Talk:Goherr: Project

GOHERR website

Permalink to discussion about the website.

GOHERR recommendations discussion

New argumentation

Reduce dioxin emissions rather than manipulate fish stocks Premises:

1. 6 Dioxin is a health hazard and thus its concentrations in fish has to be considered.
2. 7 Dioxin comes to the environment mostly from industrial and burning processes, which can be managed.
3. 8 Airborne dioxin deposits on the ground and water ecosystems and accumulates especially in marine food chains. This results in high concentrations in predatory fatty fish such as salmon and herring.
4. 9 Salmon get dioxins from herring in the predator-prey interaction
5. 10 Growth / size of fish is a definitive factor for bioaccumulation of dioxins
6. 11 Thus it can be hypothesised that by manipulating the growth of herring, dioxin concentrations in both species could be reduced.

Goherr results:

1. 12 Dioxins in fish cannot be effectively reduced by manipulating fish stocks.

Involve fisheries sector in management of dioxin problem Premises:

1. 13 Fisheries sector is affected by dioxin policies, because they restrict marketing and consumption of especially Baltic herring. They also restrict developing “high value” markets for Baltic salmon.
2. 14 The dioxin problem is currently only governed by restricting emissions and by restricting the selling of fish that contain dioxins.
3. 15 The fisheries sector, although a major stakeholder, is not involved in governing the dioxin problem of the fisheries
4. 16 The selling restrictions, exemptions from them, and the related eating recommendations differ in different countries, which confuses consumers of whether or not Baltic fish is safe to eat, and weakens the image of the fish (and potentially the whole Baltic Sea)

Goherr results:

1. 17 Involving fisheries sector in governing the dioxin problem could lead to identifying new management strategies (and harmonizing them between countries).

Increase EU fisheries contribution to food security Premises: EU Common Fisheries Policy (CFP) aims to increase the contribution of fisheries to food security. CFP identifies aquaculture as the main strategy to increase the contribution of seafood to food security and to reduce food import from non-EU countries. CFP ignores the contribution potential of small pelagic fish, such as Baltic herring, which are abundant. Baltic herring stocks are mostly in good condition. Expected yields are fairly large. The majority of Baltic herring catches are used as feed in fur and fish farming. Using small pelagic fish for human consumption rather than for industrial purposes is recommended by FAO Goherr results:

1. 19 There is a shared interest among fisheries stakeholders to use Baltic herring primarily as food.

Several pathways including actor specific actions to increase the use of Baltic herring as safe-to-eat food by 2040 were created in the first international BONUS GOHERR workshop. Include socio-cultural values explicitly in fisheries governance Premises: The way fisheries are managed (implicitly) reflects the human values associated with fish and fisheries.

1. 20 Fisheries management (as other sectors of management) should not be in conflict with important societal values.

There is a need to develop theoretical approaches to the analysis of socio-cultural values related to fisheries.

Goherr results:

1. 21 There are a lot of important societal and cultural values that relate to Baltic herring and salmon. Similarities and differences in these values between Baltic Sea countries and stakeholder groups were identified and analysed by in-depth interviews.
2. 22 Many important values are currently underrepresented in fisheries governance. The situation threatens the existence of some parts of cultural practice and the legitimacy and effectiveness of management.

Deliberative approach and theoretical framework to the analysis of values related to fish and fisheries were introduced and evaluated. Size-specific promotion of Baltic herring consumption Premises:

1. 23 Dioxins accumulate in individual fish when it gets older and start eating prey on higher trophic levels. Therefore, larger fish tends to have more dioxin than a smaller one of the same species.
2. 59 NB! This is valid only within the same area. The concentrations of a smaller fish from northern parts may be higher than a bit larger from the south. Well, anyhow it does not change the recommendation. / Annukka
3. 24 Baltic herring and salmon are species that have relative high dioxin concentrations compared with other Baltic species.

Goherr results:

1. 25 With the current dioxin concentrations and fish consumption patterns, people in Baltic Sea countries (DK, EST, FI, SWE) may exceed the EFSA tolerable weekly intake (TWI). 5-25 % exceed the current recommendation 14 pg/kg/week in different subpopulations. In contrast, if the new suggested TWI of 2 pg/kg/week is used, 30-70 % exceed the limit in different subgroups. This is most of the people that eat salmon and herring at all.
2. 26 However, despite these exceedances, there are net health benefits from eating these fish species in all subpopulations, except in young fertile women where the case is more complex. See argument #X.

Possible policy recommendations about promotion of a particular size fraction of Baltic herring. The promotion takes primarily place in food industry but also as consumer recommendation. (Exclusive and mutually exhaustive options [assuming that 17 cm is the proper cut point], i.e. one of the three logical alternatives must be chosen):

1. 27 Don’t promote the use of any specific size fraction of Baltic herring.
2. 29 DEFEND.
3. 28 Promote human consumption of small-bodied (<17 cm) Baltic herring, due to its positive net health effects and lower dioxin concentration than those of large herring.
4. 29 ATTACK. I can’t see how we can have this general recommendation.

