Final Report Summary - FLABEL (Food labelling to advance better education for life)
Nutrition labelling is meant to aid people in choosing healthier products when shopping for food. It is one of many initiatives taken to encourage people to eat more healthily. Nutrition labelling is based on the assumption that one of the reasons for less healthful choices is that consumers may find it hard to make judgments about the nutritional properties of a given foodstuff, and that therefore providing information on the nutritional properties of the food can help in making healthful choices, especially when the information is given in an easily understandable and easily comparable format. The nutrition information to be found on the back of many products (as a table or line), which can be hard to find and read, has therefore more recently been supplemented by simplified information presented on the front of the pack. There has been considerable discussion on which format is best for such information, with three main contenders: guideline daily amounts (GDA), where information on the key nutrients (fat, saturated fat, sugar, salt) and energy is given in grams / calories per portion and also converted into percentages of the guideline daily amounts for these nutrients / energy; colour-coded systems, such as the traffic light labelling, where the information on nutrients / energy per portion is overlaid with a colour red, amber, green in the case of the traffic light labelling depending on whether the amount (per 100 grams or, if serving size exceeds 100 grams, per serving) is evaluated as high, medium or low; and health logos, where products are entitled to bear a logo (e.g. the Choices International logo or the Swedish Keyhole) if they fulfill certain, category-specific criteria with regard to their nutritional profile. This discussion has resulted in a call for research that can shed light on which format is best for nutrition labelling, based on an investigation of how nutrition labelling information is processed by consumers.
The EC-funded project FLABEL, coordinated by the European Food Information Council (EUFIC) and involving a consortium of European universities and stakeholder groups, has been a major research effort in Europe to date to deal with these issues. The work in FLABEL has been structured by a model of labelling effects. If label information is indeed available on food products, any effect on food choice is contingent on that consumers pay attention to the label and that they understand the label information. It also helps if consumers like the nutrition labelling system. To which extent consumers will pay attention to labels, like, understand and use them, may depend on the label format, but will also depend on the consumers' motivation to make healthful choices. Analysing these factors in context gives insight into how different label formats affect consumers choices, but will also shed light on the importance of other potential bottlenecks like lack of consumer motivation, lack of attention, lack of liking, and lack of understanding.
Project results:
Availability of nutrition information on food products
To lay the ground for further work, FLABEL started by investigating the current availability of nutrition information on food products. For five product categories (sweet biscuits, breakfast cereals, chilled fresh ready meals, carbonated soft drinks, yoghurts), a comprehensive audit of products on supermarket shelves was carried out in all 27 European Union (EU) countries and Turkey. Results showed that 85 % of the products contained nutrition information on the back of the pack, and 48 % on the front, with some variation between countries. GDA labels were the most common form of front-of-pack (FOP) labelling, with an average penetration of 25 % across product categories and countries. The results thus indicate that there is ample nutrition information already available on food products, although most of it is on the back of the pack in the form of tabular or linear information, with additional information varying across countries by penetration and format.
Consumer liking of nutrition labels
Results from a survey of 2000 consumers in four different countries confirmed that consumers like front-of-pack nutrition labelling. When given a choice between different formats nutrient information in grams / calories only, GDAs, traffic lights, hybrid labelling combining GDAs and traffic lights, use of a health logo there was a tendency that consumers preferred formats with which they were familiar, but generally they liked the most complex and information-rich format best, i.e. the hybrid GDA-traffic light label. However, when asked to evaluate the different formats in terms of criteria such as usefulness, completeness, simplicity, ease of understanding, no major differences between the formats emerged. We concluded, in line with earlier research, that consumers like front-of-pack nutrition labelling, and while more complex formats are liked more, consumers do not have strong opinions.
Segmentation across countries has produced coherent segments irrespectively of the country. Cluster 1 (57 %) comprised respondents who perceive labels as more complete, less complex, simple overall, acting in a coercive manner and trustworthy. Cluster 2 (43 %) is the opposite. Respondents feel confused facing substantial problems in their interaction process with the labels, not trusting them, feeling they are not simple and that these labels (systems) also being incomplete. Regarding what issues exercise the greatest impact on liking, the conclusion is that three elements stand out.
These are:
a) past label use;
b) food context (complex hedonic foods like pizza appear to yield the highest coefficients, followed by simple utilitarian (yoghurt) where as simple hedonic foods like biscuits appear to yield smaller coefficients; and
c) labels use frequency.
Familiarity is thus by far the greatest determinant and no other among the large number of investigated elements, including socio-economic factors, managed to compete against. Familiarity appears thus to become a mediating factor between consumer characteristics and liking which when introduced and expanded absorbs and integrates all other influencing elements.
