It's true that some things change as we get older. But if you're a woman over 40 and you're dealing with insomnia, brain fog, moodiness, and weight gain, you don't have to accept it as just another part of aging. And with Midi Health, you can get help and stop pushing through it alone. The experts at Midi understand that all these symptoms can be connected to the hormonal changes that happen around menopause. And Midi can help you feel more like yourself again.
Many healthcare providers aren't trained to treat or even recognize menopause symptoms. Middie clinicians are menopause experts. They're dedicated to providing safe, effective, FDA-approved solutions for dozens of hormonal symptoms, not just hot flashes. Most importantly, they're covered by insurance. 91% of Middie patients get relief from symptoms within just two months.
You deserve to feel great. Book your virtual visit today at joinmidi.com. That's joinmidi.com. Welcome to the Medicine and Science Podcast from the BMJ. I'm Cameron Abassi, Editor-in-Chief. Today, we're talking about making the world healthier.
The Dutch aren't more active because they're an active culture. They're more active because in Amsterdam the fastest way to get around is cycling and it's built into life there. And if any of us moved to Amsterdam we'd probably cycle a lot more too.
Later, we'll hear about how we can take AI tools for preventative health out of the lab and into the real world. There's some really cool technologies out there already using data for patient-centric care, but it's being sold directly to the customers and not integrated with hospitals or the NHS more broadly.
First, I'm joined by Devi Sridhar, Professor and Chair of Global Public Health at the University of Edinburgh. Devi is the author of a new book, How Not to Die Too Soon, The Lies We've Been Sold and the Policies That Could Save Us. Devi, welcome to the podcast. Hi, good afternoon. Tell us more about your new book.
Yeah, so my new book I've been working on for a couple of years, and it takes on the self-help longevity literature, which is basically if you do your gratitude journals and you sleep enough hours a day and you take enough supplements, you can live to 100 or 150. It says, well, actually, what do we know from public health? Why is life expectancy actually going backwards in many parts of the world?
And what do we know from over 50 years, I look at the past 50 years, but beyond that of public health research on what makes people die too soon and the role of government and actually helping us live longer? Yeah, I suppose one thing we should perhaps clarify at the outset is the difference between extending lifespan and life expectancy. Because from my understanding of it is that try as we might.
Over the last few centuries, we've done very little to extend the possible lifespan of a human being, whereas we're constantly making improvements in terms of life expectancy. Is that understanding correct?
Yeah. And I guess there is that view of can we live longer? And, you know, Brian Johnson and multimillionaires like that who are saying, can we push the human body beyond 100 to 120 and expand lifespan? Where for me, I'm more interested in the collective approach of how do we get all of our life expectancy higher? What do we know about that? What are places in the world?
We call them blue zones, which seem to kind of have higher life expectancy. And what can we learn from that for all parts of the world or even from the way we live our lives? And that could be the communities that we're part of or the political parties we vote for. Yeah. So we're talking about how do you maximize your life expectancy in relation to a pretty static lifespan. So all those billionaires out there who want to live to 150,000.
And the social media influencers, you probably want to think again about that particular point. And there's also the, let's get the other issue out of the way, which is around, there was a theory, I mean, I think it's still a well-described kind of hypothesis that as we advance as a society in terms of our healthcare, et cetera, that we will have a,
of morbidity, which means that you will live a longer, healthier life. And I mean, the ideal in a way being that, say you were to live to 90, that you would live 89 plus a few months years of healthy life. And then the years that you're ill are very much, or the months that you're ill are very much compressed into a short period. And unfortunately, what we're seeing is that rather than that happening, we've got people, although life expectancy is
has gone up over the, you know, over the century. We're also seeing people living for longer years with illness. Yes. And also seeing even something like cancer, you know, early onset cancer. Why are we seeing that the birth cohort effect, which whereas every, you know, younger generation of people has a higher rate of cancer, you know, people in their thirties and forties now. And it's not just in Britain, that's across G7 countries, right?
