€200m and 20 years: What EU health data projects built, and what EHDS needs next (Henrique Martins)

[00:00:00] If a country is not investing in digital health, it's the citizens fault. It's not that the government's fault, the citizens chose a government that does not want to invest in digital health. And why do I say this? Because I come from a country that moved digital health very fast, very forward. In the worst moment, we had a crisis, a financial crisis, we had the intervention of FME, we had the intervention of the central bank, we were without money, and we still did digital health. Why?

[00:00:30] Because it makes a difference for health, and it saves money. So even the countries that are not so rich, they should invest in digital health. But it's citizens' decision. They should lobby, go to the streets and complain, complain that you want digital health. You know why? Because it saves money, it saves health, it might even save your life.

[00:00:50] Dear listeners, welcome to Faces of Digital Health, a podcast about digital health and how healthcare systems around the world adopt technology. With me, Tjaša Zajc. Europe has spent nearly 200 million euros over two decades on cross-border health data projects. These projects were essential for the establishment of the European health data space, which is now in progress. So I invited Henrik Martin to the

[00:01:25] associate professor at ISCTE in Lisbon and the coordinator of many of the EU projects for healthcare data exchange to join the show and talk a little bit more about where EHDS is at the moment. What's the legacy and the outcome of the past European projects? And what can we expect in terms of the upcoming milestones for EHDS?

[00:01:52] We discussed everything from concrete projects that are ending this year's expectations on what we will be able to say about EHDS at the end of next year, the role of AI in the implementation of interoperability and AI and data use in Europe, the patient role in this whole story and much, much more.

[00:02:50] Enjoy the show. If you haven't yet, check out our newsletter, which you can find at FODH.substack.com. That's FODH.substack.com. Now let's dive in today's discussion.

[00:03:17] Henrik, hi, and thank you so much for joining me on Faces of Digital Health for a discussion about EHDS. Many people in the interoperability community know you. You are someone who has been involved in European projects related to interoperability and EHDS for what, 10, 20 years?

[00:03:39] Oh, well, maybe 15, 15. The first meeting I attended was February 11, February 12, actually. Yeah. For EPSOS. Yeah. So yeah, 15 years.

[00:03:52] And that's exactly what I want to talk to you about. So EHDS and interoperability projects in Europe, nearly 200 million euros in EU funding went into cross-border interoperability projects in the last 20 years.

[00:04:13] So if we just start very easily, how would you describe this period? What seem to be the main takeaways and how are the projects that have happened, how are they complementing each other? So first of all, yes, we have researched on this and we have published an article, you know, putting together data on the different projects that's available from our group online.

[00:04:40] And so people can read more on that. But the reality is that 200 million euros seems a lot of money. But if you actually spread them into time, it's no more than 20 million per year, which for 27 member states at the time, there was one more UK, 28 is less than a million. It's less than a million euros a year to bring together the countries.

[00:05:05] And as we did that, and we did today, we have more than 16 countries already exchanging data across the borders and supporting services like dispensation of medication far away from the origin of prescription. As we did that, we also brought together different regions of some countries.

[00:05:29] So when people ask, what is the value added of cross-border efforts, you need to remember that many countries in the EU have inter-regional interoperability problems. And if we align regions and even institutions to the same European standards, we are actually making a lot of efforts synergic. So 27 benefit from the same efforts.

[00:05:57] And this is exactly what I think is the most interesting thing about the most recent effort, which is European exchange format. Because it is not just for the government or national level agencies. It is for each and every hospital and each and every IT vendor to align on things so simple like a discharge report. That, you know, today, thousands of discharge reports will be produced around Europe.

[00:06:26] And they are different. They're not interoperable. You cannot pick a bunch of data from one discharge report today, as it is today, from Budapest and inject that data into a hospital in Madrid. But it will be possible. Yeah, yeah.

[00:07:12] HL7 to FHIR, that's like a whole transformation that still needs to happen. So how does a European exchange format for discharge come into play, just to kind of put it in context? Yeah. So indeed, FHIR and many other standards are, as standards are, a way to build joint understanding and common agreement.

[00:07:37] So FHIR standards does not define exactly how for each use case, for each domain, for each data priority category, as we are now calling it in the European health data space regulation, how do you actually do it?

