Engineering safer football helmets, and the science behind drug overdoses

This podcast is supported by the Icahn School of Medicine at Mount Sinai, an international leader in research, education, and patient care. The Medical and Graduate School is part of the Mount Sinai Health System, one of the largest academic medical systems in New York City. Ranked among the top recipients of NIH funding, researchers at Mount Sinai have made breakthrough discoveries advancing the health of patients. Here, clinicians and scientists push the boundaries in cardiology, cancer, immunology, neuroscience, genomics, geriatrics, environmental medicine, and artificial intelligence. The Icahn School of Medicine at Mount Sinai. We find a way. This is a science podcast for February 5th, 2026. I'm Sarah Crespi. First this week, a very timely feature story. Staff writer Adrienne Cho is here to talk football and the latest science behind helmets engineered to reduce head injuries. Next on the show, researcher John Strang joins us to talk about the undone science of opioids. More than 100,000 people die from opioid overdoses in North America every year. And although much study has gone into addiction research, less attention has been paid to the biological details of overdose itself. this week adrian cho wrote a very seasonal feature story in honor of the upcoming super bowl we're going to talk about improving the technology of football helmets hi adrian welcome back to the podcast hi sarah it's nice to be here i i gotta admit here that i probably know for me. So this will be really similar to our other conversations that we have. Fair enough. Yeah, I really only know head injuries and touchdowns. I'm just very limited. I obviously would recognize a helmet, but it seems to be a passion of yours looking at the communications going on around this story. Is this something that you followed for a long time? Well, it's not. I'll be completely honest. I was just intrigued by the fact that a few years ago, suddenly you could see that helmets began changing a lot, right? I mean, obviously, head injuries have been an issue in football for a long, long time. And that really came to the fore in the 2000s with the discovery that a number of former NFL players who had died had this degenerative brain condition called chronic traumatic encephalopathy. But it was only relatively recently that if you're a casual football fan, you notice that the helmet started changing. And this got me curious. And you actually got to interview a player for the story. I did. Casey Kreider, who's the long snapper for the New York Giants, been in the league for 10 years and also trained as a science teacher. That's amazing. So we discussed the ins and outs of helmets and his choices. And he's changed helmets twice in the last two years. And I think you point out in your conversation with him that helmets are designed to protect your head when you're hit, but you don't have to hit people with your head. It's not mandatory that you hit people with your head, right? You could try to avoid that. The helmet is really the third line of defense in football. And the NFL has actually changed the rules and their practice regulations a lot to limit blows to the head. So the first line of defense is proper technique. You are not allowed to tackle somebody leading with your head. You have to try to keep your head out of it. And then the second line of defense is rules and practice regulations. So, for example, NFL teams, they play 17 regular season games. They only have 14 contact practices during the year. What are the harms like today? How often are people getting concussions, head injuries? Do we know what those numbers are like? Concussions in the NFL peaked in 2017, and they are now down from that peak by about 35 percent. So in the NFL, concussions are at a rate of about 0.5 concussions per game. It's still an issue. They've certainly made progress. Statistics available for high school suggest that concussions peaked around the same time and they're down by about 35%. A lot of this story focuses on this rapid change in the design of helmets over the last 14, 15 years or so. So we're saying in 2000, the recognition of CTE or the like long term brain damage that can happen. What has spurred specifically these big design changes over the last 14 years? This is the thing that really makes this a science story and not just a commercial product story, which is that the change in football helmets has been very much data driven. And there have been two major drivers of that. One is the Virginia Tech Helmet Lab, which in 2011 began rating football helmets. And they now rate every kind of helmet. They just did construction helmets, right? They do equestrian helmets. Yeah, bike helmets. Yes, everything. It gets a Virginia Tech rating. What Virginia Tech did starting out in 2003, they began outfitting Virginia Tech football players' helmets with sensors that allowed them to measure the linear and rotational accelerations and an impact. They had a big, big database of hits. They had a database of concussions. They came up with a risk function for concussion as a function