First, our results show that the size-limit is not a fixed thing - smaller herring can contain higher dioxin content than those used once upon a time for setting the 17cm limit, depending on the growth pattern. And, as Annukka pointed out, the dioxin content (and therefore important limits) vary by area. That’s why they did their spatial study, I guess. Finally, I don’t see from the GOHERR studies why we should promote eating BALTIC fish - from a health perspective, the benefit is in the omega3 and not unique to fish from the Baltic. Food security argument has never been part of the project. / Anna

1. 55 Small herring promotion can be done in two different ways: making recommendations to the public, or by targeting food industry and facilitating production and marketing of products made of small herring. We should prioritize the latter to avoid confusing customers.
2. 56 New products made of “healthy herring” would create new markets, maybe even to other parts of Europe. The larger herring could be used for feed. In the feed industry the dioxins can be removed.
3. 52 Food security argument encourages us to eat Baltic herring, so you have to balance between two harms anyway. Promoting small herring minimizes health risks while increasing food security. We didn't quantify the value of food security so this is done only qualitatively. / Jouni
4. 57 Promoting small fish has other reasons, too. Different foods can be made from small fish, and that potential is largely unused (except maybe in Estonia). / Jouni
5. 58. The policy analysis model of GOHERR (WP6) demonstrated that all the consumer groups benefit from eating smaller herring (also salmon) as the dioxin concentrations are lower in smaller fish.
6. 69 There is no zero-risk situation, but why wouldn’t we still aim for minimizing the harm, while maximizing the utilities? This can be done by promoting the use of smaller fish.
7. 30 Promote human consumption of large(>17 cm) Baltic herring.
8. 31 This has not been suggested because it would worsen dioxin exposure, but it is here as a logical alternative.

Restrictions about Baltic herring in non-sensitive subgroups Premises:

1. 24 Baltic herring contains dioxins.
2. 26 There are net health benefits in eating Baltic herring.

This recommendation applies to all other subgroups except children and young women that plan to get pregnant and have children. This recommendation comes in addition to whatever has been decided about size-specific promotion of Baltic herring. Possible options:

1. 33 Baltic herring should be eaten at most 3-4 times per year.
2. 34 The new suggested EFSA tolerable weekly intake of 2 pg/kg/week is exceeded with larger consumption.
3. 35 Baltic herring should be eaten at most 1-2 times per month.
4. 36 This is close to the current recommendations for fertile women.
5. 37 Baltic herring should be eaten according to the current recommendations, which vary from country to country.
6. 38 Variation reflects different valuations in different countries.
7. 39 Goherr results do not bring any dramatic change to our understanding, so we should keep on the status quo.
8. 40 Baltic herring can be eaten freely.
9. 41 If you want to have a recommendation on removing the ban for men > 45yr where you saw the health benefits, and for women older than fertile age, that’s fine with me. / Anna

Restrictions about Baltic herring in the sensitive subgroup Premises:

1. 24 Baltic herring contains dioxins.
2. 26 There are net health benefits in eating Baltic herring.
3. 79 Dioxin has negative health impacts on developing fetus and newborn, the most sensitive individual.

Goherr results:

1. 46 BONUS GOHERR health benefit-risk assessment (WP5) showed that Baltic fish improves public health and even in the target group (fertile women planning to have children), the benefits and risks are close to each other.
2. 47 BONUS GOHERR survey (WP5) showed that recommendations to limit Baltic herring and salmon use are ineffective on average and even counterproductive with many people.

This recommendation applies only to the sensitive subpopulation, i.e. children and young women that plan to get pregnant and have children. This recommendation comes in addition to whatever has been decided about size-specific promotion of Baltic herring. Possible options:

1. 42 Sensitive group should eat large (>17 cm) Baltic herring at most 3-4 times per year.
2. 43 This is implied by the suggested EFSA TWI from August 2018.
3. 44 Sensitive group should eat large Baltic herring at most 1-2 times per month.
4. 45 This is close to the current recommendations.
5. 46 ATTACK Health risks and benefits are close to each other in the sensitive subgroup.
6. 80 COMMENT. Note that the risks go to the child, while some of the benefits go to the mother. This may be relevant when individual net health benefits are considered.
7. 47 ATTACK Restrictive recommendations are ineffective.
8. 48 Stop health-based dioxin restrictions.
9. 49 ATTACK. Dioxin is a health concern, so we cannot just stop restrictions.
10. 50 I cannot support this very general recommendation as you in your analyses do find health risks to children and fetuses, incl serious ones like IQ. The health benefits comes not from Baltic fish, but from any source of omega3, so there is no need to promote eating dioxin-containing fish if there is any risk at all to humans. /Anna
11. 51 ATTACK. All fish contains some dioxin and some methyl mercury, so the zero-risk approach does not hold. It is true that those risks are smaller with some other fish species so it is possible to prefer those.
12. 52 I think this statement is too controversial for SLU. /Andreas
13. 53 I think the thing to keep is that we should promote is better use of small-bodied fish for human consumption, and incorporate that in the first suggested recommendation above (i.e. to keep current recommendations) / Magnus
14. arg7993.1 Dioxins have some well known harmful health effects (the tooth developmental thing) and in addition some potential, not so well known or defined. In addition there still are fish in the Baltic Sea that exceed the safe limits. For this reason I wouldn't like to say that "Hey, you can safely eat as much as you want, go ahead!" / Annukka
15. arg7998.2 The results of the integrated model showed that promoting fish eating is beneficial to elderly people, but produced harm to the young female. This result is based on the change in DALYs as the measure of utility / harm. Thus although promoting fish eating would be slightly beneficial on population level, we should not give the recommendation 1. To my mind each young female should have the right to know the existing risk, even though it is "only" something related to the teeth of their children. And wasn't one of the fish eating query results that people would like to get more transparent information? / Annukka Lehikoinen
16. 54 If we believe that dioxin risk is below concern, why would we need extra measures? This would give conflicting messages and deteriorate the main message that people can stop worrying about dioxin in their food.
17. 60 Dioxin risk has decreased to level below concern. Give up the restrictions to sell Baltic herring and salmon in EU, because they are ineffective in protecting public health. Instead, they deteriorate the viability, sustainability, and brand of Baltic fishing. / Jouni
18. 61 ATTACK. This recommendation is unacceptable.
19. 62 I don't think this recommendation follows from GOHERR's results. I think it is a better solution that THL recommends this. /Andreas
20. 63 I cannot support this statement at all. Could something closer to what was in WP5 in the third periodic report be a workable compromise (where you point out for which groups the ban could go away without any harm)? / Anna
21. 64 I agree with andreas suggestion that it is better that thl themselves go forward with this recommendation if they want but that we do not include it here (i.e. we cannot do that as we do not agree on this specific suggestion) / Magnus
22. 67. We must remember that it is not only about individual choices. Rather, the large retail companies make significant decisions on behalf of consumers. For example, if Prisma in Finland decides to remove a product from their selection, one million Finns will stop using that product. Therefore, the health-based recommendations should be clear and also guide decisions of retail companies.
23. 68 We can recommend this because there is no zero-risk situation. If people eat less fish, some people will actually die. So, I guess we agree that in general, we should recommend more fish consumption, even Baltic species that contain some pollutants.
24. 70 For more than 20 years in Finland, there has been discussion about dangerous dioxins in Baltic herring. At the same time, the Baltic herring consumption has decreased by 90 %. Of course this is not only due to the dioxin warnings, but it has probably had a significant role. So, we have lost a lot of people due to cardiovascular diseases that could have been avoided by more fish eating. We have gone too far with our restrictive recommendations. / Jouni
25. 71 ATTACK. You say “dioxin warnings, but it has probably had a significant role” - do we have any results in any GOHERR analyses that shows that this is the case? If not, we cannot recommend abandoning dioxin ban, because we do not know how this will affect consumption. / Anna
26. 72 ATTACK We don’t have data about historical impacts of recommendations. But we do have data about what people say they would do if recommendations are changed. And that is one part of the model and thus recommendations. /Jouni
27. 73 The tooth and IQ problems in children are mild, so we are not putting any identifiable people in great risk. This is in compliance with the "first, do no harm" principle by Hippocrates. Also, the total disease burden in children is not large and it is not that different from the benefits to their families, so there is a fairly good balance also on the subpopulation level. /Jouni
28. 74 One additional premise that is not in the model is that if we hope that adults eat fish in the future, they need to learn that when they are children. Therefore, we should not scare young families with horrors of fish eating, because although the risks are not zero, they are not large. / Jouni
29. 75 Recommendation related to fertile women (or even more specifically: fertile women that are planning to have children at some point) is fine. But due to the reasons described above, I think that that is only the second-best recommendation we can make. The best recommendation is that we could just stop worrying about the minor dioxin risk and focus on promoting the consumption of fish, which is healthy for public health, and in the case of Baltic herring also sustainable (unlike many ocean species). Dioxins were the problem of 1980's, but now they are largely in control. In 2010's, we should focus on sustainability, climate change, and food security. Those are the real threats today and the near future. / Jouni
30. 76 This is a good point about what we should focus on. / Päivi