So, yeah, we've made progress. I think something like dementia, we are pushing it further and further back. And, you know, that's really positive. But I think on other things, type two diabetes, hypertension, cancer, as I mentioned, other other factors linked to that. We are going backwards and it's not just Britain. What's interesting is looking at countries like Ghana,
which had a 650% increase in obesity and now having an astonishing rise in cancers. And they're a country which is trying to, you know, which was applauded for reaching the MDGs and reducing child deaths, but now they're faced with another challenge. So,
There is a it's a very complex and health situation and one that unfortunately means that we're not, as you're saying, getting healthier in those later years. We're actually needing more and more support and care and having more chronic. Yeah. Also, you made the point about both cohort effects and earlier detection. Some of that, of course, is because we're able to detect cancers earlier at an earlier stage as well, isn't it? I'm not saying that's the entire cause for this, but it is a it is a factor.
I think it's a factor, but I don't think it fully explains why we're seeing it. I think there is something in terms of if you look even at death rates of younger people to cancer or rates and they're all they all seem I mean, where I've been convinced is by the research from from Harvard, which looks at the gut microbiome and saying, is it something in the foods and the diet that shifted in every single generation? Yeah. Which means that we are seeing a rise because they are generally linked somehow to the digestive system. Yeah. And we're also seeing it across a number of
G7 countries, which you wouldn't expect that, you know, to have that if you're saying, well, actually, it's because in the UK, we have better diagnosis earlier, earlier tools. So I think there's something else going on there, which is more related to diet would be my my hypothesis, though, it's just a hypothesis so far. But tell us more about that. I mean, because that's a central theme here. What is the what is the role of diet and the digestive system, as you describe in your book?
massively. So one of the places I look at when I turn to diet is Japan. And I start with Japan, because if you actually want to learn to live to 100 or maximize human life expectancy, Japan does a remarkable job of the number of people who seem to make it to 100 in pretty good health, to the point that actually make it to 110. You know, Okinawa is kind of
looked at as one of the islands where you're like, what is happening out there? And then you look at what they eat and then you look at their diet and then you start doing experiments. And I go through them in the book in rats where they start feeding them different types of diet. And all of a sudden those rats, which are fed a typical Okinawan diet or a typical South Korean diet, all of a sudden metabolically look very different.
to the rats being fed. So I guess it comes from the observational studies. Then it goes and actually to, you know, animal model studies. And then from that, you can say, well, actually, there's something in that about what we eat affecting how long we live, our health outcomes. And
After reading all that and studying all that, I was just like, how can we be giving children in Britain in their school lunches 80% of things that are ultra-processed and which we know are bad for them? And so I think that's where it comes full circle to policy. Because they're like, the knowledge is there. The studies have been done. And it's not saying we have to make school lunches look like Japanese school lunches. That's not the point. The point is actually diet affects your health. And we know there are things that right now children are eating
why we're seeing high rates of child obesity. And we're not looking at the underlying drivers of this, which is what should we control for something like school lunches, what's made accessible. And if anything, in Britain, Brexit's made it worse. We know the price of fruits and vegetables has gone up. It's more unaffordable. I mean, when you began by talking about diet, you also...
You mentioned the role of industry. I mean, of course, there are national policies that play a role. Food policy plays a role here. There's also the role of industry. I mean, what's your view on that in terms of promoting and manufacturing ultra processed foods and non-healthy diets?
Well, they're a business and they're there to maximize profits. And I don't think, and I think this is one of the problems is that they make a profit off of selling these products, marketing them, targeting children. There's a reason why 80% of most school meals in Britain are ultra processed foods. Those are, it's a huge market, an existing market. At
At the same time, the health cost of that and the cost of that is paid by individuals and their families. They're not actually responsible for any of the cost of that product and where it falls. And this is where you need strong government. But lobbying is the issue. And this is why I come back to the policies, because it's too simplistic to say, oh, British people are culturally different than Japanese or are lazy or are uneducated or any of those kind of things you hear. You have to look at the food environment.
Because I'll tell you, whatever kids in school is put in front of them, especially here in Scotland, we have free school meals till kids are nine or 10. They're going to eat. And that's what all the other kids are eating. So if you're putting in front of them something that's deeply unhealthy, and then you have children becoming obese off the back of their diets, how can you in any way say it's the child's fault? It's government and it's lobbying and it's food manufacturers, manufacturers.
who do move when they actually have to, but that's when government comes in and says, actually, this is unacceptable and these are the health costs of it. And I think the research is pretty conclusive on ultra-processed foods right now in terms of everything I've seen over the past, especially five to 10 years, on the health implications of a diet, largely including those. Yeah. So now you've mentioned government, and I can't help resist asking you. I know you're from the US. I mean, Claire, this is...
