that isn’t actionable, and isn’t relevant to the leadership it whether it be your athletic director, athletic team, or your coach or to us to our colonels to our leadership, if they can’t action, that data, we’re collecting it for no reason. So I think that that’s, that’s a key point to the I John Cortez 18:29 think I would, in some ways disagree, they’re not the and the reason I disagree is that it’s not that we can’t collect that data that the commander of coach, someone else can’t necessarily synthesize or process at this time. Because I think there is value in having that data. And being able to, you know, we think we like to take baby steps with the coaches, you know, and get them, get them to understand the reason get them to, you know, buy in. And once they do that, if we have this other data that we’ve been collecting, while they while we’ve been trying to get them ingrained into a system or understanding, just you know, like, there’s a lot more that you can do with just seeing with your eyes, like how somebody moves, you know, there is more to it, you know, and then we have data that can start backing up as they start to buy in, we can feed them a little bit more data, and that hits the point in terms of having that longitudinal set. So I think we do have to watch out in terms of not wanting to collect data, because we can immediately turn it around. I think we can also play the long game with with data collection. Great. Emma Ostermann 19:36 Absolutely. And just what poses to the group, you know, entering whether it’s injury prediction, injury forecasting, injury, risk profile, whatever term you might use, whether you’re working with a commander or Sport Coach, do you think the term you’re using is going to help create better buy in with a coach and a lot of this is going to be in relationships, right? If we need we need to establish a relationship with a coach I’m coming in as a new athletic trainer or a new strength conditioning coach. And I want to be able to say, hey, what I’m doing is I’m going to be able to enter your risk profile these athletes, how would you go about starting that conversation, when you may not even understand it fully, like, figure out your point? These methodologies haven’t even been, you know, fine tuned. So how would you go about that? Garrett Bullock 20:24 Think having a discussion with with the culture, the manager, whoever the commander of understanding where they are of, so if trying to find what level you need to speak with them, and then having a terms of standard operating procedure with standard language that has a glossary in trying to hone your message towards towards your main stakeholders are going to be commanders or coaches. I think that’s where you begin and starting the dialogue. And you know, that might be I know, people that have made their coaches take like a week of basketball, they make them all take, they’re like, what’s your learning style for the coaches, so the strength coaches can give them the best way to give their data like one coach really likes numbers, another one likes color, and we we mesh the data the way the way they need to see it the best learning style. Emma Ostermann 21:13 Absolutely, absolutely. Perfect. Well, this has been a great conversation so far. And I can’t I can’t wait to keep it going. And, John, I’m going to turn to you for this next question is, how would you recommend someone goes about betting a technology solution that claims to be able to predict injury? John? John Cortez 21:28 Yeah, that’s, that’s actually a really great question. Um, I, one of the things I think about is because this happens so often is the person sees this new tech, and they immediately bought it. Like, because it’s flashy, they’ve got all these things going on. And then you know, it says like, injury prevention, injury risk, and, you know, it shot like, it’s showing all the happy go lucky stuff. Um, and people just immediately like, Yep, I want that tech, let’s slow down, let’s, let’s look at a little bit, the, if you’re gonna vet a any sort of technology, you have to go beyond their website or that commercial. Like, if that’s where, if that’s where it stopped for you, you haven’t bettered anything, all you’ve got you’ve, you know, you’re just you’re just drinking the Kool Aid, so to speak, the you haven’t, you need to go in a little bit further. And as you dig a little bit deeper, sometimes what you’ll notice is these, these technologies that will put out, you know, a, oh, we can do X, Y, or Z, when you look at it there, the research that they have behind it is sort of closed loop. It’s only it’s been published, it’s basically, it hasn’t been reviewed by anybody else that’s within their, their organization, they put it out. So you’re like, Well, I mean, is your organization going to put out something that says, Yeah, this didn’t work, but you should still buy our product? No, this is why, you know, when, when we look at scientifically, when we’re looking at, at, at research, when you know, we want there’s obviously those levels of evidence, and we want it to be something that that’s peer reviewed a systematic review, something that has a little bit more strength to it, in order for us to see like, is this? Is this going to capture? What what it’s