[00:07:56] So you need more details that are to be agreed upon and then documented in what we call implementation or implementing guides, IGs, and even some more detailed guidance like data set tables. So I think the most important thing people have to understand is that this is not something that you mandate, you write, and then everyone aligns and that's it.

[00:08:24] Digital health is very complex and health data is very, very heterogeneous. It reflects different practices and it reflects different languages and also different levels of maturity regarding digital transformation. So some countries are very ready to use codes like ICD codes or SNOMED codes or loin codes. They have trained their staff, they have their IT systems ready. But other countries, they are mostly using free text.

[00:08:54] So you need a balance, but it's bringing these people together and that community building is something that is now quite solid. We have different forums where people meet regularly. Some people even complain there's too many meetings. But it is that creation of a community that then defines artifacts and agrees that will make it happen. But it will still take time.

[00:09:23] And in concrete terms, yes, you may need to change some underlying standards in your system. But for the most part, in many cases, you just need a sort of a transformation layer. So imagine that you're outputting a laboratory result in a certain format, for example, HL7V2, and you use a transformer. So a small piece of software that will allow it to be wrapped up into a FHIR message.

[00:09:52] So this can be done in some cases. In other cases, it cannot be done. And I think it's finding the smart way, the intelligent way. This is why we have a project I'm coordinating called Intelligent Implementation of the Exchange Format. It's to find the intelligent ways, including using AI, to facilitate interoperability.

[00:10:15] So maybe 20 years ago, we would be talking about tables and mappings and putting everyone to use the same codes, using drop-down menus and the software so people could code the procedure. Now we talk much more on using, for example, large language models or natural language processing tools to actually extract the key data that is in a text and make it available.

[00:10:41] And then maybe the doctor, for example, or nurse can validate this pre-filled document. So we need to be smart and more intelligent as we move ahead.

[00:10:55] If we just go back to the history of the European projects, I'm just going to share an overview of projects that you prepared for one of the articles that you did on interoperability.

[00:11:11] And just, you know, looking back for one more moment, is there a project in this group that you would say has or had the biggest impact or a project that kind of brought the greatest insights or impact on how things developed further?

[00:11:36] Well, I think most people would agree that EPSOS 1 and 2 together are the sort of the cornerstones of the services related projects. So when we talk about cross-border services like e-prescription and patient summary, those concepts, those use cases were developed in EPSOS 1 and they were matured in EPSOS 2 and piloted.

[00:12:02] So perhaps that is definitely in terms of services, the most important one. But if we talk about, for example, digital health in a more broad sense, you go to JSON, which was a joint action, and you will find deliverables that cover things like registries and telehealth, global positioning of Europe on digital health. So it's much broader, the understanding.

[00:12:31] So we didn't just work on two services like patient summary and e-prescription. We also opened the way to develop many other types of things. So for example, health EIDE was looking at electronic identification applied to health, which is now more and more important because of the EIDAS regulation, the NIST directive and authentication and double factor authentication and identity.

[00:13:00] When you look at things like, for example, UNICON, which is now being continued in a joint action, which is called EU-wide, it's going to start next year with almost 20 medicines agencies applying the UNICON, UNICON, which is a standard for medicines descriptions.

[00:13:27] So it's really going back, you know, it's a bit nostalgic. Some of them I coordinated, many I participated, definitely. But there's many things. For example, e-health action, we had the first deliverable national strategies for digital health, comparing the different countries' national strategy. Today, most countries in Europe have additional digital health strategy.

[00:13:53] But at the time, even large countries, very important countries, quite mature, did not have a document that you would call a digital health strategy. So, yeah, different initiatives. And let me also highlight Trillium, because in the Trillium bridge and then Trillium II, in today's world, almost impossible. Can you imagine a project between the EU and the US to exchange health data? So that was Trillium.

[00:14:21] Yes, that was Trillium bridge and Trillium II. Two, we tested data from Portugal and Luxembourg being made available in a software in the US. Mm-hmm. You can imagine that the policy, the politics, the policy environment was completely different from today. Today, we wouldn't be able to even have funding for Trillium III, unfortunately, because we still need to work with the US.

[00:14:48] I mean, they are our biggest technological partner in terms of interoperability and health, digital health. Everyone knows that. Absolutely. Absolutely.

