"We need the structured recording of patient experiences"

For many years, Prof. Dr. Max Geraedts, M.S., Director of the Institute for Health Services Research and Clinical Epidemiology in the Faculty of Medicine at Philipps University Marburg, has been conducting quality and patient safety-related health services research in the outpatient, inpatient, and nursing care sectors based on primary and secondary data – primarily with support from ministries, the Federal Ministry of Education and Research (BMBF), and with the help of the Innovation Fund. As an experienced researcher, he knows exactly what is possible, on the one hand, and what would be possible, on the other, if the necessary data were available. "Monitor Health Services Research" spoke with Prof. Geraedts, the Congress President of the 22nd German Society for Health Services Research (DKVF), about what the scientific community can expect from October 4-6, 2023.

>> As Congress President, you
dent is the motto of this year’s German Congress for Supply
research significantly
determined that says: “Health-
health-related care for-
Research – the driving force behind evidence-based health policy." There's either a question mark missing, or a concrete classification as to whether this statement is a postulate or a vision?
Our motto is a postulate, but at the same time it is also a task for all health services researchers. The potential of health services research can be seen very closely and currently in the work of the Government Commission for Hospital Reform. Their work includes various studies from the field of health services research, which form the basis for their recommendations. Of course, attempts to establish health policy that is as evidence-based as possible always encounter headwinds, because there are simply too many, often very conflicting, interests that clash. For me, however, the path from eminence-based to evidence-based health policy is clear. This is precisely what we as health services researchers want and can already achieve to some extent, but we will certainly have to do much more in the future.

If sufficient health data were available.
That's right. We know that much of the data, which has often been available for a long time and which could further advance health research, cannot yet be used in the way that would actually be necessary to enable evidence-based health policy. This is not to say, however, that health services research is not already a driver of health policy; it simply wants to be even stronger in the future. This requires the necessary foundation of existing data, coupled with the consistent utilization of new data sources.

Which data streams do you primarily have in mind?
The link between registry and billing data from health insurance companies is one thing. A huge data source that we still largely lack in Germany is a structured recording of patient experiences and patient-reported outcomes, commonly referred to as PRO and PRE. Since the patient perspective is largely lacking in this country, it cannot be incorporated into health services research, and thus into health policy and ultimately into health care. However, if we want to not only describe this as patient-centered, but also consistently align it accordingly, we need a new data access point.

Do we need a kind of structured, patient-centered German sample for the entire health system?
We need a healthcare panel that longitudinally surveys patients' experiences with healthcare. We also need such a panel with meaningful coverage across the entire country and across regions, and—more precisely—all the way down to community-based care. Then healthcare researchers could analyze in detail where things are going particularly well or poorly, or where best practice approaches exist, which could then be scaled up as quickly as possible, given the available evidence and depending on local conditions.

If only there were a description of the current situation before an intervention or change in the law...
… we could track the effects of innovations and the influence of health policy on real care much better over time and – in the event of possible negative developments, for example in the case of mismanagement – warn much earlier.

Here again, patient experience would be helpful.
Even to a particularly high degree. What is the use, for example, of evaluating structural or process innovations solely in terms of mortality rates, when patient-reported experiences and outcomes in most healthcare settings would be much better and faster indicators that could be used to fine-tune or even redirect systems that are threatening to get out of control? This applies not only to the care of sick people, but to virtually the entire life cycle from birth to death. Only with reported patient experiences will we be able to shape healthcare even better.

Are there any experiences from abroad with this?
All Anglo-Saxon countries, as well as the Dutch and Danes, are already much further ahead than we are. Although I'm surprised that the Americans are sometimes content with small sample sizes for their government-funded programs, while the British, as part of their quality management, can rely on very large surveys in the area of patient experience, which they then use to shape their healthcare systems. What ultimately results from this is another matter entirely.

Other countries have citizens' parliaments and co-decision rights, which patients are still denied in our country – the G-BA (Federal Joint Committee on Health Care). Perhaps the system's stakeholders don't even want patients to have a say?
At least the patient representatives in the Federal Joint Committee were given the right to participate, but a change could well take place during this legislative period that would result in a right of co-determination.