Recommendation about dioxin management Premises:

1. 87 Dioxin is a man-made pollutant produced unintentionally in many incineration and industrial processes. Emissions are typically to air, from where they deposit to soil and water ecosystems.
2. 88 Dioxin accumulates in the food chain, especially in aquatic systems where there are several trophic levels.

Goherr results:

1. 78 BONUS GOHERR (WP4) showed that manipulating fish stock sizes and catching strategies is ineffective / SLU
2. 86 The integrated policy analysis of BONUS GOHERR (WP6, integrating the outputs from the other WPs) showed that reducing the dioxin load is the only effective way to (further) decrease the concentrations in herring and salmon (in comparison with fisheries and nutrient load management). / UOUL

Possible management strategies (several strategies may be applied together)

1. 77 Manage the dioxin problem at the fisheries level.
2. 78 ATTACK Fish stock manipulations are ineffective.
3. 80 ATTACK Ecological sustainability should be the first criteria for fisheries management. This is important both from the perspective of ecosystem’s stability and the societal food security.
4. 81 This is a good point and should be used as a justification. / Päivi.
5. 89 Manage the dioxin problem by reducing dioxin emissions.
6. 82 I don’t see why we need this splitting up. I’ve rephrased a recommendation on which to prioritize based on our modelling results, at the beginning of this document. / Anna
7. 83 ATTACK. I split it up due to reasons relating to the argumentation process. Some recommendations have several parts and we may agree on one but not another part. The whole hierarchy exercise is for being specific about what we talk about. When a resolution is found, we can rephrase the text as needed. / Jouni
8. 84 Many dioxin sources are still point sources and can be effectively managed. This approach reduces problems everywhere, not only in Baltic fish. / Jouni
9. 85 Dioxins come mostly from burning processes, waste incineration, and metal smelting. Cleaning these processes also helps tackling greenhouse gas and other emissions, not only dioxins. / Jouni
10. 90 Manage the dioxin problem at the food and feed sector. Implement monitoring programs and concentration limits in food and feed.
11. 91 If restrictions based on dioxin concentrations are in place (e.g. restrictive recommendations on fish consumption), monitoring of those products is needed.
12. 92 It is the current practice in the EU to monitor dioxin concentrations in food and feed. This should be continued.

Impact of uncertainties on health recommendations Premises: Uncertainties about how things are typically complicate decision making, as the impacts of a particular action are not known with certainty. Goherr results: The value of information analysis on health risk-benefit assessment (WP5) showed that the remaining scientific uncertainties are often large. However, the value of reducing that uncertainty is fairly low (ca 100 DALY/year or less), because it is unlikely that conclusions would change because of these uncertainties. Possible recommendations: Emphasise large uncertainties in the scientific basis of advice.

