December 1, 2017, Brussels
As the Dutch Nutrition in Transition (NiT) initiative has put to the fore in an editorial in the European Journal of Nutrition, nutrition sciences are meeting inherent boundaries that hamper conceptual and methodological progress, trust, and the translation of novel insights into societal benefits.
At the 4th conference of the JPI HDHL, an afternoon workshop was organized to discuss with participants their view on the limitations that the nutrition sciences are facing with respect to their capability and credibility. Approximately 20 participants joined the discussion.
Before inviting participants to actively discuss, NiT member dr. Jan de Vries gave a short presentation (that can be found here: [op NiT-website]). He analyzed the problem of nutrition research, explained goal and methods of the workshop, and introduced the three central discussion themes, notably, Concepts, Methods, and Organization.
The workshop was organized according to the World Café format with 3 discussion tables, each concentrating on either Concepts, Methods, or Organization. The session was split into 3 rounds of 20 minutes. After each round participants switched to one of the other tables. This allowed participants to contribute to all three themes and encounter new discussants. In rounds 1 and 2, participants were stimulated by the NiT table hosts to bring in a maximum of diverging views on the table. In round 3, they tried to converge the various perspectives. A subsequent plenary presentation of the output of each table allowed all participants to provide feedback on the content that was summarized.
Content of the three discussion tables
Table 1: Concepts
The discussion resulted in a focus on the concepts of “evidence” and “health”.
To start with evidence, it was seen as a concept that strongly builds on societal values. Therefore it was questioned if evidence should only be based on facts that are reported by the biomedical scientists. It was even questioned if evidence really matters, facing “alternative facts” nowadays. Nutrition science was compared to religion; everybody seems to know what is best, but based on incomplete data and many non-falsified theories. It at least was clear that evidence must be based on consensus amongst those that have the capability to understand reported facts the evidence can or should be based upon. Because the medical model is not adequate enough for nutrition sciences, evidence should come from additional scientific disciplines like public health science, the social; sciences, economics, and more. It was mentioned that nutrition sciences could damage public trust if it did not renew itself.
The concept health is receiving many definitions. Although not yet made measurable, the WHO definition was thought to cover all essential aspects of health. Awareness should be increased that the biomedical aspects are only partly covering health. Social and emotional aspects are probably underestimated in the biomedical sciences, it was said. In that perspective, much more attention has to be paid to public health underpinned with results from social sciences, behavioral sciences, public health sciences and other relevant disciplines. Biomedically, the concept of disease is considered to be measured by a diversity of biomarkers, which was thought to be much easier than measuring the concept of health. The question was raised if nutrition was mainly covering physiological needs and if a healthy physiology is just a consequence. However, a physiologically healthy diet may not be the perfect diet to health from a more holistic point of view.
Additionally, discussants linked the issues to methodological matters of researching new concepts of health and disease. Maybe the definition of health should be based on empirical research, e.g. by asking people on the street what they consider is health for them, and by distinguishing biomedical, social, and emotional aspects of nutrition and health. Combining that with measures of known and accepted (bio)markers may result in a societally acceptable definition of health which would allow generating societal relevant research questions.
Table 2: Methods
Table 2 covered the discussion on the methods that are needed to provide adequate information on the effects of nutrition on health.
Participants first discussed traditional methods (animal research, RCT’s), their strengths (e.g. clarity) and weaknesses (e.g. protocol adherence, lack of reference materials). Mentioned were also the risk of big data and artificial intelligence losing relevance when translated into personalized advice that would be ‘the mean of the mean of the mean’.
The nutrition research domain was said to have three layers, namely, diet patterns, product groups, and substances. It seemed advisable to study these holistically, and combine cohort observation studies with multisite studies and biomedical methods like biomarkers (despite the latter’s sometimes surrogate nature). Because of the public health effect of nutrition, more methods from the social sciences were said to be needed, e.g. qualitative research, economic analyses, and surveys. Awareness should be cultivated of the great differences between populations and related problems of generalizability.
Perspectives converged as follows. First, it was recommended to combine disciplines for one shared research question that should be fed by societal needs (since too many different questions in different places with different methods and different interpretations would reduce the relevance of both applied and fundamental nutrition research). The second and related convergence was about standardization as being necessary in not only biomedical and experimental, but also social sciences research.
Table 3: Organization
At Table 3 we discussed the current and future organization of nutrition research. The discussion was aimed at recognizing the positive developments from the past and current situation and the areas where new structures could contribute to future credible science.
It was recognized by the participants that in recent decades nutrition sciences were strongly influenced by a reductionist approach, with a significant role of industrial research aiming at nutrition and/or health claims (N&H) for their ingredients and/or products. On the one hand side the EFSA requirements for N&H application has blocked innovative research with its focus on the reductionist approach. On the other hand the industrial input generated new knowledge and even new technologies to establish health parameters. In general the EFSA evaluations are of public interest. It is time for a new look at concepts and methodology.
On a short overview of the current organization of nutrition research the participants came to the conclusion that the situation differs per member state. For instance, with a strongly developed food industry in The Netherlands research money is more easily available as in countries like France and Italy.
As a future perspective the discussants came to the conclusion that more research money should be available. This additional money could be spent on new innovative concepts but also and perhaps specifically on research addressing the changing societal needs.
The EU should lead the organizational harmonization. What really matters is that scientists and member states work together. This can be realized at many levels and many topics. Data sharing was one of the examples mentioned. A second one was the notoriously unreliable dietary intake data which need a EU approach. Finally an example brought forward is the need for multidisciplinary research. Scientists have to come out of their silo’s.
Quite some time was spent on the role of the food industry in driving nutrition research. Their role is unmistakable. The panelists differed of opinion. Some participants regard the influence of industry as unacceptable. Others, on the contrary, acknowledge the positive aspects, especially in public private partnerships (PPP). They recognize that PPP may be the way forward, unless independence of research is guaranteed.