FOOD LABELLING ISSUES:
Quantitative Consumer survey on Allergen labelling: Benchmark 2003
Evaluation Report Series No. 7
February 2004
The survey was designed to assess the understanding and use of food label information in food selection decisions made by the main grocery buyer in a household when shopping for foods for consumption by those who are ‘at risk’ of adverse or allergic reactions to food.
- Helpful Hint:Opening or downloading a PDF file
Full Report [ pdf format 1609 kb ]
TABLE OF CONTENTS
4 Identification and diagnosis of the most serious allergy
5 Knowledge and information about the allergy
7 The use of food labels to manage the allergy
10 Conclusions and Recommendations
11 References and background information
APPENDIX [ pdf 738 kb ]~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The Quantitative Consumer Survey on Allergen Labelling provides baseline data on consumers’:
- levels of awareness and knowledge of the allergen labelling provisions for manufactured foods;
- ability to successfully identify foods containing pertinent allergens; and
- behaviours towards food selection for those ‘at risk’ of adverse reactions to food.
The survey was designed to assess the understanding and use of food label information in food selection decisions made by the main grocery buyer in a household when shopping for foods for consumption by those who are ‘at risk’ of adverse or allergic reactions to food. A targeted sample was obtained that was not intended to be necessarily representative of gender, age, education, or ethnicity of the general populations of Australia and New Zealand, or of the ‘at risk’ population themselves. For example, the higher population of female respondents reflects the current predominance of women as household managers and carers (SNZ 2001 1 ; ABS 1995).
The data obtained from this survey will be used by Food Standards Australia New Zealand (FSANZ) to assess how well the new allergen labelling standard is working in the two countries. This benchmark survey will also assist FSANZ through a future evaluation, to track whether allergen labelling requirements meet the stated objective of providing adequate information relating to food to enable consumers to make an informed decision.
Background
The Australia New Zealand Food Standards Code (‘the Code’) became fully enforceable from December 2002 after a two year transition period. It included new labelling requirements for manufactured foods sold in Australia and New Zealand. One of the requirements was that food labels are required to provide advice for consumers on the presence of certain substances in food. This can be in the form of mandatory warning statements; mandatory advisory statements; or mandatory declarations. These requirements apply when any of the substances listed in Standard 1.2.3 Mandatory Warning and Advisory Statements and Declarations are present in food as an ingredient, an ingredient of a compound ingredient, a food additive or component of a food additive, or a processing aid or a component of a processing aid (FSANZ 20021).
The mandatory declarations, warning statements and advisory statements are intended to provide consumers with sufficient information such that they could prevent adverse/allergic reactions to a food or an ingredient in food.
Another requirement was that the source of cereals and starch (e.g. wheat, rye, barley, oats) be declared on a food label, as is the source of vegetable oils if derived from peanut, soybean or sesame (Standard 1.2.4 Labelling of Ingredients). The food labelling standards contained in the Code cover both locally produced and foods imported into Australia and New Zealand.
The survey
The survey focussed on a selection of substances listed in Standard 1.2.3 Mandatory Warning and Advisory Statements and Declarations. These were wheat (gluten containing cereals and their products); eggs and egg products; fish and fish products; milk and milk products; nuts and sesame seeds (including their products); peanuts and soybeans (including their products) and added sulphites (in concentrations of 10mg/kg or more) (FSANZ 20021).
In this survey, a broad definition of food allergy was adopted by FSANZ to target households with members who were ‘at risk’ of adverse reactions to foods, thereby obtaining a wider cross-section of respondents who may be assessing food labels critically because of concerns about the presence of allergens. ‘Food allergy’ in this survey meant ‘a reaction due to exposure to a food or ingredient in food resulting in at least one of the following symptoms: difficulty in breathing or throat swelling; swelling or itching of lips or tongue; hives, skin rashes or eczema; stomach cramps, vomiting or diarrhoea; or faintness or collapse’. While this includes anaphylactic reactions to food, it is a wider definition.
It should be noted that this survey was not intended to measure the prevalence of food allergies in Australia or New Zealand, or to assess the consistency of food labels with allergen labelling requirements of the Code.
At the time of this survey (immediately after the end of the transition period), foods were still legally available for sale that had been manufactured and labelled according to the old food standards.