of linear and rotational acceleration. They also studied where on the helmets the collisions tend to happen and the types of blows that the players experience and how often they get hit in each position. And they came up with a protocol to try to reproduce those hits using a two meter long pendulum. It's really cool. Yeah. In the lab. And so they take a helmet and they bash it in four different places at three different speeds. And then they use accelerometers in the head to measure the resultant linear and rotational accelerations, calculate a concussion risk. So basically this is like shrinking sensors, better ability to deal with lots of data. And now you have a good simulation that you can kind of test out these design innovations in the lab. The other driver in this has been the NFL decided that, you know, there has not been enough R&D in the helmet space. And so they decided that they were going to incentivize companies to innovate. A company called BioCore, which is a small sports engineering company that's working as a contractor for the NFL, has a similar testing protocol, but it's based on a very different data set. They use all the cameras and the high frames per second video available out there for games. Every game is covered by at least three dozen different cameras and they showed that they could extract the acceleration, closing speed, all this information from video and also from RFID tags, which are worn by every player where they are and how fast they're going. They then take all the video data and the RFID data and detailed knowledge of the shapes of the helmets and they try to reconstruct the accelerations experienced by each player. They hit their helmets six different positions based on this kind of study of how the hits occur. They also hit the helmet in each spot at three speeds and they hit them much harder because they have to capture the intensity of the hits in the NFL. So- The professional level. So instead of a pendulum, they use an air-powered ram. You know, when it hits a helmet, it hits a helmet pretty good, right? You wouldn't want to experience that for yourself. I don't think so. Absolutely not. That does bring up this interesting point. We talked a little bit about linear versus rotational acceleration and how those are both important components. Because on the biology side, if you look at how the brain acts when it's under these kinds of pressures, the rotation can be really important. One of the things that's really fascinating about the whole concussion issue is the concussion is mild brain trauma, right? And therefore, it's actually more subtle than the more catastrophic injuries. For example, the reason that a football helmet has a plastic hard shell, which was introduced in 39, was to keep your skull from getting broken. And a direct impact that just sort of sends your head recoiling without turning creates a linear acceleration. It creates a pressure wave. And if that pressure wave is strong enough, it can break bone. It can rupture blood vessels. And that's the basic first goal for a football helmet was to mitigate that. But at lower linear accelerations, the brain doesn't actually react that much because mostly water and water is incompressible and a pressure wave is not going to do a lot to a volume of water. However, at lower acceleration rates, a blow that makes your head turn can actually essentially make your brain twist. And your brain is much more susceptible to twisting because it's kind of this gelatinous thing. And so lower acceleration impacts, if they have this rotational component, can deform the brain more. And data suggests that this rotational component is important for concussion. These rotational blows can, you know, local parts of the brain stretch the tissue by as much as 20%. And it's thought that the concussion is somehow a kind of cascade of biochemical signaling that results from this strain in the brain tissue. I just keep imagining a jello mold That not incorrect right I mean Jell is mostly water And so if you had a volume of jello it would be you know if you don allow it to squeeze through your fingers it would be mostly incompressible, but it would shear very easily. And the brain shears much more easily than it compresses. So let's talk about some of the innovations in the helmet itself. Like what has changed the most in the last 10 years? If you go back to 2000 or even 2010, the helmet was mostly a hard shell with a foam liner and the foam absorbs energies and impact. The hard shell keeps your head from suffering a skull fracture. The helmets have changed in basically every aspect of that. So starting from the outside and working in, shells are now flexible. There have been a couple of ways to do this. A company called Vysis, their helmet actually has a shell that's not made out of stiff polycarbonate, but is made out of an elastomer, a nylon like material that will actually bend. And so it acts a bit like a car bumper, right, to absorb some of the energy. So it's not like a chunk of iron just transmitting the energy into your head. It gives Riddell, which is the biggest of the helmet makers, has taken a different