The Promises and Perils of Future #HealthTechnology - #GlobalHealthMatters podcast

The Promises and Perils of Future #HealthTechnology - #GlobalHealthMatters podcast

Show Video

Hello, I'm John-Arne Røttingen, Norway's Global Health Ambassador. I really enjoy listening to Global Health Matters, and I do it when I'm out for a run or out for a walk. It really gets me inspired.

I learn something every time I listen in. So thanks a lot for this excellent podcast, Garry. Hello and welcome to the Global Health Matters podcast. I'm your host, Garry Aslanyan. As you just heard, every month we are joined by listeners from all over the world. John-Arne Røttingen, thanks

for your message! It's always encouraging to receive feedback from our listeners. We kick off a new year with a conversation about the future and the new advances that are transforming global health practice. Like many of you, I've enjoyed playing around with ChatGPT, a large language model chatbot developed by OpenAI. It's been fascinating seeing the breadth of its functionality and humanlike responses. Investigators have already been testing the applicability of artificial intelligence to health care. A recent study in PLoS Digital Health has shown these kinds of AI algorithms to have huge potential in the early diagnostics of dementia.

In this episode, we will be discussing advances like these, among several others. I'm joined by two forerunners. My first guest is Professor Tim Mackey from the University of California in San Diego.

Along with his academic affiliations, he's also CEO and co-founder of a healthcare big data startup. My second guest is Yara Aboelwaffa, an independent digital health consultant and the co-founder of Health 2.O Egypt. Tim and Yara, welcome to the show. I'd like to ask you both, what are some of the exciting future advances that you have seen coming out of your respective regions, and how will these change global or public health in the future? So maybe I can take that question first. Thanks, Garry, for the

invitation to this great podcast. There's a lot of innovation coming forward in the future and also being developed right now, and a lot of it focuses on a suite of digital health tools that can be used to enhance public health and even generate some precision public health with data. So some of the things I'm really focused on from a research and also commercialization standpoint are things like big data, of course, machine learning, different types of additive technologies like 3D printing, also things like innovative and disruptive technologies like blockchain. So these are all technologies that tend to be used in a very commercial setting, tend to be adopted by industry first and public health later. But because of COVID-19 and the pandemic itself, there's been enhanced attention to public health use cases within these technology areas. And that's really exciting.

It's not something that we've necessarily seen all the time when a pandemic happens and technology, innovation and adoption for specific public health challenges gets rapidly used by different stakeholders around the world. So those are some of the interesting types of technologies I'm really fascinated about. The challenge, of course, now is creating technology that's really purposely driven for communities, for public health purposes, and that are not just for commercial purposes, but serve a public good.

That's great. And what about you, Yara? So also, first, thank you Garry for inviting me to this podcast. I've been listening to many of them and I'm happy to be contributing. Similar to what Tim is saying in the region and the Middle East we've been seeing massive and exciting developments in the past ten years, and with the COVID, it has accelerated these developments into public health use. The thing I'm most excited about is the data. We've moved really from the hype of different information, different tools, what is AI going to do, into understanding that data is the bedrock for providing evidence, informed interventions within the health system, and many different health systems within the region have been implementing national level systems to collect, aggregate and report on data.

So you have the UAE for example has mandated the use of EHR (electronic health record) in all of the hospitals. Iraq is implementing health information systems, vaccination monitoring, and these systems together are realizing more informed decisions for public health and also the focus on genomic data. There is a lot of focus now, specifically in the Gulf countries, for collecting genomic data, which is really new to the region.

There was an update on the genomic data publicly available and there was less than 4.08% of Arab or Middle Eastern origins. So the focus on genomic data will provide a massive progress in understanding of the disease and public health within the UAE. Great. Thanks, both of you. So interesting you've been mentioning some of the countries in the Middle East.