Let us not give up hope.
A right of co-determination in the Federal Joint Committee (G-BA) would certainly be helpful, but it is far from sufficient: Patient participation at this level does not replace a structured recording of patient experiences, which unfortunately does not yet exist.

Maybe we should encourage this together.
In the course of further developing quality assurance and quality management in the German healthcare system, this should definitely be part of the list of requirements.

You just mentioned the work of the government commission for modern and needs-based hospital care. Looking at the resulting draft bill and the political discussions about the upcoming law on nationwide hospital planning, the following impression emerges, as so often happens: politicians cherry-pick only those findings from a pool of studies that fit their own agenda or are suitable for a minimal consensus, but are miles away from what could satisfy the researchers who put so much heart and soul into developing these studies.
We live in a democracy where every voice counts. The health services research community can contribute to establishing the most evident basis possible for sound health policy decisions. The fact that these decisions are not implemented one hundred percent and one-to-one in the political process is due to the many other interests that all seek to be represented. Our healthcare system is evolving in an evolutionary rather than a revolutionary way, and perhaps it is sometimes a good thing for innovations to be tested for a while before being widely introduced.

We should illustrate this with an example.
For example, if it is determined that a particular care innovation works well in a region or in a municipal setting, this does not necessarily mean that it will work everywhere.
nized. Furthermore, replicating the results of research studies in other settings is anything but trivial. Therefore, beyond traditional research funding, we need translational funding that would enable us to investigate scalability and the necessary parameters. However, this does not mean that almost all results of health services research can eventually be transferred to real-world health care, because political considerations and social or even local circumstances simply run counter to this.

At a recent press conference in Berlin, 15-year data on primary care in Baden-Württemberg were presented – with impressive results (see MVF 03/23). How much research is needed before a healthcare innovation that has proven to work becomes standard care? Or: When is evident evident enough?
The results presented by colleagues Gerlach and Szecsenyi are impressive and should really convince political decision-makers that it is necessary to
system to this end. Unfortunately, there appears to be resistance, particularly from some actors in self-government, which is leading to the evidence being simply ignored. Because of this, a bold political decision is now needed, since the quality of healthcare provision should not depend on where you live when – considering all local structures – it could be better. But this cannot be blamed on the researchers who present the results of their hard work in numerous scientific papers, at conferences, and press conferences, but are dependent on what self-government and ultimately politicians make of it.

Where is the lobby of health services researchers?
In my opinion, we do not have an established lobby.

Perhaps health services research should be more courageous in pointing out grievances?
I can agree with that, provided it takes place within the framework of proper scientific communication, which leads to the results of health services research being made clearer and, ideally, to activities that, in turn, lead to better care. For example, at our institute, we conducted a study on patient safety in outpatient care funded by the Innovation Fund and published it internationally at the beginning of the pandemic. A positive recommendation from the Federal Joint Committee (G-BA) then encouraged representatives of several banks to take note of the results.

Do they?
They haven't yet. And this despite the fact that in this study (1), in which more than 10,000 German citizens over the age of 40 were surveyed, they reported having experienced at least one patient safety-related incident in the outpatient sector within the last year. We were able to show that, extrapolated annually, approximately two and a half million hospital stays are recorded due to patient safety-related incidents in the outpatient sector. Nevertheless, our findings have not had any political impact that would have led to error management being implemented much more extensively in the outpatient sector than before. Sure, you could blame this on the coronavirus, because our study results were either published in the wrong place or simply at the wrong time, but I can't quite believe that.

Shouldn't you take responsibility for yourself and demand that the German Network for Health Services Research, of which you are a board member, engage in much more scientific communication than before?
This is certainly an important point. We've established a working group on this topic at the DNVF, but we do all this work on a voluntary basis. We need communications professionals to disseminate the many health services research results that my colleagues and I constantly produce in a way that ensures they are received and understood—adequately prepared for politics, society, and, above all, the popular media.
And in the simplest language possible for everyone mentioned?
You're right. Information for patients and all German citizens in general should be provided in plain language. Until now, for example, it was assumed in the US that a language level corresponding to eighth grade was appropriate. Today, researchers in the US estimate that a sixth-grade language level is more effective if the goal is to reach as many people as possible, whether that's health information and patient guidelines or other vital information in an understandable format.