Children, diet, food industry is a feature of the plans around the Make America Healthy Again movement and what Robert F. Kennedy Jr. has been saying over the last weeks and months. What's your take on that analysis that children in the US are probably less healthy than they've ever been almost? And I guess you're agreeing about the role of ultra processed foods.
Yeah, I mean, the thing is, is that there's a lot of this Make America Healthy Again movement. There's like a nugget of truth and then there's a whole bunch of nonsense. And so it's difficult to just like to get rid of all of it because within it, you're like, well, they are saying some valid things. But again, the solution is very different. The policies that they're bringing in increase inequality. We know actually the problem is inequality. They, you know,
decrease, you know, the ability, for example, if you're on food benefits to actually access healthier options. And so in a way, it kind of is removing it, like, and putting it into a bubble that's completely abstract to the reality of people's lives. And actually, the reality of people's lives, and this is why in Scotland, if you go to areas where are generally, you can almost link house prices to child obesity. Mm-hmm.
I mean, it's that tightly linked. So to go into these areas in America and just be like, we don't know what's causing this and we don't know what to do. I'm like, you do know what to do. And I'll tell you, if he turns around, if our kid junior turns around and says, actually school meals across the country going to be healthier options, we're putting in the money, we're going to actually invest in it. I would be gobsmacked because right now it's just all
all about influencers, make a TikTok post, tell everyone, eat your fruit and veg. And that's what you have to do. And if you have the willpower and you're a good parent, your kids are healthy. And if you don't, you need to go watch more TikTok videos and go to the gym more. And I think that's my worry about wellness culture. It's become very individualistic and very abstract from the reality of people's lives, which is
I mean, child obesity is never a reflection of anything else than poverty, lack of opportunity and needing to make healthier options accessible, available and affordable. Yeah. Let's then talk about wellness culture that you've just mentioned, which is, of course, a very prominent feature of social media. What's your analysis of it? I know it's a feature of your book.
Yes, I talk about this because I'm actually like a trained PT and part of like, you know, the gym community. And I think a lot of it's really fake and not helpful. So for example, you know, the idea of a thigh gap and thinspiration and that you, you know, lose enough weight and you have a thigh gap, that means like you're somehow healthier and better off when the research shows that actually...
thinner thighs are actually linked with worse health outcomes because it shows you don't have enough muscle because you need a certain amount of muscle on your legs to be able to go about your day to actually be in certain health. But you won't, if you have that muscle, you don't have a thigh gap or the aspirations for a six pack. I think there's now more and more uncovering of how much is fake and
on social media. And it's great to have accounts which are authentic. But I think I'm just surprised by how much it's also I talk about in the book, like the commercialized FOMO of like, buy this supplement, and you'll look like me. Yeah. Or, you know, do this hack, and you can look like me. And I'm like, it's just selling you something. And that's not real. And I wish health was more, you know, people who would say, what can I do with my body and accounts that were
trying to promote kind of idealized versions of health and well-being and more accessible ones, which actually shouldn't be individualized. It should be the Dutch aren't more active because they're an active culture. They're more active because in Amsterdam, the fastest way to get around is cycling and it's built into life there. And if any of us moved to Amsterdam, we'd probably cycle a lot more too. That doesn't mean we suddenly change who we are. It just means that. I think that's where I get frustrated with the social media influencer world.
world, which I'm also kind of part of. I'm just as guilty as making those kind of posts and then afterwards, you know, being self-critical about what am I posting and is that actually valid and helpful and actually making people feel good and not bad. Yeah. I mean, having just been in Holland, I entirely agree with what you said. I think we better explain for anybody who's over 25, over 30, what commercialized FOMO means. I only got FOMO explained to me by my children.
a few months ago. So I use it all the time now, but perhaps some listeners won't know what it is. Okay. So FOMO is fear of missing out. And I call it commercialized FOMO because it means if you just buy something and it can give you that FOMO that if you don't have that product, you're missing out. Yeah. So it seems a hard one to kind of beat this one because misinformation, disinformation, commercialized FOMO, my new go-to phrase.
These things are out of control, aren't they, Devi? I know we might think, oh, Devi Sridhar, we should do what Devi says. Of course we should do what Devi says. But not everybody knows that. And not everybody knows when they see a TikTok video or some influencer talking about something that what they're saying is entirely misleading.
