Controlling Food Allergens

The issue of food allergy has become an increasingly important one for the food industry over the last ten years. The prevalence of allergic reactions to certain foods seems to be increasing in the population and legislation has been introduced requiring food businesses to label foods containing major allergens. This has created a significant new challenge for food manufacturers to ensure that their products are accurately labelled and do not contain potentially dangerous undeclared allergens. While a range of control measures and analytical tests for allergens in food have been developed, large numbers of allergen-related product recalls suggest that they are less than perfect. But as research fills in some of the knowledge gaps, more effective control and testing strategies should soon become a reality.

What is food allergy?

True food allergies are adverse reactions to food that have an immunological basis, which distinguishes them from food intolerance and other reactions where the immune system is not involved. Most food allergies are mediated by immunoglobulin E (IgE), with the notable exception of gluten intolerance syndrome, or coeliac disease, which is caused by an immune response, but mediated through immunoglobulin A (IgA) and immunoglobulin G (IgG). Other types of food intolerance with no immune system involvement are adverse reactions to food components, such as amines or lactose, and are sometimes referred to as non-allergic food hypersensitivity reactions. For example, lactose intolerance occurs in individuals who do not have the necessary enzymes to break down lactose in the gut. Limiting the amount of a particular food eaten can sometimes control food intolerances effectively, but food allergy sufferers may need to avoid the relevant food completely. Only true food allergies can lead to the potentially fatal reaction of anaphylaxis.

The antigens responsible for most IgE mediated food allergies are proteins, and exposure to them can lead to the development of an allergy through a two-stage process. During the sensitisation stage, exposure to the protein antigen initiates a series of metabolic reactions resulting in the production of specific IgE, a type of antibody usually only produced in response to parasitic infections. During the second stage an allergic reaction is elicited when a sensitised individual is re-exposed to the allergen. The cell-bound IgE reacts with the antigen, causing the release of inflammatory mediators such as histamine from basophils or mast cells. The mediators are responsible for the rapid development of symptoms associated with allergy, which include shortness of breath, asthma, skin irritation and rashes, gastrointestinal upsets, hypertension and increased heart rate. In some cases the person affected may go into anaphylactic shock, which can prove fatal if not treated immediately.

Theoretically, people may develop allergies to almost any protein component in food, and the FDA has identified more than 160 foods that can cause allergic reactions in sensitive individuals. But just eight food types are responsible for most food allergies. These so-called ‘major allergens’ are: cow’s milk; eggs; fish; shellfish; peanuts; tree nuts; wheat; soya. Other food allergens are reported occasionally and the European Commission has compiled a list of 14, which must be clearly labelled on any foods that contain them. The list is as follows:

  • Cereals containing gluten
  • Crustaceans
  • Eggs
  • Fish
  • Peanuts
  • Soybeans
  • Milk
  • Nuts (i.e. almonds, hazelnuts, walnuts, cashews, pecan nuts, Brazil nuts, pistachio nuts, macadamia nuts and Queensland nuts)
  • Celery
  • Mustard
  • Sesame
  • Sulfur dioxide and sulfites
  • Lupin
  • Molluscs

A number of countries around the world have introduced legislation to regulate the labelling of foods containing known allergens to help consumers avoid products that might cause an allergic reaction. But the legislation varies considerably between different countries. Since December 2008 food businesses operating in the EU have been required to label pre-packed foods containing any of the 14 allergens listed above, while in the USA, only the eight major allergens need be labelled and Japan requires labelling for just five. Uncertainty over the threshold levels of allergens needed to cause a reaction and the great variation in sensitivity between individuals means that acceptable limits for allergens are not defined in legislation, resulting in a ‘zero tolerance’ approach.

How common is food allergy?

The overall and worldwide prevalence of IgE-mediated food allergies is uncertain, but is believed to be increasing. The best recent estimates are that about 1-2% of adults and 5-7% of children suffer from some type of food allergy. The prevalence is higher amongst children because sensitivity to some allergens, especially cows’ milk and egg, may be lost during late childhood. Unfortunately, more serious food allergies, notably sensitivity to peanuts, tree nuts and shellfish, are usually life-long. Food allergy prevalence also varies in different countries. For example, peanut allergy is particularly common in the United States, where peanut butter is very widely consumed, while mustard allergy is common in France.

