Medical expert Ferdinand Gerlach contemplates the future of healthcare
Hospital reform, electronic patient record, telemedicine: Hardly a day goes by without the future of healthcare making the news. Professor Ferdinand Gerlach has advised policymakers for many years as chair of the German Advisory Council on the Assessment of Developments in the Health Care System. We join him for a conversation about outdated structures and missed opportunities – but also about the potential offered by digitalization and the important role of the general practitioner.
Forschung Frankfurt: Professor Gerlach, many articles in this issue of Forschung Frankfurt deal with the potential of increasingly individualized therapy. Do we need more specialists and fewer generalists?
Ferdinand Gerlach: That’s a nice question for a generalist. If we have lots of specialists and supramaximal specialists, we also need someone who holds all the strings. Someone who maintains an overview and reaches the right decisions with individual patients, someone who is prepared to talk to them when the going gets tough. Generalists who mediate between different specialists if patients have multiple conditions and protect them from excessive or incorrect treatment. When several specialists are involved, ideas as well as diagnostic and therapeutic approaches snowball, and there can be a lack of coordination.
What are the advantages of centralized care through general practitioners?
We call this a gatekeeping or primary care system: This is a matter of course in all Scandinavian countries, the Netherlands and Great Britain. Such a system does not exist in Germany with the exception of the State of Baden-Württemberg: Around two million patients there have voluntarily opted for GP-centered care. We have evaluated this scientifically over the past 15 years and been able to show that patient care improves when it is coordinated by primary care doctors. In a population of 119,000 diabetics, more than 11,000 serious complications were avoided in just ten years – including blindness, amputations, heart attacks and stroke.
What are the arguments against introducing such a system in Germany?
Patients in Germany have the right to choose their doctors, and the freedom to visit any specialist without any referral is perceived as a positive thing. However, when a patient with a headache goes to the wrong specialist without the prior advice of a general practitioner, it is a far less positive thing, as they can end up with a lot of unnecessary and sometimes false-positive diagnoses. In Germany, unfortunately, there are parallel structures in specialist care: We have many specialists in the outpatient sector and in hospitals, which leads to major problems as far as coordination is concerned. We need targeted referrals to the right specialist.
Yet it is often difficult to get an urgent appointment with a specialist without a prioritized referral.
Appointment bottlenecks, which particularly affect patients with statutory health insurance, are not due to a general lack of specialists, hospitals or general practitioners, but to wrong incentives and enormous inefficiency in the entire system. Let’s take quarterly billing for general practices as an example: If a doctor exhausts their budget in a quarter, taking on new patients is unprofitable, which results in too few appointments for people who are really sick, patients with chronic diseases, those with multiple conditions and urgent cases. Ultimately, poor or wrong incentives lead to overflowing practices clogged with patients who belong elsewhere in the system.
Are there any other poor incentives?
Yes. In Germany, doctors are only paid if they actually do something. If they tell the patient just to rest, exercise or change their diet, they do not earn any money. Doctors that do not present a diagnosis or offer treatment harm their own economic interests. In the outpatient sector, doctors spread as many individual services as possible over several quarters. For inpatient care, flat rates mean that as many left heart catheterizations or as many knee and hip replacements or vertebral body operations as possible are performed – even if physiotherapy would make more sense in individual cases. Doctors and hospitals are obliged to behave in accordance with poor incentives by providing services on a large scale that are lucrative from a business perspective.
Is there an alternative?
In the Scandinavian system, patients register with a GP of their choice. The practice is paid well for taking care of its patients. The healthier the patient is, the better it is for the practice. In Germany, good health is, to put it bluntly, “bad for business”. Healthy patients do not go to the doctor or the pharmacy and they do not need a diagnosis or therapy. Our system has an implicit incentive to keep people sick. This is disastrous, but unfortunately true.
Chronically ill patients are the best patients for general practitioners.
We have already established that general practitioners only earn money if they do something. Doctors can only bill services if they say you are sick or need to diagnose or rule out illness. Hospitals only earn money if they examine you, operate on you or provide any billable services. And the health insurance companies only earn money if their members are sick. The sicker they look on paper and the less they cost the practice, the better. It’s a crazy system.