Completely. And I'm not saying to say to do what Debbie says. I think all I do in my book is actually to make people think a different way. That's all I want people to take away to be like, let me rethink this in a way that I haven't thought about before. But I don't know. I mean, the rise of social media has had many positives. But in terms of health, well-being, longevity, I think it is really confusing to know what to believe, what not to believe.
And especially, you know, when posts give you that clickbait that we all want, which is get a six pack in three weeks before your holiday. And you're like, oh, I want to do that. I want to look like that. It's very tempting. It's very tempting. You see these videos and think, hang on, this sounds great. And then you think, hang on, I'm meant to be the editor of the BMJ. I need to look at this critically. No, it's very hard. I think the issue we're facing is how do we...
help people understand this is trusted information. This publication is trusted. This person is trusted. And that's what
what we try to do with everything we do at the BMJ too, and we get it wrong sometimes. Everybody gets stuff wrong. I mean, you do. I'm sure you sometimes not quite get it right as well. But it's, you know, you can trust this person more than you can trust or this publication more than you can trust other resources. And it's that challenge that I think we've, because people are seeking health information and wellness information like crazy online.
Yeah. But I think there it's so tricky because look at the anti-vax movement, the people who come out and being are tend to be very charismatic. Yeah.
you know, saying I'm on your side, I'm trying to protect your kids. And they're actually probably more eloquent than anyone in public health or in medicine could be about about vaccination. Look at the measles outbreaks in the States and just take your vitamin A and your child will be safe. So it is very difficult because even who you trust now has changed. The biggest shock for me in the past kind of year is that, you know, Joe Rogan became the voice of measles response saying, oh, everyone had measles when I was a kid.
His platform's massive and he's trusted. So what chance does anybody else have in that kind of world? And so it's, for me, that's really the challenging thing is how do you cut through? And I think all you can do is try to come back to data analysis, peer review, logic, kind of looking at this and trying to kind of get people along on that path of anchoring them
But there's no, unlike the BMJ or even newspapers or books, there's no editorial standards online. No. So you can kind of say or do anything and...
You have free reign where you wouldn't be able to do that for a newspaper or on a TV station or in books traditionally because you would just be having to do a retraction. And that just doesn't exist online, which is a problem. Yeah. OK, I'm glad you introduced data and evidence because we're going to have a talk about AI in a moment. But I've got a couple of questions to you before we turn to that. Number one, just tell us a few of the other issues that you touch upon in your book, Debbie.
So I go through mental health, which I think is a major cause of actually people having poor quality of life. I look at clean air, the rise of air pollution, clean water, access to medical care, whether it's NHS or other systems, because in the end we need to see that, as well as one that'll be less relevant to a British audience, but maybe interesting, is gun violence.
which is the leading cause of death of American children and adolescents, and talk a bit about what have we learned about how do you create policies to keep people safe from each other, especially from violence. So yes, I cover nine factors. I go through each, looking across the world, and tell a bit of a personal story in each of my interactions
and how I've tried to deal with that individually, but also say at a certain point, you know, in the chapter on medical care, when I needed the doctor, I couldn't treat myself. In the end, you're completely reliant on the state. What's your healthcare system? Can you afford it? Is there someone there? What are the waiting lists? And it makes you suddenly realize why policies are so important. It's kind of to get away from the cynicism of politics towards...
You know, most of the progress in health, if not all, has been driven by governments, by politicians. And so actually they do good, too. And they have done good over decades. And so, you know, give that a chance instead of this thinking that nothing ever changes because a lot has changed for the good, even though we have blips, you know, right now.
for example in the states where you do feel things are going backwards slightly the trajectory is generally hopeful okay okay it's good that you're sounding positive now you mentioned government and so finally when we look think about the uk government in particular you've got five arse of them in your book um so cheaper fruit and vegetables increased nhs funding for primary and public health and greater provision of mental health support i think you know
They're big areas that I think most people would entirely agree with you on and would be relatively familiar with why we need to do that. But you've got two less predictable ones. One is renationalisation of the railways and the other one is renationalisation of water. Why those, Debbie?
Well, if you read the book, you'll see why they're connected. But on... Well, I'll just say on water, in Scotland, water is nationalised. Scottish water is a public utility. Everything's better in Scotland. We know that. Well, yeah. I mean, I have to say, except in Scotland. But...
