How is Europe preparing for secondary use of data in healthcare through EHDS?

[00:00:01] Dear listeners, welcome to Faces of Digital Health, a podcast about digital health and how healthcare systems around the world adopt technology with me, Kjasa Zajc.

[00:00:18] Healthcare data has enormous potential not just for the quality of treatment for an individual, but also for secondary use for research and further healthcare improvements.

[00:00:32] Fidelia Cascini is a digital health expert engaged by the Italian Ministry of Health and involved in several European and national initiatives focused on the European health data space, primary and secondary use of data.

[00:00:51] Fidelia also published a book on secondary use of electronic health data, public health perspectives, use cases and challenges, which is an open access book so you can reach it just by going to the link in the show notes. Fidelia Cascini is a digital health expert. The book starts by categorizing electronic health data because there are several categories to differentiate from.

[00:01:18] From real world data, administrative data, clinical trials data, genomics data, patient generated data, national registries and social and behavioral determinants of health. And not forget the electronic health records data.

[00:01:34] In this discussion, I spoke to Fidelia to talk to her about her new book, the findings about secondary use of data across Europe, the European health data space and expectations in the future. So today we're going to learn more about that.

[00:01:56] And in the upcoming episode, we're going to hear a little bit more about the secondary use of data in Germany, because Germany is going to soon open its health data for research purposes. But more about that in the next episode. Before we dive in today's episode, do leave a rating or a review wherever you get your podcast.

[00:02:23] It's always good to get some feedback from you. So if you can also leave a short review, a message, anything that will also let us know what types of topics you would like to explore further. And we can take that up and do some research for you. So thank you. I appreciate any thoughts and comments that you may have already given. And I hope you will enjoy today's episode. So let's dive in.

[00:03:07] Fidelia, hi, and thank you so much for joining this discussion on Faces of Digital Health. We are going to discuss a new book that you published about the secondary use of electronic health data, the public health perspectives, use cases and challenges. And I love the first chapter of the book where you started by categorizing electronic health data.

[00:03:35] I think that the first association when we talk about health care data is the electronic health care records. But you actually outlined nine categories from real world data, administrative data, clinical trials data, genomics data, patient generated data, national registry, social and behavioral determinants of health and EHR data and more.

[00:03:58] So I think that it's good to start a discussion with the distinction between these data types, since we are often very inaccurate and very general when we talk about secondary use of data without defining which type we mean. So let's start with that. How important do you find this distinction between data types? What's your observation in terms of how people perceive it?

[00:04:26] Thank you, Tiaza, first of all, for this invitation. It's my pleasure to stay here and discuss about health data and the possible application in digital health. The idea to distinguish between all these different categories of data is because we need to understand that behind data, we have information. So different types of data and categories of data bring different kinds of information.

[00:04:56] And when you need one or the others to use them for secondary uses, you need to have clear ideas on where to find the right information that you need to find your results and to find the proper information that you need. It is true that electronic records are the most used and most available category of data.

[00:05:23] But there are important differences between the different categories of data. For instance, let's think about electronic records. They are strictly related to clinical situations, to patient recoveries, to situations where a patient is affected by illnesses or different illnesses or need healthcare services.

[00:05:45] But at the same time, we should need for our studies for the secondary use purposes, patient-generated health data, which are completely different source of information because they are spontaneously provided data.

[00:06:04] And they are related, for instance, to regular monitored information provided by the patients to health apps or wellness apps and also related to a patient's experiences or other aspects which are related to patient lifestyle or patient behaviors. So if we have these different categories of data and if we merge these different information, the results that we can reach are completely different.

[00:06:34] And also the picture that we can have in front of us is completely different. And another example could be, for instance, the difference between clinical trials, which are realized in very controlled environments and where the conditions regarding the situations and the recording of data are very controlled and prepared and expected, which is completely different from the real world.

[00:07:03] And for instance, when after a clinical trial, you have results regarding medications, for instance, and you need to apply these results in the real world, you can find that the reaction of the patients is completely different in the real world.