Methodology
About 1% of adults and less than 10% of children in the population have been estimated to be ‘at risk’ of allergic reactions to foods (Cohen 1999; Swain, Soutter & Lobley 1996). To effectively reach this target group in the population, recruitment was undertaken via three routes: immunology/allergy clinics in hospitals and medical institutions; private immunology/allergy clinics and allergy support groups, in both Australia and New Zealand. This methodology was adopted because it was considered unlikely that sufficient participants would have been recruited using random sampling of the general population due to the low proportion of consumers ‘at risk’ in the population.
Recruitment criteria were developed for use by medical specialists to ensure that both children and adults with different types of food allergies were contacted from all states, territories and regions with access to medical specialists. The three recruitment groups were controlled to ensure similar sampling proportions. When any difficulties were faced recruiting from a selected site, recruitment quotas were redistributed to other sites in the same recruitment group.
The benchmark survey was conducted using a mail-out self-completion questionnaire. It was sent to consumers who were identified by medical specialists as being ‘at risk’ of adverse/allergic reaction to certain foods or ingredients in foods or who were members of an allergy support group. The questionnaire included open and closed-ended questions to be answered primarily by the main grocery buyer in a household where there was a member 'at risk' of an adverse reaction to certain substances. Those 'at risk' who were under the age of 18 years could choose to complete the questionnaire provided that they obtained their parents’ / guardians’ consent, however none chose this option. For clarity, throughout this report participants are referred to as 'respondents' whether they were the ‘main grocery buyer’ or the person with the allergy.
In total, 1166 questionnaires were distributed via the three routes and 510 completed questionnaires were returned; 413 from Australia and 97 from New Zealand. The response rate for New Zealand was 40% and for Australia was 45%, with the overall response rate for the survey being 44%.
Results
1. Key findings
Four key issues emerged from this survey:
- respondents used food labels extensively in managing the allergies in their household. They also reported noticing changes in the way allergens were listed on food labels;
- the ability of respondents to identify food products that contained the allergen of concern varied considerably depending on the food allergy within the household and the terms used on the labels to declare the allergen;
- respondents had adopted strategies to manage their food selection when buying for ‘at risk’ consumers. For example, checking all product categories to avoid substances of concern, reading food labels carefully, buying the same brand that past experience has shown to be allergen free, and checking food lists provided by allergy support groups; and
- the labelling of foods was perceived to play a role, although not a major one, in the occurrence of a repeat allergic reaction since the allergy was first diagnosed, for example, as a result of an unlabelled or incorrectly labelled product (14% reported occurrences).
2. Summary of results
Awareness and knowledge about the allergen labelling requirements for manufactured foods
Ninety percent of the respondentsalwaysread food labels carefully, and around two thirds (67%) of respondents had already noticed some specific labelling changes. These changes included:
- greater use of 'may contain' statements;
- the use of bold print for allergens named in the ingredient list;
- more warnings about nuts;
- the use of 'made on same product line or equipment' statement;
- the use of blanket statements such as 'contains dairy / seeds / nuts …'; and
- greater use of 'made in same factory' or 'same premises' statements.
Significantly more Australians than New Zealanders had noticed specific changes, such as statements about nuts, products made on the same production equipment as nuts, and blanket statements of ‘contains dairy/nuts/eggs’ etc, whereas significantly more New Zealanders than Australians had noticed the use of bold print for allergens listed in ingredient lists.
Most respondents felt they could trust the information given on food labels. Respondents who were members of an allergy support group were significantly more likely than non-members to say they were not sure whether to trust labels. However, allergy support group members may be more critical of labelling information as they used this information more often. They were also significantly more likely to find some information, e.g. statement ‘may contain traces of…’, on food labels really useful or important, compared to non-members.
Respondents showed a clear preference for the disclaimer 'made on the same equipment as products containing …' compared to either 'may contain traces of …', and 'made in the same premises as products containing …'. However, over a third of the respondents said that the 'made on the same equipment ...' statement was not very useful.
Ability to successfully identify manufactured foods that contain allergens
The levels of awareness or ability to identify foods that contained allergens varied considerably. It depended to a large degree on the type of food allergy the respondent had to deal with, and the terms used to declare the substances of concern on the food labels. Those with tree nut, milk or egg allergies were most accurate in their assessment of ingredients whilst those with peanut and wheat allergies were less accurate.