[00:14:58] If I'm not mistaken, so was EPSOS was the project that was basically the cornerstone of My Health at EU, which is the infrastructure that's basically the basis for all the exchange that is going to also happen in the future for e-prescriptions, etc.? Yes, EPSOS was the first big pilot.

[00:15:25] And that's where we conceptualized the use cases. We developed the infrastructure. We produced the so-called OpenNCP, National Contact Point, which is a server that sits, let's say, at the door of each nation to allow exchange with other countries.

[00:15:42] And then it was refined under a project called EXPAND and then handed over those assets, those documents, those technical documents, and those software artifacts were then handed over to the European Commission for the creation of the services that are now officially called My Health at EU. Mm-hmm. Okay, wonderful. Okay, I'm going to share another timeline that I created.

[00:16:10] So I was just trying to make some sense of where EHDS is at the moment and where is it going. So there's quite a few projects that have been finalized or are in the process of being finalized this year. We saw them also on the previous timeline.

[00:16:29] So, for example, EXPAND EHR ended in May and aimed at creating the functional and technical specifications for the European Electronic Health Record Exchange format. If we stay there for just one second, so what were the kind of the key takeaways of this project and how is it going to translate into something else?

[00:16:53] And what I'm basically asking is that when we talk about European projects, they're never obligatory in any way for countries or governments to actually change something. It's all based on kind of recommendation, recommendations. So how, what is going to be the impact that you expect from this project? From ExShare, you mean? Yeah. Or from TEDx? No, so…

[00:17:23] From ExShare. ExShare is ending in November. I mean the XTHR, which ended in May. Well, I think there, there's a little bit more than just recommendations because it's a joint action. Okay. A joint action, for those that may not be well informed, is a specific type of EU project where the commission and member states jointly act.

[00:17:51] That's why it's called the joint action. And so the weight of the recommendations and their impact is much higher than a normal project led, for example, by a university or by even a national agency.

[00:18:06] So because in a joint action, most countries through their appointed national agencies or the ministries of health sometimes directly get together with the European Commission and work together on certain challenges. So in this case, this particular joint action is working with the commission to contribute to the so-called implementing acts.

[00:18:31] So the implementing acts are the legal pieces that define with more detail what is to happen under the European health status-based legislation, regulation. So they worked on logical models. They work on different sorts of recommendations. I would not be surprised that much of their recommendation material might find its way to the text, the legal text of the implementing act.

[00:19:00] So it's a very impactful joint action, I would say. Mm-hmm. Countries in Europe, as part of the EHTS, are currently setting up their digital health authorities and designated national contact points, and they need to do that by March 2027. Let's explain what that actually means. So what are the digital health authorities and national contact points? What's the point of them? What's their role? What do they need to do? Why are they so important?

[00:19:31] So the first thing is this concept that for each country, there's one entity that serves as the main door for primary use of health data. Okay? So when the data is flowing from one country to another in terms of national services. So imagine a big train station, central train station in Paris, central train station in Berlin. Okay? And they are connected.

[00:20:01] Now, this architecture is in itself in transformation because we also believe that with the exchange format, for example, there might be a time when the citizen just directly goes to a hospital in one part or another and uploads and downloads data. So he becomes another type of gate to transfer data. But even that requires some governance.

[00:20:27] So we decided as Europe that for each country, there is one digital health authority. So they have the responsibilities to, for example, imagine that a country has to decide if they are a SNOMAD member or not. This should be decided by the ministries or whatever. But then there should be some national authority that then says, okay, SNOMAD is to be used in this country in a certain way or another way. This is just an example.

[00:20:56] The same would be how do we link our digital health services with the identification services, often that are governed by the Ministry of Social, of, you know, internal administration or transformation or something. Right? Normally in the countries, the identification, electronic identification of citizens, electronic identification of professionals like doctors or nurses is run by a different ministry, a different agency.

[00:21:26] So this is the so-called trust services. So this agency tries to bring together from a national level, the different aspects that you need to organize in your country. And then serves as one out of 27 plus one because we have to add Norway that sits in these conversations and represents the country for these things. Differently, the health data access bodies, a country can have one or more.

