Connectivity does not end at the hospital exit

In June, the VDE published a position paper that I co-authored as one of fourteen authors: “Vernetzte und intelligente Medizintechnik als Treiber eines modernen Gesundheitssystems”, PDF in German, opens in a new tab, networked and intelligent medical technology as a driver of a modern healthcare system. It is the result of the ITG advisory board “Qualifizierung HealthCare” and the DGBMT expert committee “Geschäftsmodelle Intelligenter Assistenzsysteme”, fifteen pages, a year of work. The core thesis in one sentence: the devices are there, the data are there, but they do not come together. The VDE calls this a “digital rough diamond – valuable, but not cut”.

The paper looks mainly into the hospital: intensive care unit, ventilator, infusion pump, monitoring. That is where the most expensive devices stand, where the wealth of data is greatest, and where networking most often fails because of proprietary interfaces, separate responsibilities for medical technology and IT, and a lack of staff. All of this is correct, and I was glad to sign up to it in our discussions. But I come from a different corner of healthcare, and from there the same problem looks a little different.

The chain nobody has in view

People with hemophilia treat themselves at home, intravenously, often several times a week, from childhood on. Their care is a network of four parties who are rarely in the same room at the same time: the patient with their diary, the pharmacy, which has dispensed the products since 2020 and must report every dispensing to the physician, the hemophilia treatment center, which documents and reviews, and the German Hemophilia Registry, which collects the centers’ reports every year. Every factor administration must be documented batch by batch and kept for thirty years. Which regulation requires what, we recently compiled on a dedicated knowledge page.

This chain does not contain a single ventilator. But it has exactly the problems the VDE paper describes for the hospital: data originate in four places, in four systems, and without networking they are typed up four times, passed on late or never evaluated at all. The paper speaks of “redundant documentation” and “isolated solutions”. In hemophilia, for decades, this meant paper diary, fax, binder.

Five demands, translated once

The paper formulates five fields of action. I have translated them for our world, because it shows how much of this already works outside the hospital.

Interoperability. The VDE calls for open, standards-compliant interfaces instead of manufacturer-specific formats. In hemophilia this is not a demand for the future: our electronic diary eDiary passes data to the German Hemophilia Registry and to the WAPPS-HEMO calculation portal of McMaster University, and the pharmacy software ScanDoc delivers the dispensing data directly into the patient’s diary; via the same interface, another diary, florio HAEMO, is connected as well. The principle behind it is the paper’s: data belong in the process, not in a silo.

Governance. The paper laments the separate responsibilities of medical technology and IT. Our variant of this problem is separate responsibilities between institutions. Most clearly in gene therapy: a specialized center infuses, the home center takes over fifteen years of follow-up, the registry and the Federal Joint Committee want data, and so does the manufacturer. The GTH explicitly recommends a shared electronic platform for this. We built one, Gene, and learned: the technology is the smaller part. The larger part is two centers agreeing on one record. There is now a knowledge page on this as well.

Cybersecurity. Security by design across the entire life cycle, the VDE demands. For a medical device that must hold the health data of children and adults for thirty years, this is not optional. That is why our data reside exclusively on our own servers in Germany, and why eDiary is deliberately a web app without an app store: no third party sits between patient and center.

Clinical validation. The paper demands evidence instead of promises, especially for AI. I share this without reservation, and we hold ourselves to it: what eDiary achieves in care was described in 2019 in Hämostaseologie by hemophilia treatment centers, based on 663 patients from 30 centers, not by us in a brochure. And on AI, we have written down on our AI page where we use it and where we deliberately do not.

Qualification. The VDE calls for training programs for medical technology, IT and nursing. Our experience from more than a hundred pharmacies using ScanDoc: the software is explained in an hour, the process behind it needs the conversation. Whoever saves on training gets a tool nobody uses.

What I take away from working on the paper

First: the VDE’s diagnosis is correct and applies far beyond the hospital. Second: outpatient care for rare diseases is a good place to try out networking, because the parties are manageable and the benefit of every single interface becomes visible immediately. Hemophilia has a few thousand patients in Germany, not millions; here a chain can be closed before it is scaled. Third, and this is my addition to the paper: networking is not a state you establish once. Products come and go, registry obligations change, manufacturers leave markets. The patients’ data remain. A system that keeps them readable for thirty years is, in the end, more important than any single interface.

“The digitalization of care is not a technical project but a structural change.” That is how the paper ends. I would add: and the change begins where the patient is. That is rarely the intensive care unit.

The position paper is available as a PDF from the VDE, in German, opens in a new tab; the press release, in German, opens in a new tab summarizes the demands.