tack. They have actually cut these slots into the helmet to allow the shell to bend. So it's still polycarbonate, But it's got a much more sophisticated thickness profile going around the head. And it has these slots. So when the helmet is hit, it can actually deflect, actually bend. That helps it to dissipate energy. I feel like there's a lot of parallels to cars here. A lot of the test concepts have been borrowed from automotive safety. A lot of the basic parts for setting up a test, like the neck and in some cases the head form. This has all been borrowed from automotive safety. There are more things like crumple zones in the helmets as well. So if we go from the shell to the inside, right, before the liner was more or less independent from the shell, it was just what you were going to stick in there to absorb energy. It was usually some sort of foam rubber. And foam is actually pretty good because foam can compress relatively quickly, but it then expands relatively slowly. But FOMA has this basic limitation, which is if you squeeze it much below half its original thickness, you run out of empty space to fill up and it gets hard. Helmet manufacturers have gone to more sophisticated things. So Shut, one of the four helmet manufacturers, they introduced this material called thermoplastic urethane. And basically they made these sort of waffle like pads that had these little pillars. And it's a kind of a gummy material. and what happens is when you push on it, the little pillars bow, but at some point they collapse. What that means is that as you squeeze this, you know, if you were squeezing a foam, eventually it gets hard. If you squeeze the TPU pad, it goes up to some stiffness and then it just levels off, right? So it limits the amount of force that you can put on your head. Yeah, that's the goal of all of this is dissipating that force. It's not allowing it to transfer to your head as the helmet is impacted. And people have now taken this much farther because now you have 3D printed materials and 3D printed materials can be designed to buckle in all kinds of different ways and have all kinds of stiffness profiles. And this is where the tailoring comes into for position or even for player. A company called Collide makes a liner for the light helmet. It's very, very elegant because it's made out of these cells. If you squeeze a spring or a foam or something like that, it all squeezes at the same time. right? The compression is uniform, but they've designed these cells so that they actually collapse more of the way, you know, when they take down an old building, the old building collapses, right? So, you know, it collapses first at the bottom and then with a little bit more pressure, it collapses a little higher up and so on and so forth. And so they can adjust the profile of the force as a function of the amount that it's been compressed. A key thing is that they can take out so much material, right? It's mostly empty space. And because of that, the thing can compress, you know, not down to 50 percent of its original thickness, but down to 20 percent of its original thickness. That gives the head more room to slow down. The collision could be stretched out over more time. All of that reduces the accelerations in the head, which reduces the chance of some sort of injury to the brain. We talked about the light helmets, and those literally are light. The light helmet company, their goal is to make lighter helmets. And then there's other ones that are getting heavier. Is that kind of the gist there? This is one of the complaints, not surprisingly, from folks at Light, that these impact tests, there is physics there to be exploited, right? You're doing an impact test, you're measuring accelerations. Well, you know, if you make the helmet heavier for the same force, it accelerates less, right? They complain that this is an issue. It's true that other manufacturers make heavier helmets. It's also true, though, that you can't make a six-pound helmet. Nobody will wear it. There's an upper limit on that as well. I mean, the players want their helmets to be manageable, too. One of the things that is fascinating when you see a real football helmet, it's a substantial object, right? It weighs maybe four pounds. That does not sound like a lot until you stick it on your head. And then you realize this thing's actually pretty heavy. My headphones probably lay less than one pound. And it's like all I think about what I'm wearing. If you're a normal person and you put on a football helmet, you might very well have this sort of initial reaction. Certainly I had this reaction is like, hey, this thing is kind of dangerous. Like it's kind of pulling my head around here. Right. You know, my neck is not used to this, especially at the NFL level. These helmets are made for people who are in terrific shape and very strong. And it's kind of striking how when you put this thing on, you kind of think, yeah, I kind of thought it would feel like nothing, right? Like I'm just sticking on a baseball cap. This is more like armor. Armor is only for the people who should be wearing armor. Because if you're just an average