Food Allergens: assessing the risk to consumers – an update

Posted: 9 December 2015 | René Crevel, Unilever Safety & Environmental Assurance Centre, Bedford, UK | No comments yet

Managing allergens in a manner that minimises the risk to allergic consumers while maximising their choices, as well as ensuring the overall sustainability of food production processes remains the subject of considerable work. The last three years have witnessed developments in areas as diverse as regulation, hazard characterisation and risk assessment methodologies, as well as understanding of the behaviours of allergic consumers faced with allergen information that is difficult to interpret. Benchmarks for the application of precautionary allergen labelling, which will address these issues, continue to be a particular focus, with all stakeholders hoping for a quick resolution given the abundance of the emerging data. This will require not only consideration of the actual values of reference doses (management thresholds), but also the consequences for the overall usage of PAL and therefore its credibility. Since zero risk is not possible, the goal should be to minimise the number and severity of the reactions that any given choice entails…

Food Allergens: assessing the risk to consumers

Considerable progress has been made over the last 10 to 15 years in assessing the risk to consumers from unintentional exposure to food allergens1, yet they remain a major source of food safety incidents and institution of generally accepted management thresholds remains unresolved. Nevertheless in the last two to three years, much activity has taken place. This activity encompasses developments in the regulatory and risk assessment arenas, but also in the implementation of novel approaches to risk characterisation and generation of new data. Different stakeholders have also raised the profile of food allergy as an issue and demanded that it be addressed2. This paper reviews those developments and draws conclusions about the future trajectory towards benchmarks for allergen management which fulfil the needs of all stakeholders. Discussion will focus largely on developments within Europe, with references to other countries and regions where appropriate.


On 13th December 2014, the Food Information for Consumers Regulation (1169/2011/EU)3 came fully into force across the European Union and European Economic Area. It mandates several major changes to the way that regulated food allergens are handled in terms of the information provided to consumers. These concern labelling formats and the requirement to inform consumers of non-prepacked foods of the presence of any regulated allergen in such products. However, another part of the Regulation (Article 36) contains provisions that will profoundly affect the management of precautionary allergen labelling (PAL). While their form remains uncertain, they could be used to address the issues raised by stakeholders. Specifically Article 36 requires the Commission to adopt implementing acts relating to certain types of voluntary food information,

in particular

  • information on the possible and unintentional presence in food of substances or products causing allergies or intolerances

Those implementing acts relate to how this information is provided and its attributes (clarity, accuracy, evidence-based) and could serve as the basis of a framework for a harmonised approach to PAL. The Regulation does not specifically require the definition of quantitative benchmarks (reference doses/thresholds). However these are seen by most stakeholders as indispensable to a robust framework. The European Food Safety Authority (EFSA) has a critical role to play in providing the scientific basis for quantitative benchmarks to help the Commission, as the risk managers, to define appropriate ones. In late November 2014, EFSA published its updated scientific opinion on allergens4, based on a mandate by the Food Safety Authority of Ireland, which expressed concerns about the proliferation of PAL. Many stakeholders had hoped that this opinion would include a view on quantitative thresholds for implementation of PAL, which would also have helped the Commission with their implementing acts. However, in a detailed analysis5 EFSA concluded that this pertained to risk management and was therefore outside their mandate.