Wouldn’t a registration system lead to fewer diagnoses and less treatment?
To be effective, a registration system needs to be combined with performance incentives: A primary care practice would then receive a lump sum for each patient registered with it. This would make up around 80 percent of its income. The remaining 20 percent would come from special services, which might be, for example, attending to patients in nursing homes or evidence-based screening. Practices could be remunerated according to quality criteria, for example if they achieve a certain influenza vaccination rate. The electronic patient record (EPR) shows whether a patient is being treated according to the guidelines.
Why is it so difficult to change anything in the healthcare system?
This is due, among other things, to self-administration and federalism: In Germany, we have 17 health insurance associations and around 100 health insurance companies, 16 health ministers at federal state level, 17 hospital federations, and so on. And they all have their own particular interests. The hospital system is no longer sustainable, nor are the outpatient practices in their current state. We still have a long way to go as far as digital transformation is concerned.
Can you please tell us more about where we stand now in terms of digitalization?
Until now, digitalization in the healthcare sector has meant the digitalization of forms such as doctor’s certificates or prescriptions.
E-Mail instead of Fax.
Exactly. But we need a transformation. Bad analog processes need to be rethought and redesigned, and other professional groups need to network more effectively with other stakeholders. Today’s tech giants, for example, immediately implement platform-based economic models. Amazon was once a bookseller, today it is in the process of building a digital global ecosystem and making headway into digital health services.
How does this change healthcare?
In the USA, Amazon offers innovative diagnosis and therapy services, especially in the primary care sector. It initially launched telemedicine via Amazon Care and now provides this service through the Amazon One Medical Pay-per-visit platform, which in between has offered access to several video consultation providers since 2022. Amazon Diagnostics offers rapid tests and even on-site tests where a driver collects the sample and you get the result back on the same day. The third pillar is the RxPass from Amazon Pharmacy, a monthly flat rate for only five dollars that covers 53 prescription medications and includes delivery.
The same price as a coffee at Starbucks.
Yes, it’s an insanely competitive price. In Texas, Amazon are now using drones to deliver drugs. More than 500 different medications can be dropped on your doorstep within 30 to 60 minutes. In 2023, Amazon bought One Medical for $3.9 billion and acquired 182 large primary care centers, which are now connected to Amazon’s app-based telemedicine system. For an annual fee of $99, Amazon Prime customers have access to video consultation with a One Medical doctor at any time or can book an appointment at one of the 182 centers in 25 states.
Where is there still space in this framework for general practitioners?
That’s the question. With or without Amazon, we are heading in the direction of digitally supported and team-based cross-sector healthcare centers networked with maximum care providers. Cloud, AI and digital transformation will bring about far-reaching changes. Interprofessional care will be provided not only by general practitioners and specialists under one roof but also by various medical and healthcare assistants, among others. Regional primary care centers will also have observation beds.
Will we see a new type of hospital?
Not at all: About 400 to 600 small hospitals in Germany will become superfluous in their existing shape. Especially in rural areas, centers will often be created by reorganizing small hospitals, in which general practitioners and primary care specialists as well as other professional groups and social workers work together. These regional centers will be closely networked with highly specialized maximum care providers in cities, both digitally and through staff rotation in training and further education. In the future, better staffed maximum care providers, such as university hospitals, will have specialists around the clock who can help quickly and consummately with tumors, complicated operations, stroke or heart attacks. Three or four days after an operation, patients will be transferred to the regional care center, which has observation and nursing beds and is closer to home for their relatives to visit.
In one of your lectures, you said that “Amazon loves inefficient markets”. Is our healthcare system an easy target?
Amazon has clear strategies, and there are others who are interested in the healthcare market too, such as the Otto Group, the second largest digital mail order company in Europe after Amazon, although there is quite a big gap between them. A member of Otto’s executive board told me that the healthcare sector is an interesting market for them. Otto are experts in organization and logistics and, as they were unfamiliar with the market, have already acquired a majority share in the Swiss company Medgate, Europe’s largest telemedicine provider. All platform-based companies, whether Uber, Hotel Reservation Service, FlixBus or Netflix, have the same business model. These companies are matchmakers and can set the rules in their marketplace. And they earn on every transaction.
