Foto: Bax Lindhardt

If 350 grams of fish is good for you – is 700 grams twice as good?

Tuesday 25 Feb 20


Morten Poulsen
Head of Research Group, Senior Researcher
National Food Institute
+45 35 88 76 06

About research at DTU

  • DTU Food is leading in the field of risk-benefit assessment of foods and diet types. The work is carried out in the Research Group for Risk-Benefit.

  • Risk-benefit assessments make it possible to quantify, compare, and measure the health effect of individual foods or of groups of foods.

  • The research group carries out risk-benefit assessments for the Danish Veterinary and Food Administration. The assessments are included in the administration’s dietary recommendations for the population.

  • The research group heads the international risk-benefit network and cooperates with the European Food Safety Authority (EFSA). Currently, the group is also helping the Norwegian authorities make a risk-benefit assessment of fish.

Risk-benefit assessments are a new method that balances the positive and negative effects of foods on health. Senior Researcher Morten Poulsen is Head of DTU Food’s Research Group for Risk-Benefit and explains here why we do not get twice the benefits by eating twice the amount of fish a week—and elaborates on what else can be assessed by the method.


Q: If 350 grams of fish is good for you—is 700 grams twice as good?

A: No. Fish contain substances that are both beneficial and harmful to our health. In fatty fish such as salmon, herring, and trout, the beneficial substances are, among other things, fatty acids and vitamin D, while the harmful substances are heavy metals such as mercury. When we make risk-benefit assessments of foods, we compare the health benefits with the adverse health effects. Our calculations show that our health benefits from eating 350 grams of fish a week—of which 200 grams should be fatty fish. But if we eat significantly more than that, we risk ingesting too many of the harmful substances.

Q: How can you compare pros and cons?

A: If we take fatty fish as an example, the beneficial substances will reduce the risk of, e.g., cardiovascular diseases, while the harmful substances can harm brain development, for instance. In order to make an overall assessment, we convert these effects into the number of healthy years of life. In doing so, we get a sort of common currency that allows us to compare the risks and the benefits. We take into account whether the harmful effects are transient or lasting and whether they can cause life-threatening diseases such as cancer as well as early death. We can also take differences between social groups into account, such as children, women of childbearing age, and so on. It takes a lot of data to make a comprehensive assessment. It also requires us to cooperate across disciplines such as nutrition science, epidemiology, chemistry, toxicology, and microbiology.

Q: Can you assess diet types as a whole?

A: Yes. The challenge is that the time it takes to perform a risk-benefit assessment of a type of diet may mean that the diet is no longer fashionable when the assessment is ready. We would like to make an assessment of a vegan diet, as this diet type appears to be spreading, and because the motives for choosing the diet go beyond health to include considerations of foods’ environmental impact as well as of animal welfare. This indicates that a vegan diet is a long-term trend, and it might be interesting to uncover whether it has any adverse health effects.

Q: What could those adverse effects be?

A: Without having researched vegan diets yet I dare say that vegans may be at risk of developing a nutritional deficiency when they opt out of all animal foods. They may find it easy to compensate for this risk through careful dietary planning and by consuming dietary supplements. But when your diet is solely plant-based and you increase your intake of nuts and grain, for example, you could risk exposure to the harmful substances that may occur in these foods, such as carcinogenic toxins from mould fungi and heavy metals. We don’t know, but that is something that can be investigated.

Q: Can you assess the environmental impact of a diet type?

A: Yes, we can expand the risk-benefit assessment of foods to include more than health effects. The mathematical models behind risk-benefit assessments are already used to assess the disease burden in a population caused by, e.g., different chemicals and bacteria, as well as nutritional risk factors such as high sugar consumption. We would like to expand the overall assessments to also take foods’ environmental impact into account. We’ve developed a model that can include more parameters. So far, we’ve included a consumer-economic assessment that provides us with a risk-benefit assessment of fish which includes consumer economics. In this way, we ensure that our results are within the consumers’ price range.

Q: How long is a risk-benefit assessment valid for?

A: All I can say is that it isn’t forever. The content of beneficial and harmful substances in our foods may change. For example, a colleague recently demonstrated that the fatty acid composition in farm-bred salmon has changed and now contains fewer long-chain, polyunsaturated fatty acids. That can shift the balance between beneficial and harmful substances in farm-bred salmon. Every time the content of beneficial and harmful substances in our foods changes significantly, or when new knowledge emerges, an updated risk-benefit assessment should be performed.