truly saying it’s going to capture? And then on top of that, is this fitting? What what my needs are? So, you know, Dara brought up the point in terms of, you know, we were talking about athletics and military, you know, if something is being marketed to both of them, and it’s the same way, like, that’s a red flag right there, because we’ve already discussed, those are two completely different entities in the way that they operate. So we’ve, we’ve got to look at these things. And some of it is, it’s not so I would say, you know, take a step back, digest it and think about it, because if you’re if you’re like, this, this just seems too good to be true. It probably is like there’s, there are faults to all of these technologies. You know, there’s always there’s always a limitation to the technology. And we’ve got to recognize that and see, does this technology, even if it is great, does it fit within the puzzle of of my, you know, risk profile or whatever technologies I’m using to assess my group of athletes or military personnel? Darin Peterson 24:17 Right. I’d caveat off that and offer offer to the group that I think it also has to fit the practitioner, what technology you’re using the practitioners and who’s collecting the data and making, you know, decisions off of that those analytics, the tech has to fit their style, it has to fit their learning style, it has to fit their coaching style. Too many times again, to John’s point, we buy the flashy new item because it makes sounds or has cool colors. But it doesn’t give us the raw data that we really need to get after the problem set. And I would also offer at least then In the, in the military setting, we have to, we have to vet things through a lot of times through university research through things that are going on outside of the the military setting, and validate things that way. Because we don’t have the luxury of necessarily buying a piece of gear and testing it out ourselves, we tend to go and take the opinions of university type settings, or the tri services committee that have already vetted these things in different settings. And they offer, you know, the top priorities within technology of which we should buy, or where we should buy him from. Garrett Bullock 25:39 Yeah, off of that I from like a prediction algorithm of whatever software, or whatever type of risk profiling you’re looking at, as it needs to be. At the end of the day, it needs to be externally validated, and preferably with not the company that did, you know, and there’s different ways to get around, that you can have a university could be another research group, you might have to sign a nondisclosure agreement, or whatever you need to do to keep their trade secret. But still being able to externally validate it on at least another dataset of the similar similar athlete or soldier at the end of day, that is the bare minimum that needs to be done. Emma Ostermann 26:17 Absolutely, there. And I think you brought up a great point, especially coming from being able to access that raw data, you if you are going in with the technology, whatever technology you choose, and they’re running it through their algorithms and whatever it might be. If you’re not seeing how that raw data is being compared to their final outcome, how can that change the game? And how does that change a decision? Garrett, I do want to throw this out. You I know you’ve done some research in this area. And I would love for you to expand on it just a little bit, especially with the paper that you put out I think it was not too long ago. Based on what you found, especially with in terms of epidemiology, do you think there is a certain timeframe that needs to be established to be able to start creating these sort of profiles, whether it’s you need to be able to have at least 10, five, however many years of data to be able to make an injury risk profile or claim to make an injury prevention profile? I Garrett Bullock 27:15 would say that it’s less about timeframe and more about the total sample and the number of injuries. It’s specifically and then your I would say 99.9%, of prediction of injury risk prediction models out there are underpowered. And that you really need multi organization or multi, multi military base, and like a lot of the work we’re doing now is multi country, from a military standpoint with NATO have to have enough people. I mean, you’re I went to, I just did one with the military as like sample size calculation. We needed 5000 military service members to develop it. And then we needed another 10 to 12,000 to externally validate it. And that’s now and that’s still probably not enough to do deep learning off of it. Honestly pony probably more like 15 to 20,000 in development. So Emma Ostermann 28:09 absolutely. So with that knowledge, do you think a lot of these claims that tech companies are making to be able to predict these predictive models? Do you think they’re vetting with that amount of individuals? Garrett Bullock 28:19 Absolutely not. But I would say most of them are not worth what they’re claiming. Emma Ostermann 28:26 Absolutely, absolutely. John or Darrin, you guys have anything else to add on? Vadim technology claims? John Cortez 28:33 Nothing on my end? Darin Peterson 28:34 No, I don’t have