Is there any other way?
Hardly: These companies have huge data centers. Amazon is now the largest cloud service provider in the world, meaning that most of the data stored centrally somewhere is in Amazon’s hands. They know what we are interested in, what books we read, what movies we watch and soon they will know what diseases we have.
That sounds awful.
But it might be extremely convenient for the patient.
Such corporations wield unbelievable power.
The implications are incredible. But the question is: Is this really going to happen? What speaks for and against it? In any case, Amazon is extremely successful because they are uncompromising when it comes to satisfying their customers’ needs. And people don’t seem to mind revealing their data.
Why are people going along with this?
Imagine a single mother with a feverish child at home. Amazon offers her video consultation around the clock. She can therefore consult a doctor from the comfort of her own couch without having to leave the house with a sick child. And Amazon Diagnostics can offer her rapid tests and on-site tests including samples collected by an Amazon employee in her own home. Amazon Pharmacy delivers the prescribed medication within 60 minutes, possibly via a Prime Air drone. Amazon is already testing all this in the USA and could soon offer it on the German market, even if it is considered difficult. Perhaps Amazon will then set up a company health insurance fund – initially for its own staff and soon for all German citizens, who can then choose an Amazon Prime tariff – with a minimum additional contribution, 24/7 telemedicine and medication delivered within 60 minutes at no extra charge. This would be the most attractive health insurance company in Germany. Regardless of all data privacy concerns.
Could Otto get there first?
It remains to be seen. There is also AI to consider, which facilitates completely new supply chain processes and needs data. Amazon has plenty of data and can get plenty more. Just think about how people willingly install data-harvesting bugs in their homes, such as Alexa or Siri. Then, at the latest, we’re entering a whole new era. AI can do a lot of good, but it also has the potential for abuse.
What else will AI be able to do?
Forward Health from Silicon Valley is an interesting example. They tried to introduce the CarePod, the first unstaffed practice, they also call it AI Doctor’s Office. Patients open the door with their smartphone, a friendly AI avatar records their medical history, followed by AI-guided diagnosis using various sensors. There is no nurse, no medical assistant, no doctor, only a kind of “janitor”, an attendant who replenishes the consumables. In China, such video stations have been around for some time and supply the most important medicines from a vending machine. So this is already a reality. What conclusions can we draw from this for our healthcare system?
Our system is too expensive.
Highly inefficient and much too expensive.
A stupid combination. Should we be afraid of the future? Or should we say that once we have tasted the benefits, the proof is in the pudding?
In view of rapid developments and the fascinating opportunities available to us, I see many potential benefits. In this respect, I feel a deep affinity for progress. Many things can be organized differently and better. For example, AI can be used very efficiently in training or education for patients: After all, doctors always repeat the same advice. I also consider human-in-the-loop approaches, where AI systematically supports the doctor or nurse in diagnosis or therapy, to be a huge opportunity. The problem is that we need AI-enabled high-availability data centers to handle all this in real time and that the development of generative AI models is currently taking place almost exclusively outside Europe. Tech giants are taking advantage of the opportunities offered by their platform economy business models – they are competing with each other, investing unimaginable sums and it is now almost impossible to catch up with them. Amazon is investing an additional $150 billion in its cloud centers and launching more than 3,000 of its own satellites into space. In Europe, we always want to consider the ethical aspects before we do anything. We want to uphold data sovereignty. We want to regulate this and that. That’s all great, but in the end all we will have is the most highly regulated AI. The fastest will be in China. And the best in America. The European Court of Auditors recently found that Europe will lose the AI race once and for all in the foreseeable future.
Can we still change things round? Your approach to the GP model is also aimed at controlling the situation.