1. 65 Could this be formulated so that the recommendation includes: 1) the results of THL, 2) reference to food security which these results support BUT ALSO, 3) transparency of the limitations / uncertainty of the risk-benefit assessment. Surely there is uncertainty in the assessment, which is a basis for the recommendation? / Päivi
2. 66 ATTACK. There is no need to talk about uncertainties, because the conclusion is driven by value judgements, not uncertainties. In other words, the remaining uncertainties (which some are large) do not actually change our conclusion, i.e. the value of reducing these uncertainties is low. However, a value question is critical: do we want to maximise net health benefit, or do we want to avoid risks of dioxins, even if it is not the best solution for total health. / Jouni

Acknowledge uncertainties but make clear that the recommendations are robust, given the valuations used (such as emphasising net health benefit over exceedances of tolerable weekly intakes). The differences in valuations actually have a large impact on conclusions, as demonstrated by this discussion among Goherr experts. The arguments were mostly about what aspects should be emphasised (e.g. avoid even small risks if you can) rather than disagreements (i.e., uncertainties) about the risk estimates themselves. / Jounii Discussion about how recommendations should be derived Possible strategies to derive a recommendation (partly exclusive, i.e. some combinations are impossible):

1. 93 Goherr recommendations should be based on consensus among the whole project.
2. 94 Goherr recommendations should be based on majority vote.

---

arg1642: . To try if we can end up to a democratic decision: Jouni, will you do a doodle poll? Perhaps include the two alternatives you suggested, and the third that there already is (> 45 years) ?? And whatever the final recommendation will be, would it good to include some background information in the recommendation text i.e. what the recommendation is based on? (value judgment based on...(food security?)....and a sentence about the risks vs benefits assessment?) Päivi Haapasaari

⇤--arg2220: ATTACK. I’m sorry but this is not a majority vote. We need consensus. Else you can’t have it in the report. Anna Gårdmark ⇤--arg8358: ATTACK. I agree. We need consensus. But in my experience, consensus is based on discussion and compromise. So far we seem to have had the discussions, but not much compromise. Without compromise, I don’t see how we can come to a consensus and we may need to follow Paivi’s suggestion of a vote? This is an issue that is so central to the project, I don’t see how we can NOT have a recommendation for it in our final report. So simply leaving it out due to lack of agreement doesn’t seem like a tenable solution.Alyne Delaney

1. 95 Goherr recommendations should be based on first excluding everything that is not in line with the project premises (which need to be explicated) and Goherr results. What remains is then a basis for discussions about recommendations.
2. 96 Individual institutes may give separate recommendations if other Goherr institutes disagree about conclusions.
3. 97 ATTACK. These are Goherr recommendations and I don't like the idea of having disclaimers. If it gets rejected, we leave it out. However, I haven't seen any specific arguments against it. Do you disagree with some of the premises, or the reasoning? I argue here that this recommendation DOES arise from Goherr/wp5 results, backed up by the explicit values of improving public health and food security. With "any specific argument" I meant any specific, undisputed argument. We seem to all agree that eating small herring should be promoted. But that is not an argument against getting rid of ban on large herring. The argument that we could eat other fish rather than large herring has a premise that the difference between a very small risk and zero risk is very important. I think that difference is very small and thus not important. / Jouni
4. 98 Goherr recommendations should be based on scientific results only, and not include value judgements.
5. 99 ATTACK It is not possible to recommend something without some underlying value judgements. But several of those have been explicated already in the Goherr Description of Work: maximizing net health benefit, acknowledging cultural values of fishing, increasing sustainability in the Baltic Sea. So, Goherr recommendations should be based on those values.

---

arg7993: . I tend to think so that our responsibility as researchers is to report our findings honestly and objectively, not to support any political agendas. Value judgements are always related to decision making, but that is something the decision makers should do - not us! We can help the DMs to make these judgements, that's all. / Annukka

1. arg7993.3 We know the policy makers would like to get simplistic recommendations and one number -type of answers. Anyhow, as the things really are not that simple in reality, we need to make them realize that and provide as simple recommendations of the complex system as we can - but not any simpler (Occam's razor principle). / Annukka

---

arg3935: . I have to make the PDFs and submit before 3 pm today. So there is not much time anymore for this. My suggestion is: Why couldn't we do as Annukka suggests? Give the results and let decision makers draw conclusions? / Päivi Haapasaari

Recommendation about health

How to read discussions

Fact discussion: Dioxin is a concern (disc9167)
Opening statement: Dioxin is still a concern. Keep the current recommendations (consumption limits for young women), monitoring programs, and concentration limits in food. Closing statement: Resolution not yet found. (A closing statement, when resolved, should be updated to the main page.)
Argumentation: arg5270: . BONUS GOHERR health benefit-risk assessment (WP5) showed that Baltic fish improves public health and even in the target group (fertile women planning to have children), the benefits and risks are close to each other. (type: truth; paradigms: science: relevant attack) arg5271: . BONUS GOHERR survey (WP5) showed that recommendations to limit Baltic herring and salmon use are ineffective on average and even counterproductive with many people. (type: truth; paradigms: science: relevant attack)