Respondents indicated that the labelling issues that affected their ability to identify suitable foods to prevent allergic reactions were:
- lack of understanding on their part of some problematic ingredients. This was due to the complexity or lack of clarity of terms used on labels (e.g. substances like 'emulsifiers');
- the use of ‘may contain’ statements, which were perceived as being over-used or not clear, and had the effect of restricting the choice of foods that were available;
- the derivation of some ingredients in foods, e.g. the source of certain vegetable oils, was often not stated;
- issues that were the manufacturers’ responsibilities, such as unlabelled ingredients or changes to the ingredients in products without notice; and
- food labelling information, such as the location of the information on the labels or the belief that there was a difference in labelling requirements for imported foods.
Additionally, the data indicated that respondents who were members of an allergy support group were more proficient at identifying many of the substances of concern compared to non-members.
Existing behaviours of respondents in regard to food selection
Respondents reported on strategies to manage their food selection when buying for those ‘at risk’ of an adverse reaction. For example, 41% reported checking all product categories, e.g. sweet biscuits or pre-prepared meals, to avoid substances of concern. This was reported most by respondents with sesame, soy or milk allergies in their household, and least by those with shellfish, peanut, tree nut or sulphite allergies. Respondents shopping for children were also more likely to report this than those shopping for themselves.
Other strategies adopted includedalwaysreading food labels carefully (90%),alwaysbuying the same brand that past experience has shown to be allergen free (62%),neverbuying loose or bulk foods (43%),oftenbuying food in sealed packages (44%), andoftenchecking food lists provided by an allergy support group (24%).
The labelling of foods and the impact on the occurrence of adverse / allergic reaction
Forty-two percent of respondents reported that the person with the most serious food allergy within their household had had an adverse reaction after the diagnosis had first been made. The causes reported for this repeat reaction had been accidental consumption (36%), contact with the substance of concern (21%), unlabelled / incorrectly labelled food (14%) and traces of substances in unexpected foods (6%). Members of allergy support groups were significantly more likely (10%) than non-members (1%) to report that the reaction was caused by unexpected traces of substance in the product.
Most respondents with peanut, tree nut, shellfish or fish allergies said theynevereat the substance knowingly. In contrast, those with sulphite, soy or wheat allergies said they try but cannot avoid the substance completely.
Conclusions and recommendations
Clarity and understanding of allergen labelling were the two key aspects that were said to impact on the effectiveness of labels, by consumers shopping for foods for those ‘at risk’ of adverse or allergic reactions.
According to respondents, theclarityof labelling information could be enhanced by:
- adopting more meaningful or accurate labelling or ‘advisory' statements;
- ensuring that the origin or derivations of particular ingredients are stated;
- using uniform wording in plain English for allergens;
- using percentage labelling for allergens to indicate how much of the substance of concern is in the food to enable an assessment of risk; and
- further considering formatting issues, such as print size and standard placement fields on labels.
The second factor,understandingof allergen labelling, was clearly an issue for respondents. While ‘understanding’ is a difficult concept to measure, in this survey it was assessed by the respondents’ ability to use food labels to select appropriate foods, including their ability to identify different names for the substance of concern to them. The results of this survey indicated that understanding the information on food labels played a role in consumers’ ability to use food labels appropriately when selecting foods for those ‘at risk’ of adverse or allergic reactions.
This survey has found that the membership of an allergy support group assisted greatly in the respondents’ ability to identify the foods that were not suitable, and consequently to avoid foods that might otherwise trigger an allergic reaction. For example, members of a support group were more likely to contact the food manufacturer or the support group for advice on certain foods.
Clarity and understanding of food labelling information is not an issue confined to consumers shopping for the sub-population ‘at risk’ of food allergic reactions. The two key issues identified in this survey, clarity and understanding, are similar to the findings of ‘Quantitative research with consumers on food labelling issues’, a survey undertaken with the general population. That survey also indicated that consumers in Australia and New Zealand had difficulties using labelling information effectively to make informed food choices (FSANZ 2003).
This baseline survey measured the current situation with regard to the use of food labels amongst the target group of those purchasing food for a household in which there is someone 'at risk' of an adverse or allergic reaction to food. It could be argued that this particular target group is impacted most by changes to food labels. Therefore, it is important to track their progress as the changes to food labels are implemented more widely, and that their needs are considered in any further label changes.
It is recommended that as part of future evaluation activity a replicate survey be conducted in two to three years time, using the same methodology. Such a survey would enable FSANZ to track whether the allergen labelling provisions of the Code are meeting the desired objective of providing adequate information in relation to food, to enable consumers to make informed choices. If a survey is not possible, an alternative approach should be considered.
Top | Section 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | Appendixes