European Commission logo
English English
CORDIS - EU research results
CORDIS

Role of health-related claims and symbols in consumer behaviour

Final Report Summary - CLYMBOL (Role of health-related claims and symbols in consumer behaviour)

Executive Summary:
CLYMBOL (Role of health-related claims and symbols in consumer behaviour) is an EU-funded project which has explored the impact of health claims and symbols on consumers in Europe.
The aim of the project has been to study health claims and symbols in their context, e.g. as they appear on a food package, together or without additional (visual) information, and how they interplay with national (cultural) differences as well as personal factors such as motivation and/or ability to process this health-related information.
CLYMBOL showed that the prevalence of claims and symbols on foods and drinks varies between EU countries, with a quarter of all foods carrying some sort of health-related claim (nutrition claim, health claim or health-related ingredient claim). It was further found that foods bearing health claims were marginally healthier than foods without.
Consumers with a stronger personal need for health-related information, with a stronger motivation to engage in preventive health behaviours, and those who seek more stimulation during shopping have a stronger motivation to mentally process health claims. Subjective (self-assessed) knowledge about the healthiness of foods, followed by being on a diet emerged as the key drivers of consumers’ ability to process health claims and symbols. Overall, motivation to process health-related information is more important for European consumers than their cognitive ability, as a driver of the use of health claims.
At the point-of-sale, health claims have to be comprehensible; complex words or expressions should be avoided. Familiar nutrients are easier to understand for consumers. However, a too familiar combination of nutrient and function in a claim evokes less attention. As a consequence, health claims should be understandable but also need to be presented in a way that is seen as new, in order to attract attention in-store. Gaining consumer attention is crucial as it is a bottle-neck for the use of health claims. Having a health goal increases consumers’ attention to health claims. Health-related images (both claim-specific and overall health) can be helpful for people who are actively searching for products with claims. Differences in motivation and ability to mentally process health claims and symbols between countries and consumer groups should be considered when communicating about health claims and health symbols.
To improve trust, consumers should be (more) informed of the stringent requirements that health claims and health symbols have to fulfil. Effects of claims and symbols are subtle and multiple factors influence consumers’ behaviour.
The insights gained from this research can be used to inform the on-going evaluation of the EU Regulation on nutrition and health claims as well as other guidelines. To improve the relevance of claims, it is important that the health claims (and supporting information) are known, relevant, (scientifically) correct and understandable by consumers.

Project Context and Objectives:
The Regulation (EC) 1924/2006 on nutrition and health claims went into effect in 2006. It was designed to offer industry a guideline on how to use claims on food products, ensuring the effective functioning of the internal EU market, whilst protecting consumers and their right to non-misleading food information. However, the actual effect of health claims and symbols on European consumers’ understanding, purchase and consumption behaviour was largely unknown at that time. The European Commission decided to issue a call under the 7th Framework Programme for Research and Technological Development, asking researchers to contribute to a better understanding of consumer behaviour in relation to claims and symbols on food products. CLYMBOL was awarded the grant (Grant Agreement No 311963) and began its work in 2012.
The project can be seen as the successor to FLABEL, building on insights on nutrition labelling in order to strengthen knowledge on consumer understanding in the field of health claims and health-related symbols.
CLYMBOL has aimed to study how health claims and symbols influence consumer understanding, purchase and consumption behaviour. During a 4-year period, a wide range of research studies have been conducted across Europe, in order to analyse European consumer behaviour in the context of health claims and symbols. Results of the studies provide a basis for recommendations for stakeholders such as policy makers, the food industry and consumer and patient organisations.
The theoretical foundation for CLYMBOL’s structure has been a set of prerequisites for the effectiveness of claims and symbols: if health claims and symbols should support informed choice, further healthy eating, and strengthen the competitiveness of the European food industry, three conditions need to be met:
1) Consumers must understand health claims and symbols correctly
2) Effects on healthy eating will occur only if the claims and symbols actually have an effect on consumer purchasing
3) This does not necessarily imply that overall food consumption moves into a healthier direction – claims should also impact food choice and consumption behaviour
Considerable research attention has been devoted to understanding how consumers react to nutrition and health information on foods. Whereas initially studies focused mainly on the effects of nutrition labels, recently a considerable number of studies have assessed the effect of health claims, where a link is made between a nutrient and health outcome. The majority of these studies have assessed how factors related to the health claims, the products, and the consumer impact consumer understanding, attitudes, and perceptions of the health claim product (For an overview see Grunert, Scholderer, & Rogeaux, 2011; Wills et al., 2012). Little is known on how health claims and symbols affect purchase and specifically consumption behaviour. CLYMBOL proposed a research framework to outline how we expect health claims and symbols to affect purchase and consumption of health products.
The hierarchical model shown in Figure 1 summarises these processes which are reflected in the empirical research undertaken in the project.
Figure 1: Conceptual framework for effects of health-related claims and symbols on consumer behaviour

