The promise of AI to help reduce early deaths – transcript

Ideas to Innovation - Season Two

Intro: Ideas to Innovation from Clarivate.

Neville Hobson: Life expectancy at birth for humans is currently an average of 73 years, according to the World Health Organization. It means that someone born in 2023 is expected to live until 2096, just a few years shy at the start of the 22nd century. 73 doesn’t reveal the full picture though. It’s the average combined life expectancy for both men and women. An average means some people will live longer, 83 on average in Japan. while others will die younger, 50 in Lesotho. According to a WHO report in May, advances in medical science, related technologies, and other medical and societal factors have helped increase life expectancy worldwide by six years between 2000 and 2019. Such numbers are beloved by medical researchers, actuaries and pharmaceutical companies who look for certainties when attempting to predict life outcomes. Yet uncertainties are the norm. when life expectancy is seen through the lens of early deaths caused by behavioural and other preventable factors. The focus is largely but not exclusively on Western countries. Despite the reduction in exposure to many health risks such as tobacco use, unsafe water and sanitation, progress is inadequate. Risk exposure remains high, especially for factors such as alcohol consumption and hypertension, where declines began only in recent years. There’s the prevalence of obesity, which is moving in the wrong direction with no immediate sign of reversion.

Welcome to Ideas to Innovation, a podcast from Clarivate with information and insight from conversations that explore how innovation spurs incredible outcomes by passionate people in many areas of science, business, academia, technology, sport, and more. I’m Neville Hobson.  I’ve painted what looks like a bleak picture of our longevity with causes that are self-inflicted. What near future awaits us if we keep on this track of self-harm, even destruction? And is this a fair assessment? To give us some answers and insights, I’m delighted to welcome Dr. Grace Lomax to this episode. Grace founded Patient Connect with her sister, Zoe Barker, in 2008, and the company was acquired by Clarivate in 2021. Hello, Grace, thank you for joining us.

Grace Lomax: Thanks, I’m pleased to be here.

Neville Hobson: I should actually say welcome back to Ideas to Innovation, as we spoke in June, 2022 in an episode of this podcast. So that was a year ago. Would you tell us what you’re doing with Patient Connect today? How have things evolved over the past year?

Grace Lomax: Yes, glad to. So we’ve been working on providing more sophisticated solutions so that we can enable better and best care for more narrow profiles of patients Rather than herding all patients together. We’re also working on providing richer and deeper data resources that help inform and provide insight around a patient journey, where the bottlenecks in care are. where patients are being lost from care. And that’s now from primary care patients, so going to see your GP or family practitioner, but also from hospital patients, where quite often the most ill patients are treated, and therefore insight and data can be invaluable to understand their unmet needs. We, like everyone, are alert to how AI is exploding across the world. And especially here at Clarivate, we have our world-class data resources and clinical trials and real-world data. And of course, we’re using those AI to help enrich our customers’ experience of our vast data resources and enable more intuitive experience for the clients. So it’s been a great year and a more exciting year ahead.

Neville Hobson: That’s good to hear that. And thanks for bringing us up to date. I think, AI, we’re gonna talk about that again in a short while, actually. That is a hot topic, as we know. Everyone’s talking about it, and we are going to as well in a related sense. So a year ago in our conversation, we talked about how 80% of premature deaths could be avoided if patients and healthcare professionals alike had the right information at the right time to make better decisions. A major point you were making in that conversation. was about healthcare literacy and education at the point of care. It seems to me that now it’s not so much only about communication and understanding, important though they are, as it is about a highly customized approach for both healthcare professionals and patients to improve the patient experience which might positively impact life expectancy. So I can see a connection, a continuation perhaps with our conversation last year, a chapter two, so to speak. I’m mindful that the trigger for our previous topic was on avoiding 80% of unnecessary premature deaths. But the WHO findings I spoke of in the introduction to this episode suggest that progress on reducing health risks is inadequate, risk exposure remains high, and the prevalence of obesity is worsening. So what needs to happen to see meaningful progress in further reducing that 80% of unnecessary premature deaths?