[00:07:19] Because, for instance, if you think about medication administration of management, the patients could prefer orally medications or other kinds of medications to stay adherent to the therapy, which is different, for instance, from the type of study that you have in the clinical trial where you tested more injection or other situation where you needed a hospital to receive this kind of therapy,

[00:07:47] or you need a control environment. When you transfer the results of a clinical trial in the real world, you can perceive a distance between the availability and the possibility to stay adherent to the results of the trial because of barriers in the real world that you cannot calculate before when you are providing the study.

[00:08:13] This is the reason why when we use different sources of data and different categories of data, we can reach different results because we have a more clear picture in front of us of different information for the same purpose. It's a very fascinating field when you think about medication applications and how different types of that can impact basically the adherence to care.

[00:08:40] A very simple example can be the different forms of vitamin C that you can buy in the pharmacy. Some people will prefer the tablets, some people will prefer the powder that you can just add anywhere. You have candy-like types, which in essence can help some people basically to ingest more of the vitamin C than they otherwise would.

[00:09:02] And especially when it comes to more serious medications that can be very important, such as patches for dementia instead of pills. So, very complex field, as one could say. Which type of health data, given this complexity, do you see is most suitable or even most used for secondary purposes?

[00:09:31] I think that right now we can expect more results and more application regarding electronic health records because of the history of this kind of category of data. Because electronic health records are based on the need of people to receive healthcare. So, we have already the availability of different forms of electronic health records.

[00:09:57] And especially in Europe, we have a different level of maturity. We have more advanced countries, especially in the northern part of Europe, in the use of electronic health records. They are more advanced also in the secondary uses of this kind of data.

[00:10:17] But if we think about the opportunity to have value from the amount of information, I think that another important category of data is that related to the patient-generated health data. Because in this situation is something which is regularly collected through a lot of tools, electronic tools and apps.

[00:10:47] And also, people in general, not only patients, are used to use this kind of tools for their monitoring in their daily life.

[00:10:58] And it's important to receive continuously this kind of information because at the end, you can profile the behaviors of people, the lifestyles, the attitudes, the reactions concerning, for instance, environmental risks.

[00:11:18] So, I think that these two categories of data, the first, which is the electronic health records, because it's strictly related to the needs of people and patients to receive healthcare. And the other, which are patient-generated data, will be the most used in the next year. Nevertheless, we know that at the end, we will have available, especially with the HDS framework in the European Union, a lot of different categories of data.

[00:11:48] But the most used from the beginning will be these kind of categories. I have several follow-up questions related to that. And the first one is, what kind of discussions do you see among health ministries? You work closely with the Italian health ministry as the digital health advisor and also on the European level for EHDS.

[00:12:09] So, what kind of discussions are you observing in terms of how can this patient-generated data be incorporated with other health data so that it doesn't overwhelm the healthcare workforce with just too much noise?

[00:12:26] I think we've made some progress in the patient-generated data in a sense that in the past, you would just get information about how much time did you sleep or how much did you walk or how much movements did you make.

[00:12:40] But now we're slowly moving into data interpretation era, where also the providers of either wearables or apps to monitor all these signs can offer you some advice or some interpretation of what does it mean if you slept less than six hours in the last month on average. So, how do you see that the usability of the data is going to translate to how clinicians treat patients?

[00:13:09] Yes, I think that there are two levels of approach. Because the first one is related to the trustworthiness of the tools. So, not all the tools, the digital tools, are usable for health purposes.

[00:13:24] And you have to clearly distinguish if they are tools that you can use in a good way for providing health data that can be merged with the data from the electronic triggers or other official tools for recording data.

[00:13:43] And other bad tools, which are available possibly on the market, but which are not in terms of interoperability, cybersecurity, or also the liability of the information related to the health situation of the patient. And the second element is related to the integration of this kind of information.