In the current research framework, we study the effects of health claims and symbols on purchase and consumption in the context of the package. We focus on the effects of patterns in claim- symbol- package combinations, specifically the congruence between different context factors in order to be able to make more general predictions of the effects of claims and symbols on consumption and purchase behaviour of consumers.
Regarding consumption behaviour, we focus on the context of other products that are consumed along with health claim products. In realistic consumption settings, many common carrier products of health claims (e.g. yoghurt, butter spreads, bread or milk) are not consumed in isolation but are likely to be consumed along with other food products. The healthiness of other foods present may serve as a reference against which the healthiness of the claim products is perceived (and vice versa) and this may affect consumption of both the health claim product and other food products. In order to gain insight into how health claims and symbols affect the consumption pattern of consumers, it is important to assess how health claims and symbols affect not only the product itself but also the foods that are consumed alongside the claim product.
We propose that health claims and symbols most likely impact on and promote healthy choices, when health inferences are made from the product and a health goal is salient at the moment of choice. In studying the effects of health claims and symbols on choices we therefore focus on these two processes of inference making and goal accessibility.
The work programme of this project, organised in scientific work packages, was based on the conceptual framework, taking into account country differences in the history of use (WP1) and individual differences (WP2). All empirical work mirrored the three-partite structure proposed by CLYMBOL, on understanding, purchase and consumption.
The objectives of the scientific work packages are as follows:
Work package Main objective
1: Current Status of Health Claims and Symbols: Product Supply Map the presence of health claims and symbols on the market, the context in which they appear, the products they are used on, and the criteria used for assigning symbols
2: Current Status of Health Claims and Symbols: Consumer Needs & Wants Map differences in consumer motivation and ability to process health-related claims and symbols (consumer wants), as well as differences in consumers’ nutritional and health status that may have an impact on whether food products with specific health effects indeed will be beneficial to consumers (consumer needs)
3: Methodological Toolbox: Measuring Effects Develop scientifically validated, state-of-the art methods for measuring how health claims and health symbols – in their context – are understood by consumers, and how they affect consumer food purchasing and consumption
4: Empirical Investigation: Effects on understanding, purchase and consumption Provide the scientific evidence on how claims and health symbols, in their context, are understood by consumers, how they contribute to healthier food choices at the point of purchase and their potential to induce healthier consumption patterns
5: Public Policy Implications: Stimulating Healthy Food Choices for the Consumer and Innovation within the Industry Turn methodological and empirical findings into actionable implications for the different stakeholders (consumers, industry, retailers, NGOs, policy makers) with regard to the type of health claims and symbols, in their context, in order to assist consumers in making informed and healthy food choices
6: Communication, Stakeholder Engagement and Public Dissemination Disseminate and extend the results of the project to a wider audience at the European level and engage all stakeholders and networks involved

Project Results:
The CLYMBOL objectives were achieved by organising the work in a set of work packages (WP; see Figure 2 below), mapping the current prevalence of claims and symbols on the market and which products they appear on (WP 1), identifying consumers’ needs and wants with regards to claims and symbols (WP 2), developing a methodological toolbox the measure and monitor the impact of claims and symbols (WP 3), understanding the role of claims and symbols in consumer understanding, purchase and consumption (WP 4), and implications for public policy, regulators, industry and consumer organisations (WP 5). The project drew on the involvement of stakeholders from the whole food sector to ensure results with high practical relevance and was disseminated widely (WP6). Finally, the project has been managed carefully to ensure optimal use of skills and EC-funding (WP7).
Figure 2: CLYMBOL workflow overview

Work Package 1 – Current status of health claims and symbols: Product Supply
As a point of departure, the CLYMBOL consortium created a benchmark by identifying differences in the history of use of health claims and symbols across Europe. This was done by interviewing key informants in the EU 28 Member States, from the three main stakeholder groups: national food authorities, representatives of the food industry and consumer organisations. While many countries reported (at least partial) regulation of the use of health claims and symbols before 2006 (the introduction of the Health Claims Regulation 1924/2006), mandatory reporting of use was found to have only been in place in three of the EU Member States (Portugal: health symbols, Slovenia: health claims and Italy: both). Some voluntary codes of practice were in use, including pre-approval or justification when challenged. Availability of national data bases on health claims and symbols was and remains low and the data is often incomplete. Stakeholders, in general, expressed a strong interest in measuring the impact of health claims and symbols. The main areas to be investigated are the role of health claims and symbols in consumer behaviour, impact on public health and economic effects. Scepticism, however, was expressed concerning the general effectiveness of claims and symbols but also a lack of know-how, responsible authorities and/or resources in successfully monitoring them.
Based upon already existing classification schemes, a taxonomy of health claims, health symbols and their context was developed, separating out product information and (expressed) claims as food labelling components available at the point-of-purchase. Product information consists of mandatory food information (i.e. the information required according to Regulation (EU) No 1169/2011) and other information (value chain information, marketing-related information, package design and other). Claims, on the other hand, are divided into “worded nutrition and health claims” and “symbolic nutrition and health symbols”. Possible claim types are nutrition, health or health-related ingredient claims. All other claims are categorised as miscellaneous claims. For each group, detailed categorisation schemes have been developed.
Figure 3: CLYMBOL taxonomy

In order to measure the prevalence of health-related claims, a multinational survey (Germany, Spain, the Netherlands, Slovenia and the UK) involving more than 2,000 food products was conducted. The food products were collected following a randomised sampling protocol (3). Researchers found that, between countries, 20 to 35% of food products carry a claim. The UK had the highest prevalence of nutrition claims, whereas the Netherlands had the most health claims. Nutrition claims were the most frequently used claims, followed by health claims and health-related ingredient claims (non-nutrient substances which may have a nutritional or physiological effect). Nutrition and other function claims made up 47% of all health claims while only 5% were disease risk reduction claims. Regarding food categories, 78% of baby foods carried a nutrition claim and 71% carried a health claim. Regarding convenience foods, 9% carried a nutrition claim, whereas egg products didn’t carry any claim, making them the categories with the lowest prevalence.
Figure 4: Prevalence of claims by food group, in 5 EU countries

