Rapport Etude DGAL CLCV 2006 synthese anglaise

Ncvt

Description

CLCV FRENCH REPUBLIC [French Consumer Defence Association] MINISTRY OF AGRICULTURE AND FISHERIES STUDY OF NUTRITION LABELLING REQUIREMENTS IN RELATION TO NUTRITION POLICY SUMMARY September 2006 1/12 Introduction and issues One of the recitals of the European directive on nutrition labelling notes that “there is growing public interest in the relationship between diet and health and in the choice of an appropriate diet for individual needs”. It is also stated that “knowledge of the basic principles of nutrition and appropriate nutrition labelling of foodstuffs would contribute significantly towards enabling the consumer to make this choice”. In France as elsewhere in Europe, nutrition-related public health issues have assumed increasing importance in recent years. Against this background, it has become a legal requirement for nutrition labelling to perform a consumer information function. In particular, labels must make it easier for consumers to take account of nutrition and health concerns when choosing food. Unfortunately, however, this objective has manifestly not so far been achieved. Nutrition labelling still uses highly technical language and the complexity of the information given is further compounded by the multiplicity of different formats and underlying logical systems. Yet the last few months have witnessed the beginning of a debate at both ...

Informations

• Publié par: Nethug
• Nombre de lectures: 11
• Langue: English

Extrait

CLCV [French Consumer Defence Association]

FRENCH REPUBLIC

MINISTRY OF AGRICULTURE AND FISHERIES

STUDY OF NUTRITION LABELLING REQUIREMENTS IN RELATION TO NUTRITION POLICY SUMMARY

September 2006

1/12

Introduction and issues One of the recitals of the European directive on nutrition labelling notes that “there is growing public interest in the relationship between diet and health and in the choice of an appropriate diet for individual needs”. It is also stated that “knowledge of the basic principles of nutrition and appropriate nutrition labelling of foodstuffs would contribute significantly towards enabling the consumer to make this choice”. In France as elsewhere in Europe, nutrition-related public health issues have assumed increasing importance in recent years. Against this background, it has become a legal requirement for nutrition labelling to perform a consumer information function. In particular, labels must make it easier for consumers to take account of nutrition and health concerns when choosing food. Unfortunately, however, this objective has manifestly not so far been achieved. Nutrition labelling still uses highly technical language and the complexity of the information given is further compounded by the multiplicity of different formats and underlying logical systems. Yet the last few months have witnessed the beginning of a debate at both national and European level on how the current nutrition labelling provisions can be improved. This study belongs in the context of these considerations. It aspires in particular to provide concrete and reliable information on French consumers’ perceptions and comprehension of the various forms of innovative labelling. It therefore makes a substantive contribution to the drafting of realistic proposals that will conform to consumers’ needs and expectations.

2/12

1 Specific study objectives The main objectives of the study are set out below. 1.1 Obtaining representative and significant results In order to obtain valid and reliable results, it was of course of paramount importance to ensure that the sample was representative. The minimum sample size chosen was therefore 1200 persons, divided into two sub-samples each of 600 (see Section 2.3). Furthermore, to guarantee as accurate as possible a match between the sample structure and that of the population, the “quota method” was used. The INSEE [National Statistics and Economic Studies Institute] data from the 1999 census were used as a benchmark for criteria such as age, sex, training and education level and socio-occupational category. 1.2 Characterization of consumer expectations with a view to their improved satisfaction For the purposes of reforming the relevant legislation, it is of course essential to determine the nature of consumer expectations in the field of nutrition labelling and to propose an acceptable system that will meet with the approval of as many consumers as possible. The present system is too technical and does not satisfy consumer expectations and needs; these will have to be allowed for to a much greater extent in the innovations adopted. In addition to expectations concerning the type of labelling, the study was required to identify consumers’ wishes as to its implementation: Should it be stipulated for all prepacked products? Where should it appear? And so on. 1.3 Measuring comprehension of innovative forms of nutrition labelling Given that manufacturers have devised many different forms of nutrition labelling, some of which are felt to be potentially suitable as a basis for establishing a possible harmonized European system, it was essential to evaluate their comprehension by consumers. When the study objectives were set, it was decided to test consumers’ actual comprehension rather than relying on “declared” comprehension only. Both an initial evaluation and an evaluation after explanation were included, so as to provide a more subtle indication of comprehension and the need for learning. The study also includes an evaluation of consumers’ comprehension of the underlying logic of the labels – that is, the nature of the information imparted, rather than formal aspects such as the meaning of colour coding or pictograms. The CNC [National Consumer Council] working party called these two facets of labelling “mechanics” and “bodywork” respectively. This study concentrated precisely on evaluating comprehension of the “mechanics”, while as far as possible eliminating all bias due to visual attractiveness.