anything to add to that. Appreciate it. Emma Ostermann 28:38 Awesome. So just to keep this going. Gary, I’m gonna turn to you on the next question is what needs to happen moving forward in order for these predictive models to be more effective, Garrett Bullock 28:48 I would say number one is having transparency in the model development and the algorithms that they create, that’s having open access data, and which may not be which may not be able to be pure open access where anyone can, but at least having a place to be able to access the data, having complete code hyperparameters if you’re doing machine learning, or or the equation algorithms itself, if you have more statistical base model. And then second, then I think finally is having external validation from a non a third party that’s not involved with the development of it. I think those three things would go a long ways in helping develop much more robust and much more useful prediction models. John Cortez 29:29 Yeah, I think and we mentioned this earlier, like I think we need to come up come to a consensus. You know, what, we’re what we’re actually trying to achieve here and because I feel like everybody has their again their their own thought in terms of what it is and everyone’s trying to create their their niche within really what they’re, they’re all trying to do the same thing. And with that, that I think that leads into the the point Gary was making in the sense that then it becomes these data sets are closed off to other other groups. So you can can’t analyze it. And you know, that becomes a big challenge if, you know, we think about military personnel, this happens with them as happens in sport, even collegiate setting, if someone’s in that same organization, that they’re only there for a short period of time, you know, if someone’s going from one professional organization to another, and we’re not going to allow, we’re all looking at the data differently, over again, keeping that closed off it, that doesn’t help with, with that long, longitudinal long term, longitudinal data collection to be able to give us what those results are, like, we’re all looking for the same thing. So it’s basically and this I mean, this doesn’t work well in the business world. But let’s all work together, and we can actually probably achieve a lot more. But yeah, I mean, there’s, there’s definitely a challenge there in terms of being aligned in terms of what we’re looking for. And then being comfortable with, essentially sharing that. Darin Peterson 31:03 Yeah, and I think John, you just hit the nail on the head is that is alignment of efforts, right, if we can somehow get to the point where we’re sharing these data sets across all entities. And I’m not saying just open source sharing. But if we can somehow share data sets, we can get closer to these injury profiles and these injury risk models. I don’t, we don’t tend to use the word prediction, because that that raises some spider hairs on people’s backs. But if you can profile some things and put people into certain bins, that would help tremendously. And in the grand scheme of things, even inner service wide, and across the board of collegiate sports, professional sports and military, because I think there are lessons learned to deck can be learned from each that are applicable, but aligning those efforts would be key, Garrett Bullock 31:59 when one of the things that I learned or work in Australia was the term coopetition, which has been used in the tech industry for a long time to help bill like its competitive, partner competitive companies that partner in some format to build something greater. There’s actually been in the sports show that’s been done in England rub, UK rugby, and which they all share a minimum data set that they agreed upon with the entire league. And then they can all work within an hour, they want that open source, like as I’m not part of that rugby league, I can’t source it, but they have a governing body, which you each club gives it automatically, but you can apply to get the full data, I think something like that within at least each pro sport league. And then within military, which is some of the stuff we’re working on right now with the five eyes and NATO is really I think, where things need to move towards, which has already been done within NIH Wellcome Trust for medical, which is, which, you know, you have to think about identifying data, but then within physical science has been done for years. And if you don’t share your data that they don’t even believe you you’re not, you don’t have no grounding. John Cortez 33:04 So yeah, and, and I think like something this is, you know, and sort of my world now here at Fusion sport is, you know, with our system, we will, will collect the same data, one organization versus another organization, they’re using the same, let’s say, five pieces of tech. But in Derrick kind of hit on the on this point is that, you know, practitioners, the streaming mission coaches, like people have different methodologies, and that’s okay, there’s nothing wrong with that. And that the, the information that’s coming in, they’re kind of molding it in terms of their space, whether it’s, again, the military, pro sports, collegiate sports, you know, whatever, whatever