Consumer understanding of nutrition labels
Consumers ability to understand and make correct inferences from nutrition label information was studied using a variety of tasks. In the core test, consumers were asked to rate the healthfulness of three products in a given product category, and this healthfulness rating was then compared to an objective measure of healthfulness based on the same key nutrients, namely the SSAg/1 measure. Consumers evaluated the products first only based on information about the four key nutrients given in grams and energy in calories per portion. They were then given the same information supplemented with one of the standard front-of-pack labelling formats GDAs, traffic lights, hybrid GDA-traffic light label, health' logo and had to do the same task again. This allowed assessing how much each labelling format improves consumers ability to evaluate the healthiness of the products compared to the situation when the only information available are the nutrients in grams and the energy in calories per portion. Consumers generally had no difficulties in ranking the three products correctly according to healthfulness already when only given the information in calories and grams per portion. Giving additional information (GDAs, traffic lights, health logos) resulted only in small improvements of accuracy. This was true across different product categories. We concluded that correct understanding is not a major problem with any of the nutrition labelling formats.
Attention and reading of nutrition labels
If consumers both like and understand nutrition labels, will they actually notice and read them when shopping? Attention and reading were studied both in a laboratory setting and in a real-store setting. These studies clearly demonstrated that attention or rather the lack of it is a major bottleneck when it comes to the effects of nutrition labelling. The nutrition label only got a tiny share of attention of consumers 1.23 % when selecting a box of cereals. The average gaze duration was 0.021 seconds, too short for any in-depth processing. A number of factors were shown to influence attention, like familiarity, consistency of location, but also the information density on the rest of the package. While there was no consistent pattern favouring any of the label formats in attracting attention, it seems that the health logo was best in being attended to and actually influencing choice in a situation of time pressure. However, the major factor that affected attention and reading was motivation.
Motivation to make healthy choices
Motivation was studied by instructing respondents in a variety of research tasks to either choose the product that was most healthful, or choose the product they would select in a shopping situation, i.e. choose by preference. This manipulation resulted in major differences in both attention to nutrition labels and in the choices that respondents in the various studies actually made. These results confirm earlier research showing that healthfulness is only one among a range of factors influencing food choice, like expected taste, familiar brand, and family liking. We conclude that motivation to make healthful choices competes with other motives in food choice and is a major bottleneck with regard to the effects of nutrition labels on food choice.
The ideal label
Based on the results that:
a) giving nutrient / energy information in grams/calories only seems to be enough to enable consumers to rank products correctly in terms of healthiness;
b) attention is a major bottleneck in label effects;
c) consistency in format and placement of the information has a positive effect on attention; and
d) a health logo seems promising not only in getting attention but also carrying the information through to choice, we constructed a baseline ideal label.
This baseline ideal label gives the basic nutrient information in grams / calories per 100 grams in a consistent format across products, is supplemented by a health logo (also in a consistent format), and makes it visually clear when a product does not qualify for the health logo. We tested this baseline ideal label in a real-store setting and found that consistent use of this label did indeed increase attention for the label by a factor of about four, although the share of attention given to the nutrition label after introduction of the new label was, compared to the other information on the package, still not very high (between 4 and 5 % depending on the product category). There was no general effect on the choices actually made in the real-store setting, although there was a significant effect of the new label in improving the healthfulness of the choices made by those shoppers that had a lower degree of self-control (i.e. weaker in self-regulatory processes with regard to thoughts, emotions, impulsive behaviours, and performances that help consumers overcome temptations in favour of long-term goals).
Adding additional information to this baseline ideal label, such as GDA percentages, traffic light colours, texts high / medium / low, were investigated both in a laboratory setting and in a real-store setting. No effects of adding such extra information were found, except for a small tendency in two particular cases where the use of colours seemed to increase healthfulness of the choices.
Potential impact:
FLABEL, was developed in order to meet the need for further knowledge on consumers handling of nutrition labels, and thus to help basing policy decisions on evidence. The project aim was to determine how nutrition information on food labels can affect dietary choices, consumer habits and food-related health issues by developing and applying an interpretation framework incorporating both the label and other factors / influences. Furthermore, the project also aimed to provide the scientific basis on use of nutrition information on food labels, including scientific principles for assessing the impact of different food labelling schemes, to be shared with the EU institutions, the food industry, especially small and medium-sized enterprises (SMEs), and other stakeholders.
The results of the FLABEL project have filled in some of the gaps in our knowledge about nutrition labelling in Europe. They show how labels are widely available on food products across the continent, and how consumers are capable of understanding and using nutrition information under certain conditions. Across all the studies carried out as part of FLABEL, a major conclusion emerging was that attention and motivation are the major bottlenecks in trying to encourage more healthful choices by nutrition labelling. Food choice is driven by multiple motives, and health is only one of them. On top of that, choosing food products in supermarkets can be a matter of seconds, and the amount of attention given to nutrition labelling information is small. Against this background, much of the earlier discussion on finding the best form of nutrition labelling seems ill-conceived, as it was mostly based on issues about understanding or possible misunderstanding of different labelling formats.
