In managing food allergy there are two major hazards:

  • The failure to recognise a food allergen that has been included in a dish as an ingredient and
  • The accidental contamination of an otherwise allergen-free food with an allergen.

But while the failure to recognize an allergen has been the cause of the tragic, but mercifully few deaths, it is accidental contamination that currently gets most of the attention.
(For a overview of allergy fatalities see this article by Hazel Gowland and Michael Walker.)

The allergen as an ingredient

Since 2005 all the ingredients in any prepacked food must be listed on the pack and, as of December 2014, all of the 14 major allergens must be highlighted in that list.

Although this is basically a good system, mistakes do occur: the wrong dish gets packed in the wrong pack, an allergen gets left out of an ingredients list, recipes get changed without the packaging being updated and so on. But these incidents are relatively rare.

However, the system does require allergy sufferers to be sufficiently allergen aware to be able to recognize their allergens and avoid products which include them.

Food service

As of December 2014, food service providers must also be able to identify any of the 14 major allergens used in their food. This means that everyone in the service chain – front of house/waiting staff, management, chef and kitchen staff – needs to have a reasonable knowledge of those allergens and to know whether the chef has actually included any of them in that service.

However, there are a large number of food service outlets who are scarcely aware that new allergen regulations exist, or who only have the sketchiest knowledge of what they require. This does mean that for a food allergic person, eating out remains a potentially dangerous occupation. Certainly the vast majority of the deaths that have occurred as a result of food anaphylaxis have occurred in a catered situation.

But, although individual outlets need to know what is in their food and be able to convey this information clearly to their allergic/coeliac customers, those customers also need to understand about their own allergens. Not only do they need to know what they may be called and where they may be found but they must be prepared to engage with the outlet to ensure that they get a safe meal.

PeanutsThis may seem obvious, yet a significant number of deaths have been caused either by the allergy sufferer failing to understand that, for example, satay sauce is normally made from peanuts, or failing to take account of the fact that, for example, peanuts are widely used in Indian cooking and that therefore, take-away curries will always be dangerous for a peanut allergic person.

Allergic consumer education

The Food Standards Agency, the support networks and the medical fraternity do a lot to raise awareness among allergic consumers but it is arguable that a great deal more needs to be done to teach allergic people how to manage their own allergies.

Especially since on going allergy research suggests that there may be many factors other than the allergic protein consumed which contribute to the severity of a reaction: the state of health of the allergic person at that moment; if they are a woman, if they are menstruating; if they have just taken exercise; what else they are eating with the food etc.

While food manufacturers and regulators can do their best to reduce risk to an absolute minimum, it is the allergy sufferer who has the greatest interest in understanding possible risk and is in the best position to make a proper risk assessment on each eating occasion. But to do so, they must know how.

Allergen contamination

It is clear that the greatest risk to the allergic consumer is being unaware that an allergen has been included, either deliberately or accidentally, as an ingredient in a dish. But there is also the risk that a perfectly harmless, allergen-free dish might have been contaminated by an allergen.

If the same equipment is used to manufacture both allergen-free and allergen-containing foods, whether it in a factory or a food service outlet, then, unless the cleaning is extremely thorough, there is a risk that allergen residues could remain and be transferred to the allergen free dish. Good practice should reduce that risk to extremely small but there remain foods (such as chocolate) the nature of which is such that it is almost impossible to clean equipment thoroughly enough to removed all allergen traces, and where a real risk may remain.

But while this contamination remains a genuine risk, in most cases it is actually very small – far smaller than the risk posed by the wrong food being packed in the wrong box or the consumer not recognizing one of the ingredients as an allergen. But, as far as the allergic consumer (and, indeed, medical researchers) are concerned – this risk, and how great it actually is, seems to have become the main focus of their concern.

Establishing risk levels

The goal to which medical research is striving is to set ‘thresholds’ for allergens below which the vast majority of allergy sufferers will not react. The equivalent to the 20 parts per million of gluten now set as a ‘safe’ threshold for coeliacs allowing them to consume any product with less than 20ppm of gluten without fear of reaction. However, this is far from as simple as it might appear.

In order to establish those levels for allergens such as milk, nuts, soya, sesame etc, the European research community has spent the last ten years doing massive studies of the allergic population and their reactions. But the closer they get to drawing appropriate conclusions, the more complicated it becomes.