variable, it may be, that that changes it, but they’re getting the same data, and we can, we can make that fit into what we need it to be. And, you know, we’d, like I’ll build these these systems for, you know, for the aggregation to build that risk profile, it’s a for an NBA team. And then there’s one for a, you know, Olympic organization, they’re doing it in a different way with the same data. So like, let’s just the, the data is the same. So there’s nothing wrong with with sharing that data doesn’t change your, the way that you handle the data and sort of gives you that still that flexibility, again, as a professional to say like, Yeah, but I’m going to look at it this way. This is my innovative way of looking at this and how I’m going to apply it, like let that’s still the professional being the professional. And then the data is just the numbers, just let that number be a number and let’s share that around. Emma Ostermann 34:37 Absolutely. Couldn’t agree more. And, Gary, I think you brought up a really good point with, you know, lease sharing data, and I think in the States, NCAA has something similar with the injury surveillance program. Have any of you whether John Darren or Garrett worked within that program, and I’ve seen anything come out of it in terms of being able to profile student athletes or anything like that, Garrett Bullock 34:59 for that specific They do. That’s from data lists. And that’s been held at UNC for a long time. And they’ve done a really good job of being able to describe epidemiologically. You know, here’s the, here’s the prevalence of this, here’s the incidents per year, this is all this change over time. But the hard part about that is, is that you don’t get range of motion strength. So you don’t get some clinical test measures, you don’t really have injury history, um, in there, and then exposure is is which is like, you know, amount of hours or minutes you strength train or practice or playing a game that’s in there to a degree but not to the granular level that probably is needed, I’d say the closest in the States is probably the hits database, but that the same thing, they don’t have range of motion strength, they don’t have balance, they don’t have injury history is just here, the hand injuries here, the knee injuries, and exposure is not in there really at all. So there’s there’s some organizations that have moved towards that. But I think the next level was at least getting some standard clinical tests and measures and having exposure data, which at this point is very easy to track, at least at the professional level for games played or minutes played. Less on the practice side. Hmm, Emma Ostermann 36:06 absolutely. And whether Darrin John, or Garrett, could, could you explain why that exposure data is so critical? And to like these profiling? Garrett Bullock 36:14 Yeah, so this is a classic EPI, one on one right here, this is fun. Um, so if you have two patient, two athletes, and one plays, say, let’s say an NBA game 82 games in the year and they play every minute of every game, and then you have another athlete that plays in half of those games in plays 10 minutes, say they’re the eighth man or ninth man, they played 10 minutes a game or something like that. That’s, that’s a totally different amount of loading or exposure that they’ve had within the season. And so that’s really important of understanding where the risk is. So they can have the range of motion, their strength, or balance or injury history could be all the same. But how those two players are loaded. And there might be some intrinsic resilience factors in my plan to this, but the one that plays every minute of every game for all 82 games is probably gonna be at a higher risk just because the amount of exposure that have playing so much. Emma Ostermann 37:05 Absolutely. And Darrin, I do want to ask within the your studying, are you are you guys doing anything similar, where you’re tracking these exposure rates and trying to, you know, profile your, your service members. Darin Peterson 37:15 So we are, there’s a few different facets we’re trying to do that one is not necessarily related to musculoskeletal injury more for blast exposure over time. That’s one of the bigger ones we’re looking at over time. But when you look at musculoskeletal injury or injury in any sense, it gets a little tough because each duty location that a military member might be stationed, whether it’s in our service, or even within, for our instance, within the Marine Corps, if they’re in different bases, you have different access to injury histories to different alarm, electronic health records, and you can’t necessarily see everything that’s happened over time. That’s one of the things we’re working obviously, with Fusion sport and through smartabase, on with the SOCOM developments and the developments that we’re working through with the Marine Corps. So that those those injury profiles, and I guess exposures over time, can be shared. And that’s the beauty of it, because everything in there is de identified, and we can share those data sets across the board. And only at certain locations are the data sets known. But everything can be shared, and we can start to build those bigger sample sizes over time. Emma Ostermann 38:42 Absolutely. And