3/12

Much of the current diversity of nutrition labelling is attributable to the use of a wide variety of ways of expressing information. The CNC working party was nevertheless able to suggest a three-category classification of label types based on their objectives. To allow for this diversity, it was decided to test three different systems and to compare them with the current one (see Section 2.1). 1.4 Comparison of proposed and current label types The current form of labelling can serve as a benchmark for assessing the improvement that new systems might afford, as respondents needed to be able easily to compare the two options – the current and the proposed label types – and to express a preference for one or the other. 1.5 Tracking the variation of certain parameters from the 2004 study The 2004 study by the CLCV [French Consumer Defence Association] and the DGAL [Directorate-General for Food] provided information on a number of aspects of nutrition- labelling reading habits both in-store and in the home. It also allowed consumers to express declared choices as a measure of their interest in nutritional issues and to test their knowledge in this field. In addition, the combination of these criteria yielded a basic typology of consumers. The new survey offered an opportunity of confirming the 2004 data and of identifying any changes. It was therefore essential to link the two studies. 1.6 Correlating the results with consumers’ socioeconomic data and profiles Not only nutritional knowledge but also nutritional status show a high degree of variation in the population and both are closely bound up with socioeconomic variables. The aim of nutrition labelling being to supply information to the largest possible number of people, it was manifestly essential not to confine the study to overall results. Instead, we sought to highlight any links between response variability and disparities in living standards, age, training and education, socio-occupational category and other parameters.

4/12

2 Methodological options The methodology adopted had to take account not only of the above objectives but also of the constraints inherent in surveys of this type – in particular, the need to keep questionnaires to a reasonable length. The main choices concerned the systems to be tested, the forms of administering the questionnaires, the compilation of the survey material and the statistical processing of the results. Many of the relevant decisions were based on the results of a preliminary test conducted with 17 subjects by the CLCV team in Clermont-Ferrand. This trial run identified a number of flaws in the first version of the questionnaire – specifically, excessive length and complexity. 2.1 Choice of systems for testing Any aspiration to exhaustiveness in this field would have been unrealistic, and so four types of labelling were ultimately chosen in order as far as possible to allow for the existing diversity. The Gerbaulet system The Gerbaulet system, named after the French nutritionist who devised it, gives information on the amounts of five categories of foods and of sugar, salt and energy supplied by one portion of a product compared with the daily requirement. In the following example, one portion of the relevant product supplies a quarter of the daily requirement in the category “Meat, eggs and fish”. Relative to your daily requirement, one portion gives you:

5/12

Energie = Energy Céréales et autres féculents = Cereals and other starchy foods Fruits et légumes = Fruit and vegetables Viandes/Oeufs/Poissons = Meat, eggs and fish Produits laitiers = Dairy products Matières grasses = Fat Sucre = Sugar Sel = Salt Note that a common scale is not used to present the requirement of each item: one box represents the equivalent of one standard portion of the relevant food category (one item of fruit, one knob of butter, etc.), while the number of boxes varies according to the food category concerned. The information supplied by the Gerbaulet system is derived from a mathematical model for determining the nutritional composition (carbohydrates, fat, proteins, etc.) of each category of foods. A similar approach was applied for a few years by Danone. It now seems to be no longer in use on any product. This system, which is of relatively complex construction, is intended to supply basically simple and intuitive information to the consumer, as it avoids technical terms and is based instead on a widespread dietary culture. The Nutrients system This system too is based on the parameter of recommended daily allowances (RDA), but the information is given in terms of nutrients rather than food categories. In the following example, one portion of the product is seen to cover one eighth of the complex carbohydrate requirement. Relative to your daily requirement, one portion gives you:

6/12

Maxi = 4 carrés / jour = Maximum: 4 boxes/day Energie = Energy Protéines = Proteins Glucides complexes = Complex carbohydrates Sucres simples = Simple sugars Matières grasses = Fat Sel = Salt To simplify interpretation, the requirement for all items is expressed in the form of four boxes per day. Many foodstuffs are already labelled in this way, although a wide variety of formats are used. For instance, the boxes may be replaced by continuous histograms, sometimes also giving percentages. The colours, the nutrients represented, the terms and other features may also vary. The system proposed here features six items (energy and five nutrients). For simplicity, a smaller number could have been used. However, for the sake of symmetry with the Gerbaulet system, which cannot be further reduced, it was decided to retain this number. Note that a simplified terminology is used here for the nutrients: fat rather than lipids; simple sugars rather than carbohydrates; and salt rather than sodium. The CNC working party on nutrition suggests that nutrition labelling systems should be classified in three categories: factual, comparative labelling; choice-facilitating labelling; and choice-guiding labelling. The Nutrients and Gerbaulet systems constitute choice-facilitating labelling: they do not supply crude information (e.g. quantities in grams), but relate the information to consumption recommendations (RDAs). On the other hand, they do not offer a judgement on the nutritional value of a product. Traffic light systems These systems were first proposed by the United Kingdom Food Standards Agency (FSA). They are based on a score for the overall nutritional value of a product. The score is conventionally obtained from the following criteria: ·content of four “negative” nutrients: energy, saturated fats, non-dairy simple sugars and salt; ·content of three “positive” nutrients (calcium, iron and long-chain polyunsaturated fats) and fruit and vegetable content.

7/12

Depending on the result, the product carries a red, amber or green light symbol accompanied by an explanatory text such as “Eat as much as you like”, “Eat in moderation” or “Eat seldom”. Products containing high levels of fat or sugar, such as chocolate, will thus be indicated by a red light signal. The study included a test of red and green traffic light signposting as a possible addition to the Gerbaulet and Nutrients systems or to the present labelling system:

Manger à volonté = Eat as much as you like Manger rarement = Eat seldom In the classification of the CNC working party on nutrition labelling, these are consumer choice-guiding systems. Multiple traffic lights In this system, Gerbaulet or Nutrients labels are supplemented by symbols indicating whether the amounts supplied by one portion of the product are or are not particularly high for each characteristic covered. This type of labelling is inspired by the Food Standards Agency’s “multiple traffic lights” proposal. Relative to your daily requirement, one portion gives you: Relative to your daily requirement, one portion gives you:

8/12

Maxi = 4 carrés / jour = Maximum: 4 boxes/day Energie = Energy Protéines = Proteins Glucides complexes = Complex carbohydrates Sucres simples = Simple sugars Matières grasses = Fat Sel = Salt Energie = Energy Céréales et autres féculents = Cereals and other starchy foods Fruits et légumes = Fruit and vegetables Viandes/Oeufs/Poissons = Meat, eggs and fish Produits laitiers = Dairy products Matières grasses = Fat Sucre, sucreries = Sugar Sel = Salt Comparison with the current system Note that the Nutrients and Gerbaulet systems were tested in comparison with the current system of nutrition labelling: Per 100 g Energy (kcal/kJ) 150/630 Proteins (g) 6.5 Carbohydrates (g) 12.5 - of which simple sugars (g) 2 Fat (g) 8.5 Salt (mg) 1032 The current labelling system is of the factual type, without interpretation of the information. Overall, the choice of systems, when combined with a comparison with the current system, effectively reflected the diversity of approaches to nutrition labelling – in particular, the main approaches contemplated at European level and the most innovative approaches. Note that the Nutrients and Gerbaulet systems are very similar in presentation: given the objective of testing and comparing actual comprehension of the two label types, it was important to minimize any possible bias due to appreciable differences in format.