[00:14:06] So, it is not possible that data provided or recorded by the professionals are considered at the same level of data provided by the patients. For instance, patient-generated health data. We need to distinguish the two levels, even if at the end we can use both and we can complement one with the other to have a clear picture of the situation and approach the patient in the right way.

[00:14:36] But we cannot allow the patient to put his data, his or her data in the electronic health records because it could generate confusion and it also is a problem related to identifications and also the affordability of data. So, there are a lot of aspects to consider.

[00:14:56] So, these two sources are very important to consider, but they cannot be mixed because they can be used to complete the information, but we cannot mix them because at a different level. Absolutely. It's a different quality. It's like comparing a clinical grade diagnostic device to a point-of-care device that's used for monitoring at home. Exactly.

[00:15:24] The prices differ, the quality differ, but at the same time, the comment to that is usually that when you're trying to monitor something on the long run, the trend is more important than the accuracy, like the 100% accuracy of the individual data points. So, how do you imagine that this patient-generated data could be used in care?

[00:15:48] Is it just patients sharing some information that they see with themselves, with the healthcare provider? Or is it actually some sort of more concrete integration into the existing systems, IT systems used in healthcare? It depends on the situation.

[00:16:07] If you have, for instance, the need to monitor the patient at his or her home, a telemedicine tool, you can monitor through wearable devices, for instance, chronical conditions where the patients need to be continuously monitored by the doctor for understanding what is the situation, how is the illness evolving, and so on. So, it is something which is strictly related to the healthcare professional activity.

[00:16:38] So, it is something that we have to do with the healthcare provider. So, it is something that we have to do with the healthcare provider. So, it is the doctor which asks patient, please give me your data through your health app because I need to know where the situation is going, how the situation is going, and what you need, and so on.

[00:16:53] So, on the other side, we could have different tools, for instance, wellness app or other kind of health app where it's the patient that wants to record a specific kind of data to monitor his or her health status and can bring this information and experiences and symptoms or other kind of conditions to the attention of the healthcare professional.

[00:17:20] One of the healthcare professional who can decide if use or not this kind of information. So, it depends on the tools and it depends also on the conditions where the patient is monitored by doctors or is it he is choosing a different approach to provide more information to be better assisted by the doctor.

[00:17:44] One of the big discussions that some countries are already implementing is the implementation of patient reported outcomes into the electronic health records. And I think one of the discussions there is also who is then going to analyze that data if you add it to the patient portals and how is the feedback loop going to look like?

[00:18:11] How is the doctor going to respond? Is this going to be used for a different type of research? Can you maybe just briefly comment the development that you observe in that area? I think that patient reported outcomes and experiences are extremely useful. And it is something where the healthcare systems should invest because, as we told before, you cannot intend the healthcare provisions of care, the provision of healthcare,

[00:18:40] if you do not consider the experience of the patient and the reaction of the patient to prescriptions of doctors and indications of professionals. You need to know what is the reaction of the patient and you need to know how to approach the different type of patients.

[00:18:56] So it is extremely important to use at the best this kind of information and to help the patient to know and to understand how to better use these tools because not all the patients are able to manage this kind of tools. So it is very important that we facilitate this process of awareness for the patients to use these kinds of tools.

[00:19:21] And on the other side, the healthcare providers to consider these kinds of results because they are extremely useful to help people at best for the improvement of their health. Absolutely. Absolutely.

[00:19:35] When we talk about the secondary use of data in Europe, one important topic is the European health data space and this aspiration that when we harmonize the data exchange across Europe and define data standards for that exchange, we're not only going to allow clinicians to get information about the patients in front of them, regardless which country these patients are coming from,

[00:20:05] but also that we are going to create this huge data pool that's going to be available for research. I mentioned already that you are working closely with the Italian Health Ministry as the digital health advisor. Since 2021, you were involved in several EU initiatives related to primary and secondary use of data to support the framework for EHDS.