You may have heard that WHO, working in partnership with the Ministry of Health of Qatar, recently launched Florence version 2.0. Florence is the world's most extensive and freely accessible artificial intelligence health worker. So for this episode, I had a conversation with Florence.

How do we know vaccines are safe, Florence? There are a lot of opinions on this question, but the WHO and Florence have the information to show that vaccines are very safe. Would you like to know more about this? No. I want to know, Florence, how much exercise I should get a day. Hi, I'm Florence and it's so nice to meet you. Human health is my specialty and I want everyone to live a healthier life.

What interests you most? How much physical activity I should get, Florence? How much physical activity is needed for good health depends on your age and population group. I have a lot of information I can share about this. Select the age or population group you would like to learn more about.

Sorry, please type or say one of the age or population groups on the screen. Florence, will you replace all health workers in the future? I'm so glad you chose this topic. Mental health is a key part of our overall health. When we feel mentally well, we can work productively, enjoy our free time and contribute actively to our communities. I want to thank you, Florence, for helping me today. Have a great day.

Florence was engaging, but I must admit I didn't get responses sometimes that I really had questions to and it felt that she was telling me things that I needed to hear. Yara, have you talked to Florence before? What do you think about this AI health worker? I've tried to talk to Florence about weight loss, about smoking, a couple of things that she has out of the medical dictionary, and I absolutely love Florence. I think it's definitely on track to be a game changer in public health. The future versions of Florence, or generally AI powered chatbots, have many use cases they can have, like debunk medicine myths, they can respond to simple medical questions. There is a lot of possibilities the future of Florence. Mostly that it can become the first line of primary care that would initially relieve some of the pressure of the medical professionals.

I guess for Florence to realize its full potential, there are a few things that need to align together. One is that there needs to be more technology progress, of course, and this is already happening. And also evolving care model. So the care model needs to be hybrid so that it adopts Florence as a first line but then moves to a more personal type of traditionary care and user acceptance, which is happening. In short, I guess the stars are aligned for Florence. Okay. What about you, Tim,

will Florence be the future of public health? Well, I think that those type of use cases with chatbots fill a very significant role in the overall goal of a lot of public health organizations. One is universal health coverage. So can these bots replace frontline workers to a certain extent, at least to triage patients to identify where they can be routed to different services? And then, of course, are they culturally appropriate for the population of interest? These are things that Florence is piloting in a regional setting and can generate data that will be very important for very precision-based chatbots that serve specific public health roles. As Yara mentioned, great comment, it's about acceptability as well. So we'll see how acceptable people are to using a chatbot versus talking to a real person. But I think during COVID-19, we've seen that telehealth and telemedicine has gained widespread adoption.

So I think these technologies will eventually get adopted more widely. But again, they have to be culturally appropriate and they have to work. So those are some of the things we have to kind of tease out. It's hard to assess the viability of a chatbot in kind of early stages. Basically, the better technology gets is when it's fed more data. So as a chatbot like this gets more responses, get more voice recognition components to it, gets to understand what the lexicon should look like for culturally sensitive and appropriate messaging, gets to understand the logic steps to a particular question, technology will improve.

So I think the thing that's important for public health people to understand is that we can't just depend on technology to solve all our problems. We have to give it time to develop and we have to invest more in it. And that's what I'm concerned about, is a lot of the investment in public health is around prototypes, small use cases, and there's not enough investment that comes in after to really scale it to something that someone can use. But I think if you think

about the primary use cases here, improving access to health care, potentially task shifting for certain health care workers, addressing burnout to a certain extent, I think these technologies have a lot of validity to them, they just have to be continually invested in. I guess I have one more comment building on what Tim is saying, it's actually culturally appropriate. When I was talking to Florence, it uses modern standard Arabic and I think for the region where I operate, the Middle East speaks more than almost 27 dialects, which is quite different from the modern standard Arabic.

