Pharmaceutical companies and EU health data: On opportunities and challenges

Contents

Better, data-driven medicine – with this aspiration, the new Medical Register Act is intended to simplify and standardize the use of health data for research in Germany. However, the draft bill from the Federal Ministry of Health is causing heated debates about its implementation – with a few exceptions that approved the draft in their statements without comment.

However, the fundamental criticism from the Netzwerk Datenschutzexpertise goes far beyond such individual questions. In its statement (PDF), it warns that while the needs of research are being met, "even the most basic requirements of data protection (purpose limitation, transparency, safeguards, control) are being disregarded." Due to incompatibility with the Basic Law and European law, the draft must be "comprehensively revised." Yet, precisely this, arguably the sharpest statement, is not to be found on the ministry's website, unlike 66 others – although the ministry was informed about it.

The pharmaceutical industry also positions itself in this conflict between the promise of medical progress, concerns about data protection, and concrete economic interests. We spoke with Dennis Geisthardt from the Association of Researching Pharmaceutical Companies (vfa) about the plans, the criticism, and the central question of who will ultimately benefit from the valuable health data.

At an event hosted by the Technology and Methods Platform for Networked Medical Research (TMF), industry representatives welcomed the fact that companies can invoke trade secrets within the framework of the European Health Data Space (EHDS) to reject data requests. How does that fit with the common good when you receive data from a common pool but don't want to give anything back yourself?

The issue of intellectual property is naturally important to us, especially at the international level. We now have legal obligations to share data. This puts us in a situation where we have to list all datasets that fall under the EHDS scope – and that's probably 99.9 percent of all datasets we have – in a metadata catalog so they can be requested.

But the possibility of invoking trade secrets creates a loophole?

The question is: what happens to datasets that we have purchased, that we use as a basis for our research, and from which patentable results may be derived? If it becomes clear that Company A has very specific data on certain indications and Company B sees this, competition naturally arises. This can lead to distortions of competition and ultimately weaken the innovative strength of the entire industry.

In which specific cases can such claims to intellectual property, so-called IP claims, then be asserted?

We don't know that yet. That's the question that needs to be answered first. There is an industry initiative – the Innovative Health Initiative – that is trying to establish a framework and guidelines on how IP rights should be handled within the EHDS. But we don't have a final answer yet.

Where does the common good fit in?

Every research initiative is initially oriented towards the common good because we are trying to improve healthcare. In which cases this can then be substantiated with an IP claim, so that an application for the metadata set must be reviewed separately – we don't know that yet.

We want and can find a consensus that suits everyone – in specific data requests and specific data cases. But especially for legacy data – data that we collected or purchased before the EHDS came into effect – we need a mechanism that enables this consensus. Be it financial compensation or part of the fee schedule.

Compensation for whom?

If we as a company hold a dataset that we have financed ourselves – either because we purchased it or because we conducted a study ourselves – then the question is: how are the costs incurred handled? There must be a compensation mechanism for that.

In our forum, the question often came up whether patients shouldn't also get something back – after all, it's their data.

Absolutely, we share that. We also believe that, especially with regard to patients, there should be incentives for making data available for research.

In the direction of those who don't donate their data pay extra?

No, that must not be a penalty. You should not be disadvantaged by not sharing data, but you should have an advantage if you share data.

And how do you want to ensure in the long term that not sharing doesn't bring disadvantages?

That's difficult, I would agree with that. But the question is why people are willing to share some data completely willingly and others not.

Patients' willingness to share data is high – especially for non-profit research. Pharmaceutical companies have reservations due to profit motives, for example, expensive patents on medications that make access to therapies difficult and are perceived as self-serving.

We struggle with the reservations, which makes it more difficult. I wouldn't sign up for that for Europe. For example, our pricing systems in Germany and Europe work differently than in international comparisons. And that's a good thing, as long as they allow adequate innovation. Furthermore, it's like this: If I'm sick, I'm usually willing to share my data for research. And then, as a patient, I don't care whether it's used for pharmaceutical research or public research. That's always forgotten in this debate.

The rule of thumb often used in politics is: data protection is for the healthy. The question is: if we can provide data preventively that flows into general healthcare improvement, and I as an insured person perceive a direct added value – that something is being done with my data that can bring about an improvement for me or others in the long term – then the willingness will also be higher.

Wouldn't it be cheaper overall not to rely more and more on opt-out solutions for sharing health data, but on a good opt-in and reasonable information?

We had opt-in for the electronic patient record (ePA) for a long time. In 2022, I gave a keynote at a major statutory health insurer. I asked in a room full of health insurance representatives: Who here has an ePA? The number of hands was manageable. And who uses it? Perhaps two hands went up – out of 120 people in the room.

I had an ePA myself at the time because I was technically interested. But I didn't get anything out of it. I'm relatively young, have a manageable medical history. And even the chronic patients in the room didn't use it because they saw no measurable added value.

That there was no proper information is precisely the problem we have with all our TI applications to this day (Editor's note: applications of the telematics infrastructure). The added values are not perceived, not communicated correctly, not properly illuminated. This leads to absolutely low usage rates. We need to improve that.