For example, 99% of the milk allergic population may not react to a single dose of milk at, say, 5 parts per million – but what happens if they consume several doses? How do you control portion size for those doses? How does that dose relate to body size – a slim young teenager or a 15 stone middle aged man? More over, an allergen will have different reactivity depending on what it is eaten with so how do you control for that? And what happens if the allergic person is suffering from a cold or a virus – or if they have just been running – will that make them more reactive? Moreover, it ignores the fact that within each allergen (peanuts or milk, say) there are a number of different proteins to which an allergic person may react – and testing is not yet sufficiently sophisticated to identify either which protein they might react to or which protein is actually in the food.

An added problem is that these ‘thresholds’ are a European initiative so have to grind through Brussels – so it may be years before they are actually agreed.

Precautionary allergen labelling

May containMeanwhile, the allergic community, medics, industry and allergics, are left in limbo – their fall back position being to label everything which cannot be 100% guaranteed allergen free (an impossibility, no matter what the circumstances) as ‘may contain/may contain traces etc etc of x allergen’.

As far as the allergic consumer is concerned this either dramatically restricts what they are able to buy/eat or leaves them open to actual risk. Since ‘may contain’ warnings give no indication of the level of risk that particular food presents they are often ignored; a very low risk option in 90% of cases, but a very real risk in the remaining 10%.

As for the freefrom food industry, because there are currently no ‘levels’ that are deemed safe, legal teams often insist on ‘may contain’ warnings so they are not left open to prosecution if even the minutest amount of the allergen is ever found in the food.

For the medical view see this article published last year by a group of allergy research specialists.

One way in which the risk can be dramatically lessened for all concerned is if the allergen free products are manufactured (or cooked) in dedicated nut free, dairy-free, gluten-free etc factories or outlets. But while this constitutes the ‘gold standard’ for allergen control it is very expensive and likely to remain a minority option for a long time to come.

So a better way has to be found.

Changing attitudes

Over the last ten years, allergy management has done a 180 degree turn – from ‘avoidance’ to ‘inducing tolerance’. The recent LEAP study results showing that feeding infants peanuts from infancy dramatically reduced their chances of becoming peanut allergic and the successful outcome of the peanut immunotherapy trials at Addenbrookes last year, are only the most public instances of this new attitude.

Along with this new approach to inducing tolerance comes a more nuanced approach to the issue of accidental contamination. While there are no regulations in this area, the Food Standards Agency has been very specific in its guidance and its focus is very much on a risk based approach:

Advisory labelling on possible cross-contamination with allergens should be justifiable only on the basis of a risk assessment applied to a responsibly managed operation. Warning labels should only be used where there is a demonstrable and significant risk of allergen cross- contamination, and they should not be used as a substitute for Good Manufacturing Practices.

Looking at what evidence there is for contamination-based reactions (as opposed to reactions to foods where the allergen has been intentionally included but mis-identified) this seems a sensible way to go. However, it depends:

  • on industry (prepackaged and food service) both understanding and applying the principles of comprehensive risk assessment.
  • on the allergic community themselves understanding their own allergies and allergens and being able to make their own personal risk assessments in each case
  • and – the biggest ask – the allergic community trusting industry to have done their risk assessment properly.

No matter how sensible this approach may be it is asking a lot of an allergic person, let alone the parents of a severely allergic child. Effectively you are asking them to step out of the relative comfort of just excluding everything which might pose the slightest risk the themselves or their child, to taking on the responsibility for assessing the risk themselves on each eating occasion.

However, if they wish to broaden their options – and in terms of children, if they want their children to be able to manage their own allergies safely and responsibly when they grow up and leave home – then maybe it is an option that they need to consider.



Way back in 1987, just as I was starting work on a major history of English food, my eighteen-month-old son, Jonathan, and his father were diagnosed with dairy intolerances. Back then the alternatives for those on dairy-free diets were few and far between and pretty unappealing so, after some months of experimentation, I launched Berrydales Special Ices, soya based ices which were dairy and additive free – and tasted delicious! While manufacturing the ices I started a newsletter, The Inside Story, about food allergy and food intolerance and, by 1995, it was a quarterly magazine circulating to over 35,000 health professionals. In 2000 The Inside Story, re-named Foods Matter, became a subscription magazine and now all of that information, and much, much more, is accessible on the Foods Matter, Coeliacs Matter and Skins Matter sites and on our two freefrom food sites, FreeFromFoodsMatter and FreeFromRecipesMatter. You can follow me on twitter @FoodsMatter or email me at And, of course, you can also follow the exciting growth of freefrom food by checking out our annual FreeFrom Food Awards celebrating the best and the newest in freefrom foods!

Add a comment

Your email address will not be published. Required fields are marked *