Using data extracted from nutrient declarations on food labels and from a food composition database, CLYMBOL researchers further compared the nutritional quality of food products with health claims versus products without claims. The current Regulation states that a nutrient profile model should be used to regulate health claims so that only foods with the best nutritional composition may carry health related claims. Analyses determined that foods carrying health-related claims have a marginally ‘healthier’ nutritional profile than foods that do not carry such claims. CLYMBOL researchers showed that these differences would be increased if a nutrient profiling model similar to the FSANZ Nutrient Profiling Scoring Criterion (NPSC) was used to regulate health-related claims.
Work Package 2 – Current Status of health claims and symbols: Consumer needs and wants
In the second work area, the focus lied on evaluating consumers’ motivation and their ability to process health claims and symbols. CLYMBOL researchers designed a set of qualitative studies to improve our understanding of how consumers process claims and symbols. Analysing subjective causal models across 25 different nutrition and health claims showed that familiarity with the nutrient/substance and the personal relevance are the primary influence factors of consumers’ acceptance of nutrition and health claims. These factors vary strongly, depending on the individual, making it very likely that consumers perceive the same claims differently.
Another study with over 500 consumers from five EU countries (Germany, Spain, the Netherlands, Slovenia and the UK) analysed if consumers draw on their personal background knowledge to interpret health claims. It could be shown that participants indeed use personal beliefs and additional causal knowledge. Study participants for example said that saturated fat had an influence on heart health, a statement not mentioned in the examined claim. Therefore, familiar or personally relevant substances could result in an “upgrade” of a statement, showing that consumers’ assessments of the healthiness of claims does not only rely on what is actually stated in the claim.
An online survey in ten countries (Czech Republic, Germany, Denmark, Spain, France, Greece, Lithuania, the Netherlands, Slovenia and the UK) questioned over 5,000 participants about their motivation and ability to process health claims and symbols. Spanish consumers showed the highest scores regarding their motivation and ability to process health claims and symbols, while consumers in the Netherlands ranked lowest. The need for information was found to be the main driver for consumers’ motivation and it was also increased in individuals with a stronger health motivation. Subjective knowledge regarding the healthiness of food was observed as the main factor driving consumers’ self-reported ability to process health claims and symbols. Additionally, a stronger familiarity with health claims led to a higher ability to process health claims. Consumers’ motivation and ability were also shown to be clearly linked.
Further analyses were conducted to assess the public health relevance of health claims. A review of national health statistics from a variety of sources was used to identify the most prominent health needs in each country. This was then compared to the prevalence of health and nutrition claims for each country. According to the Global Burden of Disease (GBD) Study, nearly one sixth (13.6%| of the burden of disease in the European Union (EU) is due to a poor diet and the vast majority of this burden is due to over-nutrition rather than under-nutrition. Of the relatively small amount of the burden due to under-nutrition, virtually all is due to iron deficiency leading to iron deficiency anaemia. Of the large amount of the burden due over-nutrition, over a quarter is attributable to diets high in salt. Slightly lower but similar proportions are due to diets low in fruits or, vegetables, whole grains and high in processed meats. CLYMBOL compared the share of the burden of disease (measured in disease adjusted life years, DALYs) due to different diseases with the health-related functions referred to by nutrient and other function claims. The claims were classified by chapters in the International Classification of Functioning, Disability and Health (ICF). Results showed, for example, that cardiovascular disease is responsible for 18% of the total burden of disease in the EU and half of that (9%) is attributable to poor diets yet only 5% of all nutrient and other function claims are concerned with heart and blood vessel functions. Conversely, 13% of nutrient and other function claims are concerned with digestive and similar functions yet digestive diseases are only responsible for 1.3% of the disease burden and an insignificant proportion of that is diet-related. It was concluded that the prevalence of different types of health and nutrition claim bears little relationship to the burden of diet-related disease in the EU. It must, however, be taken into account that not all disease burden is directly linked to food and nutrition, thus limiting the impact the health outcomes states in health claim can have on DALYs.
Figure 5: Comparing the burden of disease (DALYS) in the EU with the prevalence of claims