My point is that it's not crazy. It's done here. And actually, if you look at accountability, water is not a commodity. You die without water, without clean water. And so to make water a commodity in the hands of for-profit companies that have largely owned foreign companies that are trying to extract its maximum profit, I don't understand why...
Anyone would want that model versus actually water being accountable to the people who live here. And Scotland, the shareholders are all of us. And on railways, the idea of this being that, how have we made it cheaper to fly than to take trains? And if you look at the places that have active travel that people can get around without relying on cars...
or flying, it's generally places that have really strong railways that are affordable and they're generally publicly owned. And so it is all connected, but these are political decisions. And I try to show in the book how your life would be different if actually these big decisions were made and how it would affect you. And give examples of actually where, you know,
you know railways are nationalized and work really well and where water is nationalized and works really well and guess what those societies function fine and it's not they're not communists they're they're just different ways of organizing yeah okay well they sound like two big incentives to read your book and lots more besides which you've been kind enough to talk to us about um debbie your book is going to be published when
So it's out June 12th. So in two weeks and yeah, I hope people just take a shot to this thing about things slightly differently in terms of how we approach our health. Or if someone's frustrating you by kind of saying to you, politics don't matter. I'm just going to be fine on my own to say to them, actually just take a look at that because yeah, just, I have had a few kind of health influencers who've had a look at an early copy and be like, Oh, I never realized my health was affected by other people. And I was like,
Yeah, it is. I hope they're posted about that. And analysis on BMJ.com is looking at the use of AI in the community to prevent elderly people being admitted to hospital. I'm joined by John Downey from the Centre of Health Technology at Peninsula Medical School and Martha Lee from NHS Devon Integrated Care Board.
Welcome to both of you. Let's first of all go to John. John, I know you're a public health expert as well. First of all, before we talk about your paper, what did you think of Devi's case for healthy living? Yeah, I think that she stressed some of the major challenges we face, particularly around the difference between life expectancy, and you mentioned that, so lifespan versus healthspan.
And that's a really important area, that gap between living longer versus living better for longer. And this is this shift that the NHS talk about and what we are talking about here, which is that shift to help people live in the community independently longer for better. And then Debbie talked about this sort of continuing burden and increasing burden of metabolic issues. So we're getting better in terms of managing the burden of disease and
But we're not getting better at managing some of these sort of long term conditions, particularly around type two diabetes and obesity. And that actually is worsening and it's worsening across the globe. So not just in the sort of high income countries as Debbie was talking about. And then those sort of health promotion or healthy lifestyle factors, which for a long time we've placed the blame with people.
And again, the sort of shift left, what we mean by that is shifting to preventative, anticipatory and cheaper versions of care is really important, but that shouldn't be to sort of
say we've given you the tools we've educated and you haven't done it, it's your fault. There are some wider determinants of health, which we need to consider. And Debbie mentioned them as well, particularly around food environments, the built environment, sort of socioeconomic status. So these are things which we talked about for a long time, but we're not necessarily doing a great deal of getting a handle on. And this is where, like you spoke about,
Moving to the integration of digital innovations can start to turn the tide on it, but not necessarily an easy fix by any means. And actually, we sort of see a widening of that gap. Okay, well, thank you. Martha, did you have any thoughts to add?
Yes, it was really interesting hearing about Debbie's piece and particularly the inequality level, the link between healthy life expectancy and poverty, because it's something that we've been really looking to tackle in our living lab down in Plymouth, which we've been setting up as a co-design space.
for people in social housing looking at assistive technologies in the home that can help people living at home and keep them there for longer in their own communities, which is obviously better for everyone. We're very aware that actually there's some really cool technologies out there already using data for patient-centric care, but it's being sold directly to the customers and not integrated with hospitals or the NHS more broadly.
So it's very much targeted at middle class patients who can provide their children with that reassurance that they're healthy and at home and living normally. Whereas actually what we don't want to create is this two tier health care system. So looking at how to integrate these new technologies, this use of data can really help provide a fairer approach to health care and social care as well.
Okay, tell us about Living Labs. What are they and what do you hope to achieve with them? So the Living Lab is essentially a laboratory in the real world. What we've got in Plymouth is one of the largest in the world. So it's 37,000 residents of Plymouth Community Homes. So that's our partner, which is a social housing provider in Plymouth.
And it basically gives us the option of looking at the effectiveness of assistive technologies or other interventions in people's homes with them.