However, this is a task for which scientists are not trained at all.
We can't do that because we don't learn it. I hope that artificial intelligence will help us here. It can already shorten an 800-word abstract to, say, 500 without significantly changing the context. The same will surely soon be possible with plain language.

Now we're about to go to the congress, which you preside over as president. This has been divided into three strands: first, the description of the reality of healthcare provision, second, evaluations, and third, translation. Let's take a closer look at each strand.
We are, of course, aware of the number of submissions received for the various strands. The first strand, which deals with the description and analysis of current healthcare provision, accounts for the majority of submissions, with 290 contributions out of over 600. The second strand, which focuses on the evaluation of healthcare outcomes, and the third, which deals with the implementation of positively evaluated healthcare innovations into routine care, accounted for approximately 60 and 140 contributions, respectively. In addition, there are approximately 120 contributions on the topic of methodological development.

The main question is: Who provides care for whom, when, and with what? And: How and why is care provided?
Correct. Because, in my opinion, the description and analysis of healthcare is an absolute focus of our work and actions.

Although it is all too often said that we already know everything and only have a problem with implementation.

We still have a deficit in both knowledge and implementation, and I believe that the longer I deal with the subject, the more we don't know.

Any details about this?
For example, we don't know which subpopulation was truly helped by vaccinations during the coronavirus pandemic. And which vulnerable groups may not have been adequately identified. And what the long-term consequences of post-COVID will be. There's a huge knowledge gap here.

This was also caused by the fact that the necessary data was not collected at the right time, which should have been collected in order to draw proper conclusions.
What the German Network for Health Services Research has been demanding almost since the beginning of the pandemic. We have repeatedly emphasized how important it would be to conduct accompanying evaluations, and especially what data is necessary for this. We have certainly tried to actively influence policymakers here, but have not been heard. That is why we now have these knowledge deficits. This, of course, applies not only to Corona, but to many other areas of healthcare as well. I have just returned from the US Health Services Research Congress in Seattle, where one of the main topics is the influence of social determinants on healthcare. In Germany, despite the established congress on "Poverty and Health," we are still at the very beginning of considering these determinants in healthcare and researching them, although important topics such as
Accessibility and health literacy
have a lot to do with social factors.

Or simply gender or age.
Exactly. For example, our institute conducted a study (2) in the area of stroke, funded by the Innovation Fund, and found that older people and women in Germany have less access to good care than younger people and men. At the same time, we were able to show that women and older people benefit just as much from appropriate treatment—especially in centers with stroke units. What we don't yet know is why these differences exist in the first place. There's still a lot of research to be done.

To continue with this example: What would need to be done to provide better care for this group, which appears to be underrepresented in stroke units?
Quite simply: equal access for everyone, regardless of age and gender. And if you want to take it even further: regardless of social background and class. Above all, however, we must raise awareness among those involved in care—from relatives to staff in nursing homes, emergency services, and even doctors—that while older people naturally have a lower chance of surviving a stroke, they can still benefit just as much from excellent, guideline-compliant care in specialized facilities.

When Federal Health Minister Lauterbach eventually completes his hospital reform, there will perhaps be 20 to 30 percent fewer hospitals, and hopefully this will affect precisely those that have already delivered unsatisfactory quality.
Complex treatments simply cannot be provided in facilities that are too small to handle enough cases to build the necessary experience, or that simply lack the necessary equipment. But this is common knowledge and has long been clearly established; no new studies are needed to prove it.

In the second strand, the DKVF is primarily concerned with the evaluation of healthcare outcomes.
And this brings us back to the topic of systematically recording patient-reported outcomes and experiences. Here, too, there is a major knowledge gap because, to date, we have assessed healthcare innovations based on their potential for clinically relevant morbidity and mortality. Important parameters such as quality of life, social participation, or even functionality after treatment are ignored. What use is it to know the mortality rate for a knee arthroplasty, which will be – let's say – 2 in 100,000, if we don't record mobility, freedom from pain, and other long-term functional and social factors, as well as post-hospital complication rates and rehospitalizations? Because we don't do this, something could go wrong somewhere in the healthcare chain, but we don't know it – and therefore can't take any countermeasures.