The amount of food allergen required to elicit an allergic response varies greatly, both between individuals and between different allergens. There is some evidence of sensitised people reacting strongly to very small (measured in micrograms) ingested doses, or even to traces of allergen on tableware, but much of this is anecdotal. Data from double-blind, placebo-controlled food challenges suggests that there are threshold doses for food allergens, below which sensitised people do not react. For peanut allergy, this threshold, or lowest observed adverse effect level (LOAEL) has been found to range from 2mg to more than 50mg. Nevertheless, there is a distinct lack of reliable data on LOAEL doses for major allergens and people with food allergies are often advised to avoid the offending food completely. Peanut allergy is seen as a particularly serious problem for the food industry. This is partly because some sensitive individuals may suffer serious, even fatal, anaphylactic shock, but also because peanut products are very common ingredients in many processed foods.

Problems facing the food industry

It is clear that allergens have become an important food safety issue for responsible manufacturers. As such, they should be controlled just as microbial and chemical contamination is controlled in food processing operations. Complying with current labelling legislation requires measures to ensure that foods not labelled with allergen declarations do not contain food allergens introduced accidentally by cross contamination – so-called ‘cross-contact’ allergens. Indeed, most authorities recommend an approach to allergen control based on the HACCP principles that have been so successful in managing other food safety hazards. For example, guidance from the UK-based Institute of Food Science and Technology (IFST) has advised that the following strategies should be adopted.

  • Implementation of a HACCP plan to analyse the entire manufacturing process in relation to allergen hazards.
  • In a multi-product company, wherever possible, segregate manufacturing operations involving the allergen-containing food into a separate building.
  • When possible, formulate foods that are free of all unnecessary major allergens as ingredients.
  • Organise raw materials supplies, storage and handling, production schedules and cleaning procedures to prevent cross-contamination of products by ‘foreign’ allergens.
  • Ensure all personnel are fully trained to understand the necessary measures and the reasons for them.
  • Comply with the relevant labelling legislation, ensuring that appropriate warnings are included on the product label warning the consumer of the presence of a major allergen.
  • Have in place an appropriate recall system for any product found to contain a major allergen not indicated on the product label.

The consensus is that adhering strictly to GMP procedures can avoid problems arising from misformulation, cross contamination, inadequate cleaning and re-work. Using separate equipment for products containing specific allergens is often recommended, while some larger manufacturers have chosen to set up separate manufacturing facilities for products containing allergens to eliminate the risk of cross contaminating other products.

Implementing these strategies will generally require drawing up an Allergen Control Plan, which will focus on such issues as identifying and tracking ingredients that contain allergens using an allergen map, dedication of processing equipment to allergenic products and strict control of rework and packaging. Effective cleaning and decontamination is also crucial where production facilities are shared between products containing allergens and those considered allergen-free. Cleaning should be validated by sampling and testing for allergen residues on equipment and surfaces. Guidance on the components needed for effective Allergen Control Plans is available from several bodies, including the US-based Food Allergy and Anaphylaxis Network (

Yet despite all the advice available and the adoption of the proven HACCP-based approach to controlling allergen contamination, undeclared allergens in food remains one of the main reasons for recalling products from the market. For example, in 2010 the FDA recorded over 120 recall incidents related to food allergens, while in the UK in 2011, the Food Standards Agency issued 50 Food Alerts, but no less than 57 Allergy Alerts. This strongly suggests that some manufacturers are struggling to control allergens, also leading to an increase in precautionary labelling and the use of ‘may contain’ warnings on many foods. Such an approach is meant for use in situations where the absence of allergens cannot be guaranteed, but it has been suggested that some manufacturers use ‘may contain’ labelling as a substitute for effective allergen control. Widespread use of precautionary labelling also presents problems for allergic consumers, who may find their choice of products severely limited and may be tempted to ignore valid warnings.

Assessing risk

It seems clear that the current approach to allergen control is not working as well as it should and some experts have suggested that a different approach may be needed. While a HACCP-based system may work well for other food safety hazards, there are some aspects of allergen control that demand a change of emphasis. A closer look at the lists of allergy alerts and recalls in the USA and Europe immediately reveals that most of them are the result of incorrect packaging or mislabelling. It appears that putting the wrong product in the right pack, or vice versa, is one of the biggest causes of allergy alerts and expensive product recalls. Adventitious contamination seems to be a relatively infrequent problem, yet the recommended control measures for allergens are focused on preventing cross contamination. Furthermore, a significant number of recalls are carried out on the basis that the absence of an undeclared allergen in the product cannot be ruled out, leading to a precautionary recall. The risk of any consumers suffering an adverse reaction may be very small in some of these cases and it could be argued that a proportion of the recalls undertaken are unnecessary.