Work Package 3 – Methodological Toolbox: measuring the effects
The objective of this work area was to develop scientifically validated, state-of-the art methods for measuring how health claims and health symbols – in their context – are understood by consumers, and how they affect consumer food purchasing and consumption. A first step was to review available methodologies before a series of studies were conducted, in order to evaluate those methodologies in terms of theoretical anchoring and applicability. Recommendations have been derived for the use of these methodologies in three contexts: a) routine use by policy and industry, b) in-depth policy-related studies and c) theory-driven consumer science studies.
On the topic of consumer understanding of health claims and symbols, the CUT (“Consumer Understanding Test”), the soft and the hard laddering method have been selected and further tested. We conclude that the richest data can be gathered from soft laddering interviews (face-to-face or over the phone). This method also results in the least ‘noise’ (i.e. unwanted or irrelevant information). However, cost associated with carrying out these interviews and analysing the data are highest, compared to the CUT method (lowest cost) and the hard laddering approach (medium cost). Ease of use also is highest for the CUT method, followed by the hard laddering approach. The soft laddering method requires the most expertise and experience in order to code and analyse the data in a meaningful and scientifically sound way. Lastly, the methods differ in the usability of their results for stakeholders: while the CUT method allows for testing of consumer understanding compared to the scientific dossier on which the specific claim is based, both the soft and the hard laddering offer broader possibilities of tracing inferences consumers make when confronted with a specific claim. Here, consequences and even links to personal values can be researched.
Regarding methods to study the effects of health claims and symbols on purchase behaviour, sales/scanner data, choice experiments (in-store and via virtual supermarkets/shelves), electrodermal measures (“arousal”) and eye-tracking were selected and tested, in combination with follow-up surveys. As a baseline measure, real sales data were collected in a German supermarket, on a product carrying various claims (nutrition, health-related or general). In a follow-up, several methods were tested in combination, in order to replicate results from the sales/scanner data as closely as possible while at the same time tracing the outcome, i.e. the choice, back to the various aspects listed above: attitude, attention and arousal. Results from the electrodermal measures showed significant differences between the claims. However, arousal could not help explain the simulated product choice. Electrodermal measures as such offer only a limited applicability in field studies. Eye-tracking, on the other hand, was shown to help explain food choice. However, there are also barriers to using this method, including higher resources (equipment, staff and time needed to carry out the study) and experience with analysing the data. As a general conclusion, in order to explain actual sales/scanner data, a combination of eye-tracking with other methods is needed to explore the impact of claims on consumers’ purchase behaviour at the Point-of-Sale.
In order to study the effects of health claims and symbols on consumption behaviour, methods using observation (covert weighting of consumed products), dietary records, biochemical markers and anthropometry were chosen based on the literature review. Based on a comprehensive study including all of the mentioned methods, the following recommendations have been derived: dietary records are a commonly used technique that does not require laboratory equipment, is less labour intensive and less reliant on memory (of the participant). They are, however, subject to bias due to self-reporting, may offer less precise portion information and are reliant on available food composition data bases in order to analyse nutrient intake. Biochemical markers are an objective reference measure (e.g. pre- and post-trial), able to represent recent or long-term intake and offer greater accuracy than traditional dietary assessments. They are, however, less suitable for regional or national measures due to being resource and labour intensive, they may be perceived as inconvenient for the participant (blood and urine samples) and there are difficulties arising from the instability of the samples (easy to contaminate if not handled properly), differences in analyses between laboratories and the interpretation of results. Using methods of anthropometry (measuring the size and shape of the body) offer the advantages of being quick, inexpensive and requiring minimal researcher training. Dietary and behavioural data, however, needs to be collected in addition to this. Overall, it is expected that actual measurable effects of the consumption of products with health claims are difficult to obtain as dietary intake and health status are influenced by a multitude of factors.
Work Package 4 – Empirical Investigation of Effects: Understanding, Purchase and Consumption
This work area aims at investigating current empirical effects of health claims and symbols on consumer understanding, purchase and consumption behaviour, using a mix of quantitative and qualitative studies across Europe. A series of on- and offline studies looked into consumer attitudes towards products with and without claims, with and without visual imagery and combinations of products and claim types. For example, one study in the Netherlands evaluated the influence of package design such as the use of images and colours on the healthiness perception of claims and symbols. It was shown that implicit cues like imagery have the same importance as explicit cues, e.g. a stated claim. Several of these studies have been repeated in or extended to other European countries. For example, online buffet studies in the Netherlands could not show the so-called licensing effect, an effect that postulates that consumers overeat as soon as some of the foods they consume carry a health logo and as such mark a healthy choice. This result could be replicated in a real-life buffet study in Spain.
The stimuli designed and used in CLYMBOL have been of a diverse nature. They ranged from drawings of product categories in combination with health claims, to photos of products that were manipulated (package colour, imagery, health claim, health symbol) to pictures of real foods that were manipulated (additional health claims or symbols).
A series of consumption studies were undertaken: a first conjoint study among a large representative sample of consumers showed that implicit package cues such as images and colours can be potentially as influential as claims and logos in determining how healthy and indulgent a consumer thinks the product is, particularly when the implicit cues are consistent with the explicit health information. A second series of studies investigated the effect of a satiety claim and symbols on consumers’ cognitive and behavioural processes towards cereal-based snacks. Results show that consumers do not seem to pay attention to the verbal information in the packages when assessing expected satiety and healthiness attributes. The results of this series of studies also highlight the importance of the methodology chosen to understand the cognitive and behavioural processes when interacting with packaging elements, in their context. A further series of consumption studies did not show that the presence of a health symbol on an assortment of healthy and less healthy foods can lead to licensing effects which result in increased consumption. Furthermore, in the interplay between internal cues of physiological nature (number of calories ingested) and external cues of psychological nature (satiety claim on the package), it was shown in a final study that internal cues have a dominant effect on food intake, while the effects a satiety claim were in general very subtle.
The objective of the purchase studies was to examine how to ensure that health claims attract attention, are understood correctly, and increase product choice among consumers for whom the claim is relevant. Two studies on understanding indicated that familiarity of nutrients also played a role; more ‘safe’ and fewer ‘risky’ inferences were made when the nutrient was familiar. Moreover, claim-specific images increased the perceived clarity of the health communication for consumers. Two choice studies showed the prominent effect of health goals: among consumers who do not have a relevant health goal choice for products with health claims is low. Health related images can increase choice, but only for people who have a relevant health goal. In an eye-tracking study, we examined attention as the underlying process, and found that images do not seem to draw attention to the claim. Overall, when it comes to choosing products with a health claim, having a specific health goal seems essential.
In in-store studies, specific claim-context combinations were tested. The results from the German in-store study showed that the combination of the two familiar expressions evokes more approach reactions than a health claim that consists of an unfamiliar element. In addition, priming with a health goal can have a positive impact on consumers’ attention to claims. However, attention was decreased when the combination was too familiar. The combination of a health claim and a congruent image increased the perceived healthiness of the product. This effect was transferred to attitude towards the brand and purchase likelihood. The results from the Slovenian in-store study showed again that the group “familiar nutrient and function” achieved the lowest attention values. Post-hoc tests showed a significant advantageousness of the “familiar nutrient/ function but unfamiliar combination” condition against other combination conditions, supporting the findings of the Germany study. The Slovenian experiments revealed an advantageousness of the products with a claim-image-combination (so irrespective whether they are related or not) compared to a situation in which the products bear no claim at all.
Lastly, household panel data analyses in Denmark and the Netherlands – both before and after the introduction of a health symbol (Nordic Keyhole in Denmark and the Dutch Choices logo in the Netherlands) showed that both across countries and across products, households with children tend to have a lower probability of purchasing labelled (health/nutrition symbol) products compared to other household types, while urbanity increases the probability. Apart from this, few identifiable sociodemographic patterns were found. However, other product characteristics might decrease the probability if labelled products are not available, e.g. in discount outlets or as organic products. The overall results suggest that consumers value products with a health/nutrition symbol higher than products without, hence the provision of the health/nutrition symbol constitutes additional value to the consumers for the majority of the products. Furthermore, consumers’ valuation of the Choices logo increased with time, and was larger three months after the product obtained the symbol than immediately after, hence there is a delayed effect of consumer valuation. Results further suggest that stated preferences are important in explaining actual purchasing behaviour; those consumers that stated preferences for the Keyhole symbol were also more likely to purchase products with the symbol and have a higher share of their total purchases that are labelled. The opposite effect did not seem to hold true; those who state that they prefer non-Keyhole products do not avoid them during their purchases. BMI was not found to have a significant effect on household purchases for most of the products analysed. Obese and overweight shoppers are just as consistent in their stated and revealed preferences for Keyhole as those with a low or normal BMI.
Overall conclusions
A better understanding of how consumers perceive health claims, their exposure to them and how they influence consumer behaviour is beneficial for many stakeholders. Policy makers need to know how effective the current legislation is in informing consumers about health claims. Especially with a revision of the current regulation on the horizon, it is important to know whether health claims actually fulfil their purpose: directing the average consumer to a healthier food choice and, ultimately, improving public health. It has been the aim of the CLYMBOL project to provide policy makers with science-based guidelines to support pan-European regulation as well as foster innovation and competitiveness among the European food industry.
At the point-of-sale, health claims have to be comprehensible; complex words or expressions should be avoided. Familiar nutrients are easier to understand for consumers. However, a too familiar combination of nutrient and function in a claim evokes less attention. As a consequence, health claims should be understandable but also need to be presented in a way that is seen as new, in order to attract attention in-store. Gaining consumer attention is crucial as it is a bottle-neck for the use of health claims. Having a health goal increases consumers’ attention to health claims. Health-related images (both claim-specific and overall health) can be helpful for people who are actively searching for products with claims. Differences in motivation and ability to mentally process health claims and symbols between countries and consumer groups should be considered when communicating about health claims and health symbols.
To improve trust, consumers should be (more) informed of the stringent requirements that health claims and health symbols have to fulfil. Effects of claims and symbols are subtle and multiple factors influence consumers’ behaviour.
The methodological toolbox further offers a set of tools and criteria to measure and monitor the impact of claims and symbols on consumer understanding, purchase and consumption behaviour. CLYMBOL’s holistic, interdisciplinary and international approach will provide a sound scientific basis for current debate and future research.
Policy implications
CLYMBOL developed recommendations for public stakeholders such as policy makers on European and national level, the food industry, and consumer and patient organisations. The complete list of stakeholders identified as relevant for CLYMBOL includes producers/manufacturers/suppliers, retail/catering, consumers, regulators, scientists, health and medical professionals, public health authorities, journalists, media in general and educators. Based on the results of the studies in the different work areas, implications from research have been collected and recommendations as well communication guidelines were developed. These were presented and put to vote during CLYMBOL’s stakeholder conference on June 15 2016, in Brussels.
In a live vote where conference attendees voted on 22 questions, the most relevant factor across all stakeholders present was to focus on ways to improve people’s motivation and interest in healthy eating. The majority of the participants also agreed that people need to have accurate information about new or less familiar nutrients and consumer awareness of existing health claims and symbols should be increased. To communicate the importance of healthy eating, finding innovative ways to communicate was voted as another top priority. However, most voters agreed that all communication has to remain simple and clear, avoiding scientific wording. CLYMBOL showed that consumers do not interpret health claims and symbols the way experts do, therefore, most stakeholders agreed that health claims and symbols have to be clearly explained: what they mean and how they are used.
All conference participants voted on how relevant certain topics are to their organisation and whether they think these topics are feasible in practice. Interestingly, in most cases, relevance to one’s organisation and feasibility in practice were similar among the stakeholder groups.
The stakeholders’ feedback was used to prioritise the policy implications and recommendations based on findings from CLYMBOL. These findings are used to support consumers in making informed and healthy food choices and also to promote industry competitiveness, taking into account individual and country differences within the EU. Another goal was to avoid misunderstanding and unwanted effects on consumer behaviour.
Figure 6. Voting on relevance of “Focus on ways to improve motivation such as creating information needs and increasing the interest in healthy eating.”