Our research shows that all the current label formats are well understood by consumers, at least in the sense that they can use them to correctly rank products within a product category according to healthfulness. When aiming to encourage healthful choices by nutrition labelling, one should rather ask which elements of nutrition labelling are most helpful to pass the attention filter. Here, consistency and familiarity seem to be most important. Whether one supplements the basic nutrient information with other elements, like daily amounts or colour-coding, appears to be a secondary consideration. Health logos may have a better chance of not only passing the filter of attention but also receiving enough processing capacity to actually influence choice, although one should only expect modest effects at best. The other basic bottleneck, motivation, cannot be solved by nutrition labelling. Motivation for healthy eating can be affected by measures of nutrition education. But motivation will be a bottleneck only as long as the health motive will lead to a different choice than the preference motive. When the preferred option becomes the healthy option, motivational bottlenecks will disappear. This is a question of product development and reformulation more than of nutrition information.
In order to derive implication of project results for public policy, FLABEL results were shared and discussed with policy makers, retailers, industry and SMEs at a consensus workshop. The goal of the workshop was to jointly arrive at a document that describes both the differing views expressed as well as the areas of agreement identified through the discussion of results between different stakeholders. In the following, the main issues on which consensus were reached are described.
1. Seeing and integrating nutrition labeling into a broader context
FLABEL research has shown that nutrition labelling can positively affect consumers in a number of ways. However, the extent of influence is rather limited. Thus, it is important to see nutrition labels as only one tool among others. Nutrition labels should be integrated into the broader context of efforts aimed at increasing healthy eating, notably information campaigns and (school) education. This is crucially important as FLABEL research has shown that the attention paid by consumers as well as the motivation of consumers to use labels are the most important bottlenecks preventing a bigger impact of nutrition labelling on healthy choices. It was mentioned in this regard that future information campaigns should be made more emotional and engaging.
In addition, also the promotion of environments that lead to the easy choice being a healthy choice should be furthered, also in the light of the FLABEL finding that a larger assortment of healthier products alone can possibly influence healthiness of choice. It was mentioned that new technologies that facilitate information access may have a potential to increase information use.
2. Wider availability of FOP nutrition labelling
Nevertheless, due to its potential benefits, most participants voiced the opinion that nutrition labels should be increasingly displayed on products. More specifically, it was stated that nutrition labelling should be available FOP on all food products in Europe. In order to ensure this, the EC might provide guidelines on the basis of the current regulation. Based on this, it is the industry stakeholders that should be taking action here, in agreement with the remaining stakeholders. The format of the label should be nutrient-based, and the additional use of health logos is supported.
3. Consistency and penetration more important than label format
There is neither clear agreement nor conclusive evidence about whether the nutrient-based label should be directive (e.g. health logo) or non-directive (e.g. tabular/linear nutrition information), and whether colour-coding should be used.
There are, however, research indications that consumers can understand most formats, and it appears that consistency and familiarity are more important than the exact type of format. It was thus agreed that it is important to present the nutrient-based label in a consistent way, especially with regard to positioning. The EC might provide guidelines with respect to the principles underlying the inclusion of forms of expression and presentation of nutrition information which are not specified in the current regulation.
Agreeing on displaying consistent nutrient-based labels at all is more important than the exact format agreed on. Therefore, it was appealed to stakeholders to focus on agreement rather than own interest or the perfect solution. Agreement might be on the dominant form or the lowest common denominator in order to reach a critical mass.
Ensuring high penetration of nutrient-based labels on the market, enough consistency of formats and coupling this with matching efforts in public and private communication and education (public awareness campaigns, food marketing and information provision by producers and retailers, education campaigns and school curricula, etc.) is expected to improve familiarity and effect of nutrient-based labels.
4. Capturing incentives for stakeholders
Many stakeholders voiced that they expect nutrient-based nutrition labels as well as health logos to encourage industry to make efforts in innovation, reformulation and changes of assortments. It was mentioned that consistent nutrient-based labels or health logos can increase transparency in this process for consumers, and they might serve branding functions for producers as well as support producers and retailers CSR efforts. In this regard, consumer skepticism towards food producers information provision was raised as a challenge, and that this aspect of trust and credibility should be tackled by the industry.
FLABEL has brought clarity into how nutrition information on food labels affects consumer purchasing decisions. The project has also developed an instrument for policy assessment that:
- raises awareness of the challenges in assessing label effectiveness by highlighting the relevant consumer behaviour theories, research challenges and results from the project's work, and
- recommends a science-based but practical tool that can be used to assess the effects of changes to current nutrition label policy on consumer behaviour, based on the methodological experiences and research results from the FLABEL project's work.
Finally, even though the project's aim was not to provide an assessment of the impact on industrial competitiveness, FLABEL research showed this is an important concern in relation to food nutrition labels, and expert surveys are suggested as a means to assess the impact of food labelling on industrial competitiveness.
More information on the FLABEL project is available from the project website at http://www.flabel.org
Contact details of the project coordinator:
European Food Information Council (EUFIC)
Rue Paul-Emile Janson 6 - 1000 Brussels
Tel. +32-250-68989
Email: info@flabel.org