This suggests that a more risk-based approach to controlling food allergens may be more effective. Conventional HACCP tends to treat all hazards equally, without making any attempt to quantify the risk to the consumer. This may not be appropriate for food allergens. For example, few would disagree that the risk presented by a misformulation resulting in undeclared peanuts being included as an ingredient is far greater than the theoretical possibility of a trace amount of lupin being present following cross contamination. Yet HACCP principles may demand that equal effort is devoted to preventing both eventualities. If allergen risks could be ranked effectively, it would surely help manufacturers focus their resources on preventing the more serious failures.

The dose-response question

Unfortunately, the idea of quantifying risk brings us back to a fundamental gap in the scientific knowledge about food allergens, namely the absence of any agreement on threshold levels. Without sound data on the amount of allergen needed to cause an adverse reaction, risk assessment is subject to severe limitations. Furthermore, conventional HACCP is hampered by the difficulty of setting critical limits at CCPs, resulting in a ‘zero tolerance’ requirement, which is virtually impossible to achieve for manufacturers without separated production facilities. This is one reason why a considerable amount of research effort is currently being targeted at determining reliable threshold data for major allergens. Both EFSA and the FDA have set up working groups to look at this question.

Although it does seem that it is possible to measure LOAELs for allergens, the methods used in past clinical trials have not been standardised and both the EFSA and FDA groups have concluded that the clinical data available so far is not sufficient to establish valid threshold values. This problem has been addressed by a number of groups, notably the US-based Food Allergy Research and Resource Programme (FARRP) and the EU-funded EUROPREVALL project, which was completed in late 2009. Nevertheless, there are still no internationally agreed threshold levels for food allergens that the food industry can use in control plans. Despite this, an Australian food industry trade organisation (The Australian Food and Grocery Council) has gone ahead with the publication of ‘action levels’ for food allergens. An industry working group called the Allergen Bureau was set up in 2005 to develop allergen control tools for industry. It came up with the Voluntary Incidental Trace Labelling (VITAL) system, which includes much of the advice common to industry guidance elsewhere, but also includes action levels to inform labelling for cross-contact allergens. The VITAL action level grid contains two action levels. Action Level 1 is a value below which a precautionary statement is not required, and Action Level 2 is the value above which a ‘may be present’ warning is required. The grid currently covers 24 individual allergens and action levels vary from 6 ppm for egg to 800 ppm for lupin. If a similar system could be agreed worldwide on the basis of sound clinical data, much precautionary labelling could be dispensed with, providing benefits for consumers and manufacturers alike.

Allergen testing methods

The lack of threshold levels is also a major problem for allergen testing since lower limits of detection will always apply. Agreement on how much allergen is acceptable in food would allow test methods to be standardised in terms of sensitivity. This would be very helpful to manufacturers because testing is of great value in validating allergen control plans.

Test methods for gluten were first commercialised in the early 1990s and Neogen launched an ELISA-based test kit for peanut in 1996. In 2012 a large number of allergen test kits utilising various technologies are available from numerous suppliers. Many are immunological tests, typically using an ELISA, or lateral flow ‘dipstick’ format. Kits for at least 13 allergens are currently available with more in development and suppliers include Neogen, Thermo Fisher Scientific and R-Biopharm. There are also PCR-based test kits and general protein tests available. However there are some problems associated with commercial test kits. Few are validated for detecting allergens at the ppm level and some are badly affected by the masking effect of the food matrix. In addition, allergens may still be present in some highly processed ingredients, especially peanuts, but may not be detectable by existing test methods.

Choice of test method is very important and will depend on exactly what the test is meant to achieve and on the nature of the food matrix. For example, general protein test kits may be useful to validate cleaning, but have low sensitivity and do not detect specific allergenic proteins. PCR on the other hand is highly sensitive, but detects DNA from the source of the allergen rather than the protein itself, and cannot be used on-site. ELISA and lateral flow test kits are rapid and much better suited to use outside the lab, but vary in their specificity and sensitivity and may not be quantitative. Immunological methods are currently preferred for validating allergen control plans and cleaning. The question of harmonised methodology is being addressed by FARRP, a working group of the European Committee for Standardisation (CEN) and by an EU-funded project called MoniQA.

Although considerable progress has been made in food allergen control over the last decade, there is still some way to go before manufacturers have all the tools available to protect sensitive consumers without recourse to precautionary labelling. But if the question of threshold levels can be resolved, then progress will be rapid, both in risk-based practical control measures and test methodology.

Notify of
Inline Feedbacks
View all comments