punter, the football helmet is going to feel heavy. How do we know that the helmets are not just better at the test? Can we see in the changes in injuries out there on the field that these updated helmets are making a difference? There is evidence that they do. In 2020, BioCorp published a paper that showed that, in fact, the helmets that rated better had lower concussion rates on the field. I think the issue is how much better is the protection? They say that they have five more years of this data and that at some point they'll publish it, but they haven't published it yet. But we do know both younger players and then also the professional players, there was a peak in concussions in 2017. Those numbers are coming down, but helmets may only play some amount of role in that because there's also these changes in regulations that you mentioned and practice. It's all good news. We just don't know yet quite how much the helmets contribute to that. I think that that is fair to say, because I think if you look at the raw scores from the Virginia Tech tests, you might expect that concussion rates would have fallen by 75 percent because the helmets are just that much better at the tests than they were before. And instead, they're down about 35 percent. And again, right, there are these these factors that are really hard to extract. Oh, and changes in detection of concussions, too, because that is also a difference. Right, which would probably drive the rate up because people are much, much more. It's messy. It's humans. But I think that the thing to keep in mind here is that even if at this point it's maybe not possible to say improvements to helmets have reduced the rate of concussion by X percent. I think it's also the case that all of this effort and all the research that has gone into developing these tests has also informed all these other changes that have helped bring the concussion rate down. For example, Steve Rosen, who's the director of the Virginia Tech Helmet Lab, one of the things that they did is they outfitted youth players with sensors in their helmets and they studied what happens to them and the kind of impacts they experience. And they found out that one particular drill known as the Oklahoma drill, which was basically one-on-one with a ball carrier and a tackler, was responsible for this disproportionate number of concussions. And so they went to the coaches and they say, hey, you know, your kids are getting hurt doing this drill. And so they they eliminated the Oklahoma drill. And that takes out a number of concussions. All this effort, whether it's attributable to the helmet or attributable to changes in practice, all of this begins with data on concussions. And so I think the thing that I hope people will appreciate from this story is that this is a serious issue. I don't think anybody involved in it is going to claim the concussion has been solved. But one thing, it's surely not an issue of people ignoring the problem. And two, it's also not an issue of people just trying arbitrary things to see what happens, right? I mean, it really has been driven by this effort to inject real data into the problem and to figure out what's actually happening on the field. This has been really fascinating. Thanks Adrian I think I learned a lot about football Thanks Sarah Adrian Cho is a staff writer for Science You can find a link to his story and the amazing graphics that illustrate some of the concepts we discussed at science slash podcast Hi, Science Podcast listeners. This is Kevin McLean. I'm one of the producers on the show. I just wanted to hop in here before we get started to ask you to consider subscribing to News from Science. Every week on the podcast, we bring you one of the stories that the News from Science Team is published, but there's so much more than what we can cover on our show here. For only about 50 cents a week, the money from subscriptions goes directly to supporting nonprofit science journalism, reporting on science policy, investigations, international news, and the latest breakthroughs from all around the world of science. Support nonprofit science journalism with your subscription at science.org slash news. You have to scroll down and click subscribe on the right side. That's science.org slash news. This week, we're going to try something new and share some of what else is going on on the site. These are all items that caught my eye in the sibling journals and on the news site and in science. There's so much good stuff. And sadly, we just can't get to it all on the podcast every week. We just do two segments usually. First, I want to mention in science, we have a book review by Robert Brooks on two books, The Intimate Animal, The Science of Sex, Fidelity, and Why We Live and Die for Love by Ayushi Agarwal, and another book, Bonded by Evolution, The New Science of Love and Connection by Tammy Blake. The reviewer notes that this pair of books focuses on the science of modern love and connection, rather than the more heavily traded topics of first impressions and sex. I just love even just reading the titles of the books that we review. There is so much fascinating