Figure 7. Voting on feasibility of “Focus on ways to improve motivation such as creating information needs and increasing the interest in healthy eating.”

Potential Impact:
Potential impact
Health claims and health symbols on food products can support informed consumer choice, further healthy eating, and strengthen the competitiveness of the European food industry. Health claims and health symbols are hence an instrument for public policy, aiming to further public health, and a tool for the food industry that can be used when competing by positioning products as healthy alternatives. From the consumer point of view, health claims and health symbols are aids in making the right choices. Health claims and symbols can therefore play an important and useful role in European food markets.

Health claims are regulated, and considerable effort goes into making sure that the health claims that appear on food products are backed by solid scientific evidence. However, making scientifically sound health claims and symbols available on food products is not enough. It is only in interaction with consumers and their behaviour that health claims and symbols will have any effect. Consumers are not nutritional scientists, and it has long been acknowledged that the lack of healthier choices is not solely – and not even mainly – a question of lack of information. CYMBOL has identified four main areas that impact on the use of health claims and symbols in consumers’ food choice:

Familiarity

• The more familiar the health claim/symbol is, the higher are attention, intention to purchase and consumption of the food product with the health claim/symbol.
• However, the health claim should contain some new information for the consumer, as if the claim is too familiar, the attention to and the purchase intention are lower.
• Besides, the presence of a health symbol on food products has a small impact on consumption for consumers familiar with the logo, but has no impact for consumers unfamiliar with it.

