An interview with Prof. dr. JJM (Hans) van Delden, Professor of Medical Ethics at the University Medical Center Utrecht, and WP 7 co-lead BigData@Heart. By Ghislaine van Thiel, Associate Professor of Medical Ethics, University Medical Center Utrecht.
Professor Van Delden is the academic lead of work package 7, concerned with Governance, ethical and legal aspects of the work in BigData@Heart. On October 29, this work package will host a webinar on datasharing and dataprivacy in international health research. In this webinar, a basic model for datasharing in BigData@Heart will be presented. What else is currently going on in WP7 and what can we expect? We met with Professor Van Delden to get his view on these matters.
Q: You are the academic lead of the Governance work package (WP7). Can you tell me a bit more about the work you are taking on in BigData@Heart?
A: I think everyone recognizes the potential of bringing together an enormous amount of data: there is huge medical promise in there. At the same time, we have to recognize that if you have that kind of datasets, you really also need good governance: you need to know who has access, you need to know when data can be shared, you need to know what kind of research can be done with these data. And so a lot of ethical and legal questions arise, in ways that may not have been envisaged at the time of data collection.
So, as you can imagine, we felt within BigData@Heart that it was important to set up a governance structure for responsibly doing this part of the work and – at the same time – also setting an example for other big data health projects. And as with all research projects, one starts with collecting the existing structures, existing literature and from there on build your own structure. The focus in this part was on the legal aspects: what is required by law? Which regulations apply to our research? We made a review of the literature on principles and guidelines for international health datasharing – which is currently under review for publication. We subsequently starting with the setup of a basic model for datasharing. This is a first step towards a more definitive model for datasharing that is going to be a part of the governance structure that we are developing for BigData@Heart. Currently we are working on this improved model for datasharing. This is however only one step towards the final goal of work package 7 –which is to integrate legal, ethical and regulatory guidance for responsible conduct of big data health research into a governance framework.
Q: One of the key elements of your work package is involving patients. Can you tell me a bit about your motivation for this part of the project? Why is this important?
A: One particularly important aspect of our work is that we immediately acknowledged that if you start using data as we do in BigData@Heart, then you have to be serious about public and patient involvement. We ask the patient and the public to trust us to use the data in a responsible way. But just asking for trust is not enough. You really have to earn that trust. And earning trust can be done in several ways - even needs to be done in several ways – but one way of course is patient and public involvement. This is very important for several reasons. One reason is: it’s the patients’ data, you know. So basically nothing should be done about them, without them. Now of course, if the consequence of that phrase would be that we have to go back to each patient each time we want to do something with their data, that would be huge and it would probably limit the things we can do in research. So instead of doing that, we need a substitute for direct involvement by individual patients. I hope - but this is not always the case – that patients are aware at the time of data collection that their data can be used for research purposes. But they also entrusted a certain body to decide on what can be done with the data. And to safeguard that trust, they need to be part of this. Not necessarily every individual patient, but at least a representation of them.
A second argument is that in using the insights that patients and the public bring to the table, we make science more socially relevant. It helps to aim for the right questions and the relevant outcomes. Third, and this may sound a bit philosophical, I think nowadays everyone recognizes that science is not without interests and science is done in a way in which interests do play a role. In the old days, we could quote for example the sociologist Merton who claimed that science was completely value-free and factual. But all of that is rubbish. Nowadays we acknowledge there are always interests at stake in scientific endeavors: of researchers, funders and others. And it is only fair to acknowledge that patients also have an interest. And therefore: why not add them to the discussion so that their interest can be heard as well? We need to involve all the stakeholders and not just the professional ones.
Q: Another question is that there seems to be a variety of methods for involving patients in research. And even though I understand you are now just starting this part of your research: can you talk a bit about what you think are potential factors for success or failure of patient involvement strategies in general?
A: I am very grateful for the words ‘a bit’ in your question, because the truth is: what I call the science of patient input is what I would call an emerging science. It is not as if we know exactly what to do and how to do it. We have ideals, we have reasons – I just stated them – and we have a bunch of methods. But there is no cookbook that tells you what to do, when and how. So I can only make a few provisional remarks. I think there are some things that can be said, for instance: involving a patient who is very much a look-a-like of yourself and then claim you have accomplished your patient involvement is probably wrong. And so we need to involve a diverse group of patients: not just one but a group. And you can involve them in a range of activities on a regular basis, for example setting the research agenda, interpreting the results or making a summary for the public.
Also I think it is really important to try to reach out to groups that a less health literate than others, and optimize the representation of these people within your group. And we should do that even if getting these people to sit with you on the table may be too high to aim for. I guess these are my best suggestions at the moment.
Q: After BigData@Heart was started, you collaborated with – among others – Amitava Banerjee and added a project to your role in BigData@Heart: on the topic of the social license to operate. What is this about?
A: It follows nicely on what I said before: what you do if you ask people to use their data is you ask them to trust you. And most doctors and researchers can be trusted, but I think it is wise to have ways to really earn that trust. And if you broaden this a bit then you can start to speak of a social license. Which is the idea of a license that a group of people – let’s say a community or a society – gives to researchers to use their data in a responsible way, and to not use them in ways that would not be acceptable to them.
If you want to see a breakdown in social license, look at what the Cambridge Analytica affair did to Facebook. They sold the data of their users for purposes that could never have been the purpose of the users and thereby ruined their social license. So this idea of a social license is in my view a very interesting idea, but this should not be a rather airy, intangible construct, but instead it should be something we can really work on. So what we will do is find empirical ways to strengthen social license and we hope to show what has an effect on social license and what not. In this way we want to put the concept on the ground and prevent is from being an idealistic but theoretical concept that everyone applauds but is never used.
Q: Well, I think this is much of what we always wanted to know about work package 7. Thank you very much!
A: Thanks for having me.