science being done. In advances, I actually thought about covering this paper for the podcast. It's called The Observation of Superballistic Brownian Motion in Liquid by Boynowitz. The one sentence summary is basically the motion of mesoscopic particles in liquid reveals non-equilibrium memory effects in fluids. If you've ever had to think about Brownian motion for your research, you will be interested in this one. Last, I wanna point to Hannah Richter's news story on the future of freezing rain. She writes about how climate change may increase freezing rain in areas. And she talks about the research that's going into predicting freezing rain zones to come. Stay tuned for our next segment about the questions researchers should be asking about opioid overdose, I had a partner with drug-addicted people to find solutions. This is an Expert Voices column. Now we have researcher and clinician John Strang. He wrote an Expert Voices column last month on the undone science of opioids. Hi, John. Welcome to the Science Podcast. Really good to meet with you. So I know there's a lot of deaths from opioid overdose in the U.S. and the North American numbers are huge, like 100,000 people per year die of opioid overdose. And that's up a lot from, you know, in 2000, for example, when it was closer to 20,000 per year. Is this a trend around the world? You know, what's the geography of this problem? We've seen a similar phenomenon in many countries in the world, but nothing on the scale that's occurred in North America. And it is North America, so it's Canada as well as the US. You can think of it as like three epidemics that have occurred over the last quarter of a century. You have an early epidemic from 2000 onwards of prescription opioids, the OxyContin story. Oh, right. And those prescription drugs were contributing to deaths. When controls brought in on that, you then had a surge in illicit heroin deaths. And just as that began to subside, you then had fentanyl coming in. And fentanyl had originally been prescription fentanyl, and then it became illicit fentanyl. So it gives scientists something a bit complicated to untangle, is which bit of the problem are you trying to deal with and which bit are you trying to intervene with? Can you talk about how does someone die from an overdose from heroin or fentanyl, any of these other opioids? You would have thought as either medics or scientists would be able to give you a really good answer to it. We can put together an answer, which is reasonable, but we haven't really studied it that much. how to study how somebody dies from an opioid overdose is tricky to design the study. That's part of what this Expert Voices contribution is about, that we should be able to do better. Globally, we have really good scientists who have the ability to design studies to understand what happens when too much opiates have been taken. you obviously don't want your subjects to die, but you can do a little bit of overdose and test. Well, what went wrong? What we're fairly sure about is that what occurs is the breathing, the centre in the brain that sends the signals so that you and I, we breathe at something like 12 breaths per minute, and we don't have to sit there thinking all the time, must remember 12 breaths per minute. It just happens. If that gets turned off, which is one of the things opiates do, your body just forgets to breathe. And if it forgets for too long, you're dead. Now, usually what happens, we now know from having created mini overdoses and studied them, is that people go into a period of maybe up to a minute of just not breathing, and then they gasp and breathe and kick back in again. They may go back into not breathing a few minutes later. There might be some mechanism, a backup that's not shut off yet or something like that. That's exactly it. And it's those aspects that we really do need to understand better. What is it that kicks in and says, enough's enough. This has gone too far. What if you could target that? Yeah, super interesting. I want to point out that there's been a lot of money that's gone into studying the public health aspects of this and addictions research, especially what makes someone addicted? How do you break up that process? But the piece of this research that you're talking about that needs to be done, you mentioned the actual mechanism of death, what might be action there. Are there other areas that just really aren't understood? Absolutely. And for example, questions that we can't really answer is, does it vary from one person to the next? Is it that anyone who gets involved with heroin use or fentanyl use, this is a risk that's consistent across all people using or some individuals more at risk than others or a regular user's more at risk or maybe even less at risk than occasional users. You really could study whether your protective system that we talked about a moment ago, are you an individual where that gets downregulated very easily, in which case that sounds pretty worrying? Or do you have a very resilient response system? Those are the sort of areas that you go, wow, I really want to know more. And then neuroscientists or brain