it sounds, you know, like, we’re on the right start, we just aren’t maybe capturing the right data sets to be able to do something with the data. But you guys both alluded to those exposure rates. And I do want to pose one more question to the group because as you guys have been talking, I’ve been you know, obviously creating questions in my head as you guys just keep keep, yeah, keep me on my toes on what I want to keep learning and I think I heard relationships in there and as you whether you’re in the tactical sense, whether you’re in the collegiate sense, whatever it might be professional setting, what key individuals or whether it’s people job titles, are crucial to be in those areas to successfully prove profile these athletes, whether it’s SMC athletic trainer, data analyst, do you guys have thoughts on that? Darin Peterson 39:32 Yeah, I’ll take a swing at this first. Sorry, John. I think one I would the first person that’s key, in my opinion is the organizer of it. whatever title you want to give that person whether it’s the director whether it’s, you know, the clown, car circular, the Wrangler Katz, that person has to be able to in order to get to these type of risk profiles, that person has to be able to have athletic trainers, talk to strength coaches, have strength coaches talk to orthopedics, have them talk to behavioral health specialists have all the different entities come together and have these conversations in organizing. One a collective event form or an event where everybody’s entering data into the same system. So that it’s not siloed. That that would be my, my initial push for for the that key person. John Cortez 40:48 Yeah, what I was saying it, and it’s similar to that is that like, it’s the, the individuals themselves, it’s sort of the boots on the ground kind of individuals, where they’re the ones who are, you know, hands on with these operators or with these athletes, that, without them, there’s such an integral part to it, because the the operator that the athletes there, they’re going out there one way or another, and it’s going to happen, if that if those providers, those professionals are not even capturing that data, not entering it in, in the case of like, let’s say, smartabase, you can’t really do a whole lot, you’d be surprised how much you can do in a system by just simply capturing that you saw someone, and you did some you did, you did maintenance with them, or you did some range of motion stuff, you we sometimes think that we have to get so granular. And sometimes all we need to do is say that we we worked with them. And what that does is that kind of builds value within yourself as a professional, when you’re looking at then the both kind of going up and down the chain going up towards stakeholders, people that are going to fund you, and fund potentially these technologies that that you want. And then also looking at, if we want to call it down, it’s not really looking down is looking at the the athlete and the the operator like we can show them, hey, look, just by coming in when more people started coming in, we had less injuries, what it has nothing to do about what we did. Like there’s that you waited talk about evidence based practices and all this other stuff, throw that stuff out the window, people were coming in, and we saw a reduction. Step one, so that’s that bias. That’s that little piece where it’s just all we got to do is just start saying that we did something and you’d be surprised how much you can show. Garrett Bullock 42:45 Yeah, we’ll just have to reiterate that I was gonna say is that the at strength coaches are kind of the first line of defense, and having buy in from them, and buying one really inputting the data, but to have of being able to be able to relay that information to the athlete, and the coaches of indiscernible way to whatever level they’re at, I think is really important. John Cortez 43:10 And I think another element this comes from my athletic training days where, you know, the you would see so often, you know, if it s&c And Dara mentioned about silos, like s&c is over here, the athletic trainers over here, and then you have you know, another, you know, the the mental health over here, is working all these silos, and it’s like, aren’t we all working for the same person? Like trying to help the same person? Why are we why are we working in parallel to each other? Like, let’s let’s collaborate because you know, that it’s like an s&c Like, you do you do a lot of great things. Like, let’s take those great things on sick, the great things an athletic trainer does, let’s take a great thing. And even the coach, this is where the coach piece can come into. It’s like they good coaches can do a lot of great things, too. Let’s all work together to be able to put that for the for the again, the end, end user and individual, because that’s where we start to get the buy in going in and out. Emma Ostermann 44:14 Absolutely perfect. John, Darrin and Garrett, these are been a great conversation so far. And as we start coming to an end, I do want to invite to either of you have anything else you guys would like to add to the conversation just to kind of wrap it up to leave our listeners with? Darin Peterson 44:29 Yeah, so I want to kind of cap on what