[00:21:56] So for secondary use, there is no obligation of having one only health data access body. You can have more than one. Why? For different types of data or? That's a very long conversation. I think no one knows the why. We know the why it's important.

[00:22:17] Because, for example, if we're thinking about cancer nodes, we have a project called Candle looking at national cancer data nodes. Maybe you want to have a node representing other types of diseases. Maybe the country has so big like France or Germany that it might have more than one organization that puts together data.

[00:22:44] So rather than being very restrictive, then the regulation decided to be a little bit more broad and allow these data holders to be more than one. Because, you see, there's no problem. You can connect more than one bank of data to this network, which is the secondary use type of network.

[00:23:07] Whereas when you're talking about authority and you want to define, for example, if we use standard A or standard B, you don't want to have two authorities, right? So this is a different thing. Yeah. Yeah, absolutely. Absolutely.

[00:23:23] If we look at the role of patients in EHDS, the promise is that we as patients are going to have access to data and also a certain level of control over data. There's another project called My Health at My Hands that aims to empower citizens with the ability to manage their health data. And that project is aimed at ending, I think, in 2029. Yeah, absolutely.

[00:23:53] So to which extent, you know, when we talk about the role of patients in various projects, are we actually just talking about patients accessing the data? And to which extent do they have more power than that? I'm wondering because one of the challenges that's already present is that if there's errors in medical records, it's impossible to change them.

[00:24:21] And to a certain degree, while it's being super useful, ambient scribes are already also causing errors in medical records if, you know, these are not the transcripts or the data capture is not really scrupulously reviewed. Yeah.

[00:24:45] I think, you know, this is, I mean, definitely I've always been an advocate of citizens access and usage of data. We have a joint paper, me and Moon, Catherine Kronacki and Giuliana Ferrari from Pfizer, where we talked about the personal health data space.

[00:25:07] And that starts with the basis of access, but then it goes all the way in the pyramid to curation, correction, deletion, uploading, visualization, and then using that data for healthy giving. So I think, I think we will go up this pyramid, but there's a sort of a logical way, right? I mean, if you've never seen your data, how can you ask for corrections?

[00:25:33] How can you govern your data and decide who gets access to it in terms of secondary use if you don't know exactly what data people are asking you to have access to? How can you do? So it's, it's a, it's a, it's definitely something we need to work. Um, most, I would say most people in the ecosystem are not ready for this. Um, starting with the patients themselves, maybe the worst.

[00:25:59] Um, we've done work in Nishté, my university on, on capacity building for patient associations. We coined the concept of digital health data activism, um, or health data activism, um, to, to say that it's not enough to give the rights to the people through the regulation, which we have finally. But now you have to know how to exert those rights.

[00:26:24] You have to demand that these rights are made, um, real for you. But, you know, look at other human rights. And there's so many other rights that are not yet fully fulfilled. So it takes decades. It will take decades, uh, realistically for every European to understand their health data, exert the right to upload, exert the right to download, exert the right to edition, to correction. But if we never start, we will never get there.

[00:26:54] So it's fundamental to start an X share with the yellow button was a very interesting project because it allows a simple, um, technologically solid solution for patients to download parts of their health data. So this is, this is the beginning of, of, of, of a, of a journey of emancipation, data emancipation, um, that the patients need to do. But to conclude, why would they do that?

[00:27:24] Hmm. To get better care. Now, if you don't plug these efforts with integration of care and quality of care initiatives in Europe, um, we might be building a gigantic, uh, building of data, but, but not for any purpose. And data for the purpose of, of data is, is, is, is not good effort.

[00:27:49] So if you don't download and upload to get more integrated care or to better get better care, you correct your record, but at the end you get the same mistaken medication because no one paid attention to your correction. Mm-hmm. Then it's doesn't fulfill the objective. The objective of data, uh, you know, um, autonomy is to make you more capable of controlling the healthcare provision.

[00:28:19] Mm-hmm. That you receive. So, so we also need to, to, to, to do a lot on, on the areas of literacy and education. There was a few projects a few years. Now there's this idea that people will automatically know what to do with the data once they have access to it. Yeah. And that is a big mistake. Mm-hmm. Um, so on that front, the project I would highlight, which is not in your, um, in your, um, slide is Gravitate Health.