Grace Lomax: That’s such an interesting subject. So let’s use heart failure as an example. The diseases are complex and difficult to manage, even if they are common and many people have them. The complexity that’s required with the physicians to understand, to ensure each patient receives best care, is high and shouldn’t be underestimated. Heart failure in the US, two million people have heart failure. And just to explain, because we have a wide experience in this audience, what heart failure is, is when the heart is not working as well as it should. And this is typically caused by an event like a heart attack, where a little patch of the muscle that is the heart has died. and that means the heart can’t work as well as it should, or it can be a valvular problem. Either way, the heart is not working. And of course, for those who want to dust off their biology lessons, the heart is essential. It’s pumping oxygen to every bit of you, and every bit of you needs oxygen, and it’s taking carbon dioxide away as well. So it’s one of the most essential organs, and when you have heart failure, very significant problems. So two million people in the US have it. That’s a high number. And I think rather alarmingly, too many of them, many of them are misdiagnosed with something else, or undiagnosed altogether. And so they’re going along struggling typically, because it’s a very symptomatic disease. So there’s breathlessness, there’s swollen ankles. There’s palpitations, there’s weight gain. So they’re quite diffuse, if you like, vague symptoms. And these patients are undiagnosed or misdiagnosed too often. And unfortunately that’s at their peril. You know, the 10% of patients who are admitted to hospital with worsening heart failure will die in hospital that visit. So one in 10 will die on that visit. And then… really quite worryingly, another 30% will die within a year. So symptomatic heart failure patients are, they’re high risk patients and we’re not looking after them as well as we should. You know, we know if you provide optimal care, you can double the life expectancy of a heart failure patient, double. And that’s just by ensuring they receive the right and best care. And so something we’ve been working on extensively this year, but also in recent years, is providing more personalised healthcare at the point of care. So instead of when a heart failure patient say presents to their physician, instead of treating them all the same, we don’t do that. We ensure that a patient who’s having a first diagnosis of heart failure is receiving the right care for that setting. So that might be an echo test, so a scan of the heart, or it might be a blood test for a natuoretic peptide, as just two examples. Or we could be looking at, again, heart failure patients who have a diagnosis, but their symptoms are getting worse. They’re quite a different profile to the first. group of patients. And then of course, as I mentioned, there’s many patients, unfortunately, who are misdiagnosed with bronchitis instead of heart failure or misdiagnosed with palpitations and arrhythmias instead of heart failure. So again, the third cluster might be those patients who don’t have a diagnosis of heart failure, have a wrong diagnosis. And again, let’s make sure those patients are getting the right care at the point of care. And so this ability to split out different profiles of patients who have distinct different needs, so important. And the data speaks loud and clear. If you provide the right care to the right groups of patients, you double the life expectancy of these patients. What more to achieve?

Neville Hobson: That’s very interesting doubling. I think you’ve painted a really interesting picture there, Grace, because it seems to me that the undiagnosed and the risk factors, the odds of surviving, they’re all ringing alarm bells, I would imagine, for everyone, particularly thinking the state of healthcare, broadly speaking, in many countries, is overwhelmingly overwhelmed, is it not, in many? The numbers of people coming, requiring treatments. So people are getting missed and all that. And earlier you talked about data and you mentioned artificial intelligence in that, kind of introducing that. And I’m thinking the data itself is obviously quite sophisticated in terms of the quantity and quality of it. But the AI element, how is that helping? I mean, it may sound an obvious question given the conversations we’re talking about. I mean, this is not chat GPT necessarily, we’re talking about this deeper than that. Is there what? What is AI doing to help you as the physician, let’s say, address this?