Consumers’ health goals

• Having specific health goals (e.g. healthy bones) increases the choice of food products with the specific health claims (e.g. good for the bones). In other words, the purchase of a food product with a health claim is low unless consumers have a specific health goal.
• The relevance of the health claim/symbol for the consumer is a factor that influences the purchase of food products with health claims/symbols.
• Then, it seems that food products with health claims might be consumed by people with health problems and goals, for curative reasons, as the health claims/symbols are relevant for them, but not by healthy or without health goals consumers for preventive reasons because the health claims/ symbols are not relevant.

Contextual factors

• Some contextual factors enhanced or limited the effect of health claims/symbols on food purchase/consumption.
• In particular:
o Visual images together with the health claim favour the choice of the food product with the health claim. This joint effect of the visual image with the health claim is higher if the consumer has a health goal (e.g. healthy bones) related to the health claim (e.g. good for the bones).
o The type of store influences the choice of products with health symbols. Consumers who purchase in a discount store are less likely to purchase food products with the symbol.

Consumers’ personal characteristics

• Some consumers’ personal characteristics also influence the purchase of food products with health symbols.
• In particular:
o Households without children, living in urban areas are more likely to purchase food products with a health symbol.
o Consumers with higher preferences and more motivated towards health food products are more prone to choose food products with a health symbol.

CLYMBOL results have shown that the influence of health claims and symbols on food purchase and consumption is relatively limited. Their influence depends on several factors such as the consumers’ familiarity with the claims, the claims’ relevance for the consumer and some contextual factors. On the other hand, no significantly different patterns across consumers’ personal characteristics and/or between countries were detected.

Nevertheless, some insights on how to improve the role of health claims and symbols on purchase and consumption have been found. Health claims and symbols could increase the purchase and consumption of healthier products if they are understandable, familiar and relevant to the specific group of consumers to whom they may be especially relevant.

As understanding the health claims/symbols by the consumer is important to enhance the role of health claims and symbols on food purchase and consumption, public authorities should increase consumers’ understanding through educational campaigns targeted to the general population, but also to children and young people to become more knowledgeable and motivated consumers, now and in the future.

As consumer’s familiarity with the health claim/symbol is also an important aspect, in order for health claims/symbols to be effective in increasing the purchase and consumption of a food product carrying them, the food industry and public health authorities should improve the familiarity of the health claims/symbols. One way of doing that could be designing and launching different communication strategies targeted to the general population in the case of the public sector and to specific consumer groups with particular health problems and goals, in the case of the food companies.

The relevance of the health claim/symbol to consumers was found important and this relevance could be more easily attained if the health claim/symbol is tailored to specific consumer segments with specific health problems and goals.

Finally, nowadays the amount of information available is tremendous. This also happens in the food market where food products carry an abundant array of information, in addition to health claims and symbols. This includes product, process and brand information (organic, natural, gluten-free but also nutrition information etc.). Consequently, consumers face a huge amount of information that they should process before their food choice/purchase decision. Most often, consumers are not willing to spend a lot of time on these decisions. As such it is logical that only consumers with health goals pay attention and take into account this information when shopping/consuming. In this regard, CLYMBOL studies found that some contextual factors such as visual images help consumers to choose food products with health claims, but that this mainly holds true for consumers with a health goal.

Main dissemination activities and exploitation of results
During its lifetime, CLYMBOL set out different dissemination activities aimed at promoting
its research and at reaching the widest and most varied audience possible.

The first dissemination activities revolved around establishing a communication plan for the project, defining the project identity (including the project logo) and creating the main dissemination tools, the CLYMBOL website and its social media accounts.

The communication plan was agreed with the consortium during the first annual project meeting on September 24-25 2012. At the same meeting, work package leader EUFIC gauged the partners’ media experience, offered training tools and has taken this experience into consideration for the subsequent media-related activities. Academic partners shared key contact details in their respective university press offices. Throughout the project, all dissemination activities have been gathered by EUFIC as the WP leader, documented internally and communicated externally via the project website (News section) and CLYMBOL social media channels (Facebook, Twitter).
Recognising the importance of show-casing the CLYMBOL results, a Publications Committee was formed (comprising all WP leaders), and guidelines were established on how to communicate shared findings, including a template to be completed by all partners for various dissemination plans. This procedure has been kept in place throughout the duration of the project and continues to be in place for upcoming publications, after the end of CLYMBOL. Its aims are to ensure accuracy and transparency in the use of CLYMBOL data, fairness in authorship and the possibility of lesser involved partners to be part of dissemination activities.
The project identity was based on the project objectives. With a clear focus on consumers, the shopping kart as well as a specific mention in the tagline were incorporated, as were “claims” and “symbols”. Health symbols were further symbolised by means of the healthy tick instead of the letter “Y”.

The project website www.clymbol.eu with its user-friendly structure and appealing graphical design went live in January 2013. A first news item on the website was the press release that had been issued for the kick-off of the project in September 2012, providing a generic overview of the project and promoted through EUFIC networks (over 40.000 health professionals, media, policy makers, consumer organisations, food and drink industry, educators, consumers) and the EUFIC website (~500.000 visits/month). The CLYMBOL website was also promoted through a quadrant created on the EUFIC website (www.eufic.org) whose main aim was to drive traffic to the CLYMBOL website.
Other press releases were produced in the course of the project on various publications arising from different work areas (WP1, health claim prevalence study, WP2, health images as claims) and at the end of the project to promote the outcomes of the stakeholder conference (September 2016).