[00:28:48] It's a very interesting project working on the electronic product information and, and teaching patients and also facilitating the access. to medication information. Mm-hmm. How to take my drugs, how to, to take my medication. Um, that's one example of a project on, on that front. Mm-hmm.

[00:29:10] Um, I think, um, one of the challenges when it comes to EHDS, uh, and kind of making sure that people understand what it is, is, um, exactly the fact that, you know, progress happens, uh, uh, gradually. So, you have to go step by step.

[00:29:31] So, for example, we're going to start with international patient summaries, e-prescriptions, and e-dispensation, and hardly a few years later, it's to be expected and mandated that that also includes medical images, reports, labs, and other diagnostic results. And I think this is where some of the disappointment, uh, from the patient community, uh, comes, uh, from.

[00:29:54] Because when you say that patient data is going to be exchanged, people automatically can assume that everything's going to be changed. And then they're disappointed because, uh, not everything is available. Um, it depends on which country you're from, how developed that country is. Are you in the country that's already, uh, working on e-prescriptions across border or not? Because some countries are and others aren't.

[00:30:21] And, for example, even the I2x project, which ends in 2029, is combining 35 pilot projects across 12 EU member states to enable e-prescriptions labs and results. And the aim of the project is to impact 3 million people. So, naturally, my, uh, my question is, what about the other 400 million, uh, or more? Uh, when can they expect an impact?

[00:30:48] Um, so, there's no, so that you avoid the impression that nothing is changing, if you know what I mean. Yeah, I, I know what you mean. Uh, but I, I always ask the question the other way around. Uh, tell me one region of the world where health data has been moving around in the last 15 years. So, let's look at just three countries.

[00:31:15] Canada, U.S., and Mexico, do they share, do they share e-prescriptions? No, they don't. Even between the, the provinces of… It depends, aha, you mean, you mean between countries? Yes, yes, between them. But even between the provinces of Canada, which have more or less the same language, well, English and French in some of them, they still do not change, exchange data. But we are. So, yes, Europe is going slow, well, according to some.

[00:31:43] But if you look at the rest of the world, um, in some parts, they're not even moving. So, so that's the first answer. The second answer is, I think that's great that people are anxious, that people want things to be fast. And that's very good. And they just need to channel that energy in the right spots. When they choose a minister of health, when they choose a government, do they check the program? Do they talk about digitalization of health?

[00:32:11] When they vote in a party, do they vote on the party that promotes two-year fast digital transformation in health? Or they're choosing a party that is supportive of high, super high, or, or in, in, in bearably high data protection rules that make it impossible to share data. This is people's decision. People, I love when people talk about democracy and then they forget it's, it's their fault.

[00:32:41] If a country is not investing in digital health, it's the citizens' fault. It's not that the government's fault. The citizens chose a government that does not want to invest in digital health. And why do I say this? Because I come from a country that moved digital health very fast, very forward in the worst moment. We had a crisis, a financial crisis. We had the intervention of FME. We had the intervention of the central bank.

[00:33:09] We were without money and we still did digital health. Why? Because it makes a difference for health and it saves money. So even the countries that are not so rich, they should invest in digital health. But it's citizens' decision. They should lobby. That's why, great. I am very happy that 400, you said 432 million are not happy because they don't have the service. Good. Go to the streets and complain.

[00:33:39] Complain that you want digital health. You know why? Because it saves money. It saves health. It might even save your life. And I think that's the message. So we, yes, we are working on this. It's very complex. But I don't think there's a good excuse. We just need to do it faster. That's great. Let's do it faster with everyone.

[00:34:05] But then when we go to you as a prescription, for example, as a doctor, prescribe, and I say, no part of paper. You have your prescription on your phone or on the app or on the portal. And the patient can ask, oh, but I like the paper. Yeah, that's definitely also going to. For me, no, then you have to be, if you want that, then let's all go for it. That's demand for good services, of course, safe services.

[00:34:34] And the final answer on that is, yes, it seems slow, but we don't want to make it all so fast that we'd make it wrong. So the Europeans are traditionally a little bit more cautious. That is how we are. And it's not going to change. It's not because we see very fast going, you know, U.S. companies or very quick Chinese adoption. We Europeans are not like that.