Grace Lomax: Yeah, it’s a really good point. And so we are using AI in the triggers. So you’ll remember from our discussion last year how we help enable a million physicians to provide better care for a billion patients. And how we do that is we’re integrated into the clinical software at the point of care. And where we bring in AI into enhancing those capabilities, we use for anyone on the… in the audience who is familiar with AI, we are using artificial narrow intelligence. And what that means for all of the rest of us is it’s basically sophisticated mathematical algorithms to identify in real time each profile of patient that we want to support. So it’s driven by maths, but the maths is driven by clinical profiles. So it’s super, super interesting. And so the AI or the artificial narrow intelligence, it’s looking back over time at what has happened to the patients in the past. So have they had a heart attack? Have they been given a beta blocker? Has their weight gone up? Has their blood pressure been measured and what was it? And if it was greater than X, then Y, you know, it’s quite sophisticated algorithms. But what they’re doing is enabling us to identify the most at risk patients or the patients with the most unmet need. Super exciting, super powerful.

Neville Hobson: Okay, does it mean as well that it does it faster and more accurately than other ways would be the case? Is that part of the kind of desire for this?

Grace Lomax: That’s a really good point. Yes, it is. Because AI is many things. One of its greatest value adds, one of its many greatest value adds in my view is its ability to remove false negatives, false positives. It provides, basically it provides greater accuracy. And that’s so important. It means more patients receive the right care, fewer patients receive the wrong care. More patients get a more timely diagnosis. You know, it gets rid of the poor care. It helps get rid of poor care. So important. Firstly, better service, more efficient service, saves lives. What more important than that?

Neville Hobson: Right, right, absolutely. Let’s switch our focus slightly for a moment. So this has all really been about heart failure itself and we’ve talked about AI. Let’s consider another factor for the moment. That’s a big piece in the jigsaw puzzle of life expectancy. And that is, as you pointed out just now, the unmet clinical need of patients with long-term diseases more broadly. What can you tell us about that, Grace?

Grace Lomax: Yeah, so the scale of unmet need is vast and the sustainability is a popular phrase these days, isn’t it? And the UN has some ambitious views there. We know there’s over 40 million premature deaths each year from chronic diseases and 80% of those could be prevented. And I think we should call out on that number that most of those, over 80% of those deaths are impoverished areas, so impoverished areas of the states, but also poor countries, you know, developing countries. So most of the unnecessary premature deaths are happening to people where health isn’t on their side, where they have less access to it, which is a travesty, it almost feels like a crime. And the fact is 80% of premature deaths can be prevented if we provide optimal care. You know, that is a massive number, a great opportunity. And where I think we as a health system still have a lot of work to do is instead of talking the talk of patient centricity, is walking the walk. We as health systems in the US, in Europe, in the developing world, and we still make patients fit into health systems. Whereas if we were doing providing patient centricity in its purest form, we would put the patient at the center of care. And I say that, that really needs to be heard. Currently, we know there are 250 billion tablets every year for hypertension globally, 250 billion that are being prescribed for patients. And we know 80% of those patients do not manage to lower their blood pressure. So that means 200 billion tablets for anti-hypertensives are going to waste or are not being optimally managed. But you can’t get away from the fact that 80% of hypertensive patients are not getting their blood pressure down. And the whole reason we manage blood pressure is to stop them dying. Half of us will die from heart disease, and half of those people will die from hypertension. And when we’re not managing to treat 80% of them, we need to be changing our health systems so that we are more patient-centric. with understanding the unmet needs of the patients, we’re understanding and listening to their fears, their misunderstandings, their worries, what’s stopping them adhering to their medicine and support them so that they have a better understanding and receive more optimal care.

Neville Hobson: Yeah, you’ve illustrated something I’m sure, Grace, many of our listeners would be nodding their heads at this because this to me is similar to what we hear about climate change. It’s almost like we know all these things, why aren’t we addressing all of this . And that’s a huge topic to talk about. And probably a good way for us to address that in this conversation is to look ahead a bit. So our final point really is a good moment to arrive at it. We’ve got the picture for today, including this part we’ve just referenced. All this going on, there’s still too many people dying that they don’t need to die. The prescripted medicines, medications aren’t getting to them or they’re not being managed well. All that’s in the picture today. What about tomorrow? What can we realistically expect to see 10 years from now in 2033? Or is that too short a timeframe? I don’t know. But it seems a good manageable chunk of time to look ahead on. How do you see it in terms of what it’s likely to be? Indeed, what would you like to see in 10 years time?