Hazard characterisation

Dose-distribution modelling, as discussed previously6 has made possible great progress in characterising the risk from food allergens. However, the models currently only permit a calculation of the likely number of reactions, without any indication of their nature in terms of the symptoms provoked, a matter of critical importance in characterising the hazard. Dose distribution models also rely largely, if not completely, on data from patients receiving oral food challenges in tertiary care facilities after pre-selection and exclusion of certain individuals. Furthermore, calculations based on incidence of reported reactions suggest that models may significantly overestimate the number potentially affected. The question thus also arose whether the same proportions of reactors would be observed in a less selected population for any given dose. A new type of study was proposed to address these issues, the single dose challenge7. The concept was to select a single dose and offer an open challenge to all qualifying patients attending a clinic. The proportion reacting would then indicate whether the selected dose would truly provoke the expected number of reactions and was therefore valid as the basis for deriving a quantitative benchmark. In addition, the pattern of symptoms observed during these challenges would provide valuable information on severity, thereby helping to characterise further the response to the tested allergens. Severity has long been acknowledged as a critical component of how tolerable a risk is8 and therefore also a critical dimension of the risk assessment9. The experimental phase of a study on peanut using the ED05 (proportion expected to cause reactions in 5% of peanut-allergic individuals) of 6mg total peanut in a cookie matrix and involving three clinics (Ireland, USA, Australia) and up to 375 patients has now been completed (Table 1, page 00) and results are being analysed. They are expected to be presented in the early part of 2016. Similar studies are included in the EU project iFAAM (reference) on hazelnut, milk and egg, all of them beginning to fill an important data gap.

New data

Major studies have been published, continuing to enhance the volume of food challenge data for dose-distribution modelling10,11, with ED values for the same allergens well within one order of magnitude of each other or less, despite the populations being drawn from different regions and a different statistical model being used (Tables 2,3, pages 00 and 00). The EuroPrevall study 11 in particular brings additional data on fish and celery for which insufficient data existed up to now for dose-distribution modelling12. It also confirms the diversity of food allergen sensitisation and reactivity across Europe.

Evolution of risk assessment

Good knowledge of the supply chain and specifically of what allergens may be present through commingling and other routes forms the basis of managing the risks they pose. Ever more sensitive analytical tests highlight the urgent need to move away from the ‘zero tolerance’ (although not zero risk) approach that some authorities consider maximally protective of allergic consumers, but which also drive ever-increasing levels of PAL, to the detriment of those same consumers. North America and Europe experienced a series of alerts relating to the presence of varying concentrations of unintended allergenic constituents such as almond and peanut in spices. Spices, by their very nature, are used in rather small amounts in prepared products. It follows that even comparatively large concentrations of unintended allergen could still be compatible with a final product that posed a negligible risk to most, if not all allergic consumers potentially reactive to that allergen. The case of peanut in cumin and almond in paprika led to the first public evaluation of the VITAL 2.0 system and its reference doses as the basis for decisions on PAL. The French agency responsible for food safety, ANSES, using data shared by the UK FSA, concluded that the VITAL 2.0 reference doses for peanut (0.2mg) and hazelnut (0.1mg) (which are both based on the ED01, so inherently protective of 99% of potentially susceptible individuals) were adequately protective of the at risk allergic population. They used them to define upper limits which would be considered not to pose a risk to public health for those allergens in different spices, based on their usage and therefore the amounts consumed13. Other bodies concerned with regulation and enforcement, such as the German ALS/ALTS, a working group of representatives of the food control authorities in the different German federal states, also examined VITAL 2.0 reference doses and adopted them as the benchmarks they would use in their enforcement activities14.

PAL and stakeholder views

As discussed before1-6, the value of PAL hinges upon its credibility among those it is intended to protect and those who advise them. This understanding is a principal driver for advocating a harmonised framework for its application, including generally accepted benchmarks, such as those of the VITAL 2.0 system. Evidence that this understanding is growing among the different groups of stakeholders continues to emerge9-15.

Concluding remarks

The volume, quality and types of data upon which quantitative benchmarks for managing allergen risks are based continue to grow rapidly. Arguably these data are already sufficient to build a framework upon which all stakeholders can agree. This framework would bring the transparency which is urgently needed, but still lacking. Of course such a framework would also be evolving. It has become clear that deriving the benchmarks not only requires consideration of the actual values of reference doses (management thresholds), based upon the degree of protection they confer and the likely health impact of the choices, but also the consequences for the overall usage of PAL and therefore its credibility. Since zero risk is not possible, the goal should be to minimise the number and severity of the reactions that any given choice entails.


