10 theses on plans for a National Health Portal

Increasingly, the internet is the place to go to for health information. One in 20 searches on Google is related to health. More than half of German internet users look for information online about illnesses and treatments at least once a year. However, if the common assumption is to be believed, they usually fail to find what they are looking for in the jungle of information and succeed only in becoming needlessly confused. Or they are confronted with factually inaccurate or misleading information. One proposed solution to this problem is the establishment of a National Health Portal. The German Federal Ministry of Health (BMG) has commissioned the Institute for Quality and Efficiency in Health Care (IQWiG) to develop a concept for such a portal by the beginning of 2018. Does this idea hold up to scrutiny? Because we would like to offer both constructive and critical input to the German government’s efforts, and after frequent requests for our own opinion in past months, this blog post gathers together our current thoughts on a National Health Portal –in ten theses.

At the Bertelsmann Stiftung, we, alongside several others, have been occupied for many years with questions regarding proper strategies for communicating health information. This work has been pursued both through our own projects such as the “Weisse Liste” and through other projects. Where is there a need for information? How should information be organized so that it is comprehensible and actionable? How do you strike the right balance between the necessary reduction of complexity and the requisite degree of correctness? And almost more importantly (but nevertheless frequently forgotten), which channels and which communication approaches could enable the information to reach as many people as possible?

Initiatives such as the Allianz für Gesundheitskompetenz (“Alliance for Health Literacy”) and the concept of a National Health Portal have, in principle, a positive effect: they lend strategic value to the subject of health information. And they move the discourse on what action should be taken to a central, national level. Our theses make no claim to absolute validity but reflect the current state of our discussion. They are based on our analyses and experiences [1] and are intended to contribute to the overall discourse. The theses are concerned with the premises for such a portal, the expectations that a portal may or may not fulfill, and the possible forms of implementation.

On the premises and expectations of a national information service

1. Any adequate communication approach must first and foremost deal with the needs of the users. In many cases, and contrary to popular belief, patients using the internet are not primarily searching for evidence-based facts in order to make decisions on health matters. Indeed, the internet can be used to address far more than purely rational needs. Rather, the motives for online research are frequently (emotional) protection or “comfort.”

A separate, as yet unpublished study [2] shows that the patients themselves are not significantly aware of the deficit described above, that is, information that is available, but which remains undiscovered, and are satisfied with the current offer of information on the Web. The online activity of patients is often psychologically motivated. A portal that focuses on the purely factual dissemination of content would be unable to fulfill this need completely. Moreover, motives and expectations differ so greatly that there can be no one-size-fits-all approach to communication. What is needed is quality-oriented diversity.

2. The information available today on the internet is in many (but not all) cases factually accurate and oriented towards current evidence. This applies in particular to the many high-reach portals. The quality of existing information services – also beyond the individual experiences of users (see Thesis 1) – is therefore an insufficient justification for a new centralized approach.

An analysis on behalf of the magazine “Ökotest” has shown that the overall quality of many prominent sites is “sound to very good.” Furthermore, the high-reach commercial media in particular, as well as public service broadcasting, apply high editorial standards and often depend on information that is equally good in terms of content, such as from specific public information providers or other system stakeholders. They frequently also have the greatest experience and competence in the dissemination and communication of information in a manner that is comprehensible and actionable for patients. However, commercial approaches become problematic when the business model calls into question the neutrality of the provided Information.

3. One centralized portal cannot, on its own, ensure that patients are no longer confronted with “bad” or misleading information. This is because the portal will not supersede the previous market logic. In addition, in an attempt to avoid cognitive dissonance, patients will continue to find the information they are searching for. The question therefore arises as to how online users can be protected from disinformation.

The most important entry point when searching for online health information is Google. In their online research, patients frequently search specifically for information that provides them with a sense of security, or which confirms their actions or attitudes. For example, if you are looking to confirm a vaccine-skeptical attitude, you will find the corresponding information – if only because the user’s search history determines the search result, and the user’s input “feeds” the search engine algorithm accordingly. A central portal will not circumvent this mechanism. Rather, the discussion should focus on how to properly handle (deliberately placed) misleading information on the internet, and how to specifically “combat” such information.

4. A centralized German portal will enjoy high levels of trust. However, this alone is not a sufficient justification as, even today, many information services achieve high levels of trust, depending on their design or the prominence of their brand.

An oft-stated hypothesis is that patients are unable to assess which information they can trust on the internet. However, many of the services on the internet, particularly high-reach commercial services, already enjoy a high level of trust among users. Among other approaches, less serious portal providers can employ sophisticated design in order to “impersonate” credibility. A centralized portal will not undermine these effects (on its own), which is another reason why a strategy is required for dealing with disinformation on the internet (see Thesis 3).

5. It is far from the case that all countries with state health portals are good role models with respect to their approach. The most successful are likely to be those who have, to the greatest extent possible, constructed their communication approach along the lines of user needs.