9/12

2.2 Individual consumer interviews Individual interviews soon proved to be the only viable option for a valid survey result: telephone, Internet or group-based surveys are not satisfactory alternatives owing to the possible biases thereby introduced. Again, the physical presence of an interviewer was found to be essential for managing the interviews (to read out the questions, to record the responses and to present the illustrations in the intended order) and in particular for the purpose of explaining the Nutrients and Gerbaulet labelling systems. The CLCV provided specific training on the various aspects for the interviewers, who were issued with a methodology guide at the end of this workshop. The surveys took the specific form of quarter-hour individual interviews. It proved impossible to reduce this time, mainly because of the need to explain the proposed systems. The interviews were conducted by local branches of the CLCV at duty desks and various events attended by large numbers of people. To guarantee that the data were sufficiently representative, the local branches were assigned quotas based on extended regions and the interviewers took care to respect the sample structure at local level as their work proceeded. 3 Results The study results must be interpreted with due regard for certain unavoidable biases and limitations: ·The sample was not entirely representative, although the correction of the variable “degrees and diplomas” allowed substantial compensation of the discrepancy. ·The questionnaires were still relatively long and complex, thus possibly calling into question the reliability of the responses to the final questions. N.B. The questionnaires were designed by a group-based approach with a view to concentration on essentials and underwent many corrections. Given the study objectives, it would therefore surely have been difficult to arrive at a shorter version. ·In view of the study constraints (available time, standardization of interview wording, etc.), the explanations given of the various systems may not have been perfect. ·The data concerning the impact of food labelling on individual behaviour are based on declared actions and intentions and do not admit of unequivocal conclusions as to possible labelling-related changes in dietary habits.

10/1 2

Subject to the above reservations, the following conclusions can be drawn: 1. A consumer typology based on interest in and knowledge of nutrition was established and, when correlated with socioeconomic data, shows thatwomen, persons with children and/or persons of high educational level are generally more interested in nutrition. 2.Consumers basically want nutrition labelling that gives factual product information.is an overall preference for this information rather than There recommendations on consumption frequency. 3.Only 30-35% of the population immediately comprehend the notion of recommended daily allowancesbased on nutrients or food categories. 4.After a brief explanation this proportion increases to 47 or 48%on depending system type.This demonstrates the significance of explanation and argues in favour of a campaign of consumer education to accompany any new system. 5. After explanation, nutrient-based labelling is better comprehended and is perceived as simpler than a system based on food categories. 6. These two forms of labelling are comprehended better by young people and persons of relatively high educational level. Hence the accompanying campaign of consumer education mentioned in item 4 should be targeted at older and/or relatively unqualified people. 7. The two forms of labelling enable nearly 85% of consumers to estimate correctly the daily intake resulting from the consumption of various products. 8. The proposed systems are preferred to the current system by a significant margin.Some 58% of consumers think they make it easier to compare products, while 76% feel that these forms of labelling, whether by themselves or as an addition to the existing system, would help them to achieve a more balanced diet. 9. Over 90% of consumers want nutrition labelling to be made compulsory on prepacked products.might be expected, those with an interest in nutrition are As even more unequivocally in favour of this measure. 10. A majority (over 49%) of respondents would like to see this form of labelling on the front of packs. 11. More than 54% of consumers do not want a logo indicating products of high nutritional value.the other hand, persons of low educational status particularly On favour such logos. 12. The vast majority (82%) of consumers are opposed to logos indicating products of low nutritional value.

11/1 2