How to read discussions

Fact discussion: Dioxin is a minor concern (disc9168)
Opening statement: Dioxin is a minor concern. Stop health-based dioxin restrictions, but promote eating small fish with less dioxin, just to be on the safe side. Closing statement: Resolution not yet found. (A closing statement, when resolved, should be updated to the main page.)
Argumentation: arg5272: . I think the statement is too controversial for SLU. Andreas Bryhn (type: relevance; paradigms: science: relevant attack) arg5273: . I cannot support this very general recommendation as you in your analyses do find health risks to children and fetuses, incl serious ones like IQ. The health benefits comes not from Baltic fish, but from any source of omega3, so there is no need to promote eating dioxin-containing fish if there is any risk at all to humans. If you want to have a recommendation on removing the ban for men > 45yr where you saw the health benefits, and for women older than fertile age, that’s fine with me. Anna Gårdmark (type: ; paradigms: science: relevant attack) arg5274: . 1 All fish contains some dioxin and some methyl mercury, so the zero-risk approach does not hold. It is true that those risks are smaller with some other fish species so it is possible to prefer those. 2 food security argument encourages us to eat Baltic herring, so you have to balance between two harms anyway. We didn't quantify the value of food security so this is done only qualitatively. Jouni Tuomisto (type: ; paradigms: science: relevant attack) arg0903: . I think the thing to keep is that we should promote is better use of small-bodied fish for human consumption, and incorporate that in the first suggested recommendation above (i.e. to keep current recommendations) Magnus Huss (type: ; paradigms: science: relevant comment) arg5275: . If we believe that dioxin risk is below concern, why do we need extra measures? This gives conflicting messages and deteriorates the main message that people can stop worrying about dioxin in their food. (type: ; paradigms: science: relevant attack) arg5276: . Small herring promotion can be done in two different ways: making recommendations to the public, or by targeting food industry and facilitating production and marketing of products made of small herring. If we do the latter, argument #5275 becomes invalid. (type: ; paradigms: science: relevant attack) arg9017: . New products made of “healthy herring” would create new markets, maybe even to other parts of Europe. The larger herring could be used for feed. In the feed industry the dioxins can be removed (type: ; paradigms: science: relevant defense) arg9018: . Promoting small fish has other reasons, too. Different foods can be made from small fish, and that potential is largely unused (except maybe in Estonia). (type: ; paradigms: science: relevant defense) arg9019: . The policy analysis model of GOHERR (WP6) demonstrated that all the consumer groups benefit from eating smaller herring (also salmon) as the dioxin concentrations are lower in smaller fish. (type: ; paradigms: science: relevant defense) arg5284: . NB! This is valid only within the same area. The concentrations of a smaller fish from northern parts may be higher than a bit larger from the south. Well... anyhow it does not change the recommendation. Annukka Lehikoinen (type: ; paradigms: science: relevant attack)

How to read discussions

Fact discussion: Dioxin is not a concern (disc9169)
Opening statement: Dioxin risk has decreased to level below concern. Stop health-based dioxin restrictions. Closing statement: Resolution not yet found. (A closing statement, when resolved, should be updated to the main page.)
Argumentation: arg5272: . I think the statement is too controversial for SLU. Andreas Bryhn (type: relevance; paradigms: science: relevant attack) arg5277: . Again: I suggest we do not give this recommendation, but WP5 can tell about the results. Päivi Haapasaari (type: ; paradigms: science: relevant attack) arg5270: . BONUS GOHERR health benefit-risk assessment (WP5) showed that Baltic fish improves public health and even in the target group (fertile women planning to have children), the benefits and risks are close to each other. (type: truth; paradigms: science: relevant defense) arg9020: . We must remember that it is not only about individual choices. Rather, the large retail companies make significant decisions on behalf of consumers. For example, if Prisma in Finland decides to remove a product from their selection, one million Finns will stop using that product. Therefore, the health-based recommendations should be clear and also guide decisions of retail companies. (type: ; paradigms: science: relevant defense) arg9021: . We can recommend this because there is no zero-risk situation. If people eat less fish, some people will actually die. So, I guess we agree that in general, we should recommend more fish consumption, even Baltic species that contain some pollutants. (type: ; paradigms: science: relevant defense) arg5278: . Yes, there is no zero-risk situation, but why wouldn’t we still aim for minimizing the harm, while maximizing the utilities? This can be done by promoting the use of smaller fish. (type: ; paradigms: science: relevant attack) arg9022: . For more than 20 years in Finland, there has been discussion about dangerous dioxins in Baltic herring. At the same time, the Baltic herring consumption has decreased by 90 %. Of course this is not only due to the dioxin warnings, but it has probably had a significant role. So, we have lost a lot of people due to cardiovascular diseases that could have been avoided by more fish eating. We have gone too far with our restrictive recommendations. (type: ; paradigms: science: relevant defense) arg5279: . You say “dioxin warnings, but it has probably had a significant role” - do we have any results in any GOHERR analyses that shows that this is the case? If not, we cannot recommend abandoning dioxin ban, because we do not know how this will affect consumption. Anna Gårdmark (type: ; paradigms: science: relevant attack) arg5280: . We don’t have data about historical impacts of recommendations. But we do have data about what people say they would do if recommendations are changed. And that is one part of the model and thus recommendations. (type: ; paradigms: science: relevant attack) arg9023: . The tooth and IQ problems in children are mild, so we are not putting any identifiable people in great risk. This is in compliance with the "first, do no harm" principle by Hippocrates. Also, the total disease burden in children is not large and it is not that different from the benefits to their families, so there is a fairly good balance also on the subpopulation level. (One additional premise that is not in the model is that if we hope that adults eat fish in the future, they need to learn that when they are children. Therefore, we should not scare young families with horrors of fish eating, because although the risks are not zero, they are not large.) (type: ; paradigms: science: relevant defense) arg9024: . Recommendation related to fertile women (or even more specifically: fertile women that are planning to get children at some point) is fine. But due to the reasons described above, I think that that is only the second-best recommendation we can make. The best recommendation is that we could just stop worrying about the minor dioxin risk and focus on promoting the consumption of fish, which is healthy for public health, and in the case of Baltic herring also sustainable (unlike many ocean species). Dioxins were the problem of 1980's, but now they are largely in control. In 2010's, we should focus on sustainability, climate change, and food security. Those are the real threats today and the near future (type: ; paradigms: science: relevant defense) arg9025: . This is a good point Päivi Haapasaari (type: ; paradigms: science: relevant defense)

Recommendation about dioxin management

Possible management strategies (several strategies may be applied together)