[00:20:29] So how fast do you expect that data could be harmonized and that the needed infrastructure to create these connections and the data pool on the EU level could be prepared realistically? One of the timelines mentioned is six years. Anyone working in healthcare IT expects that it's going to be delayed, especially for various countries, because as you said yourself,

[00:20:57] different countries have different maturity of digitalization. Yes, I can confirm that the situation in Europe is the same regarding the commitments that we have as European Union countries to reach the same goal and to go into the same direction for an harmonized division of the use of data and the secondary use of data.

[00:21:20] But the pace where we are proceeding is completely different because we have countries as Finland, Estonia or Denmark, which are really advanced in the digitalization of the healthcare systems and others which are at the medium level.

[00:21:41] Well, these kind of differences are also related to the dimension of the countries because it is completely different to digitalized the healthcare system of a country which counts 6 million of people and another which has 60 million of people versus 60. It is completely different as Italy, for instance.

[00:22:06] And we have another issue to consider that in Europe, we have also countries that have different healthcare systems in terms of federated systems. So, for instance, Italy and Spain, where the healthcare systems include different regions and you have to create an harmonized approach.

[00:22:35] Also, considering the different levels and the different management that you can see in the different regions, it is completely different. So, we know that after two years from the EHDS regulation entry into force, which is now the beginning, we have to start with the implementation of the digitalization of healthcare systems according to the EHDS regulation.

[00:23:02] For sure, we will see that some countries, we will be more advanced at the end of these two years and others which are at the starting point. And we have to consider this. But I can say that in the next eight years or 10 years, the picture will be completely different.

[00:23:22] And also, bigger countries and with more difficulties in the management of healthcare systems, we reach a very good result in terms of interoperability, digital maturity, the opportunity to be aligned to allow the secondary use of data. And especially regarding the secondary use, it will be different if we consider the different categories of data.

[00:23:49] As we told before, EHR will be more usable and faster than, for instance, genomic and other kind of omic data, which are more sensitive data and require more protections, or also wellness up data that are completely different kind of data. It is more difficult to protect the privacy of people.

[00:24:16] So we have to manage different situations that can need more time to be aligned and to give the good results to protect people and their rights. Absolutely. It seems as if we are constantly returning to the first question of the importance of differentiating between the different types of data.

[00:24:43] How do you think that expectations around secondary use of data should be managed? I'm thinking that from the patient perspective, the most inspiring part of secondary use of data is that it would help researchers discover new drugs or enable tech development and AI algorithms based on patient similarity metrics that could enable a more tailored and personalized prevention in healthcare.

[00:25:12] So how do you see this aspiration? And I also really liked the term patient similarity metrics that you outlined in the book. So maybe we can start by explaining what that is and how it's useful for the secondary use. I think that patients are positive towards the secondary uses of data.

[00:25:35] However, they are also worried about how their data will be managed, will be protected, will be used. So I think that they need more information and more, they need to be more reassured that the uses, the secondary uses of data here for good purposes.

[00:25:58] And we have noticed that in the scientific literature, that the attitudes of patients toward the secondary use could be very different, for instance, considering the different recipients of data.

[00:26:17] So if you compare, for instance, a public authority or healthcare provider, you can see that patients are more comfortable in sharing data for secondary uses.

[00:26:31] While on the other side, if you have private companies or technology providers or pharma companies, the reactions of people is more limited and restricted because they are not comfortable in sharing health data.

[00:26:49] Patients need to be more informed and needs to be how this process will function and what are the data that will be used for what purpose.

[00:27:03] And I think that right now, what we can do is to inform our best people that within the HDS framework, there are all the guarantees that people can need to be comfortable with regarding the secondary use. Because it is under the responsibility of the healthcare authority. It is something which is strictly related to purposes for public interests.

[00:27:33] It is a framework which is built to protect natural person rights. And even if the call, it is to, for sure, also allow data, the development of data market or other benefits for social improvements and economic advantages. But people rights and natural person rights are at the center, also considering the secondary use.

[00:28:01] And one example of this is also the right of the opt-out. People can decide to limit, to restrict the secondary use of some categories of data for some purposes. It will depend on single countries' approaches, but they have this right included in the regulation.