So talking chatbots in the future will need to take into account this and this technology is getting more and more available within the private sector. The thinking of those specific needs of different population groups and subgroups would be really important for the overall acceptability. Interesting. I mean, clearly, I just

had a conversation with her in English, but that probably goes for many other regionalization of languages and other cultural issues that we may not even be aware of. Next, maybe we can look in more detail, the approach to strengths and weaknesses of these new technologies. Maybe I can start with you, Tim. Both as a researcher and innovator, you've been exploring the role of big data to tackle issues ranging from counterfeit medicines, corruption in the system, or in the opioid crisis in the US. What are some of the merits and shortcomings of having so much information readily available out there? What do you think are the pitfalls here? I'll just provide one case study, and I think it's something that people talk about a lot, and that's the infodemic, the COVID-19 infodemic, which we do a lot of work with, with the WHO and with other partners on. And great that we're in an information age where there's so much available information out there. But an infodemic is

characterized by the abundance of information, a lot of it sometimes not helpful to people trying to engage in healthy behaviours. And so we're in an era now, if you look at the history of pandemics. You go back to SARS, where there was no Facebook, you go back to H1N1, where there were very few mobile apps and there were some social media platforms, until now where above 50% of people are on the Internet and a lot of them are on social media. So we have an abundance of information, but a lot of it's misinformation and a lot of it becomes viral very quickly.

So as public health professionals, we have to catch up to those tools. We have to develop tools that are specific to communities and the types of information they receive to address information voids when information is not presented by public health people and someone fills that area with information that's incorrect. And the most important thing is this technology, and whether this misinformation available in these communities, really has real world impact. So this is an example of lots of technology, lots of good use of the technology for things like precision ads, making people buy stuff. But we don't have the same technology to really combat misinformation the way we should.

So that's an example of where a lot of research has been going on over the last three years to identify different types of misinformation, different types of infodemic spread, different types of tools, whether it be machine learning, natural language processing, whether it be things that mainly focus on network analysis, etc., understanding the reality of this content, and so a lot of research has gone into this, but at the same time there's a commercial component where we have to figure out how platforms can implement these technologies and use them and really focus on public health before they think of commercial interests. We're still trying to figure out the balance between understanding what free speech is on the Internet and what is the best outcome for people from a public health standpoint when you're going through a pandemic and an infodemic at the same time. It's good that you explained the term infodemiology.

We will provide some additional resources for our listeners so they can learn more about it. Yara, you are active in the digital health space in Egypt. There's been an epidemic of mHealth, e-Health solutions. We've discussed this in our Digital Health episode in season one, actually.

However, this enthusiasm has been mismatched with the needs and realities of health systems, and we haven't really seen an extensive population level impact of these solutions. What has been your experience with evolving areas such as this? I think that question has two parts to it. One around the innovation landscape in Egypt, but also the Middle East. So the innovation, the digital health innovation landscape has been seeing massive progress. It has been attracting a lot of investment. So in the past six

years, it has attracted more than 22 times in terms of investment, and it has been the largest growing sector in the region. So in that sense, there is a healthy ecosystem that is showing progress, creating an impact within the population. We've done research, other organizations have done research, and it's basically four types of innovations; service delivery, clinical decision support, practice management software and mental health. So the fastest growing innovations in the region are Vezeeta, for example, which is an Egyptian medical service booking platform; Altibbi, which is a Jordanian telehealth platform that has done over five million consultations in the past few years; Shezlong, a mental health platform. Most of

the innovations tend to be on the service side while improving access and cutting costs. So that's because it's mostly driven by convenience and efficiency. These innovations, I'd say, are definitely targeting real demands in the health care market. That is rather different from the public health needs.

So let me give you an example. Most of the digital health they are focused on market needs that is highly profitable and scalable. And that's because innovations that do not have a proven financial ROIs (return on investment) are still extremely limited because they don't get funded. So things that look at health education, prevention, promotion, things like what Tim was talking about, epidemiology, topics like sexual reproductive health, women's health, these are not gaining traction because their financial ROI is still not enough for private investments to focus on that. So that's why you have innovations only on a certain site that is actually addressing a need but not necessarily the public health needs of the region.