imaging scientists, they'll be able to say, hey, I could come in here and I could work out how to image which bits of the brain are active and which bits seem inert. You've then got a model where you could say, hey, I wonder if we could modify this. I wonder if there are things that could be done that kept your breathing pattern more regular, or if it went wrong, you could test an intervention that reversed it. There are drugs that can reverse overdose. You know, are they, could they be better? You know, is there research that could be done there? They work pretty well. And one of the bits of good news about the opioid overdose problem is that once medics are there, we know exactly what to do. It's one of the areas of medicine where we have a real antidote. naloxone's the one that's most famous, which in North America is now under the brand name of Narcan. It's an extraordinary drug that reverses the effect of the heroin or the fentanyl. And you go, ah, okay, so what happens? Could you measure different ways in which you gave the naloxone? Or how much do you need to give to get the breathing back again? Those are areas where you think you really can recreate it in a laboratory and give better answers to the policymakers, to the clinicians, to the families, to the people actually involved with the drug use and say, hey, let's make sure you understand what the risks are and you understand what you can do to intervene. What are some of the barriers here? You talk about the different kinds of scientists that could be involved in this research, but it sounds like it's not very collaborative across disciplines? It isn't particularly a funding problem because globally as a society, we've invested hugely in improving our understanding of the development of drug problems and addictions and the complications that arise We now have a much much better understanding than we have had a quarter of a century half a century ago But this area it as if we feel squeamish about actually studying what happens when an overdose is created. And yet we absolutely need to know what happens so that we can develop better interventions. And all of the scientific disciplines that you have, there's probably a contribution they have in this area where it hasn't been captured before. It's a big challenge to study heroin overdose, but it's not impossible. How would you get a participant? Would you dose them with drugs or would you follow them if they're already kind of consistently taking the drug? The most obvious way to do such a study would be to look for people who are already actively engaged in heroin use. And that seems shocking to many people in the general public and to many people in the scientific community. But it's what we do with all sorts of disorders. If you want to study what's happening during the early stages of cancer development, you take somebody where unfortunately that's already developed and you then negotiate with them, can I look at different processes that are going on in you whilst we work out how to treat you? It would clearly be unethical to create the cancer just to do the study. But by working with those who've already been afflicted by that, you can say, let's get some science whilst we're working out what the intervention is. And that's broadly the approach that we take in looking at how we study heroin overdose, or it could be fentanyl overdose, could be other drugs. That's got challenges. We've developed in the UK, in my centre, a heroin overdose laboratory. We can bring a subject into the ward. We can wire them up to all the measures that we want to capture. And we're doing this with people who are regularly using heroin. And we're then studying what happens when they take their normal dose of heroin, what if it was a slightly larger dose or a slightly smaller dose, how does that alter their breathing patterns? And from that, we've identified these extraordinarily long apneic episodes. So essentially, episodes when you're not breathing at all. Your breathing doesn't just become shallow, it moves from a normal breathing pattern to complete silence, and then maybe 50 seconds or a minute later, you gasp and take a few more breaths and maybe drop back into not breathing again and then gasp and breathe some more. First of all, we can explore why is this occurring? And secondly, we're in the era of wearables and apps. Why can't I have a sensor that detects that suddenly I'm not breathing? My rib cage isn't moving. Maybe it wakes you up or stimulates that in some way, Yeah. And so from a pure study point of view, you go, I want the data to be able to feed into my computer. At the human compassionate level, you say, hey, couldn't this be the basis of an alarm that suddenly rings or an alert that goes to your partner or your parent or something that sends your geolocation off to the ambulance services? If we can do that with the elderly who might have a fall, why couldn't we reconceptualize that with overdose detection that, first of all, sounded an alarm so that the individual was brought back to regular breathing? Or if that didn't work, alerted either a family or an ambulance call out. So I