So we've got basically every cohort you can imagine. We've got people at risk of homelessness. We've got frail people at risk of frailty. We've got people who are type two diabetic. We've got people with multiple long term conditions. And we've set up this living lab where anyone who wants to come and test their technologies can come down, engage with us and work out how to build a real evidence base for what they've got.
And also just design better products that actually suit people in the real world. And then that's that side. My side is then going, great, this technology works. How do we make the most of it as a system, as a joint social care, health care, voluntary care, dom care system to actually enable effective outcomes for patients? Devi, how do you view the role of technology and AI?
In terms of the arguments you're making? Well, I don't really consider them in the book. But what I do talk about is screening, which I know your paper looks at, John and Martha, which is like actually how at an early stage do you pick up early signs of possible disease or when someone might need to get medical care? And I think I talk about it probably in a more
20th century model, which is actually can we use pharmacies and PTs and people to take blood pressure or waist circumference or different blood tests. But actually the idea that you could use digital devices to take this, especially in high-risk groups, the elderly feed that into AI, which could then predict to you, actually this person is at risk.
That's really positive because in the end, what we all want is people staying out of hospital. We want them healthier for longer. And the best way we know that is to strengthen primary and secondary prevention. I think about it and I talk about it in the book, like almost like a football pitch. And the medical system is the goalie just trying to save, you know, the person who's gone in. But the defense is secondary prevention.
And then the rest of the pitch should be primary prevention, who are trying to make sure. So by the time someone gets to hospital, it's because they've gotten through primary and secondary prevention. And this, to me, seems a real step forward in that secondary prevention to make sure that we detect things earlier. But yeah, I think my book will be a historical relic because it doesn't mention AI in the way that everything seems to mention it from now on and which the way the world is going. But it shows how quickly things are changing and it's how can we
integrate it and use these for the way that the paper describes to kind of predict, prescribe and prevent and look forward. Okay, Debbie, thank you. So this paper about AI in the community to prevent the admission of elderly people to hospital. Why is that a problem, first of all? Why is it a problem we should take seriously? You know, one of the biggest admissions to hospital is frail older people. We know that falls are a huge problem.
predictive factor on social care costs and not living independently. So we have this. And then obviously everyone knows the sort of news of ambulance queuing up outside of hospitals. But there is early signs that this is happening, which are getting missed coming back to Debbie's point.
you know, sending ambulances out to people who've been lying on the floor for 24 hours is avoidable. And this is an intrusive data. This is data of subtle changes in their behavior. This is the fact that they might have had two or three falls and not told anyone, which could be picked up pretty easily by motion detectors. These are sort of subtle changes where you could
anticipate when someone might need care and prevent the situation where we are overloading an already unsustainable model of acute settings. So the theory then is that we use AI to identify people at risk of falls. Is that correct?
Yeah. Should I run you through how it works? Yeah, let's do that. Brilliant. So there's some fantastic solutions out there already. There's a few academic programs looking at it at the moment too. But on the market, as we speak, there are passive sensors that will collect data points, build up the picture of somebody's day. Once...
things start to change a little. So if we note that she's getting up a bit more at night, which is a real predictor for deterioration, or she's drinking less water, which is another really big sign of changing health, that can then flag through machine learning to a team in the community and say, actually, Mary could do with a check-in. Let's deploy one of our staff. Let's actually potentially refer her to one of our extra services. And we can get in before she has a fall or before she ends up with any other problems.
So that way we can keep her healthier in her house. That's really exciting because we've picked building up a patient centric profile.
Beyond that, we can have a cohort of similar people. So if we're talking frailty, we can go, actually, this person is similar to these other 30 people, either in their neighbourhood or a bit wider. And we can build up patterns of that and start looking at the interventions we're using and actually get feedback about the effectiveness of those interventions, which is really exciting because the idea of getting feedback about how you look after patients is incredible.
worryingly new, actually. It feels quite radical somehow, particularly if you're looking at more social care type adaptations like housing changes. If you're looking at a council who naturally don't have all that many resources, knowing if a £2,000 adaptation to someone's house is as effective or even more effective than a £20,000 intervention is really valuable information to have. And that's the kind of feedback that I'm getting really excited about being able to apply. There are a number of pilot programmes that
And what have we learned from them? We've learned that they're really effective. What we unfortunately haven't got at the moment is a standard way of incorporating it into the NHS and social care. So we've got, at the moment, what tends to happen is these triggers, if we've noticed that Mary's looking poorly, we let her key relative know. And sometimes that data can get fed into the system sort of ad hoc. But the key relative knows that mum needs some help.