Two leaflets were produced during CLYMBOL life; a generic leaflet (2,500 copies) that introduced the main aims and objectives of the project, and a final leaflet (2,500 copies) that provided an accessible and quick overview of project top line results. Both leaflets have been used by partners for dissemination at conferences and seminars. They were further made available online, via the project website.

A series of short podcasts have been produced, to introduce the project and explain select findings to a lay audience, while at the same time showcasing the researchers behind these studies, by picture and biography. A series of webinars was recorded at the stakeholder conference in June 2016, offering access to the main CLYMBOL findings for anyone who could not attend the conference. A short video, showcasing some of the main findings, researchers and materials of the CLYMBOL project was produced and made available online. It was further disseminated to all subscribers of the CLYMBOL newsletter.

There has been considerable interest in the CLYMBOL project, particularly in the pan-European media. To date, numerous press clippings have been generated that make reference to the project. For example, EUFIC was invited to a podium discussion on consumer behaviour during the European Commission’s conference on 'Nutrition, Health and Food Information - Know what you eat' at the EXPO in Milan, July 3 2015. Many invitations to present CLYMBOL work have been extended over the course of the project and continue to come, due to a high visibility of the project and strong interest in the topic from all stakeholders involved.

Several articles targeted at a lay audience have been written about CLYMBOL. A generic article about the CLYMBOL project appeared in the Supplement issue 3 of EUFIC’s popular newsletter Food Today (September 2013). More than 10,000 subscribers receive the printed version in English, French, German and Spanish. A further 40,000 recipients are sent the newsletter electronically. The article is also available on www.eufic.org in Czech, English, French, German, Greek, Italian, Hungarian, Polish, Portuguese, Slovak and Spanish. Instead of a second article, an entire issue (Supplement issue 5) was dedicated to the CLYMBOL project, introducing in-store experiments and eye-tracking and explaining findings from WP4 to a lay audience (May 2016). A printed issue has been handed out to all stakeholder conference participants and further copies have been send to all project partners, to distribute nationally.

Several interviews with CLYMBOL researchers were carried out and made available online. Notably, the BBC invited Prof Monique Raats to record a podcast titled “Do you know what you’re eating” which is available on the BBC website.

CLYMBOL results were presented and mentioned at different key stakeholder meetings and conferences, held within and outside Europe, with participants representing the broad range of stakeholders the consortium aims to inform about the CLYMBOL project. A selection of those can be found as follows:

• INFORMAS (International Network for Food and Obesity/NCDs Research, Monitoring and Action Support) meeting, November 2012, Bellagio, Italy
• Dairy Council for Northern Ireland Conference: “Nutrition & Health: what’s new?”, 3 May 2013, Belfast, Northern Ireland
• “Milk, Nutritious by Nature” Symposia, 3-5 June 2013, Copenhagen, Denmark / Brussels, Belgium / The Hague, Netherlands
• International Conference “Health Claims and Functional Ingredients” by Fresenius Academy, 18 June 2013, Mainz, Germany
• Plenary meeting of the EU Platform for Action on Diet, Physical Activity and Health, 26 September 2013, Brussels, Belgium
• INFORMAS (International Network for Food and Obesity/NCDs Research, Monitoring and Action Support) meeting, March 2014, Pakiri Beach, Auckland, New Zealand
• Nordic Consumer Council conference, 13 May 2014, Oslo, Norway
• TAIEX workshop on nutrition and health claims, 26-27 May 2014, Skopje, Macedonia
• Summer School on “Food Law”, 26-28 May 2014, Warsaw, Poland
• Public Policy Exchange conference, 10 September 2014, Brussels, Belgium
• International Conference and Exhibition on Nutraceuticals and Functional Foods (ISNFF), 16 October 2014, Istanbul, Turkey
• European Childhood Obesity Group (ECOG) conference, 13-15 November 2014, Salzburg, Austria
• Food Matters Live, 20 November 2014, London, UK
• PAHO/WH Expert Consultation on Region-wide Nutrient Profiling Scheme, 9-10 February 2015, Bogota, Columbia
• World Food Programme, 22 April 2015, Rome, Italy
• EMAC (European Marketing Academy) conference, 24-27 May 2015, Leuven, Belgium
• ISBNPA (International Society of Behavioural Nutrition & Physical Activity), 3-6 June 2015, Edinburgh, UK
• Biennial conference of the Society for Applied Research in Memory and Cognition, 24-27 June 2015, Victoria, Canada
• NUTREVENT conference, 18 June 2015, Lille, France
• ISBNPA Advancing Behaviour Change Science Conference, 6 June 2015, Edinburgh, UK
• WHO-Emro Intercountry Meeting on Nutrition, 7-9 June 2015, Amman, Jordan
• Conference “Nutrition, Health and Food Information – Know what you eat” organised by DG SANTE, 3 July 2015, Milan, Italy
• EuroFIR FOOD Symposium, April 4-8 2015, Brussels, Belgium
• European Obesity Summit (EOS), 1-4 June 2016, Gothenburg, Sweden
• First Dubai International Conference on Applied Nutrition (DIFSC), November 6-8 2016, Dubai, UAE