John said earlier about not listening. Not necessarily going with the commercial, right? You see these commercials you see these things out there that make a lot of different promises. And please don’t take this at all. As a tech hater. We use tech every single day. I think it’s a great useful tool. But a lot of the commercials out there you got to dig deep into who’s saying what in what’s setting? And what I mean by that is, is that person still actively in that setting? Or is that person have they left that setting, and now maybe they’re working for that company, or maybe they’re working for a rival company. And I say that only because I think it’s only surface deep. And if you dig in to some of the claims that are being made, or ties to different entities, if you dig a little bit, you might find some value in that. That’s, that’s what I’d have to offer. Garrett Bullock 45:29 Um, I’d say the last thing from my end is that making sure that that your stakeholders, your soldier, your, your athlete, your coaching staff, understand what you’re just what you’re trying to communicate to them. Communication is key and having if you could have the best algorithm and the best tech in the world. But if they don’t understand it, then it doesn’t mean anything. John Cortez 45:54 Yeah, and that, that kind of goes right into what I was going to say, which is, and I say this often with, with a lot of the clients that I work with, you know, the the best way to get by and the best way to get people motivated a lot of times is keeping it simple. If you would keep it simple, the message doesn’t get misconstrued in one way or another. You know, even as we talk about injury prediction, prevention, forecasting, whatever it may be, is let’s, you know, again, we don’t need to get it, keep it that, well, we don’t need to make it that complicated. Let’s just keep it simple. Like we’re here to try to keep people out whether it’s an operator mission capable, or an athlete, available for for a game. That’s all we’re trying to do. What can we what can we do to help? Emma Ostermann 46:39 Absolutely. There’s a lot of great points today. And I would like to thank John Darren and Garrett for for all this information that you were able to share with us. And if listeners would like to connect with you what’s the best way Darren, do you have a social media email that people could reach out to you? Darin Peterson 46:54 So yeah, for me, the easiest way is just through email. Darren D A R. I N dot Peterson P E T E R S O N @usmc.mil. Like military. That’s the easiest way to get in touch with me. And then we do have soI East human performance Instagram page, and facebook if people want to get out and advertise or not advertise that’s the wrong word, especially in this morning. But get out there and, and communicate with us. Emma Ostermann 47:25 Absolutely. Garrett, do you have any way our listeners can connect with you? Garrett Bullock 47:29 Yeah, I’d say that my my email and my Twitter account. So my emails, G. Bullock at Wake health.edu And then my Twitter is at Dr. Gs Bullock. Emma Ostermann 47:40 Perfect and John John Cortez 47:42 Gray, either LinkedIn, or my email. John J O H N dot Cortes C O R T E S At fusion sport dot com. Emma Ostermann 47:53 Perfect. Thank you so much. We’ll make sure to provide that within our podcast description as well. First and foremost, John, Darren and Garrett, thank you so much for joining me joining us today. And if you found this conversation valuable, please follow the Vanguard roundtable podcast on your favorite platform and share with your friends and colleagues and we’ll be back next month. Thanks everyone. The Vanguard roundtable podcast is brought to you by fusion sport maker Smartabase. Smartabase is the premier human performance optimization platform for elite sport teams and military organizations. Smartabase is built on an infinitely configurable framework that allows you to create an adaptable solution to support your unique strategy, process and culture. With ultimate flexibility Smartabase helps data driven sport and military organizations continuously leveraged latest science and technology to improve athlete performance and service member combat readiness. To see how Smartabase can help you visit fusion sport.com

More From the Blog

Dangers of Diet Culture

THE DANGERS OF DIET CULTURE IN SPORTS

Playing sports often provides a sense of self-efficacy and makes participants feel confident and capable, not to mention the various ...
Vanguard Roundtable 13 Season Wrap

VANGUARD ROUNDTABLE #13: SEASON WRAP

In this episode, we give you a behind-the-scenes look at the production of the podcast and share highlights from our ...
Human Performance Analytics: Hiring the Right Technical Experts

HUMAN PERFORMANCE ANALYTICS: HIRING THE RIGHT TECHNICAL EXPERTS

By Florencio Tenllado When human performance (HP) organizations hire staff for their analytics teams, their ideal candidates would have a ...

WHY GOOD DATA MANAGEMENT OUTRANKS MORE SPORTS SCIENCE

By Alex Campanella and Marcus Colby We often see that when elite sports and military organizations are trying to optimize ...

Subscribe to Our Newsletter

Ⓒ 2022 Fusion Sport Inc. dba Smartabase. All rights reserved.

Click to access the login or register cheese