How to read discussions

Fact discussion: Fishery management (disc9170)
Opening statement: Manage the dioxin problem at the fisheries level. Closing statement: Resolution not yet found. (A closing statement, when resolved, should be updated to the main page.)
Argumentation: arg5281: . BONUS GOHERR (WP4) showed that manipulating fish stock sizes and catching strategies is ineffective (type: ; paradigms: science: relevant attack) arg5282: . Ecological sustainability should be the first criteria for fisheries management. This is important both from the perspective of ecosystem’s stability and the societal food security. (type: ; paradigms: science: relevant attack) arg9026: . This is a good point ( to be used as a justification...) Päivi Haapasaari (type: ; paradigms: science: relevant defense)

How to read discussions

Fact discussion: Emission management (disc9171)
Opening statement: Manage the dioxin problem by reducing dioxin emissions. Closing statement: Resolution not yet found. (A closing statement, when resolved, should be updated to the main page.)
Argumentation: arg5283: . I don’t see why we need this splitting up. I’ve rephrased a recommendation on which to prioritize based on our modelling results, at the beginning of this document. Anna Gårdmark (type: ; paradigms: science: relevant attack) arg0904: . I split it up due to reasons relating to the argumentation process. Some recommendations have several parts and we may agree on one but not another part. The whole hierachy exercise is for being specific about what we talk about. When a resolution is found, we can rephrase the text as needed. Jouni Tuomisto (type: ; paradigms: science: relevant comment) arg9027: . Many dioxin sources are still point sources and can be effectively managed. This approach reduces problems everywhere, not only in Baltic fish. (type: ; paradigms: science: relevant defense) arg9028: . Dioxins come mostly from burning processes, waste incineration, and metal smelting. Cleaning these processes also helps tackling greenhouse gas and other emissions, not only dioxins. (type: ; paradigms: science: relevant defense) arg9029: . The integrated policy analysis of BONUS GOHERR (WP6, integrating the outputs from the other WPs) showed that reducing the dioxin load is the only effective way to (further) decrease the concentrations in herring and salmon (in comparison with fisheries and nutrient load management). (type: ; paradigms: science: relevant defense)

How to read discussions

Fact discussion: Food management (disc9168)
Opening statement: Manage the dioxin problem at the food and feed sector. See “Recommendation about health”. Closing statement: Resolution not yet found. (A closing statement, when resolved, should be updated to the main page.)
Argumentation:

Email thread

arg3792: . Here comes the next version of the final report. It was very hard for me to conclude what to write about the wp5 policy recommendation (based on the Google docs discussion)! . Texts relating to that are highlighted yellow, for you to consider .

What do you say? Please read also the Further research and exploitation of the results Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg3792: . The final report looks fine, except for this problematic recommendation issue. I'll try to clarify how I see the recommendation about WP5. There are two possible types of recommendation:

 1.  Get rid of dioxin-based food restrictions related to Baltic herring. This means that dioxin concentrations are NOT used to remove fish catch from the market, and dioxin is NOT used as the basis for any fish consumption recommendations. Dioxins are monitored for research rather and governance purposes.
 2.  Promote Baltic herring consumption in a targeted way. This includes recommendations to eat more small herring and recommendations for particular subpopulations (everyone except young women planning to get children).

Recommendation 2 is only meaningful if recommendation 1 is NOT used. Because if you recommend type 1, that also means that there should be no restrictions that are based on dioxin concentrations, dioxin health risks etc. The logic is the "sure-thing principle": if you recommend more herring consumption even when you learn that large herring has more dioxins than small ones and even when you learn that young women are a sensitive subgroup. So, if your recommendation is the same thing in all situations, there is no point in trying to identify different situations.

Based on our discussion so far, I am the only one who has said aloud that we should recommend 1. If this was a democratic decision, we would vote and thus drop recommendation 1. But before the final vote we should be scientific in the sense that we reject ideas that are not supported by our data and premises. Our research is compatible with recommendation 2. But I am saying that it is ALSO compatible with recommendation 1, which includes recommendation 2. So, we could go with the stronger recommendation 1 and have a larger positive impact. The main differences between the two:

 *   In 1, people can stop worrying about and monitoring dioxins and stop targeting recommendations, which all are a burden. In 2, they can't.
 *   In 2, we do not solve the problem of bad reputation of herring, which is important in export efforts. In 1, we do.
 *   In 2, the Baltic fish consumption and thus food security may improve. In 1, there is more potential for that.

Someone might also think that recommendation 1 is just impossible to get accepted, as these things are decided in the EU bureaucracy. I think that is irrelevant. We should say what our data and values (net health benefit, food security, cultural values etc) drive us to tell. If the experts tell the king only those things that the king wants to hear, we are soon in deep trouble.In practice, we should NOT use the current wording such as "getting rid of dioxin-based food restrictions related to Baltic herring and salmon at least for ages > 45 years, and through this promoting human consumption of Baltic fish" because that mixes up recommendation 1 and 2. We either recommend getting rid of restrictions, or we recommend certain fish consumption in certain subpopulations.Hopefully this clarifies. Jouni Tuomisto (type: ; paradigms: science: relevant comment)

arg7993: . Based on your reasoning, can we recommend "Fish consumption should be promoted among adult men and women above fertile age"? Based on your result, this is where there are only health benefits and no health risks. And is builds on that all fish is a source of Omega3, and not just Baltic herring. Anna Gårdmark (type: ; paradigms: science: relevant comment)

arg9420: . i also think the report looks good, and I think the sentences highlighted in yellow nicely summarise the results presented to us by Jouni. Regarding your comments below, Jouni, I think the results I've seen from WP5 support your suggestion 2 but not suggestion 1. Baltic herring consumption is net benficial and virtually risk-free for adults >45 years. However, other population groups could (and possibly should) instead be advised to prefer intake of omega-3 fatty aids and vitamin D predominately from other sounrces than Baltic herring. This is nicely condensated in suggestion 2, I think. So I suppors having suggestion 2 as a GOHERR recommendation, but not suggestion 1. Anna's suggestion also looks good to me. Andreas Bryhn (type: ; paradigms: science: relevant defense)