For the foreseeable future, we humans will continue to want a human as our point of contact. Someone who knows which of several therapies is the right one in an individual case. Who can draw the strings together through analogical thinking, which AI is so far unable to do. AI can currently only reproduce the data it ingests. But it cannot transfer this data to another problem. We still need a human-machine interface, and here we especially need generalists nearby to maintain an overview, in addition to the specialists. We will still have the problem that people live in rural areas, in underserved communities with different languages. Here, AI can make a very good contribution to better care. Studies show that suitably trained machines can even express themselves more empathetically than we can. But only doctors can be empathetic.
You mentioned undesirable practices due to flat rates per case and quarterly billing. But Amazon also wants to make money: Then we will hear AI saying: “… and another hip, another hip, another hip.”
Much like: “Customers who ordered one hip replacement frequently ordered a second one” or “this seller has many repeat customers”.
Is this a future that frightens you?
I see far more advantages than problems. We have to focus systematically on the problems. One problem could be that individual platform economy companies are very profit-oriented and acquire too much power. We should prevent that. My advice to the German Association of General Practitioners is to set up a platform via which practices can offer their patients certain services, such as making appointments, vaccination monitoring, prevention, and so on. Perhaps at some point compromises will be necessary, simply because Amazon can procure medicines at significantly lower cost due to their buying power. I mention Amazon as an example so often because, according to my analyses, they are currently the furthest ahead and they will soon have the power to take on the healthcare market as an oligopoly or even monopoly supplier.
And at some point, they will be able to set the prices as they wish.
Amazon will not be involved in patient care themselves. Interdependencies will develop: We might see click doctors in call centers, who will be just as dependent on platform employment as Uber drivers. Some of these doctors might be based in Majorca, go swimming in the morning and do a few teleconsultations in the afternoon. But as with Uber drivers, the platform controls the orders and sets the prices, there is no insurance in the event of illness and no unemployment insurance.
This doesn’t match the conventional social status of a doctor.
But many are already doing this. At Medgate in Switzerland or doctor.de in Berlin, for example, there are already a lot of doctors doing video consultations from home.
And are these good doctors?
Medgate allegedly has a very sophisticated quality assurance system. For example, they use algorithms that guide to digitally supported diagnosis and documentation of symptoms: Have you got a headache? Have you had a headache for a long time? Is your headache on one side or both sides? Do you feel sick? What medications are you taking? Have you been abroad?
Isn’t that the classic type of differential diagnosis?
Maybe even better: AI won’t forget an important question. Through digital documentation, it would also be possible to see how many of the doctor’s patients came back again or suddenly had to go to the hospital, which suggests a misdiagnosis. Systems like this can contribute to highly effective quality improvement, and the doctor’s work is more transparent. Some doctors are afraid of that.
What role do GPs play in this futuristic scenario?
They are more likely to be members of a team, a human-machine interface, an empathetic companion who protects the patient from too much or wrong treatment. They will also be responsible for bedside care in cross-sectoral health centers and liaise with specialists. If this is organized intelligently, patient care will be better than it is today, especially in peripheral regions.
What role do GPs play in this futuristic scenario?
They are more likely to be members of a team, a human-machine interface, an empathetic companion who protects the patient from too much or wrong treatment. They will also be responsible for bedside care in cross-sectoral health centers and liaise with specialists. If this is organized intelligently, patient care will be better than it is today, especially in peripheral regions.
Will GPs earn more in the future?
They are already no longer at the bottom of the income scale compared to their specialist colleagues, but somewhere in the middle field. But laboratory doctors or radiologists currently earn far more. For good GP care, we must also focus on remuneration so that we attract the best people. I would not primarily reward the doctor, but the team.
Could you please explain what you mean?
So far, doctors working in outpatient departments are required to treat patients themselves and can hardly delegate tasks to other professional groups. For example, if the medical professional makes a routine home visit, which they might well do, then they hardly earn anything. But if you say that registration and care by the practice are paid for, then the practice itself can decide how to organize that care. In primary care centers in Scandinavia, for example, community nurses, midwives and other professional groups are part of the team, which relieves doctors of many tasks. In the case of the chronically ill, they see the patient independently several times a year, and these services are all reimbursed commensurately. In Germany, you only find this in GP-centered care, which so far is particularly offered in Baden-Württemberg.