A number of scientific articles have been published already and more are written up for publication in peer-reviewed journals. All articles published during the duration of the project were made available via open access, to ensure maximum distribution. To date, two invitations for book chapters have been received and are being prepared for the respective editors. The list of all CLYMBOL publications to date can be found on www.clymbol.eu. It is constantly being updated.
A special collaboration with the widely-read magazine Agro FOOD Hi Tech led to a special focus issue on health claims, with CLYMBOL researchers writing articles on project results as well as being invited to write the editorial of that issue. These are open access and as such available to all interested readers. A copy of the magazine issue was handed out to all participants during CLYMBOL’s stakeholder conference on June 15 2016:
• Hieke S, Pravst I, Grunert KG. Health claims and symbols: What role is there for health-related information to guide consumer behaviour? Agro FOOD Hi Tech 2016, 27(3). (Access here)
• Hieke S, Cascanette T, Pravst I, Kaur A, van Trijp H, Verbeke W, Grunert KG. The role of health-related claims and symbols in consumer behaviour: the CLYMBOL project. Agro FOOD Hi Tech 2016, 27(3). (Access here)
• Kaur A, Hieke S, Rayner M. Do health and nutrition claims meet consumers’ health needs? Agro FOOD Hi Tech 2016, 27(3). (Access here)

To achieve active involvement from stakeholders, a Stakeholder Advisory Board (SAB) was established that served as a valuable discussion platform to the consortium. The SAB comprised European and national food authorities, consumer organisations, as well as food and retail sector representatives (both multinational and SME). The members were:

• ANSES (French agency for food, environmental and occupational health safety)
• BEUC (The European Consumers’ Organisation)
• Coeliac UK
• Department of Health, UK
• DG SANCO (Health and Consumers Directorate General)
• EHN (European Heart Network)
• EPHA (European Public Health Alliance)
• Eurocommerce (British Retail Consortium)
• FoodServiceEurope (European federation of contract catering organisations)
• FoodDrinkEurope
• German Nutrition Society DGE
• ILSI (International Life Science Institute)
• NFA (Swedish National Food Agency) – they joined the SAB after they left the project consortium as an active partner, due to retirement of the main person working on the project

The SAB convened annually at the annual project meetings and in addition to providing general feedback on the project, they were active participants in developing and testing the policy implications and communication guidelines in WP5. SAB members also agreed to disseminate CLYMBOL results throughout the project. Three members even joined CLYMBOL in their expert panel discussion during the stakeholder conference in June 2016.

List of Websites:
The CLYMBOL project public website is: www.clymbol.eu
The CLYMBOL logo:

Consortium contact details
EUFIC - European Food Information Council (Coordinator)
Contact: Sophie Hieke
Tassel House - Rue Joseph Steven 7 - 1000 Brussels, Belgium
Tel +32 2 506 89 81 - www.eufic.org
University of Aarhus (scientific advisor)
Contact: Klaus Grunert
MAPP - Centre for Research on Customer Relations in the Food Sector
Bartholins Allé 10 - DK-8210 Aarhus V Denmark
Tel +45 89 48 66 88 - www.mapp.asb.dk
University of Surrey
Contact: Monique Raats
Consumer Behaviour and Health Research Centre, Faculty of Arts and Human Sciences
Guildford - Surrey GU2 7XH, England, United Kingdom
Tel +44 1483 689994 - www.surrey.ac.uk/SHS/fcbh.html
Wageningen University
Contact: Hans van Trijp
Department of Social Sciences, Marketing & Consumer Behaviour Group
Hollandseweg 1 - 6706 KN Wageningen, The Netherlands
Tel +31 317 4 83385 - www.mcb.wur.nl/UK/
Universität des Saarlandes
Contact: Andrea Groeppel-Klein
Institut für Konsum- & Verhaltensforschung, Im Stadtwald
Gebäude A5 4 - D-66123 Saarbrücken, Germany
Tel +49 681302-2135 - www.ikv.uni-saarland.de
University of Oxford
Contact: Mike Rayner
Nuffield Department of Population Health, British Heart Foundation Centre
New Richards Building 2/F, Old Road Campus OX3 7LG Oxford, United Kingdom
Tel +44 (0)1865 289244 - https://www.ndph.ox.ac.uk/bhfcpnp
Ghent University
Contact: Wim Verbeke
Department of Agricultural Economics, Faculty of Bioscience Engineering Coupure links 653, B-9000 Gent, Belgium
Tel + 32 9 264 59 27 - http://www.ugent.be/bw/agricultural-economics/en/department
University of Ljubljana / NUTRIS
Contact: Igor Pravst
Kongresni trg 12, 1000 Ljubljana, Slowenien
Tel +386 1 241 85 00 - https://www.uni-lj.si/eng/
Corvinus University of Budapest
Contact: Judit Simon Institute of Marketing and Media
Fõvám tér 8., H-1093 Budapest, Hungary
Tel +36 1 482-5039 - http://marketing.uni-corvinus.hu/index.php?id=25260
University of Copenhagen
Contact: Sinne Smed Department of Food and Resource Economics
Rolighedsvej 25, DK-1958 Frederiksberg C, Denmark
Tel +45 35336800 - http://ifro.ku.dk/english/

Agrifood Research and Technology Centre of Aragon
Contact: Azucena Gracia
Avda. Montañana 930, 50059 Zaragoza, Spain
Tel +34 976716300 - http://www.cita-aragon.es/
Schuttelaar & Partners
Contact: Léon Jansen
Zeestraat 84, 2518 AD Den Haag, Netherlands
Tel +31 70 318 4444 - https://www.schuttelaar-partners.com/
Swedish National Food Agency - the partner left the consortium after Mrs Laser-Reuterswärd retired, and joined the CLYMBOL SAB to ensure continuing input into the work of CLYMBOL
Contact: Anita Laser-Reuterswärd
Rosenbad 4, SE 103 33 Stockholm, Sweden
Tel +46 8 405 10 00 - http://www.government.se/government-agencies/swedish-national-food-agency/
GLOBUS SB-Warenhaus Holding
Contact: Michael Helfen
Leipziger Str. 8, 66606 St. Wendel, Germany
Tel 06851-9090 - http://www.globus.de/de/home.html