The potential is clearly there, but what can be done to overcome these challenges? First is to have a national digital health strategy so that it aligns the ecosystem partners towards a specific set of goals. The other thing is that having and releasing the public health data so innovators can actually work with data that is relevant, that is focussed, that is community-based and can actually create community specific innovations. And the third thing is incentivizing health care organizations to work on prevention, health promotion and public health institutions, because they are driving a lot of the innovation in the sector. And how we look at it is how this innovation is going to impact their bottom line.

But if there are other market incentives, then this would push different healthcare organizations to actually move more towards public health needs. Next I'd like us to explore the interrelation between communities and technology. Based on your experience, what role do communities play as custodians of these new advances? We do a lot of digital health tool development, kind of in the research context, and what we start with is a human-centred design approach, which means that you put the patient and human at the centre of how you design these digital health apps. That's not always the case, especially in commercial use cases. A lot of times people are

just engineering solutions because they think there's a problem but they're not talking to the end user. And in our case, when public health is concerned, you really have to have a community based participatory approach to developing these type of digital applications because they're not going to have acceptability if they don't. We have one project with the FDA where we're developing a tool to encourage young adult minority populations to enrol in clinical trials. And before we actually developed the tool, we did a big data analysis of where the clinical trials are actually located, so that's more big data, secondary analysis. Second, we did focus groups with minorities to find out what they were talking about, what their experiences are, how they view clinical trials, what they understand or do not understand about clinical trials. And then we did social listening, which basically we looked at how people are talking about their clinical trial experiences on different social media platforms.

And then from there, where we're at now, as we hosted these co-design sessions with young adult minority participants and had them help us design a mobile app and an Internet site that directly relates to some of the findings we had in the focus groups and the other components of the research. And then we're going to go back and evaluate whether that digital app actually is what they want. So that's a very long process of getting consensus with your target audience to figure out if we're building something that meets their needs, meets their different values, and we also have to kind of customize it to different racial and ethnic minority groups because people experience clinical trials in different ways.

When you're really focused on community based population health change, you really have to involve the community at the beginning of the design component. That's very interesting. And Yara, what's your experience with this? A lot of the methodologies and frameworks to involve target populations have been established, yet we don't see this enough, specifically with young people. Just as a background, right now, for developing digital solutions, we have the first generation of digital natives and mind you, they use the Internet completely in a different way. They have their own channels, they have different sort of influences that they follow, so involving them is an invaluable resource.

And within the Arab region we have almost more than 60% of the population is under 25. So it's a massive group of people of adolescence and youth. So I worked with UNFPA on a very interesting project, which is to improve health education around sexual and reproductive health in Upper Egypt. And this is one of the most conservative and underprivileged areas where UNFPA has youth friendly clinics, yet the challenge was to increase the footfall of the youth friendly clinics or find other channels to help or to work with adolescents and young people on sexual and reproductive health. So we developed an eight month programme to engage people within the community. So it was really interesting and the results were stunning.

First, the interest and the commitment of the young people was impressive, and the solutions that were proposed and piloted were culturally sensitive and was mostly accepted because it shows the channels within the community or within, what their peers, mostly youth, and was gaining traction with their peers quite easily. So this project and many others have already made the case for involving youth, yet we don't see this quite often within the larger digital health public health practices. And of course, because it has its difficulties, more often than not we kind of see this as some sort of a tick box exercise, but I guess the more there is an understanding of the real value that engaging youth represents, the more that they would be engaged within the whole process from planning to interpretation. What is the appetite of country governments to scale and embed successful pilots, like this one you mentioned, as a system-wide solution? I think governments in the region across the world are interested to use more digital innovations to the poor health systems and even more and more after the COVID pandemic. Broadly in the Middle East there are different approaches to adopting and spurring the innovation ecosystem.