Here, our demand could be that everyone discharged from a hospital be enrolled in a study that would reveal the true outcome after one or two years.
I believe it will be necessary to conduct precisely this type of long-term outcomes research studies, not necessarily on all patients, but at least on a meaningful sample of precisely those in whom negative long-term effects can be assumed. Only in this way will we be able to make profound statements about the quality of care across different care settings and their organizations or forms.

Accordingly, a kind of Germany cohort real care.
A national cohort for experiences and long-term outcomes in healthcare, which thus has a different focus than existing ones, which were primarily driven by indications. Something similar existed from 2001 with the Bertelsmann Foundation's Health Monitor, which unfortunately ended in 2016 but could be reactivated. If we could achieve this in the next few years, I would be very pleased, because it could truly improve healthcare.

Then let’s work on that too!
We should do that. For example, we have established a working group for regional health monitoring within the German Network for Health Services Research—a real-world health care cohort like this would be a good fit.

Now, the second strand concerns the question of which factors are particularly effective or important for achieving the goal. This is something we already addressed in the G-BA's recommendations, which do not produce any visible results.
Health care research must learn from this as well. The nature of such recommendations depends not only on political opinion-forming but also on whether the accompanying process evaluation was sufficiently meaningful. If an evaluation shows that an intervention of any kind was able to achieve certain results, but the driving attributable factors are not clearly identified and described in a generally understandable way, this indicates a shortcoming in the study implementation. I know that this is anything but trivial in complex evaluations, but in my opinion every evaluation of new forms of health care supported by the Innovation Fund should include this. If it were done this way, the subsequent part of transformation or translation into standard care would also be somewhat easier to achieve. Somewhat because other factors, especially contextual factors, also play a role.

And special funding would also be necessary.
Certainly. Anyone who wants to translate must establish sufficient transitional funding to facilitate the transition from the project to mainstream care. Otherwise, there's a risk that the knowledge gained will simply be lost after the project ends.

In a very clever way, we have already arrived at the third strand of the upcoming DKVF, which deals with the implementation of positively evaluated healthcare innovations into routine care.
One of the key questions to be answered here is what barriers stand in the way of implementing healthcare innovations. To address this question, better data availability is necessary, first and foremost.
Now, we often hear the wonderful phrase that all the data is actually available, we just don't know how to use it. If I remember the beginning of our conversation, we should ask: Do we really have the right data?
First, we have a lot of data, some of which is inaccessible to us. Second, if we want to analyze health insurance data comprehensively, for example, this is problematic because some components important for research are missing from the data set, and compatibility is also limited. However, this will hopefully change with the launch of the Research Data Center. For comprehensive healthcare analyses, we need to consolidate many other data sets, including those from registries and—as already mentioned—PROMs and PREMs that are not yet available in a structured format.

A scientific congress would be nothing without a fourth strand, which focuses on the further development of methods in health-related services research.
Correct. However, this also includes dealing with routine data. One of the questions here is how to link different data bodies, including care-related data streams such as those from registries, cohorts, etc. Likewise, we need to further develop methods for health care research even more than we have done so far. Only with the right methods will we be able to measure the right things in order to develop our health care system in a sustainable direction.

Your vision?
That's not even that far away. I'm content with a functioning, truly cross-professional and cross-sectoral electronic patient record. This must enable data access for the various service providers in the various sectors and professions, unless the patient explicitly objects to the use of the data. If what is entered into the electronic records is at least somewhat plausible, I would be very satisfied. This also includes all service providers coding the same way, which we are still miles away from achieving because the quality of documentation varies greatly depending on the profession and, above all, the culture of documentation differs greatly. Only with documentation training and harmonization will the vision of cross-professional and cross-sectoral documentation of medical findings and healthcare as the basis for healthcare research building on it become a reality one day.

Professor Geraedts, thank you for the interview.

The interview was conducted by MVF editor-in-chief Peter Stegmaier.