arg1642: . To try if we can end up to a democratic decision: Jouni, will you do a doodle poll? Perhaps include the two alternatives you suggested, and the third that there already is (> 45 years) ?? And whatever the final recommendation will be, would it good to include some background information in the recommendation text i.e. what the recommendation is based on? (value judgment based on...(food security?)....and a sentence about the risks vs benefits assessment?) Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg2220: . I’m sorry but this is not a majority vote. We need consensus. Else you can’t have it in the report. Anna Gårdmark (type: ; paradigms: science: relevant attack)

arg8358: . I agree. We need consensus. But in my experience, consensus is based on discussion and compromise. So far we seem to have had the discussions, but not much compromise. Without compromise, I don’t see how we can come to a consensus and we may need to follow Paivi’s suggestion of a vote? This is an issue that is so central to the project, I don’t see how we can NOT have a recommendation for it in our final report. So simply leaving it out due to lack of agreement doesn’t seem like a tenable solution. Alyne Delaney (type: ; paradigms: science: relevant attack)

arg1642: . So what about the new compromise I suggested in the email this morning? “Fish consumption should be promoted among adult men and women above fertile age” Can that work? I repeat it since I haven’t seen any feedback on this suggestion. Anna Gårdmark (type: ; paradigms: science: relevant comment)

arg7993: . I tend to think so that our responsibility as researchers is to report our findings honestly and objectively, not to support any political agendas. Value judgements are always related to decision making, but that is something the decision makers should do - not us! We can help the DMs to make these judgements, that's all.

We know the policy makers would like to get simplistic recommendations and one number -type of answers. Anyhow, as the things really are not that simple in reality, we need to make them realize that and provide as simple recommendations of the complex system as we can - but not any simpler (Occam's razor principle).

Dioxins have some well known harmful health effects (the tooth developmental thing) and in addition some potential, not so well known or defined. In addition there still are fish in the Baltic Sea that exceed the safe limits. For this reason I wouldn't like to say that "Hey, you can safely eat as much as you want, go ahead!"

Thus I suggest formulating the recommendation so that it brings up the different perspectives, e.g. the difference in our results if take the national population level perspective or group level perspective (and even the latter case there is plenty of individual level uncertainty). I think that we can still - by the means of the other recommendations - advance the herring consumption.

The results of the integrated model showed that promoting fish eating is beneficial to elderly people, but produced harm to the young female. This result is based on the change in DALYs as the measure of utility / harm. Thus although promoting fish eating would be slightly beneficial on population level, we should not give the recommendation 1. To my mind each young female should have the right to know the existing risk, even though it is "only" something related to the teeth of their children. And wasn't one of the fish eating query results that people would like to get more transparent information? Annukka Lehikoinen (type: ; paradigms: science: relevant comment)

arg1642: . Isn't Anna's “Fish consumption should be promoted among adult men and women above fertile age” quite near Jouni's nr 2)" Promote Baltic herring consumption in a targeted way. This includes recommendations to eat more small herring and recommendations for particular subpopulations (everyone except young women planning to get children)" ? This would also be supported by WP6 modelling I suppose? Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg1642: . I have meetings for the rest of day so it will be difficult to participate in the discussion after this. It seems obvious that my recommendation 1 does not get accepted with consensus and not with majority vote. So we just drop that. The remaining question is how to formulate recommendation 2. There is a slight difference between Anna's and my formulation. The critical details are:

 *   Preference for other fish than Baltic herring? Anna: yes, Jouni: no need but ok
 *   Preference for small herring? Both: Anna: Yes, Jouni: ok, why not
 *   Restrict consumption in which subgroup? Anna: young people, Jouni: only young women planning to get children, if any

So, the largest difference and discussion point is the subgroup. Can we formulate the subgroup so that everyone agrees? Jouni Tuomisto (type: ; paradigms: science: relevant comment)

arg3935: . I have to make the PDFs and submit before 3 pm today. So there is not much time anymore for this. My suggestion is: Why couldn't we do as Annukka suggests? Give the results and let decision makers draw conclusions? Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg3935: . What about something like this: "GOHERR results suggest that health benefits of Batic herring and salmon outweigh risks in age groups over 45 years. Benefits seem to be higher than risks even in the most sensitive subgroup, women of fertile age. In women at fertile age the critical issue is the potential effects of dioxins in their children's intelligence quotient (IQ) and tooth defects, not health impacts on the women themselves. However, there are still large uncertainties in both scientific and value-based issues. Thus, GOHERR recommends targeting information and eating recommendations to the right consumer groups." Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg7993: . All the sentences are copy-pasted from the goherr work done. But there is room to improve/specify the wording. Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg2939: . Päivi's suggestion looks ok to me. I was drafting something like this, if you want to use something from it:

 *   In older age groups (>45 years) the benefits from eating Baltic herring and salmon clearly override the potential health risks caused by the dioxins. For this reason the herring and salmon consumption should be promoted in those age groups.
 *   The negative impacts of high consumption of Baltic herring and salmon in the younger age groups (<45 years) are related to their unborn offspring, the most remarkable consequene being the tooth developmental problems (tooth enamel damage). For this reason young female planning to get children should be informed about this risk. As the dioxin exposure results from the total amount of the Baltic herring and salmon eaten, species-specific recommendations are not needed.
 *   Smaller fish regularly contain less dioxins per biomass unit than larger ones. For this reason, to maximize the total utility of herring and salmon eating, promoting the use of smaller fish is recommendable. Annukka Lehikoinen (type: ; paradigms: science: relevant defense)

arg2939: . Thanks Annukka. I would go with this kind of fact-based argumentation. But an important question is: Do we still need to speak out if this recommendation concerns the whole EU (where the selling restrictions are in place)? Or shall we leave it implicit? Päivi Haapasaari (type: ; paradigms: science: relevant defense)

arg2939: . The senior coordinator here. I have read the whole final report thse emails, and I very much agree with Alyne's comment that we shold find a compromise. I took Päivi's last text, took some elements from Annukka's text and then included, from Jouni's "agreement lines" the point where agreement should be found. So, my suggestion is as follows: "GOHERR results suggest that health benefits of Baltic herring and salmon outweigh risks in age groups over 45 years and in young male. For this reason, the small herring and young salmon consumption should not be restricted in these groups. In women at fertile age the critical issue is the potential effects of dioxins in their children's intelligence quotient (IQ) and tooth defects, not health impacts on the women themselves. However, there are still large uncertainties in the scientific basis of this advice. Thus, GOHERR recommends targeting information and eating recommendations to the right consumer groups. the consumption of freshwater fish includes no risks from dioxin." I think it's not wise to go to the EU policy. The project did not analyze it, and it is a complex risk analysys between areas and countries. If Finland (and potentially Sweden) want to get this issue to the EU agenda, that is possible and all scientific knowledge will then be collected for evaluations, and especially GOHERR results, I think. My understanding is that a practical problem in the small herring -advice is that all herring consumption is fillets. But perhaps you have solved it already?? What do you think of this compromise? Sakke Kuikka (type: ; paradigms: science: relevant defense)

arg9420: . I'm ok with Sakke's formulation. And maybe this "eat smaller fish" -recommendation is not that focal that it couldn't be left out. It's just the way to maximize the utilities while minimizing the risks. Idea we have discusses during the project is using small herring to e.g. ready-made meals, such as casseroles etc. In addition, developing some "pulled oats" type "pulled herring" and other innovative products where smaller fish could be used. That might be a way to bring Baltic herring to the European markets. Annukka Lehikoinen (type: ; paradigms: science: relevant defense)

arg9420: . I agree with this text. However, I would talk in general about children's health eefcts rathehr than specifically about IQ and tooth defects. The reasons are a) a recommendation should be short and simple b EFSA just this week said taht theh critical endpoint is sperm concentration (it done not affect our decision but it is not in our official assessment) Jouni Tuomisto (type: ; paradigms: science: relevant defense)

arg7993: . "GOHERR results suggest that health benefits of Baltic herring and salmon outweigh risks in age groups over 45 years and in young male. For this reason, the small herring and young salmon consumption should not be restricted in these groups. In women at fertile age the critical issue is potential effecrs of dioxins in their childrren's intelligence qoutient (IQ) and tooth defects, not health impacts on the women themselves. However, there are still large uncertainties in the scientific basis of this advice. Thus, GOHERR recommends targeting information and eating recommendations to the right consumer groups." This would include also the small-size fish issue: "GOHERR results suggest that at the public health level, health benefits of Baltic herring and salmon outweigh risks in age groups over 45 years. For this reason, the small herring and young salmon consumption should not be restricted in these groups. In somen of fertile age the critical issue is the potential effects of dioxins in their children's health, not healt impats on the women themselves. However, there are still large uncertainties in the scientific basis of this advice. Thus, GOHERR recommends targeting information and eating recommendations to the right consumer groups. GOHERR also recommends use of small-sized herring and salmon for human consumption, as their likely dioxin concentration is lower than large fish. The consumption of freshwater fish includes no risk from dioxin." Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg7993: . I agree with the top one. The lower one is too unspecific (what is small size?), and we know from our WP5-results that dioxin concentration in small fish can vary substantially. So I don't agree with that one. Anna Gårdmark (type: ; paradigms: science: relevant comment)

arg7993: . So we go with this: "GOHERR results suggest that health benefits of Baltic herring and salmon outweigh risks in age groups over 45 years and in young male. For this reason, the small herring and young salmon consumption should not be restricted in these groups. In women at fertile age the critical issue is the potential effects of dioxins in their children’s intelligence quotient (IQ) and tooth defects, not health impacts on the women themselves. However, there are still large uncertainties in the scientific basis of this advice. Thus, GOHERR recommends targeting information and eating recommendations to the right consumer groups. The consumption of freshwater fish includes no risks from dioxin." Thank you Sakke! Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg7993: . Dioxin concentration varies between individuals, but it is the long term average that matters. Peaks of dioxine from some individuals is not that dangerous. In our paper some years ago we made this mistake, but I learned the right argument from Jouni. One could think that in risk analysis it is the variance of mean that matters, not the variance of individuals. Sakke Kuikka (type: ; paradigms: science: relevant comment)

arg7993: . Actually also in this version we also refer to small herring and young salmon. Remove it or specify with (herring under 17cm and salmon between 40-80cm): "GOHERR results suggest that health benefits of Baltic herring and salmon outweigh risks in age groups over 45 years and in young male. For this reason, the small herring and young salmon consumption should not be restricted in these groups. In women at fertile age the critical issue is the potential effects of dioxins in their children’s intelligence quotient (IQ) and tooth defects, not health impacts on the women themselves. However, there are still large uncertainties in the scientific basis of this advice. Thus, GOHERR recommends targeting information and eating recommendations to the right consumer groups. The consumption of freshwater fish includes no risks from dioxin." Päivi Haapasaari (type: ; paradigms: science: relevant comment)

arg7993: . My view is that it is better to use them than not to use? Sakke Kuikka (type: ; paradigms: science: relevant comment)

arg7993: . Ok, I will insert the length specifications also. Päivi Haapasaari (type: ; paradigms: science: relevant comment)