So you have for the Gulf Region, for example, the innovation is driven by top down initiatives. So there are government fundings, infrastructure, data foundation, different regulations and different policies to support integrated data sharing, and all of these initiatives are very important to increase innovation within the ecosystem, but also make it easy to adopt innovations within the health system. But then you have other countries, which is like Egypt, Lebanon and Jordan. These are fast growing populations and there is patchy access to healthcare system depending on the income, location, so innovation is driven predominantly bottom up with innovators and entrepreneurs trying to make a dent in the system. Tim, your based in California, which is a hotbed for future advances. Can you tell us more about the opportunities and challenges in your innovation ecosystem? There are different resources available to digital health entrepreneurs, and the ecosystem has been really strong over the last two years.

There's been a lot of large size deals and funding opportunities from VCs (venture capitalist) and other private equity groups that are looking at digital health, and we're saying COVID's exploding. Every company has a digital health COVID use case. There was a lot of funding that came through. Most of the funding I'm

looking at Roc Health Data is in the mental health, oncology, cardiovascular, diabetes and primary care areas. These continue to be areas that are staples for basically the US health care system, this is where we have the most use rate, it's the most profitable, etc. A lot of investment is still going into drug discovery and R&D.

And so standing up for public health use case is actually very hard because it's not seen as something that an investor would ever want to invest in because it's not focused in the clinical setting. There's very little ways to make money or to monetize it. There may not be any type of reimbursement process for it unless you can put it under the umbrella of preventive care, which a lot of public health use cases are actually not always preventive care as well. So from my experience as an entrepreneur as well, we had developed technology through research to address illicit online drug sales.

And this is a huge issue in the United States. It's a more regional issue than it is a global issue because we are going through an opioid crisis where a lot of people are chemically addicted to opioids. And then at the same time we're having a crisis of counterfeit fentanyl that's entering our supply chain as well and is leading to a lot of overdoses. Any investor I talked to was not really interested in that. No one's going to pay for monitoring of the Internet to make sure that people are safe from drug sales. That's the job of the government.

That's the job of someone else. It's not a profitable area. Substance use disorder is generally not a profitable area for digital health applications. So what happens in this ecosystem is the US government provides funding for these type of public health use cases through what's called an SBIR programme. It's a small business innovation research grant that essentially takes mostly academic research and tries to translate it into commercial solutions.

And then, with the support of the government, you can hopefully gain some commercial traction with a government partner or with a commercial partner, etc. And so that's one good gateway for public health centric technology that has a public good to it, but may not be commerciable right away. Even with that support, it's relatively still hard to commercialize this technology.

So one of the things that unfortunately we have to be prepared for is to talk to investors, to VCs, to different people in the space of not just social impact investing, but actual investing in companies. And you really have to put that value proposition of why public health comes first, and you have to find a revenue stream to support it, and that is still a very difficult proposition even in the United States where we have a lot of resources. That's interesting, Tim. It seems there is really an ecosystem gap to fill when it comes to innovative solutions for public health. But not to discourage our listeners, what opportunities are there for those who want to engage in innovation within their own public health settings? One of the things that gets me excited is that a lot of this technology is available. It's open source.

During COVID 19, I think everybody I knew had a COVID-19 dashboard. Some data visualization that they had built, customized to understand different patterns of the COVID-19 pandemic. We had Johns Hopkins University running a dashboard that was used all around the world and was not from a government agency, was not from a private sector entity, but was from some group that was aggregating data in a way that people needed it.

There are really good opportunities for people in low-to-middle income countries and even high income countries, of course, to leverage free tools to them to creatively solve new challenges. People are really super innovative and they have the knowledge base to understand how their localized problems can fit into a technology solution, and they just need the tools to do it. Just a lot of space in that innovation cycle to be disruptive. I do actually think that the liberalization of these tools and the availability of them and then the training and education that comes with that, can really be a game changer because these tools weren't necessarily available to most people before. I'm encouraged.