  1. Crevel RW, Baumert JL, Baka A, Houben GF, Knulst AC, Kruizinga AG et al. Development and evolution of risk assessment for food allergens. Food Chem Toxicol 2014; 67:262-76.
  2. European Federation of Allergy and Airways Diseases Patients’ Associations (EFA). Report on the EFA Event “Eating safely: European best practices on allergens labelling”. . 2014.
  3. European Union. Regulation (EU) No 1169/2011 of the European Parliament and of the Council of 25 October 2011. O J European Union 2011; 54:L304-18-L304/63.
  4. Scientific Opinion on the evaluation of allergenic foods for labelling purposes. EFSA Journal 2014; 12(11):3894, 286pp.
  5. Outcome of a public consultation on the draft Scientific Opinion of the EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA) on the evaluation of allergenic foods and food ingredients for labelling purposes. EFSA supporting publication 2014:EN-696.202 pp. 2014.
  6. Crevel R. Food Allergy: identifying thresholds and assessing the risk to consumers. New Food 2013; 16(2):60-3.
  7. Zurzolo GA, Allen KJ, Taylor SL, Shreffler WG, Baumert JL, Tang ML et al. Peanut Allergen Threshold Study (PATS): validation of eliciting doses using a novel single-dose challenge protocol. Allergy Asthma Clin Immunol 2013; 9(1):35.
  8. Madsen CB, Hattersley S, Buck J, Gendel SM, Houben GF, Hourihane JO et al. Approaches to risk assessment in food allergy: report from a workshop ”developing a framework for assessing the risk from allergenic foods”. Food Chem Toxicol 2009; 47(2):480-9.
  9. DunnGalvin A, Chun-Han C, Crevel R, Grimshaw K, Poms R, Schnadt S et al. Precautionary allergen labelling: perspectives from key stakeholder groups. Allergy 2015; 70:1039-51.
  10. Blom WM, Vlieg-Boerstra BJ, Kruizinga AG, van der Heide S, Houben GF, Dubois AE. Threshold dose distributions for 5 major allergenic foods in children. Journal of Allergy and Clinical Immunology 2013; 131:172-9.
  11. Ballmer-Weber BK, Fernandez-Rivas M, Beyer K, Defernez M, Sperrin M, Mackie AR et al. How much is too much? Threshold dose distributions for 5 food allergens. J Allergy Clin Immunol 2015; 135(4):964-71.
  12. Taylor SL, Baumert JL, Kruizinga AG, Remington BC, Crevel RW, Brooke-Taylor S et al. Establishment of Reference Doses for residues of allergenic foods: report of the VITAL Expert Panel. Food Chem Toxicol 2014; 63:9-17.
  13. Technical and scientific support regarding the presence of almond and peanut allergens in spices. . 2015.
  14. ALTS/ALS. Arbeitskreis der auf dem Gebiet der Lebensmittelhygiene und der Lebensmittel tierischer Herkunft tätigen Sachverständigen (ALTS).
  15. Turner PJ, Allen KJ, Mehr S, Campbell DE. Knowledge, practice and views on precautionary allergen labelling for the management of patients with IgE-mediated food allergy – a survey of Australasian and UK health care professionals. J Allergy Clin Immunol.Pract . 2015.

About the author

René Crevel works as a Science Leader at Unilever’s Safety and Environmental Assurance Centre at Colworth House, Bedfordshire. He qualified initially in mammalian physiology and has postgraduate qualifications in Immunology and Toxicology. He is responsible for advice and guidance on food allergy and allergen risk assessment and management to Unilever Companies, and for leading Unilever’s food allergy research programme. He has authored papers and book chapters on various aspects of food allergy, including risk assessment and management of food allergens. He also currently chairs ILSI-Europe’s Food Allergy Task Force and is a member of the UK’s Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment.

Related topics

Send this to a friend