[00:35:02] So we also have our own ways and we need to respect the culture of our continent. But again, I would like to emphasize this point. We are not by any means delayed compared with other parts of Europe in regards to international data sharing. By no means, I can tell you that. A lot of work with other parts of the world. Absolutely.

[00:35:25] When it comes to bridging the gaps and creating connectivity across Europe, it is clearly complex. It's complex already on national levels, especially when you've got fragmented systems such as Spain or Germany. And like even inside countries, multiple authorities and rules on who decides how things will be done.

[00:35:48] So how do the discussions look like about bridging differences in system maturity, data quality and national infrastructures? So how can you, you know, make sure that that is aligned? Especially with changing governments and health ministries and everything. Yeah, yeah, yeah, no, I understand.

[00:36:12] So I would say that the most obvious example of that, I'm proud to have been in the beginning of that conversation. It's now been, you know, maintained and taken up by many others. It's called the EMSEG. EMSEG stands for eHealth Member States Expert Group.

[00:36:36] This structure that represents all member states, one or two people from each country, meets almost, I think, every four months since 2016. So it's about 10 years now. And we have had meetings regularly. So we started that group when we were setting up the governance structure for the cross-border services. At the time, it was a subgroup of the EHealth Network that I was chairing.

[00:37:06] Then I became chair of the EMSEG. And then when I became chair of the EHealth Network, this moved to a very, very amazing lady from the Czech Republic, Clara. And she has been the chair ever since. And why am I saying this? Because we agreed two things. We agreed that regardless of the maturity of the country, everyone could sit on the table.

[00:37:33] Even if you were only going to do the services in five years' time, you could still attend the meetings. And see how others prepared, how others struggled, and how others achieved going live with the services that we were talking about. We were talking about patient summary and e-prescription exchange.

[00:37:53] So it is this combination of allowing everyone to go on the bus, but allowing the bus to stop when you are ready to go out or to go in. That makes a big difference. So I think countries are more than 10 years of difference in terms of maturity. Okay? I still remember that Finland and Estonia were the first going live in 2018. There are countries that are expected to go live in 2022.

[00:38:24] So there's more than 10 years of maturity just from the cross-border dimension. But this is how we do it. We keep everyone on the bus. We stop every now and again. We have regular stops. So everyone knows they are called waves. Once a year, you decide if you go this wave or the next one. So you can also plan your journey according to the maturity and the journey of others.

[00:38:52] And to your point about political instability, exactly because this is a somehow technical group, that's why we call it experts group and not the high-level ministerial representation. It's often the people survive the political changes. And that gives continuity to this group. Yeah. I know you are short on time. So I'm just going to ask one last question.

[00:39:21] So as we mentioned earlier, next year is going to be marked by defined health data access bodies on national levels. What do you hope that we will be able to say at the end of 2027? Where do you hope that AHDS will be? Like what's the best case scenario? So we will have that, but we will have more important, in my opinion, is the publication of the implementing acts for the exchange format.

[00:39:51] So many implementing acts, so again, these smaller, more detailed bits of legislation will come out. They have to come out in March 27 because it's a deadline. So I would say that by the end of the year, we know, we will know, and some projects like I2x or My Health at My Hands will be able to show that.

[00:40:16] We either wrote them well together as European Union, the member states, the commission, the experts, or we wrote something that is inapplicable or too vague to help apply. So that's going to be, so the best scenario you're asking me. So the best scenario is that they actually are applicable.

[00:40:40] The projects make use of these implementing tales and are getting ready to demonstrate on what we call implementation and demonstration sites, which are going to be very important. Because once you have, you know, five or six hospitals around Europe, one or two national authorities, one or two services in the EU wallet that serve as a demonstrator,

[00:41:06] that serve as an implementation showcase, then everyone can see it's possible. It works. And so it's possible to get a lab report, unload it and put it in, upload it into another hospital. So then people start to believe. And that's going to be very energizing. You've been listening to Faces of Digital Health, a proud member of the Health Podcast Network.

[00:41:33] If you enjoyed the show, do leave a rating or a review wherever you get your podcast, subscribe to the show or follow us on LinkedIn. Additionally, check out our newsletter. You can find it at fodh.substack.com. That's fodh.substack.com. Stay tuned.