[00:28:22] So they, all of us, because people, or all of us, will have this kind of opportunity to protect and select data considering the different purposes. What do you think this choice and this power of sharing or not sharing means in terms of what is going to be digitized in the end? And here I'm mostly referring to the most sensitive type of data.

[00:28:51] That's the psychiatric evaluations or clinical psychologist evaluations, which is something that people are most concerned of. And I recently spoke to a clinical psychologist who said that if you're going to be forced to share all this data, then people are not going to tell us anything anymore because the clinical assessment consists of basically the person's personal history.

[00:29:20] And then there's also the clinical evaluation and the conclusion. So this clinical psychologist said that it's clinical psychologists are most open to maybe sharing the conclusion that doesn't include all the personal details about an individual's lives, which are important for the clinical psychologist to make the evaluation, but they're not necessarily as important for all the other clinical workers that come in contact with this person.

[00:29:49] So, yeah, basically, where do you see those concerns? I think that patients can have really the control of his or her data within the EHTS framework. If we consider primary use, for instance, the patient can hide and can limit the access of healthcare professionals

[00:30:12] to specific categories of data or information or situations which are related to his or her health status. It's up to the patient if he or she needs to receive healthcare services, considering or not information which are included in his or her history, medical history.

[00:30:39] For sure, this has risk because if a patient denies the access to healthcare professionals to specific information, the professionals won't know the results of that kind of information and also the provision of healthcare and the services provided by this professional will not consider this kind of information.

[00:31:04] And the risk is that the intervention will not adherent to the clinical condition of the patient. So, it is a risk that the patient have to consider if he or she wants to avoid the access to specific categories of data. But considering the secondary use, it is a different situation because the EHTS regulation

[00:31:28] is an obligation within the framework of the EHTS regulation to use data for secondary purposes only if they are anonymized. And in the case where they need to be pseudonymized for specific reasons related to the type of study or research and so on,

[00:31:48] it should be clearly justified and all the process regarding the treatment of data is under the control of the public authority and inside the secure processing environment which are built to protect the information regarding the single patients and to protect from the risk of re-identification of that specific patient.

[00:32:15] So, at the end, I can say that never before than now patients had the same opportunity to control and to limit the visibility and the access to their personal data related to their health conditions like we have now. A very advanced, a very important advancement in the management of data.

[00:32:43] So, if you would summarize, we would say that patients can have control, but they also need to be aware of that control. They need to know, have the digital health literacy to know exactly how to do that, which also means that healthcare systems, ministries, national health institutes need to invest a lot in making sure that this information is out and that basically people know and are not afraid.

[00:33:12] And I think this is often where challenges arise, where everything that's happening in terms of healthcare system development or healthcare digitalization, the marketing part is usually missing for health systems to really make sure that the awareness is there. Yes, it is perfect. Yes, it is perfect. Exactly like that.

[00:33:35] They can choose, but they have to know that this choice has consequences and they have to be aware regarding these consequences. Absolutely. What do you see as the biggest barrier or barriers to success in the UHDS ambition? A lot of work has been done already in the past. Several European projects were facilitated and basically support the ambition that the UHDS has,

[00:34:05] but we're still basically in very early stages of now implementing the changes prepared by EHDS. So, where do you think we might struggle most? Okay. We have a very important commitment by the EHDS regulation to develop this area, which is related to digital health, digitalization of health data,

[00:34:31] and also data uses for both primary and secondary uses. The barriers that I see are related to the time because we have a very short time, if you consider the change that we have to implement in our healthcare systems, and especially in those countries which have a large population and which have a federated healthcare system,

[00:34:59] because you have to involve and to align different kind of policymakers which are working on the same thing. But I can see another big issue, which is that related to limited resources, because even if the European Commission is very generous in providing funds and opportunities