And that's great. It's working as it is. But we're missing out on a really good option for pulling rich data into the system and standardising that level of care for everybody. I guess we can learn from pilot programmes. We've learned they might be helpful. What else have we learned, John, from them? So just on that point, I think digital innovation in healthcare does suffer from pilotitis.
This is what we're finding. And one of the other things we're learning is there's a lot of ad hoc knowledge in the system, but not necessarily the joint up, scaled up version of it. And that's a real issue. And the other issue is with prevention in general, as I'm sure you're aware, is
Trying to evidence or attribute cause to something you prevented is really tricky using those sort of traditional methods. So if we have prevented service utilization, Mary falling and needing a paramedic plus four other services, it's very difficult to model that because we don't know for sure whether she would have headed in that direction. So
That's an issue with knowledge accumulation in this space. And then that sort of world between the sort of business agenda versus the evidence generation agenda is another thing I've learned that I was quite new to, which is there's not, despite the fact that there is CE marks and a requirement for evidence from the NHS, there's a digital evidence framework requirements.
Actually, some of these things are being deployed without hitting those different criteria for various reasons. And that's really interesting, which is this is more about winning hearts and minds as it is producing your evidence to say, this is the effectiveness, this is the cost effectiveness, this is the safety, et cetera. So that's a level of maturity where we're at at the minute, which is interesting. It seems to be a common feature of AI, and there's lots of innovation, in fact, which is that
There are small-scale studies, small-scale pilot programs, which might hint at promise, but they're not formal evaluations. And I think what you're also saying is that undertaking a formal evaluation, like a randomized controlled trial or some other more complex,
sophisticated methodology that we might traditionally evaluate an intervention by, you know, that's a big undertaking. I mean, it still means there's no excuse, I think, when it comes to manufacturers and they're trying to market and promote and sell their devices, that if they're not being evaluated properly and getting onto the market or being used, I think that's a problem. But once they get there, of course, the evaluations have to be meaningful and
You're suggesting that there might be other ways of gathering data. I mean, I think that's what you're saying, isn't it? So gathering data in an ongoing basis. Describe that. What do you mean by that? And how does that work? So not sort of geek out, but I have a particular interest in evaluation.
And there is a real need, exactly like you said, to expect more from not just the companies who are developing and deploying this, but also the people who are procuring them to undertake more naturalistic, what I mean by that is in the wild data collection, sort of formative feedback and changes. And you make a really good point around randomized control trials. They're really important.
But they can be quite static. And there are new innovative ways to think about experimental type work with an innovation that's not a direct exposure. What I mean by that is if you take a statin of a certain amount, randomized control trials are perfect. But if you're using an app which has 20 different features and you use it, your exposure level is different at different times of the day. We need...
to do different types of data. Debbie, what's your take on this? Well, I think it's great. And I think the question that probably to get this rolled out, a lot of policymakers and politicians would ask is the cost savings from it. Meaning how much does it cost to bring this into various people's homes and how much does it save to social care and healthcare systems?
And I think there the real challenge, and John articulated this, and this is the problem of public health, is if you prevent someone from having a fall, how do you actually present that as a saving? Because you have to kind of have the counterfactual of what would have happened in the absence of that detection.
and the modeling there for that. And so I think there, it sounds incredibly compelling, but I can see the first question being, okay, about the modeling over actually how much is it saving the system? How many falls have you prevented? How many of this? And it sounds like there's a real case to be made that it's cost saving to the system.
Well, there's a long way to go. It sounds very exciting. And I'm on record as saying that we want to embrace these new technologies, AI, machine learning. We want to be supporting the development of them and
All the talk 25 years ago was about innovation. We need to innovate, do this, that and the other. And often it was small pilots and small scale studies that weren't properly evaluated. So we need people looking at it in a sort of rigorous, rounded way, as you both are. And perhaps Debbie's next book will be all about AI. It will be written by AI. Don't tell anybody that. OK, listen, thank you for joining us today, all of you.
Thanks very much. Thank you. Cheers. Bye-bye. That's it for this episode of Medicine and Science. We'll be back in a fortnight, so subscribe wherever you get your podcasts. I'm Cameron Abassi. Thanks for listening.