Grace Lomax: I think the next 10 years of medicine, we will see more change than we’ve seen in the last 100. And these are extraordinary times. So we have a collision, if you like, of three massive changes in technology and opportunity. So firstly, of course, we have AI. So I’ll come back to that. Secondly, we have the sort of maturation of digitized health data. And what I mean by that is the fact that over the last 40 years, physicians and pharmacists and healthcare professionals around the world have been documenting what’s happening to patients. So now we have a vast, rich, deep resource of real world data. And then thirdly, we have the genomic, the mapping of the human genome. You know, of course, millions, billions of alleles within the genome, but we don’t really know what the vast majority, over 99% of the genome, we don’t know what it does. We’ve just mapped it. So we have AI and the leap forward in AI, because of course AI has been around for 50 years, but the massive leap forward, two of them in the last 20 years and one in the last year, is we have been teaching machines how to read lots of data. And then in the last year, we’ve taught machines how to read words. So instead of reading algorithms and complex statistics and maths, we have taught the machines to read words. And of course, words are what we all understand, not just the computer programmers, and words are how we have documented the whole of healthcare and all of clinical trials and all of academic work. And so now we have powerful machines that are able to read what we document. They’re able to, because they’re so powerful, they’ll be able to help us interpret the gene coding that is out there, but we don’t yet understand. And they will also be able to, through deep learning, help us interrogate that digitized health systems software.

Neville Hobson: Yeah, that’s quite, that’s intriguing. You had one point, I think you mentioned before we were recording that about DNA to help individuals use their own DNA. Is that the AI role or is that another extra piece of the British Wuzzle?

Grace Lomax: that’s a super point. It’s mRNA. So when COVID came out, of course, and there was a race to develop both treatments and vaccines, one of the huge leaps forwards in healthcare was the use of mRNA to identify the optimal vaccines to develop. And that is game changing in healthcare. It means that for literally down to an individual, one person. we can take their genetic coding, develop a treatment for their illness, maybe a cancer, and then in swift order, develop the treatment that cures them, not treats them, cures them. That is how game-changing it is. And it’s done very swiftly, it takes hours, not years. And that will change healthcare, it will certainly change… the care of cancer patients notably. But it’s far more broad than just cancer, but clearly where it could add some significant value is cancer patients. I think that’s a super exciting area of medicine.

Neville Hobson: Okay, well, you’ve certainly shone some light rays of hope, I think, in terms of looking ahead these 10 years. All of this, of course, won’t be in place in 10 years, but the pathways are opening up ahead of us that we could see. And I agree with you, this is very exciting. All of this, not just the AI, although I think that in itself is particularly what you mentioned the last year in particular, so much has happened. But all of this working together, all these elements of the jigsaw puzzle. it is presenting a pretty interesting picture. Well, this has been a great conversation Grace continuation of what we started a year ago, and I’d like to thank you very much indeed for sharing your knowledge and your insights. Thank you.

Grace Lomax: Thank you, it’s been a pleasure.

Neville Hobson: You’ve been listening to a conversation about life expectancy, avoidable premature deaths and realistic outcomes with our guest Dr Grace Lomax, co-founder of Patient Connect, part of Clarivate. For information about Patient Connect, visit We’ll be releasing our next episode in a few weeks’ time. Visit slash podcast for information about ideas to innovation. And for this episode… Please consider sharing it with your friends and colleagues, rating us on your favorite podcast app, or leaving a review. Until next time, thanks for listening. So there we are. The only flub we had was me.

Outro: Ideas to innovation from Clarivate.