There's a lot of good people in this space, and I do believe in the power of creativity and human intuition, and I think these tools will help us get up and get there. Yara? So I think one of the things that we tend to use a lot of big words when it comes to public health and digital health and innovation and frameworks and theories. But at the end of the day, it's not as hard as it sounds. And a lot of people within middle- and low-income countries are actually realizing this and are developing innovative solutions. And as

Tim was actually saying, there is a lot of available open source solutions right now and also a lot of education materials that are available. Courses from the best universities in the world that are available online for free. And all of this together is not only will be used, but we see it being used within different low- and middle-income countries. And South Africa is actually a great example in terms of low resource settings, constrained but with exponential innovations happening within their health system. So final question.

How do you envision the future of public health? Looking through my imaginary lens to the future, I'd say there are five things. Public health in the future would be participatory, preventative, personalized, democratized and as you were saying Garry earlier, destigmatized. This is a long list of adjectives, but as I was saying earlier, they are easier actually than they sound. It basically means that health systems will empower people to take charge of their own health, shift to more preventative approaches to keep the population in this magic circle of wellness and provide tailored health services when required that addresses specific needs of the individual regardless of their age, sex, gender, income, or the type of indication that they might be struggling with. That's my future version of public health. And Tim, can you match that? I cannot match that because that was an excellent vision of what public health can look like in the future. I'll just add my component,

which is, I think the future of public health should really look at public health as separate from individual and precision medicine, which is really focused on optimizing health care delivery for an individual. But what we're really talking about is population health. We're talking about the fact that diseases are globalized, that pandemics are going to emerge over and over again because of the zoonotic component of these diseases and our ecological impact on the earth. And I think public health needs to move into more society based public health prevention practices; addressing not just the clinical component or the health behaviour component, but all of the political determinants, all the corporate determinants, everything that goes into what actually creates a negative public health environment. And I think as we see globalization of diseases more rapidly spread and more understanding up against the social determinants of health, we're going to see that public health has to go much farther beyond our traditional public health practices.

It has to go into these other disciplinary areas. It has to go into health diplomacy, for example, and it has to go into digital health and innovation as one of its core pillars, as a way to improve health for everyone, not just a population, not just a community, etc., but recognizing that all of our health is interconnected in some way. Having that connective tissue, I think will be very important. I think technology is one connective tissue for that, if used appropriately, we just have to make sure that technology is purposely built for public health.

Thank you both, Yara and Tim for joining me today. Good luck with your future endeavours. From the promise of AI health workers and the perils of infodemic, the advances and innovations discussed today will without a doubt shape the future of global health. It's clear that the pandemic was a significant catalyst for these new developments, yet much work remains to be done to make these an attractive investment for funders and governments in a way that can strengthen the health systems worldwide. Advances that can improve prevention, integration, participation, makes me optimistic about the future of global health.

Thank you all our listeners for joining us for this episode. To learn more about the topics discussed today, visit the episode web page where you can find additional readings, show notes and translations. Remember to be in touch with us via social media, email or by sharing a voice message with your reflections about today's episode. Tune in again next month for another engaging conversation about science and diplomacy.

Global Health Matters is produced by TDR, an infectious diseases research programme based at the World Health Organization. Garry Aslanyan, Lindi Van Niekerk and Maki Kitamura are the content producers, and Obadiah George is the technical producer. This podcast was also made possible with the support of Chris Coze, Elisabetta Dessi, Iza Suder-Dayao, Noreen O'Gallagher and Chembe Collaborative. The goal of Global Health Matters is to provide a forum for sharing perspectives on key issues affecting global health research.

Send us your comments and suggestions by email or voice message to TDRpod@who.int, and be sure to download and subscribe wherever you get your podcasts. Thank you for listening.

2023-02-04 05:35

Show Video

Other news