Are references to other countries and their centralized portals a proper justification for a National Health Portal in Germany? In a current analysis, we’re looking at the success of a range of existing national online information services, including that offered by the National Health Service (NHS) in the United Kingdom, as well as services in Denmark and Australia. The aim is to derive insights that will be of use in the debate in Germany [3]. One of our theses is that the countries that are most likely to be successful are above all those that have combined the logic of individual patient behavior with health information that is as personalized as possible (see Thesis 9). Furthermore, some health systems with national portals are centrally organized – patients are accustomed to turning to a dedicated entity. This is not the case in Germany.

6. A National Health Portal should not be overburdened with demands. On its own, such a portal could only make a limited contribution to increasing the health literacy of the population, as a purely portal-based solution would not reach those population groups that are lacking in health literacy.

One at least occasionally articulated hope in relation to the portal solution is that it can serve as a point of information for “previously excluded groups”, namely those with low health literacy. Achieving such a goal will demand integrated approaches with a particular focus on the transfer of personal information – for example by physicians in the context of “shared decision-making,” or by a case manager. If organized as such, the information found online would, at most, be supplementary or consolidatory, e especially if one considers the high proportion (14%) of the population with functional illiteracy.

On the possible solutions for implementation

7. An information service in which “good,” evidence-based information is gathered together and rendered findable can, in principle, represent an improvement on the status quo. In particular those in intermediary roles, such as patient counselling, could find information here – from a single source – that is relevant to their work.

Above all in the context of consulting on treatment decisions, it is often difficult today for patient consultants and physicians to find the information online that conforms with the current state of medical knowledge in each individual case. Furthermore, in some cases, the information services from system stakeholders are, to some extent, contradictory in their recommendations. A restructuring and consolidation could help to improve the status quo. However, this would demand a redefinition of the approach taken by the information service, also as regards target groups.

8. The logic of a classic “portal” is not consistent with the development of the information market and could well be outdated by the time a National Health Portal is completed. Information services are anticipated to become increasingly individualized, and artificial intelligence and algorithmic diagnostic tools are set to play a central role in the dissemination of information. Already in the conceptual phases, proper consideration should be given to these aspects.

A concept for a national information service should anticipate technological developments (including end-user devices) at an early stage, as well as developments in the information market. The approach must be sufficiently “adaptive” from the outset. One such negative role model is the electronic health card, for which the technical and content-related concept has been overtaken by the pace of development.

9. A National Health Portal would have to compete ongoingly with other information services. Assuming a market leader position will require a strong character that binds users together. Alternatively, the communication of health information should take place at those points where users “stay” on a regular basis, and where there is a need for information (“point of interest”). This would include platforms such as Google as well as the “personally controlled health record” (PCHA), which is still under development.

The majority of those who look for health information online use search engines as their point of departure. A “German Federal” address is likely to produce a good search engine ranking according to the criteria currently applied by search engines but will not be enough on its own. The aspect of findability in search engines must play a central role already in the design stages, while negative ramifications (e.g., caused by so-called duplicate content in an aggregate approach) must be eliminated.

Furthermore, Google is planning its own health information service in Germany and has already established similar services in the United States and other countries – partly in cooperation with not-for-profit organizations. This will represent a strong and dominant “competing” public information service. If the objective is to achieve comparable reach, we recommend dialogue and cooperation with Google. This could involve collaboration between all the German-speaking countries.

User loyalty can also be reinforced by bringing together the plans for inter-institutional Electronic Health Records with the planning for the communication of evidence-based information. This would take the form of a platform for treatment management. It is at this point – namely where information on a patient’s state of health flows together and where planning for the treatment process takes place – that the patients’ need for information arises. Here, evidence-based information could be offered in a context-sensitive manner, without the “detour” of search engine-based research on the side of the patient. If this scenario is taken further, the health information or decision-making tool could even be tailored to the patient, for example by taking into account their age, gender or previous illnesses.

10. A national information service should be developed and operated by experts in the communication and dissemination of information. Organizationally, or at least in terms of personnel, the creation of content should be separate from this task.

The operation of a consumer information service will demand highly specific expertise in the face of the rapid pace of digital development, alongside the constant creation of new informational approaches and changing user response habits. Moreover, sufficient resources for propagation must be available. In the context of public quality reporting, a model for the distribution of tasks is observable in other countries, whereby in the future, public institutions will “only” concentrate on the production of content in the sense of “raw texts.” The produced information is then made available for additional processing and mediation by third parties – for example via cross-institutional electronic health records (see Thesis 9). Alternatively, instead of “competing” with existing quality media, these could be actively and specifically promoted. Moreover, in this context, the question arises as to how far the status quo, in which numerous different entities use state resources or insurer funds to produce information on the same indications and issues, is justified.

Download (in German): 10 theses on plans for a National Health Portal (PDF, 0,37 MB)