can kind of see how some of this would eventually integrate with public health questions or addiction questions, even if you're trying to get the basics of what is happening during an overdose, what makes these drugs dangerous? You know, when you talk about sensors and responders, but you do have to deal with the stigma here. It is a challenge because within the drugs field, there is that overlap of the instinctive disapproval of the behaviour and not wanting to be condoning the behaviour, quite apart from its illegality, the inherent risks in the behaviour. But it's only in this way that we're going to be able to understand what's bringing about the risks. If I was a parent or a partner or a friend of somebody with an addiction problem, and I was aware something critical had just occurred in their bedroom, I've gone in and they're unconscious. As a family member, I want to know what can I do whilst I'm waiting, whatever length of time it is, for emergency medical services to arrive. It's one of those areas of science where you can very easily see how that would guide the interventions you then develop. If there was a lot of effort put into researching these questions, what do you hope to see come about? What would ideally come out of understanding these processes better? I think there's three immediate benefits that one could see emerging pretty quickly. First of all, just from studying it, we would get a much better understanding of which bits of brain and biology are getting thrown off kilter and leading to those sudden onset overdose deaths. Secondly, there's a real potential to develop sensors that detect that crisis has occurred. And if you could develop a sensor that detected that the rib cage wasn't moving, or if your oxygen levels had suddenly plummeted, then it's pretty easy to see how you'd hook that into an app that triggered an alarm. And thirdly, it allows us to test new medicines that are coming out. We're testing new forms of naloxone, which would be much more portable and easier to carry with you. And how well can you reverse it with smaller doses or larger doses? So those three areas, understanding the problem better, developing an alarm capability, and thirdly, actually testing some of the responses. Where do drug users fit into this? Do they need to become kind of more partners in research? I can see it being difficult if you're living with addiction to also participate in research and maybe being concerned that you'd be exposed to criminal charges. Yeah, it's a really tricky issue, but it's an important one for us to find ways of dealing with this, that individuals who are involved with this opiate use behaviour, they have so much to teach us and the partnership will allow us to be more imaginative in how those interventions could be developed. I personally find it hugely rewarding. I'm actually a clinician, I'm a medic, as well as a scientist, and I find those meetings with drug user representative groups or with individual patients, often very illuminating. One of the active ones at the moment is about how severely should we reverse an overdose to bring somebody around. And the normal response from me as a doctor or from a paramedic or from family members, just give the maximum amount you can, make sure you reverse their overdose completely. There's a counter argument which a drug user might give back to me and say, hold on, if you do that, you're going to throw me into full-blown cold turkey just in a minute. And I know how to deal with cold turkey. I just have to go out and take some more. And so you end up with a real undampened oscillation of an overdose has occurred, you've over-reversed it, they then reverse your reversal, and you go, this is not a very clever thing to do. But that joint working with users, also, if we're talking about wearables, you're going to have to work out with them. What's the way in which we can offer it to you in a way that you're willing to be tracked? There's a secretiveness to all of this that's hard to bridge. What's the point in developing the understanding and developing the technology if the entire community of people involved with using say, I'm not willing to wear it. So that co-production, I think, has to be at the heart of the way you develop the studies and work out how it should then subsequently be implemented. All right, John, thank you so much for talking with me. It's been fascinating. Okay. John Strang is a professor in the National Addiction Center in the Institute of Psychiatry, Psychology and Neuroscience at King's College London. You can find a link to the Expert Voices column we discussed at science.org slash podcast. And that concludes this edition of the Science Podcast. If you have any comments or suggestions, write to us at sciencepodcast at aaas.org. We do read everything that comes in. To find us on podcasting apps, search for Science Magazine or listen on our website, science.org slash podcast. This show was edited by me, Sarah Crespi, and Kevin McLean. We have production help from Podigy. Our music is by Jeffrey Cook and Wen Khoi Wen. On behalf of Science and its publisher, Triple S, thanks for joining us.