Apart from reducing costs, where else does digitalization benefit medical care?
We know of about 6,000 different rare diseases. More than four million people suffer from them, but many are not even aware of it. These diseases are not easily recognizable. With the help of electronic patient records and AI in doctors’ practices, specific constellations that indicate rare diseases could be filtered out in the future and brought to the attending GP’s attention. Another benefit is the possibility to recall medication. The patient doesn’t usually hear about this because batches are not currently registered and due to data protection restrictions. If your car’s brakes are faulty, the manufacturer can recall it at any time. If your medication has been withdrawn from the market because of contamination, this is not currently possible.
Will every single doctor have to acquire digital equipment and expertise in the future?
Ten years ago in Estonia, I was asked: Why do you still print this out? – “Welcome to Germany”: From a digital perspective, we are a developing country. The difficulties we are currently seeing in the implementation of e-prescriptions, for example, have to do with our outdated, decentralized and incompatible structures. More than a hundred different practice management systems have been developed in-house by different companies that have bought varying solutions from all over the place. In some cases, they still run on DOS, if that means anything to you. What we need here is international standards. The first application from the new digital world has just been approved: TI Messenger.
Never heard of it.
Famedly is the first manufacturer to be granted approval for the German healthcare system. TI Messenger enables end-to-end encrypted transmission of data between verified users based on a highly secure protocol called Matrix – similar to Threema or Signal. It is intended first of all for doctors, dentists and pharmacies. Nursing services, health insurance companies and patients will follow. We will all come into contact with it sooner or later. From 2025, every citizen will also have an electronic patient record – unless they decline, which I wouldn’t recommend. First up is the medication plan followed by the vaccination card, an emergency data record with allergies, laboratory findings, doctors’ letters, healthcare proxy, advance decisions and powers of attorney.
According to the forthcoming European Health Data Space, if you end up in hospital while on vacation, the Spanish or French doctor will be able to see that you take an anticoagulant, are allergic to penicillin, and so on. But we are not there yet. Ideally, the software should be managed centrally with standard interfaces, maintained and protected against cyber attacks: Infrastructure and software as a service – for every hospital, every practice, every physiotherapist, who would then no longer have to take care of all this by themselves. And that’s where Amazon comes in again: They are the world’s largest provider of such cloud and AI solutions. But the German providers of practice management systems naturally do not like to hear this.
Will we manage to make our healthcare system fit for the future?
I am fundamentally optimistic. It is good that two laws were passed in the Bundestag in 2023 that finally clear the way for this – the Digital Act and the Health Data Use Act, which ensure health data is not only protected but also used in the best possible way. First and foremost, for individual diagnostics and therapy, and secondly for research that serves the common good, drug and patient safety, quality assurance and control, for example in the event of a pandemic. I see far more opportunities and possibilities here than risks.
Thank you for the interesting conversation.
The interview was conducted by Dr. Anke Sauter and Dr. Markus Bernards, editors of Forschung Frankfurt.
Zur Person / Ferdinand Gerlach, born in 1961, studied medicine in Göttingen, where he also earned his doctoral degree. He later studied public health at Hannover Medical School (MHH). He qualified as a specialist in general practice there in 1992 and completed his postdoctoral degree (Habilitation) in 1998. After two years as Head of the Quality Promotion Department in the Department of General Medicine at MHH, he received a call to Kiel University, where he was director of the Institute for General Practice of Schleswig-Holstein University Hospital, Kiel Campus. He has been director of the Institute for General Practice at Goethe University Frankfurt since 2024. He also worked as a general practitioner in Bremen, Kiel and Frankfurt. For 16 years, Gerlach was a member of the German German Advisory Council on the Assessment of Developments in the Health Care System, which he chaired for eleven years (until 2023). The council advises the Bundestag, the Bundesrat and the Federal Government on a scientific basis.
gerlach@allgemeinmedizin.uni-frankfurt.de