[00:35:24] for member states for the implementation of the digital health and also the implementation of the EHDS requirements. But we have limited resources because the healthcare systems need a lot of resources, because we have a lot of investments also related to our important characteristics of our populations in the EU,

[00:35:51] as for instance, elderly people, chronical conditions, and advancement of opportunities related to high-tech tools, but that should be implemented universally within the different countries to provide the same quality and the same level of services at the different levels. Even if we have a lot of money to invest in this,

[00:36:21] but there are not so many as we need for implementing this in this short period of time. And also because we have to invest this also in human resources to increase the competencies and the capacities of these people to manage this kind of change. There are several barriers and it is very challenging, our goal right now. Yeah, absolutely. I think this is a crucial one.

[00:36:50] How do you attract people to work either in the public sector at the ministries or just increase healthcare IT personnel on the institution level and also create competitive salaries that will not disrupt the payment policies in public administration? So a huge one, as you said yourself. If we go back to the book,

[00:37:19] you write a lot about the public health perspective of secondary use of data. So the potential for disease prevention, disease control and detection, which is really important. But I think we also need to take into account the geopolitical changes and the funding disruptions that we see for WHO from the US side. So how do you see EHDS in light of these global changes?

[00:37:49] For example, if we suddenly run into another public health crisis because some of the communicable diseases or some of the diseases that we've been successfully preventing with vaccines are popping up again in some parts of the world. Do you think that EHDS could serve as a protection for Europe because we still have these early detection mechanisms in place?

[00:38:19] Yes, I think this is a very important step forward for the European Union to be prepared also for possible public health challenges that we for sure have to be... We have to know that they are possible and we have to be prepared to face them in case they will happen again. And we experimented this.

[00:38:45] We experimented the importance to have an integration of information, not only at the single country level, but in all the European Union to prevent and to manage when they happen all these kind of challenges. So I think it is very important that despite all the challenges that we are noticing because of these geopolitical complicated situations,

[00:39:12] we as Europeans will be ready to face by ourselves these kind of challenges. And it will be possible because once the EHDS regulation will be fully implemented and we will be completely aligned and we will exchange all the information and data related to healthcare systems

[00:39:39] for the management of the surveillance, of the facing of challenges and for cross-border threatens and so on, we will be for sure more capable to reach sooner and better our results. So I think that this is a very important point and it is the most important lesson that we learned after the COVID-19.

[00:40:05] And it is something that we have to preserve for the future. Maybe just one last question. When you were writing the book, did anything surprise you? What was your favorite part of the research? And how long did it take you to basically combine all the information? It is when I collected all this amount of information, which is a huge amount of information.

[00:40:33] My goal was to provide a useful and easy to read tools for all the operators and technicians and people that need to know to have an overview on this area, this field, and how to manage this for the implementation of the EHDS and the goals and the results that we need.

[00:40:59] So I look at the global situation, global picture, but my heart was bidding for the European Union and the results that we want to reach in this short time. My favorite part is the last chapter, which is related to the future perspectives and opportunities, because I think that we have to capture them. We need to do them.

[00:41:27] And this is very important for our future and for the future of our sons and generation of ourselves. So the first two chapters, which are more focused on categories of data and purposes of data, are to prepare the last chapter, which is how to capture the opportunities that we have for the future. And I hope that at the end, the reader can find the information that need

[00:41:55] to have not an exhaustive picture, but what they need, the readers to go ahead into the right direction to realize our objectives. And it's worth mentioning that this book is an open access book, so anyone can access it and read it. I'll make sure to add the link to the show notes so people can browse through the book. And for today, Fidelia, thank you so much for joining this discussion.

[00:42:25] You're also going to be at HIMSS Europe, where you're going to be talking about secondary use of data and EHDS. So I look forward to meeting you in Paris. Same for me, and it was a great pleasure. Thank you, Tessa. See you in Paris. You've been listening to Faces of Digital Health, a proud member of the Health Podcast Network. If you enjoyed the show, do leave a rating or a review wherever you get your podcasts, subscribe to the show,

[00:42:53] 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.